ORIGINAL RESEARCH published: 22 July 2022 doi: 10.3389/fbioe.2022.942304 Combining Metabolic Engineering and Multiplexed Screening Methods for 3-Hydroxypropionic Acid Production in Pichia pastoris Albert Fina 1, Stephanie Heux 2, Joan Albiol 1 and Pau Ferrer 1* 1 Department of Chemical, Biological and Environmental Engineering, Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Catalonia, Spain, 2TBI, Université de Toulouse, CNRS, INRAE, INSA, Toulouse, France Production of 3-hydroxypropionic acid (3-HP) in Pichia pastoris (syn. Komagataella phaffii) via the malonyl-CoA pathway has been recently demonstrated using glycerol as a carbon source, but the reported metrics were not commercially relevant. The flux through the heterologous pathway from malonyl-CoA to 3-HP was hypothesized as the main bottleneck. In the present study, different metabolic engineering approaches have been combined to improve the productivity of the original 3-HP producing strains. To Edited by: do so, an additional copy of the gene encoding for the potential rate-limiting step of the Xiao-Jun Ji, pathway, i.e., the C-terminal domain of the malonyl-CoA reductase, was introduced. In Nanjing Tech University, China addition, a variant of the endogenous acetyl-CoA carboxylase (ACC1S1132A) was Reviewed by: overexpressed with the aim to increase the delivery of malonyl-CoA. Furthermore, the Yongjin Zhou, Dalian Institute of Chemical Physics genes encoding for the pyruvate decarboxylase, aldehyde dehydrogenase and acetyl-CoA (CAS), China synthase, respectively, were overexpressed to enhance conversion of pyruvate into Yang Gu, Nanjing Normal University, China cytosolic acetyl-CoA, and the main gene responsible for the production of the by- *Correspondence: product D-arabitol was deleted. Three different screening conditions were used to Pau Ferrer classify the performance of the different strains: 24-deep-well plates batch cultures, Pau.Ferrer@uab.cat small-scale cultures in falcon tubes using FeedBeads® (i.e., slow release of glycerol over time), and mini bioreactor batch cultures. The best two strains from the FeedBeads® Specialty section: This article was submitted to screening, PpHP8 and PpHP18, were tested in bioreactor fed-batch cultures using a pre- Industrial Biotechnology, fixed exponentially increasing feeding rate. The strain PpHP18 produced up to 37.05 g L−1 a section of the journal Frontiers in Bioengineering and of 3-HP at 0.712 g L−1 h−1 with a final product yield on glycerol of 0.194 Cmol−1 in fed- Biotechnology batch cultures. Remarkably, PpHP18 did not rank among the 2-top producer strains in Received: 12 May 2022 small scale batch cultivations in deep-well plates and mini bioreactors, highlighting the Accepted: 22 June 2022 importance of multiplexed screening conditions for adequate assessment of metabolic Published: 22 July 2022 engineering strategies. These results represent a 50% increase in the product yield and Citation: Fina A, Heux S, Albiol J and Ferrer P final concentration, as well as over 30% increase in volumetric productivity compared to (2022) Combining Metabolic the previously obtained metrics for P. pastoris. Overall, the combination of glycerol as Engineering and Multiplexed Screening Methods for 3-Hydroxypropionic Acid carbon source and a metabolically engineered P. pastoris strain resulted in the highest 3- Production in Pichia pastoris. HP concentration and productivity reported so far in yeast. Front. Bioeng. Biotechnol. 10:942304. doi: 10.3389/fbioe.2022.942304 Keywords: 3-hydroxypropionic acid, Pichia pastoris, glycerol, malonyl-CoA, acetyl-CoA, metabolic engineering Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 1 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris INTRODUCTION ultimately, of crude glycerol, which is an abundant, inexpensive and renewable feedstock produced as a main Microbial production of platform chemicals is an increasingly by-product of the conventional biodiesel production attractive alternative to petrochemical-derived products. One process. Morevover, glycerol is very attractive for such chemical, 3-hydroxypropionic acid (3-HP), is an organic production of organic acids such as 3-HP due to its higher acid that can be chemically converted to acrylic acid, 1,3- degree of reduction compared to glucose. propanediol, and malonic acid, among others (della Pina et al., Production of 3-HP in yeast has been mainly pursued by 2011), offering a great potential for bioproduction of a wide introducing the malonyl-CoA or β-alanine pathways, e.g., range of polymers. In fact, 3-HP has been included in the list of reaching a titer of about 13.7 g L−1 and with a 0.14 ± 0.0 C- the top platform chemicals that can be obtained from biomass mol−1 yield on glucose in Saccharomyces cerevisiae fed-batch (Werpy and Petersen, 2004). cultures (Borodina et al., 2015). The methylotrophic yeast 3-HP can be naturally produced by several bacteria through Pichia pastoris (syn. Komagataella phaffii) is able to grow different biosynthetic routes. Among them, the glycerol- efficiently on glycerol. Furthermore, crude glycerol may dependant pathway, and the pathways via malonyl-CoA and contain from 1% to 25% w/w methanol (Luo et al., 2016), β-alanine intermediates have been the most extensively making this feedstock ideal for a methylotrophic investigated (Kumar et al., 2013; de Fouchécour et al., microorganism such as P. pastoris. In addition, this yeast can 2018). Multiple microorganisms have been engineered to grow at a pH as low as 3 (Cregg et al., 1993; Werten et al., 1999), produce 3-HP, including bacteria (Jers et al., 2019) and which allows performing the fermentation process at a pH below yeast (Ji et al., 2018). The highest 3-HP concentrations and the pKa of 3-HP (4.51). This would enable the use of in situ productivities achieved so far have been obtained in product recovery systems, thereby avoiding 3-HP accumulation Escherichia coli [76.2 g L−1 of 3-HP and 1.89 g L−1 h−1, to toxic levels and facilitating 3-HP export from cells (van Maris respectively (Kim et al., 2020)] and Klebsiella pneumoniae et al., 2004). The 3-HP extraction yield using a membrane- (Zhao et al., 2019). Both bacterial species were engineered assisted in situ product recovery system was increased from to produce 3-HP through the glycerol-dependent pathway. 5% to 74% when the pH of the medium was decreased from This facilitated the use of glycerol as a carbon source and, 5 to 3.2 (Chemarin et al., 2019). FIGURE 1 | Summary of metabolic engineering strategies for the optimisation of 3-HP production in P. pastoris. Strain engineering started from the 3-HP producing strain PpHP6 containing the malonyl-CoA to 3-HP pathway (Fina et al., 2021). Heterologous enzymes are depicted in yellow, endogenous enzymes being overexpressed are shown in green, and the enzymes catalyzing the reactions of genes being deleted are shown in red. The (2x) indicates that a second copy of the MCR-CCa gene was introduced. Enzyme abbreviations: Pdc1, pyruvate decarboxylase; Ald6Sc, aldehyde dehydrogenase from S. cerevisiae; ACSSeL641P, acetyl-CoA synthase from S. enterica with the mutation L641P; Acc1Yl, acetyl-CoA carboxylase from Y. lipolytica; Acc1S1132A, acetyl-CoA carboxylase from P. pastoris with the mutation S1132A; MCR-CCa, C-terminal domain of the malonyl-CoA reductase from C. aurantiacus; MCR-NCa, N-terminal domain of the malonyl-CoA reductase from C. aurantiacus; cPos5Sc, NADH kinase from S. cerevisiae relocated into the cytoplasm; ArDH, D-arabitol dehydrogenase. Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 2 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris TABLE 1 | Strain list. Strain name Parental strain Additional expression cassettes Source or deleted genes compared to parental strain X-33 Invitrogen-Thermo Fisher Scientific (MA, United States) PpHP2 X-33 pGAP_mcr-NCa_AOX1tt Fina et al. (2021) pGAP_mcr-CCa_AOX1tt PpHP6 PpHP2 pGAP_ACC1Yl_ScCYC1tt Fina et al. (2021) pGAP_cPOS5Sc_RPS3tt PpHP7 PpHP2 pGAP_mcr-CCa_TDH3tt This study PpHP9 PpHP7 pGAP_ACC1S1132A_AOX1tt This study Plasmid source (Liu et al., 2019); Addgene #126740 PpHP11 PpHP7 pTEF1_acsSeL641P_RPS3tt This study PpHP12 PpHP7 pTEF1_acsSeL641P_RPS3tt This study pMDH3_ALD6Sc_TDH3tt PpHP8 PpHP6 pGAPeN_ mcr-CCa_TDH3tt This study PpHP13 PpHP8 pTEF1_acsSeL641P_RPS3tt This study PpHP14 PpHP8 pTEF1_acsSeL641P_RPS3tt This study pMDH3_ALD6Sc_TDH3tt PpHP15 PpHP8 ΔArDH This study PpHP16 PpHP8 ΔArDH pPDC1_PDC1_PDC1tt This study PpHP17 PpHP13 ΔArDH This study PpHP18 PpHP13 ΔArDH pPDC1_PDC1_PDC1tt This study In a recent study, we addressed the use of P. pastoris to industrially-relevant, as a minimum productivity of produce 3-HP from glycerol (Fina et al., 2021). This yeast was 2.5 g L−1 h−1, 50~100 g L−1 and >0.5 g/g yield are generally metabolically engineered to express the malonyl-CoA reductase required to achieve an economically feasible bioprocess for pathway, which consists of two consecutive steps reducing carboxylic acid production (Werpy and Petersen, 2004; malonyl-CoA into 3-HP, while consuming two NADPH Mazzoli, 2021). molecules (Figure 1). Specifically, we expressed the gene Here, further metabolic engineering of the previously encoding for the bi-functional enzyme malonyl-CoA reductase developed 3-HP-producing strains (Fina et al., 2021) was (MCR) from Chloroflexus aurantiacus. This pathway provides the performed to increase the production metrics. In particular, simplest way to produce 3-HP from glycerol in P. pastoris, as the we aimed at improving both the biosynthetic pathway for 3- glycerol-dependant pathway is coenzyme B12-dependant. Since HP from its metabolic precursor (acetyl-CoA), as well as P. pastoris does not naturally produce coenzyme B-12, the improving the metabolic flux towards acetyl-CoA synthesis introduction of the glycerol-dependant pathway for 3-HP and reducing by-product (arabitol) formation. Notably, we biosynthesis pathway in this cell factory would require either used three different small-scale cultivation systems for better the addition of such expensive cofactor to the fermentation characterisation of the impact of genetic modifications on cell medium or the co-expression of the multistep B-12 growth and product yields and subsequent transfer to bioreactor- biosynthetic pathway (Fang et al., 2018). scale cultivations. In addition, several improvements were made to the original strain. First, the independent expression of the two MCR subunits was implemented, leading to a 12.5-fold increase in 3-HP MATERIALS AND METHODS production. Second, the resulting strain was further engineered to increase the availability of NADPH and malonyl-CoA, which Molecular Biology and Strains are the two substrates of this route. To this end, the acetyl-CoA All the strains used in this study are listed in Table 1. A detailed carboxylase gene from Yarrowia lipolytica (ACC1Yl) and a NADH description of the molecular biology workflow to generate each kinase from Saccharomyces cerevisiae (cPOS5Sc) were expressed in strain can be found in the Supplementary Material S1. The the cytosol. However, this strategy resulted in rather modest modular cloning vectors GoldenPics [(Prielhofer et al., 2017); improvements in terms of product yield, pointing to the delivery Addgene Kit #1000000133] and the vectors for CRISPR-Cas9 of of malonyl-CoA as a major bottleneck to further improve 3-HP the CrisPi kit (Gassler et al., 2019); Addgene Kit #1000000136) production. Notably, this strain produced up to 24.75 g L−1 of 3- were used. HP in 45.5 h in a fed-batch culture using glycerol as a carbon The strains generated using the GoldenPics plasmids were source, which is, to our knowledge, the highest productivity generated by integration of the heterologous DNA at the loci reported so far in yeast (0.54 g L−1 h−1) (Fina et al., 2021). described in the Supplementary Material S1. Despite providing the first demonstration of the potential of P. For the new strains generated by CRISPR-Cas9, the pastoris for 3-HP production from glycerol, 3-HP production integration cassette was amplified from P. pastoris genome metrics from our previous study are still far from being using a high-fidelity Q5 polymerase (New England Biolabs, Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 3 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris TABLE 2 | Oligonucleotides used for gene copy number determination analysis overnight at 30°C and 200 rpm in an incubator shaker with ddPCR. Multitron Standard (Infors HT, Bottmingen, Switzerland) with Primer name Primer sequence a 2.5 cm orbit. 50 μL of overnight-grown cultures were used to inoculate each well of a 24 deep-well plate containing 2 mL of ddPCR_mcrCCa_FW CCTAACGATGTTGCTGCTTTGGAG Buffered Minimal Glycerol (BMG) medium, containing 100 mM ddPCR_mcrCCa_RV GGATCAGGTGGATTAGGCAAGTTAGC ddPCR_act_FW TCCGGTGGTACTACTATGTTCC potassium phosphate buffer pH 6, 1.34% yeast nitrogen base ddPCR_act_RV GATAGAACCACCGATCCATACG (YNB), 1% v/v glycerol, and 0.4 mg L−1 biotin. The cultures were incubated at 28°C and 220 rpm in the same incubator shaker using a platform with a slope of 20° to improve the aeration. The cultures were grown for 48 h to ensure the full consumption of the MA, United States) and the recommended protocol (Gassler substrate. Thereafter, culture samples were centrifuged at et al., 2019). The integration loci are also indicated in the 12,000 g for 5 min and filtered through a 0.20 µm pore size Supplementary Material S1. The integrity of the inserted syringe filter (SLLGX13NK, Merck Millipore, CA, cassette was verified by Sanger sequencing at the Genomics United States). The 3-HP was quantified using HPLC. and Bioinformatics Service of the Universitat Autònoma de The parental clones PpHP2 and PpHP6 were screened in Barcelona. triplicates. Six to eight clones were screened in triplicates for the strains generated by single-homology integration (PpHP7, PpHP8, and PpHP9). One single clone was screened in Copy Number Determination by Droplet triplicates for the strains generated using CRISPR-Cas9. Digital PCR The number of copies of the gene mcr-CCa of the highest producing clones of the strains PpHP7 and PpHP8 was tested Small-Scale Screening in Falcon Tubes with droplet digital PCR (ddPCR) following a previously reported Using FeedBeads® method (Cámara et al., 2016), except that the QX200TM The inoculum was prepared following the same protocol ddPCRTM EvaGreen Supermix (Biorad, CA, United States) was described for the deep-well plates screenings. Afterwards, 50- used instead of Taqman probes. First, the genomic DNA of the mL falcon tubes were filled with 5 mL of Buffer Minimal medium highest producing clones was purified using the Wizard ® (BM; 100 mM potassium phosphate buffer pH 6, 1.34% YNB and 0.4 mg L−1 biotin), supplemented with one Glycerol FeedBeads Genomic DNA Purification Kit from Promega (WI, ® United States). Afterwards, 0.5 µg of genomic DNA were restricted using the restriction enzymes BamHI-FD and (SMFB12001, Kuhner Shaker GmbH, Germany). This FeedBead releases 40 mg of glycerol in 48 h. The cultures were inoculated ® EcoRI-FD from Thermo Fisher Scientific (MA, United States). with 50 µL of the overnight saturated cultures. The falcon tubes Subsequently, the genomic DNA was diluted to a concentration were incubated in an incubator shaker at 200 rpm and 30°C for of 1 ng μL−1. 48 h. Each clone was tested in triplicate. A triplicate control was Second, a master mix of 22.5 µL was prepared with the forward primer at 0.4 µM, the reverse primer at 0.2 µM, and the restricted ® performed by adding one FeedBead to a falcon with 5 mL of BM medium. These controls were used to determine the actual release genomic DNA at 0.08 ng μL−1. Afterwards, the master mix was of glycerol under the tested conditions. mixed with 22.5 µL of EvaGreen 2X master solution and was Culture samples were centrifuged at 12,000 g for 5 min and thoroughly mixed by vortexing. Following droplet generation filtered through a 0.20 µm pore size syringe filter (SLLGX13NK, using a Droplet generator (Biorad, CA, United States), the Merck Millipore). The 3-HP was quantified using HPLC as droplets were transferred to a 96-tubes rack and subsequently described below. sealed with a PCR Plate Heat Seal. The PCR was carried using a C1000 Touch Thermal Cycler (Biorad, CA, United States). The Mini Bioreactors Screening annealing temperature was set to 60°C. Finally, the ddPCR results The automated cultivation and sampling platform described were analysed using a QX200 Droplet Digital PCR system elsewhere was used (Heux et al., 2014). (Biorad, CA, United States). The bioreactor medium contained 2.5 g L−1 glycerol, 1.8 g L−1 The primers used for ddPCR are listed in Table 2. The primer citric acid, 0.02 g L−1 CaCl2 · 2 H2O, 12.6 g L−1 (NH4)2HPO4, pairs ddPCR_mcrCCa_FW and ddPCR_mcrCCa_RV (Table 2) 0.5 g L−1 MgSO4 · 7 H2O, 0.9 g L−1 KCl, 50 μL antifoam Glanapon were used to amplify the mcr-CCa gene. Primers ddPCR_act_FW 2000 kz (Bussetti and Co., GmbH, Vienna, Austria), 0.4 mg L−1 and ddPCR_act_RV were used to amplify the β-actin (ACT1) biotin and 4.6 mL L−1 of PTM1 trace salts (Maurer et al., 2006). endogenous gene, which was used as reference of a single- The pH was set to 5 using 1 M HCl. The medium was autoclaved copy gene. without the trace salts and the biotin, which were filter sterilized and added to the medium under sterile conditions. Each mini 24-Deep-Well Plates Screening bioreactor was filled with 15 mL of medium. The pre-inoculum P. pastoris strains were inoculated into 50 mL falcon tubes was prepared as described for the other two screening methods containing 5 mL of YPG (1% yeast extract, 2% peptone and 1% v/v glycerol) supplemented with 100 μg mL−1 zeocin (deep-well plates and falcon tubes with FeedBeads ). The overnight-saturated cultures were used to inoculate 250 mL ® (InvivoGen, CA, United States). The cells were grown shake flasks with 25 mL of YPG at a starting OD600 of 0.5–1.5. Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 4 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris The cells were grown for 8 h at 28°C and 160 rpm in a shaker µ[X(t0 )V(t0 )] [µ(t−t0 )] F(t)  e (1) incubator to ensure the cells were at the exponential phase. Each YX/S S0 bioreactor was inoculated at a starting OD600 of 0.025. Each strain X(t0 )  YX/S Sbatch (2) was tested in triplicate. The temperature of the bioreactors was set to 28°C. The Therefore, the flow rate of the feeding medium was set using dissolved oxygen (DO) and the pH were monitored Eq. 3. throughout the cell culture. Samples were automatically withdrawn from the bioreactors every 2 h for the first 16 h of µSbatch V(t0 ) [µ(t−t0 )] F(t)  e (3) culture to monitor the OD600. After 16 h, samples were S0 withdrawn hourly during 8 h for both OD600 monitoring and supernatant analysis. The 250 µL samples for culture supernatant The feeding started once the DO increased above 60%, analysis were automatically placed on 96-well plates with a indicating that all the substrate of the batch phase had been 0.45 µm pore size filter bottom. The samples were filtered by consumed. Samples were collected for biomass cell dry applying vacuum. Glycerol, 3-HP, and D-arabitol were quantified weight (CDW) determination and supernatant analysis. using NMR as described below. The CDW was quantified in triplicates. From 0.5 to 2 mL The growth rate (µmax), the specific substrate consumption of culture were filtered through pre-weighted glass rate (qSGlyc), and the specific 3-HP and D-arabitol production microfiber filters (APFF04700, Merck Millipore). The rates (qP3HP and qPAbt) were calculated during the exponential filters were then washed with 10 mL of distilled water with growth phase using the R program PhysioFit (Peiro et al., 2019). 9 g L −1 NaCl and dried overnight at 105 °C. The filters containing the dry biomass were weighted to calculate Fed-Batch Cultures in Bioreactors the CDW. The cultures were carried out using a DASGIP Parallel Bioreactor Culture samples were centrifuged 5 min at 12,000 g and the System (Eppendorf, Germany). The starting volume of each 1.3 L supernatant was filtered with a 0.2 µm pore size syringe filter reactor vessel was 400 mL. The Batch Medium described for the (SLLGX13NK, Merck Millipore). The filtered supernatant was mini bioreactors with 40 g L−1 of glycerol was used. All the analysed with HPLC for glycerol, 3-HP, and D-arabitol medium components except the biotin and the trace salts were quantification. mixed, placed into the reactor, and autoclaved. Biotin and trace salts were added through the septum port after autoclaving the HPLC Analysis reactor. The pH was controlled at 5 using 15% ammonia (only HPLC was used to quantify the glycerol, the D-arabitol, and base addition was used). Aeration was set to 1 vvm (0.4 L min−1). the 3-HP from the supernatant of the deep-well plates, the The DO was set to 30%, and it was automatically controlled using the following cascade: 1) Increasing the stirring rate from 400 to ® FeedBeads screening, and the fed-batch bioreactor samples. A previously described HPLC protocol was used (Fina et al., 1,200 rpm; 2) Compressed air was mixed with pure oxygen to 2021). The 3-HP was quantified from the UV spectrum. increase the percentage of oxygen of the inlet gas. The reactors Glycerol and D-arabitol were quantified from the Refractive were inoculated at a starting OD600 of 1. The inocula Index (RI) spectrum. As 3-HP and glycerol have the same were prepared as described elsewhere (Garcia-Ortega et al., 2013). retention time and 3-HP is also detected on the RI detector, the The feeding medium composition was 400 g L−1 glycerol, area corresponding to the 3-HP (previously quantified from 10 g L−1 KCl, 6.45 g L−1 MgSO4 · 7 H2O, 0.35 g L−1 CaCl2 · the UV signal) was subtracted to latter calculate the actual 2 H2O, 0.2 mL L−1 antifoam Glanapon 2,000 kz, 1.2 mg L−1 glycerol concentration. biotin and 15 mL L−1 PTM1 trace salts. All the feeding medium components except the biotin and the trace salts were NMR Analysis mixed and autoclaved. Biotin and trace salts were filter-sterilized Glycerol, D-arabitol, and 3-HP were quantified from the and added to the feeding medium under sterile conditions. culture supernatants of the mini bioreactor cultures using Feeding medium was added to the bioreactor using the 1D-1H Nuclear Magnetic Resonance (NMR) on a Bruker exponentially increasing feed-rate described by Eq. 1, where Advance III 800 MHz spectrometer (Bruker BioSpin, F(t) is the feeding rate at every time, t0 is the time where the Germany). Prior to the analyses, 180 µL of filtered culture feeding starts (end of the batch phase), X (t0) is the biomass supernatant samples were mixed with 20 µL of 10 mM TSP (3- concentration at the end of the batch phase, V (t0) is the starting (trimethylsilyl)-[2,2,3,3-2H4]-propionic acid sodium salt), volume of the reactor (the volume for the batch phase), YX/S is the which was used as an internal standard. The NMR biomass to substrate yield, and S0 is the concentration of substrate spectrometer was coupled to a 5 mm CQPI cryoprobe, in the feeding medium. Eq. 1 was simplified by assuming YX/S which was set to 280 K. A 30° presaturation pulse was would not vary significantly between the batch and the fed-batch recorded, followed by a relaxation delay of 7 s. The software phase (Eq. 2). In Eq. 2, Sbatch is the concentration of substrate for TopSpin 3.6.4 (Bruker BioSpin, Germany) was used for the the batch medium (40 g L−1). integration of the peaks. Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 5 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris FIGURE 2 | 3-HP production yield of the recombinant P. pastoris strains tested in deep-well plates. The genetic modifications performed to each strain are shown at the left side of the graph. The product yield in Cmol−1 is shown at the right side of the graph. RESULTS AND DISCUSSION HP production in S. cerevisiae (Kildegaard et al., 2016). In S. pombe, the addition of a second copy of mcr-CCa led to an Increasing mcr-CCa Copy Number Leads to almost 3-fold improvement in the 3-HP yield. Addition of a Higher 3-HP Production third copy of mcr-CCa led to a further 2-fold improvement in 3- Previous results showed that the co-overexpression of ACC1 HP production, while the addition of a fourth copy of the and cPOS5 in the initial 3HP-producing strain PpHP2 same gene, or the addition of a second copy mcr-NCa did (Table 1) led to a small but significant increase in the 3-HP not have an impact on the final 3-HP titre (Takayama et al., yield on glycerol in strain PpHP6 (from 0.131 Cmol Cmol−1 to 2018). 0.146 Cmol Cmol−1) (Fina et al., 2021). However, the Nevertheless, the increase in 3-HP production achieved overexpression of either ACC or cPOS5 did not result into in P. pastoris when additional copies of the mcr-C Ca gene any further significant impact on 3-HP production, were introduced was significantly lower than the one highlighting the malonyl-CoA reductase step as a major achieved in other yeasts. Notably, the MCR specific bottleneck for 3-HP production, rather than the supply of activity of cell extracts of P. pastoris expressing mcr Ca malonyl-CoA and NADPH availability. under the control of pGAP was more than one order of Thereby, a second copy of the C-terminal domain of the magnitude higher than the same gene expressed under the malonyl-CoA reductase gene (mcr-CCa) under the control of control of pTEF1 in S. cerevisiae (Chen et al., 2014; Fina the strong and constitutive GAP promoter was added to the et al., 2021). Additional bottleneck(s) upstream of the PpHP2 and PpHP6 strains, generating strains PpHP7 and malonyl-CoA reductase reaction steps are probably PpHP8, respectively. For each strain, six independently limiting the 3-HP production, so no further copies of isolated clones were tested in triplicate using deep-well mcr-C Ca were introduced. plates and BMG medium. The highest producer clone for strains PpHP7 and PpHP8 produced 1.81 ± 0.02 g L−1 and 2.29 ± 0.01 g L−1 of 3-HP (i.e., 0.143 ± 0.001 and 0.180 ± Expression of the de-regulated ACC1S1132A 0.001 Cmol−1 yield on glycerol, respectively). The gene copy Mutant Does Not Improve 3-HP Production analysis showed that both clones contained 2 copies of mcr- The acetyl-CoA carboxylase (Acc) catalyses the conversion of CCa. The 3-HP yield achieved by the strains PpHP2 and acetyl-CoA into malonyl-CoA. The enzyme Acc1S1132A from P. PpHP6 was 0.130 ± 0.005 and 0.146 ± 0.004 Cmol−1, pastoris contains a site mutation (S1132A) that prevents the respectively. Therefore, the addition of a second copy of post-translational inactivation of the enzyme by mcr-CCa to PpHP2 and PpHP6 led to a 1.38 and 1.23-fold phosphorylation when glucose is depleted (Choi and da improvement, respectively (p-values of 0.01 and 0.0001), as Silva, 2014). Expression of this endogenous ACC1 variant shown in Figure 2. under the control of the GAP promoter showed a positive These results are consistent with previous studies with other impact on the production of acetyl-CoA derived drugs in P. yeasts such as S. cerevisiae or Schizosaccharomyces pombe. The pastoris (Liu et al., 2019). The same approach was used in use of a multicopy integrative vector containing the mcrCa and PpHP7, but the new strain (PpHP9) produced less 3-HP than the ACC1 gene from S. cerevisiae led to a 3-fold increase in 3- the parental strain (0.138 ± 0.001 Cmol−1, Figure 2). Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 6 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris Acc1 is organized as a homodimer. The phosphorylation of conversion of 3-HP into the toxic compound 3- one of the two monomers inactivates the enzymatic complex hydroxypropanaldehyde (3-HPA) (Kildegaard et al., 2014). (Hunkeler et al., 2016). Therefore, it is plausible that the co- Thus, it is plausible that ALD6 overexpression in 3-HP existence of the heterologous mutated version of Acc1 and the producing P. pastoris strains leads to an increased endogenous non-mutated versions leads to the formation of accumulation of 3-HPA. inactive heterodimers, hindering the effect of Acc1S1132A Finally, the overexpression of endogenous pyruvate overexpression. This hypothesis is supported by previous decarboxylase (PDC1) was tested. To this end, an additional studies showing that replacement of the endogenous ACC1 copy of PDC1 under the control of its endogenous promoter was sequence with ACC1S1132A was required to ensure that Acc1 added to the genomes of the strains PpHP8 and PpHP13. actvity would only be controlled by Acc1S1132A expression levels Addition of a second copy of PDC1 was coupled to the (Wen et al., 2020). Deletion of the endogenous ACC1 gene in deletion of the ArDH gene, which is reported as the main strains PpHP9 may thus had a positive effect on 3-HP responsible of the D-arabitol production in P. pastoris (Melo production. Moreover, it is well reported that Acc1 is a highly et al., 2020). To distinguish the effect of the deletion of ArDH regulated enzyme (Hunkeler et al., 2016; Chen et al., 2018). In from the overexpression of PDC1, strains PpHP15 and addition to its post-translational regulation, this enzyme is also PpHP17 were generated by deleting ArDH from strains regulated by allosteric inhibition (Goodridge, 1972; Qiao et al., PpHP8 and PpHP13, respectively (without an additional 2015). Differences in the kinetic parameters between the two copy of PDC1). The strains PpHP16 and PpHP18 were also Acc1 variants might also explain the differences in the final 3-HP obtained from PpHP8 and PpHP13, and they harbour both the yield. deletion of ArDH and a second copy of the endogenous PDC1. In contrast, the overexpression of ACC1 from Y. lipolytica in P. None of the new strains produced more 3-HP than their pastoris led to a small increase in 3-HP production (Fina et al., parental strains (Figure 2). 2021). The yield in PpHP8 (harboring Acc1Yl) was around 1.25- fold higher than in PpHP7. These results are consistent with Screening Under Substrate-Limiting recent studies where the heterologous expression of ACC1 from Y. lipolytica yielded a higher accumulation of malonyl-CoA than Conditions Yields Different Ranking of the overexpression of the endogenous S. cerevisiae ACC1 (Pereira Strains Compared to Conventional et al., 2022). Substrate-Excess Screening Strategies Previous results obtained in deep-well plates showed that the overexpression of the cytosolic acetyl-CoA production pathway Metabolic Engineering for Increased did not result in a significant increase in the final 3-HP Acetyl-CoA Supply and Minimization of By- production. Moreover, accumulation of D-arabitol was not Product Formation observed in none of the strains, as was the case for The endogenous cytosolic acetyl-CoA biosynthesis pathway was PpHP6 strain in fed-batch cultivations (Fina et al., 2021). overexpressed to pull the conversion of pyruvate into acetyl-CoA. Thus, the effect of ArDH deletion could not be properly To this end, a modified Acetyl-CoA synthase from Salmonella assessed at small scale. Therefore, a representative subset of enterica was used (acsSeL641P). The gene harbours a mutation the strains constructed in this study were further cultured in (L641P) to suppress the post-translational inhibition of the two additional systems: 1) Mini bioreactors, favouring fully enzyme by acetylation (Shiba et al., 2007). Expression of aerobic conditions and easy withdrawal of multiple samples. 2) acsSeL641P was tested with or without the co-expression of the ® Falcon tubes containing FeedBeads , which release glycerol at a aldehyde dehydrogenase gene ALD6 from S. cerevisiae. The two low and constant rate (40 mg in 48 h), thereby permitting cell genes were expressed under the control of the TEF1 and growth at a low rate, i.e., under substrate-limiting conditions MDH3 promoters, respectively, which are two mid-to-high resembling a fed-batch process. expression constitutive promoters (Prielhofer et al., 2017). Use The mini bioreactors were sampled during the exponential of pGAP was dismissed to avoid the dilution of the expression of phase, which allowed to accurately determine the µmax and the the other genes, as there are already 5 heterologous genes under q-rates of the consumption of glycerol and the production of 3- the control of pGAP (Marx et al., 2009; Aw and Polizzi, 2013). HP and D-arabitol of the different strains, whereas endpoint Expression of acsSe in strains PpHP7 and PpHP8, generating samples (at 48 h) were taken from cultivations in falcon tubes strains PpHP11 and PpHP13, led to opposite effects. While PpHP11 produced significantly more 3-HP (0.153 ± ® containing FeedBeads . Notably, comparison of the results from these alternative 0.002 Cmol−1 yield) than PpHP7 (p-value of 0.0003), its screening conditions with the deep-well plate cultivations expression in PpHP13 led to a reduction of the yield (p-value (Figure 3) reveals that the ranking of the strains based on 0.0006), as shown in Figure 2. their product yield varies depending on the screening The expression of the aldehyde dehydrogenase ALD6 in both condition. PpHP8 exhibited the highest product yield for the PpHP11 and PpHP13 (PpHP12 and PpHP14, respectively) led to three screening conditions. However, the combined effect of ACS a decrease in the 3-HP yield for both strains (Figure 2). and Pdc1 overexpression and ArDH deletion (strain PpHP18) Interestingly, the deletion of ALD6 in S. cerevisiae led to a was clearly growth condition dependent. Indeed, the 3-HP reduced susceptibility to 3-HP, as Ald6 might catalyse the production of the two strains, PpHP8 and PpHP18 was very Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 7 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris FIGURE 3 | Comparison of the most relevant 3-HP producing P. pastoris strains under three different screening conditions. (A) 3-HP production yield in deep-well plates (grey solid bars), FeedBeads® (blue dotted bars), and mini bioreactors (orange striped bars). (B) Growth rate (yellow vertically striped bars) and arabitol production yield (green horizontally striped bars) for each strain grown in mini bioreactors. significantly at µmax (i.e., in the mini bioreactor experiment). ® similar at a low growth rate (i.e., using FeedBeads ), but differed the product yield of strains PpHP8 and PpHP18 are almost identical when the cells are grown at a low growth rate (i.e., using Interestingly, 3-HP yields were generally higher in deep-well plates and falcon tubes, where oxygen transfer is usually ® FeedBeads ). This is also reinforced by the differential D-arabitol production of strains PpHP8 and PpHP13. The presence of arabitol considered as rate-limiting during cultivation, than in fully denotes a redox imbalance (Fina et al., 2021). When the delivery of aerobic mini bioreactor cultures. This is consistent with the acetyl-CoA is improved by overexpressing ACS (PpHP13), the fact that the pGAP promoter is upregulated in P. pastoris cells D-arabitol production decreases. growing on glucose under hypoxic conditions (Baumann et al., Notably, the mini bioreactors data reveals that the deletion of 2011). ArDH from strain PpHP8 (strain PpHP15) resulted in a Moreover, strains PpHP13, PpHP15, PpHP17, and decrease in the D-arabitol yield and an improvement of the PpHP18 showed a much lower 3-HP product yield than µmax. Production of D-arabitol from glycerol results in net ATP PpHP8 when the cells were grown in mini bioreactors. consumption. Thus, the reduction in the D-arabitol production Conversely, the µmax of strain PpHP8 was the lowest of this leads to higher energetic yield from the substrate, as well as a subset of stains. An inverse correlation between µmax and 3-HP better carbon conservation (as less by-product D-arabitol is product yield was observed. Despite lower growth rates would be produced), thereby potentially increasing acetyl-CoA supply generally expected for the strains with a larger number of and supporting higher growth rates. However, this heterologously overexpressed genes, overexpression of ACS and might result in reduced Acetyl-CoA availability for 3-HP Pdc1 and the deletion of ArDH led to a higher µmax. production. Acc1 overexpression has been reported to cause growth Altogether, the observed results indicate the supply of impairment in yeast (Shi et al., 2014; Kildegaard et al., 2016). cytosolic acetyl-CoA as a major limiting factor to further This reduction in the growth rate was attributed to a metabolic increase the 3-HP yield. burden. The fact the cell growth improves when the cytosolic acetyl- CoA pathway is overexpressed suggests that the lower growth rate of Fed-Batch Cultures PpHP8 might be caused by a limitation in acetyl-CoA supply for the biosynthetic reactions. Such hypothesis is supported by the fact that ® The use of FeedBeads mimics the conditions at a fed-batch culture, where cells grow under substrate-limiting conditions, Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 8 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris sustain the growth rate in fed-batch cultivations at a pre-set µ of 0.10 h−1 when 3-HP concentrations reached concentrations above 15 g L−1, leading to the accumulation of glycerol and D-arabitol in the reactor medium (Fina et al., 2021). This phenomenon is still observed in PpHP8 strain growing at 0.075 h−1, where 17.41 g L−1 of glycerol and 6.01 g L−1 of D-arabitol accumulated into the reactor, while no glycerol nor D-arabitol accumulation were observed for PpHP18. For this reason, the PpHP18 cultivation was finalised just after stopping the addition of the feeding. Overall, the fed-batch for PpHP18 lasted 52 h, while the fed-batch of PpHP8 lasted 56.5 h. Thus, the volumetric productivity of PpHP18 was significantly higher than the volumetric productivity of PpHP8 (0.712 ± 0.010 g L−1 h−1 compared to 0.627 ± 0.033 g L−1 h−1). The overall product yield on glycerol was 0.194 ± 0.004 Cmol−1 for PpHP18 and 0.186 ± 0.010 Cmol−1 for PpHP8. The productivity obtained for the strain PpHP18 is the highest productivity reported in yeast, and, to our knowledge, it is also the highest productivity reported using the malonyl-CoA pathway to produce 3-HP in any microorganism. Furthermore, 37.05 g L−1 is FIGURE 4 | Profiles of fed-batch cultivations of the PpHP8 (A) and the highest 3-HP concentration reported in yeast, highlighting PpHP18 (B) strains. Concentration of glycerol, biomass, 3-HP, and D-arabitol the potential of the combined use of the P. pastoris cell factory and are represented using the left-side y-axis. The total amount of glycerol added glycerol as feedstock for 3-HP production. Notably, the to the reactor, normalized by the actual volume of the reactor at every metabolically engineered PpHP18 strain showed a 50% time is represented using the y-axis at the right side. The average result of two independent cultivations is shown. The error bars correspond to the standard increase in the final 3-HP titre and yield, and a 31.9% increase deviation for the duplicate. in volumetric productivity compared to the initial strain (PpHP6). i.e., at growth rates below the µmax. The strains PpHP8 and PpHP18, which showed the highest 3-HP yields in FeedBeads cultures, were thus further tested in a fed-batch ® CONCLUSION bioreactor culture using a pre-programmed exponential Production of 3-HP in P. pastoris through the malonyl-CoA glycerol feeding strategy to maintain the growth rate at reductase has been reported. The aim of this work was to improve 0.075 h−1, which is approximately half the value of the µmax its production by several metabolic engineering strategies. of the two strains (0.133 h−1 and 0.165 h−1, respectively). The Improvement of the 3HP yield was obtained by adding a feeding phase finished when the equivalent to 195 g L−1 of second copy of mcr-CCa . However, 1) increasing the glycerol were added to the reactor, to compare the results with availability of malonyl-CoA by overexpressing a post- our previous experiments where the same overall amount of translationally unregulated Acc1 from P. pastoris; 2) substrate was used (Fina et al., 2021). Thereafter, the overexpressing endogenous acetyl-CoA pathway (ACS, Ald6, cultivation was finalised when the DO signal increased, and Pdc1) or 3) limiting the production of arabitol by deleting indicating the depletion of any residual glycerol that might ArDH did not improve 3-HP production in small-scale cultures. have been accumulated during the late phase of the fed-batch Nevertheless, deletion of ArDH improved the 3-HP production in culture. The cultivation profiles for both strains are shown in fed-batch bioreactor cultures. Figure 4. The work of this study demonstrates that the system and the At the end of the batch phase, PpHP8 produced 4.92 ± mode of cultivation clearly affected the phenotype of the 0.17 g L−1 3-HP, while the strain PpHP18 produced 4.11 ± strains. For instance, the strain PpHP18 showed the 0.17 g L−1 3-HP. These results agree with the observations from highest production of 3-HP in fed-batch cultivation mode the mini bioreactors, where the yield of PpHP8 was higher than but not in deep-well plates and mini bioreactors operated in the yield of PpHP18. However, 3-HP concentration at the end batch. This provides clear evidence on the importance to of the fed-batch culture of PpHP18 reached 37.05 ± 0.73 g L−1, implement robust and reliable small-scale cultivation while the PpHP8 strain reached 35.40 ± 1.85 g L−1. Thus, the methods allowing for the mid/high throughput yield during the feeding phase was higher for PpHP18 than characterisation of strain performance under bioprocess-like PpHP8. Moreover, the fact PpHP18 µmax is higher than the one conditions. However, the results obtained with the of PpHP8 resulted in a shorter batch phase (22 h compared to 24 h). Previous studies revealed that P. pastoris cannot ® FeedBeads were similar to the ones obtained in fed-bath cultures, thereby validating this methodology as a reliable Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 9 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris method for strain screening/ranking and fast translation to bioreactor-scale. ® FeedBeads and the fed-batch experiments. AF and SH performed the mini bioreactors screening. AF drafted the Overall, our study showed that the multiplexed screening manuscript. All authors critically reviewed the manuscript. methodologies allowed to improve the information content, thereby supporting the formulation of hypotheses of previously unidentified metabolic bottlenecks such as that the FUNDING supply of cytosolic acetyl-CoA may be limiting 3-HP production. Nonetheless, further studies, e.g., using 13C-based metabolic flux This work was supported by project PID2019-104666GB- analysis and metabolomics are needed to design novel metabolic 100 from the Spanish Ministry of Science and Innovation; engineering strategies. 2017-SGR-1462 from the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR) of the Catalan Government. AF was supported with a FPU DATA AVAILABILITY STATEMENT scholarship (FPU17/05434) from the Spanish Ministry of Science and Innovation and a Short-Term Fellowship The original contributions presented in the study are included in (8853) from the European Molecular Biology Organization the article/Supplementary Material, further inquiries can be (EMBO). directed to the corresponding author. SUPPLEMENTARY MATERIAL AUTHOR CONTRIBUTIONS The Supplementary Material for this article can be found online at: AF, JA, and PF conceived and designed the research project. AF https://www.frontiersin.org/articles/10.3389/fbioe.2022.942304/ constructed all strains, performed the deep-well plates, the full#supplementary-material Hydroxypropionic Acid. Biotechnol. Adv. 36, 1207–1222. doi:10.1016/j. REFERENCES biotechadv.2018.03.020 della Pina, C., Falletta, E., and Rossi, M. (2011). A Green Approach to Chemical Aw, R., and Polizzi, K. M. (2013). Can Too Many Copies Spoil the Broth? Microb. Building Blocks. The Case of 3-Hydroxypropanoic Acid. Green Chem. 13, Cell Fact. 12, 128. doi:10.1186/1475-2859-12-128 1624–1632. doi:10.1039/c1gc15052a Baumann, K., Dato, L., Graf, A. B., Frascotti, G., Dragosits, M., Porro, D., et al. Fang, H., Li, D., Kang, J., Jiang, P., Sun, J., and Zhang, D. (2018). Metabolic (2011). The Impact of Oxygen on the Transcriptome of Recombinant S. Engineering of Escherichia coli for De Novo Biosynthesis of Vitamin B12. Nat. cerevisiae and P. pastoris - A Comparative Analysis. BMC Genomics 12, 218. Commun. 9, 4917. doi:10.1038/s41467-018-07412-6 doi:10.1186/1471-2164-12-218 Fina, A., Brêda, G. C., Pérez-Trujillo, M., Freire, D. M. G., Almeida, R. V., Albiol, J., Borodina, I., Kildegaard, K. R., Jensen, N. B., Blicher, T. H., Maury, J., Sherstyk, S., et al. (2021). Benchmarking Recombinant Pichia pastoris For 3- et al. (2015). Establishing a Synthetic Pathway for High-Level Production of 3- Hydroxypropionic Acid Production from Glycerol. Microb. Biotechnol. 14, hydroxypropionic Acid in Saccharomyces cerevisiae via β-alanine. Metab. Eng. 1671–1682. doi:10.1111/1751-7915.13833 27, 57–64. doi:10.1016/j.ymben.2014.10.003 Garcia-Ortega, X., Ferrer, P., Montesinos, J. L., and Valero, F. (2013). Fed-Batch Cámara, E., Albiol, J., and Ferrer, P. (2016). Droplet Digital PCR-Aided Screening Operational Strategies for Recombinant fab Production with Pichia pastoris and Characterization of Pichia pastoris multiple Gene Copy Strains. Biotechnol. Using the Constitutive GAP Promoter. Biochem. Eng. J. 79, 172–181. doi:10. Bioeng. 113, 1542–1551. doi:10.1002/bit.25916 1016/j.bej.2013.07.013 Chemarin, F., Athès, V., Bedu, M., Loty, T., Allais, F., Trelea, I. C., et al. (2019). Gassler, T., Heistinger, L., Mattanovich, D., Gasser, B., and Prielhofer, R. Towards an In Situ Product Recovery of Bio-Based 3-hydroxypropionic Acid: (2019). CRISPR/Cas9-Mediated Homology-Directed Genome Editing in Influence of Bioconversion Broth Components on Membrane-Assisted Pichia pastoris. Methods Mol. Biol. 1923, 211–225. doi:10.1007/978-1- Reactive Extraction. J. Chem. Technol. Biotechnol. 94, 964–972. doi:10.1002/ 4939-9024-5_9 jctb.5845 Goodridge, A. G. (1972). Regulation of The Activity of Acetyl Coenzyme A Chen, X., Yang, X., Shen, Y., Hou, J., and Bao, X. (2018). Screening Carboxylase By Palmitoyl Coenzyme A and Citrate. J. Biol. Chem. 247, Phosphorylation Site Mutations in Yeast Acetyl-CoA Carboxylase Using 6946–6952. doi:10.1016/s0021-9258(19)44677-2 Malonyl-CoA Sensor to Improve Malonyl-CoA-Derived Product. Front. Heux, S., Poinot, J., Massou, S., Sokol, S., and Portais, J.-C. (2014). A Novel Microbiol. 9, 47. doi:10.3389/fmicb.2018.00047 Platform for Automated High-Throughput Fluxome Profiling of Metabolic Chen, Y., Bao, J., Kim, I.-K., Siewers, V., and Nielsen, J. (2014). Coupled Variants. Metab. Eng. 25, 8–19. doi:10.1016/j.ymben.2014.06.001 Incremental Precursor and Co-factor Supply Improves 3-hydroxypropionic Hunkeler, M., Stuttfeld, E., Hagmann, A., Imseng, S., and Maier, T. (2016). The Acid Production in Saccharomyces cerevisiae. Metab. Eng. 22, 104–109. doi:10. Dynamic Organization of Fungal Acetyl-Coa Carboxylase. Nat. Commun. 7, 1016/j.ymben.2014.01.005 11196. doi:10.1038/ncomms11196 Choi, J. W., and da Silva, N. A. (2014). Improving Polyketide and Fatty Acid Jers, C., Kalantari, A., Garg, A., and Mijakovic, I. (2019). Production of 3- Synthesis by Engineering of the Yeast Acetyl-CoA Carboxylase. J. Biotechnol. Hydroxypropanoic Acid from Glycerol by Metabolically Engineered 187, 56–59. doi:10.1016/j.jbiotec.2014.07.430 Bacteria. Front. Bioeng. Biotechnol. 7, 124. doi:10.3389/fbioe.2019.00124 Cregg, J. M., Vedvick, T. S., and Raschke, W. C. (1993). Recent Advances in the Ji, R.-Y., Ding, Y., Shi, T.-Q., Lin, L., Huang, H., Gao, Z., et al. (2018). Metabolic Expression of Foreign Genes in Pichia pastoris. Nat. Biotechnol. 11, 905–910. Engineering of Yeast for the Production of 3-Hydroxypropionic Acid. Front. doi:10.1038/nbt0893-905 Microbiol. 9, 2185. doi:10.3389/fmicb.2018.02185 de Fouchécour, F., Sánchez-Castañeda, A.-K., Saulou-Bérion, C., and Spinnler, H. Kildegaard, K. R., Hallström, B. M., Blicher, T. H., Sonnenschein, N., Jensen, N. B., É. (2018). Process Engineering for Microbial Production of 3- Sherstyk, S., et al. (2014). Evolution Reveals A Glutathione-Dependent Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 10 July 2022 | Volume 10 | Article 942304 Fina et al. 3-HP Production in Pichia pastoris Mechanism of 3-Hydroxypropionic Acid Tolerance. Metab. Eng. 26, 57–66. Applied Synthetic Biology in the Yeast Pichia pastoris. BMC Syst. Biol. 11, 123. doi:10.1016/j.ymben.2014.09.004 doi:10.1186/s12918-017-0492-3 Kildegaard, K. R., Jensen, N. B., Schneider, K., Czarnotta, E., Özdemir, E., Qiao, K., Imam Abidi, S. H., Liu, H., Zhang, H., Chakraborty, S., Watson, N., Klein, T., et al. (2016). Engineering and Systems-Level Analysis of et al. (2015). Engineering Lipid Overproduction in the Oleaginous Yeast Saccharomyces cerevisiae For Production of 3-Hydroxypropionic Acid Yarrowia lipolytica. Metab. Eng. 29, 56–65. doi:10.1016/j.ymben.2015. Via Malonyl-Coa Reductase-Dependent Pathway. Microb. Cell Fact. 15, 02.005 53. doi:10.1186/s12934-016-0451-5 Shi, S., Chen, Y., Siewers, V., and Nielsen, J. (2014). Improving Production of Kim, J. W., Ko, Y. S., Chae, T. U., and Lee, S. Y. (2020). High-Level Production of Malonyl Coenzyme A-Derived Metabolites by Abolishing Snf1-Dependent 3-Hydroxypropionic Acid From Glycerol as a Sole Carbon Source Using Regulation of Acc1. mBio 5, e01130. doi:10.1128/mBio.01130-14 Metabolically Engineered Escherichia coli. Biotechnol. Bioeng. 117, Shiba, Y., Paradise, E. M., Kirby, J., Ro, D.-K., and Keasling, J. D. (2007). 2139–2152. doi:10.1002/bit.27344 Engineering of the Pyruvate Dehydrogenase Bypass in Saccharomyces Kumar, V., Ashok, S., and Park, S. (2013). Recent Advances in Biological cerevisiae For High-Level Production of Isoprenoids. Metab. Eng. 9, Production of 3-Hydroxypropionic Acid. Biotechnol. Adv. 31, 945–961. 160–168. doi:10.1016/j.ymben.2006.10.005 doi:10.1016/j.biotechadv.2013.02.008 Takayama, S., Ozaki, A., Konishi, R., Otomo, C., Kishida, M., Hirata, Y., et al. (2018). Liu, Y., Bai, C., Liu, Q., Xu, Q., Qian, Z., Peng, Q., et al. (2019). Engineered Ethanol- Enhancing 3-Hydroxypropionic Acid Production in Combination With Sugar Supply Driven Biosynthetic System for Improving Production of Acetyl-CoA Derived Engineering by Cell Surface-Display and Metabolic Engineering of Schizosaccharomyces Drugs in Crabtree-Negative Yeast. Metab Eng. 54, 275–284. doi:10.1016/j. pombe. Microb. Cell Fact. 17, 176. doi:10.1186/s12934-018-1025-5 ymben.2019.05.001 Wen, J., Tian, L., Xu, M., Zhou, X., Zhang, Y., and Cai, M. (2020). A Synthetic Luo, X., Ge, X., Cui, S., and Li, Y. (2016). Value-Added Processing of Crude Malonyl-Coa Metabolic Oscillator in Komagataella phaffii. ACS Synth. Biol. 9, Glycerol into Chemicals and Polymers. Bioresour. Technol. 215, 144–154. 1059–1068. doi:10.1021/acssynbio.9b00378 doi:10.1016/j.biortech.2016.03.042 Werpy, T., and Petersen, G. (2004). Top Value Added Chemicals from Biomass Maris, A. J. A. v., Konings, W. N., Dijken, J. P. v., and Pronk, J. T. (2004). Microbial Volume I - Results of Screening for Potential Candidates from Sugars and Export of Lactic and 3-Hydroxypropanoic Acid: Implications for Industrial Synthesis Gas. Golden, CO, USA: U.S. Department of Energy. Available at: Fermentation Processes. Metab. Eng. 6, 245–255. doi:10.1016/j.ymben.2004.05.001 http://www.osti.gov/bridge. Marx, H., Mecklenbräuker, A., Gasser, B., Sauer, M., and Mattanovich, D. (2009). Werten, M. W. T., van den Bosch, T. J., Wind, R. D., Mooibroek, H., and de Wolf, F. Directed Gene Copy Number Amplification in Pichia pastoris by Vector A. (1999). High-Yield Secretion of Recombinant Gelatins By Pichia pastoris. Integration into the Ribosomal DNA Locus. FEMS Yeast Res. 9, 1260–1270. Yeast 15, 1087–1096. doi:10.1002/(sici)1097-0061(199908)15:11<1087::aid- doi:10.1111/j.1567-1364.2009.00561.x yea436>3.0.co;2-f Maurer, M., Kühleitner, M., Gasser, B., and Mattanovich, D. (2006). Versatile Zhao, P., Ma, C., Xu, L., and Tian, P. (2019). Exploiting Tandem Repetitive Modeling and Optimization of Fed Batch Processes for the Production of Promoters for High-Level Production of 3-Hydroxypropionic Acid. Secreted Heterologous Proteins with Pichia pastoris. Microb. Cell Fact. 5, 37. Appl. Microbiol. Biotechnol. 103, 4017–4031. doi:10.1007/s00253-019- doi:10.1186/1475-2859-5-37 09772-5 Mazzoli, R. (2021). Current Progress in Production of Building-Block Organic Acids by Consolidated Bioprocessing of Lignocellulose. Fermentation 7, 248. Conflict of Interest: The authors declare that the research was conducted in the doi:10.3390/fermentation7040248 absence of any commercial or financial relationships that could be construed as a Melo, N. T. M., Pontes, G. C., Procópio, D. P., de Gois e Cunha, G. C., Eliodório, K. potential conflict of interest. P., Costa Paes, H., et al. (2020). Evaluation of Product Distribution in Chemostat and Batch Fermentation in Lactic Acid-Producing Komagataella Publisher’s Note: All claims expressed in this article are solely those of the authors phaffii Strains Utilizing Glycerol as Substrate. Microorganisms 8, 781. doi:10. and do not necessarily represent those of their affiliated organizations, or those of 3390/microorganisms8050781 the publisher, the editors and the reviewers. Any product that may be evaluated in Peiro, C., Millard, P., de Simone, A., Cahoreau, E., Peyriga, L., Enjalbert, B., et al. this article, or claim that may be made by its manufacturer, is not guaranteed or (2019). Chemical and Metabolic Controls on Dihydroxyacetone Metabolism endorsed by the publisher. Lead To Suboptimal Growth of Escherichia coli. Appl. Environ. Microbiol. 85, e00768. doi:10.1128/AEM.00768-19 Copyright © 2022 Fina, Heux, Albiol and Ferrer. This is an open-access article Pereira, H., Azevedo, F., Domingues, L., and Johansson, B. (2022). Expression of distributed under the terms of the Creative Commons Attribution License (CC BY). Yarrowia lipolytica Acetyl-CoA Carboxylase in Saccharomyces cerevisiae and its The use, distribution or reproduction in other forums is permitted, provided the Effect on In-Vivo Accumulation of Malonyl-CoA. Comput. Struct. Biotechnol. J. original author(s) and the copyright owner(s) are credited and that the original 20, 779–787. doi:10.1016/j.csbj.2022.01.020 publication in this journal is cited, in accordance with accepted academic practice. Prielhofer, R., Barrero, J. J., Steuer, S., Gassler, T., Zahrl, R., Baumann, K., et al. No use, distribution or reproduction is permitted which does not comply with these (2017). GoldenPiCS: A Golden Gate-Derived Modular Cloning System for terms. Frontiers in Bioengineering and Biotechnology | www.frontiersin.org 11 July 2022 | Volume 10 | Article 942304