Enhancing the efficiency of the Pichia pastoris AOX1 promoter via the synthetic positive feedback circuit of transcription factor Mxr1
- 438 Downloads
The methanol-regulated AOX1 promoter (PAOX1) is the most widely used promoter in the production of recombinant proteins in the methylotrophic yeast Pichia pastoris. However, as the tight regulation and methanol dependence of PAOX1 restricts its application, it is necessary to develop a flexible induction system to avoid the problems of methanol without losing the advantages of PAOX1. The availability of synthetic biology tools enables researchers to reprogram the cellular behaviour of P. pastoris to achieve this goal.
The characteristics of PAOX1 are highly related to the expression profile of methanol expression regulator 1 (Mxr1). In this study, we applied a biologically inspired strategy to reprogram regulatory networks in P. pastoris. A reprogrammed P. pastoris was constructed by inserting a synthetic positive feedback circuit of Mxr1 driven by a weak AOX2 promoter (PAOX2). This novel approach enhanced PAOX1 efficiency by providing extra Mxr1 and generated switchable Mxr1 expression to allow PAOX1 to be induced under glycerol starvation or carbon-free conditions. Additionally, the inhibitory effect of glycerol on PAOX1 was retained because the synthetic circuit was not activated in response to glycerol. Using green fluorescent protein as a demonstration, this reprogrammed P. pastoris strain displayed stronger fluorescence intensity than non-reprogrammed cells under both methanol induction and glycerol starvation. Moreover, with single-chain variable fragment (scFv) as the model protein, increases in extracellular scFv productivity of 98 and 269% were observed in Mxr1-reprogrammed cells under methanol induction and glycerol starvation, respectively, compared to productivity in non-reprogrammed cells under methanol induction.
We successfully demonstrate that the synthetic positive feedback circuit of Mxr1 enhances recombinant protein production efficiency in P. pastoris and create a methanol-free induction system to eliminate the potential risks of methanol.
KeywordsPichia pastoris Recombinant protein expression AOX1 promoter Methanol Methanol expression regulator 1 Synthetic gene circuit Transcriptional reprogramming
Buffered dextrose-complex medium
Buffered glycerol-complex medium
Buffered methanol-complex medium
Buffered complex medium without a carbon source
green Fluorescent protein
Mxr1-reprogrammed KM71 strains
Methanol expression regulator 1
Alcohol oxidase 1 promoter
Single-chain variable fragment
Yeast extract-peptone-dextrose with zeocin
The methylotrophic yeast Pichia pastoris has been extensively used in the production of recombinant proteins because it provides the advantages of post-translational modification in a eukaryotic single-cell system. Protein production in P. pastoris is typically driven by the AOX1 promoter (PAOX1), which occurs in response to methanol induction due to its strong and regulatable characteristics [1, 2]. To date, more than 5000 recombinant proteins have been successfully produced in P. pastoris [3, 4]. Despite the potential of the P. pastoris expression system, tightly regulated PAOX1 limits expression to restrictive conditions, with the presence of repressing carbon sources significantly decreasing recombinant protein expression during the methanol-induction phase . Residual carbon sources can be removed by medium replacement prior to methanol induction, though this process is not applicable in large-scale production . In addition, methanol is a toxic and flammable compound that presents some potential problems as the inducer or carbon source [7, 8, 9, 10, 11].
The development of synthetic biology tools has enabled researchers to reprogram cellular behaviour in P. pastoris to avoid the drawbacks of tight PAOX1 regulation. The use of alternative promoters or depressed PAOX1 variants both provided solutions for methanol-independent production [12, 13, 14]. The improvement of induction efficiency of PAOX1 under non-methanol carbon sources was achieved by reprogramming the carbon metabolic pathway of P. pastoris [15, 16]. However, such a strategy might interrupt carbon metabolism and result in growth defects. An alternative approach is to reprogram transcriptional regulation of PAOX1. Although the regulation mechanism of PAOX1 is not fully understood, several transcription factors involved in PAOX1 regulation have been identified. In response to different carbon sources, PAOX1 has three regulated stages of gene expression including repression, derepression, and activation . Among carbon regulation of PAOX1, the transcriptional activator Mxr1 is constitutively expressed and plays a crucial role in PAOX1 derepression and activation processes [18, 19, 20]. The Nrg1 repressor participates in the inhibition mechanism by competing for Mxr1 binding elements in PAOX1 ; Nrg1 can be down-regulated by switching the carbon source from glycerol to methanol , though the detailed expression pattern remains unclear. During the activation process, the activators Prm1 and Mit1 are up-regulated by methanol to activate PAOX1 expression [19, 20]. In previous studies, the PAOX1-based methanol free expression system could be achieved by deletion of three repressors (Nrg1, Mig1 and Mig2) and overexpression of one activator (Mit1) , or by derepressed overexpression of Mxr1 or Mit1 . However, NRG1 deletion might lead to the potential risk of growth defects . Hence, developing the synthetic circuits to provide an efficient approach for controlling the characteristics of PAOX1, remains a challenge.
In this study, we reprogrammed regulatory networks in P. pastoris through biological inspiration of another methylotrophic yeast: Hansenula polymorpha. Unlike PAOX1 in P. pastoris, glycerol does not interfere with the methanol-induced efficiency of PMOX, the alcohol oxidase promoter in H. polymorpha. In addition to induction by methanol, PMOX can express recombinant genes via a carbon starvation strategy [17, 24]. The regulatory difference between PAOX1 and PMOX results from upstream transcriptional networks in cells rather than due to promoter sequences alone . Interestingly, this phenomenon might be related to the different expression pattern between P. pastoris Mxr1 and its orthologous gene (HPODL00650) in H. polymorpha, as HPODL00650 is up-regulated by methanol . We speculate the existence of positive feedback regulation of HPODL00650 in H. polymorpha, which might contribute to the flexible activation of PMOX1. Therefore, to mimic the expression pattern of HPODL00650, a reprogrammed P. pastoris strain was constructed by inserting a synthetic positive autoregulation circuit of Mxr1. In addition to endogenous Mxr1, exogenous Mxr1 was driven by the methanol-regulated AOX2 promoter (PAOX2), a promoter that is weaker than PAOX1. This novel strategy did not affect cells in the repression condition, thus maintaining tight regulation of PAOX1 and preventing growth defects. We demonstrate herein that the transcriptional efficiency of PAOX1 was enhanced and that the interference due to residual repressing carbon sources was reduced. These Mxr1-reprogrammed cells show great potential for broader applications.
Mxr1-reprogrammed cells had altered GFP expression but did not show growth defects
PAOX1 remained controllable in Mxr1-reprogrammed cells
Positive feedback of Mxr1 increased the transcriptional efficiency of PAOX1 and broke the Mxr1 titration effect
According to the study of Camara et al., the Mxr1 titration effect is a plausible explanation for the observed transcriptional attenuation of methanol-induced genes in increasingly used PAOX1-regulated expression cassettes . To verify whether this Mxr1 titration effect can be disrupted by exogenous expression of Mxr1, the methanol utilization capacities of P. pastoris KM71H, P. pastoris KM71/GFP, and P. pastoris KM71m/GFP cells were determined (Fig. 3c). As evidenced by a colorimetric assay, AOX activity in P. pastoris KM71/GFP cells was weaker than that in P. pastoris KM71H cells in response to methanol. However, the defect in methanol utilization was recovered in P. pastoris KM71m/GFP cells, suggesting that the expression level of Mxr1 was a bottleneck for heterologous gene expression. Hence, breaking the Mxr1 titration effect by a synthetic Mxr1 circuit is expected to increase the potential of using a high copy-number strategy.
Mxr1-reprogrammed cells were inducible under non-restricted conditions
Although the fluorescence intensity of P. pastoris KM71/GFP was enhanced by medium replacement (Fig. 4b), the intensity of P. pastoris KM71m/GFP with or without medium replacement was significantly higher than that of P. pastoris KM71/GFP with medium replacement. These results suggested that Mxr1 reprogramming overcame the interference of residual repressive carbon and resulted in a smooth transition between glycerol and methanol. In addition to the above advantage, Mxr1 reprogramming showed great potential in the development of a methanol-free induction system. As shown in Fig. 4c, the fluorescence intensity of P. pastoris KM71m/GFP increased significantly by the daily addition of 0.33% glycerol, whereas only baseline intensity was detected for P. pastoris KM71/GFP. These results indicated that Mxr1-reprogrammed P. pastoris is inducible under glycerol depletion condition.
Application of Mxr1-reprogrammed cells in the production of scFv
In addition to Mxr1, the role of the repressor Nrg1 was also investigated (Additional file 2: Figure S2). The expression level of Nrg1 was down-regulated in response to methanol. However, there was no significant difference in Nrg1 expression levels between the glycerol group and the no-carbon-source group. It was interesting that the engineered cells could reach high Mxr1 expression level in carbon-free medium without down-regulation of Nrg1. Hence, we speculate the existence of repressor X, which is regulated by glycerol. Under glycerol-starvation conditions, down-regulation of repressor X has a lower inhibitory effect on PAOX1; however, PAOX1 is still repressed in non-reprogrammed cells due to weak constitutive expression of endogenous Mxr1. In contrast, down-regulation of repressor X would generate the different strength of repression between glycerol repression and derepression conditions, which explains why PAOX1 in reprogrammed cells was only activated in glycerol depletion condition but not in glycerol repression condition. Repressor X is likely to be Mig1 or Mig2, as their response to glycerol or methanol is consistent with that of repressor X [3, 22]. Regardless, the detailed expression patterns of Mig1 and Mig2 under glycerol, methanol and carbon depletion require further investigation.
Overexpression or deletion of transcription factors is the simplest way to control PAOX1 expression; however, these strategies might result in cellular growth defects or break the tight regulation of PAOX1 in response to repressive carbon sources [23, 32, 33]. Recently, Vogl et al. converted a PAOX1-based expression strain into a methanol-free production system without breaking the inhibitory effect of glucose through derepressed overexpression of Mxr1 or Mit1 under CAT1 promoter regulation . Interestingly, a methanol-free production system was also achieved in our study by generating switchable Mxr1 expression via a synthetic positive feedback circuit, even though PAOX2 is not a naturally derepressed promoter [17, 34]. Compared with the derepressed overexpression of Mxr1, the positive feedback circuit of Mxr1 showed better improvement of productivity under methanol induction and derepression condition. Meanwhile, extra Mxr1 regulated by weak PAOX2 could prevent the detrimental effects of strong Mxr1 expression . However, both strategies indicate that a synthetic circuit resulting in different levels of transcription factor expression under various conditions is a viable and flexible approach to controlling the characteristics of PAOX1. Although only the MutS strain was used in this study, Vogl et al. reported that the Mut phenotype was independent to the methanol regulation machinery . The strategy of transcriptional reprogramming is expected to be effective on both Mut+ and MutS strains.
In addition to the transcriptional process, the secretion pathway is also a bottleneck of recombinant protein production in P. pastoris [35, 36]. Previous studies have shown that using a weak promoter could lower the production rate of recombinant proteins, which was favorable for the secretion of proteins with complex folding . Mxr1-reprogrammed P. pastoris with the glycerol starvation strategy could achieve the similar goal of slowing down the expression level and get a better productivity of secreted protein. On the other hand, co-expression of target proteins with chaperone protein Kar2 , protein disulphide isomerase (PDI) or transcription factors such as Aft1  and Hac1  could enhance the productivity of certain proteins in P. pastoris . Hence, the combination of Mxr1 reprogramming with other secretion-related gene circuits is necessary to solving the problem of intracellular target protein accumulation.
We demonstrated that the recombinant protein production driven by PAOX1 was greatly enhanced by the synthetic positive feedback circuit of Mxr1 in P. pastoris under methanol induction. Breaking the Mxr1 titration effect by added Mxr1 increased the potential of using a high copy-number strategy. In addition, Mxr1 reprogramming reduced interference from the residual repressing carbon source, allowing a smooth transition between different carbon sources. This platform also provided an alternative approach to expressing target genes driven by PAOX1 under glycerol starvation, thereby eliminating the potential risks of methanol. These Mxr1-reprogrammed cells are expected to have great potential for broader applications.
Plasmids and strains
Strains used in this study
E. coli EPI300
Gene cloning host
Epicentre Technologies Corp, USA
P. pastoris KM71
Gene expression host with histidine deficiency
P. pastoris KM71H
Gene expression host
P. pastoris KM71/GFP
GFP expressed in P. pastoris KM71 and histidine deficiency was restored by the empty pAOX2 vector
In this study
P. pastoris KM71m/GFP
GFP expressed in P. pastoris KM71m containing the synthetic Mxr1 circuit
In this study
P. pastoris KM71H/scFv
scFv expressed in P. pastoris KM71H
In this study
P. pastoris KM71Hm/scFv
scFv expressed in P. pastoris KM71Hm containing the synthetic Mxr1 circuit
In this study
Media and culture conditions
The media used in this study are listed in Additional file 4: Table S2. The protein expression procedure for P. pastoris was conducted according to the manufacturer’s instructions (Invitrogen, Carlsbad, California, USA). Cells were cultured in 3 mL YPDZ for 20 h as seed culture and inoculated into 100 mL BMGY medium to an optical density at 600 nm (OD600) of 0.15. The cells were grown at 30 °C, 250 rpm for 24 h and harvested by centrifugation at 3000×g for 10 min to remove residual repressive carbon sources in the supernatant. The pellet was resuspended in 20 mL of a different carbon source medium, BMMY, BMGY, or BMNY. BMGY was a fresh medium with a different concentration of glycerol. BMMY was a fresh medium with 0.5% methanol added every 24 h to induce protein expression. BMNY was a fresh medium without a carbon source.
Protein expression analysis
Quantification of GFP expression was monitored using a SpectraMax M2e Microplate Reader (Molecular Device, Sunnyvale, California, USA). The excitation wavelength was 488 nm, with an emission wavelength of 509 nm. The GFP expression level was normalized to the cell density, which was monitored by optical density at 600 nm. Extracellular expression of scFv was analysed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE), followed by Coomassie Brilliant Blue G-250 staining. The relative scFv titre was determined using UVP image analysis software (Analytik Jena AG, Jena, Germany). For intracellular protein analysis, cell pellets were harvested by centrifugation and washed once with the same volume of lysis buffer (50 mM monosodium phosphate pH 7.4, 1 mM ethylenediaminetetraacetic acid (EDTA), 5% glycerol, 1 mM phenylmethylsulfonyl fluoride (PMSF)). The cell pellets were resuspended in breaking buffer and ground in liquid nitrogen. Total intercellular proteins were transferred to a polyvinylidene fluoride (PVDF) membrane (PerkinElmer, Waltham, Massachusetts, USA) after electrophoresis and detected by specific antibodies. The primary antibody was a rabbit anti-his polyclonal antibody (Bioman, New Taipei City, Taiwan), and a horseradish peroxidase (HRP)-conjugated antibody (PerkinElmer) was used as the secondary antibody. Both antibodies were diluted 5000× with gelatine-NET (0.15 M NaCl, 5 mM EDTA, 0.05% Tween in 50 mM Tris-HCl, pH 8.0) before use. Colorimetric detection was performed using an enhanced chemiluminescence substrate (PerkinElmer). The total intracellular protein content was assessed using a rabbit anti-GAPDH polyclonal antibody (GeneTex, Irvine, California, USA).
RNA expression level analysis
The mRNA expression levels of transcription factor genes and GFP were verified by real-time PCR. Total RNA derived from cells was extracted using a NautiaZ Bacteria/Fungi RNA Mini Kit (Nautia Gene, Taipei, Taiwan) in accordance with the manufacturer’s procedure. Reverse transcription was performed using ARROW-Script Reverse transcriptase III with Radom hexamers (ARROWTEC, Taipei, Taiwan), and the products were used for subsequent real-time PCR performed with the StepOne™ System (Applied Biosystems, Foster, California, USA) using 2xIQ2 SYBR Green FAST qPCR System Master Mix-HIGH ROX (Bio-Genesis Technologies, Taipei, Taiwan). The primers used for real-time PCR are listed in Additional file 5: Table S3. After the cycle threshold values (CT) were determined, relative fold differences were calculated using the 2−ΔΔCT method with 18S rRNA as the endogenous reference gene.
AOX activity assay
Cells were cultured in 3 mL YPDZ for 16 h as seed culture and inoculated into 3 mL BMDY, BMGY, BMMY, BMNY, or BMGMY medium to an optical density of 1at 600 nm (OD600). The cells were grown at 30 °C, 250 rpm for 10 h, and a total of 5 × 107 cells was harvested by centrifugation. The cell pellets were resuspended in AOX activity reagent and incubated at 30 °C for 30 min .
This work was supported by Ministry of Science and Technology, Taiwan, ROC (MOST-105-2313-B-002-043).
Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information files, including the raw data in spreadsheets (Additional file 6).
CHC and CTH are responsible for the project planning and experimental design. CHC and HAH constructed the strains and performed the transcriptional and expression analyses. KLH contributed to the experimental operation. CHC, KLH and HAH analysed the results and wrote the manuscript. CTH revised and produced the final manuscript. All authors contributed to scientific discussion. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 13.Prielhofer R, Reichinger M, Wagner N, Claes K, Kiziak C, Gasser B, Mattanovich D. Superior protein titers in half the fermentation time: promoter and process engineering for the glucose-regulated GTH1 promoter of Pichia pastoris. Biotechnol Bioeng. 2018.Google Scholar
- 18.Lin-Cereghino GP, Godfrey L, de la Cruz BJ, Johnson S, Khuongsathiene S, Tolstorukov I, Yan M, Lin-Cereghino J, Veenhuis M, Subramani S, et al. Mxr1p, a key regulator of the methanol utilization pathway and peroxisomal genes in Pichia pastoris. Mol Cell Biol. 2006;26(3):883–97.PubMedPubMedCentralCrossRefGoogle Scholar
- 27.Bawa Z, Routledge SJ, Jamshad M, Clare M, Sarkar D, Dickerson I, Ganzlin M, Poyner DR, Bill RM. Functional recombinant protein is present in the pre-induction phases of Pichia pastoris cultures when grown in bioreactors, but not shake-flasks. Microb Cell Factories. 2014;13(1):127.CrossRefGoogle Scholar
- 32.Zhan C, Yang Y, Zhang Z, Li X, Liu X, Bai Z. Transcription factor Mxr1 promotes the expression of Aox1 by repressing glycerol transporter 1 in Pichia pastoris. FEMS Yeast Res 2017;17(4), fox015.Google Scholar
- 34.Vogl T, Sturmberger L, Kickenweiz T, Wasmayer R, Schmid C, Hatzl AM, Gerstmann MA, Pitzer J, Wagner M, Thallinger GG, et al. A toolbox of diverse promoters related to methanol utilization: functionally verified parts for heterologous pathway expression in Pichia pastoris. ACS Synth Biol. 2016;5(2):172–86.PubMedCrossRefGoogle Scholar
- 36.Gasser B, Saloheimo M, Rinas U, Dragosits M, Rodriguez-Carmona E, Baumann K, Giuliani M, Parrilli E, Branduardi P, Lang C, et al. Protein folding and conformational stress in microbial cells producing recombinant proteins: a host comparative overview. Microb Cell Factories. 2008;7:11.CrossRefGoogle Scholar
- 39.Guerfal M, Ryckaert S, Jacobs PP, Ameloot P, Van Craenenbroeck K, Derycke R, Callewaert N. The HAC1 gene from Pichia pastoris: characterization and effect of its overexpression on the production of secreted, surface displayed and membrane proteins. Microb Cell Factories. 2010;9:49.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.