Thomas O. Eichmann

Biogenesis and Catabolism of Diacylglycerols

- Role of Stereochemistry -

DOCTORAL THESIS/Dissertation

submitted to the Faculty of Natural Sciences

at the University of Graz (Austria)

for attainment of the degree

Doctor of Natural Sciences

Doctor rerum naturalium

(Dr. rer. nat.)

October 2012

Sometimes the questions are complicated and the answers are simple.

Theodor Seuss Geisel, US author

Gewidmet jenen, die mich vorbehaltlos unterstützen.

- Meine Familie, Freundin und Freunde -

Preface

Herewith I declare that this doctoral thesis has been written independently and without any

assistance from third parties. Moreover, references employed for the composition of this manuscript

are cited in the bibliography and no other sources were used.

____________________________ _______________

Mag. rer. nat. Thomas O. Eichmann Date (22.10.2012)

Acknowledgement/Danksagung

I want to express my sincerest thanks to…

… Rudolf Zechner for giving me the great opportunity to work in this outstanding laboratory.

Moreover, I want to thank for his supervision, motivation and support.

… Achim Lass for all the discussions and his invaluable guidance and permanent

encouragement throughout all these years.

… Robert Zimmermann, Günter Hämmerle, Karin Preiss-Landl and Fritz Spener for their

encouragement.

… the Doktoratskolleg (DK) Molecular Enzymology for funding my Ph.D. work and giving me

the opportunity to exchange and discuss ideas and problems with other great scientists.

… Robert V. Farese Jr. and Günter Hämmerle for contributing to this work by providing

enzymes and mouse models, respectively.

… all members of our laboratories that made this institute a wonderful place to work. In

particular, I want to thank those people who became more than just colleagues, Martina

Schweiger, Gabriele Schoiswohl, Franz Radner, Christoph Heier, Manju Kumari,

Chandramohan Chitraju, Matthias Romauch, Tarek Moustafa, Harald Hofbauer, Martin

Kreim, Kathrin Zierler, Ulrike Taschler, and Renate Schreiber for a great time.

Mein größter Dank aber gilt meiner ganzen Familie und meiner Freundin. Meinen

Eltern Walter und Inge und meinen Geschwistern Michael, Katja, Jörg und Teresa, welche

mich in jeder Lebenslage vorbehaltlos unterstützen und immer hinter mir stehen. Großer

Dank gilt auch meiner Freundin Anja, die mir immer zur Seite steht und auch in schweren

Zeiten eine unschätzbare Stütze ist. Ich danke auch meinen Freunden, Florian, Patrick,

Manfred und Rupert, welche mich seit Langem begleiten und mir stets Rückhalt bieten.

Abstract of the Doctoral Thesis Submitted to the Faculty of Natural Sciences at the University of Graz for Attainment of the Degree Doctor of Natural Sciences

Biogenesis and Catabolism of Diacylglycerols

- Role of Stereochemistry -

Mag. rer. nat. Thomas O. Eichmann

Institute of Molecular Biosciences

University of Graz

In mammals, excessive energy is stored in form of energy-dense triacylglycerol (TAG). A tight balance

between TAG synthesis and degradation is instrumental for maintaining energy homeostasis.

Dysregulation of energy homeostasis is strongly associated with obesity and related disorders, like

insulin resistance (IR) and type 2 diabetes mellitus (T2DM). Adipose triglyceride lipase (ATGL) is rate-

limiting for the initial step of TAG hydrolysis, generating diacylglycerol (DAG) and fatty acids (FAs).

Accumulations of both lipids are correlated to defective insulin signaling, hence provoking IR and

T2DM. DAG exist in three different stereochemical isoforms (sn-1,2; sn-1,3; sn-2,3), of which

exclusively sn-1,2 DAG affects insulin signaling via activation of protein kinase C (PKC). So far the FA-

and positional- (stereo-) selectivity of ATGL-dependent TAG hydrolysis is unknown. Yet a direct link

between an imbalance in TAG hydrolysis and defective insulin signaling is likely. The objective of this

study focused on the eludication of the FA- and stereoselectivity of ATGL as well as the

stereoselectivity of enzymes involved in further DAG utilization. Results reveal that ATGL exhibits a

strong preference for the hydrolysis of long-chain FA esters at the sn-2 position of TAG. This

selectivity broadens to the sn-1 position upon stimulation of the enzyme by its co-activator

comparative gene identification-58. Furthermore, ATGL-derived sn-1,3 DAGs are the preferred

substrate for the consecutive hydrolysis by hormone-sensitive lipase. Interestingly, also

diacylglycerol-O-acyltransferase 2 (DGAT2), which is a key enzyme of TAG synthesis, preferentially

esterifies sn-1,3 DAG. This suggests, that ATGL and DGAT2 act coordinately in the hydrolysis/re-

esterification of TAG and that ATGL creates a distinct pool of sn-1,3 DAG. The inability of ATGL to

generate sn-1,2 DAG suggests that TAG-derived DAG cannot directly impair insulin signaling via PKC

activation.

Table of contents

Introduction ................................................................................................................................... 1

Stereochemistry of DAG ........................................................................................................... 3

Metabolic formation of DAG .................................................................................................. 6

A) Formation of DAG by intracellular lipases ............................................................................ 7

B) De novo synthesis of DAG ..................................................................................................... 12

Utilization of DAG ...................................................................................................................... 15

DAG and DAG-derived signals .............................................................................................. 20

Aim of the Thesis ..................................................................................................................... 29

Results.............................................................................................................................................. 31

I) ATGL selectivity ...................................................................................................................... 32

A) Stereo/regioselectivity ........................................................................................................... 32

B) FA selectivity ........................................................................................................................... 37

C) Substrate selectivity ............................................................................................................... 47

II) Selectivity of DAG hydrolysis and re-esterification ................................................ 51

A) Stereo/regioselectivity of DGAT enzymes ........................................................................... 51

B) Stereo/regioselectivity of HSL-dependent DAG hydrolysis ............................................... 60

Discussion ...................................................................................................................................... 64

Materials & Experimental Procedures ..................................................................... 76

I) Materials ................................................................................................................................... 77

II) Experimental Procedures .................................................................................................. 78

Publications ................................................................................................................................. 86

First author .................................................................................................................................. 87

Co-author ..................................................................................................................................... 88

Appendix ........................................................................................................................................ 90

Abbreviations & Acronyms ........................................................................................................ 91

Bibliography ................................................................................................................................ 95

Introduction

Introduction

2

Energy homeostasis constitutes a primary necessity of living organisms to ensure constant metabolic

fluxes and physiological flexibility. The storage of excessive energy metabolites displays an

evolutionary highly conserved strategy, to mobilize energy reserves and consequently facilitate

survival in times of increased energy demand or inadequate nutrient supply. In mammals, a dietary

surplus of either carbohydrates or fat is converted to inert and energy-dense triacylglycerol (TAG). In

TAG the fatty acids (FAs), which are esterified to the glycerol backbone constitute the main source of

energy substrates. In higher animals, the highly hydrophobic TAG molecules are embedded within

glycerophospholipid (PL)-coated lipid droplets (LDs) and stored in almost every cell type. In large

quantities such LDs are found in adipocytes of the white adipose tissue (WAT). An efficient

mobilization of these TAG stores is required to maintain a constant whole body energy supply. Thus

metabolic pathways which balance TAG synthesis and degradation need to be precisely regulated.

This evolutionary important ability to store excessive energy has evolved to a human burden. In the

western world countries times of caloric scarcity are rare and the oversupply of inexpensive, calorie-

dense food often goes along with little to no caloric demands. Metabolic alterations following

overnutrition lead to dysregulations of the energy equilibrium. The resulting obesity may eventually

also cause disorders such as non-alcoholic fatty liver disease, coronary heart diseases, and typ 2

diabetes mellitus (T2DM), all summarized in the term metabolic syndrome [1, 2].

The cellular metabolism of TAG is accomplished by a variety of enzymes including lipases and

acyltransferases. The intermediates diacylglycerol (DAG), FAs and coenzyme A-activated FAs (FA-CoA)

constitute indispensable, bioactive precursors that are involved in a variety of metabolic processes.

Adipose triglyceride lipase (ATGL) [3-5] represents the major lipase responsible for the initiation of

TAG mobilization in adipocytes. FAs of this reaction are thought to be released to circulation to

supply peripheral tissues either fuelling mitochondrial beta-oxidation, act as precursor for newly

synthesized lipids, or act as ligands for a variety of nuclear receptor transcription factors. The

hydrolysis of TAG by ATGL also generates DAG which may function as substrate for further

degradation, as precursor for TAG and PL synthesis, or as potential signaling molecule.

Dysregulations of ATGL, which lead to either blunted or excessive lipolysis, may also result in altered

levels of this metabolic-active lipid species in various tissues. Such fluctuations in bioactive lipid

content may have non-conceivable consequences in regard to cellular signaling and are often

interrelated with the development of diseases that are part of the metabolic syndrome.

For the understanding of the interrelation between altered energy metabolism and disturbed cellular

signaling, the biochemistry of lipid intermediates, namely DAGs and FAs, as well as their physiological

fate is of great importance.

Introduction

3

Stereochemistry of DAG

Lipolysis is the catabolic arm of TAG metabolism, which liberates FAs from inert TAG stored in LDs. It

is catalyzed by lipases, which represent a special class of serine hydrolases. Generally, hydrolases are

characterized as enzymes catalyzing the reversible, hydrolytic cleavage of various chemical bonds

including carboxylic ester, ether, amide and peptide bonds. Lipases exclusively catalyze the hydrolysis

of carboxylic esters in aqueous environment and the reverse reaction in organic milieu [6]. One

fundamental difference between hydrolases/esterases and lipases is the fact that esterases

encounter water-soluble substrates, whereas lipases are activated by oil/water interfaces (e.g. lipid

emulsions in water) [7]. This unique ability was recognized early in studies with pancreatic lipase [8,

9]. The structural factor determining interface-activation is a mobile element, called lid, which is

detectable in almost all lipases [10]. This mobile lid caps the substrate binding site in the absence of

an interface. The open form, which is thought to be stabilized in a hydrophobic environment, makes

the substrate binding site accessible and hence elevates enzymatic activity [11]. Follow-up studies

verified the fundamental difference of esterases and lipases in that esterases show normal Michaelis-

Menten kinetic whereas lipases sharply increase activity upon exceedance of the substrate solubility

limit, which yields in an interface formed by emulsified substrate molecules [12]. This privileges

lipases to release FAs during hydrolysis of lipids (e.g. glycerolipids, cholesterylesters), which are

usually stored in PL-coated micelles or are embedded within biological membranes. A second

characteristic, which distinguishes lipases from other esterases is the feature of particular

stereochemistry. Lipases can encounter both chiral and prochiral substrates. This ability makes them

unique compared to other hydrolases, like proteases, phospholipases, and nucleases, which can only

hydrolyze one optical form of their respective substrate [13, 14].

Isomers (from greek: isos – equal, mèros – part) are molecules sharing identical molecular formulas

but differ in structure. Isomerism can be divided into two main groups. On the one hand structural

isomers, which exhibit differentially linked atoms and functional groups. On the other hand spatial

isomers (stereoisomers), which display same linkage of atoms and functional groups but differ in

their geometrical position in space. A special group of stereoisomers, named enantiomers, is related

by reflection, which implies that two enantiomers are not superimposable. Furthermore,

enantiomers are characterized by an asymmetric or chiral carbon atom, featured by four different

ligands (Fig. 1).

Introduction

4

FIGURE 1. Compilation of the different forms of isomerism on the basis of DAG. DAGs feature different forms of

isomerism and can differ either in structural or spatial conformation.

DAGs arise during a variety of metabolic reactions and are important signaling molecules. They

illustrate a lipid class that exhibits different isomeric properties. TAG, one possible metabolic

precursor of DAG, contains three FAs esterified to a glycerol backbone. This implicates that TAG

provides three possible sites for lipase-dependent hydrolysis, which potentially lead to three

different DAG isoforms. Herein, the stereospecific numbering (sn) indicates the position of the FA at

the glycerol backbone (sn-1, sn-2, sn-3). Besides chirality, TAGs exhibit another important property of

lipase substrates, namely prochirality. Prochirality describes the condition that an achiral molecule

can be converted into a chiral molecule by a single step reaction. In case of TAG esterified with a

single FA species the achiral carbon atom at sn-2 position becomes a chiral center by removal of the

attached FA at either sn-1 or sn-3 position (Fig. 2).

FIGURE 2. TAG lipases can produce a new center of chirality during TAG hydrolysis. Sequence rule order: CH2-O-

COR>R2>R1 (modified after [15])

Introduction

5

Certain TAG species exhibit chemically identical but enantiotopic reactive groups (pro-R and pro-S;

e.g. oleic acid (C18:1) at sn-1 and sn-3 position), which are chirally discriminated during lipase-

dependent hydrolysis reaction, yielding a chiral DAG product. Lipase-dependent cleavage of a FA at

either sn-1 or sn-3 position of a TAG molecule leads to one of the two enantiomers, sn-2,3 or sn-1,2

DAG, respectively. These DAG enantiomers face themselves as reflection and are not

superimposable. Furthermore, enantiomeric DAGs can be classified in respect to the R/S

configuration nomenclature based on the Cahn-Ingold-Prelog (CIP) system [16, 17]. The CIP system is

used to uniquely specify enantiomeric molecules. Therefore, priorities are assigned to all groups

attached to the chiral center (CIP-rules) [16, 17]. Subsequently, the lowest ranked group is set below

the image plane and the other groups are counted starting at highest priority substituents. The

counted sequence can be either clockwise or counterclockwise and specifies the present molecule as

either R-configurated (from latin: rectus - right) or S-configurated (from latin: sinister - left). In case

of DAG, sn-1,2 DAG reflects the S-configuration whereas sn-2,3 DAG is R-configured (Fig. 3).

FIGURE 3. R/S nomenclature of DAG enantiomers according to Cahn-Ingold-Prelog convention. sn-1,2 DAG represents the

S-configuration whereas sn-2,3 DAG represents the R-configuration.

Upon hydrolysis of sn-2 bound FA esters the generated DAG exhibits sn-1,3 conformation and owns

no chiral center. In regard to the different region of hydrolysis (sn-1/sn-3: esters of a primary alcohol;

sn-2: ester of a secondary alcohol) sn-1,3 DAG is a so called regiomer (Fig. 1). Thus, lipases can

regioselectively differentiate between sn-2 and sn-1/sn-3 position or enantioselectively differ

between sn-1 and sn-3 position. Similarly, other enzymes, like acyltransferases can potentially

discriminate different isoforms of DAG.

These differences regarding isomerism of DAG as well as the selectively of metabolic DAG-

generating/consuming enzymes could be an important issue when lipids, like DAG, which display

intersections between lipid and signaling metabolism, are investigated. So, the selectivity of cellular

Introduction

6

lipases/acyltransferases as well as the impact of DAG isomerism on cellular metabolism is obviously

very important.

Metabolic formation of DAG

Intracellularly, several reactions which contribute to the generation of DAG are located at different

subcellular compartments including the endoplasmic reticulum (ER), LDs, and the plasma-membrane

(Fig. 4). Therefore, either TAGs stored in cytoplasmic or ER-associated LDs or PLs, which assemble

cellular membranes can act as source material for lipase-dependent generation of DAG (Fig. 5).

Additionally, DAGs can arise during de novo synthesis of TAG either generated by acyltransferases or

phosphohydrolases. The stereo/regioselectivity of enzymes involved and thus the isomerism of the

formed DAGs is widely unknown but might play a crucial role for subsequent cellular reactions. The

following sections describe known biochemical characteristics and stereo/regiochemical properties

of enzymes involved in the formation of DAG.

FIGURE 4. Catabolic and anabolic reactions leading to the formation of different DAG isoforms. DAG can be generated by

hydrolysis of either TAG or PLs, by dephosphorylation of phosphatidic acid, or by the esterification of monoacylglycerol

catalyzed by certain acyltransferases.

Introduction

7

A) Formation of DAG by intracellular lipases

Within most cell types, TAG turnover is crucial to balance energy storage and distribution. In whole

body energy metabolism this function is mainly achieved by specialized cells, named adipocytes,

present in WAT. In adipocytes, excessive energy is stored in form of TAG departed in cytoplasmic LDs.

Interestingly, the size of LDs is different in adipocytes as compared to other cell types. Whereas

adipocytes harbor usually a single LD in a size range about 100 µm, non-adipose tissue cells exhibit

usually multiple LDs with diameters of around 1 µm. Nevertheless, all LDs share basically the same

architecture. The core is strictly assembled of hydrophobic lipid esters, like TAG and cholesteryl

esters (CEs), and the surface is formed by a PL monolayer [18]. This monolayer harbors a variety of

anchored or embedded proteins and serves as an amphipathic shield against the aqueous milieu,

which is present in the cell [19]. The most important function of adipocyte LDs is the storage and

lipase-dependent release of energy metabolites, primarly in form of FAs. This tightly regulated

process of TAG degradation, known as lipolysis, generates stepwise lipid intermediates like FAs, DAG,

and monoacylglycerol (MAG) and is executed by a cascade of lipases. Therein, HSL was first identified

and thus thought to be essential for the initial step of TAG hydrolysis generating potential signaling

molecules, more precisely DAG and FAs.

FIGURE 5. Formation of DAG by intracellular lipases. DAG can be generated by hydrolysis of TAG or PLs. Intracellularly,

ATGL and HSL are the main TAG lipases. The generation of DAG from PLs is catalyzed by PLC. ATGL, adipose triglyceride

hydrolase; DAG, diacylglycerol; HSL, hormone-sensitive lipase; PLC, phospholipase C; TAG, triacylglycerol.

Hormone sensitive lipase (HSL)

Following the observation that WAT lipolysis is strongly inducible by hormonal stimulation, HSL was

the first characterized TAG lipase in WAT [20, 21]. HSL is known to exhibit a uniquely-broad substrate

spectrum, which includes TAG, DAG, MAG, CE, retinyl ester (RE) as well as water-soluble short chain

Introduction

8

esters [21-25]. However, in vitro studies showed that the specific activity of HSL is highest against

DAG, which exceeds those against TAG and MAG around 10-fold [23, 26]. Earlier studies described a

sn-1/(3) specificity of HSL for DAG [26], whereas most recently HSL was identified to be quite sn-3

specific [27]. Concerning substrate specificity, HSL exhibits preference for polyunsaturated FAs

(PUFAs; n-3/n-6), which was shown using crude preparations of rat HSL [28]. Although the preference

for the hydrolysis of DAG was noticed, HSL was long considered to be the rate-limiting lipase in TAG

mobilization of adipose and non-adipose tissues. Albeit HSL is expressed in many tissues, protein as

well as mRNA expression of HSL are highest in WAT and brown adipose tissue (BAT) [29]. The COOH-

terminal region of HSL harbors the α/β hydrolase fold domain, containing an active site serine (Ser423)

as part of a catalytic triad (Asp703, His733) responsible for hydrolytic activity [22, 30-32]. The lipid

binding site as well as the site responsible for protein dimerization was found at the NH2-terminal

region of HSL [33].

HSL activity is strongly regulated by hormones. A variety of phosphorylation events control HSL

activity by affecting both intracellular localization and protein-protein interactions. The major

positive stimulus is caused by catecholamines, which bind to β-adrenergic receptors during periods

of nutritional deprivation (fasting). The contrary nutritional condition (feeding) inhibits HSL action via

insulin. The activation of HSL upon β-adrenergic stimulation is primarly mediated by protein kinase A

(PKA)-dependent phosphorylation [34-40]. Additional phosphorylations, involved in the regulation of

HSL, are catalyzed by AMP-activated kinase (AMPK), extracellular signal-regulated kinase (ERK),

glycogen synthase kinase-4 as well as Ca2+/calmodulin-dependent kinase [41]. The major

phosphorylation sites controlling increased activity have been shown to be Ser659 and Ser660

(numbered for rat HSL), which are phosphorylated by either PKA or ERK [36, 42, 43]. In contrast,

phosphorylation of Ser565 by AMPK leads to antilipolytic effects by potentially antagonizing PKA-

dependent phosphorylation [34, 44]. All above mentioned phosphorylation sites are located in the

regulatory module (~150 amino acid (AA) loop) within the COOH-terminal region of the enzyme. Even

though phosphorylations of HSL are crucial, they cause only moderate changes in enzymatic activity.

The more important factor in HSL activation is the binding of the enzyme to LDs. In adipocytes, LDs

are shielded by perilipin-1, which surrounds LDs, forming a barrier between lipases and respective

lipid substrates [45]. In basal condition this LD barrier keeps lipolysis at a low rate [46]. Upon β-

adrenergic stimulation perlipin-1 is phosphorylated by PKA at six consensus serine residues (Ser81,

Ser222, Ser276, Ser433 Ser492, and Ser517; numbered for murine perilipin-1) [47-50]. This facilitates

binding of HSL to perilipin-1 at the NH2-terminal region, leading to a translocation of HSL to the LD

surface [51, 52] where it deploys full activity. The inactivation of HSL in adipocytes is triggered by

insulin as a consequence of nutrient uptake. In the cyclic AMP (cAMP)-dependent pathway, insulin

activates a variety of phosphodiesterases, which hydrolyze cAMP resulting in reduced cAMP levels

Introduction

9

and the loss of PKA activation [53, 54]. In the cAMP-independent pathway, insulin induces protein

phosphatase-1, which leads to HSL dephosphorylation and inactivation [55].

The long standing dogma that HSL acts as pacemaker of lipolysis, thereby hydrolyzing TAGs and

DAGs, was disproven when mice, carrying a global deletion of HSL (HSLko) showed no signs of obesity

on a high-fat diet (HFD) but demonstrated normal bodyweight and reduced fat mass as compared to

wildtype (wt) mice [56, 57]. Decreased adipose mass was partially explained by reduced FA

esterification counteracting decreased lipolytic activity [58]. Furthermore, HSL-deficient adipocytes

still showed a catecholamine dependent increase in FA release from WAT, which suggested an

additional lipase catalyzing lipid degradation [56, 59, 60]. The most intriguing finding in HSLko mice

was the drastic accumulation of DAG in several tissues [59] suggesting that HSL is responsible for

DAG hydrolysis.

Adipose triglyceride lipase (ATGL)

In 2004, a lipase which fulfilled all postulated requirements was identified independently by three

laboratories. The former denoted transport secretion protein-2.2 was renamed as ATGL/patatin-like

phospholipase domain containing A 2 (PNPLA2) [3], desnutrin [4], and calcium independent

phospholipase A2ζ [5].

ATGL is one of nine PNPLA family members found in humans (PNPLA1-9) [61]. The PNPLA protein

family is named after the patatin-domain, which was first identified in hydrolases of plants and

denominated after the most abundant protein of the potato tuber, patatin. Lipid hydrolases

containing this domain catalyze the non-selective hydrolysis of a variety of lipids, including PLs,

glycolipids, DAGs, and MAGs [62-64]. In addition to ATGL also other members of the PNPLA family

possess hydrolase (PNPLA3/adiponutrin; PNPLA4/gene sequence 2, GS2; PNPLA5/GS2-like),

phospholipase (PNPLA8 and 9), or lyso-phospholipase (PNPLA6 and 7) activity [5, 65, 66]. Within the

NH2-terminal half of human ATGL, the patatin domain (180 AA) is embedded in a 250 AA sized α/β/α

sandwich structure and includes a conserved serine lipase consensus sequence motif (GXSXG), which

contains the active site serine (Ser47) [66-68]. In contrast to the catalytic triad of classical lipases, the

enzymatic activity of ATGL relies on a catalytic dyad formed by Ser47 and Asp166, both identified to be

essential for hydrolytic activity [62, 69]. The COOH-terminal half of ATGL contains a lipid-binding

domain formed by a hydrophobic AA stretch (AA 315-360) [70, 71] as well as two putative

phosphorylation sites (Ser404, Ser428 in human ATGL or Ser406, Ser430 in murine ATGL) [72, 73].

Orthologs of ATGL exist in almost every eukaryotic species including invertebrates, fungi, and plants.

In contrast to the broad substrate spectrum of HSL, ATGL is highly specific for TAG and shows only

Introduction

10

weak or no activity against DAG, MAG, CE, or RE [3]. Furthermore, phospholipase A2 (PLA2) as well as

DAG transacylase activity was reported for ATGL [5, 74]. Yet, the physiological relevance of these

relatively minor activities is not established.

The key-role of ATGL in degradation of TAG is demonstrated by ATGL-deficient mice (ATGLko).

ATGLko mice exhibit 2-fold higher whole body fat mass as well as drastically enlarged adipose tissues

[75]. TAG accumulation is observable in all tissues reaching up to a 10-fold increase in TAG content

[75]. Furthermore, a dramatic accumulation of TAG in cardiomyocytes of ATGLko mice led to a

premature death of these animals due to severe cardiac dysfunction. In accordance to TAG

accumulation, isoproterenol-stimulated lipolytic activity of WAT explants is decreased around 70% as

compared to wt mice. Additionally, in vitro TAG hydrolase activity assay demonstrated the crucial

role of ATGL in the mobilization of TAG in WAT, BAT, cardiac muscle (CM), skeletal muscle (SM),

testis, and liver [75].

In mice, ATGL mRNA is expressed in all examined tissues. Highest expression is observed in WAT and

BAT. Lower expression levels are detectable in SM, CM, liver, and testis [3, 4, 66, 76]. In adipocytes,

ATGL expression is markedly upregulated during differentiation and highest when LD accumulation is

observable [3, 4]. Furthermore, ATGL shows increased expression upon fasting, treatment with

glucocorticoids [4], i.e. dexamethasone, or peroxisome proliferator activated receptor γ (PPARγ)

agonists, like thiazolidinedione [76-78]. Despite a downregulation of ATGL expression in murine

obesity models, like leptin- or leptin receptor-deficient mice (ob/ob; db/db), ATGL expression is

decreased upon feeding and by hormones associated with obesity like insulin, isoproterenol, and

TNF-α [79-82]. However, mRNA levels of ATGL do not always correlate with lipase activity, which

could be explained by substantial posttranslational regulation of ATGL. Posttranslational regulation

may include phosphorylation of Ser406 and Ser430 but reports regarding phosphorylation of ATGL are

controversial. Ser406 is phosphorylated by PKA [83] and/or AMPK [73], thereby increasing ATGL´s

hydrolase activity. In contrast, other publications showed an inhibitory effect of AMPK on TAG

hydrolysis [84] as well as a PKA-independent phosphorylation of ATGL [3].

Besides regulation via phosphorylation or nutritional condition, full stimulation of ATGL´s activity

requires the presence of an LD-associated protein named comparative gene identification-58 (CGI-

58; or α/β hydrolase fold domain containing protein 5, ABHD5) [68]. CGI-58 is highly expressed in

testis and shows lower expression in WAT, CM, SM, and liver [68]. Additionally, in vitro TAG-

hydrolase activity of cell lysates expressing murine ATGL as well as of WAT lysates is 20-fold and 2-

fold increased upon addition of exogenous CGI-58, respectively [68, 85]. Although CGI-58 belongs to

a lipase subfamily characterized by α/β hydrolase folds it does not possess intrinsic hydrolase activity

since a required putative active site serine within the canonical lipase motif (GXSXG) is replaced by an

Introduction

11

asparagine (Asp153) [86]. Noteworthy, two independent laboratories described CGI-58 as an acyl-CoA-

dependent lysophosphatidic acid acyltransferase (LPAAT) in vitro [87, 88] but metabolic implications

are so far unknown.

The molecular mechanism behind CGI-58 dependent stimulation of ATGL remains elusive. However,

a number of studies identified structural key features of both proteins. Whereas truncations of the

COOH-terminal part of ATGL increases the co-activation of ATGL by CGI-58 [70, 89], mutations within

the NH2-terminal region of ATGL result in defective CGI-58 co-activation [70]. In case of CGI-58,

truncation mutants lacking NH2-terminal part failed both to localize to the LD and to co-activate ATGL

[90]. The finding that a removal of three Trp-residues within the NH2-terminal part of CGI-58 results

in defective co-activation of ATGL but did not influence CGI-58/ATGL protein interaction suggests

that both LD binding and protein interaction is essential for activation.

A second protein has been recently identified to be strongly involved in the inhibition of ATGL

activity. The 103 AA sized G0/G1 switch gene 2 (G0S2) originally shown to be involved in cell cycle

transition selectively inhibits ATGL-dependent hydrolase activity [91]. G0S2 is found in many tissues

with highest expression levels in WAT and liver. Without competing CGI-58, G0S2 directly interacts

with ATGL, whereat the hydrophobic region of G0S2 as well as the patatin domain of ATGL are crucial

structural features [91, 92]. Similarly to CGI-58, the biochemical mechanism behind G0S2-depedent

inhibition of ATGL is unknown.

Both ATGL and HSL together are responsible for more than 90% of lipolytic (TAG-hydrolase) activity

in cultured adipocytes and WAT [85]. So, ATGL and HSL, together with monoglyceride lipase (MGL),

which was identified as potent MAG hydrolase [93, 94], represent the three major lipases responsible

for adipocyte lipolysis (Fig. 6).

FIGURE 6. Stereo/regioselectivity of enzymes involved in lipolysis. ATGL and ATGL/CGI-58 hydrolyze TAG with so far

unknown selectivity. In the second step HSL degrades DAG specifically at the sn-3 position. The resulting mixture of MAG

isoforms is further degraded to glycerol by MGL, which exhibits no selectivity. ATGL, adipose triglyceride hydrolase; CGI-58,

comparative gene identification-58; DAG, diacylglycerol; FA, fatty acid; G, glycerol; HSL, hormone-sensitive lipase; MAG,

monoacylglycerol; MGL, monoglyceride lipase; TAG, triacylglycerol.

Introduction

12

Phospholipase C

Phospholipases specifically hydrolyze PLs at different chemical position. The four major classes of

phospholipases are distinguished by the type of catalyzed reaction. Phospholipase A1 (PLA1) and A2

(PLA2) catalyze the hydrolytic cleavage of the acyl chains at respective sn-1 or sn-2 position. In

contrast, phospholipase C (PLC) and D (PLD) cleave phospho-glycerol and phospho-headgroup esters.

Hence, only PLCs contribute to the intracellular formation of DAG. So far, thirteen PLC isozymes were

identified and assigned to six subclasses, namely β (1-4), γ (1-2), δ (1,3,4), ε, ζ, and η (1,2) [95-98].

Virtually all PLC isozymes are highly expressed in different brain regions and only a few (β3, δ1, δ3,

δ4, ε) are distributed to other, peripheral tissues, like liver, SM, or CM [99]. Intracellularly, the

soluble PLC proteins are localized mainly in the cytoplasm. Upon cell activation PLCs translocate to

the plasma membrane and develop catalytic activity [99]. The PLC dependent hydrolysis of

phosphatidylinositol 4,5-bisphosphate (PIP2), a plasma membrane-associated PL, describes a key

event during regulation of a variety of cellular functions. By producing two intracellular messengers,

namely sn-1,2 DAG and inositol 1,4,5-triphosphate (IP3), this reaction mediates activation of protein

kinase C (PKC) as well as intracellular Ca2+ release, respectively [99]. Additionally, PIP2, which usually

exhibits an arachidonic acid (C20:4) at sn-2 position, represents the precursor for 2-

arachidonoylglycerol (2-AG) that is strongly involved in endocannabinoid signaling [100, 101]. Due to

the strict sn-3 position of the phosphate residue, PLC generates exclusively sn-1,2 DAG.

B) De novo synthesis of DAG

In addition to the catabolic formation of DAG, two other pathways contribute to anabolic generation

of DAG (see Fig. 4). In those pathways DAG arises as intermediates of the de novo biosynthesis of

TAG and PLs (Fig. 7). The glycerol-3-phosphate (G3P) pathway is one major pathway of TAG and PL

synthesis and takes place in most tissues, predominantly in liver and WAT. The G3P-pathway begins

with the consecutive acylation of G3P by acyl-CoA dependent glycerol-3-phosphate acyltransferase

(GPAT) and acyl-CoA acylglycerol-3-phosphate acyltransferase (AGPAT, LPAAT). The product of this

reaction is phosphatidic acid (PA), which is dephosphorylated to DAG by PA phosphatase (PAPase,

lipins) [102-104]. In the so called MAG-pathway, MAGs are esterified to DAG by monoacylglycerol-

acyltransferase (MGAT). This pathway plays a predominant role in enterocytes upon feeding and is

also involved in the storage of TAG in adipocytes [105, 106].

Introduction

13

FIGURE 7. De novo synthesis of DAG. DAG can be formed by either G3P or MAG pathway. AGPAT, acyl-CoA acylglycerol-3-

phosphate acyltransferase; DAG, diacylglycerol; G3P, glycerol-3-phosphate; GPAT, glycerol-3-phosphate acyltransferase;

LPA, lysophosphatidic acid; MAG, monoacylglycerol; MGAT, monoacylglycerol-acyltransferase; PA, phosphatidic acid;

PAPase, phosphaditic acid phosphatase.

The G3P pathway

It was recognized quite early that the liver exhibits enzymatic activities to generated PA from glycerol

and that DAG acts as precursor for both phosphatidylcholine (PC) and TAG. In the 1950´s, Kennedy

and coworkers described an enzymatic activity that dephosphorylates PA to form DAG in vitro [107].

This finding completed the enzymatic sequence of TAG and PL synthesis from glycerol. This pathway

is now known as Kennedy-pathway [108, 109].

In mammals, PA dephosphorylation is catalyzed by three recently identified members of the Lipin

family, namely Lipin1, 2, and 3. Among these, Lipin1 is the most extensively studied. It is highly

expressed in tissues with high rates of lipid flux, like CM, WAT, and SM [110, 111]. In WAT and SM,

Lipin1 was identified to account for the entire PAP activity, whereas the other two members

contribute essentially to total PA dephosphorylation of liver, brain and placenta [111, 112].

Furthermore, early studies revealed that lipins locate within the cytoplasm and translocate rapidly to

ER-membranes upon elevated levels of intracellular FAs [113]. Loss-of-function mutations within

Lipin1 cause dramatic metabolic impairments, like hypertriglyceridemia and severe hepatic steatosis,

as observed in fatty liver dystrophic mice [110]. The opposite effect is observed in transgenic mice

overexpressing Lipin1 in adipocytes. These mice exhibit increased amounts of TAG, which fits to

Lipin1 dependent generation of DAG as precursor for TAG [114]. Since G3P and PA are

phosphorylated at the sn-3 position of the glycerol backbone, their dephosphorylation generates

exclusively sn-1,2 DAG.

Introduction

14

The MAG-pathway

The esterification of MAG catalyzed by MGAT enzymes forms the first step in TAG synthesis following

dietary absorption by enterocytes. So far, three enzymes are known to possess MGAT activity,

MGAT1, MGAT2, and MGAT3. All three isoforms are located at the ER [115-119]. Besides similar

intracellular localization, MGAT isoforms differ in their tissue-specific expression pattern as well as in

their specific catalytic activity. In contrast to MGAT1, which is mainly expressed in stomach, AT and

kidney, MGAT2 and MGAT3 exhibit highest expression in small intestine [115-117, 119, 120]. MGAT3,

which is found exclusively in higher mammals, exhibits significantly higher specific DAG-

acyltransferase activity as compared to MGAT1 and MGAT2. This indicates that MGAT3 functions as

TAG synthase [121]. Furthermore, MGAT3 prefers sn-2 MAG as acyl-acceptor and activated palmitic

acid (C16:0) or C18:1 as acyl-donor [117]. Thus, the generated DAGs exhibit either sn-1,2 or sn-2,3

isoform. The prior mentioned lipid intermediates, sn-2 MAG, C16:0 and C18:1, are the major

hydrolytic products of the pancreatic lipase (PAL) dependent TAG hydrolysis during intestinal

digestion [122-124]. Hence, MGAT3 is supposed to be mainly involved in the re-esterification of

dietary absorbed fat within the small intestine.

MGAT2 was identified to possess little or no selectivity regarding chain-length or saturation-level of

FA-CoAs [115]. Furthermore, incubation of MGAT2 with rac-1/3 MAG results in the generation of sn-

1,3 and rac-1,2/2,3 DAG implicating that all position of the glycerol backbone can be esterified by

MGAT2 [115]. In contrast, MGAT1 displays a stricter selectivity for glycerol position as well as for

utilized FA-CoA species. MGAT1 shows highest activity when incubated with long-chain unsaturated

FAs, with the utmost with arachidonic acid (C20:4) [116]. Incubation with either sn-1 or sn-3 MAG

yields in the generation of sn-1,3 DAG suggesting that MGAT1 preferentially esterifies sn-1 or sn-3

position [116]. In line with these findings, incubation of sn-2 MAG results exclusively in the

generation of the rac-1,2/2,3 DAG [116].

In summary, intracellular reactions, which contribute to DAG formation, can generate all DAG

isoforms (Fig. 4). While the stereo/regioselectivity of most TAG lipases is presently unknown, PLC

isozymes generate exclusively sn-1,2 DAG. Additionally, sn-1,2 DAG is also formed during de novo

synthesis via the G3P-pathway. In contrast, DAGs generated by different MGAT enzymes can exhibit

other isomerism, like sn-1,3, depending on the catalyzing MGAT enzyme. The potential significance

of different DAG isoforms in regard to further utilization is addressed within the next sections.

Introduction

15

Utilization of DAG

In general, not only DAG generating enzymes, but also enzymes, which are involved in the utilization

of DAG might exhibit selectivity for specific DAG isoforms (Fig. 8).

FIGURE 8. Several enzyme classes are potentially involved in the utilization of different DAG isoforms. Different DAG

isoforms are assumable substrates for several enzyme classes, including transferases, kinases, and lipases.

Thus, isomerism of DAG species could influence (i) their degradation by lipases which leads to MAG

formation, (ii) their re-esterification to TAG by DAG-specific acyltransferases, (iii) their conversion to

PC by CDP-choline:1,2-diacylglycerol cholinephosphotransferases (CPTs), and (iv) their

phosphorylation to PA by diacylglycerol kinases (DGKs) (Fig. 9). So far, little is known about the

potential impact of DAG isomerism on the activity of the consuming enzymes.

FIGURE 9. Different enzyme classes utilize DAG. DAG is a potential substrat for lipase, kinase and acyltransferase reactions.

CPT, CDP-choline:1,2-diacylglycerol cholinephosphotransferases, DAG, diacylglycerol; DAGL, DAG lipase; DGAT, DAG-O-

acyltransferase; DGK, DAG kinase; MAG, monoacylglycerol; PA, phosphatidic acid; PL, phospholipid; TAG, triacylglycerol.

Introduction

16

DAG acyltransferases

The acylation of DAG is the final step of TAG synthesis. The esterification reaction consumes DAG and

FA-CoA and is catalyzed by diacylglycerol-O-acyltransferases (DGATs). So far, two mammalian DGAT

enzymes have been identified, DGAT1 and DGAT2.

DGAT1 was identified due to high sequence similarity with acyl-CoA:cholesterol-acyltransferase

(ACAT) enzymes. DGAT1 belongs to the large family of membrane-bound O-acyltransferases

(MBOAT) whose members catalyze the transfer of FAs onto thiol or hydroxyl groups of either lipid or

protein acceptors [125]. DGAT1 is highly expressed in small intestine, AT, SM, CM, skin, spleen, and

testis, where it localizes strictly to ER-membranes [126]. DGAT1 contains three transmembrane-

spanning domains and an active site within the COOH-terminal region facing the ER lumen [127]. The

NH2-terminal region, located in the cytoplasm, allows formation of tetramers and binds long-chain

FA-CoAs [128], but is not required for acyl transfer [129]. Whether the active site of DGAT1 localizes

lumenal (latent activity) is conrtoversial. Yet, several studies found mild, latent activity [130-132].

DGAT1 can catalyze a diversity of different acyltransferase reactions including MGAT, monoester wax

synthase, and retinol acyltransferase [133, 134].

DGAT2 shares little similarity with DGAT1. It belongs to a seven-member family including former

mentioned MGAT1, 2, and 3 [118, 135]. All members of this family contain the highly conserved

amino acid sequence HPHG, which in case of DGAT2 is supposed to be part of the active site [118,

125, 136]. Additionally, DGAT2 contains a FLXLXXXn consensus sequence, which displays a neutral

lipid binding domain, found in other neutral lipid metabolizing proteins, like plasma cholesteryl ester

transfer protein, HSL or Triacylglycerol hydrolase/Carboxylesterase 3 (TGH/Ces3) [137, 138]. Within

DGAT2 this domain is responsible for DAG binding, and mutations in this region markedly reduce in

vitro acyltransferase activity [136]. Expression of DGAT2 mRNA is highest in liver, AT, mammary

gland, testis, peripheral leukocytes, and CM [126]. In cultured cells DGAT2 localizes to the ER under

basal conditions. Upon supplementation of FAs and resulting induction of TAG-synthesis, DGAT2

partly localizes to mitochondria-associated membranes (ER domains tightly interacting with

mitochondria) and to LDs [139]. Unlike DGAT1, DGAT2 contains two ER-membrane spanning domains

and both the COOH- as well as the NH2-terminal domain face the cytoplasm [118], suggesting a

spatial difference of DGAT1 and DGAT2-dependent TAG synthesis. Furthermore, DGAT2 does not

possess activity towards substrates other than DAG.

Both DGAT enzymes have been shown in functional studies to be involved in intracellular TAG-

synthesis. Overexpression of either DGAT1 or DGAT2 in mammalian cells causes an increase of in

vitro DGAT activity, whereat specific activity of DGAT1 is significantly higher as compared to that of

Introduction

17

DGAT2 [140]. In contrast, cells expressing DGAT2 show higher TAG mass, bigger LDs, as well as

enhanced glycerol incorporation as compared to cells expressing DGAT1 [140], indicating differences

between in vitro activity and their importance in vivo. DGAT2 shows no preferences regarding

saturation level of FA-CoAs but prefers medium-chained FA-CoAs (C12:0) and short- and medium-

chained DAG species (e.g. C6:0, C8:0, C12:0) as substrate [118, 135]. In contrast, DGAT1 prefers

monounsaturated acyl-donors, like C18:1-CoA, as compared to saturated FA-CoAs, like C16:0-CoA

[118]. Noteworthy, studies on human DGAT1 of the small intestinal revealed equal preference for

both C16:0 and C18:1 [141]. An additional difference is the sensitivity of both enzymes against

magnesium, which was demonstrated by in vitro experiments. High concentrations (> 50mM) are

described to suppress DGAT2 activity whereas DGAT1 activity is much less affected [118].

The differences in subcellular localization as well as in enzymatic activity of DGATs suggest different

intracellular functions. This hypothesis is supported by the phenotypes of DGAT-deficient mice

(DGAT1ko and DGAT2ko). Interestingly, the enzymes cannot functionally compensate for each other

[140, 142]. DGAT1ko mice exhibit a moderate phenotype characterized by reduced TAG levels in

several tissues (e.g. WAT, SM, liver) and resistance against diet-induced obesity [142]. In contrast,

DGAT2ko mice die within the first days of life and suffer from a major skin dysfunction and severe

lipopenia [140]. The exact reasons for these drastic differences are unknown. Recently, distinct

functions of DGAT1 and DGAT2 in hepatic TAG synthesis were reported. DGAT2 is described to

mediate esterification of newly synthesized FAs, whereas DGAT1 catalyzes the synthesis via

utilization of exogenously supplied FAs [143, 144]. Whether this finding holds true for other tissues

needs to be tested.

The regio/stereospecificity of DGAT enzymes is currently unknown. Conclusive studies regarding this

topic have not been performed. Since neutral lipid metabolism leads to the formation of all DAG

isoforms it is conceivable that DAG isomerism influences the activity of DGAT enzymes. A diverging

stereo/regioselectivity of DGAT enzymes might partially explain the differences observed in DGAT1ko

and DGAT2ko mice.

DAG lipases

In addition to HSL, several other cellular lipases possess hydrolytic selectivity for DAG, generating

MAG and free FAs. In humans, two additional sn-1 specific DAG lipases were identified, named DAG

lipase α and β (DAGLα and β) [145]. Both enzymes share ~30% homology and comprise four

predicted transmembrane-spanning domains as well as a serine lipase motif [145]. DAGLα is highly

expressed in brain and pancreas and to a lower extend in AT. DAGLβ shows highest expression in

Introduction

18

bone marrow and the liver [145, 146]. Enzymatic characterization revealed that both are specific

DAG lipases and exhibit a 3 to 8-fold higher selectivity for the sn-1 over the sn-2 position of DAG

[145]. Additionally, DAGLβ prefers DAG species, containing linoleic acid (C18:2) > C18:1 > C20:4 >

stearic acid (C18:0) at sn-2 position, whereas DAGLα shows equal activity against all examined DAG

species [145]. Both lipases strictly localize to the plasma membrane where sn-1,2 DAG is generated

during PLC-dependent hydrolysis of PIP2. DAGL-mediated breakdown of sn-1,2 DAG results in 2-AG,

which is the most abundant endocannabinoid in tissues and acts as ligand for cannabinoid-receptors

(CB1, CB2) [100, 101]. The role of DAGLα and DAGLβ in the biosynthesis of the endocannabionid 2-

AG was established from the phenotype of mice, either deficient in DAGLα or DAGLβ (DAGLαko,

DAGLβko) [146]. In line with the tissue expression pattern, DAGLαko mice display an 80% reduction

in 2-AG levels in brain, whereas DAGLβko mice show 90% reductions in 2-AG levels in liver [146]. To

date it is unknown if plasma membrane-bound, sn-1 specific DAGL enzymes are additionally involved

in the degradation of DAGs, which derive from lipolysis of cytoplasmic TAG stores.

DAG kinases

DGKs catalyze the formation of PA by phosphorylating the free hydroxyl (-OH) group of DAG.

Together with PAPases/Lipins, DGKs are crucially involved in the maintenance of intracellular DAG

and PA levels. So far, ten DGKs isozymes have been identified in mammals [147, 148]. All mammalian

DGK isozymes share two common structural features, the cysteine-rich C1 domain, which is

responsible for DAG binding and potentially involved in protein-protein interaction and a catalytic

domain, responsible for enzymatic activity [149]. Almost every tissue expresses at least one member

of the DGK family. Moreover, numerous tissues express several different DGK isozymes, e.g. all ten

DGKs can be found in brain extracts [150]. DGKs localize to multiple cellular compartments, including

the nucleus, plasma membranes, the cytoskeleton, the golgi apparatus, and the ER [151]. Little is

known about the specific functions of different DGK enzymes with regard to their organelle

distribution. Whereas some isoforms translocate to the plasma membrane upon activating stimuli

(e.g. DGKδ1 upon exposure to phorbol esters, most likely due to attenuate DAG concentration raised

by PLCs), others stay located inside the nucleus potentially regulating nuclear DAG level (e.g. DGKα,

DGKζ) [149].

DGKs act sn-1,2 selective, since their C1 domain shares high homology to sn-1,2 DAG specific binding

motifs C1A and C1B of PKCs [152-154]. Up to now it is questionable if DGK isoforms exhibit activity to

also phosphorylate sn-1,3 or sn-2,3 DAG species. Some more information about DAG selectivity is

available for particular DGKs, like DGKε. DGKε was identified as smallest DGK isoform, strictly located

Introduction

19

at plasma- or ER-membranes [155]. Furthermore, DGKε exhibits specificity for DAG containing C20:4

at sn-2 position [156], which is the product of PIP2 hydrolysis by PLC [157].

DAG choline/ethanolamine phosphotransferases

All tissues and cell types can synthesize PC via the CDP-choline or the “Kennedy” pathway [108, 109].

In an analogue manner, phosphatidylethanolamine (PE) can be formed via the CDP-ethanolamine

pathway. The final step of both pathways, namely the direct conversion of DAG to either PC or PE is

catalyzed by CPT or CDP-ethanolamine:1,2-diacylglycerol ethanolaminephosphotransferase (EPT),

respectively.

In humans, two proteins exhibiting CPT activity have been identified. CPT, which acts as CDP-choline

specific enzyme and CEPT which can utilize CDP-choline as well as CDP-ethanolamine [158, 159].

Both CPT and CEPT proteins are integral membrane proteins and localize at the golgi apparatus and

the ER, respectively [160]. CPT expression is most abundant in colon, intestine, CM, and spleen,

whereas CEPT is expressed in all tissues examined [159]. Due to a cellular excess of CPT activity, both

enzymes are not supposed to be rate-limiting in PC synthesis [161]. Accordingly, overexpression of

either CPT or CEPT does not lead to elevated PC synthesis [162, 163]. Since no purified mammalian

CPT enzyme is available, information about substrate specificity originates from structure/function

analysis of CPT1 of S. cerevisiae. This protein prefers sn-1,2 DAG, more precisely dipalmitolein

(2xC16:1) > diC16:0 = C18:1/C16:0 > C16:0/C18:1 DAG as substrates in vitro [164]. Whether the

mammalian ortholouges exhibit similar stereospecificity needs to be shown.

EPT1 encodes for an enzyme, which in contrast to CPT utilizes CDP-ethanolamine as substrate,

thereby generating PE [165]. Little is known about expression or biochemical properties of EPT1.

Notably, incubation of hepatocytes with radiolabeled ethanolamine results in a highly specific

accumulation of radioactivity in C16:0/C22:6 PE in vivo indicating that EPT activity prefers

C16:0/C22:6 DAG as substrate [166].

Taken together, CPT, CEPT, and EPT are involved in the synthesis of PLs, thereby consuming

specifically sn-1,2 DAG.

Introduction

20

DAG and DAG-derived signals

Most of the reactions afore described either consume or generate FAs and DAGs. DAGs and FAs itself

as well as DAG- and FA-derived lipid species, like MAGs, FA-CoAs and ceramides are bioactive lipid

species, which act as second messengers within different signaling pathways (Fig. 10). Upon

dysregulation of the generation or utilization of DAGs and FAs, such lipid intermediates may

adversely affect cellular signaling.

FIGURE 10. DAG and DAG-derived lipid species are involved in various signaling pathways. Sn-1,2 DAG and ceramides can

activate novel/conventional and atypical PKC isoforms, respectively. FA and FA-CoA are involved in the activation of

different PPARs. The MAG species 2-AG binds to CBRs and activates endocannabinoid signaling. CBR, cannabinoid receptor;

DAG, diacylglycerol; FA, fatty acid; FA-CoA, FA-coenzyme A; MAG, monoacylglycerol; PKC, proteinkinase C; PPAR,

peroxisome proliferator-activated receptor.

A landmark study by Randle and co-workers [167] postulated the negative effects of FAs on insulin-

stimulated glucose oxidation in muscle. According to Randle´s theory, excessive uptake of FAs in the

muscle and concomitant increase of fat oxidation leads to a combined inhibition of glycolytic key

enzymes including pyruvate dehydrogenase, phosphofructokinase, and hexokinase. The ensuing

intracellular accumulation of glucose-6-phosphate and glucose inhibits further glucose uptake [167].

This interpretation was partially challenged when several studies revealed that increased FA levels in

the circulation were associated with a defect in glucose transport, provoked by an impaired insulin-

mediated translocation of glucose transporter 4 (GLUT4) rather than a defect in glycolysis [168, 169].

In addition, elevated levels of plasma FAs can result in an ectopic accumulation of lipids, like TAG,

DAG, and FAs. Furthermore, such lipid accumulations can lead to an altered insulin response of

insulin-sensitive tissues (e.g. muscle, liver).

Introduction

21

In muscle cells and adipocytes insulin activates the plasma membrane-bound insulin receptor, a

tyrosine kinase, which further phosphorylates the intracellular family of insulin receptor substrates

(IRS 1-4) on several tyrosine residues [170, 171]. Activated IRSs serve as docking station for proteins,

like phosphatidylinositide-3-kinase (PI3K) and downstream effectors, like protein kinase B (PKB/Akt)

[172, 173]. The subsequent generation of phosphatidylinositol-3,4,5-triphosphate (PIP3) by PI3K

facilitates the recruitment of PKB/Akt to the plasma membrane where it is phosphorylated by 3-

phosphoinositide-dependent protein kinase 1 (PDK1) [171]. This event enables further signal

transmission, which results in the release of GLUT4 to the plasma membrane [174]. Membrane-

associated GLUT4 facilitates glucose uptake. Additionally, activation of PKB/Akt in muscle and liver

cells leads to phosphorylation and inhibition of glycogen synthase kinase 3 [172, 175]. This inhibition

promotes glycogen synthesis and inhibits gluconeogenesis (Fig. 11). Defects within this signaling

cascade result in a loss of insulin-sensitivity and can lead to insulin resistance of affected

cells/tissues.

FIGURE 11. Intracellular pathway of insulin signaling. Insulin binds to insulin receptor that further activates IRS. This leads

to the activation of PI3K and furthermore to phosphorylation/activation of PKB/Akt by PDK1. Consequently GLUT4

translocate to the plasma membrane and enables glucose uptake in SM and adipocytes. Furthermore, activation of PKB/Akt

promotes glycogen synthesis in SM and liver. Additionally, insulin promotes lipogenesis and inhibits gluconeogenesis in

liver. GLUT4, glucose transporter 4; IRS, insulin receptor substrate; PDK1, 3-phosphoinositide dependent protein kinase 1;

PI3K, phosphoinositide-3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP3, phosphatidylinositol 3,4,5-triphosphate;

PKB/Akt, protein kinase B.

Introduction

22

DAG signal

The intracellular accumulation of DAG is assumed to be tightly connected to an altered insulin

response. More recently, ectopic DAG accumulation was found to be associated with disturbed

insulin signaling. These DAG effects are thought to derive from the action of the PKC-family members

which are known to play a crucial role in many signaling events, cellular differentiation, and cell

growth [176]. The PKC-family comprises three different subgroups, namely conventional (α, β1, β2

and γ; cPKC), novel (δ, ε, η and θ; nPKC), and atypical (ζ and λ/ι; aPKC) PKCs [176]. cPKS and nPKCs

display lipid-sensitive isoforms and are usually activated by PLC-dependent generation of sn-1,2 DAG,

either Ca2+-dependent (cPKC) or -independent (nPKC). In contrast, aPKC are mainly activated by

protein/protein interactions and are insensitivity towards DAG or Ca2+ [176]. These differences are

attributed to regulatory domains, designated as C1 and C2, which account for lipid binding and Ca2+-

sensing, respectively. The C1 domain of both conventional and novel PKCs binds DAG and phorbol

esters, whereas C1 domain of aPKCs binds PIP3 and ceramides. Only conventional isoforms contain a

functional C2 domain which binds anionic PLs in a Ca2+-dependent manner [176]. The activity of

cPKCs and nPKCs is highly influenced by intracellular levels of DAG. Noteworthy, earlier studies

showed that only the sn-1,2 DAG isoform has the ability to activate PKCs. The other isoforms, sn-1,3

and sn-2,3 are inactive [152-154].

Of all PKC subgroups mainly nPKC, more precisely PKCε and PKCθ, adversely affect insulin signaling

[177]. Earlier studies showed that PKCs are activated in diabetic rodent models [178, 179] and that

activation of PKCs by phorbol esters cause IR [180-182]. In rodents, an infusion of intralipid/heparin

leads to enhanced plasma FA concentrations causing impaired insulin signaling associated with

activation of PKCθ in SM [183]. The development of IR was thereby associated with an increase in

intramuscular DAG levels and independent of TAG or ceramide content [184]. In accordance, PKCθ-

deficient mice are protected from acute SM-IR after lipid infusion [185]. Additionally, PKCε is also

implicated in the development of hepatic IR. In rodents, 3 days of high-fat feeding cause hepatic

steatosis and hepatic IR without peripheral lipid accumulation or IR. The requirement of PKCε in this

process is evident by enhanced hepatic insulin response in PKCε-antisense oligonucleotide (ASO)

treated fat-fed rats [186]. Consistent with this, mice lacking PKCε exhibit slightly increased hepatic

lipid content but are resident against diet-induced IR following 1 week of high-fat feeding [187].

Up to now several mechanisms have been identified by which nPKCs impair insulin action. Recent

studies showed that PKCθ can phosphorylate IRS1 at Ser1101 which blocks insulin stimulated tyrosine

phosphorylation [188] and activation of PI3K [177]. In rat liver, PKCε and insulin receptor reside in

close proximity and inhibition of PKCε expression protects against HFD-induced reduction of insulin

receptor kinase activation [186] (Fig. 12).

Introduction

23

FIGURE 12. Inactivation of insulin signaling by DAG via novel PKC isoforms. sn-1,2 DAG activates novel PKC isoforms,

PKCε/θ, which leads to phosphorylation of IRS at serine residues. This event inhibits downstream effector signaling. GLUT4,

glucose transporter 4; IRS, insulin receptor substrate; PDK1, 3-phosphoinositide dependent protein kinase 1; PI3K,

phosphoinositide-3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; PKB/Akt, protein kinase B; PKC, protein kinase C.

Phenotypes of a variety of genetic murine models lead to the hypothesis that enhanced DAG levels in

insulin responsive tissue ultimately result in an impaired insulin signaling. Mice lacking mitochondrial

GPAT accumulate FA-CoA, but not TAG or DAG when set on a high-fat diet [189]. Besides this

elevation of FA-CoA, these mice are protected from diet-induced hepatic IR. In contrast,

overexpression of mitochondrial GPAT does not change FA-CoA levels but leads to hepatic IR, which

is associated with increased levels of lysophosphatidic acid (LPA), DAG, and TAG [190]. Similarly, in

obese Zucker rats (non-functional leptin receptor) IR is associated with increased amounts of hepatic

and muscle ceramide and DAG contents [191]. Another model arguing for DAG as mediator of IR are

mice fed high-ketogenic diet. These mice develop severe hepatic steatosis and profound hepatic IR

which is associated with a 350% increase in hepatic DAG content [192]. This increase is followed by

activation of PKCε and decreased insulin-mediated tyrosine-phosphorylation of IRS2. In mice

overexpressing DGAT2 in liver, hepatic TAG as well as DAG and ceramide levels are markedly

increased [193, 194]. These mice were first reported to exhibit normal hepatic insulin sensitivity

[193] but were recently identified to exhibit enhanced PKCε activation accompanied by severe

hepatic IR [194]. Furthermore, the key interaction between DAG, PKCε activation and hepatic IR is

confirmed in numerous other rodent models [195]. Interestingly, a recent study discovered that

hepatic DAG content of cytoplasmic LDs is the best predictor of IR in obese, non-diabetic individuals

[196]. In the same study they observed distinct localization of PKCε at cytoplasmic LDs as well as

Introduction

24

enhanced activation of this PKC isoform in these patients [196]. So far, this study is unique in

connecting LD turnover with PKC-signaling. However, the question concerning the intracellular

localization as well as the primary metabolic source of DAGs, which are involved in PKC-signaling is

still under dispute.

A number of murine model argue against a causative role of DAG accumulation in the development

of IR. For example, HSLko animals accumulate large amounts of DAG in adipose and non-adipose

tissues [56, 59] but do not develop pronounced alterations in insulin signaling [57, 197-199].

Although, depending on genetic background, some HSLko strains show signs of impaired insulin

signaling [199, 200], others do not or even show increased insulin sensitivity [197, 198]. ASO-

dependent decrease of CGI-58 expression, the co-activator of ATGL, leads to a markedly increase of

hepatic TAG, DAG, and ceramide content in fat-fed mice [201]. However, CGI-58 knockdown in mice

is associated with improved insulin sensitivity and glucose tolerance [201] suggesting additional

mechanism and factors to be involved in insulin signaling. So far, little evidence points towards a

direct association between ATGL-dependent lipolysis and altered insulin signaling. ATGLko mice

exhibit markedly increased ectopic lipid levels, mainly TAG, but display improved glucose tolerance

[75]. A recent study observed that overexpression of ATGL in cultured myotubes leads to an

increased content of both DAG and ceramides, which is associated with impaired insulin signaling

[202]. Nevertheless, mice overexpressing ATGL in adipose tissue appear to be protected from IR

[203]. Thus, to date it is unclear whether ATGL- derived DAGs activate PKC and may thereby interfere

with insulin signaling.

In summary, above mentioned studies suggest a link between DAG, PKC-signaling, and the

development of IR. In this context, DAG isomerism as well as intracellular DAG compartmentation

might play a crucial role. As outlined above, a number of metabolic reactions degrade or consume

DAG. Yet, they are located at different cellular compartments and the resulting DAG isoforms are

mostly unknown. Earlier studies clearly demonstrated that DAG-binding of the C1 domain of PKCs is

highly specific for sn-1,2 DAG. Other isoforms, like sn-2,3 or sn-1,3 DAG, completely fail to activate

PKCs [152-154] and hence display no signaling properties. Due to the inability of DAG isoforms, other

than sn-1,2, to activate PKCs it is crucial to determine the DAG isoforms which are generated and

further utilized during different metabolic reactions. The predominant reactions, accounting for DAG

generation and utilization in adipocytes, are catalyzed by ATGL/HSL as well as DGAT1/DGAT2,

respectively. Hence, the identification of the stereo/regioselectivity of these lipases and

acyltransferases would markedly augment the understanding of connections between DAG

metabolism and signaling and my provide explanations to previously unclear data regarding this

topic.

Introduction

25

FA and ceramide signal

Besides mentioned FA-induced alterations of glucose uptake, FAs can influence a variety of other

cellular processes. Since TAG displays the major storage form of FAs, TAG metabolism has a large

influence on the intracellular amount and composition of FAs and also FA-CoA. Both FAs and FA-CoAs

can interfere with cellular signaling pathways via activation of peroxisome proliferator-activated

receptors (PPARs). PPARs are nuclear hormone receptors involved in the regulation of genes

associated with inflammation, energy homeostasis as well as lipid, and lipoprotein metabolism. The

PPAR family is composed of PPARα, PPARδ, and three isoforms of PPARγ [204]. Oxidative tissues, like

CM, SM, or liver express mainly PPARα. PPARα activates the expression of genes involved in FA

transport and oxidation as well as keto- and gluconeogenesis [205]. PPARδ, which is responsible for

the activation of genes involved in glucose and FA utilization displays ubiquitous expression [206].

PPARγ is mainly active in cells associated with lipid storage and act as inducer of genes involved in

lipogenesis [207]. Activation of all PPARs requires co-activation by PPARγ coactivator-1α or β (PGC1α

or PGC1β), which follows binding of lipid ligands and dimerization with retinoid X receptor [208-210].

Both PGCs are involved in the regulation of uptake, transport, and oxidation of energy substrates

[205, 206]. Hence, dysregulation of these genes leads to severe metabolic disorders. Currently the

specificity of PPAR activation, in respect to endogenous ligands, is not entirely clear. However, the

established hypothesis lists FA and FA-CoAs either as direct (ligand) or indirect (precursor for other

lipid ligands) PPAR activator. In this regard unsaturated but not saturated FAs exhibit high signaling

potential for PPARs [211]. In particular mono-unsaturated FAs (MUFAs), like C16:1 and C18:1, are

potent activators for PPARα, much less for PPARβ or PPARγ [211-213]. PUFAs, like C18:2 and linolenic

acid (C18:3) exhibit activation potential for all PPAR species [211, 212].

Recently, ATGLko and HSLko mice were investigated regarding cellular signaling as a consequence of

TAG hydrolysis. These studies revealed different effects of either ATGL or HSL on the expression

pattern of genes involved in oxidative metabolism [214]. Deficiency of ATGL or HSL leads to a

decreased expression of oxidative genes in BAT [214]. However, only ATGL deficiency attenuates

oxidative gene expression in other tissues, like CM and SM suggesting a role of ATGL-dependent

lipolysis in the transcriptional control of FA metabolism [214]. Compatible with this theory,

overexpression of ATGL in either adipocytes or hepatocytes leads to increased gene expression of

PPARα and δ and enhanced FA oxidation as well as to enhanced PPARα activity and further target

gene expression, respectively [203, 215]. Conversely, knockdown of either ATGL or its co-activator

CGI-58 in hepatocytes leads to a suppression of PPARα target gene expression in vivo [201, 216].

Additionally, a recent study uncovered the role of ATGL in PPAR activation in cardiomyocytes [217].

Therein, it was shown that ATGL deficiency leads to a drastic decrease of PGC1α and β and to a

Introduction

26

dysfunction of mitochondrial substrate oxidation and respiration within cardiomyocytes. The

resulting excessive lipid accumulation leads to cardiac insufficiency and a lethal cardiomyopathy.

These results indicate that ATGL-dependent TAG hydrolysis generates essential mediators involved in

the generation of PPAR ligands. Moreover, ATGLko mice treated with PPARα agonists display

completely restored mitochondrial function and are spared from premature death [217].

Besides PPAR activation, which can be triggered by virtually all FAs, other bioactive lipid species

require specific FAs. In this context sphingolipids, more precisely ceramides, have been shown to be

negatively involved in the regulation of insulin signaling [218-221]. Since the sphingolipid synthesis

requires C16:0 [222, 223], the synthetic pathway of sphingolipids depends on the availability of this

FA species [224, 225]. Similar as for DAG, also ceramide accumulation was observed in SM of insulin-

resistant animals [191, 226], lipid-infused humans [227], and patients suffering from T2DM [220].

Furthermore, several studies reported that an accumulation of ceramides, induced by excessive

C16:0, is accountable for the initiation of IR in lean tissues [228, 229]. Ceramides efficiently block the

translocation of PKB/Akt to plasma membranes [221], which is a key step in insulin signaling and

promotes GLUT4 translocation to the plasma membrane resulting in glucose uptake [230]. The

detrimental effect of ceramides on PKB/Akt translocation was observed in several cell types,

including muscle cells [226, 231, 232], and adipocytes [232].

FIGURE 13. Inactivation of insulin signaling by ceramides via atypical PKC isoforms. Ceramides activate PKCζ which in turn

phosphorylates PKB/Akt. This phosphorylation inhibits translocation of PKB/Akt and consequently prevents translocation of

GLUT4 to the plasma membrane. GLUT4, glucose transporter 4; IRS, insulin receptor substrate; PDK1, 3-phosphoinositide-

dependent protein kinase 1; PI3K, phosphoinositide-3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP3,

phosphatidylinositol 3,4,5-triphosphate; PKB/Akt, protein kinase B; PKC, protein kinase C.

Introduction

27

In some cells ceramides block PKB/Akt translocation by direct activation of phosphatases which are

responsible for the dephosphorylation of PKB/Akt [175]. However, recent studies demonstrated that

the atypical, ceramide-binding PKCζ [233] is involved in phosphorylation of PKB/Akt. This

phosphorylation prevents translocation of PKB/Akt [226, 231] (Fig. 13). Additionally, a repression of

ceramide synthesis, by inhibition of serine palmitoyltransferase results in diminished lipotoxicity,

enhanced glucose regulation, and improved insulin response in mice [175, 234]. Together, these data

indicate that excessive amounts of C16:0, which result in an accumulation of ceramides represents at

least one possible reason for the onset of IR.

In summary, FAs affect directly or indirectly a variety of signaling events. Since some of the observed

effects strictly depend on single FA species, the composition of FA species within the cell plays a

crucial role. Since TAG hydrolysis is highly involved in the homeostasis of FAs and ATGL is rate-

limiting in this process, the identification of FA species, which are specifically released by ATGL during

lipolysis, is of great interest. A clear picture of the FA species composition, which derives from ATGL-

mediated lipolysis, would consequently enlarge the understanding of the interplay between energy

metabolism and FA-associated cell signaling.

MAG signal

MAG derives from either TAG- or PL-breakdown as direct product of DAG hydrolysis. Specifically 2-AG

was found to be the most abundant, endogenous ligand of cannabinoid receptors (CBRs; CBR1, CBR2)

[100, 101]. 2-AG derives mainly from the hydrolysis of arachidonic acid-containing PLs through the

combined actions of PLC and DAGL [100, 101]. In the nervous system, 2-AG is released from

postsynaptic neurons and causes retrograde inhibition of the presynaptic neurotransmitter release

[235]. Following binding and activation of presynaptic CBR1, 2-AG is internalized and degraded by

MGL. Endogenous compounds activating CBRs are commonly summarized as endocannabinoids

(ECs). Together, EC-metabolizing enzymes, CBRs, and ECs form the endocannabinoid system (ECS)

[236], which is active in neurons and non-neuronal cells, including hepatocytes, adipocytes, and cells

of the immune system. The ECS regulates numerous neuronal processes, including emotional

behavior, cognition, and pain. Furthermore, it is also involved in the regulation of food intake, energy

balance and lipid turnover [235].

Physiological implications of the ECS have been demonstrated in mice and humans. Obese patients

exhibit decreased food intake, reduced lipogenesis, and increased energy expenditure after

treatement with the CBR1-antagonist rimonabant [237, 238]. The same effects are observable in

mice lacking CBR1. Conversely, treatment with CBR-agonists results in a hyperactive ECS, which

Introduction

28

causes a central orexogenic effect and a concomitant reduction of energy expenditure [239]. This

leads to an increased lipid deposition in peripheral tissues, including WAT and the liver. A hyperactive

ECS may be one reason of increased appetite (food intake) and enhanced lipid deposition in obese

patients [240-242].

Due to the effects of the ECS on the energy metabolism, it has been linked to the pathogenesis of

metabolic deseases, like obesity and T2DM. Whether ATGL/HSL-dependent LD lipolysis contributes to

the generation of arachidonic acid-containing MAG signals or HSL is involved in the degradation of 2-

AG at cellular membranes is not known.

Aim of the Thesis

Aim of the Thesis

30

ATGL is the rate-limiting enzyme in TAG degradation. Its activity has been shown to be crucial for

lipid and energy homeostasis but also for intracellular lipid signaling, in particular PPARα activation.

The first aim of this study was to elucidate the enzymatic characteristics of ATGL-dependent TAG

hydrolysis. The characterization includes ATGL´s substrate, FA-, and stereo/regioselectivity and

should clarify the preference of ATGL for FA species, the position of hydrolyzed FA ester within TAG

and, thus, the stereo/regiospecific conformation of produced DAG. Since ATGL belongs to a large

protein family, which in part also comprises of phospholipases, it appears feasible that ATGL might

also hydrolyze any of the FA ester bonds of LD-forming PLs.

The second aim of this study was to investigate possible stereo/regiochemical requirements of DAG-

utilizing enzymes, more precisely DGAT1, DGAT2, and HSL. Selectivities of these enzymes should

delineate which cellular reactions can directly utilize lipolysis-derived DAGs, the direct product of

ATGL-dependent TAG hydrolysis.

Results

Results modified after Eichmann et al. [243] are indicated (§).To describe the investigation of the stereo/regioselectivity of

ATGL consistently, results obtained during forgone diploma thesis [244] were recapitulated (indicated by §§

).

Results

32

I) ATGL selectivity

This section focuses on the substrate-, FA- and stereo/regioselectivity of ATGL-dependent TAG

hydrolysis.

A) Stereo/regioselectivity

TAG consists of three FAs which are esterified to the three carbon atoms of the glycerol backbone.

According to stereochemical rules the carbon atoms are stereospecifically numbered (sn) 1-3. The

selectivity of a lipase is denoted depending on the position of hydrolysis.

ATGL regioselectively hydrolyzes sn-2 FA esters of TAG generating sn-1,3 DAG

in vitro

To investigate if ATGL exhibits selectivity for the FA position on the glycerol backbone, TAG hydrolysis

experiments with subsequent analysis of generated DAG isoforms were performed. Therefore, Cos7-

cells expressing ATGL (Fig. 14A) were homogenized and cytosolic fractions were incubated with 14C-

glycerol-labeled triolein in the presence of HSL-specific inhibitor (76-0079) with or without purified

GST-tagged CGI-58. Since the regioisomeric forms of DAG, sn-1,2/2,3 and sn-1,3 DAG, can be

separated by thin layer chromatography (TLC), the generation of this species was investigated first.

Incubation of ATGL led to an almost exclusive generation of sn-1,3 DAG, over a time period of 60 min.

In contrast, ATGL co-activated by CGI-58 led, besides an increase in activity, to the generation of both

regioisomeric forms of DAG (Fig. 14B). Experiments were repeated with slight modifications;

prolonging incubation period as well as changing the radiolabeled TAG tracer to 3H-FA-labeled

triolein. In addition, experiments were performed for different time periods and DAG formation was

measured. Data obtained reconfirm that ATGL alone generates exclusively sn-1,3 DAG, whereas ATGL

co-activated by CGI-58 results in an overall increase in generated DAGs and in the generation of both

DAG regioisomers (Fig. 14C, D). Results indicate that ATGL cleaves TAG selectively at sn-2 position

and extends selectivity upon CGI-58 co-activation to sn-1/3 position.

Results

33

FIGURE 14. ATGL cleaves TAG at sn-2 position and expands selectivity to sn-1/3 upon co-activation by CGI-58. A,

Expression of his-tagged ATGL in Cos7-cells was assessed by immunoblotting. Cytosolic fractions of Cos7-cells expressing

ATGL were incubated with a HSL-specific inhibitor (76-0079) in the presence (B, D) or absence (B, C) of purified GST-tagged

CGI-58 with either 14

C-glycerol (B) or 3H-FA labeled (C, D) triolein emulsified with PC for either 40 (C, D), 60 (B) or 120 min

(C, D) at 37°C. Lipids were extracted, separated by TLC and radioactivity in DAG bands was determined by scintillation

counting. Data are normalized to LacZ and presented as means +/- S.D. and are representative for 2 independent

experiments. Statistical significance was determined applying Student´s unpaired t-test (***,p<0,001; **, p<0,01). n.s.…no

significance. (Fig. 14C,D)§,§§

.

ATGL co-activated by CGI-58 generates sn-1,3 and sn-2,3 DAG

To investigate if ATGL looses sn-2 selectivity upon CGI-58 co-activation, DAG species were analyzed

using chiral-phase HPLC. This method allows the discrimination between all DAG species namely sn-

1,3 DAG, sn-1,2 DAG and sn-2,3 DAG. Therefore, TAG-hydrolysis experiments were performed using

cytosolic fractions of Cos7-cells expressing ATGL in combination with purified GST-tagged CGI-58 in

the presence of HSL-specific inhibitor (76-0079). Samples were incubated with non-labeled triolein

for 60 min. Lipids were extracted and separated using TLC. Following isolation, DAGs were

0

100

200

300

400

500

600

ATGL ATGL/CGI-58

DA

G (

cpm

/µg

pro

tein

*h)

sn-1,3

sn-1,2/2,3

**

0

20

40

60

80

100

120

140

160

180

40 120

DA

G (

cpm

/µg

pro

tein

*h)

time [min]

ATGLsn-1,3

sn-1,2/2,3

n.s.

0

500

1000

1500

2000

2500

40 120

DA

G (

cpm

/µg

pro

tein

*h)

time [min]

ATGL/CGI-58sn-1,3

sn-1,2/2,3***

***

50 -

37 -

ATGL kDa

B

C D

A

Results

34

derivatized to their corresponding 3,5-dinitrophenylurethanes and separated by chiral-phase HPLC.

First, as “proof of principal”, a racemic diolein reference compound, which comprises all three DAG

species, was analyzed. The analysis proved the suitability of this method by separating all three

species into distinct peaks over a time range of 16 min (Fig. 15A). Next, as a positive control, “egg-

yolk”-lecithin was digested using phospholipase C (b. cereus). Breakdown of lecithin resulted in

exclusively sn-1,2 DAG (Fig. 15B). Subpeaks within the sn-1,2 DAG peak reflect the FA composition of

this biological sample and correspond to sn-1,2 DAGs comprising either C16:0, C18:1 or C18:2 in

different combinations. When triolein was hydrolyzed by ATGL co-activated by CGI-58 exclusively sn-

1,3 and sn-2,3 DAGs were generated (Fig. 15C). These data indicate a sn-2 selectivity of ATGL which is

enlarged to sn-1 position by CGI-58.

FIGURE 15. ATGL co-activated by CGI-58 generates sn-1,3 and sn-2,3 DAG. A-C, Chiral-phase HPLC resolution of different

DAG species. DAGs were analyzed as corresponding 3,5-dinitrophenylurethanes by chiral-phase HPLC. Analysis of a racemic

dioleoylglycerol reference mix (A), the reaction products of „egg yolk lecithin” hydrolyzed by purified PLC (b. cereus, the

three peaks at the retention time range 11.5-13min display the different FA composition of the resulting DAGs; I: 16:0-

18:1+18:1-18:1; II 16:0-18:2+18:1-18:2; III:18:2-18:2; B), and the reaction products of triolein hydrolyzed by ATGL contained

in cytosolic fraction of Cos7-cells, in the presence of purified GST-tagged CGI-58 and HSL-specific inhibitor (76-0079, C). Data

are representative for 2 independent experiments. x…unknown compound. (Fig. 15A,B,C)§,§§

.

A B

C

Results

35

Conversion of DAG substrates is not detectable

The next question was, whether sn-1,2 DAG is preferentially consumed in follow-up reactions

occuring in cellular cytosolic fractions, or if spontaneous occurring non-enzymatic transesterification

(racemization) could be a reason for the lack of sn-1,2 DAG. Hence a racemic mixture of diolein was

incubated with cytosolic fractions of Cos7-cells in the presence of HSL-specific inhibitor (76-0079) for

120 min. DAGs were isolated from extracted lipids and analyzed before and after incubation using

chiral-phase HPLC. No alterations in DAG species composition were observed over an incubation

period of 120 min (Fig. 16A). Furthermore, diolein isoforms were determined by TLC (Fig. 16B). Also

with this method, no evidence of transesterification was observed. This proves the stability of DAG

isoforms in this experimental setup.

FIGURE 16. DAG substrates show no transesterification or decomposition. A, A mixture of sn-1,3, sn-1,2 and sn-2,3 DAG or

either sn-1,3 (B) or rac-1,2/2,3 DAG (B) emulsified with PC was incubated with cytosolic fraction (A) or lysates (B) of Cos-7

cells in the presence of a HSL-specific inhibitor (76-0079) for 120 min at 37°C. Lipids were extracted before and after

incubation according to Folch et al, separated by TLC and DAGs were analyzed as corresponding 3,5-dinitrophenylurethanes

by chiral-phase HPLC (A) or visualized by iodine staining (B). Data are presented as means +/- S.D. and are representative for

2 independent experiments. (Fig. 16B)§.

HSLko mice show drastic accumulation of sn-1,3 DAG in WAT

To investigate, if the selectivity of ATGL is also present in an in vivo model, acylglycerol composition

of WAT from HSLko mice was examined. The deficiency of HSL is known to lead to a drastic

accumulation of DAGs, due to the fact that HSL represents the predominant intracellular DAG-lipase

in lipolysis. Lipid analysis revealed that total acylglycerol levels (including TAG, DAG, MAG) of HSLko

mice were unaltered compared to wt littermates (Fig. 17A). Further analysis revealed the expected

0

10

20

30

40

50

60

70

sn-1,3 sn-1,2 sn-2,3

DA

G is

om

ers

(%

of

tota

l)

timepoint: 0min

timepoint: 120min

A B

Results

36

substantial accumulation of DAG (~6-fold), which was accompanied by a moderate decrease in TAG

level (~10 %) (Fig. 17B).

FIGURE 17. Accumulation of sn-1,3 DAG in WAT of HSL-deficient mice. A, Lipids were extracted according to Folch et al.

and total acylglycerol levels in WAT of wt and HSLko mice were determined using Infinity-TAG kit. C, Neutral lipids of wt and

HSL-deficient WAT were separated by TLC using chloroform/acetone/acetic acid (90/8/1; v/v/v). B, D, Bands corresponding

to TAG and DAG were scraped off, extracted and acylglycerol content was determined using Infinity-TAG kit (B and D) and

DAG isomers were analyzed by chiral-phase HPLC (C, right half). Data are presented as means +/- S.D. Statistical significance

was determined applying Student´s unpaired t-test (***, p<0,001), n=4 (each genotype). (Fig. 17A,B,C,D)§.

Next, TLC analyses were performed and showed that sn-1,3 DAG was not detectable in lipid extracts

of wt mice. In contrast, sn-1,3 DAG constitutes a major DAG species in acylglycerols of HSLko mice

(Fig. 17C). The detailed species composition of isolated DAG from wt and HSLko WAT was analyzed

using chiral-phase HPLC. In wt WAT, acyglycerol content consisted of 1.7% of sn-1,2 DAG (80% of

total DAG). 0.06% of sn-1,3 (2% of total DAG) and 0.32% of sn-2,3 DAG (18% of total DAG) constitute

only a minor portion of total acylglycerol in wt mice (Fig. 17D left). Composition analysis confirmed

the drastic accumulation of sn-1,3 DAG in WAT of HSLko mice, which accounted for 7.7% (65% of

0

10

20

30

40

50

60

70

80

90

100

wildtype HSLko

acyl

glyc

ero

l (%

of

tota

l )

triacylglycerol

diacylglycerol97.9%

88.2%

2.1%

11.8%

0

50

100

150

200

250

wildtype HSLko

tota

l acy

lgly

cero

l (µ

mo

l/g

WA

T)

A B

C D

0

1

2

3

4

5

6

7

8

9

wildtype HSLko

DA

G (

% o

f to

tal a

cylg

lyce

rol)

sn-1,3 DAG

sn-1,2 DAG

sn-2,3 DAG

***

Results

37

total DAG) of total acylglycerols. Furthermore, it showed that sn-2,3 DAG was more abundant than

sn-1,2 DAG and accounted for 2.8% (24% of total DAG) of HSLko acylglycerols. The 1,3% of sn-1,2

DAG (11% of total DAG) were comparable to the sn-1,2 DAG content of wt mice (Fig. 17D right).

Together, data of this section clearly demonstrate that ATGL hydrolyzes TAG selectively at sn-2

position, yielding sn-1,3 DAG. Furthermore, co-activation of ATGL by CGI-58 leads to the concomitant

generation of sn-2,3 DAG but not sn-1,2 DAG (Fig. 18). The accumulation of sn-1,3 and sn-2,3 DAG in

HSLko mice corroborates the stereo/regioselectivity of ATGL in vivo.

FIGURE 18. Schematic depiction of the stereo/regioselectivity of ATGL. ATGL hydrolyzes TAG at sn-2 position, or at sn-1 or

sn-2 position when co-activated by CGI-58 yielding sn-1,3 DAG or sn-1,3 and sn-2,3 DAG, respectively. ATGL, adipose

triglyceride hydrolase; CGI-58, comparative gene identification-58; DAG, diacylglycerol; FA, fatty acid; TAG, triacylglycerol.

B) FA selectivity

TAGs occurring in biological systems are usually not uniformly esterifies with only one kind of FA but

instead contain different FAs with varying chain-length and saturation. Naturally abundant FAs

contain 8 to 22 carbon atoms and up to 6 double bonds. Annotation of FAs includes number of

carbon atoms followed by number of double bonds (saturation level) e.g. palmitic acid (C16:0, 16

carbon atoms:0 double bonds). Besides stereo/regioselectivity, lipases usually display substrate

selectivity against chain-length as well as saturation level of bound FA species.

This section focuses on the FA-selectivity of ATGL in vitro and in vivo.

Results

38

ATGL differentiates between chain-length of TAG-bound FAs in vitro

The next aim was to investigate, if ATGL exhibits selectivity for distinct FA species during TAG

hydrolysis. Therefore, murine ATGL and CGI-58 were expressed in Cos7-cells (Fig. 19A) and in vitro

TAG hydrolase assays were performed using various TAG species as substrate. In the first set of

experiments, the used TAG substrates were uniformly esterified with naturally abundant FA species,

namely palmitoleic acid (C16:1), C18:1, C18:2 and C18:3. The high melting points (>40°C) of TAGs

consisting of three saturated FAs (C12:0, C14:0, C16:0, C18:0) precluded the preparation of suitable

lipid emulsions. Furthermore, also liquid soluble TAG emulsions (unsaturated FAs) differ in their

solubility. To prevent any effects, caused by solubility, all substrates were diluted to the same final

concentration (0.25 mM). ATGL displayed highest activity against tripalmitolein (3xC16:1), followed

by triolein (3xC18:1), trilinolenin (3xC18:3) and trilinolein (3xC18:2) (Fig. 15B). ATGL co-activated by

CGI-58 showed an expected increase in hydrolase-activity. As a consequence differences in activities

against used substrates became more obvious (Fig. 19B). Results indicate that ATGL is capable of

hydrolyzing virtually all used TAG substrates with a marked preference for C16:1 esters.

FIGURE 19. ATGL hydrolyzes TAG with FAs of different chainlength in vitro. A, Recombinant proteins were expressed in

Cos7-cells and expression of his-tagged proteins was assessed by immunoblotting. B, Homogenates of Cos7-cells expressing

ATGL were incubated in the absence or presence of CGI-58 with different, homogeneously esterified TAG species emulsified

with PC as substrates for 1 h at 37°C. Released FAs were measured using NEFA-C kit. Data are normalized to LacZ and

presented as means +/- S.D. and are representative for 2 independent experiments. Statistical significance was determined

applying Student´s unpaired t-test (***, p<0,001; **, p<0,01). (Fig. 19A,B)§.

0

100

200

300

400

500

600

700

800

FA (

nm

ol/

mg

pro

tein

*h)

ATGL/LacZ

ATGL/CGI-58

***

***

*** ***

*****

150 -

100 -

75 -

50 -

LacZ

37 -

ATGL CGI-58 kDa kDa

A B

Results

39

ATGL cleaves saturated FAs of mixed-labeled TAG species in vitro

Next, the ability of ATGL to hydrolyze saturated FA esters within mixed TAG species was investigated.

First, ATGL-dependent hydrolysis of TAG containing C16:0 at the sn-1/3 or sn-2 and C18:1 at the

remaining positions was investigated. These TAGs reflect a substrate class mixed in chain-length as

well as saturation-level. Virtually all of these TAG species were hydrolyzed at the same rates by either

ATGL alone or ATGL co-activated by CGI-58 (Fig. 20A). A minor, yet significant decrease in the

hydrolytic activity of ATGL was found when C16:0 was esterified at sn-2 position, indicating that a

saturated FA at the preferred position of ATGL-dependent hydrolysis results in a slightly decreased

activity. Next, substrates composed of FAs exhibiting same chain-length (18 carbon atoms) but

different saturation-level were tested. ATGL alone as well as ATGL co-activated by CGI-58 showed

equal hydrolase activities against the different TAG substrates (Fig. 20B). The small decrease in

activity using TAG containing C18:2 at sn-2 position most likely reflects a weak preferential selectivity

of ATGL for C18:1 over C18:2 (Fig. 20B). All in all, these data suggest that the saturation-level of TAG-

bound FAs is not a crucial factor of ATGL-dependent TAG hydrolysis.

FIGURE 20. ATGL hydrolyzes TAGs with FAs differing in saturation-level in vitro. A, B, Homogenates of Cos7-cells

expressing ATGL were incubated in the absence or presence of CGI-58 with different, heterogeneously esterified TAG

species as substrate for 1 h at 37°C. Released FAs were measured using NEFA-C kit. Data are normalized to LacZ and

presented as means +/- S.D. and are representative for 2 independent experiments. Statistical significance was determined

applying Student´s unpaired t-test (***, p<0,001; **, p<0,01). (Fig. 20A)§.

0

50

100

150

200

250

300

350

400

FA (

nm

ol/

mg

pro

tein

*h)

ATGL/LacZ

ATGL/CGI-58***

18:1/18:1/18:1 18:1/18:1/16:016:0/18:1/18:1

18:1/16:0/18:1

triglyceride

0

50

100

150

200

250

300

350

400

FA (

nm

ol/

mg

pro

tein

*h)

ATGL/LacZ

ATGL/CGI-58**

18:1/18:1/18:1

triglyceride

18:1/18:0/18:1 18:1/18:2/18:1

A B

Results

40

FA-analysis by gaschromatography with flame-ionization detection was used

to assess the FA-selectivity of ATGL in vivo

To examine FA selectivity of ATGL in vivo, ATGLko mice and their wt littermates were used to

determine FA composition of WAT-TAG as well as FA composition of plasma lipids. This data should

indicate whether ATGL differentially hydrolyzes FA species in vivo. For this purpose, derivatization of

FAs to their corresponding methylesters (FAMEs) and analysis by gaschromatography with flame-

ionization detection (GC-FID) was performed. GC-FID exhibits the advantage that the

concentration/signal ratio is constant over a broad analyte concentration range and is equal for all

detectable FA species. To test the linearity of the method, different concentrations of FA species

were analyzed. Measurements of different concentrations of e. g. C18:1 and C18:3 showed accurate

linearity of the received signals, which made this method suitable for FA determination (Fig. 21A). A

second advantage of GC-FID is the excellent separation over a broad range of FA species in a short

time frame. The naturally most abundant FA species were clearly peak-separated within a time range

of around 17 min (Fig. 21B).

FIGURE 21. GC-FID analysis yields linear signal/concentration ratios as well as clear separation of different FAs analyzed

as FAMEs. A, Different concentrations of either oleic (upper panel) or linoleic acid (lower panel) were analyzed using GC-

FID. B, Chromatogram of a mixture of most abundant saturated and unsaturated FAs as FAMEs.

y = 26,49x + 18,111R² = 0,9979

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

pe

ak a

rea

(AU

/10

0)

oleic acid (C18:1)

y = 25,458x + 7,4352R² = 0,9996

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

0 50 100 150 200

pe

ak a

rea

(AU

/10

0)

pmol/µl

linolenic acid (C18:3)

A B

Results

41

Accumulating TAGs in ATGLko WAT contain all FAs with highest increase in

C16:1

To elucidate the FA-selectivity of ATGL in vivo, FA composition of WAT-TAG from wt and ATGLko mice

was determined. As previously described [75], ATGLko mice exhibited drastically increased TAG

content in WAT (~4-fold) (Fig. 22A). FA composition analysis of WAT-TAG expressed as “% of total”

showed an altered FA-pattern of ATGLko animals compared to wt, with a relative increase in C16:0

and C16:1 and a relative decrease in FAs containing 18 carbon atoms (Fig. 22B).

FIGURE 22. WAT-TAGs of ATGLko mice accumulate virtually all FA species. A, WAT lipids of non-fasted wt and ATGLko

mice were extracted according to Folch et al. and TAG content was measured using Infinity-TAG kit. B, C, Extracted WAT

lipids were separated by TLC and TAGs were isolated and transesterified. FAMEs were separated and analyzed using

GC/FID. FA composition demonstrated as percentage of total (B) or as amounts per tissue weight (C). D, Calculated changes

in FA species of WAT-TAGs in ATGLko versus wt mice. Data are presented as means +/- S.D. Statistical significance was

determined applying Student´s unpaired t-test (***, p<0,001; **, p<0,01; *, p<0,05). n=4 (each genotype). (Fig. 22C,D)§.

The molar depiction showed that the absolute amounts of all FA species massively accumulate in

ATGLko WAT-TAG (Fig. 22C). Calculation of the ratio of FAs comparing ATGLko and wt showed that all

0

3

6

9

12

15

WA

T TA

G-F

A (

fold

ch

ange

, A

TGLk

o t

o w

ildty

pe

)

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:1meanchange

***

***

*** ***

***

0

100

200

300

400

500

600

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:1

WA

T TA

G-F

A (

µm

ol/

g ti

ssu

e)

wildtype

ATGLko

***

**

***

***

***

***

*** ***

0

5

10

15

20

25

30

35

40

45

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:1

WA

T TA

G-F

A (

% o

f to

tal)

wildtype

ATGLko

**

***

***

**

**

**

**

B

C

D

A

0

50

100

150

200

250

300

350

400

450

500

WA

T-TA

G (

µm

ol/

g ti

ssu

e)

wildtype

ATGLko

*

Results

42

FA species are increased around 6 times, except C16:1 which accumulated up to 12-fold in ATGLko

WAT-TAG (Fig. 22D). This data fit to results from in vitro TAG hydrolase assays where ATGL shows

highest activity against C16:1. Hence, 16:1 accumulated most prominently in WAT-TAGs of ATGLko

mice.

Feeding/fasting-dependent changes in plasma-FA composition of wt mice

inversely correlate with FA-accumulation in WAT-TAGs of ATGLko mice

Next, plasma-FA composition of wt mice after 8 h fasting and in non-fasted state was determined.

Since, the metabolic switch, following starvation leads to an elevated, ATGL-mediated release of FAs

from WAT, determination of released FA species should reflect the preference of ATGL to hydrolyze

various FA species of WAT-TAGs.

FIGURE 23. Wt mice show a distinct pattern of FAs released into plasma. Plasma lipids of non-fasted and 8 h fasted wt

mice were extracted according to Folch et al. and separated by TLC. FAs were isolated and transesterified. FAMEs were

analyzed using GC/FID. A, Concentration of plasma FAs was calculated and expressed as sum of all measured FAs. B, Plasma

FA composition demonstrated as amounts per plasma volume. C, Calculated changes in plasma FA species of non-fasted

and 8 h fasted wt mice. Data are presented as means +/- S.D. Statistical significance was determined applying Student´s

unpaired t-test (***, p<0,001; **, p<0,01; *, p<0,05). n=4; nd…not detectable. (Fig. 23B,C)§.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

pla

sma-

FA (

mm

ol/

l)

non-fasted

fasted

***

***

0

0,05

0,1

0,15

0,2

0,25

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:4 C22:6

pla

sma-

FA (

mm

ol/

l)

non-fasted

fasted

***

***

***

***

***

* nd nd

0

1

2

3

4

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:4 C22:6

pla

sma-

FA (

fold

ch

ange

, fas

ted

/no

n-f

aste

d)

meanchange

***

***

***

*****

nd nd

A B

C

Results

43

In agreement with published data [75], FA concentration in plasma of wt mice was doubled after 8 h

of food deprivation (Fig. 23A). Furthermore, composition analysis showed that the majority of

abundant FA species were increased around 2-fold (Fig. 23B). Calculation of the release changes for

each FA species upon fasting revealed that C16:1 as well as C18:1 and C18:2 displayed significantly

higher ratios, comparing fasted and non-fasted FA levels in plasma (Fig. 23C).

Fasting-induced FA release of WAT is blunted in ATGLko mice

To test if FA release observed in wt mice is caused by ATGL-dependent WAT-TAG hydrolysis the same

experiment was performed with ATGLko mice. FA release of WAT was completely diminished in

these mice. No increase in concentration of plasma FAs were observed upon 8 h fasting (Fig. 24A).

According to that, no changes in the pattern of plasma FAs were detected (Fig. 24B, C). These results

demonstrate that ATGL is crucial for WAT lipolysis and the release of FAs into plasma.

FIGURE 24. ATGLko mice display blunted release of FA into plasma. Plasma lipids of non-fasted and 8 h fasted ATGLko

mice were extracted according to Folch et al. and separated by TLC. FAs were isolated and transesterified. FAMEs were

separated and analyzed using GC/FID. A, Concentration of plasma FAs was calculated as sum of all measured FA

concentrations. B, Plasma FA composition demonstrated as amounts per plasma volume. C, Calculated changes in plasma

FA species of non-fasted and 8 h fasted ATGLko mice. Data are presented as means +/- S.D. n=4.

0

0,05

0,1

0,15

0,2

0,25

0,3

pla

sma-

FA (

mm

ol/

l)

non-fasted

fasted

***

0

0,02

0,04

0,06

0,08

0,1

0,12

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2

pla

sma-

FA (

mm

ol/

l)

non-fasted

fasted

0

1

2

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2

pla

sma-

FA (

fold

ch

ange

, fas

ted

/no

n-f

aste

d)

meanchange

A B

C

Results

44

To compare the fasting response of ATGLko and wt mice, both data sets were combined. Consistent

with the blunted FA release in ATGLko mice, the plasma FA levels were decreased around 3-fold as

compared to wt mice (Fig. 25A). Changes in FA composition of ATGLko plasma mirrored the ATGL-

dependent fasting-response of wt mice. Nearly all FA species were decreased in plasma of fasted

ATGLko mice and long chain FAs were undetectable (Fig. 25B). Calculation of the release ratio

between wt and ATGLko mice showed that especially C16:1, C18:1 and C18:2 were drastically

diminished in plasma of ATGLko mice (Fig. 25C). Taken together, ATGL-dependent hydrolase activity

in WAT leads to a release of predominantly unsaturated FA species, in particular C16:1, C18:1 and

C18:2 into circulation. Furthermore, these data provide clear evidence that ATGL acts as an essential

lipase during fasting-stimulated WAT lipolysis and, hence, determines FA release.

FIGURE 25. ATGL in WAT is essential for the fasting-induced FA release into plasma. Plasma lipids of 8 h fasted wt and

ATGLko mice were extracted according to Folch et al. and separated by TLC. FAs were isolated and transesterified. FAMEs

were analyzed using GC/FID. A, Concentration of plasma FAs was calculated as sum of all measured FAs. B, Plasma FA

composition demonstrated as amounts per plasma volume. C, Calculated changes in plasma FA species of fasted wt mice

compared to ATGLko littermates. Data are presented as means +/- S.D. Statistical significance was determined applying

Student´s unpaired t-test (***, p<0,001; *, p<0,05). n=4 (each genotype); nd…not detectable. (Fig. 25B,C)§.

0

0,05

0,1

0,15

0,2

0,25

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:4 C22:6

pla

sma-

FA, f

aste

d (

mm

ol/

l)

wildtype

ATGLko

***

***

***

*** ***

*

nd nd nd

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

pla

sma-

FA, f

aste

d (

mm

ol/

l)

wildtype

ATGLko

***

***

0

1

2

3

4

5

6

7

8

9

C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:4 C22:6

pla

sma-

FA, f

aste

d (

fold

ch

ange

, wild

typ

e/A

TG

Lko

)

meanchange

***

***

***

***

nd nd nd

A B

C

Results

45

ATGL affects FA composition of plasma TAG

To assess possible effects of ATGL-dependent WAT-FA release on the FA composition of VLDL-TAG,

the FA composition of plasma TAG of 8 h fasted ATGLko and wt mice was determined using GC-FID.

In wt, plasma TAG levels were 4-times higher compared to ATGLko (Fig. 26A). Analysis of FA

composition revealed that most of the FA species were in about 3-fold higher (Fig. 26B, C) in plasma

TAG of wt mice as compared to ATGLko mice. Furthermore, plasma TAG-FA composition of wt mice

showed highest specific increase in C16:1 (6-fold) followed by C18:1 (5-fold). This indicates that the

loss of ATGL leads to most prominent decrease in unsaturated FAs, which presumably is a result of

decreased hydrolytic activity in WAT of these mice.

FIGURE 26. FA composition of plasma TAG is changed in ATGLko mice. Plasma lipids of 8 h fasted wt and ATGLko mice

were extracted according to Folch et al. and separated by TLC. TAGs were isolated and transesterified. FAMEs were

analyzed using GC/FID. A, Concentration of plasma TAGs were measured using Infinity-TAG kit. B, Plasma TAG-FA

composition shown as concentration per plasma volume. C, Numeric changes in plasma TAG-FA species of fasted wt mice

compared to ATGLko littermates. Data are presented as means +/- S.D. Statistical significance was determined applying

Student´s unpaired t-test (***, p<0,001). n=4 (each genotype).

0

1

2

3

4

5

6

7

pla

sma

TAG

-FA

, fas

ted

(m

mo

l/l)

wildtype

ATGLko

***

***

0

1

2

3

4

5

6

7

8

C 14:0 C 16:0 C 16:1 C 18:0 C 18:1 C 18:2 C 18:3 C 20:4 C 20:6

pla

sma

TAG

-FA

, fa

ste

d (

fold

ch

ange

, wild

typ

e/A

TGLk

o)

meanchange

***

***

0

1

2

3

4

5

6

C 14:0 C 16:0 C 16:1 C 18:0 C 18:1 C 18:2 C 18:3 C 20:4 C 20:6

pla

sma

TAG

-FA

, fas

ted

(m

mo

l/l)

wildtype

ATGLko

***

***

***

***

A B

C

Results

46

ATGL affects TAG content in the brain

To additionally investigate the effect of ATGL on brain lipid metabolism, neutral and polar lipid

fraction of whole brains, from 8 h fasted ATGLko and wt mice, were analyzed.

FIGURE 27. TAGs accumulate most abundant FA species in brain of ATGLko mice. Brain lipids of wt and ATGLko mice were

extracted according to Folch et al. and separated by TLC. Total lipid extract or TAGs isolated after TLC were transesterified.

FAMEs were analyzed using GC/FID. FA-composition of PLs (A), CEs (B), DAGs (C) and total brain lipids (D) shown as

percentage of total. E, Brain TAG content was calculated as sum of all measured FA species. F, Brain TAG-FA composition

shown as amount per tissue weight. Data are presented as means +/- S.D. Statistical significance was determined applying

Student´s unpaired t-test (***, p<0,001; *, p<0,05). n=5 (each genotype). (Fig. 27E,F modified after [245])

100

200

300

400

500

600

700

800

(nm

ol/

g ti

ssu

e)

wildtype

ATGLko***

***

***

*

0

20

40

60

80

100

C 14:0 C 16:0 C 16:1 C 18:0 C 18:1 C 18:2 C 18:3 C 20:1 C 20:4 C 22:6

bra

in T

AG

-FA

, fas

ted

*

0

5

10

15

20

25

30

35

C 14:0 C 16:0 C 16:1 C 18:0 C 18:1 C 18:2 C 18:3 C 20:1 C 20:4 C 22:4 C 22:6

bra

in li

pid

-FA

, fas

ted

(%

of

tota

l)

wildtype

ATGLko

0

100

200

300

400

500

600

700

800

bra

in T

AG

, fas

ted

(n

mo

l/lg

tis

sue

)

wildtype

ATGLko***

0

10

20

30

40

50

60

C 14:0 C 16:0 C 18:0 C 18:1 C 18:2 C 20:4

bra

in D

AG

-FA

(%

of

tota

l)

wildtype

ATGLko

0

10

20

30

40

50

60

70

C 16:0 C 18:0 C 18:1

bra

in C

E-FA

(%

of

tota

l)

wildtype

ATGLko

0

5

10

15

20

25

30

C 14:0 C 16:0 C 16:1 C 18:0 C 18:1 C 18:2 C 18:3 C 20:1 C 20:4 C 22:4 C 22:6

bra

in P

L-FA

(%

of

tota

l)

wildtype

ATGLko

A

C

E F

D

B

Results

47

Total brain lipids or single lipid species were isolated by TLC and FA compositions were determined as

FAMEs by GC-FID. FA composition of PLs, CEs, and DAGs were unaltered in ATGLko mice as compared

to wt mice (Fig. 27A, B, C). Similarly the FA composition of total brain lipids from ATGLko and wt mice

was identical (Fig. 27D). In contrast, large differences were found in total TAG content and TAG-FA

composition. TAG content in ATGLko brain exceeded that of wt about 14-fold (Fig. 27E). FA

composition analysis revealed that all TAG-FAs of brain were increased between 5 and 30-fold in

ATGLko mice as compared to wt mice. Besides that, many FA species, including C16:1 as well as long-

chained docosahexanoic acid (C22:6), which were not detectable in wt mice, were highly increased in

ATGLko mice (Fig. 27F). Although the role of ATGL and TAGs in brain is not known, the changes in

TAGs clearly speak for an active, ATGL-dependent TAG turnover in brain (results published in [245]).

In summary, in vitro and in vivo results of this section show that ATGL is able to hydrolyze all major

saturated und unsaturated FAs occuring in TAG. ATGL exhibits moderate preference for unsaturated

FA in particular C16:1. Even if C16:1 is not very abundant, increases in C16:1 levels were ubiquitous

when neutral lipids of ATGLko mice were investigated.

C) Substrate selectivity

Amphiphatic PLs form the lipid droplet surface monolayer, thereby enclosing lipophilic TAGs and

emulsifying them in the aqueous environment. Since ATGL belongs to a protein family which also

comprises phospholipases [65] it was tested whether ATGL also hydrolyzes LD-associated PLs.

ATGL hydrolyzes PLs independent of CGI-58

First, the hydrolytic activity of ATGL against PC, which is the major PL species on LD surface, was

investigated. Therefore, murine ATGL was expressed in Cos7-cells and hydrolase activity assays using

PC as substrate were performed. Purified PLA2 (naja mossambica m.) served as positive control. PC-

hydrolase assays were performed in the presence and absence of 1 mM Ca2+, since many, including

snake venom PLA2, hydrolyze PLs in a Ca2+-dependent manner. Little activity was observed for snake

venom PLA2 in the absence of Ca2+. Addition of Ca2+ led to a 32-fold increase in phospholipase

activity (Fig. 28A). To investigate the phospholipase activity of ATGL, PC-hydrolase assays in the

presence or absence of Ca2+ and purified GST-tagged CGI-58 were performed. ATGL exhibited

Results

48

hydrolytic activity against PC, which was independent of Ca2+ (Fig. 28B). Furthermore, the PC

hydrolase activity of ATGL was not co-activated by CGI-58 (Fig. 28B) which is in contrast to effects

observed for ATGL-dependent TAG hydrolase activity.

FIGURE 28. ATGL hydrolyzes PC independent of calcium and CGI-58. Purified phospholipase A2 (1 IU, naja mossambica m.)

(A) or homogenates of Cos7-cells expressing ATGL (B) were incubated in the absence or presence of purified GST-tagged

CGI-58 (B) or 1 mM Ca2+

(A,B) with PC as substrate for 1 h at 37°C. Released FAs were measured using NEFA-C kit. Data are

normalized to LacZ and presented as means +/- S.D. and are representative for 2 independent experiments.

Phospholipase activity of ATGL is co-activated by CGI-58 when PLs are mixed

with TAG

To investigate whether a LD-like lipid mixture is a better substrate for ATGL-dependent

phospholipase activity, experiments were performed using different neutral lipids emulsified with PC

as substrate. First, 14C-labeled PC micelles were used as a substrate and incubated with homogenates

of Cos7-cells expressing ATGL. As expected, ATGL hydrolyzed PC and was not co-activated by CGI-58

(Fig. 29A). Next, non-radiolabeled CE was emulsified with non-radiolabeled PC. This substrate should

reflect liposomes, which contain a hydrophobic core that cannot be hydrolyzed by ATGL. In this

setup, ATGL showed hydrolase activity against PC independent of CGI-58 (Fig. 29B). So far, the

discrepancy in absolute values of hydrolase activity using either radiolabeled or non-radiolabeled PC

substrates is unclear. Next, experiments were performed using 3H-labeled triolein emulsified with

14C-labeled PC. Both lipids are potential substrates of ATGL, and so, generated liposomes more or less

reflect the composition of a cytosolic LD. Unexpectedly, FAs released of PC showed that ATGL

exhibited the same activity against PC but increased activity 2-fold when co-activated by CGI-58 (Fig.

29C).

0

2

4

6

8

10

12

14

16

18

20

-Ca2+ +1mM Ca2+

FA (

nm

ol/

mg

pro

tein

*h)

ATGL ATGL/CGI-58

0

2

4

6

8

10

12

14

16

18

-Ca2+ +1mM Ca2+

FA (

mm

ol/

mg

pro

tein

*h)

phospholipase A2 (najamossambica m.)

A B

Results

49

FIGURE 29. The co-activation of the phospholipase activity of ATGL by CGI-58 depends on the neutral lipid species

emulsified within PC liposomes. Homogenates of Cos7-cells expressing ATGL were incubated in the absence or presence of

purified GST-tagged CGI-58 with 14

C-labeled PC (A) or CE emulsified with non-labeled PC (B) or 3H-labelled TAG emulsified

with 14

C-labeled PC (C,D) or non-labeled TAG emulsified with 14

C-labeled PC (E) as substrate for 1 h at 37°C. Released FAs

were determined by either NEFA-C kit (B) or scintillation counting (A, C, D). Data are normalized to LacZ and presented as

means +/- S.D. and are representative for 2 independent experiments. Statistical significance was determined applying

Student´s unpaired t-test (***, p<0,001; **, p<0,01).

0

1

2

3

4

5

6

7

ATGL ATGL/CGI-58

FA (

nm

ol/

mg

pro

tein

*h)

**

14C-phosphatidylcholine3H-triacylglycerol

14C-FA release

0

50

100

150

200

250

ATGL ATGL/CGI-58

FA (

nm

ol/

mg

pro

tein

*h)

***

14C-phosphatidylcholine3H-triacylglycerol

3H-FA release

0

1

2

3

4

5

6

ATGL ATGL/CGI-58

FA (

nm

ol/

mg

pro

tein

*h)

14C-phosphatidylcholine micelle

0

2

4

6

8

10

12

14

16

18

20

ATGL ATGL/CGI

FA (

nm

ol/

mg

pro

tein

*h)

phosphatidylcholinecholesterylester

0

2

4

6

8

10

12

ATGL ATGL/CGI-58

FA (

nm

ol/

mg

pro

tein

*h)

***

14C-phosphatidylcholinetriacylglycerol

14C-FA release

A B

C D

E

Results

50

Determination of 3H-labeled FA release revealed that the TAG hydrolase activity of ATGL was 9-fold

co-activated upon addition of CGI-58 (Fig. 29D). To exclude an error of measurement caused by an

interference of simultaneously used 3H- and 14C-labeling of FAs, experiments were performed using

non-labeled triolein emulsified with 14C-labeled PC. As observed in experiments using labeled TAG

and PC, ATGL hydrolyzed PC and increased activity 3-fold upon CGI-58 co-activation (Fig. 29E). These

data indicate that ATGL hydrolyzes PC and that the PC hydrolase activity of ATGL can be co-activated

by CGI-58 when PC is mixed with TAG. Furthermore, TAG hydrolase activity of ATGL and ATGL co-

activated by CGI-58 is 10-fold and 45-fold higher, respectively, as compared to PC hydrolase activity.

ATGL hydrolyzes most abundant PL species

To investigate if ATGL is able to hydrolyze PL species other than PC, PL hydrolase assays were

performed using homogenates of Cos7-cells expressing ATGL and PC, phosphatidylserine (PS), PE, PA

and phosphatidylglycerol (PG) as substrates. ATGL hydrolyzed all investigated PLs in micellar form

and showed comparable activities against PC, PS and PE as well as increased activity against PA and

PG (Fig. 30). These data suggest that ATGL exhibits phospholipase activity against a broad spectrum

of PLs, thereby slightly preferring PA and PG.

FIGURE 30. ATGL hydrolyzes highly abundant PLs, favoring PA and PG. Homogenates of Cos7-cells expressing ATGL were

incubated with different PL species as substrate for 1 h at 37°C. Released FAs were determined by NEFA-C kit. Data are

normalized to LacZ and presented as means +/- S.D. and are representative for 2 independent experiments. Statistical

significance was determined applying Student´s unpaired t-test (***, p<0,001).

0

5

10

15

20

25

30

PC PA PS PE PG

FA (

nm

ol/

mg

pro

tein

*h)

***

***

Results

51

Thus, ATGL can act as phospholipase, with preference for PG and PA as substrates. Thereby, the

position selectivity of ATGL as well as the impact of the phospholipase activity of ATGL on cellular

lipid homeostasis needs to be investigated.

II) Selectivity of DAG hydrolysis and re-esterification

This section investigates the isomer-selectivity of possible downstream reactions of DAGs generated

by ATGL during lipolysis. On the one hand, the re-esterification of DAG by either DGAT1 or DGAT2,

and on the other hand, the degradation of DAG by HSL (Fig. 31).

FIGURE 31. Potential hydrolysis and re-esterification reactions of DAG generated on cytoplasmic LDs. DAG can act as

precursor for TAG-synthesis catalyzed by either DGAT1 or DGAT2. Furthermore, DAG can be a substrate for degradation by

HSL yielding MAG.

A) Stereo/regioselectivity of DGAT enzymes

The regeneration of TAG by esterification of DAG with FA-CoA is one possible reaction of lipolysis-

derived DAGs. This reaction is catalyzed by either DGAT1 or DGAT2. If one of these enzymes exhibit

certain selectivity for DAG isomers is unknown and topic of this section.

Results

52

DGAT1 and DGAT2 differentiate between DAG isoforms and are highly active

on nano-sized liposomes

To investigate stereo/regioselectivity of DGAT enzymes in vitro, murine DGAT1 and DGAT2 were

expressed in Cos7-cells (Fig. 33A). Acyltransferase assays were performed using different DAG

isoforms as well as different DAG/PC compositions. The published method of DGAT activity assays

describes a DAG/PC ratio of ~1/4 [246] as substrate, which results in liposomes with a distinct size.

Given that cellular LDs differ in size, it was of interest if liposomal size affects DGAT activity.

Therefore, a constant substrate concentration of DAG (200 µM) was used while PC concentration

was decreased from 800 to 200 and 50 µM, giving DAG/PC ratios of 0.25, 1 and 4, respectively. To

examine the size of generated liposomes the substrate solutions were incubated with lipophilic

BodiPy® and liposomes were visualized using fluorescence microscopy. The pictures showed that an

increase in DAG/PC ratio, led to an increase in liposomal size. DAG/PC ratios of 0.25, 1 and 4 resulted

in a mean liposome size of 50-15 nm, 500-1000 nm and 2-3 µm, respectively (Fig. 32A, B, C).

FIGURE 32. Micelles size of DAG/PC substrate increases with increasing DAG/PC ratio. 200 µM of DAG substrate was

emulsified with either 800 µM (DAG/PC – 0.25, A), 200 µM (DAG/PC – 1, B) or 50 µM (DAG/PC – 4, C) of PC. Micelle size was

visualized by staining with BodiPy® and imaged by fluorescence microscopy.

Furthermore, either sn-1,2, rac-1,2/2,3 or sn-1,3 DAG was emulsified with PC in the mentioned ratios

and used as substrate. Both DGAT enzymes showed a reduction in activity for all provided DAG

isoforms when DAG/PC ratio was increased from 0.25 to 4 (Fig. 33B). Unexpectedly, DGAT1 and

DGAT2 exhibited different selectivity for DAG isoforms. Using a DAG/PC ratio of 0.25, at which both

DGATs were most active, DGAT1 showed highest activity against sn-1,2 and rac-1,2/2,3 DAG whereas

DGAT2 showed highest activity in esterifying sn-1,3 DAG (Fig. 33C). For further experiments a

DAG/PC ratio of 0.25 was used. Since expression levels of both DGATs differ, the selectivity quotient

for both enzymes was calculated. Results showed a 80/20 ratio of sn-1,2/sn-1,3 DAG for DGAT1 and,

in contrast, a 30/70 ratio of sn-1,2/sn-1,3 DAG for DGAT2 (Fig. 33D). Thus, DGAT1 as well as DGAT2

A B C

Results

53

prefer nano-liposomes as substrate and DGAT1 prefers sn-1,2 and sn-2,3 DAG as acceptor, while

DGAT2 favours sn-1,3 DAG.

FIGURE 33. DGAT1 and DGAT2 display different activities against DAG isoforms and micelle size. A, Expression of his-

tagged LacZ and flag-tagged murine DGAT1 and DGAT2 in Cos7-cells was assessed by immunoblotting. B, Homogenates of

Cos7-cells expressing DGAT1 or DGAT2 were incubated with either sn-1,2, rac-1,2/2,3 or sn-1,3 DAG emulsified with PC in

different molar ratios and 14

C-labeled C18:1-CoA as substrate for 10 min at 37°C. Reaction was stopped by addition of

CHCl3/MeOH (2/1, v/v). Lipids were extracted, separated by TLC and radioactivity in TAG bands was determined by

scintillation counting. C, Comparison of DGAT1 and DGAT2 activities against different DAG isoforms emulsified with PC in a

molar ratio of 800/200 (PC/DAG, µM/µM). D, Regioselectivity of recombinant DGAT enzymes against sn-1,2 or sn-1,3

diolein, expressed as percentage of total acyltransferase activity. Data are normalized to LacZ and presented as means +/-

S.D. and are representative for 2 independent experiments. Statistical significance was determined applying Student´s

unpaired t-test (***,p<0,001). (Fig. 33A,C,D)§

Selective inhibiton of DGAT enzymes is required to assign endogenous DGAT

activity to either DGAT1 or DGAT2

To additionaly investigate the stereo/regioselectivity of endogenous DGAT1 and DGAT2 in WAT, a

specific inhibition of either DGAT1 or DGAT2 is required. Therefore, a variety of DGAT inhibitors was

0

10

20

30

40

50

60

70

80

90

DGAT 1 DGAT2

rati

o (

% o

f to

tal a

cylt

ran

sfe

rase

act

ivit

y)

sn-1,2 diolein

sn-1,3 diolein

***

***

0

20

40

60

80

100

120

140

sn-1,2 rac-1,2/2,3 sn-1,3

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

DGAT1

DGAT2

***

***

diolein

0

20

40

60

80

100

120

140

0.25 1 4 0.25 1 4

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

sn-1,2 diolein

rac-1,2/2,3 diolein

sn-1,3 diolein

ratio diolein:phospatidylcholine

DGAT2DGAT1

D

A

C

B

Results

54

tested. DGAT activity assays were performed in vitro using reported inhibitors and a 50/50 mixture of

sn-1,2/sn-1,3 DAGs as substrate. Niacin, proposed as DGAT2-selective inhibitor [247] showed a 90%

inhibition of DGAT2 whereas DGAT1 activity was unaffected (Fig. 34A).

FIGURE 34. The effects of DGAT-inhibitors on the enzymatic activity of DGAT1 and DGAT2 in cell homogenates. A, B, C,

Homogenates of Cos7-cells expressing DGAT1 or DGAT2 were incubated in the absence or presence of different inhibitory

compounds with either sn-1,2 or sn-1,3 DAG or a 1/1 mixture of sn-1,2/sn-1,3 DAG substrate emulsified with PC in a molar

ratio of 800/200 (PC/DAG, µM/µM) and 14

C-labeled C18:1-CoA for 10 min at 37°C. Reaction was stopped by addition of

CHCl3/MeOH (2/1, v/v). Lipids were extracted, separated by TLC and radioactivity in TAG bands was determined by

scintillation counting. Data are presented as means +/- S.D. and are representative for 2 independent experiments.

Statistical significance was determined applying Student´s unpaired t-test (***,p<0,001).

0

20

40

60

80

100

120

140

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

--

Niacin (5mM)DGAT1 inh. (5µM)

LacZ DGAT1 DGAT2

***

***

--

+-

-+

++

--

+-

-+

++

0

20

40

60

80

100

120

140

160

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

--DGAT1 inh. (5µM)

LacZ DGAT1 DGAT2

***

***

***

MgCl2 (100mM) --

+-

-+

++

--

+-

-+

++

0

5

10

15

20

25

30

35

LacZ DGAT1 DGAT2

14C

-TA

G fo

rme

d (

nm

ol/

mg

pro

tein

*h) 10mM MgCl2

100mM MgCl2

sn-1,3 diolein

***

***

0

20

40

60

80

100

120

140

DGAT2DGAT1LacZ

14C

-TA

G fo

rme

d (

nm

ol/

mg

pro

tein

*h)

sn-1,2 diolein

***

C

A

B

Results

55

Conversely, with the same efficiency, DGAT1-specific inhibitor (2-((1s,4s)-4-(4-(4-amino-7,7-dimethyl-

7H-pyrimido[4,5-b][1,4]oxazin-6-yl)phenyl)cyclohexyl) acetic acid, [139]) inhibited DGAT1 but not

DGAT2 (Fig. 34A). Next, the effect of MgCl2 on DGAT activity was investigated. Low concentrations

are required for DGAT activity while high concentrations (>100 mM) should selectively inhibit DGAT2

[118]. As expected, a concentration of 100 mM MgCl2 led to a complete inhibition of DGAT2 activity.

Unexpectedly, the same concentration also significantly inhibited 60% of DGAT1-dependent activity.

The residual activity of DGAT1 was eliminated by using DGAT1-specific inhibitor (Fig. 34B). Since

published data only use sn-1,2 DAG as substrate, it was conceivable that enzyme inhibition depends

on substrate composition. To test this, either sn-1,3 or sn-1,2 DAG was used as substrate in

acyltransferase experiments. With sn-1,3 DAG as substrate, a higher activity of DGAT2 was observed

as compared to DGAT1 (Fig. 34C). Furthermore, 100 mM MgCl2 led to a complete inhibition of DGAT2

and a ~70% inhibition of DGAT1. In contrast, using sn-1,2 DAG as substrate, DGAT1 showed a 4-fold

higher activity compared to DGAT2. Using sn-1,2 DAG as substrate, 100mM MgCl2 had no effect on

DGAT1 activity while again completely inhibiting DGAT2. This led to the conclusion that due to the

substrate-dependent inhibition of DGAT1, MgCl2 is inappropriate to discriminate activities of

endogenous DGAT enzymes.

TAG hydrolase activity in WAT homogenates affects in vitro acyltransferase

measurements

To investigate the stereo/regioselectivity of endogenous DGAT enzymes in WAT of wt mice, the

previously tested inhibitors were applied to acyltransferase experiments using WAT homogenates.

Acyltransferase assays were performed in the presence of niacin as well as the DGAT1-specific

inhibitor [118]. Notably, results differed between niacin treatment of WAT and Cos7-cell

homogenates. In WAT homogenates DGAT1-specific inhibitor led to a 90% reduction of TAG

formation. In contrast, niacin enhanced DGAT activity, resulting in a 40% increase of TAG (Fig. 35A).

This may be due to an effect of niacin on enzymes involved in lipolysis, which may degrade TAGs

formed during acyltransferase assays. If niacin modulates lipase activity of ATGL or HSL in

acyltransferase experiments is unknown. To investigate this hypothesis, hydrolase inhibitors like HSL-

specific inhibitor (76-0079), bromoenol lactone (BEL) or orlistat were used. To test whether these

hydrolase inhibitors additionaly interfere with DGAT enzymes, homogenates of Cos7-cells expressing

DGAT1 and DGAT2 as well as described inhibitors were used in combination. Neither the two

unspecific hydrolase inhibitors BEL and orlistat, nor the HSL-specific inhibitor resulted in significant

differences of DGAT activity in homogenates of Cos7-cells expressing DGAT1 or DGAT2 (Fig. 35B).

Results

56

Next, hydrolase inhibitors as well as the DGAT1-specific inhibitor were used alone and in combination

in acyltransferase experiments using WAT homogenates.

FIGURE 35. The effects of different lipase and/or DGAT-inhibitors on the enzymatic activity of DGAT1 and DGAT2 in WAT

and cell homogenates. Homogenates of WAT from wt mice (A, C) or homogenates of Cos7-cells expressing DGAT1 or

DGAT2 (B) were incubated in the absence or presence of different inhibitory compounds with a 1/1 mixture of sn-1,2/sn-1,3

DAG substrate emulsified with PC in a molar ratio of 800/200 (PC/DAG, µM/µM) and 14

C-labeled C18:1-CoA for 10 min at

37°C. Reaction was stopped by addition of CHCl3/MeOH (2/1, v/v). Lipids were extracted according to Folch et al, separated

by TLC and radioactivity in TAG bands was determined by scintillation counting. Data are presented as means +/- S.D. and

are representative for 2 independent experiments. Statistical significance was determined applying Student´s unpaired t-

test (***,p<0,001). n=3 (pooled).

0

10

20

30

40

50

60

70

80

90

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

BEL (5µM)

Orlistat (20µM)

76-0079 (12,5µM)DGAT1 inh. (5µM)

-

WAT homogenate

---

+---

-+--

--+-

-++-

+-+-

+-++

-+++

***

***

0

10

20

30

40

50

60

70

80

90

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

DGAT1 inh. (5µM)

Niacin (5mM) --

***

***

WAT homogenate

+-

-+

0

20

40

60

80

100

120

140

160

180

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

)

Orlistat (20µM)

76-0079 (12,5µM)

LacZ DGAT1 DGAT2

BEL (5µM) ---

---

+--

-+-

--+

---

+--

-+-

--+

A

C

B

Results

57

Unexpectedly, inhibition of lipolytic enzymes in WAT clearly affected acyltransferase experiments.

BEL, orlistat, and HSL-specific 76-0079 resulted in a 1.7-fold increase of TAG formation. In the

presences of lipase inhibitors, DGAT1-specific inhibition led to a 90% reduction of TAG formation (Fig.

35C). These results lead to the conclusion that inhibition of TAG hydrolases is crucial for an accurate

determination of DGAT-dependent acyltransferase activity of WAT samples.

Endogenous DGAT1 and DGAT2 exhibit preference for sn-1,2 and sn-1,3 DAG,

respectively

In the next set of experiments the stereo/regioselectivity of endogenously expressed DGAT enzymes

of WAT was investigated. WAT homogenates of non-fasted C57BL/6J mice were separated into

microsomal (including plasma membranes), cytplasmic and LD fraction using ultracentrifugation.

Microsomal fractions showed highest DGAT1 and DGAT2 expression (Fig. 36A). In contrast, DGAT1

and DGAT2 were undetectable in the cytosolic fraction and in the LD fraction. Each cellular fraction

was subjected to acyltransferase assays using combinations of orlistat/76-0079 and DGAT1-specific

inhibitor and either sn-1,2 or sn-1,3 DAG as substrate.

FIGURE 36. DGAT1 and DGAT2 of WAT display different specific activities against DAG isoforms. A, Expression of

endogenous DGAT1 and DGAT2 in microsomal fraction of WAT was assessed by immunoblotting. B, Microsomal fraction of

WAT from wt mice were incubated in the absence or presence of different inhibitory compounds with either sn-1,2 or sn-

1,3 DAG substrate emulsified with PC in a molar ratio of 800/200 (PC/DAG, µM/µM) and 14

C-labeled C18:1-CoA for 10 min

at 37°C. Reaction was stopped by addition of CHCl3/MeOH (2/1, v/v). Lipids were extracted according to Folch et al,

separated by TLC and radioactivity in TAG bands was determined by scintillation counting. C, Calculated regioselectivity of

DGAT enzymes in microsomal fraction of WAT against sn-1,2 or sn-1,3 DAG expressed as percentage of total acyltransferase

activity. Data are presented as means +/- S.D. and are representative for 2 independent experiments. Statistical significance

was determined applying Student´s unpaired t-test (***,p<0,001). n=3. (Fig. 36A,B,C)§

0

10

20

30

40

50

60

70

DGAT1 DGAT2

rati

o (

% o

f to

tal a

cylt

ran

sfe

rase

act

ivit

y)

sn-1,2 diolein

sn-1,3 diolein

***

***microsomal fraction

0

20

40

60

80

100

120

140

160

180

14C

-TA

G f

orm

ed

(n

mo

l/m

g p

rote

in*h

) ***

***

76-0079 (12,5µM)DGAT1 inh. (5µM)

Orlistat (5mM)++

+sn-1,2 diolein sn-1,3 diolein

***

***

microsomal fraction

+-

+--

-++

++-

+--

-

C B A

Results

58

Microsomal fraction exhibited highest endogenous DGAT activity against both DAG isoforms.

Addition of lipase inhibitors resulted in a 1.6-fold increased formation of TAG when sn-1,2 DAG was

used as substrate. Upon inhibition of DGAT1, specific endogenous acyltransferase activity for sn-1,2

DAG was decreased by around 70% (Fig. 36B). DGAT activity of microsomal fraction was 20% lower

using sn-1,3 DAG as substrate and 1.6-fold increased upon lipase inhibiton. Addition of the DGAT1-

specific inhibitor led to a 50% reduction (Fig. 36B). Assuming that residual DGAT activity can be

attributed to DGAT2 and taken into account that expression levels of DGAT enzymes were different,

the selectivity quotient of microsomal fraction was calculated. It showed a 65/35 ratio for sn-1,2/sn-

1,3 DAG for DGAT1 and a 40/60 ratio for sn-1,2/sn-1,3 DAG for DGAT2 (Fig. 36C).

FIGURE 37. DGAT1 and DGAT2 of WAT display preference for different DAG isoforms. A, Cytoplasmic and C, LD fraction of

WAT from C57BL/6J mice were incubated in the absence or presence of different inhibitory compounds with either sn-1,2

or sn-1,3 DAG substrate emulsified with PC in a molar ratio of 800/200 (PC/DAG, µM/µM) and 14

C-labeled C18:1-CoA for 10

min at 37°C. Reaction was stopped by addition of CHCl3/MeOH (2/1, v/v). Lipids were extracted according to Folch et al,

separated by TLC and radioactivity in TAG bands was determined by scintillation counting. Calculated regioselectivity of

DGAT enzymes in B, cytoplasmic, and D, LD fraction of WAT against sn-1,2 or sn-1,3 DAG expressed as percentage of total

acyltransferase activity. Data are presented as means +/- S.D. and are representative for 2 independent experiments.

Statistical significance was determined applying Student´s unpaired t-test (***,p<0,001). n=3.

0

10

20

30

40

50

60

70

80

90

100

DGAT1 DGAT2

rati

o (

% o

f to

tal a

cylt

ran

sfe

rase

act

ivit

y)

sn-1,2 diolein

sn-1,3 diolein

***

***

cytoplasmic fraction

0

1

2

3

4

5

6

7

14C

-TA

G f

orm

ed

(n

mo

l/h

*mg)

***

***

-76-0079 (12,5µM)

DGAT1 inh. (5µM) -

Orlistat (5mM) -+-

+++

+--

-+-

+++

+

sn-1,2 diolein sn-1,3 diolein

cytoplasmic fraction

0

5

10

15

20

251

4C

-TA

G f

orm

ed

(n

mo

l/h

*mg)

***

***

-76-0079 (12,5µM)

DGAT1 inh. (5µM) -

Orlistat (5mM) -+-

+++

+--

-+-

+++

+

sn-1,2 diolein sn-1,3 diolein

******

lipid droplet fraction

0

10

20

30

40

50

60

70

80

DGAT1 DGAT2

rati

o (

% o

f to

tal a

cylt

ran

sfe

rase

act

ivit

y)

sn-1,2 diolein

sn-1,3 diolein

***

***lipid droplet fraction

A

B

C

D

Results

59

Next, DGAT activity of the cytoplasmic fraction and the LD fraction was assessed. Despite much lower

DGAT activities, comparable effects were observed using sn-1,2 DAG as substrate in combination

with lipase and DGAT1-specific inhibitors. Addition of lipase inhibitors resulted in a 3-fold increase of

TAG formation. Combined inhibition of lipases and DGAT1 decreased DGAT activity by 90%. In

contrast, no inhibitory effects were detectable using sn-1,3 DAG as substrate (Fig. 37A). Calculation

of the selectivity quotient for sn-1,2/sn-1,3 DAG led to a 90/10 ratio for DGAT1 and to a 20/80 ratio

for DGAT2 (Fig. 37B). Addition of lipase inhibitors to DGAT assays using LD fraction had no effects,

whereas addition of DGAT1-specific inhibitor led to >90% reduction of TAG formation using either sn-

1,2 or sn-1,3 DAG as substrate (Fig. 37C). The calculated selectivity quotients of DGAT activity within

LD fraction were similar to that obtained for microsomal and cytosolic fraction. DGAT1 exerts 70%

preference for sn-1,2 DAG whereas DGAT2 prefers sn-1,3 DAG (70%, Fig. 37D).

Taken together, results of this section demonstrate that DGAT1 as well as DGAT2 exhibit

regioselectivity for DAG species. DGAT1 prefers sn-1,2 DAG as substrate, which is mainly generated

during de novo lipid synthesis, located at the ER. In contrast, DGAT2 predominantly esterifies sn-1,3

DAG, which is generated by ATGL-dependent hydrolysis at the LD (Fig. 38).

FIGURE 38. Cellular acyltransferase-reactions utilizing DAG. sn-1,3 DAG, which is generated by ATGL in the course of

lipolysis displays the preferred substrate for DGAT2-dependent re-esterification. In contrast, DGAT1 predominantly

esterifies sn-1,2 DAG, which is mainly produced during de novo lipid synthesis at the ER.

Results

60

B) Stereo/regioselectivity of HSL-dependent DAG hydrolysis

In the process of lipolysis the breakdown of DAG on cytoplasmic LDs is catalyzed by HSL.

This section investigates the selectivity of HSL to degrade specific DAG isoforms. Since HSLko mice

accumulate large amounts of DAG, which can influence insulin signaling, insulin tolerance was

additionally assessed.

HSL preferentially hydrolyzes sn-1,3 DAG

First, the stereo/regioselectivity of the second consecutive reaction in the breakdown of TAG, namely

the degradation of DAG by HSL, was investigated. Hydrolase experiments were performed using

Cos7-cell homogenates containing murine HSL (Fig. 39A) and different DAG isoforms as well as TAG

as substrates. As previously described [23], HSL showed around 10-fold higher activity against

racemic DAG as compared to TAG. HSL exhibited highest hydrolase activity against sn-1,3 DAG when

sn-1,2, rac-1,2/2,3, and sn-1,3 DAG were separately used as substrate. DAG hydrolase activity of HSL

with sn-1,3 DAG as substrate was 1.5-fold increased as compared to hydrolase activity using either

sn-1,2 or rac-1,2/2,3 DAG. (Fig. 39B). This result indicates that sn-1,3 DAG displays a preferred

substrate for HSL.

FIGURE 39. HSL preferentially hydrolyzes sn-1,3 DAG. A, Expression of his-tagged LacZ and HSL in Cos7-cells was assessed

by immunoblotting. B, Homogenates of Cos7-cells expressing HSL were incubated with either TAG or isomeric different DAG

species as substrate, emulsified with PC for 1 h at 37°C. Generated FAs were measured using NEFA-C kit. Data are

normalized to LacZ and presented as means +/- S.D. and are representative for 2 independent experiments. Statistical

significance was determined applying Student´s unpaired t-test (***,p<0,001). (Fig. 39A,B)§

0

50

100

150

200

250

300

350

triolein rac sn-1,2 rac-1,2/2,3 sn-1,3

FA (

nm

ol/

mg

pro

tein

*h)

***

diolein

***

B A

Results

61

HSLko mice show signs of increased insulin sensitivity fed a normal chow diet

To address the question if DAG accumulation in HSLko mice causes IR, studies were performed using

male HSLko mice and wt littermates fed a normal chow diet (CD). NMR analysis revealed that body

composition of CD fed HSLko mice did not differ from the body composition of wt mice. Both,

showed comparable body weights (HSLko: 22.8±1.9 g; wt: 24.5±0.7 g) as well as 75% lean mass

versus 20% of fat mass (Fig. 40A). To test the effect of insulin, an intraperitoneal insulin tolerance

test (IPITT) was performed. Mice were fasted for 2 h following an intraperitoneal bolus of bovine

insulin (0.6 IU/kg body weight). Subsequently, blood glucose levels were measured at different time

points. Results of this IPITT showed that HSLko mice displayed increased insulin response as

compared to wt mice. This is evident by a significant decrease of blood glucose after 50 min of

injection as compared to wt mice (Fig. 40B). Calculating the area under the curve for both mice

showed that HSLko mice displayed significantly increased insulin tolerance (Fig. 35C). Accordingly,

the accumulation of DAG in tissues of HSLko mice does not promote the development of IR, but

instead increases insulin sensitivity.

FIGURE 40. HSLko mice show increased insulin sensitivity fed a normal CD. A, Body composition of male HSLko and

wildtype mice fed a normal CD was determined using NMR. B, Mice were fasted from 8:00 to 10:00 am and received an

intraperitoneal bolus of bovine insulin (0.6 IU/kg body weight). Blood glucose was determined at indicated time points

using Accu-Check glucometer. C, Areas under the curves were calculated using trapezoid method. Data are presented as

means +/- S.D. Statistical significance was determined applying Student´s unpaired t-test (*,p<0,05); n=4 (each genotype).

0

10

20

30

40

50

60

70

80

90

wildtype HSLko

we

igh

t (%

of

tota

l)

lean mass

fat mass

A B

0

20

40

60

80

100

120

140

160

180

200

0 50 100 150 200

glu

cose

(m

g/d

l)

time (min)

wildtype

HSLko

0

5

10

15

20

25

30

wildtype HSLko

are

a u

nd

er

the

cu

rve

(A

U/1

00

0)

*

C

Results

62

HSLko mice show no signs of impaired insulin signaling fed a HFD

To further investigate insulin signaling, male HSLko and wt mice were fed a HFD. Subsequently, body

composition was determined using NMR. Wt mice showed an extensive, 2.5-fold increase of fat mass

whereas HSLko mice showed a moderate 1,4-fold increase in fat mass as compared to that of mice

fed normal CD (Fig. 41A). To study insulin signaling, mice were fasted for 8 h following a refeeding

period of 1 h. After that, mice were sacrificed and liver as well as SM (m. gastrocnemius) samples

were collected. Tissues were homogenized and separated into cytoplasmic and microsomal fraction

(including plasma membranes) using ultracentrifugation. Samples were then subjected to

immunoblot analysis of proteins potentially involved in insulin signaling. GAPDH, which displays a

cytoplasmic protein, was not detectable in microsomal fractions but showed comparable expression

in cytoplasmic samples, which confirmed purity of fractionation.

FIGURE 41. HSLko mice show no signs of impaired insulin signaling fed a HFD. A, Body composition male HSLko and

wildtype mice fed a HFD was determined using NMR. B, Mice were fasted from 12:00 to 8:00 am and reefed for 1 h.

Subsequently mice were sacrificed and liver and gastrocnemius muscle were collected. Tissue homogenates were separated

into cytoplasmic and microsomal fraction (including plasma membranes) by ultracentrifugation. 40 µg of cellular fractions

were used to determine expression of indicated proteins by immunoblotting. Data are presented as means +/- S.D.; n=4

(each genotype).

0

10

20

30

40

50

60

70

80

wildtype HSLko

we

igh

t (%

of

tota

l)

lean mass

fat mass

B A

Results

63

Furthermore, HSL expression was missing in tissues of HSLko mice, proving correct genotypes (Fig.

41B). Tissue specific PKC isoforms, namely PKCθ in skeletal muscle and PKCε in liver showed

increased expression in cytoplasmic fraction of wt compared to HSLko mice and were much less

expressed in microsomal fractions. Both PKC isoforms are known to be positively correlated with

defective insulin signaling. PKCα which is induced by insulin showed slightly increased expression in

microsomal fractions of HSLko tissues. PKCα showed higher expression in microsomal fraction

(translocation) of SM as compared to the expression in cytoplasmic fraction. This translocation was

not observable in liver, were expression in cytoplasmic fraction was higher as compared to

microsomal fraction (Fig. 41B).

Together, the results of IPITT and analysis of insulin signaling in wt and HSLko mice showed that

HSLko mice are more insulin responsive as compared to wt mice and do not show defects in insulin

signaling even on a HFD.

Discussion

Discussion

65

During the last decades the incidence of obesity and concomitant T2DM and IR in the western world

increased alarmingly. Elevated levels of circulating free FAs are known to be involved in the

development of IR in peripheral tissues [167]. These circulating free FAs are deposited in non-adipose

tissues, a phenomenon called ectopic fat deposition [248]. The lipid overload of these tissues also

promotes the accumulation of intracellular signaling molecules, which causes cell dysfunction and

damage (generally termed as lipotoxicity) [249, 250]. Specifically the enrichment of DAGs and

ceramides is suggested to result in an inhibition of insulin signaling [230], caused by an activation of

c/nPKC and aPKC, respectively. This activation blocks the phosphorylation and/or translocation of

important proteins involved in insulin-signaling and inhibits cellular insulin response [230].

In cells, DAG can be generated by three different pathways. (i) The breakdown of PLs catalyzed by

PLC, (ii) the glycerolipid de novo synthesis (by the consecutive reaction of GPATs, AGPATs and

PAPases/lipins) or (iii) the hydrolysis of TAG, catalyzed by TAG hydrolases, like ATGL or HSL. All of the

mentioned pathways are located in different compartments of the cell. Degradation of PLs catalyzed

by phospholipases occurs mainly at the plasma membrane. In contrast, most of the enzymes involved

in DAG synthesis are located at ER membranes. Finally, TAG hydrolysis occurs mainly at the surface of

LDs. DAG generation in different cellular compartments suggests that their metabolic fate as well as

their signaling potential may differ. The signaling potential of DAGs also depends on their

stereospecific conformation. Several studies delineated that only sn-1,2 but not the two other DAG

isoforms, sn-2,3 and sn-1,3, are able to activate novel and conventional PKC isoforms [152-154].

Moreover, several studies have shown that increased levels of DAG in peripheral tissues but not of

adipose tissue are associated with IR in both rodents and humans [177]. Together, these findings

argue for distinct DAG pools in non-adipose and adipose tissues, which may differ in their

stereochemistry as well as in their localization. PKC-activating sn-1,2 DAG is the intermediate of both

PL/TAG de novo synthesis and the hydrolysis of PLs by PLC-dependent activity. In contrast, the

stereochemistry of DAG, which derives from the lipolytical breakdown of TAG, catalyzed mainly by

ATGL, is unknown and may explain why adipose tissues do not develop IR.

In this study the stereo/regioselectivity of ATGL was investigated to unravel DAG isoforms, which are

formed in the initial step of TAG hydrolysis. In addition, the FA-preference of ATGL as well as the

activity of ATGL against PL substrates was studied. Results showed that ATGL hydrolyzes all, in

rodents highly abundant, species of long-chain, saturated, and unsaturated FAs in vitro. Highest

activity was observed for C16:1 and C18:1 esters. The same selectivity of ATGL was concluded from

Discussion

66

the analysis of neutral lipids, which accumulate in adipose tissue of ATGLko mice in vivo.

Interestingly, ATGL also hydrolyzes LD-associated PLs. Yet, in comparison to TAG hydrolysis the rate

for PL hydrolysis is much lower. However, the regioselectivity of ATGL for PL-bound FAs is currently

unknown. This study also demonstrated that ATGL by itself (without co-activation by CGI-58)

hydrolyzes TAG highly selective at sn-2 position, thereby generating sn-1,3 DAG. Interestingly, the sn-

2 regioselectivity of ATGL expands to sn-1 position upon co-activation by CGI-58, generating

additionally sn-2,3 DAG. Notably, ATGL did not hydrolyze sn-3 esters, hence did not generate sn-1,2

DAG in detectable amounts. Additionally, HSL and DGAT2, which catalyze important, subsequent

reactions of DAG, also prefer the generated sn-1,3 DAG isomer as substrate.

Results of this study, in combination with published data, reveal the entire selectivity of the enzymes

involved in TAG catabolism of adipose tissue LDs, based on stereo/regiochemical considerations (Fig.

42).

FIGURE 42. Stereo/regioselectivity of enzymes involved in the lipolytic breakdown of TAG. TAG is hydrolyzed at sn-2

position by ATGL, or at sn-1 or sn-2 position by ATGL co-activated by CGI-58 yielding sn-1,3 DAG or sn-1,3 and sn-2,3 DAG,

respectively. All of the generated DAG species exhibit a FA ester at sn-3 position, which is preferentially hydrolyzed by HSL

yielding either sn-1 or sn-2 MAG. MGL does not exhibit preference for MAG isomers and hydrolyzes both sn-1 and sn-2

MAG. ATGL, adipose triglyceride hydrolase; CGI-58, comparative gene identification-58; DAG, diacylglycerol; FA, fatty acid;

G, glycerol; HSL, hormone-sensitive lipase; MAG, monoacylglycerol; MGL, monoglyceride lipase; TAG, triacylglycerol.

Discussion

67

In basal state, ATGL generates sn-1,3 DAG. ATGL additionally generates sn-2,3 DAG upon co-

activation by CGI-58. Both generated DAG species are preferred substrates for sn-3 selective HSL

[27], which generates sn-1 and sn-2 MAG. MGL finally catabolizes both provided MAG isoforms [93].

Thus, the stereo/regioselectivity of the entire enzymatic cascade catalyzing TAG breakdown at

cytoplasmic LDs suggests, that the involved enzymes have co-evolutionary developed in consecutive

reactions of lipolysis to optimize hydrolytic efficiency (Fig. 42).

The stereo/regioselectivity of ATGL in the absence of the co-activator CGI-58 is unique in that ATGL is

so far the only mammalian lipase described to hydrolyze TAG selectively at sn-2 position. Earlier

studies, which focused on the stereochemical characterization of a variety of animal and microbial

lipases clearly showed that the large majority of lipases exhibit sn-1 or sn-3 selectivity. Only very few

microbial lipases are able to catalyze the hydrolysis of the secondary ester bond at sn-2 position of

TAG. Out of those, only Candida Antarctica A lipase was found to exhibit a clear positional selectivity

for the sn-2 position [14]. Furthermore, all stereochemically characterized TAG lipases involved in

digestion and lipid absorption, like lingual lipase (LL), PAL, and gastric lipase (GL) exhibit preference

for sn-1 or sn-3 position of TAG. PAL, GL, and LL hydrolyze TAG specifically at the sn-3 position but

also hydrolyze DAGs. Thus, they break down TAG into FAs and MAG [13, 14, 251-253]. An example

for a positionally unspecific lipase is bile-salt stimulated lipase (BSSL), which hydrolyzes all positions

of TAG, generating glycerol and FAs [254]. In contrast to intestinal lipases, lipoprotein lipase (LPL),

which depletes circulating lipoproteins from TAGs, exhibits sn-1 specificity for TAG and also

hydrolyzes DAG at the sn-2 position, generating FAs and sn-3 MAG [13, 14, 255].

Intracellularly, several other lipases are supposed to hydrolyze TAG. In this context, a number of

studies focused on triacylglycerol hydrolase/carboxylesterase 3 (TGH1/Ces3), which is mainly

expressed in liver, and to a lower extent in WAT, kidney, and heart [256, 257]. TGH/Ces3 localizes to

the ER, especially to areas surrounding cytosolic LDs and mitochondria [258], and catalyzes the

hydrolysis of long-, medium-, and short-chained TAG substrates with so far unknown

stereoselectivity [257, 259]. Besides TGH1/Ces3, several members of PNPLA protein family, like

adiponutrin (PNPLA3), GS2 (PNPLA4), and GS2-like (PNPLA5) as well as triacylglycerol hydrolase

2/carboxylesterase ML1 (TGH2), are described to possess TAG-hydrolase activity, even though

activities are not comparable with that of ATGL. For most of them, like TGH2, GS2-like, or adiponutrin

the stereo/regioselectivity, if any, is unknown. Only human GS2 was shown to hydrolyze TAG at rac-

1/3 and sn-2 position [74]. So far, more detailed studies on the physiological role of GS2 are lacking,

since the mouse genome lacks GS2 and the rat GS2 lacks activity [66, 74]. Besides ATGL, HSL is the

only other intracellular TAG hydrolase involved in TAG-turnover. HSL exhibits also DAG, MAG, CE, and

Discussion

68

RE hydrolase activity. Thus, if HSL hydrolyzes TAG, it most likely also degrades the product (DAG) to

MAG and FAs. HSL exhibits sn-1/3 preference for TAG and hydrolyzes DAG predominantly at sn-3

position, thereby generating sn-2 MAG [27, 260]. Consistent with previous reports, in vitro studies of

this work revealed that HSL prefers sn-1,3 DAG and not sn-1,2 or rac-1,2/2,3 DAG as substrate. In

contrast, ATGL exhibits clear sn-2 preference for TAG hydrolysis and does not hydrolyze DAG. Upon

co-activation by CGI-58, ATGL becomes much more active and expands its selectivity (to sn-1). The

mechanism by which CGI-58 co-activation of ATGL broadens the regioselectivity of the enzyme

remains to be elucidated. The regioselectivity of ATGL may have important physiological relevance. It

can be assumed that during basal hydrolysis of TAG, when ATGL is not co-activated by CGI-58, only

MUFAs and PUFAs are released from the sn-2 position of TAG since unsaturated FAs are very

abundant at the sn-2 position of TAG [261]. In contrast, activation of lipolysis by hormones, like ß-

adrenergic receptor agonists, leads to stimulation of lipolysis and co-activation of ATGL by CGI-58.

During stimulated lipolysis ATGL broadens its stereo/regioselective spectrum from sn-2 to sn-1

position, thus generating both sn-1,3 and sn-2,3 DAGs. In addition, stimulated lipolysis also leads to

increased FA mobilization, which are predominately released into circulation and used from

peripheral tissues for energy production.

Another aim of this study was to investigate the FA preference of ATGL. In vitro measurements

showed that ATGL exhibits highest activities against tripalmitolein, followed by triolein, trilinolenin,

and trilinolein. As expected, upon co-activation by CGI-58 ATGL hydrolase activity increased but FA-

selectivity remained unchanged. Furthermore, no major differences in ATGL and ATGL/CGI-58-

dependent activity were found when one of the C18:1 esters of trioleate was substituted by a

saturated FA, like C16:0 or C18:0. These results show that ATGL is able to hydrolyze various TAG-

substrates, even if they contain unsaturated FAs. In agreement, FA-composition analysis of plasma-

FAs and of WAT-TAG from ATGLko and wt mice indicated that ATGL hydrolyzes all major saturated

and unsaturated FA species. Highest differences in FA composition of WAT-TAG between wt and

ATGLko mice were found for unsaturated FAs, in particular for C16:1 and C18:1. The observed FA

preference for length and degree of saturation of ATGL might be a direct result of its

stereo/regioselectivity for TAG hydrolysis, because early studies of Brockerhoff et al. showed that in

WAT-TAG more than 50% of total C16:1 are located at the sn-2 position and more than 75% of all FAs

at sn-2 position of TAG are unsaturated FA species [261]. Thus, it is conceivable that ATGL-mediated

hydrolysis of TAG releases preferentially unsaturated FAs because they are very abundant at sn-2

position of TAG and not because ATGL preferentially cleaves unsaturated FA at the sn-2 position.

Discussion

69

In addition to the FA-composition of WAT-TAG, also the FA-composition of circulating lipid species

was investigated. Upon fasting, WAT lipolysis releases FAs into the circulation. A large proportion of

FAs is further utilized for VLDL-TAG synthesis within hepatocytes. Thus, the FA-selectivity of ATGL in

adipose tissue lipolysis may indirectly influence the FA-composition of circulating VLDL-TAG. If the FA

release of WAT is blunted, such as in ATGLko mice, then also the plasma levels of VLDL-TAG are

drastically reduced. Furthermore, FA-composition analysis revealed that, besides an overall decrease

of plasma FAs in ATGLko mice, especially unsaturated FAs like C16:1 and C18:1 were largely

diminished in plasma FAs and VLDL-TAG. This goes along with the FA-preference of ATGL observed

during in vitro experiments since ATGL showed highest activity for C16:1 and C18:1 esters. The direct

involvement of hepatic ATGL in VLDL production is unclear, since liver-specific ATGLko mice show

drastic accumulation of TAG in cytosolic LDs but no alterations in circulating VLDL-TAGs [262].

The FA preference of ATGL is further relevant in the perspective that FAs are known ligands involved

in intracellular signaling, e.g. via activation of PPARs [211, 212]. Interestingly, unsaturated, but not

saturated FAs have strong signaling potential, because they exhibit high activation potential for

PPARs [211]. MUFAs, like C16:1 and C18:1, which are preferentially hydrolyzed by ATGL, are potent

activators for PPARα. In contrast, activation potential of MUFAs is much less for PPARβ or PPARγ

[211-213]. Studies regarding PPAR-activation of PUFAs, like C18:2 and C18:3, showed that PUFAs can

activate all PPAR species to similar extends [211, 212]. PUFAs in plasma and WAT TAG were not

determined in this study since they are not very abundant and were most of the times below

detection limit of the method. However, since unsaturated FAs are preferentially release by ATGL-

mediated TAG hydrolysis it seems likely that these FA species may in part also act as ligands and

facilitate PPAR signaling. Thus, if ATGL-mediated TAG breakdown is defective (e.g. ATGLko), the

impaired MUFA release may be causative for reduced PPARα activation as observed in ATGLko

animal model [217].

Additionally, the hydrolysis of specific FA species could induce the synthesis of bioactive molecules,

e.g. ceramides. Ceramide synthesis highly depends on C16:0 supply and ceramide levels are accepted

to be tightly connected to alterations in aPKC-dependent signaling. It is known that ceramides act as

ligand and activator of e.g. PKCζ which leads to an inhibition of PKB/Akt and consequently disturbed

insulin response and glucose uptake [230]. The fact that C16:0 levels are largely unaltered in ATGLko

animals and that ATGL exhibits no specific preference for C16:0 in vitro suggests that ATGL is not

directly involved in the provision of C16:0. Thus, ATGL may not directly affect intracellular ceramide

levels.

Discussion

70

Intracellular LDs are formed by a core of hydrophobic TAGs surrounded by a monolayer of

amphipathic PLs. This serves as mediator between TAG and the aqueous cellular environment. In

agreement with previous studies, which reported PLA2 activity for ATGL [5], this study demonstrates

that ATGL is able to hydrolyze PLs. Testing a variety of common PLs, it was evident that ATGL

hydrolyzes all investigated PL species, even though activities against PLs are much smaller than

against TAG. Interestingly, the rate of hydrolysis of micellar PLs by ATGL could not be co-activated by

CGI-58. The finding that ATGL can hydrolyze PLs independent of the co-activator CGI-58 raises the

questions about the metabolic implications of this hydrolytic activity. From results of this study, it

can be deduced that ATGL can hydrolyze PLs if presented in micellar form (with no hydrophobic core)

as well as in liposomal form, surrounding other neutral lipids. The latter liposomal substrate may

more reflect cellular LDs. Interestingly, addition of CGI-58 to ATGL led to an increase of ATGL´s

phospholipase activity, when TAG but not CE liposomes were used as substrate. The mechanism

underlying this substrate-dependent effect is so far unclear.

The phospholipase activity of ATGL as well as the positional selectivity against TAG is consistent with

enzymatic activities of other PNPLA protein family members. Next to ATGL, this protein family also

includes characterized sn-1 and sn-2 specific phospholipases (PNPLA8/9) and sn-1 specific lyso-

phospholipases (PNPLA6/7) [263-266]. The observed similarities in stereo/regioselectivity suggest a

common ancestry of these enzymes. To date it is unknown whether PNPLA6-9 contribute directly to

lipid turnover on cytoplasmic LDs. The fact that phospholipase activity of ATGL is much weaker than

TAG-hydrolase activity could match the abundance of the two substrates at LDs. Since PLs form the

surface of LDs, the molar amount of PLs is much smaller than that of TAG of the core. An explanation

why ATGL also exhibits phospholipase activity could be that the enzyme gains better access to the

hydrophobic TAGs by hydrolyzing PLs of the LD monolayer. Another explanation might be a spatial

consideration that during LD degradation also PLs must by hydrolyzed since LDs loose TAG and

become smaller. This may also implicate an explanation for the CGI-58-dependent enhancement of

ATGL´s phospholipase activity when TAG liposomes were used. Since TAG-hydrolysis is the main

activity of ATGL a supply of both TAG and PLs yields primary in TAG-degradation. Consequently the

size of used liposomes decreases, thus, surface area increases and provides potentially more PL

substrate, thereby enhances ATGL´s activity against PLs. The phospholipase activity of ATGL also

raises the possibility that in the process of lipolysis, lipid intermediates, such as arachidonic acid (a

very abundant FA at the sn-2 of PLs) and lyso-PLs, which could act as signaling molecules in cells, are

generated. In any case, further studies are necessary to confirm and clarify the phospholipase activity

of ATGL and its cellular function.

Discussion

71

The stereochemical analysis of DAG, which are formed by ATGL hydrolysis of TAG showed that ATGL-

mediated catabolism of TAG generates sn-1,3 and sn-2,3 DAGs. Interestingly, the stereochemistry of

DAG isoforms, which derive from ATGL catabolism is also corroborated by the accumulation of sn-1,3

DAG (65% of total DAG) in WAT of HSLko mice. In addition to sn-1,3 DAG, HSLko mice also

accumulate sn-2,3 DAG, constituting approximately 25% of the residual DAG. Interestingly, these

DAG isoforms are apparently not the “right” stereochemical species, to act as ligand for PKC [152-

154]. This conclusion is supported by the phenotype of HSLko mice. These mice do not develop a

clear defect in insulin response, despite massive accumulation of DAGs in many tissues, including SM

[56, 59]. In this regard the genetic background of the HSLko strains may play a role. Some strains

show indications for impaired insulin signaling [199, 200], whereas other, including the HSLko strain

used in this study, showed the opposite, increased hepatic insulin sensitivity [197, 198]. So far, this

discrepancy is entirely unsolved. Results obtained in this study show no indications of impaired

insulin sensitivity/signaling. In contrast, insulin tolerance tests indicated rather increased insulin

sensitivity in HSLko mice as compared to wt littermates. Similarly, Voshol et al. [197] and Park et al.

[198] monitored increased hepatic insulin sensitivity during hyperinsulinemic-euglycemic clamp

studies. In contrast, both clamp studies of Mulder et al. [199] and glucose tolerance tests of Roduit et

al. [200] suggest a decrease in insulin sensitivity. Additionally, results of this study show that the

expression levels of SM and liver specific protein kinase C isoforms, PKCθ and PKCε, respectively,

were tentatively decreased in HFD-fed HSLko mice as compared to wt animals. Both PKC isoforms

target IRS proteins, one of the initial events in insulin signaling. PKCθ can directly phosphorylate IRS1

on Ser1101 and Ser307 and PKCε increases IRS1 phosphorylation at Ser636/639 [230]. Such serine

phosphorylations of IRS1 are described to prevent insulin-induced IRS1 tyrosine phosphorylation,

which leads to inhibition of PI3K activity and PKB/Akt phosphorylation and hence blocks insulin-

induced cellular glucose uptake [230]. Since both PKC isoforms are described to negatively regulate

insulin signaling [177] these data indicate that in the present study, wt mice are more prone to HFD-

induced IR than HSLko mice. All in all, data of this study give no indication that DAGs, which

accumulate due to HSL deficiency, lead to an impairment of insulin signaling. Additionally, the fact

that TAG-derived DAGs are sn-1,3 and sn-2,3 isoforms and are generated at LDs and, thus, may never

end up at the plasma membrane where PKC resides also argues against a role of lipolysis-derived

DAGs in PKC signaling.

Intracellular DAGs can potentially undergo several anabolic reactions. HSL can catabolize them in the

process of lipolysis. DGKs can convert them to PA by phosphorylation of the free OH-group. CPT can

transform them to PC and the acyltransferases DGAT1 and DGAT2 can acylate them thereby forming

Discussion

72

TAG. CPT and DGKs represent membrane-bound enzymes. Hence, these reactions are unlikely to

occur on cytoplasmic LDs. Additionally, neither sn-1,3 DAG nor sn-2,3 DAG are substrates for these

enzymes (only sn-1,2 DAG is the intermediate of PL metabolism and glycerophospholipid de novo

synthesis). Both, the wrong stereochemistry of TAG-derived DAGs and the “mislocalization” on LDs

makes it unlikely that these DAGs are endogenous substrates for PL synthesis without prior

transesterification to sn-1,2 isomer and DAG transport systems. To exclude a misinterpretation of

obtained data, caused by DAG remodeling in vitro, transesterification of DAG was additionally

investigated. Results show that no conversion of rac-1,2/2,3 to sn-1,3 and vice versa was detectable.

Consistently, no mammalian DAG-transport proteins and no in vivo occurring transesterification of

DAG have been described yet. This suggests that the DAG pool, formed by ATGL, is not a precursor

for PL synthesis.

Another aim of this study was to determine the stereo/regioselectivity of the re-esterification of

DAG, which is catalyzed by DGAT1 and DGAT2. It turned out that DGAT2, but not DGAT1, exhibits

clear preference for sn-1,3 DAG. This finding was independent of the enzyme sources and confirmed

in acyltransferase assays using either cell homogenates containing DGAT1 or DGAT2 or homogenates

of murine WAT. DGAT1 converts rac-1,2/2,3 DAG much more efficiently than sn-1,3 DAG, whereas

DGAT2 exhibits the opposite substrate selectivity. These results were rather unexpected since the

role of DGATs on LDs has not been investigated so far. Until now, both DGAT enzymes are thought to

catalyze the same, final reaction of de novo TAG biosynthesis by esterifying sn-1,2 DAG with long-

chain FA-CoAs at the ER. The observed differences in substrate selectivity may be due to structural

differences of these two enzymes. Both belong to unrelated protein families. DGAT1 is a member of

the ACAT/DGAT1 gene family which is a subfamily of the MBOAT superfamily, whereas DGAT2

belongs to the DGAT2 gene family [126]. These two enzymes show no homology and it is known that

they cannot functionally compensate for each other [140, 267]. To date, the reaction mechanism

behind the observed metabolic differences, apart from structural differences, is elusive. Both DGATs

localize to the ER but only DGAT2 has been also reported to be associated within cytosolic LDs [139,

268]. Our finding that DGAT2 prefers sn-1,3 DAG as acyl-acceptor, generated by ATGL on LDs, marks

a new difference with regard to biochemical properties of both DGAT enzymes. Furthermore, the

finding that DGAT2 localizes on LDs together with the preference for sn-1,3 DAG suggests that ATGL

and DGAT2 may act coordinately on LDs to facilitate TAG hydrolysis and re-esterification, thereby

remodeling the degree of saturation of sn-2 FA esters. If this holds true, two distinct DAG pools for

DGAT enzymes would exist (Fig. 43).

Discussion

73

Both DGAT enzymes would esterify DAG pool 1 consisting of sn-1,2 DAGs, generated during de novo

glycerolipid synthesis at the ER. On the other hand, a sn-1,3 DAGs (pool 2), generated by ATGL on

cytoplasmic LDs would serve as substrate for DGAT2, exclusively. Such a local separation as well as

stereochemical distinction between different DAG pools could explain the differences observed in

mice carrying a global DGAT1 or DGAT2 deficiency. In line, a recent study overexpressing either

DGAT1 or DGAT2 in McA-RH7777 cells described surprisingly different phenotypes [140]. Despite

higher in vitro acyltransferase-activity when overexpressing DGAT1, cells expressing DGAT2 show

significant higher TAG mass, as well as glycerol incorporation into cellular TAG moiety. Furthermore,

LDs of cells overexpressing DGAT1 were considerably smaller compared to LDs of DGAT2 expressing

cells [140]. Another recent publication suggested different substrate sources of DGATs in HepG2

hepatoma cells. Studies using stable isotope-labeled substrates show that DGAT1 mainly

incorporates exogenously FAs to glycerol, whereas DGAT2 primary uses endogenously synthesized

FAs for TAG generation [144]. In combination with the observations of this study, it can be concluded

that DGAT1 and DGAT2 exhibit different functions in cellular TAG synthesis as evident by their

different localization, substrate source, and stereo/regioselectivity.

The observation that high doses of MgCl2 (100 mM) can result in an inhibition of either DGAT2 alone

or DGAT1 and DGAT2 in in vitro acyltransferase experiments was rather unexpected. High doses of

MgCl2 (> 100 mM) are described to inhibit specifically DGAT2, which has been demonstrated in mice

lacking DGAT1 as well as in cells overexpressing DGAT2 [118]. Results of this study show that the

presence of 100 mM MgCl2 leads to an exclusive inhibition of DGAT2 when sn-1,2 DAG is used as

substrate. In contrast to the expected inhibition of DGAT2, inhibition of DGAT1 (~50%) is additionally

observable when sn-1,3 DAG is used as substrate. Thus, MgCl2-dependent inhibition of DGAT activity

depends on the involved enzyme and the supplied substrate isoform. A comparison with the former

mentioned study of Cases et al. [118] is difficult because they did not use cells overexpressing DGAT1

do confirm their assumed selectivity of inhibition. In addition, they exclusively used sn-1,2 DAG as

substrate and thus may have overlooked the finding that MgCl2 also inhibits DGAT1 when sn-1,3 is

used as acyl-acceptor. In summary, data indicate that MgCl2 can be used as specific inhibitor for

DGAT2 in experiments containing sn-1,2 DAG. In experiments using sn-1,3 DAG the additional

inhibition of DGAT1 can falsify results. So far, the the mechanism by which MgCl2 differentially

inhibits DGAT enzymes is unknown.

In summary, results of this study demonstrate that ATGL hydrolyzes TAG selectively at the sn-2

position, thereby generating sn-1,3 DAG in vitro and in vivo. Generated sn-1,3 DAG constitutes the

preferred substrate for further hydrolysis by HSL or re-esterification catalyzed by DGAT2.

Discussion

74

Furthermore, co-activation of ATGL by CGI-58 leads to enhanced hydrolytic activity and to the

expansion of stereo/regioselectivity to sn-1 but not sn-3 position, therby additionally generating sn-

2,3 DAG but not sn-1,2 DAG. Since ATGL-dependent TAG hydrolysis does not generate sn-1,2 DAG,

the isoform that is known to activate PKCs, it is unlikely that lipolysis-derived DAGs regulate PKC-

activity. The preference of ATGL for long-chain unsaturated FAs, which are known PPAR ligands,

suggests that ATGL activity may contribute to PPAR activation in that way. Interestingly, ATGL is also

able to hydrolyze a variety of PLs, though with low activity. Moreover, results obtained from HSLko

mice showed that these mice cumulate sn-1,3 and sn-2,3 DAGs that are selectively generated by

ATGL-dependent lipolysis. Furthermore studies on HSLko mice suggest that accumulation of TAG-

derived DAG is not crucially involved in disturbed insulin signaling.

FIGURE 43. “3-pool” model of intracellular DAG distribution. Pool I consists of sn-1,2 DAG formed at the ER during de novo

lipogenesis. DGAT1, DGAT2 as well as CPT can metabolize newly generated sn-1,2 DAG. Pool II consists of sn-1,3 ± sn-2,3

DAGs, which are generated on cytoplasmic LDs by ATGL ± CGI-58-dependent TAG hydrolysis. Both DAG species are targets

for subsequent hydrolysis catalyzed by HSL or re-esterification catalyzed by DGAT2. sn-1,2 DAGs of pool III function as

activators of various PKCs and are generated by PLC at the plasma membrane. Furthermore, these DAGs are substrates for

DGKs and DAGLs. ATGL, adipose triglyceride lipase; CPT, choline:1,2-diacylglycerol cholinephosphotransferase; CGI-58,

comparative gene identification-58; DAG, diacylglycerol; DGK, diacylglycerol kinase; DAGL, diacylglycerol lipase; DGAT1/2,

diacylglycerol-O-acyltransferase 1/2; HSL, hormone-sensitive lipase; MAG, monoacylglycerol; PA, phosphatidic acid; PKC,

protein kinase C; PL, phospholipid; PLC, phospholipase C; TAG, triacylglycerol.

Observations of the present study in combination with previously published data support a “3-pool”

model of intracellular DAG compartmentation. Pool 1 is located at the ER and consists of sn-1,2 DAGs

that are generated during de novo lipogenesis by either MGATs or PAPases/lipins. This pool is

accessible and metabolized by DGAT enzymes and CPT (Fig. 43). TAG-derived DAGs form pool 2 at

Discussion

75

cytoplasmic LDs. This pool comprises sn-1,3 DAGs ± sn-2,3 DAGs and is generated by ATGL ± CGI-58.

DAGs of this pool are catabolized by HSL-mediated hydrolysis or re-esterified to TAG by DGAT2 (Fig

43). Pool 3 contains sn-1,2 DAGs, which are generated by PLC-dependent hydrolysis of PLs located at

the plasma membrane. DAGs of this pool can either serve as precursor for de novo synthesis of PL via

phosphorylation catalyzed by DGKs, or be degraded by membrane-bound DAGLs. So far, only sn-1,2

DAGs of pool 3 are widely accepted to be involved in cellular signaling events by activating novel and

conventional isoforms of PKC (Fig 43). The local separation of these pools within the cell as well as

the clear differences in stereo/regiochemistry of included DAGs may provide explanations to

previously confusing date regarding DAG-induced signaling and lipid metabolism.

Materials & Experimental Procedures

Materials & Experimental Procedures

77

I) Materials

Chemicals and buffers

Materials, chemicals and radiochemicals were obtained from either Sigma-Aldrich (St. Louis, MO) or

GE Healthcare (Waukesha, WI). Lipids were supplied by either Larodan Fine Chemicals (Malmō,

Sweden; TAGs), Sigma-Aldrich (DAGs) or Avanti Polar Lipids (Alabaster, AL; PLs).

Solution A: 0.25 M sucrose, 1 mM EDTA, 1 mM dithiothreitol, 20 µg/ml leupeptine, 2 µg/ml antipain

and 1 µg/ml pepstatin, pH of 7.0

Phosphate buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 4.7 mM Na2HPO4, 1.4 mM KH2PO4, pH

7.4

Animals

Mice were housed on a regular light/dark cycle (12 h/12 h) and had ad libitum access to water and

normal CD (fat: 4.5%, protein: 22.1%, starch: 35.8%, sugar: 5.0%; w/w fat from Ssniff, Soest,

Germany) or HFD (fat: 34.0%, protein: 24.1%, starch: 1.1%, sugar: 23.8%; w/w fat from Ssniff, Soest,

Germany). 8-10 weeks-old male mice on CD or 20-22 weeks old male mice kept the last 12 weeks on

a HFD were used for studies. Mice with a global deletion of HSL (HSLko) or ATGL (ATGLko) were

generated by targeted homologous recombination and backcrossed to C57BL/6J at least 5 times, as

described previously [59, 75]. Blood and WAT of non-fasted and 8 h (12:00 – 08:00 am) fasted mice

on CD were used for lipid analyses. Tissue samples of liver and m. gastrocnemius (SM) of HFD-fed

mice fasted for 8 h (12:00 – 08:00 am), following 1 h of refeeding, were used for immunoblot

analysis. Body mass composition was assessed using TD-NMR minispec Live Mice Analyzer system

(LF90II, Bruker Optik GmbH, Ettlingen, Germany). Blood was collected via retro-orbital puncture from

isoflurane-anesthetized (Baxter, Deerfield, IL) mice. WAT, LIV, and SM samples were surgically

removed from cervically-dislocated mice. Tissue samples were washed in PBS, containing 1 mM

EDTA, 100 IU/ml heparin and disrupted using an Ultra Turrax® (IKA, Staufen, Germany), either in ice-

cold solution A for enzymatic activity measurements and immunoblotting or in methanol for further

lipid extraction and analysis. For analysis of brain lipids, mice were treated as described previously

[245].

Materials & Experimental Procedures

78

II) Experimental Procedures

cDNA cloning of recombinant proteins

pcDNA4/HisMaxC vector (Invitrogen, Carlsbad, CA) constructs containing the entire open reading

frame of murine ATGL, murine CGI-58, and murine HSL were generated as described previously [70].

Constructs of flag-tagged DGAT1 and DGAT2 were kindly provided by Robert V. Farese Jr. (Gladstone

Institute of Cardiovascular Disease, and Departments of Medicine and Biochemistry & Biophysics,

University of California, SF) and generated as described [140]. As control, pcDNA4/HisMaxC

containing β-galactosidase was used (Invitrogen). For purification of CGI-58, respective coding

sequence was subcloned into pYex4T-1 vector (CLONTECH Laboratories Inc., Mountain View, CA) as

previously described [68].

Purification of GST-tagged CGI-58

GST-tagged CGI-58 was expressed in S. cerevisiae BY4742 strain. Affinity purification of the GST-

fusion protein was performed as described [68]. pYex4T-1 vector was transformed into the S.

cerevisiae BY4742 strain. Copper promoter-driven expression of GST-CGI was induced by raising S.

cerevisiae in YNB-urea media containing 0.5 mM CuSO4. Cells were harvested and protoplasts were

generated using zymolyase. Protoplasts were disrupted by sonication using a Virsonic 475 (Virtis,

Gardiner, NJ) in the presence of 0.2% NP-40. GST-tagged CGI-58 within the supernatant was purified

using Glutathione-Sepharose beads (GE Healthcare) and further dialyzed overnight with 150 mM KCl,

10 mM potassium phosphate buffer (pH 7.0), and 0.01% NP-40.

Expression of recombinant proteins

SV-40 transformed monkey embryonic kidney cells (Cos7; ATCC, CRL-1651) were cultivated in

Dulbecco´s modified eagle medium (DMEM; GIBCO, Invitrogen) containing 10% fetal calf serum (FCS)

supplemented with penicillin (100 IU/ml, GIBCO) and streptomycin (100 µg/ml, GIBCO) at standard

conditions (95% humidified atmosphere, 37°C, 7.5% CO2). Transfection of cloned constructs was

performed using 1 µg DNA complexed with Metafectene (Biontex, Munich, Germany) in FCS-free

medium. After 4 h media were changed to DMEM containing 10% FCS. Finally, after 48 h cells were

washed twice in PBS and collected using a cell scraper.

Materials & Experimental Procedures

79

Preparation of tissue and cell homogenates

Cells were disrupted in solution A by sonication (3x6 sec on ice, 20% output; Virsonic 475). Both, cell

and tissue homogenates were centrifuged to remove nuclei and unbroken cells (1,000 x g, 4°C, 30

min). For microsomal-free cytoplasmic fraction, homogenates were further centrifuged (100,000 x g,

4°C, 60 min). The microsomal pellet was resuspended in ice-cold solution A to obtain microsomal

fraction.

Protein determination

Protein concentrations of homogenates and prepared cellular fractions were determined using Bio-

Rad protein assay reagent (BioRad laboratories, Hercules, CA) and bovine serum albumin (BSA) as

standard.

Immunoblotting

Cos7-cell homogenates (20 µg protein) or WAT microsomal fraction (30 µg protein) or microsomal or

cytosolic fraction of LIV and SM samples (40 µg protein) were separated according to their molecular

weight by sodium-dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a 10%

polyacrylamide gel and tris/glycine (1.6 M glycine, 0.8% SDS, 200 mM Tris-HCl, pH 8.3) as

electrophoresis buffer. Then, proteins were transferred onto polyvenylidenfluorid (PVDF) membrane

(Carl Roth GmbH, Karlsruhe, Germany) using CAPS buffer (1 mM 3-cyclohexylamino-1-propansulfonic

acid). Unspecific binding sites were blocked using 10% non-fat dry milk (Carl Roth) dissolved in TST

buffer (0.15 M NaCl, 0.1% tween-20 v/v, 50 mM Tris-HCl, pH 7.4) for 2 h. His- or Flag-tagged proteins

(His-tag: ATGL, CGI-58, β-galaktosidase (LacZ) and HSL; Flag-tag: DGAT1 and DGAT2) were detected

by hybridization with either primary anti-His (GE-Healthcare) or anti-Flag antibodies (1:5000, 2% milk

in TST buffer), following anti-mouse horseradish-peroxidase (HRP)-linked secondary antibody (GE

Healthcare) (1:10000, 2% milk in TST buffer). Endogenous expression of DGAT1 and DGAT2 in murine

WAT was detected using polyclonal DGAT1 or DGAT2 primary antibody (ProSci, Poway, CA) (1:1000,

2% milk in TST buffer) and anti-rabbit HRP-linked secondary antibody (Vector laboratories Inc.,

Burlingame, CA) (1:10000, 2% milk in TST buffer). Endogenous expression of PKCα, PKCε, PKCθ,

GapDH, and HSL in LIV or SM samples was detected using respective primary antibodies (Cell

Signaling, Danvers, MA, USA) (1:1000, 2% milk in TST buffer) and anti-rabbit HRP-linked secondary

Materials & Experimental Procedures

80

antibody. Chemiluminescence was induced by ECLplus Kit (GE Healthcare) and detected by exposure

to X-ray films (Hyperfilm™ ECL, GE Healthcare).

Total lipid extraction and separation

Lipids of homogenized tissue samples (WAT, brain [245], or blood) were extracted using

chloroform/methanol/water (2/1/0,6; v/v/v; 1 % acetic acid). For GC-FID measurements, extracts

were supplemented with 20 µg margaric acid (C17:0) and 20 µg trimargarinate per ml as internal

standard. Extraction was performed under steady shaking for 1 h at room temperature (RT). After

centrifugation (1,000 x g, 4°C, 15 min), the organic phase was collected, dried under nitrogen, and

dissolved in 500 μl chloroform. Aliquots (30 μl) were separated by TLC using silica-gel coated plastic

TLC plates (solvent for TAG and FFA separation: hexane/diethyl ether/acetic acid (70/29/1; v/v/v) and

solvent for DAG separation: chloroform/acetone/acetic acid (95/4/1; v/v/v)). Lipids on TLC plates

were stained using iodine vapor. Bands corresponding to selected lipid species were scraped off and

either dissolved in 500 μl toluene, containing 500 μg butylated hydroxytoluene (1 mM in toluene) as

antioxidant and used for determination of FA composition (GC-FID) or lipids were re-extracted with

chloroform and used for HPLC lipid analysis (see below).

Fatty acid determination by gaschromatopgraphy with flame-ionization

detection

FA species were analyzed by GC-FID according to Sattler et al. [269] with some modifications: For

transesterification, 2 ml of borone trifluorid (BF3) were added to lipids, dissolved in 500 µl toluene,

and incubated for 1 h at 110°C in an incubator. Reactions were stopped by addition of 1 ml ice-cold

H2O. Resulting fatty acid methylesters (FAMEs) were extracted twice by addition of 2ml of

hexane/chloroform (4:1, v/v) and shaking for 10 min at RT. After centrifugation (1,000 x g, RT, 10

min) the upper-phase was collected. The combined phases were evaporated under nitrogen and

FAMEs were dissolved in 100 µl of hexane. The GC conditions were set to: split injection (split flow:

15 ml/min, split ratio: 1/5, injection volume: 2 µl), using an injector temperature of 230 °C, and a

wall-coated open tubular fused silica column (25 m, 0.32 mm inner diameter, FFA phase coated, film

thickness = 0.3, Agilent technologies, Santa Clara, CA). The carrier gas consisted of helium. As a

temperature gradient two consecutive ramps from 150 to 260 °C (ramp 1: 5°C/min to 250°C hold for

2 min, ramp 2: 10°C/min to 260 hold for 5 min) were used. FID (Trace-GC 2000series, ThermoQuest

corp., Atlanta, GA) conditions were as follows: base temperature 150°C; gas flows: 200 ml/min air, 30

Materials & Experimental Procedures

81

ml/min hydrogen, and 20ml/min helium. Data acquisition and analysis was done with Xcalibur 2.0

software (Thermo Fisher Scientific, Waltham, MA). For quantitative analysis the corresponding peaks

of FAMEs were integrated and peak areas were calculated using C17:0 as internal standard. FAME

concentrations were calculated as percentage of total FAMEs in a given sample and/or as amounts

per wet tissue weight (nmol/g) or plasma volume (nmol/l).

Enzymatic generation of DAG

The triolein substrate (0.3 mM final concentration in the used buffer) containing no tracer or either

1.0 x 106 cpm 3H-FA [9,10] labeled triolein or 0.5 x 106 cpm 14C-glycerol labeled triolein as tracer (per

sample) and PC (45 µM final concentration in the used buffer) were emulsified on ice in 100 mM

potassium phosphate buffer (pH 7.0) by sonication using a sonicator (3x60 sec on ice, 20% output;

Virsonic 475). Then, substrates were adjusted to 2% bovine serum albumin (BSA; FA free) and 100 µl

of substrate were incubated with cytoplasmic fractions of Cos7-cell lysates overexpressing ATGL (50

µg of protein in 100 µl solution A) without or with 200 ng of purified GST-tagged CGI-58 (in 10 mM

potassium phosphate buffer (pH 7.0), 0.01% NP-40) and HSL-specific inhibitor (12.5 µM of 76-0079 in

DMSO; Novo Nordisk, Kopenhagen, Denmark) in a water bath at 37°C for different periods (40 min,

60 min, or 120 min). The reaction was terminated by extracting the lipids according to Folch et al.

[270] using 1 ml chloroform/methanol (2/1; v/v). Subsequently, lipids were separated by TLC using

chloroform/acetone/acetic acid (95/4/1, v/v/v) as solvent. Bands corresponding to DAG were

scraped off and either re-extracted using chloroform and used for further analysis or radioactivity

was determined by liquid scintillation counting (Tri-Carb 2300 TR; Packard, Meridan, CT).

Determination of hydrolase activity

Different TAG (different FA composition) or diolein (sn-1,2; sn-1,3 or rac-1,2/2,3) substrates were

prepared by emulsification (sonication) with di-C18:1 PC (45 µM final concentration in the used

buffer) in 100 mM potassium phosphate buffer (pH 7.0) on ice using a sonicator (3x60 sec on ice, 20%

output; Virsonic 475). Final concentration was set to 0.25 mM for TAG substrates and to 0.3 mM for

DAG substrates using 100 mM potassium phosphate buffer (pH 7.0). Then, substrates were adjusted

to 2% bovine serum albumin (BSA; FA free). Cos7-cell homogenates (50 µg protein) containing either

HSL or ATGL were incubated without or with cell homogenates containing CGI-58 (50 µg protein) to

give a total volume of 100 µl in solution A to which 100 µl of indicated substrate was added and

incubated for 60 min in a water bath at 37°C. Then 200 µl of 0.1% Triton X-100 were added and

Materials & Experimental Procedures

82

samples were agitated for 10 min at RT. Released FAs were determined enzymatically using NEFA-C

kit (Principle: Activation of FA by acyl-CoA synthetase; Oxidation of FA-CoA by acyl-CoA oxidase;

Peroxidase dependent colour reaction; Wako chemicals, Neuss, Germany) after 10 min of shaking at

RT.

Determination of DAG rearrangement/transesterification

Racemic diolein (mixture of sn-1,2/sn-2,3/sn-1,3 or rac-1,2/2,3) or either sn-1,3 or sn-1,2 diolein

substrate (0.3 mM final concentration in the used buffer) was prepared by emulsification (sonication)

with PC (45 µM final concentration in the used buffer) in 100 mM potassium phosphate buffer (pH

7.0) on ice using a sonicator (3x60 sec on ice, 20% output; Virsonic 475). Then, substrates were

adjusted to 2% bovine serum albumin (BSA; FA free). 100 µl of substrate were incubated either with

100 µg cytosolic protein or 100 µg lysate protein of Cos7-cells in the presence of a HSL-specific

inhibitor (12.5 µM of 76-0079 in DMSO) (total volume: 100 µl in solution A) in a water bath at 37 °C

for 120 min. Before and after incubation, lipids were extracted according to Folch et al [270] by

adding 1 ml chloroform/methanol (2/1; v/v) and separated by TLC (silica gel-coated plastic TLC

plates) using chloroform/acetone/acetic acid (95/4/1, v/v/v) as solvent. Bands corresponding to DAG

were either stained using iodine vapor or scraped of, re-extracted with 10 ml chloroform, dried

under nitrogen and used for chiral analysis.

Determination of DGAT activity

WAT cellular fraction (microsomal, cytosolic, lipid droplet) or Cos7-cell homogenates containing

either DGAT1 or DGAT2 (50 µg protein) were incubated with HSL-specific inhibitor (12,5 µM of 76-

0079 in DMSO), tetrahydrolipstatin/orlistat (20 µM in DMSO, xenical®), niacin (5 mM in water), MgCl2

(100 mM in water), bromoenol lactone (BEL, 5 µM in DMSO, [271]), or DGAT1-specific inhibitor (5 µM

of (2-((1s,4s)-4-(4-(4-amino-7,7-dimethyl-7H-pyrimido[4,5-b][1,4]oxazin-6-yl)phenyl)cyclohexyl)

acetic acid [139] in DMSO) and different diolein substrates (final volume: 100 µl). Substrates were

composed of different diolein isoforms (0.2 mM of sn-1,2, sn-1,3 or rac-1,2/2,3 final concentration in

the used buffer) in Tris-buffer (50 mM, 20 mM MgCl2, pH 7.4) and either 0.8 mM (DAG/PC ratio:

0.25), 0.2 mM (DAG/PC ratio: 1) or 0.05 mM (DAG/PC ratio: 4) PC (final concentration in the used

buffer) and were emulsified on ice by sonication (3x60 sec on ice, 20% output; Virsonic 475).

Subsequently, C18:1-CoA (30 µM final concentration within substrate) and 14C-labeled C18:1-CoA (55

µCi/µmol; 20 µM final concentration within substrate) were added. Substrate was mixed 1:1 with

Materials & Experimental Procedures

83

samples to give a final volume of 200µl and incubated for 10 min at RT. Reactions were stopped and

lipids extracted by addition of chloroform/methanol (2:1, v/v) and separated on TLC (silica gel-coated

plastic TLC plates) using hexane/diethylether/acetic acid (70/29/1, v/v/v) as solvent. Bands

corresponding to TAG were scraped off and radioactivity was determined by liquid scintillation

counting (Tri-Carb 2300 TR).

Acylglycerol determination

Acylglycerol content (TAG, DAG) of lipid extracts from homogenized tissue preparations (WAT, brain

[245]), plasma as well as different non-radiolabeled TAG substrates were determined enzymatically

using INFINITY Triglyceride kit (Principle: Lipase-dependent hydrolysis of acylglycerol to glycerol;

Phosphorylation of glycerol by glycerol kinase; Oxidation of G3P by glycerolphosphate oxidase;

Colour reaction catalyzed by peroxidase using hydrogen peroxide; Thermo Scientific Fisher,

Middletown, VA). Plasma lipids or lipids extracted from TLC bands were measured directly, whereas

TAG substrates where diluted 1:1 with Triton X-100 to give a final concentration of 0.05 % Triton X-

100. Samples were measured photometrical after 10 min of shaking at RT.

Enzymatic hydrolysis of lecithin

Five mg of lecithin (egg yolk, average molecular weight ~800 g/mol) were emulsified in 10 ml of

potassium-phosphate buffer (100 mM, pH 7.4) by sonication (3x60 sec on ice, 20% output; Virsonic

475). Then, emulsion was adjusted to 2% bovine serum albumin (BSA; FA free). Lipid emulsion (500

µl, ~0.7 mM) was incubated with 100 µl of recombinant PLC (B. cereus, 100 IU/ml) for 2 h at 37°C.

Lipids were extracted according to Folch et al. [270] and separated by TLC using

chloroform/acetone/acetic acid (95/4/1, v/v/v) as solvent. Bands corresponding to DAG were

scraped off, re-extracted using chloroform, and subjected to chiral-phase HPLC analysis.

Determination of DAG isomers by chiral-phase HPLC

DAGs obtained either from rearrangement/transesterification experiments, of WAT tissue extracts,

or from TAG hydrolase assays were dried under nitrogen and derivatized according to Itabashi et al.

[272] with slight modifications. In detail, dried DAGs were dissolved in 400 µl dry toluol, 40 µl

pyridine and 2 mg 3,5-dinitrophenylisocyanate were added. The reaction mixture was shaken for 1 h

Materials & Experimental Procedures

84

at RT, dried under nitrogen, resolved in chloroform, and reaction products were separated by TLC

using UV254 silica plates and chloroform/methanol (95/5, v/v) as solvent. Bands corresponding to 3,5-

dinitrophenyl-urethanes were visualized under UV-light, scraped off, and extracted twice in

chloroform. The solvent was evaporated, samples were dissolved in 100 µl

hexane/dichloroethane/ethanol (80/20/2, v/v/v), and subjected to HPLC analysis. The HPLC system

consisted of Waters alliance e2695 separation module (Millford, MA) and an UV/VIS 2489 detector.

For separation a YMC-Pack A-K03 column (YMC Inc, Kyoto, Japan) containing (R)-(+)-1-(1-

naphtyl)ethylamine as stationary phase was used. Samples were analyzed under isocratic conditions

using hexane/dichloroethane/ethanol (80/20/2, v/v/v; flow rate: 1 ml/min) as mobile phase. DAG

derivatives were detected at 226 nm using an UV/VIS 2489 detector (Waters). Peaks corresponding

to DAG species were integrated and expressed as percentage of total DAG using Empower pro

software (Waters).

Intraperetoneal insulin tolerance test (IPITT)

Mice were fasted for 2 h and anestethized using isoflurane following an intraperetoneal bolus of 0.6

IU/kg bodyweight bovine insulin. Blood glucose levels were determined at indicated time points

using an Accu-Check glucometer (Roche Diagnostics, Basel, Switzerland). For every measurement

fresh blood was sampled by tail cutting. Area under the curve was calculated from absolute glucose

values using trapezoid method.

Determination of phospholipase activity

250 µM of 14C-labeled or non-radiolabeled PC was used as substrate for phospholipase activity

analysis. For determination of PL species selectivity, 500 µM non-labeled PC, PS, PE, PG, or PA was

used as substrates. For determination of phospholipase activity against mixed micelles either 50 µM

of 14C-labelled PC and 300 µM of 3H-labelled triolein or 50 µM non-labelled PC and 300 µM

cholesterylpalmitate were used as substrates. All substrates were prepared by emulsification in 100

mM potassium phosphate buffer (pH 7.0) in the presence or absence of 2 mM Ca2+ on ice using a

sonicator (3x60 sec on ice, 20% output; Virsonic 475). Then, substrates were adjusted to 2% bovine

serum albumin (BSA; FA free). All used PLs contained two C18:1 residues. Either purified PLA2 (1 IU;

Naja mossambica m.) or 50 µg protein of Cos7-cell homogenates containing ATGL without or with

purified GST-tagged CGI-58 (total volume: 100 µl in solution A) were incubated with 100 µl of

Materials & Experimental Procedures

85

respective substrate in a water bath at 37 °C for 60 min. Resulting FAs were determined either by

NEFA-C kit (Wako chemicals) or by scintillation counting.

Visualization of DAG substrate size by laser scanning microscopy

One ml of DAG substrates emulsified with PC in the DAG/PC ratios, 0.25 (50 µM/200 µM), 1 (200

µM/200 µM), and 4 (800 µM/200 µM) were incubated with the neutral lipid-specific dye BodiPy®

558/568 C12 (BodiPy, 15 µg/ml, Invitrogen) for 20min at RT. Subsequently, imaging of fluorescently

labeled structures was performed on a Leica SP2 confocal microscope (Leica microsystems, Wetzlar,

Germany) using a ×100, NA 1.40 oil immersion objective. BodiPy fluorescence was exited at 514 nm

and detected at 570 nm.

Statistics

Data are shown as means ± S.D. Statistical significance between two groups was determined by

unpaired Student´s two-tailed t-test. Following levels of statistical significance were used: *, p<0.05;

**, p<0.01; ***, p<0.001.

Publications

Publications

87

First author

Studies on the substrate and stereo/regioselectivity of adipose triglyceride lipase, hormone-

sensitive lipase, and diacylglycerol-O-acyltransferases

Thomas O. Eichmann, Manju Kumari, Joel T. Haas, Robert V. Farese Jr., Robert Zimmermann, Achim

Lass, and Rudolf Zechner

J. Biol. Chem. 2012.

Adipose triglyceride lipase (ATGL) is rate-limiting for the initial step of triacylglycerol (TAG) hydrolysis,

generating diacylglycerol (DAG) and fatty acids (FAs). DAG exists in three stereochemical isoforms.

Here we show that ATGL exhibits a strong preference for the hydrolysis of long-chain FA esters at the

sn-2 position of the glycerol backbone. The selectivity of ATGL broadens to the sn-1 position upon

stimulation of the enzyme by its co-activator CGI-58. sn-1,3 DAG is the preferred substrate for the

consecutive hydrolysis by hormone-sensitive lipase (HSL). Interestingly, diacylglycerol-O-

acyltransferase 2 (DGAT2), present at the endoplasmic reticulum and on lipid droplets (LDs),

preferentially esterifies sn-1,3 DAG. This suggests that ATGL and DGAT2 act coordinately in the

hydrolysis/re-esterification cycle of TAGs on LDs. Since ATGL preferentially generates sn-1,3 and sn-

2,3 it suggests that TAG-derived DAG cannot directly enter glycerophospholipid synthesis or activate

protein kinase C without prior isomerization.

Publications

88

Co-author

FAT SIGNALS - Lipases and Lipolysis in Lipid Metabolism and Signaling.

Zechner R, Zimmermann R, Eichmann TO, Kohlwein SD, Haemmerle G, Lass A, Madeo F.

Cell Metab. 2012 Mar 7;15(3):279-91.

Adipose triglyceride lipase affects triacylglycerol metabolism at brain barriers.

Etschmaier K, Becker T, Eichmann TO, Schweinzer C, Scholler M, Tam-Amersdorfer C, Poeckl M,

Schuligoi R, Kober A, Chirackal Manavalan AP, Rechberger GN, Streith IE, Zechner R, Zimmermann R,

Panzenboeck U.

J Neurochem. 2011 Dec;119(5):1016-28. doi: 10.1111/j.1471-4159.2011.07498.x. Epub 2011 Oct 20.

ATGL-mediated fat catabolism regulates cardiac mitochondrial function via PPAR-α and PGC-1.

Haemmerle G, Moustafa T, Woelkart G, Büttner S, Schmidt A, van de Weijer T, Hesselink M, Jaeger D,

Kienesberger PC, Zierler K, Schreiber R, Eichmann TO, Kolb D, Kotzbeck P, Schweiger M, Kumari M,

Eder S, Schoiswohl G, Wongsiriroj N, Pollak NM, Radner FP, Preiss-Landl K, Kolbe T, Rülicke T, Pieske

B, Trauner M, Lass A, Zimmermann R, Hoefler G, Cinti S, Kershaw EE, Schrauwen P, Madeo F, Mayer

B, Zechner R.

Nat Med. 2011 Aug 21;17(9):1076-85. doi: 10.1038/nm.2439.

Publications

89

Growth retardation, impaired triacylglycerol catabolism, hepatic steatosis, and lethal skin barrier

defect in mice lacking comparative gene identification-58 (CGI-58).

Radner FP, Streith IE, Schoiswohl G, Schweiger M, Kumari M, Eichmann TO, Rechberger G, Koefeler

HC, Eder S, Schauer S, Theussl HC, Preiss-Landl K, Lass A, Zimmermann R, Hoefler G, Zechner R,

Haemmerle G.

J Biol Chem. 2010 Mar 5;285(10):7300-11. Epub 2009 Dec 18.

Adipose triglyceride lipase plays a key role in the supply of the working muscle with fatty acids.

Schoiswohl G, Schweiger M, Schreiber R, Gorkiewicz G, Preiss-Landl K, Taschler U, Zierler KA, Radner

FP, Eichmann TO, Kienesberger PC, Eder S, Lass A, Haemmerle G, Alsted TJ, Kiens B, Hoefler G,

Zechner R, Zimmermann R.

J Lipid Res. 2010 Mar;51(3):490-9. Epub 2009 Nov 25.

Neutral lipid storage disease: genetic disorders caused by mutations in adipose triglyceride

lipase/PNPLA2 or CGI-58/ABHD5.

Schweiger M, Lass A, Zimmermann R, Eichmann TO, Zechner R.

Am J Physiol Endocrinol Metab. 2009 Aug;297(2):E289-96. Epub 2009 Apr 28.

Appendix

Appendix

91

Abbreviations & Acronyms

2-AG 2-arachidonoyl glycerol

AA amino acid

AGPAT 1-acylglycerol-3-phosphate-O-acyltransferase

AMPK AMP-activated kinase

ASO antisense oligonucleotide

AT adipose tissue

ATGL adipose triglyceride lipase

BAT brown adipose tissue

BSA bovine serum albumin

BSSL bile-salt stimulated lipase

cAMP cyclic AMP

CD chow diet

CE cholesteryl ester

Ces carboxylesterase

CGI-58 comparative gene identification-58

CIP Cahn-Ingold-Prelog

CM cardiac muslce

CoA coenzyme A

CPT CDP-choline:1,2-diacylglycerol cholinephosphotransferases

DAG diacylglycerol

DAGL diacylglycerol lipase

DGAT diacylglycerol-o-acyltransferase

Appendix

92

DGK diacylglycerol kinase

EPT CDP-ethanolamine:1,2-diacylglycerol ethanolaminephosphotransferase

ER endoplasmic reticulum

ERK extracellular signal-regulated kinase

FA fatty acid

FAME fatty acid methylester

G0S2 G0/G1 switch gene 2

G3P glycerin 3-phosphate

GAPDH glycerinaldehyde 3-phosphate dehydrogenase

GC-FID gas chromatography with flame-ionization detector

GL gastric lipase

GLUT4 glucose transporter 4

GPAT glycerol-3-phosphate acyltransferase

GS2 gene sequence 2

HFD high fat diet

HSL hormone-sensitive lipase

IP3 inositol 1,4,5-triphosphate

IPITT intraperetoneal insulin tolerance test

IR insulin resistance

IRS insulin receptor substrate

LacZ β-galaktosidase

LD lipid droplet

LL lingual lipase

LPA lyso-phosphatidic acid

Appendix

93

LPAAT lyso-phosphatidic acid acyltransferase

LPL lipoprotein lipase

MAG monoacylglycerol

MBOAT membrane-bound O acyltransferases

MGAT monoacylglycerol acyltransferase

MGL monoglyceride lipase

mTOR mammalian target of rapamycin

MUFA mono-unsaturated fatty acid

PA phosphatidic acid

PAL pancreatic lipase

PAPase phosphatidic acid phosphatase

PBS phosphate-buffered saline

PC phosphatidyl choline

PDK 3-phosphoinositide-dependent protein kinase 1

PE phosphatidyl ethanolamine

PGC peroxisome proliferator-activated receptor gamma co-activator

PI3K phosphatidylinositide-3-kinase

PIP2 phosphatidylinositol 4,5-bisphosphate

PKA proteinkinase A

PKB/Akt protein kinase B

PKC, c/n/a protein kinase C, conventional/novel/atypical

PL glycerophospholipid

PLA phospholipase

PNPLA patatin-like phospholipase domain containing A

Appendix

94

PPAR peroxisome proliferator-activated receptor

PS phosphatidyl serine

PUFA poly-unsaturated fatty acid

RE retinylester

Rictor rapamycin-insensitive companion of mammalian target of rapamycin

SM skeletal muscle

sn stereospecific numbering

T2DM Type 2 diabetes mellitus

TAG triacylglycerol

TGH triacylglycerol hydrolase

TLC thin layer chromatography

WAT white adipose tissue

wt wild type

Bibliography

Bibliography

96

1. Guilherme, A., Virbasius, J.V., Puri, V., and Czech, M.P. (2008) Adipocyte dysfunctions linking

obesity to insulin resistance and type 2 diabetes. Nat.Rev.Mol.Cell Biol. 9, 367-377

2. Misra, A., Singhal, N., and Khurana, L. (2010) Obesity, the metabolic syndrome, and type 2 diabetes

in developing countries: role of dietary fats and oils. J.Am.Coll.Nutr. 29, 289S-301S

3. Zimmermann, R., Strauss, J.G., Haemmerle, G., Schoiswohl, G., Birner-Gruenberger, R., Riederer,

M., Lass, A., Neuberger, G., Eisenhaber, F., Hermetter, A., and Zechner, R. (2004) Fat mobilization in

adipose tissue is promoted by adipose triglyceride lipase. Science. 306, 1383-6

4. Villena, J.A., Roy, S., Sarkadi-Nagy, E., Kim, K.H., and Sul, H.S. (2004) Desnutrin, an adipocyte gene

encoding a novel patatin domain-containing protein, is induced by fasting and glucocorticoids:

ectopic expression of desnutrin increases triglyceride hydrolysis. J.Biol.Chem. 279, 47066-75

5. Jenkins, C.M., Mancuso, D.J., Yan, W., Sims, H.F., Gibson, B., and Gross, R.W. (2004) Identification,

cloning, expression, and purification of three novel human calcium-independent phospholipase A2

family members possessing triacylglycerol lipase and acylglycerol transacylase activities. J.Biol.Chem.

279, 48968-75

6. Schmid, R.D., and Verger, R. (1998) Lipases: Interfacial Enzymes with Attractive Applications.

Angew. Chem. Int. Ed. 37, 1608-1633

7. Jensen, M.O., Jensen, T.R., Kjaer, K., Bjornholm, T., Mouritsen, O.G., and Peters, G.H. (2002)

Orientation and conformation of a lipase at an interface studied by molecular dynamics simulations.

Biophys.J. 83, 98-111

8. Holwerda, K., Verkade, P.E., and de Willigen, A.H.A. (1936) Vergleichende Untersuchungen über

die Verseifungsgeschwindigkeit einiger einsäuriger Triglyceride unter Einfluss von Pankreasextrakt.

Recl. Trav. Chim. Pays-Bas. 55, 43-57

9. Schonheyder, F., and Volqvartz, K. (1945) On the affinity of pig pancreatic lipase for tricaproin in

heterogeneous solution. Acta Physiol.Scand. 9, 57-67

10. Brady, L., Brzozowski, A.M., Derewenda, Z.S., Dodson, E., Dodson, G., Tolley, S., Turkenburg, J.P.,

Christiansen, L., Huge-Jensen, B., and Norskov, L. (1990) A serine protease triad forms the catalytic

centre of a triacylglycerol lipase. Nature. 343, 767-770

Bibliography

97

11. van Tilbeurgh, H., Egloff, M.P., Martinez, C., Rugani, N., Verger, R., and Cambillau, C. (1993)

Interfacial activation of the lipase-procolipase complex by mixed micelles revealed by X-ray

crystallography. Nature. 362, 814-820

12. SARDA, L., and DESNUELLE, P. (1958) Actions of pancreatic lipase on esters in emulsions.

Biochim.Biophys.Acta. 30, 513-521

13. Rogalska, E., Ransac, S., and Verger, R. (1993) Controlling lipase stereoselectivity via the surface

pressure. J.Biol.Chem. 268, 792-794

14. Rogalska, E., Cudrey, C., Ferrato, F., and Verger, R. (1993) Stereoselective hydrolysis of

triglycerides by animal and microbial lipases. Chirality. 5, 24-30

15. Faber, K. (2011) Biotrasnformations in Organic Chemistry

16. Prelog, V., and Helmchen, G. (1982) Basic principles of the CIP system and proposals for a

revision. Angewandte Chemie, International edition in english. 21, 567-583

17. Cahn, R.S., Ingold, C., and Prelog, V. (1966) Specifications of molecular chirality. Angewandte

Chemie, International edition in english. 5, 385-415

18. Tauchi-Sato, K., Ozeki, S., Houjou, T., Taguchi, R., and Fujimoto, T. (2002) The surface of lipid

droplets is a phospholipid monolayer with a unique Fatty Acid composition. J.Biol.Chem. 277, 44507-

44512

19. Murphy, D.J., and Vance, J. (1999) Mechanisms of lipid-body formation. Trends Biochem.Sci. 24,

109-115

20. Hollenberg, C.H., Raben, M.S., and Astwood, E.B. (1961) The lipolytic response to corticotropin.

Endocrinology. 68, 589-598

21. Vaughan, M., Berger, J.E., and Steinberg, D. (1964) Hormone-Sensitive Lipase and Monoglyceride

Lipase Activities in Adipose Tissue. J.Biol.Chem. 239, 401-409

22. Osterlund, T., Danielsson, B., Degerman, E., Contreras, J.A., Edgren, G., Davis, R.C., Schotz, M.C.,

and Holm, C. (1996) Domain-structure analysis of recombinant rat hormone-sensitive lipase.

Biochem.J. 319 ( Pt 2), 411-20

Bibliography

98

23. Fredrikson, G., Stralfors, P., Nilsson, N.O., and Belfrage, P. (1981) Hormone-sensitive lipase of rat

adipose tissue. Purification and some properties. J.Biol.Chem. 256, 6311-20

24. Wei, S., Lai, K., Patel, S., Piantedosi, R., Shen, H., Colantuoni, V., Kraemer, F.B., and Blaner, W.S.

(1997) Retinyl ester hydrolysis and retinol efflux from BFC-1beta adipocytes. J.Biol.Chem. 272, 14159-

14165

25. Tsujita, T., Ninomiya, H., and Okuda, H. (1989) P-Nitrophenyl Butyrate Hydrolyzing Activity of

Hormone-Sensitive Lipase from Bovine Adipose Tissue. J.Lipid Res. 30, 997-1004

26. Fredrikson, G., and Belfrage, P. (1983) Positional specificity of hormone-sensitive lipase from rat

adipose tissue. J.Biol.Chem. 258, 14253-6

27. Rodriguez, J.A., Ben Ali, Y., Abdelkafi, S., Mendoza, L.D., Leclaire, J., Fotiadu, F., Buono, G.,

Carriere, F., and Abousalham, A. (2010) In vitro stereoselective hydrolysis of diacylglycerols by

hormone-sensitive lipase. Biochim.Biophys.Acta. 1801, 77-83

28. Gavino, V.C., and Gavino, G.R. (1992) Adipose hormone-sensitive lipase preferentially releases

polyunsaturated fatty acids from triglycerides. Lipids. 27, 950-954

29. Holm, C., Osterlund, T., Laurell, H., and Contreras, J.A. (2000) Molecular mechanisms regulating

hormone-sensitive lipase and lipolysis. Annu.Rev.Nutr. 20, 365-393

30. Osterlund, T., Contreras, J.A., and Holm, C. (1997) Identification of essential aspartic acid and

histidine residues of hormone-sensitive lipase: apparent residues of the catalytic triad. FEBS Lett.

403, 259-62

31. Osterlund, T., Beussman, D.J., Julenius, K., Poon, P.H., Linse, S., Shabanowitz, J., Hunt, D.F.,

Schotz, M.C., Derewenda, Z.S., and Holm, C. (1999) Domain identification of hormone-sensitive lipase

by circular dichroism and fluorescence spectroscopy, limited proteolysis, and mass spectrometry.

J.Biol.Chem. 274, 15382-8

32. Langin, D., Laurell, H., Holst, L.S., Belfrage, P., and Holm, C. (1993) Gene organization and primary

structure of human hormone-sensitive lipase: possible significance of a sequence homology with a

lipase of Moraxella TA144, an antarctic bacterium. Proc.Natl.Acad.Sci.U.S.A. 90, 4897-901

33. Shen, W.J., Patel, S., Hong, R., and Kraemer, F.B. (2000) Hormone-sensitive lipase functions as an

oligomer. Biochemistry (N.Y.). 39, 2392-8

Bibliography

99

34. Stralfors, P., and Belfrage, P. (1983) Phosphorylation of hormone-sensitive lipase by cyclic AMP-

dependent protein kinase. J.Biol.Chem. 258, 15146-52

35. Shen, W.J., Patel, S., Natu, V., and Kraemer, F.B. (1998) Mutational analysis of structural features

of rat hormone-sensitive lipase. Biochemistry (N.Y.). 37, 8973-9

36. Anthonsen, M.W., Ronnstrand, L., Wernstedt, C., Degerman, E., and Holm, C. (1998) Identification

of novel phosphorylation sites in hormone-sensitive lipase that are phosphorylated in response to

isoproterenol and govern activation properties in vitro. J.Biol.Chem. 273, 215-21

37. Belfrage, P., Fredrikson, G., Olsson, H., and Stralfors, P. (1984) Regulation of adipose tissue

lipolysis through reversible phosphorylation of hormone-sensitive lipase. Adv.Cyclic Nucleotide

Protein Phosphorylation Res. 17, 351-359

38. Belfrage, P., Fredrikson, G., Nilsson, N.O., and Stralfors, P. (1981) Regulation of adipose-tissue

lipolysis by phosphorylation of hormone-sensitive lipase. Int.J.Obes. 5, 635-641

39. Stralfors, P., Bjorgell, P., and Belfrage, P. (1984) Hormonal regulation of hormone-sensitive lipase

in intact adipocytes: identification of phosphorylated sites and effects on the phosphorylation by

lipolytic hormones and insulin. Proc.Natl.Acad.Sci.U.S.A. 81, 3317-21

40. Garton, A.J., and Yeaman, S.J. (1990) Identification and role of the basal phosphorylation site on

hormone-sensitive lipase. Eur.J.Biochem. 191, 245-250

41. Lass, A., Zimmermann, R., Oberer, M., and Zechner, R. (2011) Lipolysis - a highly regulated multi-

enzyme complex mediates the catabolism of cellular fat stores. Prog.Lipid Res. 50, 14-27

42. Stralfors, P., Bjorgell, P., and Belfrage, P. (1984) Hormonal regulation of hormone-sensitive lipase

in intact adipocytes: identification of phosphorylated sites and effects on the phosphorylation by

lipolytic hormones and insulin. Proc.Natl.Acad.Sci.U.S.A. 81, 3317-3321

43. Greenberg, A.S., Shen, W.J., Muliro, K., Patel, S., Souza, S.C., Roth, R.A., and Kraemer, F.B. (2001)

Stimulation of lipolysis and hormone-sensitive lipase via the extracellular signal-regulated kinase

pathway. J.Biol.Chem. 276, 45456-61

44. Garton, A.J., Campbell, D.G., Carling, D., Hardie, D.G., Colbran, R.J., and Yeaman, S.J. (1989)

Phosphorylation of bovine hormone-sensitive lipase by the AMP-activated protein kinase. A possible

antilipolytic mechanism. Eur.J.Biochem. 179, 249-54

Bibliography

100

45. Blanchette-Mackie, E.J., Dwyer, N.K., Barber, T., Coxey, R.A., Takeda, T., Rondinone, C.M.,

Theodorakis, J.L., Greenberg, A.S., and Londos, C. (1995) Perilipin is located on the surface layer of

intracellular lipid droplets in adipocytes. J.Lipid Res. 36, 1211-1226

46. Brasaemle, D.L., Rubin, B., Harten, I.A., Gruia-Gray, J., Kimmel, A.R., and Londos, C. (2000)

Perilipin A increases triacylglycerol storage by decreasing the rate of triacylglycerol hydrolysis.

J.Biol.Chem. 275, 38486-93

47. Miyoshi, H., Perfield, J.W.2., Souza, S.C., Shen, W.J., Zhang, H.H., Stancheva, Z.S., Kraemer, F.B.,

Obin, M.S., and Greenberg, A.S. (2007) Control of adipose triglyceride lipase action by serine 517 of

perilipin A globally regulates protein kinase A-stimulated lipolysis in adipocytes. J.Biol.Chem. 282,

996-1002

48. Miyoshi, H., Souza, S.C., Zhang, H.H., Strissel, K.J., Christoffolete, M.A., Kovsan, J., Rudich, A.,

Kraemer, F.B., Bianco, A.C., Obin, M.S., and Greenberg, A.S. (2006) Perilipin promotes HSL-mediated

adipocyte lipolysis via phosphorylation-dependent and independent mechanisms. J.Biol.Chem.

49. Tansey, J.T., Huml, A.M., Vogt, R., Davis, K.E., Jones, J.M., Fraser, K.A., Brasaemle, D.L., Kimmel,

A.R., and Londos, C. (2003) Functional studies on native and mutated forms of perilipins. A role in

protein kinase A-mediated lipolysis of triacylglycerols. J.Biol.Chem. 278, 8401-6

50. Sztalryd, C., Xu, G., Dorward, H., Tansey, J.T., Contreras, J.A., Kimmel, A.R., and Londos, C. (2003)

Perilipin A is essential for the translocation of hormone-sensitive lipase during lipolytic activation.

J.Cell.Biol. 161, 1093-103

51. Egan, J.J., Greenberg, A.S., Chang, M.K., Wek, S.A., Moos, M.C.J., and Londos, C. (1992)

Mechanism of hormone-stimulated lipolysis in adipocytes: translocation of hormone-sensitive lipase

to the lipid storage droplet. Proc.Natl.Acad.Sci.U.S.A. 89, 8537-41

52. Su, C.L., Sztalryd, C., Contreras, J.A., Holm, C., Kimmel, A.R., and Londos, C. (2003) Mutational

analysis of the hormone-sensitive lipase translocation reaction in adipocytes. J.Biol.Chem. 278,

43615-43619

53. Enoksson, S., Degerman, E., Hagstrom-Toft, E., Large, V., and Arner, P. (1998) Various

phosphodiesterase subtypes mediate the in vivo antilipolytic effect of insulin on adipose tissue and

skeletal muscle in man. Diabetologia. 41, 560-568

Bibliography

101

54. Kitamura, T., Kitamura, Y., Kuroda, S., Hino, Y., Ando, M., Kotani, K., Konishi, H., Matsuzaki, H.,

Kikkawa, U., Ogawa, W., and Kasuga, M. (1999) Insulin-induced phosphorylation and activation of

cyclic nucleotide phosphodiesterase 3B by the serine-threonine kinase Akt. Mol.Cell.Biol. 19, 6286-

6296

55. Stralfors, P., and Honnor, R.C. (1989) Insulin-induced dephosphorylation of hormone-sensitive

lipase. Correlation with lipolysis and cAMP-dependent protein kinase activity. Eur.J.Biochem. 182,

379-85

56. Osuga, J., Ishibashi, S., Oka, T., Yagyu, H., Tozawa, R., Fujimoto, A., Shionoiri, F., Yahagi, N.,

Kraemer, F.B., Tsutsumi, O., and Yamada, N. (2000) Targeted disruption of hormone-sensitive lipase

results in male sterility and adipocyte hypertrophy, but not in obesity. Proc.Natl.Acad.Sci.U.S.A. 97,

787-92

57. Harada, K., Shen, W.J., Patel, S., Natu, V., Wang, J., Osuga, J., Ishibashi, S., and Kraemer, F.B.

(2003) Resistance to high-fat diet-induced obesity and altered expression of adipose-specific genes in

HSL-deficient mice. Am.J.Physiol.Endocrinol.Metab. 285, E1182-95

58. Zimmermann, R., Haemmerle, G., Wagner, E.M., Strauss, J.G., Kratky, D., and Zechner, R. (2003)

Decreased fatty acid esterification compensates for the reduced lipolytic activity in hormone-

sensitive lipase-deficient white adipose tissue. J.Lipid Res. 44, 2089-99

59. Haemmerle, G., Zimmermann, R., Hayn, M., Theussl, C., Waeg, G., Wagner, E., Sattler, W., Magin,

T.M., Wagner, E.F., and Zechner, R. (2002) Hormone-sensitive lipase deficiency in mice causes

diglyceride accumulation in adipose tissue, muscle, and testis. J.Biol.Chem. 277, 4806-15

60. Wang, S.P., Laurin, N., Himms-Hagen, J., Rudnicki, M.A., Levy, E., Robert, M.F., Pan, L., Oligny, L.,

and Mitchell, G.A. (2001) The adipose tissue phenotype of hormone-sensitive lipase deficiency in

mice. Obes.Res. 9, 119-28

61. Wilson, P.A., Gardner, S.D., Lambie, N.M., Commans, S.A., and Crowther, D.J. (2006)

Characterization of the human patatin-like phospholipase family. J.Lipid Res. 47, 1940-9

62. Rydel, T.J., Williams, J.M., Krieger, E., Moshiri, F., Stallings, W.C., Brown, S.M., Pershing, J.C.,

Purcell, J.P., and Alibhai, M.F. (2003) The crystal structure, mutagenesis, and activity studies reveal

that patatin is a lipid acyl hydrolase with a Ser-Asp catalytic dyad. Biochemistry (N.Y.). 42, 6696-708

Bibliography

102

63. Senda, K., Yoshioka, H., Doke, N., and Kawakita, K. (1996) A cytosolic phospholipase A2 from

potato tissues appears to be patatin. Plant Cell Physiol. 37, 347-53

64. Strickland, J.A., Orr, G.L., and Walsh, T.A. (1995) Inhibition of Diabrotica Larval Growth by Patatin,

the Lipid Acyl Hydrolase from Potato Tubers. Plant Physiol. 109, 667-674

65. Kienesberger, P.C., Oberer, M., Lass, A., and Zechner, R. (2008) Mammalian patatin domain

containing proteins: a family with diverse lipolytic activities involved in multiple biological functions.

J.Lipid Res.

66. Lake, A.C., Sun, Y., Li, J.L., Kim, J.E., Johnson, J.W., Li, D., Revett, T., Shih, H.H., Liu, W., Paulsen,

J.E., and Gimeno, R.E. (2005) Expression, regulation, and triglyceride hydrolase activity of Adiponutrin

family members. J.Lipid Res. 46, 2477-87

67. Smirnova, E., Goldberg, E.B., Makarova, K.S., Lin, L., Brown, W.J., and Jackson, C.L. (2006) ATGL

has a key role in lipid droplet/adiposome degradation in mammalian cells. EMBO Rep. 7, 106-13

68. Lass, A., Zimmermann, R., Haemmerle, G., Riederer, M., Schoiswohl, G., Schweiger, M.,

Kienesberger, P., Strauss, J.G., Gorkiewicz, G., and Zechner, R. (2006) Adipose triglyceride lipase-

mediated lipolysis of cellular fat stores is activated by CGI-58 and defective in Chanarin-Dorfman

Syndrome. Cell Metab. 3, 309-19

69. Dessen, A., Tang, J., Schmidt, H., Stahl, M., Clark, J.D., Seehra, J., and Somers, W.S. (1999) Crystal

structure of human cytosolic phospholipase A2 reveals a novel topology and catalytic mechanism.

Cell. 97, 349-60

70. Schweiger, M., Schoiswohl, G., Lass, A., Radner, F.P., Haemmerle, G., Malli, R., Graier, W.,

Cornaciu, I., Oberer, M., Salvayre, R., Fischer, J., Zechner, R., and Zimmermann, R. (2008) The C-

terminal region of human adipose triglyceride lipase affects enzyme activity and lipid droplet binding.

J.Biol.Chem. 283, 17211-17220

71. Duncan, R.E., Wang, Y., Ahmadian, M., Lu, J., Sarkadi-Nagy, E., and Sul, H.S. (2010)

Characterization of desnutrin functional domains: critical residues for triacylglycerol hydrolysis in

cultured cells. J.Lipid Res. 51, 309-317

72. Bartz, R., Zehmer, J.K., Zhu, M., Chen, Y., Serrero, G., Zhao, Y., and Liu, P. (2007) Dynamic activity

of lipid droplets: protein phosphorylation and GTP-mediated protein translocation. J Proteome Res.

6, 3256-65

Bibliography

103

73. Ahmadian, M., Abbott, M.J., Tang, T., Hudak, C.S., Kim, Y., Bruss, M., Hellerstein, M.K., Lee, H.Y.,

Samuel, V.T., Shulman, G.I., Wang, Y., Duncan, R.E., Kang, C., and Sul, H.S. (2011) Desnutrin/ATGL Is

Regulated by AMPK and Is Required for a Brown Adipose Phenotype. Cell Metab. 13, 739-748

74. Gao, J.G., and Simon, M. (2007) A comparative study of human GS2, its paralogues, and its rat

orthologue. Biochem.Biophys.Res.Commun. 360, 501-6

75. Haemmerle, G., Lass, A., Zimmermann, R., Gorkiewicz, G., Meyer, C., Rozman, J., Heldmaier, G.,

Maier, R., Theussl, C., Eder, S., Kratky, D., Wagner, E.F., Klingenspor, M., Hoefler, G., and Zechner, R.

(2006) Defective lipolysis and altered energy metabolism in mice lacking adipose triglyceride lipase.

Science. 312, 734-7

76. Kershaw, E.E., Schupp, M., Guan, H.P., Gardner, N.P., Lazar, M.A., and Flier, J.S. (2007) PPAR

gamma regulates adipose triglyceride lipase in adipocytes in vitro and in vivo.

Am.J.Physiol.Endocrinol.Metab.

77. Liu, L.F., Purushotham, A., Wendel, A.A., Koba, K., Deiuliis, J., Lee, K., and Belury, M.A. (2008)

Regulation of adipose triglyceride lipase by rosiglitazone. Diabetes Obes.Metab.

78. Festuccia, W.T., Laplante, M., Berthiaume, M., Gelinas, Y., and Deshaies, Y. (2006) PPARgamma

agonism increases rat adipose tissue lipolysis, expression of glyceride lipases, and the response of

lipolysis to hormonal control. Diabetologia. 49, 2427-36

79. Kim, J.Y., Tillison, K., Lee, J.H., Rearick, D.A., and Smas, C.M. (2006) The adipose tissue triglyceride

lipase ATGL/PNPLA2 is downregulated by insulin and TNF-alpha in 3T3-L1 adipocytes and is a target

for transactivation by PPARgamma. Am.J.Physiol.Endocrinol.Metab. 291, E115-27

80. Kralisch, S., Klein, J., Lossner, U., Bluher, M., Paschke, R., Stumvoll, M., and Fasshauer, M. (2005)

Isoproterenol, TNFalpha, and insulin downregulate adipose triglyceride lipase in 3T3-L1 adipocytes.

Mol.Cell.Endocrinol. 240, 43-9

81. Hotamisligil, G.S., Shargill, N.S., and Spiegelman, B.M. (1993) Adipose expression of tumor

necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 259, 87-91

82. Jocken, J.W., Langin, D., Smit, E., Saris, W.H., Valle, C., Hul, G.B., Holm, C., Arner, P., and Blaak,

E.E. (2007) Adipose TriGlyceride Lipase (ATGL) and Hormone-Sensitive Lipase (HSL) protein

expression is decreased in the obese insulin resistant state. J.Clin.Endocrinol.Metab.

Bibliography

104

83. Pagnon, J., Matzaris, M., Stark, R., Meex, R.C., Lance Macaulay, S., Brown, W., O'Brien, P.E.,

Tiganis, T., and Watt, M.J. (2012) Identification and Functional Characterization of Protein Kinase A

Phosphorylation Sites in the Major Lipolytic Protein, Adipose Triglyceride Lipase. Endocrinology.

84. Daval, M., Diot-Dupuy, F., Bazin, R., Hainault, I., Viollet, B., Vaulont, S., Hajduch, E., Ferre, P., and

Foufelle, F. (2005) Anti-lipolytic action of AMP-activated protein kinase in rodent adipocytes.

J.Biol.Chem. 280, 25250-25257

85. Schweiger, M., Schreiber, R., Haemmerle, G., Lass, A., Fledelius, C., Jacobsen, P., Tornqvist, H.,

Zechner, R., and Zimmermann, R. (2006) Adipose triglyceride lipase and hormone-sensitive lipase are

the major enzymes in adipose tissue triacylglycerol catabolism. J.Biol.Chem. 281, 40236-41

86. Lefevre, C., Jobard, F., Caux, F., Bouadjar, B., Karaduman, A., Heilig, R., Lakhdar, H., Wollenberg,

A., Verret, J.L., Weissenbach, J., Ozguc, M., Lathrop, M., Prud'homme, J.F., and Fischer, J. (2001)

Mutations in CGI-58, the gene encoding a new protein of the esterase/lipase/thioesterase subfamily,

in Chanarin-Dorfman syndrome. Am.J.Hum.Genet. 69, 1002-12

87. Ghosh, A.K., Ramakrishnan, G., Chandramohan, C., and Rajasekharan, R. (2008) CGI-58, the

causative gene for Chanarin-Dorfman syndrome, mediates acylation of lysophosphatidic acid.

J.Biol.Chem. 283, 24525-33

88. Montero-Moran, G., Caviglia, J.M., McMahon, D., Rothenberg, A., Subramanian, V., Xu, Z., Lara-

Gonzalez, S., Storch, J., Carman, G.M., and Brasaemle, D.L. (2010) CGI-58/ABHD5 is a coenzyme A-

dependent lysophosphatidic acid acyltransferase. J.Lipid Res. 51, 709-719

89. Kobayashi, K., Inoguchi, T., Maeda, Y., Nakashima, N., Kuwano, A., Eto, E., Ueno, N., Sasaki, S.,

Sawada, F., Fujii, M., Matoba, Y., Sumiyoshi, S., Kawate, H., and Takayanagi, R. (2008) The lack of the

C-terminal domain of adipose triglyceride lipase causes neutral lipid storage disease through

impaired interactions with lipid droplets. J.Clin.Endocrinol.Metab. 93, 2877-84

90. Gruber, A., Cornaciu, I., Lass, A., Schweiger, M., Poeschl, M., Eder, C., Kumari, M., Schoiswohl, G.,

Wolinski, H., Kohlwein, S.D., Zechner, R., Zimmermann, R., and Oberer, M. (2010) The N-terminal

region of comparative gene identification-58 (CGI-58) is important for lipid droplet binding and

activation of adipose triglyceride lipase. J.Biol.Chem. 285, 12289-12298

91. Yang, X., Lu, X., Lombes, M., Rha, G.B., Chi, Y.I., Guerin, T.M., Smart, E.J., and Liu, J. (2010) The

G(0)/G(1) switch gene 2 regulates adipose lipolysis through association with adipose triglyceride

lipase. Cell Metab. 11, 194-205.

Bibliography

105

92. Lu, X., Yang, X., and Liu, J. (2010) Differential control of ATGL-mediated lipid droplet degradation

by CGI-58 and G0S2. cell cycle. 9, 2719-2725

93. Tornqvist, H., and Belfrage, P. (1976) Purification and some properties of a monoacylglycerol-

hydrolyzing enzyme of rat adipose tissue. J.Biol.Chem. 251, 813-819

94. Karlsson, M., Contreras, J.A., Hellman, U., Tornqvist, H., and Holm, C. (1997) cDNA cloning, tissue

distribution, and identification of the catalytic triad of monoglyceride lipase. Evolutionary

relationship to esterases, lysophospholipases, and haloperoxidases. J.Biol.Chem. 272, 27218-27223

95. Rhee, S.G. (2001) Regulation of phosphoinositide-specific phospholipase C. Annu.Rev.Biochem.

70, 281-312

96. Cockcroft, S. (2006) The latest phospholipase C, PLCeta, is implicated in neuronal function. Trends

Biochem.Sci. 31, 4-7

97. Bunney, T.D., and Katan, M. (2006) Phospholipase C epsilon: linking second messengers and small

GTPases. Trends Cell Biol. 16, 640-648

98. Suh, P.G., Park, J.I., Manzoli, L., Cocco, L., Peak, J.C., Katan, M., Fukami, K., Kataoka, T., Yun, S.,

and Ryu, S.H. (2008) Multiple roles of phosphoinositide-specific phospholipase C isozymes. BMB Rep.

41, 415-434

99. Fukami, K., Inanobe, S., Kanemaru, K., and Nakamura, Y. (2010) Phospholipase C is a key enzyme

regulating intracellular calcium and modulating the phosphoinositide balance. Prog.Lipid Res. 49,

429-437

100. Mechoulam, R., Ben-Shabat, S., Hanus, L., Ligumsky, M., Kaminski, N.E., Schatz, A.R., Gopher, A.,

Almog, S., Martin, B.R., and Compton, D.R. (1995) Identification of an endogenous 2-monoglyceride,

present in canine gut, that binds to cannabinoid receptors. Biochem.Pharmacol. 50, 83-90

101. Sugiura, T., Kondo, S., Sukagawa, A., Nakane, S., Shinoda, A., Itoh, K., Yamashita, A., and Waku,

K. (1995) 2-Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain.

Biochem.Biophys.Res.Commun. 215, 89-97

102. Bell, R.M., and Coleman, R.A. (1980) Enzymes of glycerolipid synthesis in eukaryotes.

Annu.Rev.Biochem. 49, 459-487

Bibliography

106

103. Coleman, R.A., and Lee, D.P. (2004) Enzymes of triacylglycerol synthesis and their regulation.

Prog.Lipid Res. 43, 134-176

104. Coleman, R.A., and Mashek, D.G. (2011) Mammalian triacylglycerol metabolism: synthesis,

lipolysis, and signaling. Chem.Rev. 111, 6359-6386

105. Phan, C.T., and Tso, P. (2001) Intestinal lipid absorption and transport. Front.Biosci. 6, D299-319

106. Polheim, D., David, J.S., Schultz, F.M., Wylie, M.B., and Johnston, J.M. (1973) Regulation of

triglyceride biosynthesis in adipose and intestinal tissue. J.Lipid Res. 14, 415-421

107. Kennedy, E.P., Smith, S.W., and Weiss, S.B. (1956) New synthesis of lecithin in an isolated

enzyme system. Nature. 178, 594-595

108. Kennedy, E.P. (1957) Biosynthesis of phospholipides. Fed.Proc. 16, 847-853

109. Gibellini, F., and Smith, T.K. (2010) The Kennedy pathway--De novo synthesis of

phosphatidylethanolamine and phosphatidylcholine. IUBMB Life. 62, 414-428

110. Peterfy, M., Phan, J., Xu, P., and Reue, K. (2001) Lipodystrophy in the fld mouse results from

mutation of a new gene encoding a nuclear protein, lipin. Nat.Genet. 27, 121-124

111. Donkor, J., Sariahmetoglu, M., Dewald, J., Brindley, D.N., and Reue, K. (2007) Three mammalian

lipins act as phosphatidate phosphatases with distinct tissue expression patterns. J.Biol.Chem. 282,

3450-3457

112. Harris, T.E., Huffman, T.A., Chi, A., Shabanowitz, J., Hunt, D.F., Kumar, A., and Lawrence, J.C.,Jr

(2007) Insulin controls subcellular localization and multisite phosphorylation of the phosphatidic acid

phosphatase, lipin 1. J.Biol.Chem. 282, 277-286

113. Cascales, C., Mangiapane, E.H., and Brindley, D.N. (1984) Oleic acid promotes the activation and

translocation of phosphatidate phosphohydrolase from the cytosol to particulate fractions of isolated

rat hepatocytes. Biochem.J. 219, 911-916

114. Phan, J., and Reue, K. (2005) Lipin, a lipodystrophy and obesity gene. Cell Metab. 1, 73-83

115. Yen, C.L., and Farese, R.V.,Jr (2003) MGAT2, a monoacylglycerol acyltransferase expressed in the

small intestine. J.Biol.Chem. 278, 18532-18537

Bibliography

107

116. Yen, C.L., Stone, S.J., Cases, S., Zhou, P., and Farese, R.V.,Jr (2002) Identification of a gene

encoding MGAT1, a monoacylglycerol acyltransferase. Proc.Natl.Acad.Sci.U.S.A. 99, 8512-8517

117. Cheng, D., Nelson, T.C., Chen, J., Walker, S.G., Wardwell-Swanson, J., Meegalla, R., Taub, R.,

Billheimer, J.T., Ramaker, M., and Feder, J.N. (2003) Identification of acyl coenzyme

A:monoacylglycerol acyltransferase 3, an intestinal specific enzyme implicated in dietary fat

absorption. J.Biol.Chem. 278, 13611-13614

118. Cases, S., Stone, S.J., Zhou, P., Yen, E., Tow, B., Lardizabal, K.D., Voelker, T., and Farese, R.V.,Jr

(2001) Cloning of DGAT2, a second mammalian diacylglycerol acyltransferase, and related family

members. J.Biol.Chem. 276, 38870-38876

119. Cao, J., Lockwood, J., Burn, P., and Shi, Y. (2003) Cloning and functional characterization of a

mouse intestinal acyl-CoA:monoacylglycerol acyltransferase, MGAT2. J.Biol.Chem. 278, 13860-13866

120. Cao, J., Burn, P., and Shi, Y. (2003) Properties of the mouse intestinal acyl-CoA:monoacylglycerol

acyltransferase, MGAT2. J.Biol.Chem. 278, 25657-25663

121. Cao, J., Cheng, L., and Shi, Y. (2007) Catalytic properties of MGAT3, a putative triacylgycerol

synthase. J.Lipid Res. 48, 583-591

122. Lykidis, A., Mougios, V., and Arzoglou, P. (1995) Kinetics of the two-step hydrolysis of

triacylglycerol by pancreatic lipases. Eur.J.Biochem. 230, 892-898

123. Brindley, D.N., and Hubscher, G. (1966) The effect of chain length on the activation and

subsequent incorporation of fatty acids into glycerides by the small intestinal mucosa.

Biochim.Biophys.Acta. 125, 92-105

124. Brindley, D.N., and Hubscher, G. (1965) The intracellular distribution of the enzymes catalysing

the biosynthesis of glycerides in the intestinal mucosa. Biochim.Biophys.Acta. 106, 495-509

125. Hofmann, K. (2000) A superfamily of membrane-bound O-acyltransferases with implications for

wnt signaling. Trends Biochem.Sci. 25, 111-112

126. Yen, C.L., Stone, S.J., Koliwad, S., Harris, C., and Farese, R.V.,Jr (2008) Thematic review series:

glycerolipids. DGAT enzymes and triacylglycerol biosynthesis. J.Lipid Res. 49, 2283-2301

Bibliography

108

127. McFie, P.J., Stone, S.L., Banman, S.L., and Stone, S.J. (2010) Topological orientation of acyl-

CoA:diacylglycerol acyltransferase-1 (DGAT1) and identification of a putative active site histidine and

the role of the n terminus in dimer/tetramer formation. J.Biol.Chem. 285, 37377-37387

128. Siloto, R.M., Madhavji, M., Wiehler, W.B., Burton, T.L., Boora, P.S., Laroche, A., and Weselake,

R.J. (2008) An N-terminal fragment of mouse DGAT1 binds different acyl-CoAs with varying affinity.

Biochem.Biophys.Res.Commun. 373, 350-354

129. Cheng, D., Meegalla, R.L., He, B., Cromley, D.A., Billheimer, J.T., and Young, P.R. (2001) Human

acyl-CoA:diacylglycerol acyltransferase is a tetrameric protein. Biochem.J. 359, 707-714

130. Abo-Hashema, K.A., Cake, M.H., Power, G.W., and Clarke, D. (1999) Evidence for triacylglycerol

synthesis in the lumen of microsomes via a lipolysis-esterification pathway involving carnitine

acyltransferases. J.Biol.Chem. 274, 35577-35582

131. Waterman, I.J., Price, N.T., and Zammit, V.A. (2002) Distinct ontogenic patterns of overt and

latent DGAT activities of rat liver microsomes. J.Lipid Res. 43, 1555-1562

132. Waterman, I.J., and Zammit, V.A. (2002) Activities of overt and latent diacylglycerol

acyltransferases (DGATs I and II) in liver microsomes of ob/ob mice. Int.J.Obes.Relat.Metab.Disord.

26, 742-743

133. Yen, C.L., Stone, S.J., Koliwad, S., Harris, C., and Farese, R.V.,Jr (2008) Thematic review series:

glycerolipids. DGAT enzymes and triacylglycerol biosynthesis. J.Lipid Res. 49, 2283-2301

134. Yen, C.L., Monetti, M., Burri, B.J., and Farese, R.V.,Jr (2005) The triacylglycerol synthesis enzyme

DGAT1 also catalyzes the synthesis of diacylglycerols, waxes, and retinyl esters. J.Lipid Res. 46, 1502-

1511

135. Lardizabal, K.D., Mai, J.T., Wagner, N.W., Wyrick, A., Voelker, T., and Hawkins, D.J. (2001) DGAT2

is a new diacylglycerol acyltransferase gene family: purification, cloning, and expression in insect cells

of two polypeptides from Mortierella ramanniana with diacylglycerol acyltransferase activity.

J.Biol.Chem. 276, 38862-38869

136. Stone, S.J., Levin, M.C., and Farese, R.V.,Jr (2006) Membrane topology and identification of key

functional amino acid residues of murine acyl-CoA:diacylglycerol acyltransferase-2. J.Biol.Chem. 281,

40273-40282

Bibliography

109

137. Alam, M., Gilham, D., Vance, D.E., and Lehner, R. (2006) Mutation of F417 but not of L418 or

L420 in the lipid binding domain decreases the activity of triacylglycerol hydrolase. J.Lipid Res. 47,

375-383

138. Au-Young, J., and Fielding, C.J. (1992) Synthesis and secretion of wild-type and mutant human

plasma cholesteryl ester transfer protein in baculovirus-transfected insect cells: the carboxyl-

terminal region is required for both lipoprotein binding and catalysis of transfer.

Proc.Natl.Acad.Sci.U.S.A. 89, 4094-4098

139. Stone, S.J., Levin, M.C., Zhou, P., Han, J., Walther, T.C., and Farese, R.V.,Jr (2009) The

endoplasmic reticulum enzyme DGAT2 is found in mitochondria-associated membranes and has a

mitochondrial targeting signal that promotes its association with mitochondria. J.Biol.Chem. 284,

5352-5361

140. Stone, S.J., Myers, H.M., Watkins, S.M., Brown, B.E., Feingold, K.R., Elias, P.M., and Farese,

R.V.,Jr (2004) Lipopenia and skin barrier abnormalities in DGAT2-deficient mice. J.Biol.Chem. 279,

11767-11776

141. Hiramine, Y., and Tanabe, T. (2011) Characterization of acyl-coenzyme A:diacylglycerol

acyltransferase (DGAT) enzyme of human small intestine. J.Physiol.Biochem. 67, 259-264

142. Smith, S.J., Cases, S., Jensen, D.R., Chen, H.C., Sande, E., Tow, B., Sanan, D.A., Raber, J., Eckel,

R.H., and Farese, R.V.,Jr (2000) Obesity resistance and multiple mechanisms of triglyceride synthesis

in mice lacking Dgat. Nat.Genet. 25, 87-90

143. Wurie, H.R., Buckett, L., and Zammit, V.A. (2012) Diacylglycerol acyltransferase 2 (DGAT2) acts

upstream of DGAT1, and utilises nascent diglycerides and de novo synthesised fatty acids in HepG2

cells. FEBS J.

144. Qi, J., Lang, W., Geisler, J.G., Wang, P., Petrounia, I., Mai, S., Smith, C., Askari, H., Struble, G.T.,

Williams, R., Bhanot, S., Monia, B.P., Bayoumy, S., Grant, E., Caldwell, G.W., Todd, M.J., Liang, Y.,

Gaul, M.D., Demarest, K.T., and Connelly, M.A. (2012) The use of stable isotope-labeled glycerol and

oleic acid to differentiate the hepatic functions of DGAT1 and -2. J.Lipid Res. 53, 1106-1116

145. Bisogno, T., Howell, F., Williams, G., Minassi, A., Cascio, M.G., Ligresti, A., Matias, I., Schiano-

Moriello, A., Paul, P., Williams, E.J., Gangadharan, U., Hobbs, C., Di Marzo, V., and Doherty, P. (2003)

Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid

signaling in the brain. J.Cell Biol. 163, 463-468

Bibliography

110

146. Gao, Y., Vasilyev, D.V., Goncalves, M.B., Howell, F.V., Hobbs, C., Reisenberg, M., Shen, R., Zhang,

M.Y., Strassle, B.W., Lu, P., Mark, L., Piesla, M.J., Deng, K., Kouranova, E.V., Ring, R.H., Whiteside,

G.T., Bates, B., Walsh, F.S., Williams, G., Pangalos, M.N., Samad, T.A., and Doherty, P. (2010) Loss of

retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-

out mice. J.Neurosci. 30, 2017-2024

147. Merida, I., Avila-Flores, A., and Merino, E. (2008) Diacylglycerol kinases: at the hub of cell

signalling. Biochem.J. 409, 1-18

148. Sakane, F., Imai, S., Kai, M., Yasuda, S., and Kanoh, H. (2007) Diacylglycerol kinases: why so many

of them?. Biochim.Biophys.Acta. 1771, 793-806

149. Topham, M.K., and Epand, R.M. (2009) Mammalian diacylglycerol kinases: molecular

interactions and biological functions of selected isoforms. Biochim.Biophys.Acta. 1790, 416-424

150. Topham, M.K., and Prescott, S.M. (1999) Mammalian diacylglycerol kinases, a family of lipid

kinases with signaling functions. J.Biol.Chem. 274, 11447-11450

151. Goto, K., Hozumi, Y., Nakano, T., Saino-Saito, S., and Martelli, A.M. (2008) Lipid messenger,

diacylglycerol, and its regulator, diacylglycerol kinase, in cells, organs, and animals: history and

perspective. Tohoku J.Exp.Med. 214, 199-212

152. Boni, L.T., and Rando, R.R. (1985) The nature of protein kinase C activation by physically defined

phospholipid vesicles and diacylglycerols. J.Biol.Chem. 260, 10819-10825

153. Rando, R.R., and Young, N. (1984) The stereospecific activation of protein kinase C.

Biochem.Biophys.Res.Commun. 122, 818-823

154. Nomura, H., Ase, K., Sekiguchi, K., Kikkawa, U., Nishizuka, Y., Nakano, Y., and Satoh, T. (1986)

Stereospecificity of diacylglycerol for stimulus-response coupling in platelets.

Biochem.Biophys.Res.Commun. 140, 1143-1151

155. Kobayashi, N., Hozumi, Y., Ito, T., Hosoya, T., Kondo, H., and Goto, K. (2007) Differential

subcellular targeting and activity-dependent subcellular localization of diacylglycerol kinase isozymes

in transfected cells. Eur.J.Cell Biol. 86, 433-444

Bibliography

111

156. Tang, W., Bunting, M., Zimmerman, G.A., McIntyre, T.M., and Prescott, S.M. (1996) Molecular

cloning of a novel human diacylglycerol kinase highly selective for arachidonate-containing

substrates. J.Biol.Chem. 271, 10237-10241

157. Prescott, S.M., and Majerus, P.W. (1981) The fatty acid composition of phosphatidylinositol

from thrombin-stimulated human platelets. J.Biol.Chem. 256, 579-582

158. Henneberry, A.L., and McMaster, C.R. (1999) Cloning and expression of a human

choline/ethanolaminephosphotransferase: synthesis of phosphatidylcholine and

phosphatidylethanolamine. Biochem.J. 339 ( Pt 2), 291-298

159. Henneberry, A.L., Wistow, G., and McMaster, C.R. (2000) Cloning, genomic organization, and

characterization of a human cholinephosphotransferase. J.Biol.Chem. 275, 29808-29815

160. Henneberry, A.L., Wright, M.M., and McMaster, C.R. (2002) The major sites of cellular

phospholipid synthesis and molecular determinants of Fatty Acid and lipid head group specificity.

Mol.Biol.Cell. 13, 3148-3161

161. McMaster, C.R., and Bell, R.M. (1997) CDP-choline:1,2-diacylglycerol

cholinephosphotransferase. Biochim.Biophys.Acta. 1348, 100-110

162. Wright, M.M., Henneberry, A.L., Lagace, T.A., Ridgway, N.D., and McMaster, C.R. (2001)

Uncoupling farnesol-induced apoptosis from its inhibition of phosphatidylcholine synthesis.

J.Biol.Chem. 276, 25254-25261

163. Sriburi, R., Bommiasamy, H., Buldak, G.L., Robbins, G.R., Frank, M., Jackowski, S., and Brewer,

J.W. (2007) Coordinate regulation of phospholipid biosynthesis and secretory pathway gene

expression in XBP-1(S)-induced endoplasmic reticulum biogenesis. J.Biol.Chem. 282, 7024-7034

164. Hjelmstad, R.H., Morash, S.C., McMaster, C.R., and Bell, R.M. (1994) Chimeric enzymes.

Structure-function analysis of segments of sn-1,2-diacylglycerol choline- and

ethanolaminephosphotransferases. J.Biol.Chem. 269, 20995-21002

165. McMaster, C.R., and Bell, R.M. (1997) CDP-ethanolamine:1,2-diacylglycerol

ethanolaminephosphotransferase. Biochim.Biophys.Acta. 1348, 117-123

Bibliography

112

166. Samborski, R.W., Ridgway, N.D., and Vance, D.E. (1990) Evidence that only newly made

phosphatidylethanolamine is methylated to phosphatidylcholine and that phosphatidylethanolamine

is not significantly deacylated-reacylated in rat hepatocytes. J.Biol.Chem. 265, 18322-18329

167. Randle, P.J., Garland, P.B., Hales, C.N., and Newsholme, E.A. (1963) The glucose fatty-acid cycle.

Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet. 1, 785-789

168. Ciaraldi, T.P., Abrams, L., Nikoulina, S., Mudaliar, S., and Henry, R.R. (1995) Glucose transport in

cultured human skeletal muscle cells. Regulation by insulin and glucose in nondiabetic and non-

insulin-dependent diabetes mellitus subjects. J.Clin.Invest. 96, 2820-2827

169. Garvey, W.T., Maianu, L., Zhu, J.H., Brechtel-Hook, G., Wallace, P., and Baron, A.D. (1998)

Evidence for defects in the trafficking and translocation of GLUT4 glucose transporters in skeletal

muscle as a cause of human insulin resistance. J.Clin.Invest. 101, 2377-2386

170. Litherland, G.J., Hajduch, E., and Hundal, H.S. (2001) Intracellular signalling mechanisms

regulating glucose transport in insulin-sensitive tissues (review). Mol.Membr.Biol. 18, 195-204

171. Pearce, L.R., Komander, D., and Alessi, D.R. (2010) The nuts and bolts of AGC protein kinases.

Nat.Rev.Mol.Cell Biol. 11, 9-22

172. Hajduch, E., Litherland, G.J., and Hundal, H.S. (2001) Protein kinase B (PKB/Akt)--a key regulator

of glucose transport?. FEBS Lett. 492, 199-203

173. Hajduch, E., Alessi, D.R., Hemmings, B.A., and Hundal, H.S. (1998) Constitutive activation of

protein kinase B alpha by membrane targeting promotes glucose and system A amino acid transport,

protein synthesis, and inactivation of glycogen synthase kinase 3 in L6 muscle cells. Diabetes. 47,

1006-1013

174. Sano, H., Kane, S., Sano, E., Miinea, C.P., Asara, J.M., Lane, W.S., Garner, C.W., and Lienhard,

G.E. (2003) Insulin-stimulated phosphorylation of a Rab GTPase-activating protein regulates GLUT4

translocation. J.Biol.Chem. 278, 14599-14602

175. Holland, W.L., and Summers, S.A. (2008) Sphingolipids, insulin resistance, and metabolic disease:

new insights from in vivo manipulation of sphingolipid metabolism. Endocr.Rev. 29, 381-402

176. Steinberg, S.F. (2008) Structural basis of protein kinase C isoform function. Physiol.Rev. 88,

1341-1378

Bibliography

113

177. Samuel, V.T., Petersen, K.F., and Shulman, G.I. (2010) Lipid-induced insulin resistance:

unravelling the mechanism. Lancet. 375, 2267-2277

178. Avignon, A., Yamada, K., Zhou, X., Spencer, B., Cardona, O., Saba-Siddique, S., Galloway, L.,

Standaert, M.L., and Farese, R.V. (1996) Chronic activation of protein kinase C in soleus muscles and

other tissues of insulin-resistant type II diabetic Goto-Kakizaki (GK), obese/aged, and obese/Zucker

rats. A mechanism for inhibiting glycogen synthesis. Diabetes. 45, 1396-1404

179. Considine, R.V., Nyce, M.R., Allen, L.E., Morales, L.M., Triester, S., Serrano, J., Colberg, J., Lanza-

Jacoby, S., and Caro, J.F. (1995) Protein kinase C is increased in the liver of humans and rats with non-

insulin-dependent diabetes mellitus: an alteration not due to hyperglycemia. J.Clin.Invest. 95, 2938-

2944

180. Haring, H., Kirsch, D., Obermaier, B., Ermel, B., and Machicao, F. (1986) Tumor-promoting

phorbol esters increase the Km of the ATP-binding site of the insulin receptor kinase from rat

adipocytes. J.Biol.Chem. 261, 3869-3875

181. Jacobs, S., Sahyoun, N.E., Saltiel, A.R., and Cuatrecasas, P. (1983) Phorbol esters stimulate the

phosphorylation of receptors for insulin and somatomedin C. Proc.Natl.Acad.Sci.U.S.A. 80, 6211-6213

182. Karasik, A., Rothenberg, P.L., Yamada, K., White, M.F., and Kahn, C.R. (1990) Increased protein

kinase C activity is linked to reduced insulin receptor autophosphorylation in liver of starved rats.

J.Biol.Chem. 265, 10226-10231

183. Griffin, M.E., Marcucci, M.J., Cline, G.W., Bell, K., Barucci, N., Lee, D., Goodyear, L.J., Kraegen,

E.W., White, M.F., and Shulman, G.I. (1999) Free fatty acid-induced insulin resistance is associated

with activation of protein kinase C theta and alterations in the insulin signaling cascade. Diabetes. 48,

1270-1274

184. Yu, C., Chen, Y., Cline, G.W., Zhang, D., Zong, H., Wang, Y., Bergeron, R., Kim, J.K., Cushman,

S.W., Cooney, G.J., Atcheson, B., White, M.F., Kraegen, E.W., and Shulman, G.I. (2002) Mechanism by

which fatty acids inhibit insulin activation of insulin receptor substrate-1 (IRS-1)-associated

phosphatidylinositol 3-kinase activity in muscle. J.Biol.Chem. 277, 50230-50236

185. Kim, J.K., Fillmore, J.J., Sunshine, M.J., Albrecht, B., Higashimori, T., Kim, D.W., Liu, Z.X., Soos,

T.J., Cline, G.W., O'Brien, W.R., Littman, D.R., and Shulman, G.I. (2004) PKC-theta knockout mice are

protected from fat-induced insulin resistance. J.Clin.Invest. 114, 823-827

Bibliography

114

186. Samuel, V.T., Liu, Z.X., Wang, A., Beddow, S.A., Geisler, J.G., Kahn, M., Zhang, X.M., Monia, B.P.,

Bhanot, S., and Shulman, G.I. (2007) Inhibition of protein kinase Cepsilon prevents hepatic insulin

resistance in nonalcoholic fatty liver disease. J.Clin.Invest. 117, 739-745

187. Raddatz, K., Turner, N., Frangioudakis, G., Liao, B.M., Pedersen, D.J., Cantley, J., Wilks, D.,

Preston, E., Hegarty, B.D., Leitges, M., Raftery, M.J., Biden, T.J., and Schmitz-Peiffer, C. (2011) Time-

dependent effects of Prkce deletion on glucose homeostasis and hepatic lipid metabolism on dietary

lipid oversupply in mice. Diabetologia. 54, 1447-1456

188. Li, Y., Soos, T.J., Li, X., Wu, J., Degennaro, M., Sun, X., Littman, D.R., Birnbaum, M.J., and

Polakiewicz, R.D. (2004) Protein kinase C Theta inhibits insulin signaling by phosphorylating IRS1 at

Ser(1101). J.Biol.Chem. 279, 45304-45307

189. Neschen, S., Morino, K., Hammond, L.E., Zhang, D., Liu, Z.X., Romanelli, A.J., Cline, G.W.,

Pongratz, R.L., Zhang, X.M., Choi, C.S., Coleman, R.A., and Shulman, G.I. (2005) Prevention of hepatic

steatosis and hepatic insulin resistance in mitochondrial acyl-CoA:glycerol-sn-3-phosphate

acyltransferase 1 knockout mice. Cell Metab. 2, 55-65

190. Nagle, C.A., An, J., Shiota, M., Torres, T.P., Cline, G.W., Liu, Z.X., Wang, S., Catlin, R.L., Shulman,

G.I., Newgard, C.B., and Coleman, R.A. (2007) Hepatic overexpression of glycerol-sn-3-phosphate

acyltransferase 1 in rats causes insulin resistance. J.Biol.Chem. 282, 14807-14815

191. Turinsky, J., O'Sullivan, D.M., and Bayly, B.P. (1990) 1,2-Diacylglycerol and ceramide levels in

insulin-resistant tissues of the rat in vivo. J.Biol.Chem. 265, 16880-5

192. Jornayvaz, F.R., Jurczak, M.J., Lee, H.Y., Birkenfeld, A.L., Frederick, D.W., Zhang, D., Zhang, X.M.,

Samuel, V.T., and Shulman, G.I. (2010) A high-fat, ketogenic diet causes hepatic insulin resistance in

mice, despite increasing energy expenditure and preventing weight gain.

Am.J.Physiol.Endocrinol.Metab. 299, E808-15

193. Monetti, M., Levin, M.C., Watt, M.J., Sajan, M.P., Marmor, S., Hubbard, B.K., Stevens, R.D., Bain,

J.R., Newgard, C.B., Farese RV, S., Hevener, A.L., and Farese, R.V.,Jr (2007) Dissociation of hepatic

steatosis and insulin resistance in mice overexpressing DGAT in the liver. Cell Metab. 6, 69-78

194. Jornayvaz, F.R., Birkenfeld, A.L., Jurczak, M.J., Kanda, S., Guigni, B.A., Jiang, D.C., Zhang, D., Lee,

H.Y., Samuel, V.T., and Shulman, G.I. (2011) Hepatic insulin resistance in mice with hepatic

overexpression of diacylglycerol acyltransferase 2. Proc.Natl.Acad.Sci.U.S.A. 108, 5748-5752

Bibliography

115

195. Jornayvaz, F.R., and Shulman, G.I. (2012) Diacylglycerol activation of protein kinase Cepsilon and

hepatic insulin resistance. Cell Metab. 15, 574-584

196. Kumashiro, N., Erion, D.M., Zhang, D., Kahn, M., Beddow, S.A., Chu, X., Still, C.D., Gerhard, G.S.,

Han, X., Dziura, J., Petersen, K.F., Samuel, V.T., and Shulman, G.I. (2011) Cellular mechanism of insulin

resistance in nonalcoholic fatty liver disease. Proc.Natl.Acad.Sci.U.S.A. 108, 16381-16385

197. Voshol, P.J., Haemmerle, G., Ouwens, D.M., Zimmermann, R., Zechner, R., Teusink, B., Maassen,

J.A., Havekes, L.M., and Romijn, J.A. (2003) Increased hepatic insulin sensitivity together with

decreased hepatic triglyceride stores in hormone-sensitive lipase-deficient mice. Endocrinology. 144,

3456-3462

198. Park, S.Y., Kim, H.J., Wang, S., Higashimori, T., Dong, J., Kim, Y.J., Cline, G., Li, H., Prentki, M.,

Shulman, G.I., Mitchell, G.A., and Kim, J.K. (2005) Hormone-sensitive lipase knockout mice have

increased hepatic insulin sensitivity and are protected from short-term diet-induced insulin

resistance in skeletal muscle and heart. Am.J.Physiol.Endocrinol.Metab. 289, E30-9

199. Mulder, H., Sorhede-Winzell, M., Contreras, J.A., Fex, M., Strom, K., Ploug, T., Galbo, H., Arner,

P., Lundberg, C., Sundler, F., Ahren, B., and Holm, C. (2003) Hormone-sensitive lipase null mice

exhibit signs of impaired insulin sensitivity whereas insulin secretion is intact. J.Biol.Chem. 278,

36380-36388

200. Roduit, R., Masiello, P., Wang, S.P., Li, H., Mitchell, G.A., and Prentki, M. (2001) A role for

hormone-sensitive lipase in glucose-stimulated insulin secretion: a study in hormone-sensitive lipase-

deficient mice. Diabetes. 50, 1970-1975

201. Brown, J.M., Betters, J.L., Lord, C., Ma, Y., Han, X., Yang, K., Alger, H.M., Melchior, J., Sawyer, J.,

Shah, R., Wilson, M.D., Liu, X., Graham, M.J., Lee, R., Crooke, R., Shulman, G.I., Xue, B., Shi, H., and

Yu, L. (2010) CGI-58 knockdown in mice causes hepatic steatosis, but prevents diet-induced obesity

and glucose intolerance. J.Lipid Res.

202. Badin, P.M., Louche, K., Mairal, A., Liebisch, G., Schmitz, G., Rustan, A.C., Smith, S.R., Langin, D.,

and Moro, C. (2011) Altered skeletal muscle lipase expression and activity contribute to insulin

resistance in humans. Diabetes. 60, 1734-1742

203. Ahmadian, M., Duncan, R.E., Varady, K.A., Frasson, D., Hellerstein, M.K., Birkenfeld, A.L., Samuel,

V.T., Shulman, G.I., Wang, Y., Kang, C., and Sul, H.S. (2009) Adipose overexpression of desnutrin

promotes fatty acid use and attenuates diet-induced obesity. Diabetes. 58, 855-866

Bibliography

116

204. Madrazo, J.A., and Kelly, D.P. (2008) The PPAR trio: regulators of myocardial energy metabolism

in health and disease. J.Mol.Cell.Cardiol. 44, 968-975

205. Lefebvre, P., Chinetti, G., Fruchart, J.C., and Staels, B. (2006) Sorting out the roles of PPAR alpha

in energy metabolism and vascular homeostasis. J.Clin.Invest. 116, 571-580

206. Barish, G.D., Narkar, V.A., and Evans, R.M. (2006) PPAR delta: a dagger in the heart of the

metabolic syndrome. J.Clin.Invest. 116, 590-597

207. Tontonoz, P., and Spiegelman, B.M. (2008) Fat and beyond: the diverse biology of PPARgamma.

Annu.Rev.Biochem. 77, 289-312

208. Lehman, J.J., Barger, P.M., Kovacs, A., Saffitz, J.E., Medeiros, D.M., and Kelly, D.P. (2000)

Peroxisome proliferator-activated receptor gamma coactivator-1 promotes cardiac mitochondrial

biogenesis. J.Clin.Invest. 106, 847-856

209. Puigserver, P., Wu, Z., Park, C.W., Graves, R., Wright, M., and Spiegelman, B.M. (1998) A cold-

inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell. 92, 829-839

210. Rowe, G.C., Jiang, A., and Arany, Z. (2010) PGC-1 coactivators in cardiac development and

disease. Circ.Res. 107, 825-838

211. Xu, H.E., Lambert, M.H., Montana, V.G., Parks, D.J., Blanchard, S.G., Brown, P.J., Sternbach, D.D.,

Lehmann, J.M., Wisely, G.B., Willson, T.M., Kliewer, S.A., and Milburn, M.V. (1999) Molecular

recognition of fatty acids by peroxisome proliferator-activated receptors. Mol.Cell. 3, 397-403

212. Krey, G., Braissant, O., L'Horset, F., Kalkhoven, E., Perroud, M., Parker, M.G., and Wahli, W.

(1997) Fatty acids, eicosanoids, and hypolipidemic agents identified as ligands of peroxisome

proliferator-activated receptors by coactivator-dependent receptor ligand assay. Mol.Endocrinol. 11,

779-791

213. Hostetler, H.A., Petrescu, A.D., Kier, A.B., and Schroeder, F. (2005) Peroxisome proliferator-

activated receptor alpha interacts with high affinity and is conformationally responsive to

endogenous ligands. J.Biol.Chem. 280, 18667-18682

214. Pinent, M., Hackl, H., Burkard, T.R., Prokesch, A., Papak, C., Scheideler, M., Hammerle, G.,

Zechner, R., Trajanoski, Z., and Strauss, J.G. (2008) Differential transcriptional modulation of

Bibliography

117

biological processes in adipocyte triglyceride lipase and hormone-sensitive lipase-deficient mice.

Genomics. 92, 26-32

215. Sapiro, J.M., Mashek, M.T., Greenberg, A.S., and Mashek, D.G. (2009) Hepatic triacylglycerol

hydrolysis regulates peroxisome proliferator-activated receptor alpha activity. J.Lipid Res. 50, 1621-

1629

216. Ong, K.T., Mashek, M.T., Bu, S.Y., Greenberg, A.S., and Mashek, D.G. (2011) Adipose triglyceride

lipase is a major hepatic lipase that regulates triacylglycerol turnover and fatty acid signaling and

partitioning. Hepatology. 53, 116-126

217. Haemmerle, G., Moustafa, T., Woelkart, G., Buttner, S., Schmidt, A., van de Weijer, T., Hesselink,

M., Jaeger, D., Kienesberger, P.C., Zierler, K., Schreiber, R., Eichmann, T., Kolb, D., Kotzbeck, P.,

Schweiger, M., Kumari, M., Eder, S., Schoiswohl, G., Wongsiriroj, N., Pollak, N.M., Radner, F.P., Preiss-

Landl, K., Kolbe, T., Rulicke, T., Pieske, B., Trauner, M., Lass, A., Zimmermann, R., Hoefler, G., Cinti, S.,

Kershaw, E.E., Schrauwen, P., Madeo, F., Mayer, B., and Zechner, R. (2011) ATGL-mediated fat

catabolism regulates cardiac mitochondrial function via PPAR-alpha and PGC-1. Nat.Med. 17, 1076-

1085

218. Samad, F., Hester, K.D., Yang, G., Hannun, Y.A., and Bielawski, J. (2006) Altered adipose and

plasma sphingolipid metabolism in obesity: a potential mechanism for cardiovascular and metabolic

risk. Diabetes. 55, 2579-2587

219. Straczkowski, M., Kowalska, I., Baranowski, M., Nikolajuk, A., Otziomek, E., Zabielski, P.,

Adamska, A., Blachnio, A., Gorski, J., and Gorska, M. (2007) Increased skeletal muscle ceramide level

in men at risk of developing type 2 diabetes. Diabetologia. 50, 2366-2373

220. Haus, J.M., Kashyap, S.R., Kasumov, T., Zhang, R., Kelly, K.R., Defronzo, R.A., and Kirwan, J.P.

(2009) Plasma ceramides are elevated in obese subjects with type 2 diabetes and correlate with the

severity of insulin resistance. Diabetes. 58, 337-343

221. Hajduch, E., Balendran, A., Batty, I.H., Litherland, G.J., Blair, A.S., Downes, C.P., and Hundal, H.S.

(2001) Ceramide impairs the insulin-dependent membrane recruitment of protein kinase B leading to

a loss in downstream signalling in L6 skeletal muscle cells. Diabetologia. 44, 173-183

222. Williams, R.D., Wang, E., and Merrill, A.H.,Jr (1984) Enzymology of long-chain base synthesis by

liver: characterization of serine palmitoyltransferase in rat liver microsomes. Arch.Biochem.Biophys.

228, 282-291

Bibliography

118

223. Merrill, A.H.,Jr (1983) Characterization of serine palmitoyltransferase activity in Chinese hamster

ovary cells. Biochim.Biophys.Acta. 754, 284-291

224. Merrill, A.H.,Jr, Wang, E., and Mullins, R.E. (1988) Kinetics of long-chain (sphingoid) base

biosynthesis in intact LM cells: effects of varying the extracellular concentrations of serine and fatty

acid precursors of this pathway. Biochemistry. 27, 340-345

225. Messmer, T.O., Wang, E., Stevens, V.L., and Merrill, A.H.,Jr (1989) Sphingolipid biosynthesis by

rat liver cells: effects of serine, fatty acids and lipoproteins. J.Nutr. 119, 534-538

226. Powell, D.J., Turban, S., Gray, A., Hajduch, E., and Hundal, H.S. (2004) Intracellular ceramide

synthesis and protein kinase Czeta activation play an essential role in palmitate-induced insulin

resistance in rat L6 skeletal muscle cells. Biochem.J. 382, 619-629

227. Straczkowski, M., Kowalska, I., Nikolajuk, A., Dzienis-Straczkowska, S., Kinalska, I., Baranowski,

M., Zendzian-Piotrowska, M., Brzezinska, Z., and Gorski, J. (2004) Relationship between insulin

sensitivity and sphingomyelin signaling pathway in human skeletal muscle. Diabetes. 53, 1215-1221

228. Chavez, J.A., Knotts, T.A., Wang, L.P., Li, G., Dobrowsky, R.T., Florant, G.L., and Summers, S.A.

(2003) A role for ceramide, but not diacylglycerol, in the antagonism of insulin signal transduction by

saturated fatty acids. J.Biol.Chem. 278, 10297-10303

229. Schmitz-Peiffer, C., Craig, D.L., and Biden, T.J. (1999) Ceramide generation is sufficient to

account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal muscle cells

pretreated with palmitate. J.Biol.Chem. 274, 24202-24210

230. Turban, S., and Hajduch, E. (2011) Protein kinase C isoforms: mediators of reactive lipid

metabolites in the development of insulin resistance. FEBS Lett. 585, 269-274

231. Powell, D.J., Hajduch, E., Kular, G., and Hundal, H.S. (2003) Ceramide disables 3-

phosphoinositide binding to the pleckstrin homology domain of protein kinase B (PKB)/Akt by a

PKCzeta-dependent mechanism. Mol.Cell.Biol. 23, 7794-7808

232. Hajduch, E., Turban, S., Le Liepvre, X., Le Lay, S., Lipina, C., Dimopoulos, N., Dugail, I., and

Hundal, H.S. (2008) Targeting of PKCzeta and PKB to caveolin-enriched microdomains represents a

crucial step underpinning the disruption in PKB-directed signalling by ceramide. Biochem.J. 410, 369-

379

Bibliography

119

233. Wang, G., Krishnamurthy, K., Umapathy, N.S., Verin, A.D., and Bieberich, E. (2009) The carboxyl-

terminal domain of atypical protein kinase Czeta binds to ceramide and regulates junction formation

in epithelial cells. J.Biol.Chem. 284, 14469-14475

234. Holland, W.L., Brozinick, J.T., Wang, L.P., Hawkins, E.D., Sargent, K.M., Liu, Y., Narra, K., Hoehn,

K.L., Knotts, T.A., Siesky, A., Nelson, D.H., Karathanasis, S.K., Fontenot, G.K., Birnbaum, M.J., and

Summers, S.A. (2007) Inhibition of ceramide synthesis ameliorates glucocorticoid-, saturated-fat-,

and obesity-induced insulin resistance. Cell Metab. 5, 167-179

235. Alger, B.E., and Kim, J. (2011) Supply and demand for endocannabinoids. Trends Neurosci. 34,

304-315

236. Di Marzo, V. (2009) The endocannabinoid system: its general strategy of action, tools for its

pharmacological manipulation and potential therapeutic exploitation. Pharmacol.Res. 60, 77-84

237. Leite, C.E., Mocelin, C.A., Petersen, G.O., Leal, M.B., and Thiesen, F.V. (2009) Rimonabant: an

antagonist drug of the endocannabinoid system for the treatment of obesity. Pharmacol.Rep. 61,

217-224

238. Christopoulou, F.D., and Kiortsis, D.N. (2011) An overview of the metabolic effects of

rimonabant in randomized controlled trials: potential for other cannabinoid 1 receptor blockers in

obesity. J.Clin.Pharm.Ther. 36, 10-18

239. Kirkham, T.C., Williams, C.M., Fezza, F., and Di Marzo, V. (2002) Endocannabinoid levels in rat

limbic forebrain and hypothalamus in relation to fasting, feeding and satiation: stimulation of eating

by 2-arachidonoyl glycerol. Br.J.Pharmacol. 136, 550-557

240. Kunos, G., Osei-Hyiaman, D., Liu, J., Godlewski, G., and Batkai, S. (2008) Endocannabinoids and

the control of energy homeostasis. J.Biol.Chem. 283, 33021-33025

241. Cota, D. (2008) Role of the endocannabinoid system in energy balance regulation and obesity.

Front.Horm.Res. 36, 135-145

242. Perkins, J.M., and Davis, S.N. (2008) Endocannabinoid system overactivity and the metabolic

syndrome: prospects for treatment. Curr.Diab Rep. 8, 12-19

Bibliography

120

243. Eichmann, T.O., Kumari, M., Haas, J., Farese, R.V.J., Zimmermann, R., Lass, A., and Zechner, R.

(2012) Studies on the substrate and stereo/regioselectivity of adipose triglyceride lipase, hormone-

sensitive lipase, and diacylglycerol-O-acyltransferase enzymes. J.Biol.Chem.

244. Eichmann, T.O. (2008) Die Untersuchung der stereospezifischen Konformation von enzymatisch

gebildeten Diacylglyzerinen

245. Etschmaier, K., Becker, T., Eichmann, T.O., Schweinzer, C., Scholler, M., Tam-Amersdorfer, C.,

Poeckl, M., Schuligoi, R., Kober, A., Chirackal Manavalan, A.P., Rechberger, G.N., Streith, I.E., Zechner,

R., Zimmermann, R., and Panzenboeck, U. (2011) Adipose triglyceride lipase affects triacylglycerol

metabolism at brain barriers. J.Neurochem. 119, 1016-1028

246. Cases, S., Smith, S.J., Zheng, Y.W., Myers, H.M., Lear, S.R., Sande, E., Novak, S., Collins, C., Welch,

C.B., Lusis, A.J., Erickson, S.K., and Farese, R.V.,Jr (1998) Identification of a gene encoding an acyl

CoA:diacylglycerol acyltransferase, a key enzyme in triacylglycerol synthesis. Proc.Natl.Acad.Sci.U.S.A.

95, 13018-13023

247. Ganji, S.H., Tavintharan, S., Zhu, D., Xing, Y., Kamanna, V.S., and Kashyap, M.L. (2004) Niacin

noncompetitively inhibits DGAT2 but not DGAT1 activity in HepG2 cells. J.Lipid Res. 45, 1835-1845

248. Schaffer, J.E. (2003) Lipotoxicity: when tissues overeat. Curr.Opin.Lipidol. 14, 281-287

249. DeFronzo, R.A. (2010) Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the

missing links. The Claude Bernard Lecture 2009. Diabetologia. 53, 1270-1287

250. Samuel, V.T., and Shulman, G.I. (2012) Mechanisms for insulin resistance: common threads and

missing links. Cell. 148, 852-871

251. Carriere, F., Rogalska, E., Cudrey, C., Ferrato, F., Laugier, R., and Verger, R. (1997) In vivo and in

vitro studies on the stereoselective hydrolysis of tri- and diglycerides by gastric and pancreatic

lipases. Bioorg.Med.Chem. 5, 429-435

252. Paltauf, F., Esfandi, F., and Holasek, A. (1974) Stereospecificity of lipases. Enzymic hydrolysis of

enantiomeric alkyl diacylglycerols by lipoprotein lipase, lingual lipase and pancreatic lipase. FEBS Lett.

40, 119-123

253. Jensen, R.G., DeJong, F.A., Clark, R.M., Palmgren, L.G., Liao, T.H., and Hamosh, M. (1982)

Stereospecificity of premature human infant lingual lipase. Lipids. 17, 570-572

Bibliography

121

254. Wang, C.S., Kuksis, A., Manganaro, F., Myher, J.J., Downs, D., and Bass, H.B. (1983) Studies on

the substrate specificity of purified human milk bile salt-activated lipase. J.Biol.Chem. 258, 9197-9202

255. Morley, N., and Kuksis, A. (1972) Positional specificity of lipoprotein lipase. J.Biol.Chem. 247,

6389-6393

256. Lehner, R., Cui, Z., and Vance, D.E. (1999) Subcellullar localization, developmental expression

and characterization of a liver triacylglycerol hydrolase. Biochem.J. 338 ( Pt 3), 761-768

257. Gilham, D., and Lehner, R. (2004) The physiological role of triacylglycerol hydrolase in lipid

metabolism. Rev.Endocr Metab.Disord. 5, 303-309

258. Wang, H., Wei, E., Quiroga, A.D., Sun, X., Touret, N., and Lehner, R. (2010) Altered lipid droplet

dynamics in hepatocytes lacking triacylglycerol hydrolase expression. Mol.Biol.Cell. 21, 1991-2000

259. Lehner, R., and Verger, R. (1997) Purification and characterization of a porcine liver microsomal

triacylglycerol hydrolase. Biochemistry. 36, 1861-1868

260. Fredrikson, G., Stralfors, P., Nilsson, N.O., and Belfrage, P. (1981) Hormone-sensitive lipase of rat

adipose tissue. Purification and some properties. J.Biol.Chem. 256, 6311-6320

261. Brockerhoff, H. (1965) Stereospecific analysis of triglycerides: an analysis of human depot fat.

Arch.Biochem.Biophys. 110, 586-592

262. Wu, J.W., Wang, S.P., Alvarez, F., Casavant, S., Gauthier, N., Abed, L., Soni, K.G., Yang, G., and

Mitchell, G.A. (2011) Deficiency of liver adipose triglyceride lipase in mice causes progressive hepatic

steatosis. Hepatology. 54, 122-132

263. van Tienhoven, M., Atkins, J., Li, Y., and Glynn, P. (2002) Human neuropathy target esterase

catalyzes hydrolysis of membrane lipids. J.Biol.Chem. 277, 20942-20948

264. Kienesberger, P.C., Lass, A., Preiss-Landl, K., Wolinski, H., Kohlwein, S.D., Zimmermann, R., and

Zechner, R. (2008) Identification of an insulin-regulated lysophospholipase with homology to

neuropathy target esterase. J.Biol.Chem. 283, 5908-5917

265. Yan, W., Jenkins, C.M., Han, X., Mancuso, D.J., Sims, H.F., Yang, K., and Gross, R.W. (2005) The

highly selective production of 2-arachidonoyl lysophosphatidylcholine catalyzed by purified calcium-

independent phospholipase A2gamma: identification of a novel enzymatic mediator for the

generation of a key branch point intermediate in eicosanoid signaling. J.Biol.Chem. 280, 26669-26679

Bibliography

122

266. Tang, J., Kriz, R.W., Wolfman, N., Shaffer, M., Seehra, J., and Jones, S.S. (1997) A novel cytosolic

calcium-independent phospholipase A2 contains eight ankyrin motifs. J.Biol.Chem. 272, 8567-8575

267. Chen, H.C., Smith, S.J., Ladha, Z., Jensen, D.R., Ferreira, L.D., Pulawa, L.K., McGuire, J.G., Pitas,

R.E., Eckel, R.H., and Farese, R.V.,Jr (2002) Increased insulin and leptin sensitivity in mice lacking acyl

CoA:diacylglycerol acyltransferase 1. J.Clin.Invest. 109, 1049-1055

268. McFie, P.J., Banman, S.L., Kary, S., and Stone, S.J. (2011) Murine diacylglycerol acyltransferase-2

(DGAT2) can catalyze triacylglycerol synthesis and promote lipid droplet formation independent of its

localization to the endoplasmic reticulum. J.Biol.Chem. 286, 28235-28246

269. Sattler, W., Puhl, H., Hayn, M., Kostner, G.M., and Esterbauer, H. (1991) Determination of fatty

acids in the main lipoprotein classes by capillary gas chromatography: BF3/methanol

transesterification of lyophilized samples instead of Folch extraction gives higher yields.

Anal.Biochem. 198, 184-190

270. Folch, J., Lees, M., and Sloane Stanley, G.H. (1957) A simple method for the isolation and

purification of total lipides from animal tissues. J.Biol.Chem. 226, 497-509

271. Ackermann, E.J., Conde-Frieboes, K., and Dennis, E.A. (1995) Inhibition of macrophage Ca(2+)-

independent phospholipase A2 by bromoenol lactone and trifluoromethyl ketones. J.Biol.Chem. 270,

445-450

272. Itabashi, Y., Kukis, A., Marai, L., and Takagi, T. (1990) HPLC resolution of diacylglycerol moieties

of natural triacylglycerols on a chiral phase consisting of bonded (R)-(+)-1-(1-naphthyl)ethylamine.

J.Lipid Res. 31, 1711-1717