Embed Size (px)
Citation preview
Redox engineering by ectopic overexpression of NADH kinase 1
in recombinant Pichia pastoris (Komagataella phaffii): Impact 2
on cell physiology and recombinant production of secreted 3
proteins 4
5
Màrius Tomàs-Gamisansa, Cristiane Conte Paim Andradeb, Francisco Marescaa, Sergi Monfortea, 6
Pau Ferrera, Joan Albiola# 7
a Departament d’Enginyeria Química, Biològica i Ambiental, Universitat Autònoma de Barcelona, 8
Bellaterra (Cerdanyola del Vallès), Catalonia, Spain. 9
b Current affiliation: Centre of Biological Engineering, University of Minho, Campus de Gualtar, 10
4710-057, Braga, Portugal 11
Running Title: NADH kinase Redox engineering in Pichia pastoris. 12
#Corresponding author: Joan Albiol. [email protected] 13
14
AEM Accepted Manuscript Posted Online 22 November 2019Appl. Environ. Microbiol. doi:10.1128/AEM.02038-19Copyright © 2019 American Society for Microbiology. All Rights Reserved.
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Abstract 15
High-level expression and secretion of heterologous proteins in yeast causes an increased 16
energy demand, which may result in altered metabolic flux distributions. Moreover, 17
recombinant protein overproduction often results in ER-stress and oxidative stress, causing 18
deviations from the optimal NAD(P)H regeneration balance. In this context, overexpression of 19
genes encoding enzymes catalyzing endogenous NADPH producing reactions, such as the 20
oxidative branch of the pentose phosphate pathway, has been previously shown to improve 21
protein production in Pichia pastoris (syn. Komagataella spp). 22
In this study, we evaluate the overexpression of the S. cerevisiae POS5-encoded NADH 23
kinase in a recombinant P. pastoris strain as an alternative approach to overcome such redox 24
constraints. Specifically, POS5 was co-overexpressed in a strain secreting an antibody fragment, 25
either by directing Pos5 to the cytosol or to the mitochondria. The physiology of the resulting 26
strains was evaluated in continuous cultivations with glycerol or glucose as sole carbon source, 27
as well as under hypoxia (on glucose). Cytosolic targeting of Pos5 NADH kinase resulted in lower 28
biomass-substrate yields but allowed for a 2-fold increase in product specific productivity. In 29
contrast, Pos5 NADH kinase targeting to the mitochondria did not affect the growth physiology 30
and recombinant protein production significantly. Growth physiological parameters were in 31
silico evaluated using the recent upgraded version (v3.0) of the P. pastoris consensus genome-32
scale metabolic model iMT1026, providing insights on the impact of POS5 overexpression on 33
metabolic flux distributions. 34
35
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Importance 36
Recombinant protein overproduction often results in oxidative stress, causing deviations 37
from the optimal redox cofactor regeneration balance. This becomes one of the limiting factors 38
in obtaining high levels of heterologous protein production. Overexpression of redox affecting 39
enzymes has been explored in other organisms such as Saccharomyces cerevisiae as a mean to 40
fine tune the cofactor regeneration balance in order to obtain higher protein titers. In the 41
present work this strategy is explored in P. pastoris. In particular one NADH kinase enzyme from 42
S. cerevisiae (Pos5) is used, either in the cytosol or in mitochondria of P. pastoris, and its impact 43
in the production of a model protein (antibody fragment) is evaluated. A significant 44
improvement in the production of the model protein is observed, when the kinase is directed to 45
the cytosol. These results are significant in the field of heterologous protein production in 46
general and in particular in the development of improved metabolic engineering strategies for 47
P. pastoris. 48
49
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Introduction 50
High-level expression and secretion of heterologous proteins in yeast has been reported 51
to cause a metabolic burden that can significantly impact on energy metabolism and alter the 52
central carbon metabolism flux distribution (1, 2). Producing strains may not cope with the 53
additional demand of ATP, NADPH and precursors for de novo biosynthesis of amino acids, 54
thereby leading to a suboptimal cell fitness and reduced production yields (3). In addition, the 55
folding, posttranslational modifications and secretion processes of complex proteins is avid of 56
resources, particularly of NADPH, which is required for disulphide bound formation and 57
alleviating ER oxidative stress (4). Thus, overproduction of recombinant proteins would result in 58
altered redox cofactor state and, specifically, a reduction in NADPH availability. Such alterations 59
in redox cofactor balance have a strong impact on cell metabolism (5). Therefore, strain 60
engineering strategies targeting redox metabolism have been successfully applied to improve E. 61
coli (6, 7), S. cerevisiae (8, 9) and P. pastoris (10–12) strains for a range of different applications. 62
NADPH availability is tightly related to biomass yields and recombinant protein 63
production yields (13). Driouch et al. (2012) reported that Aspergillus niger strains 64
overproducing recombinant proteins show higher fluxes through the oxidative branch of 65
pentose phosphate pathway (PPP), which is the main cytosolic NADPH generation pathway. 66
Also, Nocon et al.(2016) overexpressed genes coding for enzymes of the oxidative branch of 67
PPP, obtaining higher productivities of recombinant cytosolic human superoxide dismutase 68
(hSOD). Indeed, a preceding study identified several metabolic engineering targets for 69
improving recombinant protein production using a genome-scale metabolic model (11). 70
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Interestingly, about the 50% of the identified targets pointed towards NADPH supply reactions 71
(15). 72
Based on the key metabolic role of NADPH on protein synthesis and secretion and the 73
supporting evidence from previous studies pointing at the positive effect that its increased 74
supply seems to have on recombinant protein production, we have investigated the impact of 75
the overexpression of a heterologous gene encoding a NADH kinase on a recombinant P. 76
pastoris strain secreting an antibody fragment (Fab). Previous studies in our research group 77
reported an increase of Fab specific productivity under reduced oxygen availability conditions 78
(16). This was concomitant with a shift to a respiro-fermentative metabolism, as reflected in the 79
generation of ethanol and arabitol for cofactor reoxidation (17). Furthermore, parallel 80
metabolomics analyses revealed that the reduced oxygen availability for electron transport 81
chain leads to higher NADH/NAD+ ratios under such hypoxic conditions (18). In this context, we 82
postulate that the NADH excess found under hypoxic conditions could be a potential source of 83
electrons for NADPH regeneration and, therefore, the physiological effects of the ectopic NADH 84
kinase might be boosted under hypoxia in comparison to the reference normoxic condition. In 85
order to test our hypothesis, redox-engineered strains were grown on glycerol and glucose 86
under normoxic conditions as well as on glucose under hypoxic conditions. Overall, we aimed to 87
investigate the combined effect of a process strategy (hypoxic conditions) and metabolic 88
engineering strategy to improve production of a secreted recombinant protein. 89
Moreover, we have used the genome-scale metabolic model to evaluate the experimental 90
physiological datasets obtained in chemostat cultivations and assist the metabolic 91
interpretation of the observed macroscopic changes. 92
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Results 93
Cytosolic and mitochondrial overexpression of POS5 and its effect on 94
recombinant Fab secretion 95
The codon optimised POS5 gene and its truncated version devoid of its native 96
mitochondrial signal peptide encoding sequence was overexpressed under the control of the 97
constitutive GAP promoter into a P. pastoris X-33 strain expressing the genes encoding for the 98
2F5 antibody fragment under the control of the same promoter, aiming for the targeted co-99
overexpression of NADH kinase in the mitochondria or in the cytosol (mPOS5 and cPOS5, 100
respectively). A set of twelve independent clones of each co-overexpressing strain were verified 101
for integration of mPOS5 or cPOS5 prior to the screening in baffled shake flasks. 102
The two series of Fab-producing clones overexpressing each of the POS5 forms were first 103
tested in a small scale screening at shake flask scale. Product titters and biomass were 104
measured after 24 h of cultivation to check the effect of co-expression of mPOS5 or cPOS5 on 105
recombinant Fab secretion. The average Fab yields of each series of 12 clones were normalised 106
to those obtained from the reference strain X-33/2F5, as shown in Figure 2A. Co-expression of 107
the two POS5 gene forms demonstrated a largely unchanged recombinant protein secretion 108
capacity. Interestingly, the average Fab yields of cPOS5 clones showed a high standard 109
deviation as a result of two distinct clone populations: one with the same behaviour of the 110
control strain (FC = 1.07 ± 0.21) and another showing significantly higher product yields (FC = 111
2.34 ± 0.52). A plausible explanation for the observed clonal variation would be that isolated 112
transformants differ in the dosage of the co-expressed cPOS5 gene. In order to test this 113
hypothesis, the recombinant cPOS5 gene dosage was determined by droplet digital PCR 114
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
(ddPCR) for a representative sample of each clone subpopulations, i.e. those giving significantly 115
increased Fab yields and the other showing no significant effect on product yield compared to 116
the reference strain. Notably, the cPOS5 clone population showing significantly increased 117
product yields corresponded to those clones harbouring 2 copies of the cPOS5 expression 118
cassette (Figure 2B). 119
Physiological characterization of the POS5 co-overexpressing strains growing in 120
chemostat cultures 121
In order to study the effects of altering NADPH generation on cell physiology and Fab 122
extracellular production under different environmental conditions, the reference strain X-123
33/2F5, as well as representative clones of the mPOS5 (mitochondrial expression of POS5), 124
cPOS5 and 2cPOS5 (cytosolic expression with one and two copies of POS5, respectively) were 125
cultivated in carbon-limited chemostat bioreactors using glycerol or glucose as carbon source 126
under normoxic conditions, as well as under hypoxia when using glucose (see Table S2 for a 127
summary of physiological growth parameters). Since Pos5p catalyses a NADPH-generating 128
reaction, an alteration of the redox cofactor balance can be expected. In order to check 129
whether POS5 overexpression has a significant impact on the redox cofactor balance, the 130
NADPH/NADP+ ratios were measured and calculated for all strains under the tested growth 131
conditions (Figure 3). The reference strain X-33/2F5 showed the lowest ratios in all the three 132
growth conditions. Strains co-overexpressing mPOS5 and cPOS5 had comparable 133
NADPH/NADP+ values to the reference strain when growing on glycerol, whereas this value was 134
2-fold higher in the 2cPOS5 strain. Co-overexpression of POS5 had a higher impact in cells 135
growing on glucose compared to glycerol, as shown in Figure 3A and Figure 3B. In this case, 136
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
both cPOS5 and mPOS5 strains showed a trend of increasing NADPH/NADP+ ratios of about a 137
20% and 30% compared to the reference strain, respectively, even though this effect was not 138
statistically significant. The 2cPOS5 strain also had a higher impact in the NADPH/NADP+ ratio in 139
glucose-grown than in glycerol-grown cells. Notably, when the oxygen supply was reduced, this 140
adaptation pattern was further accentuated, i.e. NADPH/NADP+ ratios were comparatively 141
higher for all four strains (Figure 3C). Such increase in the NADPH/NADP+ ratio observed in 142
hypoxic conditions may be a consequence of the NADH excess or accumulation observed under 143
these conditions (18), which is generally converted to ethanol or other by-products such as 144
arabitol (16, 19). Thus, an increased availability of NADH under hypoxic conditions would allow 145
higher conversions to NADPH by Pos5p and, consequently, increased NADPH/NADP+ ratios. 146
Such cofactor balance alteration due to POS5 overexpression had a further impact on the 147
physiological growth profile of the POS5-engineered strains (Figure 4). Although such impact 148
was not significant in all strains and growth conditions, some tendencies can be appreciated 149
(Figure 4). In particular, the specific oxygen uptake rate (qO2) showed a tendency to increase in 150
all growth conditions in both cytosolic- and mitochondrial-directed POS5 overexpressing 151
strains. Notably, 2cPOS5 showed higher qO2 than the other strains in most of the tested 152
conditions, probably reflecting an increased activity of the respiratory chain due to the 153
additional ATP demand for consumption in the NADH kinase reaction (see sections 3.4. and 4.3 154
for further discussion). Similarly, the specific CO2 production rate (qCO2) showed an increasing 155
trend as a result of higher cytosolic NADH kinase activity levels, both in normoxic and hypoxic 156
conditions, whereas mPOS5 seemed to show an opposite trend, i.e. reduced qCO2 compared to 157
the reference strain (Figure 4). Notably, the variation in CO2 specific production rates, although 158
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
not statistically significant, appear to be at expenses of biomass yields (YXS), which show the 159
opposite pattern: a statistically significant reduction for 2cPOS5 and an increase in mPOS5 in 160
hypoxic conditions (Table S2). 161
As expected, by-product formation was detected only under respiro-fermentative 162
conditions attained under hypoxia. In particular, while both cPOS5 and 2cPOS5 showed similar 163
or higher arabitol and ethanol specific production rates, ethanol formation decreased in mPOS5 164
compared to the reference strain (Figure 4C). 165
POS5 overexpression also had a clear impact on Fab production (Figure 4D). When 166
growing on glycerol, only 2cPOS5 improved significantly Fab productivity (qP,Fab) with respect to 167
the reference strain. However, all three redox-engineered strains showed significantly 168
enhanced qP,Fab when growing on glucose under normoxic conditions, being 2cPOS5 the strain 169
with the best performance (1.55-fold increase compared to the reference strain). In oxygen 170
limiting conditions, although mPOS5 showed a decrease in productivity, the cytosolic POS5 171
overexpressing strains improved up to 1.8-fold Fab production compared to the reference 172
strain grown under the same condition, i.e. showing an additive effect of hypoxia and cPOS5 173
overexpression. 174
Cytosolic POS5 overexpression further enhances glycolysis upregulation caused 175
by hypoxia 176
To gain further insight on the additive effect of hypoxia and cytosolic POS5 177
overexpression observed in chemostat cultures, the relative transcriptional levels of TDH3 (as 178
marker gene for glycolysis) and POS5 were determined by ddPCR for all strains and growth 179
conditions tested (Figure 5). Previous studies demonstrate that, in contrast to S. cerevisiae, 180
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
glycolysis is transcriptionally upregulated by hypoxia in P. pastoris (17, 20). Glucose-grown cells 181
showed an increase in TDH3 transcripts when POS5 was overexpressed, compared to reference 182
strain. Specifically, TDH3 was upregulated 1.44- and 1.24-fold in the 1- copy and 2-copies cPOS5 183
strains, respectively, under normoxic conditions, whereas the mPOS5 strain showed only a 184
slight increase (1.12-fold) in TDH3 mRNA levels. Such upregulation of TDH3 was further 185
enhanced by hypoxic conditions, with a 1.93-, 3.16- and 2.46-fold increase observed in the 186
1cPOS5, 2cPOS5 and mPOS5, respectively, compared to the reference strain. Notably, such 187
results correlate with the increased Fab secretion observed in this condition. In contrast, no 188
significant changes were observed in TDH3 transcriptional levels when glycerol was used as 189
carbon source in 1cPOS5 and 2cPOS5 strains compared to the reference strain, except for 190
mPOS5 strain, with a 1.3-fold increase. 191
Concerning POS5 transcription, the 2cPOS5 strain showed the highest POS5 192
transcriptional levels in all the tested conditions. As expected, POS5 mRNA levels in glucose-193
grown 2cPOS5 cells were at least 2-fold higher than in the 1cPOS5 strain. Moreover, as POS5 194
was expressed under the control of the GAP promoter, i.e. the TDH3 promoter, POS5 195
expression was 1.8 fold higher in 2cPOS5 cells grown under glucose/hypoxia compared to 196
glucose/normoxic-grown cells. Such hypoxic boost effect was not observed in glucose-grown 197
1cPOS5 cells, suggesting that ectopic POS5 mRNA levels are too low to cause a significant 198
metabolic effect in these cells. Similarly, POS5 expression levels in glucose-grown mPOS5 cells 199
did not show significant differences between cultivation conditions. 200
Notably, POS5 transcriptional levels in glycerol-grown cells appeared to be on the same 201
range as in glucose/normoxic-grown cells. This is in contradiction to previous studies reporting 202
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
that TDH3 promoter expression level is around two-thirds of that in glucose (21, 22). Such 203
differences could be related to the different growth conditions used in previous studies (shake 204
flask cultures) and the different analytical method employed in our study (ddPCR). 205
In silico interpretation of POS5 overexpression physiological impact 206
One of the applications of genome-scale metabolic models (GSMM) is to assist the 207
interpretation of biological data to reinforce or discard hypothesis (23). We used the iMT1026 208
v3.0 genome-scale metabolic model for P. pastoris (23), which enables improved prediction 209
capabilities for growth on glycerol compared to the original version (24). 210
The effect of Fab overproduction on flux distribution was tested by performing a series of 211
simulations successively increasing Fab production constrains (employing FSEOF). NADPH 212
turnover ratios were calculated for the resulting predicted fluxes. A clear correlation between 213
Fab production and this cofactor turnover was obtained due to the increased demand of amino 214
acid biosynthesis for Fab production (Figure S1). 215
The metabolic impact of POS5 overexpression was also modelled by constraining 216
successively increasing fluxes through the specific NADH kinase reaction (cytosolic or 217
mitochondrial) and using MOMA for assessing the new distribution. As a result, metabolic 218
fluxes through some pathways are predicted to be up or downshifted concomitantly with an 219
increased flux through NADH kinase reaction. Mitochondrial- or cytosolic-directed POS5 220
overexpression impacts differently on metabolic flux distribution, as it is affecting different 221
compartmentalised NAD(P)H pools. For example, cytosolic Pos5p synthesis supplies an 222
important fraction of NADPH under glucose growing conditions, (Figure 6B, C) and, 223
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
consequently, the flux through the oxidative branch of the pentose phosphate pathway 224
decreased. In contrast, mitochondrial NADPH kinase synthesis would not be able to supply 225
enough cytosolic NADPH and therefore the flux of the NADPH generating reactions of the 226
pentose phosphate pathway is enhanced (Figure 6D, E). Another predicted and important 227
difference between cytosolic and mitochondrial Pos5p overexpression is the increased activity 228
of glycolysis pathway (coherent with experimental ddPCR data) and TCA cycle when the 229
cytosolic NADPH kinase is overexpressed, while mitochondrial overexpression is predicted to 230
cause a decreased flux through this pathway. However, some predicted changes are common 231
between both cPOS5 and mPOS5 strains: in both cases, there is an increase of the activity of the 232
oxidative phosphorylation that supplies the additional ATP demanded in the NADPH kinase 233
reaction. The increased electron transfer to the respiratory chain is predicted to be through 234
NADH oxidation in mPOS5 but, for cPOS5, succinate oxidation to fumarate would be the 235
additional electron supply. Overall, an increase of oxygen consumption, CO2 production, by-236
product formation as well as Fab production is predicted for all the conditions tested in 237
simulations for the cPOS5 strain. Similarly, an increase of qO2and qCO2 during growth on glycerol 238
and glucose-normoxia is predicted for mPOS5, whereas no significant changes are predicted in 239
terms of by-product formation or Fab production for this strain. Conversely, mitochondrial 240
overexpression of NADH kinase leads to a reduction in CO2, ethanol and Fab production in 241
hypoxic conditions, coherent with the experimentally observed behaviour. 242
Discussion 243
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
POS5 overexpression increases NADPH availability and recombinant protein 244
production 245
Several studies have addressed the impact of heterologous protein expression on 246
metabolism and cell physiology of P. pastoris (25) and how environmental conditions can 247
modulate the product yields (17, 26). Heyland et al.(2010) postulated and Nocon et al.(2016, 248
2014) provided strong evidence that increasing flux through the PPP allows supplying additional 249
NADPH, thereby compensating the extra demand caused by biosynthetic processes involved in 250
heterologous protein production. In fact, our in silico simulations indicate that an increase of 251
Fab extracellular production correlates with higher NADPH turnover rates. Redox-engineered 252
strains are able to generate additional NADPH compared to the reference X-33/2F5 strain, 253
resulting in increased NADPH/NADP+ ratios. Moreover, higher POS5 gene dosage further 254
increased POS5 transcriptional levels, correlating with higher NADPH/NADP+ ratios and Fab 255
protein production. Higher NADPH/NADP+ ratios may reflect more NADPH available for cellular 256
processes, not only biosynthesis of recombinant protein, but also for its potential use in protein 257
folding and ER oxidative stress response processes involved in protein secretion (14). Therefore, 258
although POS5 overexpression leads to a drain in cell energy resources (ATP consumption), it 259
ensures non-limiting supply NADPH, which has been demonstrated to allow for increased 260
recombinant protein production yields (12). Our in silico calculations show increased Fab 261
secretion when the ectopic Pos5p NADH kinase is targeted to the cytosol. This may indicate 262
that Fab biosynthesis and secretion could compensate the cofactor perturbation (boosted 263
NADPH levels) by draining such NADPH excess, thereby restoring the redox cofactor original 264
state. 265
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Metabolic impact of POS5 overexpression 266
Our results strongly suggest that Pos5p NADH kinase overexpression perturbs cofactor 267
balance. This is further reflected in the differences observed in growth profiles of POS5 268
overexpressing strains compared to the reference strain, pointing to an impact in the 269
distribution of fluxes through the cell’s metabolic network. Previous studies in other yeast and 270
fungi have reported that the overexpression of NADH kinases has a strong effect on metabolic 271
flux distribution (27–29), as also predicted by the simulations performed with iMT1026 v3.0. 272
Noteworthy, the impermeability of organelle membranes to NAD(P)H leads to cofactor 273
reoxidation in the same compartment where they are reduced (30). Thus, different effects of 274
the cofactor perturbation are expected in cPOS5 and mPOS5 strains. POS5 overexpression 275
provides a source of NADPH in addition to the oxidative branch of the PPP, which is the main 276
cytosolic NADPH-generating pathway in yeast (31). As extra NADPH is supplied by the NADH 277
kinase, flux through the oxidative branch of the PPP is predicted to decrease, as observed when 278
POS5 overexpression is targeted to the cytosol of S. cerevisiae (28). Indeed, in S. cerevisiae 279
NADPH inhibits ZWF1, the first step in oxidative PPP (32). Therefore, the increased levels of 280
NADPH in cPOS5 (and 2cPOS5), would be coherent with the predicted reduction of flux through 281
the oxidative branch of the PPP. Conversely, mitochondrial overexpression of POS5 would 282
result in an increased flux through the PPP. Part of the carbon flux would be redirected to the 283
mitochondria for generating the required NADH surplus at expenses of cytosolic NADH yields. 284
This reduction in cytosolic NADH generation would be compensated by the additional NADPH 285
synthesised. Thus, enzymes able to use both NADH and NADPH would turn its specificity to 286
NADPH. Therefore, despite the production of additional mitochondrial NADPH, the 287
impermeability of mitochondrial membrane to redox cofactors would force cytosolic NADPH 288
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
generation pathways to supply the required NADPH for compartment-specific biosynthetic 289
processes. The correlation between the flux through the oxidative branch of the pentose 290
phosphate pathway and biomass yields has been widely reported (31, 33). Accordingly, the 291
predicted flux increase and decrease through the PPP as a consequence of mPOS5 and cPOS5 292
overexpression, respectively, are in agreement with the biomass yields observed in chemostat 293
cultivations: an increase of YXS yield in mPOS5 and a reduction in cPOS5 strains (Table S2) 294
Similarly, in other species, while mitochondrial POS5 overexpressing strains show an increase in 295
YXS (5, 27, 29), the cytosolic overexpression of the NADH kinase results in reduced biomass 296
yields. It is noteworthy to mention that the scaled reduced costs of Fab production over 297
biomass generation are narrow (1.01 10-4) and, therefore, differences in biomass yields are 298
mainly consequence of POS5 overexpression and the concomitant redistribution of metabolic 299
fluxes and energy consumption, rather than increased Fab yields. 300
Redox cofactor balance and energy metabolism are very closely linked in the respiratory 301
chain. It is therefore plausible that a perturbation in redox cofactor levels would cause a 302
metabolic flux redistribution to restore the energy supply capacity of the cells. Moreover, in 303
addition to the redox cofactor imbalance created by a surplus of NADPH at the expenses of 304
NADH, the Pos5p-catalyzed NADH kinase reaction is ATP consuming. According to the 305
performed simulations, higher fluxes in oxidative phosphorylation would compensate this ATP 306
drain. These predictions are also in agreement with the increased oxygen consumption rates of 307
the mutant strains during the chemostat cultivations. Nevertheless, in the cPOS5 strain, part of 308
the cytosolic NADH generated is consumed in the NADH kinase reaction and cannot be neither 309
used for biosynthetic purposes nor transported by mitochondrial redox shuttles to further 310
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
deliver its electrons into the respiratory chain. Despite the higher TCA cycle activity and 311
consequent increase in mitochondrial NADH generation, it would not provide enough reducing 312
power for the electron transport chain and a reduction in the NADH electron transfer as well as 313
an increase in alternative electron delivery mechanisms (i.e. succinate oxidation to fumarate) 314
are predicted for cPOS5. Since the NADH demand is located in mitochondria in mPOS5, the 315
compensation of the drain would rely on flux readjustments in mitochondrially-located 316
reactions. In this case, an activation of the malic enzyme as well as an increase in the 317
anaplerotic feed of TCA cycle intermediates would supply the additional NADH enabling 318
increased electron transfer to the respiratory chain. As in S. cerevisiae and A. nidulans, we do 319
not observe a reduction in biomass yields (27, 28). 320
When growing in hypoxic conditions, the additional supply of ATP required by Pos5p is 321
limited due to the restricted oxygen availability constraining the respiratory chain activity. 322
Simulations predict that when cPOS5 cells grow under hypoxia, the ATP drain caused by Pos5p 323
activity cannot be completely compensated, leading to a decrease of cell fitness and increased 324
hypoxic effects. Accordingly, experimental data shows an increase of by-product formation 325
when cPOS5 is overexpressed. These results are also supported by the ddPCR analyses 326
performed, showing a transcriptional adaptation of the glycolytic pathway under hypoxic 327
conditions positively correlated with gene copy number (and transcriptional levels) of POS5. 328
TDH3 has been reported to increase its transcription levels in hypoxia (Baumann et al., 2010). 329
Our results strongly support that increasing NADPH availability by overexpressing POS5 330
enhances this hypoxic effect. Consequently, since the ectopic expression of POS5 is also under 331
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
the transcriptional control of the TDH3 promoter, POS5 positively feeds-back its own 332
overexpression at low oxygen availability conditions. 333
Conversely, despite the oxygen limitations, the mPOS5 strain would be able to 334
compensate the drained ATP by supplying additional mitochondrial NADH to the respiratory 335
chain, thereby increasing ATP production. This strain, similarly to A. nidulans, is able to 336
overcome the ATP drain and increase biomass yields respect to the control strain, even under 337
hypoxic conditions (27). Although P. pastoris is commonly classified as Crabtree negative yeast, 338
it can produce certain amount of ethanol and other by-products (33), particularly under hypoxic 339
conditions (17). By-product formation is a consequence of limitations in TCA cycle and oxidative 340
phosphorylation capacities, leading to an excess of reduced NAD(P)H that the cell is not able to 341
reoxidise by the respirative pathway (33, 34). The mPOS5 strain showed a decrease in ethanol 342
secretion due to the reduction in available mitochondrial NADH, while arabitol production 343
remained comparable to the reference X-33/2F5 strain. The cPOS5 strains showed increased 344
arabitol and ethanol production both in experimental data and simulations. In hypoxic 345
conditions, increased NADH kinase levels would convert the excess of NADH to NADPH (as 346
reflected in the experimentally determined increased NADPH/NADP+ ratio); this NADPH surplus 347
would be subsequently reoxidised through the generation of arabitol. In addition, simulations 348
indicate an increase in TCA cycle flux leading to enhanced NADH generation. Due to the 349
reduced capacity of oxidative phosphorylation caused by the oxygen limitation, the additional 350
NADH generated has to be reoxidised forming ethanol, in agreement with the experimental 351
observation. 352
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
To conclude, in this study, the S. cerevisiae POS5 gene, encoding for a NADH kinase, was 353
overexpressed in P. pastoris, either directed to the cytosol or to the mitochondria. The 354
physiological characterisation of these strains in chemostat cultivations showed a clear effect of 355
POS5 overexpression on redox cofactor balance. Indeed, POS5 overexpression increased 356
NADPH/NADP+ ratio in all the strains and conditions tested. Furthermore, the strain containing 357
two copies of POS5 integrated in the cell’s genome (2cPOS5) showed the highest increase in 358
NADPH/NADP+ ratios compared to the reference strain. This strongly supports a positive 359
correlation between POS5 gene dosage and NADPH availability. Moreover, this correlation can 360
also be observed when comparing the strain performance: 2cPOS5 cells showed the greatest 361
fold change increase in Fab productivity. These results are also in agreement with the 362
performed simulations, which show a positive correlation between NADPH turnover and Fab 363
production as well as increased Fab productivities when flux through the cytosolic NADH kinase 364
reaction is increased. 365
The different behaviour of mPOS5 and cPOS5 strains indicates the complexity of cell 366
metabolism with organelle membranes impermeable to redox cofactors and highlights the 367
importance of directing enzymes to the appropriate compartment when designing metabolic 368
engineering strategies. 369
As a result of POS5 overexpression, metabolic fluxes through the central carbon 370
metabolism redistribute. Notably, redox engineered strains showed higher oxygen 371
requirements concomitant with increased oxidative phosphorylation in order to replenish the 372
ATP pools drained in the reaction catalysed by the NADH kinase. Consequently, these strains, 373
particularly cPOS5, are more sensitive to O2 (i.e. show a lower threshold for the onset of 374
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
respiro-fermentative metabolism) and showed more extreme/drastic hypoxic effects (increased 375
by-product formation). In fact, the effects of POS5 overexpression are boosted under hypoxic 376
conditions, and redox-engineered strains show higher NADPH/NADP+ ratios and Fab 377
productivities. Moreover, this effect is particularly notorious when gene copy number of cPOS5 378
is increased, remarking how important gene dosage is while performing strain modifications at 379
a metabolic level. The gene dosage effect is supported by the ddPCR results, with 2cPOS5 380
showing the highest fold-change in TDH3 transcription (the glycolytic marker) and the highest 381
POS5 transcription level. Notably, both in silico flux distributions and macroscopic growth 382
parameters of the POS5-engineered strains were coherent in terms of increased demand of 383
oxygen, CO2 and by-product generation profiles, as well as Fab productivities, revealing 384
iMT1026 v3.0 as a useful tool for consistently assessing the interpretation of the cultivation 385
results, as well as taking into account the effect of growth conditions on the metabolic 386
phenotype of the engineered strains. 387
Overall, we demonstrated the impact of redox cofactor perturbation in cell metabolism 388
and provided further evidence of NADPH metabolism as key cell engineering target for 389
improved recombinant protein production. Nonetheless, further studies are needed in order to 390
dissect the actual contribution of protein folding and secretion (vs protein synthesis) to the 391
increased NADPH demand. In this respect, future comparative studies of the impact of NADPH 392
metabolism perturbation on the production of the same model protein produced in the cytosol 393
or secreted should bring further insights. In addition, future development of production 394
processes (i.e. high cell density fed-batch cultivations) using POS5-engineered strains would 395
also benefit from regulated POS5 expression (e.g. co-induced with the recombinant product 396
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
gene, expressed under the control of an easily tuneable promoter), to minimize the potential 397
metabolic burden associated to the GAP-driven constitutive expression of this gene. 398
Materials and methods 399
Strain generation 400
A Pichia pastoris X-33 (ThermoFisher Scientific) derived strain expressing multiple copies 401
of the genes encoding the human antigen-binding fragment (Fab) 2F5 under the transcriptional 402
control of the constitutive GAP promoter and with the α-mating factor secretion signal 403
sequence from Saccharomyces cerevisiae (35) was used in this study. 404
The S. cerevisiae POS5 gene, encoding for the mitochondrial NADH kinase Pos5p (36, 37) 405
was codon-optimised for heterologous expression in P. pastoris and synthetized by Geneart 406
(ThermoFisher Scientific), cloned into a pPUZZLE vector (38) under the control of GAP 407
promoter, thereby generating vector pPUZZLE_mPOS5 (Figure 1). Similarly, an analogous 408
construction, pPUZZLE_cPOS5, was constructed expressing a 5’-truncated POS5 excluding the 409
first 48 bp coding for the N-terminal 16 amino acids, allowing for cytosolic Pos5p localisation. 410
Escherichia coli DH5α was used for plasmid propagation. 411
P. pastoris X-33/2F5 strain transformation and recombinant clone isolation were 412
performed as described in (38). The presence of integrated expression cassette into the host 413
genome was confirmed by colony PCR (39) using the primer pairs described in Table S1. 414
Clone screening at small scale 415
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
A set of 12 recombinant clones for each strain construct were screened for growth and 416
Fab 2F5 production in triplicate baffled shake flasks cultures using glucose minimal medium, as 417
described by Baumann et al.(2011). 418
Chemostat cultivations 419
Two independent carbon-limited chemostat cultivations were performed for each strain 420
in three different growing conditions, using as carbon source either glycerol, glucose under 421
normoxic conditions (100% air in the inlet gas composition) or glucose under hypoxic conditions 422
(25:75 of air:N2 in the inlet gas composition), as previously described (40). Cultivations were 423
performed at a working volume of 1-L in a 2-L bench-top Biostat B (B. Braun Biotech 424
International) bioreactor. Operational conditions were set to 25°C, 700 rpm, 1 vvm inlet gas 425
flow, 0.2 bar overpressure, 0.1 h-1 dilution rate (D) and pH 5.0 controlled by addition of 15% 426
(v/v) NH4. Samples were taken at 3rd, 4th and 5th residence times for cell density monitoring, Fab 427
titers, extracellular metabolites and dry cell weight (DCW) analyses. The off-gases were cooled 428
dawn in a condenser at 4C and further desiccated in two silica gel columns. The off-gas O2 and 429
CO2 concentrations were measured by BCP-O2 (zirconium dioxide) and BCP-CO2 (infrared) 430
BlueSens Gas Analyser, respectively. 431
For reactor inoculation, strains were cultivated in 1-L baffled Erlenmeyer flask containing 432
150 mL YPG broth (1% w/v yeast extract, 2% w/v peptone, 1% w/v glycerol) and antibiotic (100 433
µg·L-1zeocin for control strain or 500 µg·L-1geneticin G418 for NADH kinase recombinant clones 434
selection) at an optical density (OD600) between 0.15 and 0.30. Pre-cultures were incubated at 435
25°C under 130 rpm for 16-24 h. 436
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Batch medium composition was as previously described (16). Briefly, it contained: 40 g·L-1 437
glycerol, 1.8 g·L-1 citric acid, 12.6 g·L-1 (NH4)2HPO4, 0.5 g·L-1 MgSO4.7H2O, 0.9 g·L-1 KCl, 0.02 g·L-1 438
CaCl2.2H2O, 4.6 mL·L-1 trace salts stock solution, 2 mL·L-1 of biotin solution (0.2 g·L-1) and 250 439
µL·L-1 of Glanapon 2000 antifoam (Bussetti). Chemostat medium was also adapted from (16). 440
Briefly, it contained: 50 g·L-1 carbon source – glycerol or glucose –, 0.92 g·L-1 monohydrate citric 441
acid; 4.35 g·L-1(NH4)2HPO4; 0.65 g·L-1 MgSO4.7H20; 1.7 g·L-1KCl; 0.01 g·L-1 CaCl2.2H20; 1.6 mL 442
trace salt solution, 1 mL biotin solution (0.2 g·L-1) and 200 µL·L-1Glanapon antifoam. Trace salt 443
solution was composed of: 6.0 g·L-1 CuSO4.5H2O; 0.08 g·L-1NaI; 3.36 g·L-1 MnSO4.H2O; 0.2 g·L-1 444
Na2MoO4.2H2O; 0.02 g·L-1 H3BO3; 0.82 g·L-1 CoCl2.6H2O; 20 g·L-1 ZnCl2; 65 g·L-1 FeSO4.7H2O and 445
5.0 mL H2SO4 (95-98% w/w). Media pH was adjusted to 5.0 with 6 N HCl. 446
Analytical methods 447
Biomass concentration 448
Cell density was monitored by optical density in a DR3900 spectrophotometer (Hach 449
Lange GMBH) at 600 nm. Dry cell weight (DCW) was measured by gravimetric methods as 450
follows: A volume of 2 to 10 mL of sample was filtered in Glass Fibre Prefilters (Merck 451
Millipore), pre-weighted after drying at 105°C for 24 h. Each filter was washed twice with 10 mL 452
of distilled water; dried at 105°C for 24 h, cooled in a desiccator and weighted. 453
Fermentation products analysis 454
Citric acid, glucose, glycerol, arabitol, succinic acid, acetic acid and ethanol were analyzed 455
by HPLC in an UltiMate 3000 Liquid Chromatography Systems (Dionex) using an ICSep ICE-456
COREGEL 87H3 (Transgenomic) ion exchange column and a Waters 2410 (Waters) refraction 457
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
index detector. 6 mM sulphuric acid was used as continuous phase at 0.5 mL/min flow and 20 458
μL sample injection volume. Data were analysed in CROMELEON software (Dionex). 459
Quantification of Fab 460
Fab 2F5 was quantified by ELISA in 96-well Immuno Plates (Nunc, Thermo Scientific) as 461
described by Gasser and co-workers (35). Briefly, plates were subjected to an overnight pre-462
coating of Fab specific Anti-Human IgG (I5260, Sigma) primary antibody in PBS buffer (1:1000). 463
Then, plates were washed three times with PBS 1% Tween 20 and samples and Fab standard 464
(Bethyl Inc.) were diluted in PBS buffer containing 10% (w/v) BSA (Sigma) and 0.1% (v/v) Tween 465
80. Plates were incubated for 2 h, washed again with PBS 1% Tween 20 three times and 466
incubated for 1 h after addition of Anti-Human Kappa Light Chains (bound)-Alkaline 467
Phosphatase (Sigma) secondary antibody. Plates were washed with PBS 1% Tween 20 three 468
times, treated with pNPP phosphatase substrate (Sigma) and the absorbance was measured at 469
405 nm using a Multiskan™ FC Microplate reader (Thermo Scientific). 470
ddPCR (droplet digital PCR) analysis 471
ddPCR was used to quantify recombinant POS5 gene copy number and transcriptional levels of 472
TDH3 and POS5. 473
For gene dosage determination, genomic DNA was purified by Wizard Genomic DNA 474
Purification Kit (Promega), according to manufacturer instructions, and quantified in NanoDrop 475
2000 Spectrophotometer (Thermo Scientific). 0.5 µg DNA were digested by EcoRI and BamHI 476
FastDigest Enzymes (Thermo Scientific) to produce DNA fragments of size lower than 5 kb and 477
purified with Wizard® SV Gel and PCR Clean-Up System (Promega). Reaction conditions (1× 478
Supermix ddPCR TaqMan, 300 nM of each primer, 200 nM of each probe and 0.02 ng·μL-1 479
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
digested genomic DNA) and operational conditions were performed as suggested by BioRad 480
and optimised for P. pastoris as described in (41). The annealing temperature was set to 57°C, 481
after temperature gradient determination. Primers used for ddPCR are listed in (Table S1). 482
Expression levels of TDH3 and POS5 were quantified by ddPCR as described in (42) with 483
minor modifications. Briefly, 1 ng of cDNA was used for the reaction mixture instead of 0.4 ng 484
and annealing temperature was set to 56.5°C in the PCR reaction. The house-keeping gene β-485
actin (ACT1) was selected to normalize data. Primers used are described in Table S1. 486
NADPH/NADP+ ratio determination 487
Samples for NADPH and NADP+ quantification were taken and rapidly quenched with cold 488
60% v/v methanol (43, 44). Cell suspensions were centrifuged and washed twice with 489
quenching solution as described in Ortmayr et al., ( 2014) (4000 g, 10C, 10 min in a Centrifuge 490
5804 R, Eppendorf). Finally, pellets were stored at 80C. NADPH and NADP+ concentrations 491
were determined using EnzyChromTM NADP+/NADPH Assay Kit (BioAssay Systems) and the 492
optical densities were measured by means of a Multiskan™ FC Microplate reader (Thermo 493
Scientific) at a wavelength of 595 nm. Analyses were performed in duplicate. The relative 494
standard deviation (RSD) of the analytical method was 20%. 495
Statistical analysis 496
Chemostat cultivation data was checked for consistency and standard reconciliation 497
procedures were applied (45). A statistical consistency test, based on h-index as described in 498
(45) was passed with a confidence level of 95%. Consequently, there was no evidence of gross 499
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
measurement errors. A statistical comparison of the macroscopic growth profiles of the 500
different strains was performed using the Microsoft Excel 2-tailed Student’s t-Test. 501
Metabolic modelling 502
The iMT1026 v3.0 metabolic model (biomodels database MODEL1612130000) of P. 503
pastoris (24) was used in the COBRA Toolbox v2.0.6 (46) under Matlab 2014 (Mathworks, USA) 504
with SBML toolbox v4.1.0 (47) and libSBML library v5.12.0 (48) with IBM® ILOG® CPLEX® 505
Optimization Studio 12.7 as solver. The prediction of flux redistribution due to Fab 506
overexpression was performed employing Flux Scanning based on Enforced Objective Function 507
(FSEOF) (49) by maximizing the biomass production at a constrained range of Fab secretion (0 – 508
0.12 mg·gDCW-1·h-1). Particularly, redox cofactor turnover rates were calculated using the flux-509
sum analysis (50) on each resulting flux distribution. A cytosolic NADH kinase reaction was 510
incorporated into the model (the corresponding mitochondrial reaction was not added, as the 511
iMT1026 v3.0 already contained the endogenous mitochondrial NADH kinase reaction). The 512
perturbation of the NADH kinase activity on flux distribution was calculated by Minimization of 513
Metabolic Adjustment (MOMA) (51) performing a series of simulations enforcing a minimal flux 514
through the NADH kinase reaction (cytosolic or mitochondrial) constraining the uptake of 515
carbon source to the control strain experimental values in the case of normoxia (glycerol and 516
glucose), and additionally constraining the oxygen uptake rate for the simulations in hypoxic 517
conditions. The resulting flux distributions at different NADH kinase reaction fluxes (0 – 2 518
mmol·gDCW-1·h-1) were compared against the control strain (X-33/2F5) in which the NADH kinase 519
reaction flux is 0. 520
521
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
Acknowledgements 522
This work was supported by the project CTQ2013-42391-R and CTQ2016-74959-R 523
(AEI/FEDER, UE) of the Spanish Ministry of Economy, Industry and Competitiveness, 2014-SGR-524
452 and Reference Network in Biotechnology (XRB) (Generalitat de Catalunya), the grant FPU 525
FPU12/06185 (M.T.) and the grant FPI BES-2014-067935 (S.M.) of the Spanish Ministry of 526
Education, Culture and Sport, as well as the postdoctoral grant from the Ciencia sem Fronteiras 527
Program (CNPq, Brazil, Process 249872/2013-7) of Cristiane C P Andrade. We thank MSc 528
student Ane Quesada for supporting small scale clone screening experiments. 529
References 530
531
1. Ferrer P, Albiol J. 2014. 13C-Based Metabolic Flux Analysis in Yeast: The Pichia pastoris 532
Case, p. 209–232. In Mapelli, V (ed.), Methods in molecular biology. Springer New York, 533
New York, NY. 534
2. Klein T, Niklas J, Heinzle E. 2015. Engineering the supply chain for protein 535
production/secretion in yeasts and mammalian cells. J Ind Microbiol Biotechnol 42:453–536
464. 537
3. Wu G, Yan Q, Jones JA, Tang YJ, Fong SS, Koffas MAG. 2016. Metabolic Burden: 538
Cornerstones in Synthetic Biology and Metabolic Engineering Applications. Trends 539
Biotechnol 34:652–664. 540
4. Delic M, Rebnegger C, Wanka F, Puxbaum V, Haberhauer-Troyer C, Hann S, Köllensperger 541
G, Mattanovich D, Gasser B. 2012. Oxidative protein folding and unfolded protein 542
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
response elicit differing redox regulation in endoplasmic reticulum and cytosol of yeast. 543
Free Radic Biol Med 52:2000–2012. 544
5. Hou J, Lages NF, Oldiges M, Vemuri GN. 2009. Metabolic impact of redox cofactor 545
perturbations in Saccharomyces cerevisiae. Metab Eng 11:253–61. 546
6. Lee HC, Kim JS, Jang W, Kim SY. 2010. High NADPH/NADP+ ratio improves thymidine 547
production by a metabolically engineered Escherichia coli strain. J Biotechnol 149:24–32. 548
7. Siedler S, Bringer S, Bott M. 2011. Increased NADPH availability in Escherichia coli: 549
Improvement of the product per glucose ratio in reductive whole-cell biotransformation. 550
Appl Microbiol Biotechnol 92:929–937. 551
8. Kim S, Hahn JS. 2015. Efficient production of 2,3-butanediol in Saccharomyces cerevisiae 552
by eliminating ethanol and glycerol production and redox rebalancing. Metab Eng 31:94–553
101. 554
9. Geertman J-MA, van Maris AJA, van Dijken JP, Pronk JT. 2006. Physiological and genetic 555
engineering of cytosolic redox metabolism in Saccharomyces cerevisiae for improved 556
glycerol production. Metab Eng 8:532–542. 557
10. Krainer FW, Dietzsch C, Hajek T, Herwig C, Spadiut O, Glieder A. 2012. Recombinant 558
protein expression in Pichia pastoris strains with an engineered methanol utilization 559
pathway. Microb Cell Fact 11:22. 560
11. Nocon J, Steiger MG, Pfeffer M, Sohn SB, Kim TY, Maurer M, Rußmayer H, Pflügl S, Ask M, 561
Haberhauer-Troyer C, Ortmayr K, Hann S, Koellensperger G, Gasser B, Lee SY, 562
Mattanovich D. 2014. Model based engineering of Pichia pastoris central metabolism 563
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
enhances recombinant protein production. Metab Eng 24:129–138. 564
12. Nocon J, Steiger M, Mairinger T, Hohlweg J, Rußmayer H, Hann S, Gasser B, Mattanovich 565
D. 2016. Increasing pentose phosphate pathway flux enhances recombinant protein 566
production in Pichia pastoris. Appl Microbiol Biotechnol 100:5955–5963. 567
13. Heyland J, Fu J, Blank LM, Schmid A. 2010. Quantitative physiology of Pichia pastoris 568
during glucose-limited high-cell density fed-batch cultivation for recombinant protein 569
production. Biotechnol Bioeng 107:357–368. 570
14. Driouch H, Melzer G, Wittmann C. 2012. Integration of in vivo and in silico metabolic 571
fluxes for improvement of recombinant protein production. Metab Eng 14:47–58. 572
15. Mattanovich D, Sauer M, Gasser B. 2016. Industrial Microorganisms: Pichia pastoris, p. 573
687–714. In Industrial Biotechnology. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 574
Germany. 575
16. Baumann K, Maurer M, Dragosits M, Cos O, Ferrer P, Mattanovich D. 2008. Hypoxic fed-576
batch cultivation of Pichia pastoris increases specific and volumetric productivity of 577
recombinant proteins. Biotechnol Bioeng 100:177–183. 578
17. Baumann K, Carnicer M, Dragosits M, Graf AB, Stadlmann J, Jouhten P, Maaheimo H, 579
Gasser B, Albiol J, Mattanovich D, Ferrer P. 2010. A multi-level study of recombinant 580
Pichia pastoris in different oxygen conditions. BMC Syst Biol 4:141. 581
18. Carnicer Heras M. 2012. Systematic metabolic analysis of recombinant Pichia pastoris 582
under different oxygen conditions A Metabolome and Fluxome Based Study. Universitat 583
Autònoma de Barcelona. 584
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
19. Carnicer M, Baumann K, Töplitz I, Sánchez-Ferrando F, Mattanovich D, Ferrer P, Albiol J. 585
2009. Macromolecular and elemental composition analysis and extracellular metabolite 586
balances of Pichia pastoris growing at different oxygen levels. Microb Cell Fact 8:65. 587
20. Baumann K, Dato L, Graf AB, Frascotti G, Dragosits M, Porro D, Mattanovich D, Ferrer P, 588
Branduardi P. 2011. The impact of oxygen on the transcriptome of recombinant S . 589
cerevisiae and P . pastoris - a comparative analysis. BMC Genomics 12:218. 590
21. Waterham HR, Digan ME, Koutz PJ, Lair S V, Cregg JM. 1997. Isolation of the Pichia 591
pastoris glyceraldehyde-3-phosphate dehydrogenase gene and regulation and use of its 592
promoter. Gene 186:37–44. 593
22. Prielhofer R, Cartwright SP, Graf AB, Valli M, Bill RM, Mattanovich D, Gasser B. 2015. 594
Pichia pastoris regulates its gene-specific response to different carbon sources at the 595
transcriptional, rather than the translational, level. BMC Genomics 16:167. 596
23. Tomàs-Gamisans M, Ferrer P, Albiol J. 2018. Fine-tuning the P. pastoris iMT1026 597
genome-scale metabolic model for improved prediction of growth on methanol or 598
glycerol as sole carbon sources. Microb Biotechnol 11:224–237. 599
24. Tomàs-Gamisans M, Ferrer P, Albiol J. 2016. Integration and validation of the genome-600
scale metabolic models of Pichia pastoris: A comprehensive update of protein 601
glycosylation pathways, lipid and energy metabolism. PLoS One 11:e0148031. 602
25. Jordà J, Jouhten P, Cámara E, Maaheimo H, Albiol J, Ferrer P. 2012. Metabolic flux 603
profiling of recombinant protein secreting Pichia pastoris growing on glucose:methanol 604
mixtures. Microb Cell Fact 11:57. 605
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
26. Dragosits M, Stadlmann J, Albiol J, Baumann K, Maurer M, Gasser B, Sauer M, Altmann F, 606
Ferrer P, Mattanovich D. 2009. The effect of temperature on the proteome of 607
recombinant Pichia pastoris. J Proteome Res 8:1380–1392. 608
27. Panagiotou G, Grotkjær T, Hofmann G, Bapat PM, Olsson L. 2009. Overexpression of a 609
novel endogenous NADH kinase in Aspergillus nidulans enhances growth. Metab Eng 610
11:31–39. 611
28. Hou J, Vemuri GN, Bao X, Olsson L. 2009. Impact of overexpressing NADH kinase on 612
glucose and xylose metabolism in recombinant xylose-utilizing Saccharomyces cerevisiae. 613
Appl Microbiol Biotechnol 82:909–919. 614
29. Qiao K, Wasylenko TM, Zhou K, Xu P, Stephanopoulos G. 2017. Lipid production in 615
Yarrowia lipolytica is maximized by engineering cytosolic redox metabolism. Nat 616
Biotechnol 35:173–177. 617
30. Bakker BM, Overkamp KM, van Maris AJ, Kötter P, Luttik M a H, van Dijken JP, Pronk JT. 618
2001. Stoichiometry and compartmentation of NADH metabolism in Saccharomyces 619
cerevisiae. FEMS Microbiol Rev 25:15–37. 620
31. Blank LM, Lehmbeck F, Sauer U. 2005. Metabolic-flux and network analysis in fourteen 621
hemiascomycetous yeasts. FEMS Yeast Res 5:545–558. 622
32. Llobell A, Lopez-Ruiz A, Peinado J, Lopez-Barea J. 1988. Glutathione reductase directly 623
mediates the stimulation of yeast glucose-6-phosphate dehydrogenase by GSSG. 624
Biochem J 249:293–296. 625
33. Heyland J, Fu J, Blank LM, Schmid A. 2011. Carbon metabolism limits recombinant 626
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
protein production in Pichia pastoris. Biotechnol Bioeng 108:1942–1953. 627
34. Vemuri GN, Eiteman M, McEwen JE, Olsson L, Nielsen J. 2007. Increasing NADH oxidation 628
reduces overflow metabolism in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 629
104:2402–7. 630
35. Gasser B, Maurer M, Gach J, Kunert R, Mattanovich D. 2006. Engineering of Pichia 631
pastoris for improved production of antibody fragments. Biotechnol Bioeng 94:353–361. 632
36. Outten CE, Culotta VC. 2003. A novel NADH kinase is the mitochondrial source of NADPH 633
in Saccharomyces cerevisiae. EMBO J 22:2015–24. 634
37. Strand MK, Stuart GR, Longley MJ, Graziewicz MA, Dominick OC, Copeland WC. 2003. 635
POS5 Gene of Saccharomyces cerevisiae Encodes a Mitochondrial NADH Kinase Required 636
for Stability of Mitochondrial DNA. Eukaryot Cell 2:809–820. 637
38. Baumann K, Adelantado N, Lang C, Mattanovich D, Ferrer P. 2011. Protein trafficking , 638
ergosterol biosynthesis and membrane physics impact recombinant protein secretion in 639
Pichia pastoris. Microb Cell Fact 10:93. 640
39. Bergkessel M, Guthrie C. 2013. Chapter 25: Colony PCR., p. 299–309. In Laboratory 641
Methods in Enzymology. 642
40. Adelantado N, Tarazona P, Grillitsch K, García-Ortega X, Monforte S, Valero F, Feussner I, 643
Daum G, Ferrer P. 2017. The effect of hypoxia on the lipidome of recombinant Pichia 644
pastoris. Microb Cell Fact 16:86. 645
41. Cámara E, Albiol J, Ferrer P. 2016. Droplet digital PCR-aided screening and 646
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
characterization of Pichia pastoris multiple gene copy strains. Biotechnol Bioeng 647
113:1542–1551. 648
42. Cámara E, Landes N, Albiol J, Gasser B, Mattanovich D, Ferrer P. 2017. Increased dosage 649
of AOX1 promoter-regulated expression cassettes leads to transcription attenuation of 650
the methanol metabolism in Pichia pastoris. Sci Rep 7:44302. 651
43. Ortmayr K, Nocon J, Gasser B, Mattanovich D, Hann S, Koellensperger G. 2014. Sample 652
preparation workflow for the liquid chromatography tandem mass spectrometry based 653
analysis of nicotinamide adenine dinucleotide phosphate cofactors in yeast. J Sep Sci 654
37:2185–2191. 655
44. Carnicer M, Canelas AB, Ten Pierick A, Zeng Z, van Dam J, Albiol J, Ferrer P, Heijnen JJ, 656
van Gulik W. 2012. Development of quantitative metabolomics for Pichia pastoris. 657
Metabolomics 8:284–298. 658
45. Noorman HJ, Romein B, Luyben KC a M, Heijnen JJ. 2000. Classification, error detection, 659
and reconciliation of process information in complex biochemical systems. Biotechnol 660
Bioeng 49:364–376. 661
46. Schellenberger J, Que R, Fleming RMT, Thiele I, Orth JD, Feist AM, Zielinski DC, Bordbar A, 662
Lewis NE, Rahmanian S, Kang J, Hyduke DR, Palsson B. 2011. Quantitative prediction of 663
cellular metabolism with constraint-based models: The COBRA Toolbox v2.0. Nat Protoc 664
6:1290–1307. 665
47. Keating SM, Bornstein BJ, Finney A, Hucka M. 2006. SBMLToolbox: An SBML toolbox for 666
MATLAB users. Bioinformatics 22:1275–1277. 667
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
48. Bornstein BJ, Keating SM, Jouraku A, Hucka M. 2008. LibSBML: An API library for SBML. 668
Bioinformatics 24:880–881. 669
49. Choi HS, Lee SY, Kim TY, Woo HM. 2010. In Silico Identification of Gene Amplification 670
Targets for Improvement of Lycopene Production. Appl Environ Microbiol 76:3097–3105. 671
50. Chung B, Lee D-Y. 2009. Flux-sum analysis: a metabolite-centric approach for 672
understanding the metabolic network. BMC Syst Biol 3:117. 673
51. Segrè D, Vitkup D, Church GM. 2002. Analysis of optimality in natural and perturbed 674
metabolic networks. Proc Natl Acad Sci U S A 99:15112–15117. 675
676
677
Figures and tables 678
Figure 1. Plasmid maps for pPUZZLE_mPOS5 and pPUZZLE_cPOS5. In red, the restriction 679
enzymes used: SbfI and SfiI for cloning mPOS5 and cPOS5; AscI for plasmid linearization; BglII 680
for plasmid ligation verification. pPUZZLE contains the kanMX gene encoding for kanamycin 681
resistance (E. coli) and geneticin G418 resistance (P. pastoris). 682
683
Figure 2 Representation of the results of clone screening and selection. (A) Specific Fab 684
production was measured and normalised to the reference strain X-33/2F5; (B) A 685
representative clone of each population was analysed for determining POS5 gene copy number. 686
Error bars show the standard deviation of the average from at least three different clones. 687
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
688
Figure 3. Representation of NADPH/NADP+ molar ratios in all the strains for growth on glycerol 689
(A) or glucose (B) in normoxic conditions and hypoxic conditions (C). 690
691
Figure 4. Main macroscopic growth parameters for all strains and chemostat cultivation 692
conditions. A: specific O2 consumption rate; B: specific CO2 production rate; C: specific by-693
product generation rate in glucose–hypoxic conditions; D: fold change of Fab productivity of the 694
generated strains in comparison to the X-33/2F5 strain. Each measure +/- standard deviation of 695
two different experiments. 696
697
Figure 5. Transcriptional analysis of chemostat cultivations. A: Relative changes of TDH3 698
mRNA levels in POS5 strains are expressed as fold-change referred to TDH3 levels in the 699
reference strain. B: POS5 mRNA levels are expressed as relative expression levels to ACT1 700
(POS5/ACT1 signal) for each individual strain and condition (*p-val < 0.05, **p-val < 0.01) 701
compared to control strain. Error bars show the standard deviation between three technical 702
replicates. 703
704
Figure 6. Graphical representation of the predicted metabolic flux redistribution when 705
overexpressing the NADH kinase in the cytosol (A-C) or mitochondria (D-F). Cells were 706
growing in glycerol (A, D), glucose under normoxic conditions (B, E) and glucose under 707
hypoxia (C, F). Blue, red and black lines denote reactions with increased, decreased or non-708
changed metabolic fluxes, respectively. 709
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
710 Table 1. List of primers used in this study. 711
Primer name Sequence (5’-3’) Tm
(°C)
POS5-F a,b ATCCGCGCCTGCAGGAATGTTTGTTAGAGTTAAGTTGAACAAGCCAGTTAA 67
POS5_cyt-F a,b ATCCGCGCCTGCAGGAATAATGTCCACTTTGGACTCCCATTCCTTGAA 68
POS5-R a,b ATGACTAGGCCGAGGCGGCCTTAGTCGTTGTCAGTCTGTCTC 68
POS5_int-R a,b AGCAACACCGTCAGCAGTAG
POS5_amp-F c GGAGTGTCACTTGAAGAA 38.6
POS5_amp-R c CGTCAGCAGTAGTTCTAG 36.6
POS5 probe c ACTCCAACTCCTCCATCGTTACTCA (5’: 6-Fam / 3’: BHQ-1) 57.1
a Primer used for cloning POS5 into pPUZZLE. 712
b Primer used for clone verification. 713
c Primer used for gene copy number determination. 714
715
716
on Novem
ber 4, 2020 by guesthttp://aem
.asm.org/
Dow
nloaded from
