The identification of novel promoters and terminators for protein expression and metabolic engineering applications in Kluyveromyces marxianus

Abstract

The K. marxianus has emerged as a potential yeast strain for various biotechnological applications. However, the limited number of available genetic tools has hindered the widespread usage of this yeast. In the current study we have expanded the molecular tool box by identifying novel sets of promoters and terminators for increased recombinant protein expression in K. marxianus. The previously available transcriptomic data were analyzed to identify top 10 promoters of highest gene expression activity. We further characterized and compared strength of these identified promoters using eGFP as a reporter protein, at different temperatures and carbon sources. To examine the regulatory region driving protein expression, serially truncated shorter versions of two selected strong promoters were designed, and examined for their ability to drive eGFP protein expression. The activities of these two promoters were further enhanced using different combinations of native transcription terminators of K. marxianus. We further utilized the identified DNA cassette encoding strong promoter in metabolic engineering of K. marxianus for enhanced β-galactosidase activity. The present study thus provides novel sets of promoters and terminators as well as engineered K. marxianus strain for its wider utility in applications requiring lactose degradation such as in cheese whey and milk.

Highlights

• •Novel promoters and terminators for constitutive gene expression in K. marxianus.
• •The promoters show constitutive expression at varying temperature and carbon source.
• •K. marxianus strain with improved production of β-galactosidase.

1. Introduction

The use of microorganisms for various biotechnological applications has been greatly expanded from producing beverages to food, enzymes, biofuels, flavors and pharmaceuticals. The increased use of these organisms is primarily due to the advancement of various genetics and molecular biology tools, which has facilitated metabolic engineering of biochemical pathways for desired applications (Lian et al., 2018; Nielsen and Jewett, 2008). The yeast-based system has gained more potential as it is amenable to genetic manipulations and also possesses various biochemical pathways found in higher eukaryotes such as mammalian or plant cells (Chakraborty et al., 2016; Mohammadi et al., 2015). Some of the commonly employed yeasts for research as well as industrial processes are Saccharomyces cerevisiae, Pichia pastoris, Hansenula polymorpha, Kluyveromyces lactis and Yarrowia lipolytica. Recently another yeast Kluyveromyces marxianus, a sister species of K. lactis, has also emerged as a valuable strain for biotechnological applications (Karim et al., 2020; Lane and Morrissey, 2010).

The K. marxianus is a non-conventional, non-pathogenic and GRAS (generally regarded as safe) organism. K. marxianus possesses various desirable properties which is absent in many other commonly used yeast, and thus the strain is a potential substitute for biotechnological needs. It’s a fastest growing organism as compared to any of the other known eukaryotes (Groeneveld et al., 2009). It’s a thermotolerant yeast that could grow up to 52 ​°C (Banat et al., 1992). K. marxianus can assimilate various sugars such as glucose, lactose, mannose, galactose and xylose (Fonseca et al., 2013) (Rodrussamee et al., 2011). The availability of its whole genome sequence could be used for desired genetic manipulations.

In contrast to extensively studied S. cerevisiae, not much is established in K. marxianus. The lack of efficient genetic tools, limited number of auxotrophic markers, limited choice of constitutive or inducible promoters is the major drawback for its metabolic engineering, and usage for biotechnological applications. The presence of dominant nonhomologous end joining (NHEJ) over homologous recombination in K. marxianus is another limitation for construction of specific gene knock-out strain (Abdel-Banat et al., 2010). The limitation has been partially overcome by the development of CRISPR-Cas9 system that has improved the efficiency to construct specific gene knockout strain (Löbs et al., 2017). Similarly, a CRISPR interference (CRISPRi) method based upon inactive Cas9, has been successfully used for simultaneous suppression of multiple genes for improved ethyl acetate production in K. marxianus (Löbs et al., 2018). The K. marxianus has shown its potential usage in production of various biomolecules including proteins, ethanol, oligosaccharides and oligopeptides (Belem et al., 1999; Cheng et al., 2006; Yanase et al., 2010). Various enzymes, such as Inulinase, aminopeptidases, β-galactosidase have been successfully produced in their active state from K. marxianus (Bansal et al., 2008; Ramírez-Zavala et al., 2004; Zhou et al., 2014). To fully exploit the unique properties of K. marxianus, it is desirable to expand the tool-box for carrying out genetic manipulations. The ability to modulate a biochemical pathway by varying the expression level of its enzymes and proteins is required for the development of yeast strain for desired application. Thus the identification of novel promoters that could be used in the presence of various sugars and different temperatures would further enhance the potential of K. marxianus as a substitute for P. pastoris or other yeasts.

Various approaches such as optimization of signal sequences and exploration of strong promoters are currently underway for metabolic engineering applications in K. marxianus. Both types of promoters, native as well as those derived from budding yeast S. cerevisiae have been examined in K. marxianus. Promoters of S. cerevisiae origin used for expression in K. marxianus are of galactokinase (P_ScGAL1) (Almeida et al., 2003), glyceraldehyde-3-dehydrogenase (P_ScTDH3) (Nonklang et al., 2009), 3-phosphoglycerate kinase (P_ScPGK1) (Ball et al., 1999; Pecota et al., 2007) and alcohol dehydrogenase (P_ScADH1) (Lee et al., 2013). Among these promoters, the P_ScGAL1 promoter is galactose regulatable, and other such as P_ScTDH3, P_ScPGK1 and P_ScADH1 are constitutively expressed. Similarly, native promoters of K. marxianus origin such as P_KmINU1 (promoter of inulinase enzyme), P_KmGAL1, P_KmTDH3, P_KmPGK1 and P_KmADH1 have been employed to express recombinant proteins (Akada et al., 2014; Bergkamp et al., 1993; Yang et al., 2015; Zhang et al., 2015). The expression from P_KmINU1 and P_KmGAL1 is carbon source dependent. Furthermore, attempts have been made to improve strength of native promoters such as of P_KmINU1 (Zhou et al., 2018). Studies show that under growth conditions where glucose is used as carbon source, the native promoters of K. marxianus show higher strength than homologous promoter derived from S. cerevisiae paving way for further exploration of native promoters of K. marxianus (Yang et al., 2015).

In addition to promoters, terminators are known to play critical role in protein expression by modulating the stability of mRNA. Only limited number of terminators sequences have been examined in K. marxianus. The different gene-terminators of K. marxianus and S. cerevisiae origin have been used to modulate promoter strength (Rajkumar et al., 2019). These studies suggest that further expansion of number of terminators sequences would provide additional control for gene expression for metabolic engineering applications in K. marxianus.

Overall there are only limited number of promoters as well as terminator sequences available for protein production and metabolic engineering applications in K. marxianus, and thus there is a need to identify novel promoters and terminators for modulation of desirable biochemical pathways. Thus in the current study, we carried out analysis of available transcriptomic profile of K. marxianus, and identified novel promoters for high expression of proteins, enzymes and metabolic engineering applications. Also, using different sets of terminator sequences, we further enhanced promoter strength. Using the identified combination of these promoters and terminators, we expressed HA antigen of influenza virus. We further used the promoter cassette to engineer K. marxianus with improved β-galactosidase production. The study thus provides novel sets of promoters and terminators for high level production of heterologous proteins and thus contributes to novel genetic tools for use of K. marxianus for metabolic engineering and biotechnological applications. The genetic tools developed in the present study would aid to further understanding of cellular biology of K. marxianus.

2. Materials and methods

2.1. Plasmid, strains and growth media

The K. marxianus NBRC1777 was obtained from Biological Resource Centre, NITE (NBRC), Japan. The S. cerevisiae strain BY4741 (MATa his3Δ1, leu2Δ0, met15Δ0, ura3Δ0) was used. The plasmid used in the study are listed in Table 3. SD (Synthetic-defined) media was composed of yeast nitrogen base (YNB) with ammonium sulfate (6.7 ​g/l) and 2% dextrose (w/v). SX media composition is similar to SD media except that 2% (w/v) xylose instead of dextrose provided carbon source. Amino acids were supplemented as required. YPD media is composed of 1% yeast extract, 2% peptone and 2% dextrose (w/v). YPX and YPL media is same as YPD except that 2% xylose or lactose was used instead of dextrose.

Table 3.

Plasmids constructed in the present study.

S. No.	Plasmid	Marker
1.	pPKM316-PKmIMTCP1-eGFP-TScCYC1	URA3
2.	pPKM316-PKmIMTCP2-eGFP-TScCYC1	URA3
3.	pPKM316-PKmIMTCP3-eGFP-TScCYC1	URA3
4.	pPKM316-PKmIMTCP4-eGFP-TScCYC1	URA3
5.	pPKM316-PKmIMTCP5-eGFP-TScCYC1	URA3
6.	pPKM316-PKmIMTCP6-eGFP-TScCYC1	URA3
7.	pPKM316-PKmIMTCP7-eGFP-TScCYC1	URA3
8.	pPKM316-PKmIMTCP8-eGFP-TScCYC1	URA3
9.	pPKM316-PKmIMTCP9-eGFP-TScCYC1	URA3
10.	pPKM316-PKmIMTCP10-eGFP-TScCYC1	URA3
11.	pPKM316-PKmTDH3-eGFP-TScCYC1	URA3
12.	pPKM316-PKmPGK1-eGFP-TScCYC1	URA3
13.	pPKM316-PKmIMTCP1-787-eGFP-TScCYC1	URA3
14.	pPKM316-PKmIMTCP1-612-eGFP-TScCYC1	URA3
15.	pPKM316-PKmIMTCP1-400-eGFP-TScCYC1	URA3
16.	pPKM316-PKmIMTCP1-194-eGFP-TScCYC1	URA3
17.	pPKM316-PKmIMTCP2-699-eGFP-TScCYC1	URA3
18.	pPKM316-PKmIMTCP2-499-eGFP-TScCYC1	URA3
19.	pPKM316-PKmIMTCP2-408-eGFP-TScCYC1	URA3
20	pPKM316-PKmIMTCP2-299-eGFP-TScCYC1	URA3
21.	pPKM316-PKmIMTCP2-199-eGFP-TScCYC1	URA3
22.	pPKM316-PKmIMTCP2-111-eGFP-TScCYC1	URA3
23.	pPKM316-PKmIMTCP2-eGFP-TKmIMTT1	URA3
24.	pPKM316-PKmIMTCP2-eGFP-TKmIMTT2	URA3
25.	pPKM316-PKmIMTCP1-FLuc-TKmIMTT1	URA3
26.	pPKM316-PKmIMTCP1-FLuc-TKmIMTT2	URA3
27.	pPKM316-PKmIMTCP2-FLuc-TKmIMTT1	URA3
28.	pPKM316-PKmIMTCP2-FLuc-TKmIMTT2	URA3
29.	pPKM316-PKmIMTCP1-FLuc-TScCYC1	URA3
30.	pPKM316-PKmIMTCP2-FLuc-TScCYC1	URA3
31.	pPKM316-PKmTDH3-FLuc-TScCYC1	URA3
32.	pPKM316-PKmPGK1-FLuc-TScCYC1	URA3
33.	pPKM316-PKmIMTCP2-ssInu1Km-eGFP-TKmIMTT1	URA3
34.	pPKM316-PKmIMTCP2-ssPho1Pp-eGFP-TKmIMTT1	URA3
35.	pPKM316-PKmIMTCP2-ssαMFSc-eGFP-TKmIMTT1	URA3
36.	pPKM316-PKmIMTCP2-ssαMFKm-eGFP-TKmIMTT1	URA3
37.	pPKM316-PKmIMTCP2-HA-TKmIMTT1	URA3
38.	pPKM316-PKmPGK1-β-Gal-TScCYC1	URA3
39.	pPKM316-PKmTDH3- β-Gal-TScCYC1	URA3
40.	pPKM316-PKmIMTCP2- β-Gal-TKmIMTT1	URA3
41.	pRS316-PScGPD-FLuc-TScCYC1	URA3
42.	pRS316-PKmIMTCP1-FLuc-TScCYC1	URA3
43.	pRS316-PKmIMTCP2-FLuc-TScCYC1	URA3
44.	pRS316-PScGPD-FLuc-TKmIMTT1	URA3
45.	pRS316-PScGPD-FLuc-TKmIMTT2	URA3
46.	pRS316-PKmIMTCP1-FLuc-TKmIMTT1	URA3
47.	pRS316-PKmIMTCP1-FLuc-TKmIMTT2	URA3
48.	pRS316-PKmIMTCP2-FLuc-TKmIMTT1	URA3
49.	pRS316-PKmIMTCP2-FLuc-TKmIMTT2	URA3
50.	pRS316-PScGPD-HA-TScCYC1	URA3

2.2. Development of K. marxianus expression vectors

The integrating shuttle vector pRS306 was used as a backbone for construction of recombinant yeast expression vector. First, 1265 bp comprising of ARS/CEN sequence was PCR amplified from the genomic DNA of K. marxianus strain NBRC1777. Amplified product was inserted into the vector pRS306 by Transfer-PCR method (Erijman et al., 2011). The resultant expression vector is referred to as pPKM316. Subsequently, the DNA encoding eGFP and T_ScCYC1 terminator (of S. cerevisiae) was subcloned into pPKM316 using restriction sites BamHI and KpnI to construct pPKM316-eGFP-T_ScCYC1. The DNA encoding desired promoter (Table S2) was PCR amplified from genomic DNA of K. marxianus strain NBRC1777 using promoter specific forward and reverse primers (Table S1). The amplified DNA sequences were digested with restriction enzymes SacI and XbaI. The digested product was ligated into pPKM316-eGFP-T_ScCYC1 to develop expression vector encoding eGFP under the control of desired promoter. The plasmid encodes in 5′ to 3’ direction, the desired promoter, eGFP and T_ScCYC1 terminator.

For studying the role of different terminators, the T_ScCYC1 terminator was later replaced by desired terminator sequences. The desired terminator sequences (T_KmIMTT1 or T_KmIMTT2) were PCR amplified from K. marxianus genome, and subcloned into pPKM316-P_KmIMTCP2-eGFP plasmid using XhoI and KpnI restriction sites.

Four different signal sequences (signal sequence of Inulinase with P10L substitution (ssInu1_Km) (Zhou et al., 2018) and mating α-factor of K. marxianus (ssαMF_Km), Pho1 signal sequence of P. pastoris (ssPho1_Pp), mating α-factor signal sequence with D83E of S. cerevisiae (ssαMF_Sc)) were examined in the present study. The DNA encoding ssαMF_Sc was PCR amplified from P. pastoris plasmid pPic9 (Invitrogen #K1710-01). The DNA encoding ssαMF_Km was PCR amplified from the K. marxianus genomic DNA. The DNA encoding ssInu1_Km and ssPho1_Pp signal sequence were constructed using overlap PCR. The DNA cassette encoding from 5′ to 3′, P_KmIMTCP2 promoter and desired signal sequences were constructed using overlap PCR and further subcloned using restriction sites SacI and XbaI into expression plasmid at 5’ end of gene encoding eGFP.

For β-galactosidase expression, the β-galactosidase gene sequence was PCR amplified from the genomic DNA of K. marxianus strain NBRC1777 and subcloned at SpeI and XhoI restriction sites under desired promoter and terminator.

For PCR amplification, the initial denaturation was carried out for 5 ​min ​at 98 ​°C, followed by additional 30 cycles of denaturation for 30 ​s ​at 98 ​°C, primer annealing for 30 ​s ​at 55–60 ​°C and extension at 72 ​°C. Then one final extension time of 10 ​min ​at 72 ​°C was carried out.

2.3. Construction of K. marxianus strain with uracil auxotrophy

The uracil auxotrophy of K. marxianus was constructed using homologous recombination of DNA cassette lacking Ura3 encoding gene at Ura3 locus. The DNA cassette lacking URA3 and having homology of 250 base pair (bp) upstream and 250bp downstream sequence of URA3 gene was synthesized from GenScript USA, Inc. The cassette was transformed into K. marxianus strain. The transformants with uracil auxotropy was identified by their inability to grow on solid media lacking uracil as well as growth onto 5-fluoroorotic acid (5-FOA) containing growth media.

2.4. Immunoblot analysis

NBRC1777 ura3Δ cells (10 ​ml culture volume in 50 ​ml flasks) harboring desired plasmid were grown until mid-log phase. The cells were harvested by centrifugation and washed with PBS buffer. The cells were lysed, and cell-debris was removed by centrifugation at 3000g for 2 ​min. The lysate supernatant was used for immunoblotting. About 10–15 ​μg of total protein from cell lysate was subjected to electrophoresis onto a 12% SDS–polyacrylamide gel. The proteins were then transferred onto PVDF membrane. A 1:5000 dilution of an anti-GFP (cat. No. MA5-15256; thermo fisher scientific) or anti-Pgk1 (Life Technologies, USA-459250) or anti- His₆ (Pierce, USA-MA1-21315) antibody was used for detection.

To detect eGFP in culture growth media, the cells were separated from growth media by centrifugation at 12000g for 30 ​min. The supernatant was collected, and 20 ​μl of supernatant was used for immunoblotting.

2.5. In vivo luciferase assay

The gene encoding luciferase (FLuc) was PCR amplified and subcloned into plasmid downstream of desired promoter cassette using restriction sites BamHI and XhoI. The designed plasmids encoding luciferase were transformed into yeast K. marxianus or S. cerevisiae. The 3–4 transformants were pooled and grown overnight into SD liquid growth media lacking uracil. The overnight grown primary culture was diluted to 0.05 O.D.₆₀₀ using 20 ​ml of sterile growth media (in 100 ​ml flask) with dextrose or xylose as carbon source. Cells were grown at either 30 ​°C or 45 ​°C at 200 ​rpm until cell O.D.₆₀₀ reached 0.8–1.0. The cells were harvested by centrifugation and re-suspended into 1 ​ml of YNB media. For measuring luciferase activity, 200 ​μl of 0.3 O.D.₆₀₀ ​cells were added with 50 ​μl of 1 ​mg/ml D-Luciferin (Sigma-Aldrich Cat #L9504), and the luminescence was measured in Multimode Plate Readers (TECAN Infinite M200 PRO, Switzerland).

2.6. Quantitative real-time PCR

Cells, grown in selective SD media (20 ​ml in 100 ​ml flasks) at 30 ​°C and 200 ​rpm until mid-log phase, were harvested and total RNA was isolated using HiPurA Yeast RNA Purification Kit (HiMedia, India-MB611). The 100 ​ng of isolated RNA was used to prepare cDNA using a cDNA synthesis kit (BIO-RED Cat#170–8891). For quantitative Real-Time PCR (qRT-PCR), cDNA (1 ​μl) was amplified using iTaq™ Universal SYBR® Green Supermix qPCR kit (BIO-RED, CAT#172–5124) by using eGFP specific primer (Table S1) and following manufacturer’s protocol. The reaction was subjected to an initial denaturation at 95 ​°C for 2 ​min and 40 repetitive cycles of 15 ​s ​at 95 ​°C, 30 ​s ​at 60 ​°C in CFX96 Touch Real-Time PCR System (BIO-RED). The 18 ​S rRNA was selected as a reference gene.

2.7. β-Galactosidase activity assay

The K. marxianus cells harboring plasmid encoding β-galactosidase were grown until mid-log phase as 10 ​ml culture in 50 ​ml flasks (shaking speed 200 ​rpm). 1 O.D._600nm cells were treated with 3 rounds of freeze thaw cycle for permeabilization. Cells were further incubated with 0.2 ​ml of ortho-Nitrophenyl-β-galactoside (ONPG 4 ​mg/ml) for 15 ​min. The reaction was stopped with 0.3 ​ml of 1 ​M Na₂CO₃. Cells were centrifuged at 12000g for 10 ​min and the enzyme activity was monitored in supernatant by measuring absorbance at 420 ​nm.

3. Results

3.1. The transcriptomic analysis of gene expression reveals the promoters of higher strength

As promoter act as a regulatory region for gene transcription, the relatively higher number of a gene transcript is likely to be expressed from a stronger promoter. Thus, to identify strong promoters in K. marxianus, we analyzed the transcription profile of cellular genes as a reporter of the relative strength of promoters. A previous study has reported transcriptomic data of K. marxianus cells grown at 30 ​°C in liquid YPD media under shaking condition (Lertwattanasakul et al., 2015). To understand the relative increase in the number of transcripts, we normalized the available gene expression data of each gene by dividing its number of transcripts by the lowest positive transcript number (0.03 ​ppm ​at 30 ​°C).

Table 1 provides fold increase of the top 10 highly expressed gene transcripts. Interestingly the top 2 highly expressed genes, IMTCP1 and IMTCP2, encode for yet uncharacterized proteins. One of the gene, IMTCP3 codes for a subunit of 40 ​S ribosomal protein. The ribosomal proteins are known to be abundantly present in the cell (Ho et al., 2018). The gene products driven by previously used strong promoters such as glyceraldehyde 3-phosphate dehydrogenase 1, inulinase, enolase were also found to be among top 10 identified transcripts suggesting that these identified promoters are likely to be strong promoters for use in K. marxianus.

Table 1.

Fold change in mRNA level of 10 most abundant transcripts. The fold change was calculated by dividing the transcript value of the gene by the lowest non-zero transcript value (0.03) reported in (Lertwattanasakul et al., 2015).

Promoter/Gene notation	Name	CDS product	30 ​°C (Fold change)	45 ​°C (Fold change)	Length of Promoter (bp)
PKmIMTCP1/IMTCP1	KLMA_10040	Uncharacterized protein YDR524C–B	17169530	7149637	1000
PKmIMTCP2/IMTCP2	KLMA_40174	Uncharacterized cell wall protein YDR134C	1362106	1221608	999
PKmIMTCP3/IMTCP3	KLMA_40094	40 ​S ribosomal protein S13	213216.4	438075.7	438
PKmIMTCP4/IMTCP4	KLMA_10830	Ribosyldihydronicotinamide dehydrogenase [quinone]	101348.9	451185.9	594
PKmIMTCP5/IMTCP5	KLMA_20547	Protein BTN2	131705.2	602783.1	479
PKmIMTCP6/IMTCP6	KLMA_10518	Inulinase	431333.3	20887.56	930
PKmIMTCP7/IMTCP7	KLMA_40218	Glyceraldehyde-3-phosphate dehydrogenase 1	420417	79451.56	1008
PKmIMTCP8/IMTCP8	KLMA_70281	Asparagine synthetase [glutamine-hydrolyzing] 1	23655.22	353828.3	1000
PKmIMTCP9/IMTCP9	KLMA_10179	Glycerol-3-phosphate dehydrogenase [NAD+] 1	76492.56	40412	1000
PKmIMTCP10/IMTCP10	KLMA_10462	Enolase	143882.2	26598.33	977
PKmTDH3/TDH3	KLMA_80059	Glyceraldehyde-3-phosphate dehydrogenase 3	489093.7	81713.78	1060
PKmPGK1/PGK1	KLMA_10540	Phosphoglycerate kinase	15041.11	3238.889	1035

3.2. Promoter driving IMTCP1 or IMTCP2 shows the highest expression of reporter protein eGFP

To examine the relative strength of the above-identified promoters, we constructed an expression plasmid compatible for use in K. marxianus. The plasmid was further used to express eGFP encoded under the identified promoters.

For the construction of the recombinant yeast expression vector, the ARS/CEN sequence of K. marxianus strain NBRC1777 was inserted into integrating shuttle vector pRS306 (ATCC 77141) to generate plasmid pPKM316 (Fig. 1) (Ball et al., 1999). To examine the strength of individual promoter, the gene encoding reporter protein eGFP was subcloned into pPKM316. The DNA encoding each of the identified promoters was further subcloned into the plasmid at region 5’ to gene encoding eGFP. Previous studies show that 500–700 bp upstream region is sufficient to drive gene expression (Lang et al., 2020; Lee et al., 2015; Ohler and Niemann, 2001). Thus a minimum length of 500bp was selected as promoter region. We selected either intergenic or 1000 bp upstream region (except for P_KmIMTCP4) if intergenic region is more than 1000bp, as the promoter length. For P_KmIMTCP4, instead of 1000bp only 594 bp were selected (due to the presence of a desired restriction site just upstream of 594bp) as the promoter region.

Fig. 1.

Schematics showing the methodology used for development of K. marxianus expression plasmid. The CEN/ARS origin sequence of K. marxianus was inserted into the vector pRS306 to generate pPKM316. The pPKM316 was further used for insertion of cassette encoding desired promoters and terminators.

The plasmids encoding eGFP under desired promoters were transformed into K. marxianus cells. The transformants were grown into liquid SD media lacking uracil at 30 ​°C. The cellular lysate prepared from equal O.D._600nm cells was probed with anti-GFP antibodies. The Pgk1 expressed from chromosomal PGK1 encoded under its native promoter was used as an internal control. As seen in Fig. 2, though Pgk1 is expressed at a similar level, the cellular abundance of eGFP varied with different promoters. We observed that as compared to previously known strong constitutive promoter P_KmTDH3, the eGFP expression from P_KmIMTCP2 was about 2 fold higher. The P_KmIMTCP1 showed second highest expression as compared to other promoters selected from transcriptomic analysis. The other chosen promoters showed much lower expression than P_KmIMTCP2.

Fig. 2.

P_KmIMTCP1and P_KmIMTCP2show relatively higher expression of reporter protein eGFP. (A) Immunoblot analysis to compare the strength of different promoters using eGFP as a reporter protein and T_ScCYC1 terminator. The cells were transformed with plasmid encoding indicated promoters. Transformants were grown in liquid growth media containing dextrose (SD) at 30 ​°C or 45 ​°C. The cellular lysate was immunoblotted with anti-GFP or anti-PGK1 antibody. Previously known P_KmTDH3 and P_KmPGK1 promoters of K. marxianus origin were also examined for comparison. (B) The promoter strength was analyzed similar to as descrived in Panel “A” except that xylose was used as carbon source instead of dextrose. (C) The eGFP expression from indicated promoters. Cells were grown at 30 ​°C in media containing dextrose or xylose as carbon source. The endogenous Pgk1 showed lower expression when cells were grown in xylose containing media.

To examine whether the relative strength of identified promoters varies with temperature, we examined eGFP level from cells grown under shaking conditions in SD media at 45 ​°C. The cellular lysate was prepared similarly to the method described above, and the eGFP level was measured on immunoblot using anti-GFP antibodies. Similar to as seen above for cell growth at 30 ​°C, among all promoters that were examined, P_KmIMTCP2 was found to be the strongest promoter followed by P_KmIMTCP1 whose strength was found be similar to as of previously used P_KmTDH3. The strong activity of P_KmIMTCP2 is in agreement with a recent study that identified NC1 (smaller fragment of P_KmIMTCP2) as one of the strong promoters in K. marxianus (Lang et al., 2020).

3.3. P_KmIMTCP1 and P_KmIMTCP2 show the strongest eGFP expression in growth media containing xylose as carbon source

The type of carbon source varies depending upon the biotechnological applications which influences microbial growth. For example, xylose is one of the primary sources of carbon when lignocellulose hydrolysates is used for biofuel production. We thus examined the relative strength of our identified promoter in media containing xylose instead of dextrose as carbon source, and compared with that of previously used strong promoters P_KmTDH3 and P_KmPGK1. The cells harboring plasmid encoding eGFP under various promoters were grown into liquid growth media with xylose (SX) as the carbon source and the cellular lysate was examined for eGFP expression. We observed that in xylose containing media, the expression of eGFP was higher from promoters P_KmIMTCP1, P_KmIMTCP2, P_KmIMTCP6, and P_KmIMTCP9 as compared to that from previously used strong promoters, P_KmTDH3 and P_KmPGK1 (Fig. 2B). The results from SD and SX media showed that Inulinase promoter (P_KmIMTCP6) has relatively higher activity in the presence of xylose than dextrose. Among the newly identified promoters, the eGFP expression was highest from P_KmIMTCP2 followed by from P_KmIMTCP1. We further compared the ability of two strongest promoters (P_KmIMTCP1 and P_KmIMTCP2) to drive eGFP expression in dextrose or xylose containing growth media. As seen in Fig. 2C, though the strength of P_KmTDH3 varied with the carbon source, the identified strong promoters showed similar eGFP expression in both media suggesting that the strength of these newly identified promoters is independent of dextrose or xylose as carbon source. Overall above data suggest that the identified P_KmIMTCP1 and P_KmIMTCP2 promoters are more efficient at different temperatures and carbon sources than the previously known promoters in K. marxianus (Fig. S1).

3.4. The 388 bp at 5’ end of P_KmIMTCP1 are indispensable for its activity

We further focused on P_KmIMTCP1 and P_KmIMTCP2 promoters that showed relatively higher eGFP expression as compared to the other promoters. Promoter region binds to RNA polymerase as well as with other regulatory factors required for the polymerase recruitment. To determine the regulatory region in P_KmIMTCP1 and P_KmIMTCP2, we constructed various derivatives of these promoters by sequentially truncating stretch of DNA bases from the 5′ region of the promoter as shown in Fig. 3A and 3B. For P_KmIMTCP1, 4 truncated derivatives of different lengths were PCR amplified, and subcloned 5′ to gene encoding eGFP in plasmid pPKM316-eGFP-T_ScCYC1. The capability of truncated promoters to drive eGFP expression was monitored in K. marxianus as described above for full-length promoter at 30 ​°C in media containing dextrose as a carbon source. The varying level of eGFP expression was observed with promoters of different length (Fig. 3C and 3D). The results show that P_{KmIMTCP1-1000}, P_KmIMTCP1-787 and P_KmIMTCP1-612 have relatively similar strength. Further truncation of promoter length led to a progressive decrease in eGFP cellular abundance with P_KmIMTCP1-194 showing least expression of the protein. Overall, the data suggest that the extreme 388 bp at 5’ region of the P_KmIMTCP1 does not contribute much to its ability to drive gene expression.

Fig. 3.

eGFP expression varies with the promoter length. (A and B) Schematics of designed derivatives of P_KmIMTCP1 and P_KmIMTCP2. (C) Cells were transformed with plasmid encoding varying length of P_KmIMTCP1 promoter. Transformants were grown in SD liquid media, and cellular lysate was immunoblotted with anti-GFP antibody. Pgk1 was probed as loading control. (D) Cells were examined for eGFP expression using method similar to as mentioned above in Panel “C”. The eGFP was expressed under indicated derivatives of P_KmIMTCP2.

Similarly, six truncated promoters of varying lengths based upon the parent P_KmIMTCP2 promoter were constructed and examined for their strength using eGFP as reporter protein. The expression of eGFP decreased with an increase in truncation of 5’ region of P_KmIMTCP2 suggesting the full-length promoter was required for the maximal protein expression.

3.5. The terminators T_KmIMTT1 and T_KmIMTT2 enhance promoter activity by stabilizing mRNA

The cellular abundance of a protein not only depends upon the strength of promoter but also on the terminator sequence that follows 3′ to the encoding gene. The terminator sequence affects gene expression by modulating mRNA stability (Curran et al., 2013; Rajkumar et al., 2019). Thus the relatively higher transcript level of the chromosomally encoded native gene under P_KmIMTCP1 or P_kmIMTCP2 could be dependent upon its native terminators. To examine the effect of native terminators of P_kmIMTCP1 (T_KmIMTT1) and P_kmIMTCP2 (T_KmIMTT2) on P_kmIMTCP2 strength, we replaced T_ScCYC1 terminator from 3’ end of gene encoding eGFP with that of T_KmIMTT1 and T_KmIMTT2. It is reported that terminator sequences generally contains characteristic elements such as AT-rich efficiency elements and A-rich positioning elements (AAG/TAA), and about 200bp downstream sequence from stop codon is sufficient for termination (Helden et al., 2000). To include such characteristic features, we used minimum of 250bp downstream of stop codon (269bp and 290bp for T_KmIMTT1 and T_KmIMTT2 respectively) as the terminator sequence.

The plasmid encoding from 5′ to 3’ direction P_kmIMTCP2, eGFP, T_KmIMTT1/T_KmIMTT2 was transformed into K. marxianus strain. The transformants were grown in liquid SD media at 30 ​°C, and examined for eGFP expression using immunoblot as described above. Interestingly, the expression of eGFP from P_kmIMTCP2 was enhanced when T_KmIMTT1 was used as terminator instead of T_ScCYC1 (Fig. 4A). Similar increase of eGFP expression was observed with T_KmIMTT2 terminator as compared to T_ScCYC1. To examine whether increase in eGFP level is because of enhanced mRNA stability, we carried out qRT-PCR to examine the transcript level (Fig. 4B). Similar to as seen for increase in eGFP protein abundance, about 4 fold increase of mRNA level was observed when T_ScCYC1 was replaced with either T_KmIMTT1/T_KmIMTT2. Further the eGFP expression from P_KmIMTCP2-T_KmIMTT1 was found to similar at different temperatures (18 ​°C, 30 ​°C and 37 ​°C) used for K. marxianus growth (Fig. 4C). We also examined the effect of the protein overexpression on cells growth, and seen in Fig. S2, no significant growth defect was observed upon eGFP overexpression from P_kmIMTCP2-T_KmIMTT1 cassette. Overall, the designed sets of promoters and terminators showed significantly higher expression than any of the other native promoters used for protein expression in K. marxianus.

Fig. 4.

The eGFP expression improves when P_KmIMTCP2is used in combination with T_KmIMTT1or T_KmIMTT2(A) Cells were transformed with plasmids encoding indicated combinations of promoters and terminators. Transformants were grown in SD liquid media at 30 ​°C, and examined for eGFP expression in cellular lysate using anti-GFP antibodies. Pgk1 level was examined as loading control. (B) Cells were grown as in Panel A. The mRNA was isolated, and eGFP transcript level was examined using qRT-PCR with primers specific for the genes encoding eGFP or 18 ​S (internal control). (C) The strength of indicated DNA cassette was examined by monitoring eGFP level expressed at 18 ​°C, 30 ​°C or 37 ​°C in K. marxianus.

3.6. The cassette P_KmIMTCP2-T_KmIMTT1 shows relatively higher expression of luciferase gene

We further examined the effect of T_KmIMTT1/T_KmIMTT2 terminator sequences on the expression of gene encoding firefly luciferase driven by P_KmIMTCP1 and P_KmIMTCP2 promoters. The luciferase encoding gene was subcloned in-between 5′ and 3′ region of desired promoter and terminator respectively. The plasmid encoding from 5′ to 3’ direction, the desired promoter, gene encoding firefly luciferase, and desired terminator was transformed into K. marxianus strain. The transformants were examined for in vivo luciferase activity by monitoring increase in luminescence as a measure of luciferase expression. As seen in Fig. 5, when cells were grown at 30 ​°C (Fig. 5A) or 45 ​°C (Fig. 5B) in dextrose containing liquid growth media, varying level of luciferase activity was observed from different combinations of promoters and terminators. The strongest luminescence signal was obtained when luciferase was expressed from P_KmIMTCP2 promoter in combination with T_KmIMTT1 or T_KmIMTT2 terminator sequences, suggesting that among the different combinations, P_KmIMTCP2-T_KmIMTT1 or P_KmIMTCP2-T_KmIMTT2 drive the strongest expression of the protein. Similarly, there was ~1.7 fold increase in luminescence when the luciferase was expressed from P_KmIMTCP1-T_KmIMTT1 or P_KmIMTCP2-T_KmIMTT1 as compared to P_KmIMTCP1-T_ScCYC1 or P_KmIMTCP2-T_ScCYC1. The data shows that the order of promoter-terminator combination strength with regard to their ability to drive luciferase expression is P_KmIMTCP2-T_KmIMTT1 ​~ ​P_KmIMTCP2-T_KmIMTT2 ​> ​P_KmIMTCP1-T_KmIMTT1 ​~ ​P_KmIMTCP1-T_KmIMTT2 ​> ​P_KmIMTCP2-T_ScCYC1 ​~ ​P_KmIMTCP1-T_ScCYC1 ​> ​P_KmTDH3-T_ScCYC1 >P_KmPGK1-T_ScCYC1.

Fig. 5.

The T_KmIMTT1enhances luciferase expression. Cells were transformed with plasmid encoding luciferase under the control of different combinations of indicated promoters and terminators. Transformants were grown until OD_600nm of 0.8 in liquid synthetic media containing dextrose or xylose as carbon source. About 200 ​μl of 0.3 O.D._600nm cells were added with D-Luciferin as luciferase substrate, and increase in luminescence was monitored. Shown is the luciferase activity from cells grown (A) at 30 ​°C with dextrose or (B) 45 ​°C with dextrose as carbon source or (C) at 30 ​°C with xylose as carbon source.

In order to explore whether the relative strength of above DNA cassette with identified promoters and terminators is affected by a different carbon source, we grew cells harbouring plasmid encoding these cassettes in media containing xylose instead of dextrose as source of carbon. The luciferase activity was measured as described above. Similar to as observed for dextrose containing media, when cells were grown in media containing xylose, P_KmIMTCP2-T_KmIMTT1 ​~ ​P_KmIMTCP2-T_KmIMTT2 showed highest luciferase activity followed by that shown by P_KmIMTCP1-T_KmIMTT1 ​~ ​P_KmIMTCP1-T_KmIMTT2 (Fig. 5C). Further, the identified promoters, P_KmIMTCP1 and P_KmIMTCP2 with any of the three terminators T_ScCYC1, T_KmIMTT1 or T_KmIMTT2 showed relatively higher luciferase expression than P_KmPGK1-T_ScCYC1 or P_KmTDH3-T_ScCYC1.

3.7. The terminators T_KmIMTT1 and T_KmIMTT2 further improved promoter strength in S. cerevisiae

The promoters of S. cerevisiae origin have been successfully used to express proteins in K. marxianus suggesting that promoters from one yeast species could also be active in others. We thus examined the ability of identified K. marxianus promoter to drive protein expression in S. cerevisiae. Further as terminators provide an additional regulation for protein expression, we examined whether the above identified terminators affect P_ScGPD in S. cerevisiae. Various combinations of promoters and terminators (Fig. 6), were constructed as described in Table 2. Overall, we used three promoters, P_ScGPD from S. cerevisiae, P_KmIMTCP1 and P_KmIMTCP2 from K. marxianus; and three terminator sequences, T_ScCYC1 from S. cerevisiae, T_KmIMTT1 and T_KmIMTT2 from K. marxianus. The firefly luciferase was used as reporter protein to examine the strength of these expression cassettes.

Fig. 6.

Comparison of the activity of luciferase expressed using indicated combination of promoter and terminator in S. cerevisiae. Cells harboring plasmid encoding luciferase under different promoter and terminator combinations were grown in liquid SD media at 30 ​°C. The luciferase activity was monitored as described in Materials and Methods. The luciferase substrate D-luciferin was added into 200 ​μl of culture volume (O.D._600nm ​= ​0.3) and increase in luminescence was monitored.

Table 2.

Promoter and terminator combinations used in this study.

S. No.	Combination
1.	PKmIMTCP1-TScCYC1
2.	PKmIMTCP2-TScCYC1
3.	PKmIMTCP3-TScCYC1
4.	PKmIMTCP4-TScCYC1
5.	PKmIMTCP5-TScCYC1
6.	PKmIMTCP6-TScCYC1
7.	PKmIMTCP7-TScCYC1
8.	PKmIMTCP8-TScCYC1
9.	PKmIMTCP9-TScCYC1
10.	PKmIMTCP10-TScCYC1
11.	PKmTDH3-TScCYC1
12.	PKmPGK1-TScCYC1
13.	PKmIMTCP1-TKmIMTT1
14.	PKmIMTCP1-TKmIMTT2
15.	PKmIMTCP2-TKmIMTT1
16.	PKmIMTCP2-TKmIMTT2
17.	PScGPD-TScCYC1
18.	PScGPD-TKmIMTT1
19.	PScGPD-TKmIMTT2

The pRS316 either alone or encoding any of above expression cassettes was transformed into S. cerevisiae strain BY4741, and transformants were examined for the luciferase activity using method similar to as used above for K. marxianus strain. Both P_KmIMTCP1-T_ScCYC1 and P_KmIMTCP2-T_ScCYC1 showed activity in S. cerevisiae as seen by enhanced luciferase activity as compared to cells harboring vector alone however the luciferase activity was slightly lower than when luciferase was expressed from native promoter and terminator of S. cerevisiae (P_ScGPD-T_ScCYC1). The substitution of T_ScCYC1 in P_ScGPD-T_ScCYC1 with T_KmIMTT1 or T_KmIMTT2 improved the promoter ability to drive luciferase expression as seen by further increase in luminescence by 1.6 and 1.8 fold respectively. Similar increase in luciferase activity was detected when T_ScCYC1 was substituted with T_KmIMTT1 or T_KmIMTT2 in P_KmIMTCP1-T_ScCYC1 and P_KmIMTCP2-T_ScCYC1. The cells encoding designed P_ScGPD-T_KmIMTT1 or P_ScGPD-T_KmIMTT2 showed marginally higher luminescence than those encoding P_KmIMTCP2-T_KmIMTT1 or P_KmIMTCP2-T_KmIMTT2.

3.8. The cassette P_KmIMTCP2-T_KmIMTT1 showed higher strength than P_scGPD-T_ScCYC1 for influenza HA antigen expression

We further examined the application of above designed expression cassette from K. marxianus for its ability to drive expression of influenza protein Hemagglutinin (HA). The HA antigen is a potential candidate for the design of influenza vaccine. We examined HA expression driven by the designed DNA cassette P_KmIMTCP2-T_KmIMTT1 in K. marxianus, and compared with that produced from widely used strong constitutive promoter cassette P_ScGPD-T_ScCYC1 in S. cerevisiae. The gene encoding His₁₀-HA was subcloned with desired combination of promoters and terminators in centromeric plasmid for expression in K. marxianus or S. cerevisiae. The cells harboring expression plasmid were grown until mid log phase, and cellular lysate was probed with anti-His tag antibodies on an immunoblot. As seen in Fig. 7, the HA expression from K. marxianus was found to be 2.5 fold higher than that from P_ScGPD promoter in S. cerevisiae, further suggesting that the identified promoter cassette could not only drive stronger expression of reporter proteins such as eGFP or firefly luciferase but also of other heterologous proteins.

Fig. 7.

Comparative expression of HA protein in different yeast strains. Immunoblot analysis of expression level of His tagged HA protein in S. cerevisiae and K. marxianus. The His₁₀-HA was expressed from strong GPD promoter in S. cerevisiae. The P_KmIMTCP2-T_KmIMTT1 was used to express His₁₀-HA in K. marxianus. Yeast cells were grown at 30 ​°C with dextrose as carbon source. The cells were harvested and cellular lysate was probed with anti-His₆ or anti-Pgk1 (loading control) antibody.

3.9. S. cerevisiae α-factor signal sequence efficiently secretes eGFP in K. marxianus

We further examined the ability of the identified P_KmIMTCP2-T_KmIMTT1 cassette on expression of eGFP secreted out into extracellular growth media. The secretion of a protein into extracellular space requires the presence of secretory signal sequence that traffics the protein into media supernatant through endoplasmic reticulum (ER).

To examine the secretion of eGFP in extracellular growth media, the protein was expressed in fusion with various signal sequences under P_KmIMTCP2. The various signal sequence that are compared in the present study are; Inulinase signal sequence (ssInu1_Km) with P10L substitution of K. marxianus, Pho1 signal sequence (ssPho1_Pp) of P. pastoris, mating α-factor signal sequence (ssαMF_Sc) containing D83E substitution of S. cerevisiae and mating α-factor signal sequence (ssαMF_Km) of K. marxianus. The plasmid encoding eGFP in fusion with desired signal sequence was transformed in K. marxianus. The media supernatant was collected and examined for eGFP expression. As seen in Fig. 8A, among the various signal sequences examined, eGFP in media supernatant was significantly higher when expressed in fusion with signal sequence from mating α factor of S. cerevisiae.

Fig. 8.

Comparison of secretory capacity of various signal sequences in K. marxianus. (A) Cells harboring plasmid encoding eGFP fused with indicated signal sequences under P_KmIMTCP2-T_KmIMTT1 cassette was grown for 24 ​h and 48 ​h. The culture was spun at 12000g for 15 ​min, and media supernatant was collected, and further probed with anti-GFP antibody. The signal sequences examined are: Inulinase signal sequence (ssInu1_Km) with P10L substitution of K. marxianus, ssPho1_Pp signal sequence of P. pastoris, mating α-factor signal sequence (ssαMF_Sc) with D83E of S. cerevisiae and mating α-factor signal sequence (ssαMF_Km) of K. marxianus(B) expression of extracellular eGFP expressed in fusion with ssαMF_Sc encoded under the control of P_KmIMTCP2-T_KmIMTT1. The DNA P_KmIMTCP2-ssαMF_Sc-eGFP-T_KmIMTT1 cassette was genome integrated at HIS3 locus in K. marxianus genome. Cells were grown in media containing dextrose (YPD), xylose (YPX) and lactose (YPL). The eGFP expression was examined as above in Panel (A).

We further examined the effect of changes in carbon source on promoter strength as well as ability of αMF_Sc for extracellular secretion of proteins. The DNA cassette P_KmIMTCP2-ssαMF_Sc-eGFP-T_KmIMTT1 encoding fusion of eGFP with ssαMF_Sc expressed under the control of P_KmIMTCP2 promoter was genomically integrated in K. marxianus at HIS3 locus. The genomically integrated strain was grown in liquid growth media containing dextrose, xylose or lactose as carbon source. The eGFP expression was confirmed in extracellular growth media using immnoublot analysis. As seen in Fig. 8B, the eGFP was expressed similarly in all the three types of growth media suggesting the strength of promoter as well as the secretion signal sequence is independent of the type of carbon source used for culturing cells.

3.10. The cassette P_KmIMTCP2-T_KmIMTT1 showed higher β-galactosidase expression

The K. marxianus has potential usage for the enzymatic mitigation of lactose in products such as cheese whey and milk, due to its ability to produce β-galactosidase. To further improve the utility of K. marxianus for lactose utilization, we engineered the strain for enhanced production of β-galactosidase.

Above data show that the ability of designed P_KmIMTCP2-T_KmIMTT1 for protein production is independent of carbon sources such as dextrose, xylose or lactose. We subcloned gene encoding K. marxianus β-galactosidase under the P_KmIMTCP2-T_KmIMTT1 in Ura3 based plasmid. The β-galactosidase encoding plasmid was transformed into K. marxianus, and examined for β-galactosidase activity as mentioned in Materials and Methods. As seen in Fig. 9, wt K. marxianus with empty vector showed lower β-galactosidase activity of 10 miller units. The enzyme activity of the β-galactosidase produced from P_KmIMTCP2-T_KmIMTT1 showed ~1800 miller units higher activity as compared to that produced chromosomal borne gene. Further the P_KmIMTCP2-T_KmIMTT1 produced 1.6 and 24 fold higher activity than previously identified P_KmTHD3 and P_KmPGK1 promoters. Overall our data suggest that the K. marxianus engineered to produce β-galactosidase from P_KmIMTCP2-T_KmIMTT1 produces significantly higher β-galactosidase then the wt organism.

Fig. 9.

The engineered K. marxianus with β-galactosidase encoded gene under P_KmIMTCP2-T_KmIMTT1shows significantly higher enzyme activity. Cells were transformed with plasmid encoding β-Galactosidase under the control of different combinations of indicated promoters and terminators. The β-galactosidase activity was measured as described in Materials and Methods. As seen, the engineered K. marxianus cells that express β-galactosidase from combination of P_KmIMTCPP2-T_KmIMTT1 show significantly higher enzyme activity than those that express from chromosomally encoded enzyme.

4. Discussion

For wider utilization of K. marxianus, there is a need to expand the genetic toolbox required for various applications such as to improve the production of heterologous proteins, engineering of its glycosylation pathways and secretory capacity. Many of such synthetic biology approaches require genetic tools for genomic integration as well as expression of desired proteins in different carbon sources and temperatures. The present study provides novel sets of promoters and terminators, and further used it for metabolic engineering of K. marxianus for higher β-galactosidase activity.

The analysis of available transcriptomic data revealed relative strength of various promoters in K. marxianus. The transcript level of previously used strong promoter P_KmPGK1 in K. marxianus was found to be lower than the most abundant transcripts at 30 ​°C. In agreement, the relative eGFP expression driven from identified top 2 promoters was found to be higher than from P_KmPGK1 suggesting that transcriptomics could potentially provide novel strong promoters with varying level of protein expression. This is also in agreement with previous studies focused to identify novel promoters based upon level of various transcripts (Yuan et al., 2020).

The eGFP or luciferase expression from identified strong promoters P_KmIMTCP1 and P_KmIMTCP2 at 30 ​°C and 45 ​°C showed consistent higher expression than previously known P_KmTDH3 or P_KmPGK1 promoter suggesting that their ability to drive relatively higher protein expression is independent of alteration of growth temperature. The constitutive expression from P_KmIMTCP1 and P_KmIMTCP2 is also consistent with the reported transcriptomic data obtained under various conditions (Table S3) (Lertwattanasakul et al., 2015). As evident from Table S3, IMTCP1 and IMTCP2 have highest transcript value than the rest of the genes (IMTCP3-IMTCP10) under all four conditions (Lertwattanasakul et al., 2015). The promoter strength at higher temperature is particularly crucial in processes where cell growth is required at higher temperature such as during alcohol fermentation.

The substitution of dextrose with xylose for carbon source showed alteration of eGFP expression level suggesting that strength of some of the identified promoters are dependent upon the nature of the carbon source. The effect of change in carbon source to xylose was more pronounced for P_KmIMTCP6 and P_KmIMTCP9 than P_KmIMTCP1 or P_KmIMTCP2. It is possible that the effect of xylose on promoter strength is due to cellular adaptation of metabolic pathways for better utilization of the pentose sugars. This is in agreement with the known influence of change in carbon source on metabolic perturbation and reprograming of gene transcription (Diniz et al., 2017; Paulo et al., 2015). Interestingly, the strength of the best two identified promoters, P_KmIMTCP1 and P_KmIMTCP2 remain independent of change in the carbon source whereas the previously used P_KmTDH3 show relatively lower expression in xylose containing media.

It is known that various regulatory elements interact with the promoter region and govern its ability to drive gene expression. The study examining DNA fragments of P_KmIMTCP1 or P_KmIMTCP2 that are still capable to drive protein expression showed that though P_KmIMTCP1 could be truncated by 388 bp without affecting its ability to regulate protein expression, full length P_KmIMTCP2 is required for maximal activity. The smaller fragment of P_KmIMTCP1 having 612 bp showing similar activity as of full length 1000bp promoter is in agreement with previous studies showing that in general 500-700bp upstream sequence is sufficient to drive protein expression (Lang et al., 2020; Lee et al., 2015; Ohler and Niemann, 2001). However since full-length P_KmIMTCP2 is required, it is possible that different regulatory factors such as transcription factors binds to the extreme 5′ end of the promoter P_KmIMTCP2 and allosterically influences RNA polymerase binding to the 3’ region of the promoter. Similar upstream enhancer/regulatory sequences has been observed before for promoters such POX2 in Y. lipolytica (Shabbir Hussain et al., 2016).

The P_KmIMTCP1 or P_KmIMTCP2 were found to be strongest among the selected 10 promoters. Promoters generally consist of conserved functional elements such as a core promoter region (consisting of TATA box and the transcriptional start site) and upstream regulatory sequences. In general, these regions lie within first 200bp upstream sequence (Basehoar et al., 2004). We looked into the presence of TATA boxes in all 10 promoters and found TATA boxes in less than 250bp upstream of translation start site of all the promoters. The upstream regulatory elements of these promoters are still not defined. Thus we did not find any apparent correlation between amplified promoter regions of the remaining 8 promoters and their low activities. A detailed study to identify the upstream regulatory elements and their role in promoter strengths would be required to further understand the underlying basis of varying activities of these identified promoters.

The further increase in the ability of P_KmIMTCP1 and P_KmIMTCP2 to drive protein expression upon substitution of T_ScCYC1 with T_KmIMTT1 or T_KmIMTT2 indicates that the protein abundance not only depends upon the promoter but also upon terminator which is in agreement with previous studies. The increase in protein expression with identified terminators is not only specific for P_KmIMTCP1 or P_KmIMTCP2 but more general as they also enhanced promoter strength of other promoter such as P_ScGPD. The enhanced expression is not specific to eGFP but is more general for other proteins such as luciferase. As the effect is not specific to a particular promoter or substrate, it is possible that the ability of the identified terminators to enhance protein expression could be related to enhanced mRNA stability. This is in agreement with increase in transcript level of eGFP when T_KmIMTT1 or T_KmIMTT2 were used as terminator sequence.

Our data show that the identified promoter cassette P_KmIMTCP2-T_KmIMTT1 functions efficiently at wide range of sugars as carbon source such as dextrose, xylose and lactose. The metabolically engineered K. marxianus harboring this cassette showing more than 181 fold higher β-galactosidase activity as compared to the strain expressing from chromosomally encoded enzyme, further enhances the wider utility of the strain in applications requiring lactose degradation such as from cheese whey and milk.

The ability to metabolically engineer microorganism has greatly enhanced their potential for diverse applications. K. marxianus provides a unique opportunity in various biotechnological applications and thus number of ongoing studies are focused to develop novel genetic tools and to understand more about its biology such as its glycosylation pathway and metabolomics. The present study has now provided various DNA cassettes consisting of novel strong promoters and terminators and further showed the application of these tools in metabolic engineering of K. marxianus for enhanced β-galactosidase activity. We also show that these tools could be efficiently used for the production of potential therapeutic agents such as influenza antigen HA. The promoters and terminators as well as the engineered K. marxianus strain identified in the present study has enhanced the potential of K. marxianus for various research, industrial and biotherapeutic applications.

5. Conclusions

In the present study, we have identified novel strong native promoters of K. marxianus. The ability of identified promoters for protein production was further enhanced by novel sets of terminators. The identified promoters showed similar strength at wider range of temperatures and different carbon sugars. The identified promoters and terminators were also found to be active in S. cerevisiae. Further, we used the identified promoter and terminator DNA cassette to engineer K. marxianus for increased production of β-galactosidase. The engineered strain would have wider applications requiring mitigation of lactose such as in cheese whey and milk. Thus, the designed DNA cassettes of novel promoters and terminators has increased the genetic tool box for basic research, protein production and metabolic engineering applications in K. marxianus.

Author contributions

P.K., D.K.S. and D.S. designed research. P.K. performed research. D.S., and D.K.S., arranged for reagents and tools. P.K., D.K.S. and D.S.analyzed results. P.K., D.K.S. and D.S. wrote manuscript.

Financial disclosure

The work is supported by the Council of Scientific and Industrial Research (CSIR), India, and the Ramalingaswami Re-entry Fellowship (BT/RLF/RE-entry/33/2010) from the Department of Biotechnology (DBT), India to Deepak Sharma. The current study was also supported by funds received from BIRAC, Department of Biotechnology, India.

Additional information

A patent on part of this study has been filed: “VECTOR FOR HIGH-EXPRESSION OF PROTEINS IN YEAST” Application No. PCT/IN 2020/050640.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is/are the supplementary data to this article:

Figure S1. The relative strength of four strong promoters, identified in the current study. Immunoblot analysis to compare the strength of four selected promoters using eGFP as a reporter protein and T_ScCYC1 terminator. The cells were transformed with plasmid encoding indicated promoters. Transformants were grown in liquid SD media at 30 ​°C or 45 ​°C and SX at 30 ​°C. The cellular lysate was immunoblotted with anti-GFP or anti-PGK1 antibody. Previously known promoters of K. marxianus origin, P_KmTDH3 and P_KmPGK1, were also examined for comparison.

Figure S2. Effect of eGFP overexpression on cell growth. Cells were transformed with empty vector (EV) and vector having eGFP under the control of P_KmIMTCP2 -T_KmIMTT1. The cell growth was monitored by measuring OD₆₀₀ at indicated time intervals.