The strains used in the present study are listed in Table I (25,26). All strains constructed in this experiment were engineered with fluorescent labels via spore production, as later described.

YE5S Liquid Medium was prepared by adding 2.5 g Yeast Extract (Thermo Fisher Scientific, Inc.), 0.01125 g Amino Acids (Ade/Leu/Ura/His/Lys) (Merck KGaA), 15 g Dextrose (Shanghai Aladdin Biochemical Technology Co., Ltd.) and 500 ml ultrapure Water. YE5S Solid Medium was prepared by adding 8.5 g agar (BioFroxx; neoFroxx GmbH) to the liquid medium. G418, HygR or NatMx Resistance Selection Medium were prepared by adding 0.3 mg/ml G418 (Guangzhou Saigou Biotech Co., Ltd.), 0.3 mg/ml HygR (Shanghai Macklin Biochemical Co., Ltd.) or 0.1 mg/ml NatMx (Beijing Solarbio Science & Technology Co., Ltd.) to YE5S solid medium. Leu-deficient medium was prepared by removing Leu from YE5S solid medium. 1X snail enzyme solution was prepared by mixing 100X snail enzyme buffer with 10 g/ml snail enzyme (Shanghai Yuanye Bio-Technology Co., Ltd.) and 990 µl ultrapure water. The 100X Snail Enzyme Buffer contains 1 mol/l sorbitol (Shanghai Yuanye Bio-Technology Co., Ltd.), 100 mmol/l EDTA (Shanghai Macklin Biochemical Co., Ltd.) and 14 mmol/l β-mercaptoethanol (Merck KGaA). All media and reagents were prepared at room temperature and sterilized prior to use.

The wild-type strain with fluorescent marker (PT2514, PT3850, YL20, YL24, YL26) and the opposite mating-type rok1Δ strain (2125-A or 2125-B) (Table I) were inoculated onto YE5S solid medium and were activated at 25˚C for 3 days. The activated h⁺ and h^- strains were then mixed on EMM-N nitrogen-deficient medium for sporulation (27) and incubated at 25˚C for 2 days. Colonies were picked and examined using an OLYMPUS BX51 microscope (Olympus Corporation) to monitor sporulation. Upon confirmation of spore formation, the cells were treated with 1 ml prepared 1X snail enzyme working solution (28,29) to release the spore suspension. The suspension was spread onto YE5S solid medium and incubated at 25˚C for 2 days to allow spore germination and single colony formation. Colonies were subsequently transferred using the replica plating method onto nutrient-deficient or antibiotic-containing media (30). Positive clones obtained were expanded through serial passages, stored with 30% glycerol as cryoprotectant in an 80˚C ultra-low temperature freezer (Thermo Fisher Scientific, Inc.) for future use.

Live cell imaging of the experimental strains was performed using a TCS-SP8 laser confocal microscope (Leica Microsystems GmbH) at 25 and 37˚C. The parameters were set as follows: Green fluorescence received in the wavelength range of 493-545 nm, red fluorescence received in the wavelength range of 584-736 nm, pixels of 512x512 µm, seven optical slices with a pitch of 6.02 µm, exposure time of 400 msec, shooting interval of 2 min and total shooting time of 120 min.

Measurements of the dynamics of cellular mitotic spindle, actin, kinetochore, and centrosome proteins were analyzed using ImageJ (version 1.51s; National Institutes of Health). Potential outliers were identified and excluded based on the Z-score method. The Z-score was calculated as Z=(χ-µ)/σ, where χ is the measured value, µ is the mean of measurements and σ is the population standard deviation. Data points with Z>2 were considered outliers (31). To ensure data consistency across all metrics including microtubule, actin, kinetochore and centrosomal protein data, each initially collected from 20 cells (No. 1-No. 20), if a cell was identified as an outlier in any metric, all measurement data from that cell were excluded.

For example, using actin ring data from rok1Δ strains cultured at 25˚C, six metrics were analyzed: Actin ring length, total time from assembly to complete disappearance, assembly time, contraction time, total contraction rate from assembly to complete disappearance and contraction rate during the contraction phase. After Z-score normalization of measurements from the 20 cells, three outliers (Z>2) were identified: The contraction rate of the No. 4 cell and both the total duration and assembly time of the actin ring of the No. 14 cell. Consequently, all six parameters from the No. 4 and the No. 14 cells were excluded. Following this procedure, data from 5 cells were excluded for each metric, ultimately retaining a uniform set of 15 cells per metric for subsequent analysis.

Data analysis was performed using SPSS Statistics 26 (International Business Machines Corporation). The Shapiro-Wilk test assessed data normality, and the unpaired Student's t-test analyzed differences between wild-type and rokI∆ strains. Differentially expressed genes were identified using an adjusted P<0.05, log₂ fold change >0(32) as screening criteria. RT-qPCR data were analyzed using the 2^-∆∆Cq relative quantification method (33) to describe the transcriptional expression changes of target genes in the rok1Δ strain relative to the wild-type strain. P<0.05 was considered to indicate a statistically significant difference.

The PT287 and 2125 strains were cultured at 25˚C and 37˚C until the cells entered the logarithmic growth phase, and then the organisms were collected and frozen. Total RNA was extracted using the Yeast total RNA kit (Omega Bio-Tek, Inc.) according to the manufacturer's standard protocol, and sent to Beijing Novogene technology Co., Ltd. for quality testing and RNA-Seq. Reference genes were compared using Hisat2 v2.0.5 software (34); gene expression was quantified using Feature Counts and Stringtie (1.3.3b) software (35); and differential significance of gene expression of samples was analyzed using DESeq2 (1.20.0) and EdgeR (3.22.5) software (36). Differential genes were processed by Gene Ontology (GO) function enrichment (37), Kyoto Encyclopedia of Genes and Genomes (KEGG) (38) pathway enrichment analysis was performed using Cluster Profile software (version 3.8.1) (39).

PT287 and 2125 strains were cultured at 25 and 37˚C until the cells reached the logarithmic growth phase, then collected and frozen. Total RNA was extracted using the Yeast Total RNA Kit and reverse-transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Inc.) at 42˚C for 60 min. The qPCR conditions were as follows: 95˚C pre-denaturation for 30 sec, followed by 39 cycles of 95˚C denaturation for 5 sec and 60˚C annealing/extension for 30 sec. The melting curve analysis stage consisted of 95˚C for 10 sec, 65˚C for 5 sec and 95˚C for 5 sec. The act1 gene was used as the internal control, and qPCR was performed using the CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.) to analyze gene expression levels. The primer sequences used are listed in Table II.

The growth rates of the wild-type strain and the rok1Δ strain were compared and the results showed that there was no significant difference in growth rates between the two over the period of 0-4 h at 25˚C and 37˚C (all P>0.05). After 6 h, the wild-type strain entered the logarithmic growth phase and the growth rate of the rok1Δ strain significantly decreased. At 25˚C, optical density (OD)₅₉₅ at 12 h was 0.68±0.00 and 0.33±0.02 for the wild-type and rok1Δ strains, respectively. At 37˚C, OD₅₉₅ at 12 h was 0.89±0.01 and 0.61±0.07 for the wild-type and rok1Δ strains, respectively (Fig. 2A and B), which were significantly different (P<0.01). The aforementioned results indicated that rok1 gene deletion significantly decreased the cell growth rate.

The results of spore number production showed that 99.33±0.31% of the wild-type strains produced 4 spores, whereas the percentage of the rok1Δ strain producing 4 spores was 96.60±0.92%, which was a significant difference. In addition, there were 2.00±0.72% of the rok1Δ strain producing 3 spores (Fig. 2C and D). The results showed that the rok1Δ strain produced an abnormal number of spores compared to the wild-type strain, with a significant decrease in the number of spores. It is noteworthy that the high temperature of 37˚C significantly impacts spore production. Previous studies have reported that at 37˚C, gene-deleted fission yeast either fail to produce spores or exhibit a low probability of spore formation (40). Therefore, spore production was investigated exclusively at 25˚C.

A significant decrease in the growth rate of the rok1Δ strain was found by growth rate measurements; therefore, the mitosis-related protein dynamics of the rok1Δ strain was investigated.

During mitosis, microtubule proteins assemble to form the spindle, which is responsible for the accurate segregation of chromosomes during cell division (41). Live-cell imaging of wild-type and rok1Δ cells with green fluorescent protein (GFP)-Atb2-tagged microtubule proteins was used to assess spindle properties. The analysis of spindle dynamics at 25˚C showed that the spindle elongation lengths of the wild-type strains at the prophase, metaphase and anaphase were 0.68±0.27, 0.63±0.23 and 8.82±0.62 µm and for the rok1Δ strain they were 0.75±0.22, 0.61±0.19 and 8.57±0.80 µm, respectively (Fig. 3A-D). The rok1Δ strain exhibited no significant changes in spindle elongation length compared to the wild-type strain in all phases (all P>0.05). Further analysis revealed (Fig. 3E and F) that the spindle elongation times of the wild-type strains at prophase, metaphase and anaphase were 3.33±1.45, 4.80±1.47 and 11.07±1.03 min and for the rok1Δ strain they were 4.27±1.83, 5.07±1.67 and 10.40±1.12 min, respectively. The spindle elongation rates of the wild-type and rok1Δ strains at the prophase, metaphase and anaphase were 0.22±0.10, 0.14±0.05 and 0.80±0.10 µm/min and 0.19±0.08, 0.13±0.04 and 0.83±0.09 µm/min, respectively. There was no significant difference in the elongation time and elongation rate of the spindle in all periods between the wild-type and rok1Δ strains (all P>0.05). These results indicate that the deletion of the rok1 gene does not affect spindle dynamics at 25˚C.

The spindle elongation lengths at 37˚C showed that compared with the wild-type there was a significant decrease in the spindle elongation length of the rok1Δ strains at anaphase (10.38±1.02 vs. 9.33±0.83 µm, respectively); however, there were no significant changes in spindle elongation length at prophase or metaphase. Further analysis revealed that this trend was consistent for the spindle elongation time with wild-type strains displaying significantly longer elongation times compare with rok1Δ strains at anaphase (12.13±2.56 vs. 10.40±1.12 min, respectively), with no significant changes observed at prophase and metaphase (P>0.05) (Fig. 3E). In addition, there was no significant difference in spindle elongation rates between the wild-type and rok1Δ strains at prophase (0.26±0.06 vs. 0.23±0.05 µm/min, respectively), metaphase (0.10±0.06 vs. 0.08±0.05 µm/min, respectively) and anaphase (0.88±0.17 vs. 0.90±0.09 µm/min), respectively (both P>0.05) (Fig. 3F).

These results indicate that the rok1 gene deletion decelerated the spindle elongation process at 37˚C. Comparison of 25˚C and 37˚C revealed that a high temperature stress of 37˚C had an inhibitory effect on the spindle elongation process.

During mitosis, actin filaments co-assemble with myosin II to form an actomyosin ring that help the cell complete cytoplasmic division (42). Live-cell imaging was performed of wild-type and rok1Δ cells with Pact1-LifeAct-mGFP-tagged actin. Analysis of actin rings at 25˚C indicated that the actin ring length was significantly decreased (3.75±0.17 vs. 3.52±0.15 µm), total time from assembly to complete disappearance was significantly increased (28.70±2.84 vs. 31.40±1.29 min) and total contraction rates were significantly decreased (0.11±0.02 vs. 0.09±0.01 µm/min), in the rok1Δ strain compared with the wild-type stain (Fig. 4A-F). Further analysis results indicated that the assembly times of the actin rings in the wild-type and rok1Δ strains were 13.31±2.46 and 16.70±1.78 min, respectively, which was a significant difference (Fig. 4E). Entering the contraction phase, the actin rings of the wild-type and rok1Δ strains contracted for 15.39±0.85 and 14.70±1.92 min, respectively, and the contraction rates were 0.19±0.02 and 0.18±0.02 µm/min, respectively (Fig. 4F); there were no significant differences in contraction time and contraction rate (both P>0.05). These results indicate that rok1 gene deletion prolongs the actin ring assembly process at 25˚C.

At 37˚C the contraction lengths were 4.09±0.38 vs. 3.50±0.26 µm and the total contraction rate of the actin ring was 0.12±0.02 vs. 0.09±0.01 µm/min for the wild-type and rok1Δ strains, respectively. Both actin ring length and total contraction rate were significantly decreased in the rok1Δ strain compared with the wild-type. However, the actin ring assembly time was 11.07±2.25 and 11.20±1.47 min for the wild-type and rok1Δ strains, respectively, which was not a significant difference (Fig. 4A-F). Entering the contraction phase, the contraction time and rate of the actin ring for the wild-type and rok1Δ strains were 15.60±3.22 vs. 15.33±1.95 min and 0.20±0.03 vs. 0.17±0.03 µm/min, respectively. The actin ring contraction rate of the rok1Δ strain was significantly decreased compared with the wild-type, and there was no significant difference in the contraction time (Fig. 4E and F). These results indicate that rok1 gene deletion affected the contraction process of actin rings at 37˚C. Comparison of results at 25˚C and 37˚C revealed that rok1 gene deletion both decreased the initiation length and contraction length of the actin ring and decreased the contraction rate. Unlike at 37˚C, rok1 gene deletion prolonged the actin ring assembly process at 25˚C.

The kinetochores interact with spindle microtubules to assist in the correct segregation of chromosomes during mitosis. Live-cell imaging of wild-type and rok1Δ cells with Mis12-GFP-tagged kinetochore proteins. The analysis of kinetochore dynamics at 25˚C showed that the time from the separation to localization at the spindle poles of kinetochores in the wild-type and rok1Δ strains was 6.67±0.07 vs. 8.40±0.83 min, respectively, which was a significant difference (Fig. 5A-D). Further analysis showed that the separation distance (9.44±0.78 vs. 10.06±0.67 µm) and separation time (11.87±1.60 vs. 13.87±2.07 min) were significantly increased in rok1Δ strains compared with wild-type, but there was no significant change in separation rates (0.81±0.11 vs. 0.74±0.10 µm/min) of kinetochores (Fig. 5E-G). These results indicate that rok1 gene deletion prolonged the separation distance and separation time of kinetochores as well as the time for kinetochores to reach the poles of the spindle at 25˚C.

The time from the separation to localization at the spindle poles of kinetochores in the wild-type and rok1Δ strains was 5.60±1.12 and 6.63±0.92 min, respectively, at 37˚C, which was a significant difference (Fig. 5A-D). Further analysis indicated that the kinetochores separation distances of wild-type and rok1Δ strains were 10.08±0.78 and 10.45±0.08 µm, respectively, which was not significant difference (Fig. 5E). The separation time was significantly increased (11.60±1.12 vs. 13.73±2.12 min; Fig. 5F) and the rate of kinetochores was significantly decreased (0.88±0.09 vs. 0.78±0.13 µm/min; Fig. 5G) in rok1Δ strains compared with wild-type. These results indicate that rok1 gene deletion delayed the kinetochore separation process at 37˚C. At 25˚C and 37˚C, rok1 gene deletion prolonged the kinetochore segregation time, the time from the separation to localization at the spindle poles of kinetochores and thus delayed the kinetochore segregation process. The inhibition of the kinetochore segregation process was more effective with temperature stress at 37˚C.

During mitosis, centrosomes assist in the completion of microtubule assembly and influence the organization and function of the spindle. Live-cell imaging of wild-type and rok1Δ cells with Sid4-GFP-tagged centromere proteins was used to investigate changes in the centrosomes. The analysis of centrosome dynamics at 25˚C indicated that the total separation time was unchanged (19.87±0.52 vs. 19.87±1.92 min); however, the final separation distances (10.40±0.43 vs. 11.17±1.06 µm) and the total centrosome separation rates (0.52±0.03 vs. 0.56±0.02 µm/min) were significantly increased in rok1Δ strains compared with wild-type strains (Fig. 6A-F). Further analysis of the centrosome separation process revealed that the centrosome separation time of the wild-type and rok1Δ strains at the prophase were 6.00±1.07 and 5.07±1.28 min, at metaphase they were 4.13±0.52 and 4.40±1.12 min and at anaphase they were 9.73±1.03 and 10.40±1.35 min, respectively (Fig. 6E). The centrosome separation rates at prophase were 0.18±0.03 and 0.23±0.05 µm/min, at metaphase they were 0.11±0.06 and 0.16±0.08 µm/min and at anaphase the rates were 0.89±0.13 and 0.88±0.07 µm/min in the wild-type and rok1Δ strains, respectively. The centrosome separation time of rok1Δ strains at prophase was significantly decreased and the separation rate at prophase was significantly increased when compared with the wild-type. These results indicated that at 25˚C, rok1 gene deletion resulted in increased centromere separation distance and increased separation rate.

When the experiments were conducted at 37˚C, the total separation time was not significantly altered (22.80±4.20 vs. 22.53±1.77 min); on the other hand, there was a significant decrease in both the final separation distance (12.90±1.44 vs. 10.82±0.63 µm) and the total centrosome separation rates (0.57±0.06 vs. 0.48±0.04 µm/min) in the rok1Δ strains compared with wild-type. Further analysis of the centrosome separation process revealed that the centrosome separation time of the wild-type and rok1Δ strains at different phases were as follows: Prophase, 4.53±1.41 and 7.07±2.12 min; metaphase, 4.80±1.82 and 3.87±1.60 min; and anaphase, 13.47±2.88 and 11.60±1.12 min, in wild-type and rok1Δ strains, respectively. The centrosome separation rates were as follows: Prophase, 0.23±0.06 and 0.16±0.05 µm/min; metaphase, 0.19±0.06 and 0.10±0.04 µm/min; and anaphase, 0.81±0.10 and 0.79±0.10 µm/min, in wild-type and rok1Δ strains, respectively. The centrosome separation time at prophase in the rok1Δ strain was significantly increased, but the separation time at anaphase was significantly decreased compared with the wild-type. Furthermore, the separation rate of the rok1Δ strain was significantly decreased at prophase and metaphase compared with the wild-type. These results indicated that at 37˚C, rok1 gene deletion shortened the centrosome separation distance and decreased the separation rate. Comparison of results at 25 and 37˚C revealed that 37˚C high temperature stress reversed the effect of rok1 gene deletion on the centrosome segregation process.

Combining the three replicates, the total number of bases in the high-quality analyzed data for the wild-type-25˚C strain, the wild-type-37˚C strain, the rok1Δ-25˚C strain and the rok1Δ-37˚C strain were 6.80 Giga bases (G), 6.67 G, 6.88 G and 6.74 G, respectively. The Q_phred20 (percentage of bases with a phred score >20, where Phred=-10·log10(e) and e represents the sequencing error rate) (43) value of all four groups of samples was >98.00%, the Q_phred30 (percentage of bases with a Phred score >30) value was >94%, the GC content was between 41-43% and the sequencing error rate was <0.03%. These results suggest that the sequencing data were accurate and reliable and could be analyzed and studied subsequently.

Gene expression levels were quantitatively analyzed, with FPKM <1 as low or no gene expression, FPKM >1 as genes expressed, FPKM >60 as genes highly expressed and FPKM >5,000 as genes very highly expressed (35,44). The results indicated that there were 4,950, 7,757, 4,997 and 7,488 genes expressed in the wild-type-25˚C strain, the wild-type-37˚C strain, the rok1Δ-25˚C strain and the rok1Δ-37˚C strain, respectively, which accounted for 39.00, 61.11, 39.37 and 59.00% of the total genes capable of expression in fission yeast. In addition, there were 1,966, 2,181, 1,811 and 2,176 highly expressed genes, accounting for 15.17, 17.18, 14.27 and 17.14% of the total genes, respectively. From these highly expressed genes, representative genes exhibiting statistically significant expression differences between the rok1∆ strain and the wild-type strain were further selected for display (Fig. 7A and B; Tables III and IV).

Among the genes expressed to a very high level in both wild-type and rok1∆-25˚C strains, the FPKM value of the zym1 gene decreased by 1.629-fold compared to the wild-type strain (P<0.01). The FPKM values of the tdh1 and pgk1 genes increased by 1.1621- and 1.5047-fold, respectively (both P<0.01). Among the genes that were expressed at very high levels in both wild-type and rok1∆-37˚C strains, the FPKM value of the rpl3202 gene increased 1.6323-fold in the wild-type strain (P<0.05). The FPKM values of trx1 and plr1 genes were decreased by 0.4418-fold (P<0.05) and 0.5132-fold (P<0.01), respectively.

The differentially expressed genes of the wild-type strain and the rok1Δ-25˚C and rok1Δ-37˚C strains were analyzed. The results indicated that there were 3,034 and 1,695 differentially expressed genes in the rok1Δ-25˚C and rok1Δ-37˚C strains, respectively, compared with the wild-type strains, which included 1,482 and 705 upregulated genes and 1,552 and 990 downregulated genes, respectively (Fig. 7C and D).

Compared with the wild-type strain, in the rok1Δ-25˚C strain, mei2, map2, map3 and psc3 gene expression were upregulated by 1.2866-, 1.8764-, 2.7593- and 1.3973-fold, respectively (Table V). In the rok1Δ-37˚C strain, gpa2, rgs1, myo51 and blt1 gene expression was upregulated by 1.9694-, 2.1606-, 0.7366- and 1.0046-fold, respectively (Table VI). In the rok1Δ-25˚C strain, ddx27 and pas1 gene expression were downregulated by 1.7320- and 1.6148-fold, respectively (Table VII). In the rok1Δ-37˚C strain, rho5, wos2, apc14, and pas1 genes were downregulated by 1.4299-fold, 1.1209-fold, 0.7467-fold, and 1.3259-fold, respectively. arp1, dil1, cmk1 and mfr1 gene expression were downregulated by 0.9087-, 1.6737-, 1.3417- and 1.1414-fold, respectively (Table VIII). Notably, although arp1, dli1, cmk1, and mfr1 genes were all significantly downregulated, the downregulation factor for arp1 was 0.9087-fold, lower than that of the other genes (>1). This result indicates that while the deletion of the rok1 gene broadly suppresses the transcription of these genes, its inhibitory effect on arp1 is relatively weaker.

Analysis of differentially expressed genes from the transcriptomic data revealed that psc3 and psm1 were key genes at 25˚C, whereas myo51 and blt1 were key genes at 37˚C. To validate the expression changes of these genes, specific primers were designed using act1 as the reference gene and verified for specificity through the NCBI Primer-BLAST tool. The results demonstrated that in the S. pombe (taxid: 4896) genome, the primers for psc3, psm1, myo51, blt1 and act1 all specifically matched unique target sequences, confirming the validity of the primer design and their suitability for RT-qPCR experiments.

The RT-qPCR results showed that the expression levels of psc3 and psm1 were significantly upregulated at 25˚C (Fig. 8A); in addition, myo51 and blt1 were significantly upregulated at 37˚C (Fig. 8B) with rok1 deletion compared with the wild-type. These findings were consistent with the RNA-Seq data, confirming the reliability of the transcriptomic results. These results suggest that rok1 regulate the expression of these key genes, thereby influencing the dynamics of mitotic progression.

GO enrichment analysis of differentially expressed genes was performed in wild-type and rok1Δ strains, and the 10 categories with the most significant up- and downregulated differentially expressed genes were selected and plotted as bar graphs. The analysis revealed that at 25˚C, the rok1Δ strain was enriched for differential genes up to 303 GO branches (P≤0.05), including 230 biological processes, 41 cellular components and 32 molecular functions, compared with the wild-type strain. Among the biological processes, upregulated genes were enriched in ‘sister chromatid segregation’ and ‘reproductive process’. In cellular components, upregulated genes were enriched in ‘chromosomes, centromeric region’ and ‘condensed chromosome inner kinetochore’ regions. In molecular function, upregulated genes were enriched in ‘nucleoside-triphosphatase activity’ and ‘translation elongation factor activity’ (Fig. 9A). In addition, downregulated differential genes were enriched in ‘purine-containing compound metabolic process’ and ‘actin cortical patch organization’, ‘actin cortical patch’ and ‘plasma membrane’ regions in cellular components, with ‘transmembrane transporter activity’ and ‘oxidoreductase activities’ in molecular functions (Fig. 9B).

At 37˚C, the differential genes of the rok1Δ strain were enriched in 28 GO branches (P≤0.05), including 3 biological processes and 25 cellular components. compared with the wild-type strain. upregulated genes were enriched in ‘transmembrane transport’ and ‘ribosome biogenesis’ during biological processes; in cellular components, upregulated genes were enriched in the ‘ribosome’; in molecular function, upregulated genes were enriched in ‘transmembrane transporter activity’ and ‘RNA polymerase II transcription factor activity, sequence-specific DNA binding’ (Fig. 10A). In addition, downregulated differential genes were enriched in ‘protein folding ‘and ‘actin filament-based’ processes, ‘mitochondrial outer membrane’ and ‘proteasome core complex’ regions in cellular components and ‘hydrolase activity, acting on glycosyl bonds’ in molecular functions (Fig. 10B). The GO enrichment results indicate that the ribosome biogenesis process of the rok1Δ strain was affected, the mitotic cytoskeleton was abnormal and the normal cell growth and reproduction process was disrupted.

KEGG enrichment analysis was performed on the differentially expressed genes of the wild-type strain and the rok1Δ strain, and the top 10 pathways of selected upregulated genes and downregulated genes were plotted as bar graphs (Fig. 11A and B). The results indicated that 1,634 differential genes were enriched in 80 pathways in the rok1Δ-25˚C strain. The upregulated genes were enriched in pathways such as ‘ribosome’, ‘cell cycle-yeast’ and ‘proteasome’, whereas the downregulated genes were enriched in pathways such as ‘MAPK signaling pathway-yeast’ and ‘citrate cycle (TCA cycle)’. The rok1Δ-37˚C strain was enriched for 787 differential genes in 89 pathways. Upregulated genes were enriched in pathways such as ‘ribosome’, ‘ribosome biogenesis in eukaryotes’ and ‘DNA replication’, whereas downregulated genes were mainly enriched in pathways such as ‘autophagy-yeast’ and ‘protein processing in endoplasmic reticulum’. At 25˚C, the ‘cell cycle-yeast’ pathway was enriched for 27 upregulated genes (Fig. 11A). The genes that were upregulated in this pathway included spo4, rad17, psm1, mis4 and mad1. At 37˚C, the ‘autophagy-yeast’ pathway was enriched for 18 downregulated genes (Fig. 11B). The major genes with downregulated expression in this pathway included pas1, atg17, atg13, atg15 and arc1.