Alien Chromosome Serves as a Novel Platform for Multiple Gene Expression in Kluyveromyces marxianus

TL;DR

Scientists used a yeast species to express multiple genes by inserting a foreign chromosome, enabling stable and efficient gene expression for biotechnology applications.

Contribution

The study introduces an alien chromosome as a stable platform for multi-gene expression in Kluyveromyces marxianus.

Findings

Replacing genes on the alien chromosome with heterologous genes led to active expression and enhanced enzyme production.

Multiple gene insertions at specific loci resulted in higher expression levels compared to single insertions.

The alien chromosome platform supports stable propagation without selective pressure.

Abstract

Kluyveromyces marxianus is an emerging yeast cell host for diverse products, but multiple-gene expression in K. marxianus faces challenges due to limited current knowledge of cis-regulatory elements and insertion loci. Our previous study transferred an alien Saccharomyces cerevisiae chromosome I (R1) into K. marxianus, resulting in the creation of the monochromosomal hybrid yeast KS-R1. All R1 genes were actively transcribed, providing a series of loci with varying transcriptional activities. Here, we explore the use of R1 as a novel platform for stable, multi-gene integration and expression. By deleting three essential K. marxianus genes while complementing their functions with orthologs on R1, we achieved stable propagation of R1 in the absence of selective pressure. We characterized several loci on R1 that exhibit stable transcriptional activities under various conditions. GFP inserted in place of genes at six such loci demonstrated varying expression levels. Strains with GFP at two loci exhibited significantly higher expression than those with GFP at a single locus. Furthermore, we replaced five R1 genes with disulfide bond formation genes from Pichia pastoris at distinct loci, resulting in the active expression of all five genes and significantly enhanced production of heterologous glucoamylases BadGLA and TeGlaA. Our findings demonstrate that alien chromosomes offer a stable and versatile platform for the coordinated expression of multiple heterologous genes, serving as valuable tools for metabolic engineering and synthetic biology.

Figures

Complementation between essential genes on R1 and deletion of their counterparts in K. marxianus. (A) Schematic view of R1 engineering. (B) List of Sc essential genes on R1. Expression levels were indicated by read counts in the transcriptomic analysis. The evolutionary rate was obtained from OrthoDB [31]. (C) Identification of genotypes for KS-R1-Δhk, KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E. Three selective markers on R1 (URA3, kanR, hygR), seven R1 position markers, as well as the KmPOP5, KmPRP45, and KmMAK16 loci in the Km genome were characterized by PCR. The positions of primers relative to the loci and the expected sizes of the PCR products are shown in Figure S1A. (D) Growth curves of KS-R1, KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E in YPD medium. Cells were grown in YPD for 72 h, and OD600 was measured at the indicated times. Values represent the mean (n = 3). (E) Stability of R1 in KS-R1 and KS-R1E. Cells were grown in YPD for 5 days at 30 °C or 37 °C. The stability of R1 in KS-R1 was determined by the ratio of colonies on SC-Ura plates to those on YPD plates. The stability of R1 in KS-R1E was measured by the percentage of colonies maintaining all seven position markers via PCR. Values represent the mean ± standard deviation (SD) (n = 3).

Full text

1. Introduction

Kluyveromyces marxianus (Km) is a homothallic budding yeast belonging to the Saccharomyces subclade within hemiascomycetes. Km harbors various favorable features for industrial production, including “generally regarded as safe” (GRAS) status, thermotolerance, fast growth, and a broad spectrum of utilizable carbon sources [1,2,3]. Since 1980, Km has become an emerging yeast cell host for diverse products. To date, over 50 recombinant proteins, including industrial enzymes, virus-like particles, and food additive proteins, have been successfully expressed by Km [4,5,6,7], with the highest yield reaching 16.8 g/L [8]. The highest titers of bioethanol produced by Km reached 110 g/L [9]. Km has also achieved the production of various chemicals, such as lactic acid [10], xylitol [11], 2-phenylethanol [12], hexanoic acid [13], astaxanthin [14], and polyketides [15]. Apart from that, Km possesses a high potential to produce aroma compounds, including fruit esters, furanones, ketones, carboxylic acids, and aromatic hydrocarbons [16,17].

Genetic engineering of Km always involves the regulated expression of multiple heterologous genes. To display the cellulosome on the cell surface, five fungal cellulase genes were inserted into the genome of Km [18]. To increase the yield of soy leghemoglobin, six heterologous heme synthesis-related genes were introduced into Km via yeast artificial chromosomes (YAC) [7]. To remodel the disulfide bond formation pathway, seven Pichia pastoris genes were introduced into Km using YAC [8].

The current bottleneck in multi-gene expression in Km lies in the lack of knowledge about the cis-elements with varying transcriptional activities, which have not been as well-characterized as they have been in Sc, particularly promoters and terminators. Previous studies have identified a series of native Km promoters with varying transcriptional activities, such as TEF3 (strong), PDC1 (strong), GPD1 (medium), INU1 (medium), MED4 (weak), and SRB7 (weak) [19,20,21]. However, native promoters, when placed upstream of heterologous genes, might cause recombination with native loci, leading to genetic instability. Additionally, native promoters may compete for transcriptional factors, affecting the functions of native genes. Heterologous promoters can reduce the risk of recombination and competition for trans-acting factors. However, only a few heterologous promoters, including the Saccharomyces cerevisiae (Sc) ADH1, ADH2, TEF1, PGK1, TDH3, and GAL1 promoters; the Pichia pastoris AFT1 promoter; and the Ashbya gossypii TEF promoter, have been identified as functional in Km [15,19,20,21,22]. Moreover, all these heterologous promoters are strong or medium promoters, which cannot meet the demand for the expression of multiple genes at various levels. Terminators provide additional means to fine-tune gene expression [20]. A series of Km terminators have been characterized, but these terminators might cause undesirable recombination with native loci [20]. A few heterologous terminators, including the A. gossypii TEF terminator, the ScADH1 terminator, and some short synthetic terminators, have been proven to be functional in Km [21,23]. The limited number of available heterologous terminators hinders multiple-gene engineering in Km.

The limited availability of insertion loci is another problem for multiple heterologous gene expression. The insertion of heterologous genes at genomic loci exhibits positional effects, which might interfere with the expression of the inserted genes, as well as the expression of contextual genes around the locus [24]. Plasmids provide episomal loci for gene insertion, but the stability of plasmid propagation is low [25,26]. Yeast artificial chromosomes of Km (KmYACs) exhibit high stability comparable to natural chromosomes and can carry multiple genes without affecting the native genome [8]. However, the YAC backbone lacks cis-elements for integrated genes. Gene cassettes, including promoters, open reading frames (ORFs), and terminators, have to be assembled before being integrated into the YAC [8]. Therefore, the issue of insufficient cis-elements still needs to be addressed during the insertion of multiple genes into YACs.

Our previous study reported the transfer of alien Sc chromosomes 1 (R1) or 3 (R3) into Km, resulting in the creation of the monochromosomal hybrid yeasts KS-R1 and KS-R3, respectively [27]. R1 and R3 can replicate and segregate normally in KS-R1 and KS-R3, respectively. The introduction of R1 and R3 did not cause any noticeable growth defects. R1 and R3 contain 89 and 151 genes, respectively, all of which can be expressed in the hybrids. Since the Sc chromosomes are heterologous, the genes on them are dispensable for the hybrid and can be safely disrupted or deleted. Therefore, the gene loci on the alien chromosomes can serve as safe integration sites for heterologous genes. Moreover, the expression levels of R1 and R3 genes were diverse in the hybrid, thus providing groups of loci with different transcriptional activities.

Here, by using KS-R1 as a model, we investigate the strategy of utilizing an alien chromosome from S. cerevisiae as a platform for multiple-gene insertion and expression. Three essential Km genes were deleted while the growth of the hybrid was maintained by their orthologous genes on R1. The resulting strain exhibited stable propagation of R1 without selective pressure and provided markers for subsequent engineering. We characterized loci on R1 that exhibit stable expression levels under different conditions. GFP was inserted into six stable loci to determine the varying expression levels among loci. Strains harboring GFP at two loci showed markedly higher expression compared to those with GFP at a single locus. Finally, five disulfide bond formation genes from P. pastoris were inserted into five different loci on R1. All five genes were actively expressed and significantly enhanced the expression of heterologous proteins.

2. Materials and Methods

2.1. Strains, Media and Plasmids

The yeast strains used in this study are listed in Supplementary Table S1. The plasmids used in this study are listed in Supplementary Table S2, and primers are listed in Supplementary Table S3. The following media were used in this study: YPD (2% peptone, 1% yeast extract, 2% glucose, 2% agar for plates), YD (2% yeast extract, 4% glucose), supplemented synthetic medium lacking uracil (SC-Ura) (0.67% yeast nitrogen base without amino acids (291930, BD, Franklin Lakes, NJ, USA), 2% glucose, 2 g/L DO Supplement-Ura (630416, Takara, Beijing, China), 2% agar for plates), and SC medium (SC-Ura medium supplemented with 200 mg/L uracil, 2% agar for plates). Nourseothricin (96736-11-7, RuiTaiBio, Beijing, China) and 5-FOA (F8420, Solarbio, Beijing, China) were added to YPD medium at final concentrations of 100 mg/L and 400 mg/L, respectively, to prepare YPDN and YPD + 5-FOA media. The recipe for SC-Ura + GH plates is as follows: 0.17% yeast nitrogen base without amino acids and ammonium sulfate (233520, BD, Franklin Lakes, NJ, USA), 0.1% glutamate (L817833-500g, Macklin, Shanghai, China), 2% glucose, 2 g/L DO Supplement-Ura, 200 mg/L G418 (A600958-0005, Sangon, Shanghai, China), 300 mg/L hygromycin B (H8080, Solarbio, Beijing, China), and 2% agar.

The construction of FIM-1ΔU was described previously [25]. The engineered chromosome 1 of Sc strain W303-1A was transformed into FIM-1ΔU to obtain KS-R1 [27]. Other strains used in this study were constructed using a CRISPR/Cas9 strategy with an ARS1-based CRISPR plasmid LHZ531 [23] or LHZ1660. Primers containing a 20 bp target sequence were annealed and inserted into the SapI site of LHZ531 to obtain CRISPR plasmids LHZ1654, LHZ1655, LHZ1657, and LHZ1659. The open reading frame (ORF) of natMX4, along with the TEF promoter and terminator amplified from pFA6a-A1:natMX4 (Addgene: 134428), was inserted into the SpeI and SmaI sites of LHZ531 to obtain CRISPR plasmid LHZ1660. Primers containing a 20 bp target sequence were annealed and inserted into the SapI site of LHZ1660 to obtain CRISPR plasmids LHZ1650-LHZ1653, LHZ1656, and LHZ1658. To obtain a donor sequence for gene deletion, 500 bp upstream and downstream sequences of the gene to be deleted were amplified and ligated by overlap PCR. KS-R1 was co-transformed with LHZ1650 and the *hyg^R^*Δ donor using the lithium acetate method [28], and transformants were selected in YPDN medium. A positive clone was named KS-R1-Δh. Similarly, kan^R^ was deleted in KS-R1-Δh to obtain KS-R1-Δhk using LHZ1651. KmPOP5 and KmPRP45 were deleted together in KS-R1-Δhk to obtain KS-R1-Δhkpp using LHZ1652. KmMAK16 was deleted in KS-R1-Δhkpp to obtain KS-R1-Δhkppm using LHZ1653. URA3 was deleted in KS-R1-Δhkppm by counterselection on YPD + 5-FOA plates to obtain KS-R1E.

To obtain the donor sequence to replace an R1 gene with GFP, 500 bp upstream and downstream sequences of the target gene were fused with the 5′ end and 3′ end of the GFP ORF by overlap PCR. To replace CYS3 with GFP, KS-R1E was co-transformed with LHZ1654 and the GFP donor, and transformants were selected in SC-Ura medium. Similarly, PAU7, FLC2, and ADE1 were replaced with GFP using LHZ1655, LHZ1657, and LHZ1659, respectively. To replace ERV46 with GFP, KS-R1E was co-transformed with LHZ1656 and the GFP donor, and transformants were selected in YPDN medium. Similarly, LDS1 was replaced with GFP using LHZ1658. To construct strains with GFP at two loci, in KS-R1E/lds1Δ::GFP, either CYS3 or ERV46 was replaced with an additional GFP using the CRISPR plasmid, as mentioned above. The donor sequences to replace R1 genes with P. pastoris genes were constructed in the same manner as GFP donors. In KS-R1, FLC2 was replaced with PpYPT1 using CRISPR plasmid LHZ1657. Subsequently, LDS1 was replaced with PpMPD1-2 using LHZ1658. Next, PAU7 was replaced with PpPDI1 using LHZ1655. Next, CYS3 was substituted with PpMPD1-1 via LHZ1654, and finally, ERV46 was replaced with PpMPD1-3 using LHZ1656. The CRISPR plasmid was removed before the next gene replacement. The substitution efficiency at each step was measured. The resultant strain containing five substitutions was named KS-R1E5P.

Plasmids for expressing BadGLA (LHZ1020) and TeGlaA (LHZ1021) were described previously [8].

2.2. Growth Curves and Chromosome Stabilities

KS-R1, KS-R1-Δhkpp, and KS-R1-Δhkppm were cultured in SC-ura liquid medium, while KS-R1E was cultured in YPD liquid medium overnight. The overnight cultures were then diluted into YPD medium to an initial OD_600_ of 0.2 and grown at 30 °C for 72 h, during which OD_600_ measurements were recorded. To assess the chromosome stability of KS-R1 and KS-R1E, the cultures were cultivated at either 30 °C or 37 °C. Every 24 h, the cultures of KS-R1 and KS-R1E were diluted into fresh YPD medium with an initial OD_600_ of 0.2. After five days of continuous passage (approximately 7 generations per day), the cultures of KS-R1 were diluted and plated onto YPD and SC-Ura + GH plates. The stability of R1 was determined by dividing the colony-forming unit count on SC-Ura + GH plates by that on YPD plates. For KS-R1E, the maintenance of R1 in eight different clones was verified by PCR of seven position markers.

2.3. RNA-Seq Data Analysis

RNA-seq data were extracted from a previous study [27]. Cells were grown in SC-Ura + GH medium overnight, then diluted into 50 mL fresh YPD medium to an initial OD_600_ of 0.2. When the OD_600_ reached 0.6, the cells were collected and washed with ddH₂O. To induce stress, the medium was supplemented with tunicamycin at a final concentration of 1 μg/mL or NaCl at a final concentration of 1 M. The cells were treated for 1 h, collected, washed with ddH_2_O, and then subjected to RNA extraction and RNA-seq [27]. To identify gene loci exhibiting stable transcriptional activities, the coefficients of variation (CV) were calculated by dividing the standard deviation by the mean. Sc-specific genes were identified previously [27]. The relative FPKM level refers to the ratio of the average FPKM value of an R1 gene across three conditions to the average FPKM value of three housekeeping genes (SWC4, TRK1, MPE1) across the same conditions.

2.4. Measurement of GFP Gene Expression

Cells were grown in SC medium overnight, then diluted into fresh SC medium to an initial OD_600_ of 0.2. To induce stress, the medium was supplemented with tunicamycin at a final concentration of 0.2 μg/mL or NaCl at a final concentration of 0.5 M. Cells were then cultured at 30 °C and 220 rpm for 24 h. A total of 100 μL of culture was collected, and the fluorescence intensities were measured at 485–528 nm using a Synergy2(BioTek, Winooski, VT, USA). The relative green fluorescence intensity was calculated as the ratio of green fluorescence intensity to OD_600_. For the subculturing of cells with GFP at two loci, an overnight culture was diluted into fresh SC medium with an initial OD_600_ of 0.2, and the cells were then grown at 30 °C. Dilution was performed every 24 h. After three days of continuous passage, the culture was collected to measure GFP intensity. The culture was also restreaked onto YPD medium to obtain single clones for PCR assays. The experiments were performed with three parallel cultures.

2.5. qPCR

KS-R1E strains with GFP integrated at different loci and KS-R1E5P were cultured overnight in YPD liquid medium. The cultures were diluted into YPD to an initial OD_600_ of 0.2. After 12 h of cultivation, 1 mL of the cultures was collected and washed with ddH_2_O. Total RNA was extracted using the ZR Fungal/Bacterial RNA MiniPrep kit (R2014, ZymoResearch, Irvine, CA, USA). Samples were reverse transcribed using HiScript III All-in-One RT SuperMix Perfect for qPCR (R333-01, Vazyme, Nanjing, China). The qPCR was performed using a ChamQ Universal SYBR qPCR Master Mix (Q711-02, Vazyme, Nanjing, China). The mRNA level of a specific gene was normalized to the mRNA level of a housekeeping gene, SWC4. The primers used in qPCR are listed in Supplementary Table S3. The experiments were performed with three parallel samples.

2.6. Enzymatic Assays

KS-R1E was transformed with LHZ1020 or LHZ1021, and transformants were selected on SC-Ura plates. Cells were grown in YD liquid medium for 72 h. We collected 100 μL of supernatant for enzymatic assays. The enzymatic activities in the supernatant were determined as described previously [8]. In brief, the activities of BadGLA and TeGlaA were determined by using 1% (w/v) soluble starch as a substrate. A total of 5 µL of diluted supernatant was mixed with 45 µL of 20 mM NaAc (pH 5.5) and 50 µL of 1% soluble starch in 20 mM NaAc (pH 5.5). After 5 min incubation at 60 °C, 100 µL of DNS (D7800, Solarbio, Beijing, China) was added, and the mixture was incubated at 98 °C for 5 min. Absorbance at A540 was then measured with 100 µL of the sample. One unit of activity of these enzymes was defined as the amount of enzyme required to release 1 µmol glucose per minute [21].

2.7. Statistical Analysis

Each assay subjected to statistical analysis comprised three biological replicates. Statistical significance was determined using a t-test. A p-value of < 0.05 was considered statistically significant.

3. Results

3.1. Maintenance of R1 by Complementation Between Essential Genes on R1 and Deletion of Their Counterparts in K. marxianus

As described in our prior work [27], chromosome 1 was circularized to obtain R1 by removing the telomeres and joining the ends with KmURA3 using CRISPR/Cas9. During the circularization, the telomere-associated long repetitive sequence on chromosome 1 was also deleted to prevent homologous recombination [29]. To ensure effective selection of R1, the R1 carries three selective markers: hyg^R^, kan^R^, and URA3 (Figure 1A). When KS-R1 grows in non-selective YPD medium, the loss rate of R1 is low, at only 0.11% per generation [27]. However, to maintain R1 during long-term culturing or under stress conditions, it is necessary to add two antibiotics (hygromycin and G418) to a nutrient-deficient medium (SC-Ura) for screening, which increases cultivation costs. Moreover, R1 occupies three common genetic markers, potentially affecting further engineering of KS-R1.

To stabilize R1 without relying on selective markers, we plan to utilize the essential genes on R1 to functionally compensate for the deletion of their counterparts in the K. marxianus genome. Currently, the essential genes of K. marxianus have not been fully annotated; therefore, we inferred these essential genes based on those present on R1. In total, there are 12 S. cerevisiae essential genes on R1 [30]. We compared the protein sequence identity and expression levels between these essential genes and their K. marxianus orthologs (Figure 1B). From this comparison, we selected three K. marxianus genes for deletion: KmPOP5, KmPRP45, and KmMAK16. ScMak16 and KmMak16 share a high identity and a low evolutionary rate, suggesting that their functions are likely conserved. KmPOP5 and KmPRP45 exhibited expression levels similar to those of ScPOP5 and ScPRP45, indicating that the regulatory mechanisms of POP5 and PRP45 are relatively conserved. In addition, since KmPOP5 and KmPRP45 are adjacent in the K. marxianus genome, both genes can be deleted simultaneously, which simplifies the engineering process.

First, we sequentially deleted the hyg^R^ and kan^R^ markers on R1 using CRISPR. The deletion of these two markers was confirmed by PCR (Figure 1C). The resultant strain was named KS-R1-Δhk. Subsequently, we simultaneously deleted KmPOP5 and KmPRP45 in the KS-R1-Δhk strain (Figure 1C). The resultant strain was named KS-R1-Δhkpp. Next, we deleted KmMAK16 in the KS-R1-Δhkpp strain, resulting in KS-R1-Δhkppm (Figure 1C). During the construction of R1, we installed direct repeats flanking the URA3 gene [27]. Recombination between these direct repeats would remove the URA3 marker. Therefore, we used 5-FOA to counterselect ura3 mutants in KS-R1-Δhkppm. The loss of URA3 in the positive clone was validated by PCR (Figure 1C). Subsequently, we confirmed that the engineered R1 retained all seven position markers (Figure 1C). These results indicate that removing all three selective markers from R1 did not affect its integrity. The resultant strain was named KS-R1E (E for “maintenance by essential genes”).

When a Km essential gene is deleted, its function is provided by its ortholog on R1. Therefore, the growth status of the engineered strains reflects the extent of functional complementation between the Km gene and its Sc ortholog. To investigate this, we compared the growth curves of the KS-R1 strain with those of KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E in YPD medium. The results showed that, compared to KS-R1, the initial growth rates of KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E were slightly slower. However, in the later phases, the biomass of all three engineered strains was close to that of KS-R1 (Figure 1D). These findings indicate that ScPOP5, ScPRP45, and ScMAK16 can effectively compensate for the deletion of their Km counterparts, thereby maintaining the normal growth of the hybrid yeast.

In KS-R1E, all selective markers on R1 have been removed. The maintenance of R1 relies on functional compensation for the deletion of essential Km genes by their counterparts on R1. Theoretically, this compensation can maintain R1 without additional selective pressure. To verify this idea, we measured the chromosome stability of R1 in KS-R1 and KS-R1E in non-selective YPD medium (Figure 1E). The results showed that at 30 °C, the stability of R1 in KS-R1 was 91.9% after 5 days (approximately 35 generations), consistent with our previous reports [27], while the loss rate of R1 was zero in KS-R1E. At 37 °C, the stability of R1 in KS-R1 was 28.7% after 5 days, indicating that high temperature causes unstable inheritance of R1. In contrast, the loss rate of R1 in KS-R1E remained zero. These results indicate that the strategy of essential gene complementation can efficiently maintain R1 under both normal and stress conditions.

Complementation between essential genes on R1 and deletion of their counterparts in K. marxianus. (A) Schematic view of R1 engineering. (B) List of Sc essential genes on R1. Expression levels were indicated by read counts in the transcriptomic analysis. The evolutionary rate was obtained from OrthoDB [31]. (C) Identification of genotypes for KS-R1-Δhk, KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E. Three selective markers on R1 (URA3, kanR, hygR), seven R1 position markers, as well as the KmPOP5, KmPRP45, and KmMAK16 loci in the Km genome were characterized by PCR. The positions of primers relative to the loci and the expected sizes of the PCR products are shown in Figure S1A. (D) Growth curves of KS-R1, KS-R1-Δhkpp, KS-R1-Δhkppm, and KS-R1E in YPD medium. Cells were grown in YPD for 72 h, and OD600 was measured at the indicated times. Values represent the mean (n = 3). (E) Stability of R1 in KS-R1 and KS-R1E. Cells were grown in YPD for 5 days at 30 °C or 37 °C. The stability of R1 in KS-R1 was determined by the ratio of colonies on SC-Ura plates to those on YPD plates. The stability of R1 in KS-R1E was measured by the percentage of colonies maintaining all seven position markers via PCR. Values represent the mean ± standard deviation (SD) (n = 3).

3.2. Characterization of Gene Loci on R1 Displaying Stable Transcriptional Activities

To identify suitable sites on R1 for gene insertion, we analyzed the transcriptomic data of R1 genes under three conditions: YPD medium, YPD containing 1 μg/mL tunicamycin, and YPD containing 1 M NaCl (Figure 2A). Among the 89 genes, 63 genes have orthologs in K. marxianus, while the remaining 26 genes are considered Sc-specific genes. Km and Sc orthologs were identified using OrthoVenn3 [32]. The expression levels of orthologous R1 genes were not significantly different from those of Sc-specific genes (Figure 2B). To identify the gene loci exhibiting stable transcriptional activities, the coefficients of variation (CV) of expression levels across the three conditions for each R1 gene were calculated (Figure 2A). The CVs of Sc-specific genes were significantly higher than those of orthologous R1 genes. This might be due to the faster divergence of cis-elements in Sc-specific genes, leading to greater fluctuations in gene expression under various conditions (Figure 2B).

We focused on genes that maintained stable expression across the three conditions (CV < 0.5), as these gene loci might offer consistent expression for inserted genes. Among 63 orthologous R1 genes, 46 genes (73%) were stably expressed. Among the Sc-specific genes, 13 genes (50%) were stably expressed, yielding a total of 59 stably expressed genes. Compared to the average expression level of three housekeeping genes (SWC4, TRK1, and MPE1), 2 of the stably expressed genes were high-expression genes (ratio > 10), 39 were medium-expression genes (1 < ratio < 10), and 15 were low-expression genes (ratio < 1) (Figure 2C). ScPOP5, ScPRP45, and ScKmMAK16 are the remaining three stably expressed genes. Since they are required to maintain the growth of KS-R1E, their loci are not suitable for insertion and are not listed in Figure 2C. Therefore, the alien chromosome R1 contains 56 gene loci with varying and stable transcriptional activities, offering valuable insertion sites for multiple heterologous gene expressions.

Next, we compared the transcription levels and stability of the stably expressed R1 genes with their Km orthologs. Compared to the Km orthologs, 25 stably expressed R1 genes displayed higher expression levels (Figure 2C, labeled in red). This might be due to the inactivation of repressive elements of these R1 genes in the Km background, leading to subsequent derepression. Meanwhile, compared to their Km orthologs, 27 R1 genes exhibited greater stability, as reflected by their lower CV values (Figure 2C, underlined). In these R1 genes, stress-responsive cis-elements might not be functional in Km, resulting in stable expression under various stress conditions. These R1 gene loci are likely to provide more stable expression for inserted genes than the orthologous Km gene loci.

3.3. GFP Replacing R1 Genes at Different Loci Displayed Various Expression Levels

To verify whether loci on R1 can serve as insertion sites for heterologous genes, we inserted GFP into six loci with varying transcriptional activities: CYS3 and PAU7 (strong), ERV46, FLC2, and LDS1 (medium), and ADE1 (weak) (Figure 3A). We constructed a GFP donor containing the GFP ORF flanked by 500 bp of upstream and downstream sequences of the target R1 gene ORF. KS-R1E was transformed with a plasmid co-expressing gRNA and Cas9 along with the GFP donor. Through recombination between the donor and the R1 loci, the GFP ORF replaced the target gene ORF in situ. Subsequently, GFP expression was driven by the native promoter and terminator of the targeted R1 gene (Figure 3B).

We verified GFP insertion using one primer upstream of the targeted locus and another primer within the GFP ORF. Positive transformants yielded PCR bands of the expected size (Figure 3C). Next, we measured the mRNA levels of GFP expressed from different loci. In YPD medium, the mRNA levels of GFP expressed from the CYS3, PAU7 (strong), ERV46, FLC2, and LDS1 (medium) loci were higher than those at the ADE1 (weak) locus (Figure 3D). The mRNA level of GFP expressed from the CYS3 locus was lower than that from PAU7, and the mRNA level of GFP expressed from ERV46 was lower than that from LDS1 and FLC2 (Figure 3D). This pattern differs from the ranking of transcriptional levels at these loci shown in Figure 2C. This discrepancy might be due to the fact that the transcriptional level of a locus is influenced not only by the promoter and terminator but also by mRNA sequence characteristics, such as mRNA stability [33,34]. Subsequently, we measured the intensity and stability of GFP fluorescence under different conditions (Figure 3E). SC medium was used in these experiments to reduce the GFP background signal observed in YPD medium. In SC medium, the ranking of GFP fluorescence intensities from different loci was consistent with the ranking of their mRNA levels (Figure 3E). The differences in GFP intensity between strains could be observed directly using a fluorescence microscope (Figure 3E, upper panel). Notably, the coefficient of variation (CV) of GFP levels across the six loci ranged from 0.2 to 0.4 under the three conditions, reflecting stability in expression (Figure 3F). These results demonstrate that R1 loci can achieve stable expression of inserted genes at varying levels.

In K. marxianus, transcriptional activities of gene loci with strong promoters, such as TEF3, PDC1, or HXT4, are 44~488 times the activity of the housekeeping SWC4 locus in YPD [21,27]. In comparison, CYS3, PAU7, ERV46, LDS1, and FLC2 loci exhibited transcriptional activities 34, 31, 13, 5, and 11 times that of the SWC4 locus in YPD, respectively. Therefore, a single locus on R1 might not be sufficient to achieve the required level of overexpression. To address this issue, a heterologous gene can be inserted at multiple stable loci to achieve higher expression. To validate this idea, in the strain with GFP integrated at the LDS1 locus, an additional GFP gene was integrated at either the CYS3 or ERV46 locus. The fluorescence intensity of strains with GFP at two loci was significantly higher than that of strains with GFP at a single LDS1 locus, reflecting a roughly additive effect (Figure 3G). Next, the stability of GFP loci in strains carrying dual GFP was measured, since recombination might occur between identical ORFs. Strains were subcultured in SC medium for 3 days (approximately 21 generations), and the fluorescence intensity of the culture remained the same as that of the original culture, suggesting that both GFP copies were conserved (Figure 3G). We also examined the status of GFP loci in three clones restreaked from each parallel subculture. The results showed that GFP integrated at the LDS1, CYS3, and ERV46 loci was retained in all clones after subculturing. The results for one representative clone are shown in Figure 3H. These findings indicate that the integration of the same ORF at two loci improved expression levels without readily causing rearrangement of R1.

3.4. Substitution of R1 Genes by P. pastoris Disulfide Bond Formation Genes Improved the Expression of Heterologous Proteins

To verify whether loci on R1 can be used for the co-expression of multiple genes, we sequentially replaced five R1 genes with five P. pastoris disulfide bond formation genes using a strategy similar to the GFP substitution (Figure 4A). The ORFs of CYS3, LDS1, ERV46, PAU7, and FLC2 were replaced with the ORFs of PpMPD1-1, PpMPD1-2, PpMPD1-3, PpPDI1, and PpYPT1, respectively (Figure 4B). Successful replacement of the R1 genes by P. pastoris genes was verified by PCR (Figure 4C). This engineering did not impair the integrity of R1, as R1 retained all the position markers and deletion of selective markers (Figure 4D). The substitution efficiencies at the CYS3, PAU7, FLC2, ERV46, and LDS1 loci were 50%, 25%, 33.3%, 45.8%, and 62.5%, respectively (Figure 4E), indicating that these loci can stably integrate different heterologous genes. The resulting strain was named KS-R1E5P.

We performed qPCR to measure the expression levels of the P. pastoris genes in KS-R1E5P. The results showed that the expression levels of PpMPD1-1, PpMPD1-2, PpMPD1-3, PpPDI1, and PpYPT1 were, respectively, 43, 5, 12, 19, and 8 times those of the control KmSWC4 gene (Figure 4F). Notably, the relative mRNA level of the P. pastoris gene expressed from the CYS3 or ERV46 locus was higher than that of GFP mRNA expressed from the same locus (Figure 3D), reflecting the influence of the ORF sequence on mRNA levels.

Finally, we tested the effects of the insertion of multiple genes on the expression of heterologous proteins. We compared the expression of two glucoamylases, BadGLA and TeGlaA, in the FIM-1ΔU and KS-R1E5P strains. BadGLA and TeGlaA were predicted to contain one and four pairs of disulfide bonds, respectively [8]. These structural features require a robust disulfide bond formation machinery, making them ideal candidates for evaluating the impact of co-expressing P. pastoris genes involved in disulfide bond formation. Genes encoding BadGLA or TeGlaA were carried by a multicopy plasmid containing a URA3 marker. Because KS-R1E5P did not occupy the URA3 marker, it readily accepted these plasmids. The results showed that the secretory activities of BadGLA and TeGlaA in KS-R1E5P were significantly higher than those in FIM-1ΔU (Figure 4G). These findings indicate that the alien chromosome from S. cerevisiae can serve as a novel platform to provide sites for the expression of multiple heterologous genes, thereby enhancing the performance of the host.

4. Discussion

Our results indicate that an alien chromosome can serve as a novel platform for the expression of multiple heterologous genes. Compared to inserting heterologous genes into the genome, integrating them onto an alien chromosome does not disrupt the native genomic structure and avoids affecting the expression of nearby native genes. Plasmids and artificial chromosomes are two other options for heterologous gene insertion, but alien chromosomes offer greater stability. In this study, we maintained the alien chromosome using functional compensation between essential genes on the alien chromosome and deletion of their orthologs in the genome, resulting in a near-zero loss rate of the alien chromosome in non-selective conditions. The KS-R1E strain containing the modified alien chromosome, compared to the parental KS-R1 strain, showed a slightly reduced initial growth rate. However, biomass between KS-R1E and KS-R1 in the later growth phase was comparable. This result suggests that although the common ancestor of S. cerevisiae and K. marxianus diverged 115 million years ago [35], the regulatory expression and protein functions of Sc genes such as POP5, PRP45, and MAK16 remain highly conserved with their Km counterparts, allowing these Sc orthologs to effectively compensate the deletion of their counterparts. In future research, replacing ScPOP5, ScPRP45, and ScMAK16 on R1 with the corresponding Km genes may achieve complete functional compensation, and the slight growth defect of KS-R1E is expected to be redeemed.

Regardless of whether multiple heterologous genes are expressed via genome integration, plasmids, or artificial chromosomes, each gene requires suitable promoters and terminators. A major advantage of an alien chromosome is that it provides a series of loci with varying transcriptional activities, along with their native cis-regulatory elements. Typically, a gene cassette for replacement through recombination is composed of an upstream homologous arm, a promoter, a heterologous gene ORF, a terminator, and a downstream homologous arm [36]. On an alien chromosome, the regions flanking gene ORFs are natural promoters and terminators, so the composition of the gene cassette simplifies to: upstream homologous arm, heterologous gene ORF, downstream homologous arm (Figure 3B and Figure 4A). Our results showed that after in situ replacement of R1 genes with P. pastoris genes, the original transcriptional activities of the loci were almost fully recapitulated (Figure 4F). These findings demonstrate the effectiveness of R1 loci in providing differential expression of multiple genes.

This study focused on loci with stable expression levels. In addition to these stable loci, some genes show altered expression under specific stress conditions—for example, KIN3 is specifically induced by NaCl, while RFA1 and PEX22 are specifically highly expressed in the presence of tunicamycin (Figure 2A). These loci might be used for condition-dependent expression.

In summary, alien chromosomes not only provide suitable insertion sites for multi-gene expression in the K. marxianus but also offer a variety of cis-expression control elements like promoters and terminators with different activities. This new platform facilitates the engineering and optimization of K. marxianus in metabolic engineering and synthetic biology.