The Role of Hexokinases in Epigenetic Regulation: Altered Hexokinase Expression and Chromatin Stability in Yeast

Abstract Background . Human hexokinase 2 ( HK2 ) plays an important role in regulating Warburg effect, which metabolizes glucose to lactate acid even in the presence of ample oxygen and provides intermediate metabolites to support cancer cell proliferation and tumor growth. HK2 overexpression has been observed in various types of cancers and targeting HK2 -driven Warburg effect has been suggested as a potential cancer therapeutic strategy. Given that epigenetic enzymes utilize metabolic intermediates as substrates or co-factors to carry out post-translational modification of DNA and histones in cells, we hypothesized that altering HK2 expression-mediated cellular glycolysis rates could impact the epigenome and, consequently, genome stability in yeast. To test this hypothesis, we established genetic models with different yeast hexokinase 2 ( HXK2) expression in Saccharomyces cerevisiae yeast cells and investigated the effect of HXK2 -dependent metabolism on parental nucleosome transfer, a key DNA replication–coupled epigenetic inheritance process, and chromatin stability. Results . By comparing the growth of mutant yeast cells carrying single deletion of hxk1Δ , hxk2Δ , or double-loss of hxk1Δ hxk2Δ to wild-type cells, we demonstrated that HXK2 is the dominant HXK in yeast cell growth. Surprisingly, manipulating HXK2 expression in yeast, whether through overexpression or deletion, had only a marginal impact on parental nucleosome assembly, but a noticeable trend with decrease chromatin instability. However, targeting yeast cells with 2-deoxy-D-glucose (2-DG), a HK2 inhibitor that has been proposed as an anti-cancer treatment, significantly increased chromatin instability. Conclusion . Our findings suggest that in yeast cells lacking HXK2 , alternative HXK s such as HXK1 or glucokinase 1 ( GLK1 ) play a role in supporting glycolysis at a level that adequately maintain epigenomic stability. While our study demonstrated an increase in epigenetic instability with 2-DG treatment, the observed effect seemed to occur independently of Hxk2-mediated glycolysis inhibition. Thus, additional research is needed to identify the molecular mechanism through which 2-DG influences chromatin stability.


Background
Hexokinases (Hks), the rate-limiting enzyme responsible for glucose metabolism, catalyzes the phosphorylation of glucose to glucose-6-phosphate (Glc-6-P).Glc-6-P subsequently enters glycolysis or alternative metabolic pathways, serving as a vital substrate for adenosine triphosphate (ATP) production and the biosynthesis of various metabolites.There are four Hk isozymes (Hk1-4) in humans.And each has a different tissue and organ distribution, as well as distinct metabolic functions (1).Expression of some Hks is associated with worse prognosis in several tumor types.For example, HK1 plays an oncogenic role in bladder cancer (2), and HK2 enhances live cancer stemness (3).High HK2 expression has been linked to a variety kind of cancer and is associated with poor overall survival in cancer patients (4)(5)(6).For these reasons, targeting Hks has been suggested as a potential strategy for cancer therapy (7,8).In support of this approach, deleting HK2 can decrease cancer cell proliferation without prominent side effects in animal models, suggesting HK2 is a promising target for cancer therapy (5).However, the mechanism by which altered HK expression on other non-glycolysis cellular functions is poorly understood.To learn more about the utility of targeting HKs, it is important to characterize the effects of modulating HK expression on eukaryotic cells.
As a key node in glucose metabolism and the main producer of energy, acetyl-coenzyme A (Acetyl-CoA) is also the substrate of globe histone acetylation (9).Furthermore, Epigenetic enzymes employ several glucose metabolic intermediates as co-factors to carry out post-translational modi cations of DNA and histones (10).Thus, altering the expression of cellular metabolism genes, such as the HKs, may change global epigenetic states.This, in turn, may create conditions that promote cancer cell proliferation, consistent with HKs overexpression effects in cancer cells.
In this study, we investigated how abnormal yeast Hexokinase 2 (HXK2) expression affects epigenetics and genomic instability in a simple, genetically tractable model system for eukaryotic cell function: the budding yeast Saccharomyces cerevisiae.In this organism, three genes encode proteins that phosphorylate glucose to Glc-6-P: HXK1, HXK2, and the glucokinase gene GLK1.HXK2 appears to play the main role in glucose phosphorylation in vivo (11,12).Therefore, we investigated the effects of HXK2 deletion and overexpression on parental histone H3-H4 transfer, an essential step in epigenetic inheritance (13).We hypothesized that the DNA-replication coupled epigenetic inheritance process is altered because DNA replication is an energy consuming process.We found both types of altered HXK2 expression had only a minor effect on parental histone H3-H4 distribution on replicating chromatin and appeared to decrease chromatin instability.However, 2-deoxy-D-glucose (2-DG), an Hxk2 inhibitor that has been used as a cancer therapy (14), increased chromatin instability in yeast.The data indicates that genetically altered Hxk2 expression has only a minor impact on the epigenome and genomic stability.
However, the pharmacological use of the Hxk2 inhibitor 2-DG treatment does lead to an increase in chromatin instability.

Materials and methods
Yeast strains, plasmids, and growth conditions S. cerevisiae yeast strains were grown in YPD (2% peptone, 1% yeast extract, 2% glucose) or yeast nitrogen base minus uracil (AA-Ura) media, to prevent eviction of plasmids at 30°C (15).Yeast strains used in this study are listed in Supplemental Table 1.Plasmids were transformed into yeast strains using the standard lithium acetate transformation method (16).AA-Ura or YPD plates containing 2% glucose were supplemented with 20% 2-DG (20 g 2-DG dissolved in 100 ml sterile water) to a nal concentration of 0.2% w/v.

HXK cloning and yeast strain construction
To amplify yHXK1 and yHXK2 from yeast genomic DNA, we employed speci c primer pairs (Kpn1-HXK1/Xba1-HXK1 and Kpn1-HXK2/Xba1-HXK2; primer sequences provided in Supplemental Table 2).Following ampli cation, the resulting DNA fragments were gel-puri ed using a Qiagen gel puri cation kit and subsequently cleaved with Kpn1 and Xba1 restriction enzymes.The resulting fragments were inserted into the yeast expression vector PSF-TEF1-URA3 (OGS534, Milipore-Sigma), exploiting the corresponding restriction enzyme sites.
The yeast-codon optimized human HK genes hHK1 and hHK2 were synthesized directly, with the sequences described in the Supplemental Data.These gene fragments were enzymatically cut using the restriction enzymes EcoRV and XhoI and integrated into the PSF-TEF1-URA3 vector, utilizing the matching restriction enzyme sites.
To introduce the HXK-containing vectors and vector controls into yeast, we followed the standard lithium acetate transformation protocol.Colonies were selected on yeast synthetic complete supplement mixture minus uracil (SC-Ura) plates.

Cell cycle analysis
Yeast strains were grown in YPD medium at 30° C until the cells reached mid-log phase (~ 0.6 OD at 600 nm), at which time they were treated with 2-DG (0.2% w/v).A total of ~ 1X10 7 cells were harvested at various time points by centrifugation, washed with water, and xed with 95% ethanol overnight at 4° C. Fixed cells were collected by centrifugation, washed with 50 mM sodium citrate buffer (pH 7.5), and resuspended in 500 µL sodium citrate buffer (pH 7.5) with RNAase A (0.025 mg/mL).After cells were incubated for 1 h at 50° C, 25 µL proteinase K (20 mg/mL) was added to the cell suspension and the mixture was incubated at 50° C for another hour.Cells were subsequently stained overnight at 4° C using propidium iodide at a nal concentration of 0.02 mg/mL.Finally, samples were analyzed in a uorescence-activated cell sorting (FACS) ow cytometer (BD Fortessa cytometer).A minimum of 50,000 cells per sample were acquired.

Analysis of silencing-loss at the HML locus using the CRASH assay
The yeast strains WT (cyc1250), hxk2Δ (cyc1253), WT + PSF-yHXK2 (cyc1247), and hxk2Δ + PSF-yHXK2 (cyc1251) were used to measure the apparent silencing-loss rate at the HML locus.Brie y, 10 colonies of each strain were grown separately overnight in SC media, diluted to 0.01 OD at 600 nm in SC media, and grown for 5 hours at 30° C. Cells were treated with 20 µM nicotinamide for the green uorescent protein (GFP) + control; cells were grown in hygromycin (200 µg/ml) for the red uorescent protein (RFP) + control.The apparent silencing-loss rate at the HML locus was calculated by dividing the number of RFP + GFP + cells (cells that have recently undergone Cre-mediated recombination express GFP but not RFP) by the total number of cells with the potential to lose silencing (RFP + GFP-and RFP + GFP+).For each colony, 50,000 events were analyzed using a BD Fortessa cytometer.

Enrichment and sequencing of protein-associated nascent (eSPAN)
The yeast strains WT (cyc1023), hxk2Δ (cyc1025), WT + PSF-HXK2 (cyc1022), and hxk2Δ + PSF-HXK2 (cyc1024) were used in this experiment.Brie y, yeast cells were grown in SC-URA-medium to exponential growth phase (~ 0.5 OD at 600 nm).Two doses of α factor (5 µg/ml; EZBiolab) were added for 3 hours at 25° C to arrest cell growth at the G1 phase.Cells were washed twice with cold water and then released into fresh YPD medium containing 400 mg/L BrdU and 200 mM hydroxyurea for 45 min at 30° C. Hydroxyurea stalls the replication fork but does not interfere with assembly of newly synthesized and parental histones (19,20).
After washing the beads extensively, ChIP DNA was recovered using the Chelex-100 protocol (21).
ChIP DNA was denatured by incubating it at 100° C for 5 min and then on ice for 5 min.DNA was diluted with BrdU IP buffer [1X PBS, 0.0625% Triton X-100 (v/v)].BrdU antibody (0.17 µg/ml; 555627, BD Biosciences) was added, and samples were incubated at 4° C for 2 hours.Next, 20 µl prewashed Protein G beads (17-0618-02, GE Healthcare) were added to each sample and incubated for an additional hour at 4° C. The beads were extensively washed, then DNA was eluted with 100 ul 1X TE buffer containing 1% SDS and puri ed using a QIAGEN MinElute PCR Puri cation kit.Single-stranded DNA libraries were prepared using an Accel-NGS 1S Plus DNA library kit (10096, Swift Biosciences).

Sequence mapping and data analysis
Sequence mapping, nucleosome mapping, and eSPAN analysis were performed similarly to what has been previously described (19,20).Brie y, reads were mapped to the Saccharomyces Genome Database (http://www.yeastgenome.org/)reference genome with Bowtie2 software (22).Only paired-end reads with both ends mapped correctly were selected for continued analysis.We determined nucleosome occupancy using 120-170 bp DNA fragments calculated from paired-reads, using Python programs we developed ourselves.To calculate the eSPAN bias pattern, we separated forward (Watson strand) and reverse (Crick strand) reads following the reference genome.Nucleosome positions around DNA replication origins were determined previously (23).Total eSPAN sequence reads at ± 10 nucleosomes surrounding the DNA replication origins were counted.The log2 ratio of Watson over Crick strand reads at each nucleosome position was used to obtain the average eSPAN bias pattern.

Western blot to detect yeast Rad53-P
The sample collection and process with Rad53-P followed previous publication (40).Blotting was then performed with anti-Rad53 antibody (ab104232, Abcam).

Results
Human HK1 or HK2 cannot complement yeast HXK deletion mutant, as determined by yeast cell growth In yeast, two HXKs, HXK1 and HXK2, and one glucokinase (GLK1) phosphorylate glucose, the rst irreversible step in the intracellular metabolism of glucose (11).Among these three enzymes, HXK2 is the dominant player in cell growth, as evidenced by the slow growth phenotype exhibited by hxk2Δ and hxk1Δ, hxk2Δ mutant cells; the growth of hxk1Δ cells is comparable to that of wild-type (WT) cells (Fig. 1A, left).Importantly, as shown previously (24), the HXK inhibitor 2-DG speci cally targets Hxk2 but not Hxk1 (Fig. 1A, right).This conclusion was based on the observation that the hxk1Δ cells, grows weaker than in hxk2Δ cells on the 2-DG plate (Fig. 1A, right).
To investigate the effects of HXK expression, we overexpressed both yeast HXKs (yHXK1, yHXK2) and their human orthologs (hHK1, hHK2) in hxk1Δ hxk2Δ double mutant yeast cells.The HXK genes were placed under the control of the yeast constitutive TEF1 promoter (21) in a plasmid, and we grew each transformed strain under both conditions that selected for the plasmid (AA-URA medium) and did not select for the plasmid (YPD medium).We hypothesized that expressing both human and yeast versions of the HXK genes could complement the slow growth phenotype of hxk1Δ hxk2Δ double mutant.Surprisingly, overexpressing human HKs in the yeast mutants cultured in YPD or AA-URA media had no discernible impact on yeast proliferation relative to vector-only control yeast (Fig. 1B).This outcome strongly suggests that human HK1 and HK2 do not function effectively in yeast cells.By contrast, expression of yeast HXK1 and HXK2 exhibited a complementary effect on yeast proliferation, evident in both YPD and AA-URA media (Fig. 1B).At here, we want to explain a little about the media difference.
Because cell could frequently lose the vectors (HXKs-containing or empty control) during proliferation process, the cells on the non-selective media (YPD) are mixture of cells with or without vectors.On the selective media (AA-URA), only vectors containing cells can grow.
Next, we investigated the effects of chemically inhibiting the overexpressed HXKs.The purpose is to further test the 2-DG's speci city on the Hxks.When the transformed yeast strains were cultured on YPD supplemented with 0.2% 2-DG (non-selective media with an Hks inhibitor), no signi cant differences in proliferation were observed relative to the vector control strain.This nding can likely be explained by loss of the transformed plasmid under the non-selective conditions.When the strains were cultured on AA-URA + medium supplemented with 0.2% 2-DG (selective media with an Hks inhibitor), the inhibitory effect of 2-DG was speci c to cells containing yHXK2 (Fig. 1B).This result underscores the pivotal role played by yeast HXK2 in cellular proliferation and highlights the speci city of 2-DG targeting yHxk2p.

Impact of HXK2 expression on global histone modi cation levels
To explore the relationship between HXK2 expression and the chromatin structure/epigenome, we employed four yeast strains characterized by varying HXK2 expression levels (ranked highest to lowest): WT + HXK2, hxk2Δ + HXK2, WT, and hxk2Δ (Fig. 2).First, we con rmed that expression levels for the strains were as expected (Fig. 2A).Overexpression of yeast HXK2 in the hxk2Δ yeast strain signi cantly upregulated proliferation (Fig. 2A on AA-URA plate), indicating complementation hxk2Δ, whereas overexpression of yeast HXK2 in the WT yeast strain appeared to have no bene cial impact on cell proliferation.The results are consistent with our former observation that the yeast HXK2 expression can complement the yeast hxk1Δ hxk2Δ (Fig. 1).
Next, we conducted immunoblot analysis to assess the effects of HXK2 expression on global histone modi cations, as well as the abundance of the major chromatin silencing protein, Sir2, in the four yeast strains (11).We examined modi cations such as H3K36me3, H3K27Ac (associated with active gene transcription), H3K56AC, and H4K5AC (linked to speci c histone functions) (25).As the HXK2 expression level may affect the epigenome, which may show different histone modi cation levels.However, no substantial differences were detected in the levels of these histone marks or in the abundance of Sir2 across the four yeast strains (Fig. 2B).This nding indicates HXK2 expression does not exert global effects on histone tail modi cations.It is important to note, however, that these results do not rule out the possibility of speci c changes in histone modi cations occurring at particular loci, warranting further investigation into the potential context-dependent impact of HXK2 on histone modi cations.

Impact of HXK2 expression on parental histone transfer
One crucial aspect of stable epigenetic inheritance is the successful replication of epigenetic information during DNA replication.Nucleosome assembly, including the appropriate transfer of both parental and newly synthesized histones, plays a pivotal role in this process.Given the energy-intensive nature of this step, we investigated whether HXK2 expression in uences parental histone transfer and assembly.In WT yeast, parental H3-H4 tetramers are normally distributed nearly evenly between the leading and lagging strands of DNA replication forks, with a slight bias toward the lagging strand (19).To assess the impact of HXK2 expression on parental histone distribution, we conducted an eSPAN assay, employing H3K4me3 as a marker for parental H3-H4 histones, as in our previous studies (13,26).Brie y, yeast cells were synchronized at the G1 phase and released into early S phase in the presence of BrdU and hydroxyurea.
Cell harvesting occurred 45 minutes after release (Fig. 3A).Throughout the cell culture process, we employed selective culture medium (AA-URA) to prevent the loss of the HXK2-containing plasmid.FACS analysis of DNA content con rmed the effectiveness of our cell synchronization procedure (Fig. 3B), and Western blot analysis con rmed the well-controlled expression of HXK2 (Fig. 3C).In the hypothesis, we would expect to observe the changes of parental histone distribution pattern on the two replicating strands by eSPAN analysis.
H3K4me3 ChIP, BrdU-IP-ssSeq, and H3K4me3 eSPAN sequence data were mapped to both the Watson and Crick strands of the yeast reference genome.No substantial differences were observed among the four strains for H3K4me3 ChIP and BrdU-IP-ssSeq data at typical mapping sites, such as ARS1309 (Fig. 4A, 4B).H3K4me3 eSPAN peaks at ARS1309 (Fig. 4C) displayed a nearly symmetric bias pattern, with a slight preference for the lagging strand in all four yeast strains.These ndings suggest that parental histones H3-H4 were distributed in a nearly symmetrical manner between the leading and lagging strands at hydroxyurea-stalled replication forks, regardless of HXK2 expression levels.
To determine whether the results we obtained at ARS1309 were characteristic of replication origins around the genome, we calculated the average strand bias ratio of H3K4me3 eSPAN peaks surrounding all 134 early DNA replication origins in yeast.The average bias ratio re ects the relative abundance of histones on the leading vs lagging strand.Consistent with our ndings for H3K4me3 eSPAN data at ARS1309 (Fig. 4C), all four strains exhibited a bias toward the lagging strand at these early DNA replication origins, regardless of HXK2 expression levels (Fig. 4D-G).However, we did observe signi cant variation in the average strand bias ratio at different nucleosome locations around the replication origins, particularly in the hxk2Δ strain (Fig. 4F, G).This variability may be attributable to differences in cell growth rates or low BrdU incorporation rates, factors that can signi cantly in uence eSPAN bias calculations (26).Overall, we conclude HXK2 expression has only a minor effect on the parental histone transfer process.

Effect of HXK2 expression and inhibition on chromatin stability
Even the HXK2 expression level does not has any clear impact on the parental histone transfer process by, it could affect the chromatin stability by other ways.To test this possibility genetically, we performed a CRASH assay to monitor the transient loss of heterochromatin silencing at the HML locus.Flow cytometry analysis was utilized to quantify the rate of silencing loss in each of the four yeast strains (Fig. 2-4).We hypothesized that overexpression or deletion of HXK2 increases chromatin stability because the glucose metabolism alteration.HXK2 overexpression (WT + HXK2) and deletion (hxk2Δ or hxk2Δ + HXK2) both resulted in a lower silencing loss rate than observed in the WT strain (Fig. 5A, B).
When WT strains were treated with 2-DG to inhibit HXK2, the silencing loss rate increased signi cantly by student t-test (Fig. 5B).In summary, these results indicate that abnormal HXK2 levels reduce yeast chromatin instability.However, 2-DG treatment can increase chromatin instability.They also shed light on the potential of 2-DG to modulate chromatin dynamics in yeast cells, underscoring the multifaceted effects of this HXK2 inhibitor on cellular processes.

Impact of HXK2 inhibition on cell cycle progression and the DNA damage response
To gain further insights into the effects of HXK2 inhibition by 2-DG on cell physiology, we assessed its in uence on cell cycle progression and its potential to induce systemic DNA damage.First, we investigated the effects of HXK2 inhibition on cell cycle progression.Logarithmically growing yeast cells (both WT and hxk2Δ) were subjected to 2-DG treatment, and we conducted FACS at various time points to measure changes in DNA content.We hypothesized that the cell cycle is predominately arrested at one of cell cycle stages because the glucose metabolism is inhibited.In the WT yeast strain, characterized by normal HXK2 expression, 2-DG treatment predominantly led to cell cycle arrest at the G2 phase (Fig. 6A).No signi cant effects were observed in hxk2Δ yeast cells following 2-DG treatment; the G1 enrichment observed in this strain at the 8-hour time point was attributed to prolonged culture and nutrient depletion.
Because the 2-DG treatment changed the cell cycle progress, which is frequently associated with DNA damage response, we next investigated whether inhibiting HXK2 induced systemic DNA damage.Rad53, a vital cell cycle checkpoint protein, is activated and phosphorylated (p-Rad53) in response to DNA replication stress.Thus, p-Rad53 indicates a systemic DNA damage response (27).For instance, exposure to methyl-methanesulfonate induces detectable levels of p-Rad53 (Fig. 6B).We analyzed p-Rad53 levels in strains treated with 2-DG via Western blot analysis.Our results indicate that neither 2-DGtreated WT nor 2-DG-treated hxk2Δ yeast strains displayed activated Rad53 (Fig. 6B).These ndings suggest that 2-DG-induced cell cycle arrest occurs primarily at the G2 phase but does not lead to widespread DNA damage.These observations provide valuable insights into the speci c impact of 2-DG on cell cycle dynamics and establish the absence of a prominent DNA damage response, further contributing to our understanding of the cellular effects of this glycolytic inhibitor.

Discussion
In this study, we explored the impact of altering expression of the glycolytic enzyme HXK2 on parental nucleosome transfer and epigenetic stability in yeast.The nucleosome assembly pathway is an energyintensive process (11).Epigenetic enzymes employ several metabolic intermediates as substrates or cofactors to carry out post-translational modi cations of DNA and histones (9), and numerous studies have underscored the crucial role of histone tail acetylation in nucleosome assembly (28,29).Disrupting acetylation modi cations can result in genome instability (28, 29).Thus, we hypothesized that altering HXK2 expression might impact glycolysis rates in the cell and thus the cell's global epigenetic signature and genome stability.However, our ndings indicate that changes in HXK expression, whether resulting from overexpression or deletion, have a minimal effect on parental nucleosome assembly and seem to decrease chromatin instability.Surprisingly, inhibiting HXK2 via treatment with 2-DG increased chromatin instability.
These ndings imply that in yeast cells with HXK2 deletions, the minimal ATP levels generated through glycolysis, driven by alternative HXKs, such as HXK1 or GLK1, are su cient to sustain epigenetic stability.Interestingly, recent research has shown that a high-glucose culture medium can enhance chromatin instability, with the NAD-dependent histone deacetylase Sir2 playing a role in this process (30).It appears that energy availability, rather than enzyme levels, is the primary rate-limiting factor for chromatin stability.
The proliferation of many cancer cells heavily relies on aberrant energy metabolism, such as the Warburg effect or aerobic glycolysis (31).Therefore, inhibiting Warburg effect has emerged as a potential strategy to target and eliminate tumor cells, with the HXK inhibitor 2-DG being one potential candidate drug of this type.As mentioned, our study suggests that 2-DG treatment increases epigenetic instability.However, this effect appears to occur independent of glycolysis inhibition, as lower HXK2 levels increase epigenetic stability.Thus, the precise mechanism underlying 2-DG's impact on chromatin stability warrants further investigation.
In our experiments, overexpressing the human-derived hHK1 and hHK2 genes did not rescue the growth defect phenotype of a yeast hxk1Δ hxk2Δ mutant, despite yeast HXK1 and HXK2 genes in the same vector functioning effectively.This observation suggests the slow growth phenotype of the hxk2Δ strain may not solely result from its reduced glucokinase activity.Alternatively, hHK1 and hHK2 may require additional factors to function that are absent in yeast cells.Interestingly, the human GlkB (hHK4) protein, which has a similar molecular weight as the yeast HXK2 protein, can complement the growth phenotype of a yeast hxk1Δ/hxk2Δ/glk1Δ mutant (32).Likewise, several plant HXKs have been shown to complement growth phenotypes in yeast HXK mutants (33)(34)(35).

Conclusions
Hexokinases (HXKs) are the key enzymes regulating glycolysis in cells.Overexpression of HXKs has been linked to numerous types of cancers and targeting HXKs has been suggested as a potential strategy of cancer therapy.In this study, our result suggests the altered HXKs expression level decreases the chromatin instability.However, 2-Deoxy-D-glucose (2-DG), a HXK inhibitor, treatment cells show increased chromatin instability with unknown mechanisms.

Declarations
Ethics approval and consent to participate.applicable.Consent for publication.Not applicable.

Figures
Figure 1 In yeast, HXK2 is the dominant hexokinase; overexpressing yeast HXK1 or HXK2, but not human HXK1 or HXK2, complements the slow cell growth phenotype yeast cells lacking HXK2.(A) Ten-fold serial dilutions of yeast cells of the indicated genotypes were plated onto YPD medium with or without 0.2% 2-deoxy-D-glucose (2-DG, an HXK2 inhibitor).This result suggests that HXK2 is the dominant player in glucose metabolism.(B) Yeast HXK1 or HXK2 can complement the cell growth phenotype in hxk1Δ hxk2Δcells, but human HXK1 and HXK2 cannot.Ten-fold serial dilutions of yeast cells of the indicated genotypes were plated onto YPD medium with or without 0.2% 2-DG (nonselective for the plasmid containing the HXK genes) and AA-URA medium with or without 0.2% 2-DG (selective for the plasmid).These results suggest that the 2-DG is speci cally targeting HXK2 and that overexpression of HXK1 can partially complement the growth phenotype of HXK2 deletion, but 2-DG has no effect on the HXK1.
Impact of HXK2 expression on yeast cell growth and total histone modi cation levels.In this panel, we con rm the HXKs expression level by western blot and HXKs level on total histone modi cation levels.(A) Cell growth rate of WT, WT+HXK2, hxk2Δ+HXK2, and hxk2Δ yeast strains.Ten-fold serial dilutions of yeast cells were plated onto YPD medium and AA-URA (conditions selective for the vector containing cells) with or without 0.2% 2-deoxy-D-glucose (2-DG, an HKsinhibitor) medium.The purpose of these dot assay is phenotype con rming of the genotypes.(B)Total levels of histone modi cations (H3K36me3, H3K27Ac, H3K4me3, H3K56Ac and H4K5,8,12Ac) has subtle change in the yeast strains from (A) with Western blot analysis.The exposure growth cells on AA-URA medium were collected for Western blot analysis.Western blot analysis monitoring HXKand H3K4me3 levels in the yeast strains from (B).The log phase hxk1Δ hxk2Δ strain sample serves as a negative control for the Hxk antibody.This data suggests the HXK2 expression level is well controlled.

Figure 5 Effect
Figure 5