Microsporidian EnP1 alters host cell H2B monoubiquitination and prevents ferroptosis facilitating microsporidia survival

Significance

Microsporidia are obligate intracellular pathogens. They have emerged as important opportunistic pathogens in immune-compromised hosts as well as significant pathogens in agriculture and aquaculture. Microsporidia have been shown to alter their host’s physiology, e.g., infection can result in the formation of xenomas and juvenilizing their hosts; however, the mechanism(s) by which they alter host cells is unknown. In this study, we found that microsporidia can secrete EnP1 into the host nucleus which interacts with host histone H2B. This interaction disrupts H2Bub, inhibiting p53 transcription. Consequently, there’s an upregulation of SLC7A11 expression, suppressing host ferroptosis, thereby facilitating microsporidian proliferation. This elucidates a pivotal survival mechanism within host cells, shedding light on the intricate interplay between microsporidia and their hosts.

Abstract

Microsporidia are intracellular eukaryotic pathogens that pose a substantial threat to immunocompromised hosts. The way these pathogens manipulate host cells during infection remains poorly understood. Using a proximity biotinylation strategy we established that microsporidian EnP1 is a nucleus-targeted effector that modifies the host cell environment. EnP1’s translocation to the host nucleus is meditated by nuclear localization signals (NLSs). In the nucleus, EnP1 interacts with host histone H2B. This interaction disrupts H2B monoubiquitination (H2Bub), subsequently impacting p53 expression. Crucially, this inhibition of p53 weakens its control over the downstream target gene SLC7A11, enhancing the host cell’s resilience against ferroptosis during microsporidian infection. This favorable condition promotes the proliferation of microsporidia within the host cell. These findings shed light on the molecular mechanisms by which microsporidia modify their host cells to facilitate their survival.

Epigenetic regulation of host cell responses induced by intracellular infection is a common strategy employed by diverse pathogens, including protozoan parasites, and prokaryotic and viral pathogens, which facilitate pathogen survival and disease progression (1). The dynamic interplay between pathogens and host cells is one of the most adaptable and ever-evolving systems in the natural world (2). Intracellular pathogens face daunting challenges of surviving within a potentially hostile environment. To overcome the selective pressures imposed by their hosts, many pathogens have evolved a wide array of strategies. These manipulative strategies can lead to alterations in host physiology, morphology, and behavior, ultimately maximizing the survival and transmission of the pathogens (3, 4). Pathogens have been found to employ various molecules to modify their hosts including those involved in DNA methylation, chromatin remodeling, histone modification, and noncoding RNAs (ncRNAs) with currently unknown functions. These molecules can effectively modulate the epigenetic status of the host cells, thereby playing pivotal roles in regulating host cell physiology and the host’s immune response (5–8). Despite these intriguing insights, our understanding of the extent and significance of epigenetic variation in host-pathogen interactions, especially during microsporidian infections, remains limited. Furthermore, evidence of direct interactions between proteins produced by microsporidia and the host cell’s epigenetic machinery has yet to be described.

Microsporidia are considered a sister clade of Fungi and are related to the Cryptomycota. They are opportunistic intracellular pathogens and microsporidiosis has been reported in virtually all invertebrates and vertebrates (9). About 17 species of microsporidia from ten genera have been identified to cause infections in humans, with Encephalitozoon spp. and Enterocytozoon bieneusi causing the majority of infections (10). The clinical manifestations of microsporidiosis are dependent on the infecting species and the immune status of the host but can range from self-limiting inflammation in immune-competent persons to fatal infections with a reported overall prevalence rate of 10.1% to 13.4% in immune-compromised individuals such as those with AIDS (11).

Microsporidia primarily invade host cells using a unique organelle termed the polar tube. Upon appropriate environmental conditions, microsporidian spores germinate and eject their polar tube which interacts with the host cell membrane allowing the sporoplasm, which travels through the hollow polar tube, to establish infection and subsequently proliferate within the host cell (12). During its prolonged intracellular parasitic life cycle, microsporidia do not immediately disrupt the functionality and integrity of the host cells, but rather, through an extensive process of evolution, have developed a sophisticated set of strategies to regulate the stable internal environment of the host cells (13). Because of the intimate relationship between microsporidia and their hosts, they are heavily dependent on host resources and have undergone extensive genomic reduction (14, 15). However, genomic reduction was followed by a large expansion in microsporidia-specific and genus-specific gene families including some host-exposed proteins such as polar tube and spore wall proteins (15). Various microsporidian host-exposed proteins may contain targeting signals that would direct them to specific host cell compartments allowing them to manipulate host function (16, 17).

EnP1, initially identified as endospore protein 1 (EnP1), is localized to both the endospore wall and the exospore wall, with enrichment at the anchoring disc site, and plays a crucial role in the interactions between microsporidian and host cells (18, 19). EnP1 harbors a signal peptide (SP), suggesting it is also a secreted protein. EnP1 is abundantly expressed during the intracellular proliferation phase of microsporidia (18). In other eukaryotic pathogens previously identified “structural” proteins have been found to have other functions, such as functioning as host cell effector molecules [e.g. MAG1 in Toxoplasma gondii (20)].

In this study, EnP1 was identified to function as a nucleus-targeted factor that is secreted by Encephalitozoon hellem into its host cells during infection. It interacts with nuclear histone H2B, and attenuates H2Bub; thereby modulating the host cell environment to better support microsporidian proliferation. This epigenetic modification is achieved through upregulation of the ferroptosis regulator SLC7A11, mediated by p53, which results in inhibition of host cell ferroptosis. This study represents the exploration of the mechanisms underlying microsporidia modulation of host epigenetic regulation and provides insights into the relationship between microsporidian proliferation and ferroptosis. These findings offer insights into potential candidate targets for preventing and controlling microsporidiosis.

Results

Microsporidian EnP1 Is a Nucleus-Targeted Effector.

Proximity labeling techniques have been widely employed to investigate protein interactions due to their efficiency and sensitivity (21). A biotin-based proximity labeling approach was used to screen for microsporidia effectors that are targeted to the host cell nucleus (Fig. 1) (22, 23). A novel fusion protein named 3HA-TurboID-NLS (biotin ligase TurboID with a nuclear localization signal) was constructed in pcDNA3.1 and transfected into HEK293T cells (Fig. 1A). Stable recombinant protein expression was confirmed by Immunoblot (IB) (SI Appendix, Fig. S1A). An immunofluorescence assay (IFA) using αHA confirmed the localization of 3HA-TurboID-NLS fusion protein to the host cell nucleus (Fig. 1A). Following the addition of biotin labeling of the host cell nucleus (Fig. 1B) and biotinylating of host cell nucleus proteins (SI Appendix, Fig. S1B) were demonstrated in cells expressing the 3HA-TurboID-NLS fusion protein. The vector and model system were therefore utilized to screen host cells for host cell nucleus-targeted microsporidian proteins.

Fig. 1.

EnP1 is identified as a potential nucleus-targeted effector of microsporidia by proximity labeling approach. (A) Schematic representing plasmid transfection and nuclear accumulation of 3HA-TurboID-NLS proteins and the IFA detection of the heterologous expression of 3HA-TurboID-NLS in HEK293T cells using αHA monoclonal antibody (mAb). (B) Schematic representing nuclear accumulation of biotinylated proteins after supplementing with biotin, and the IFA detection of the biotinylated proteins in HEK293T cells using Alexa Fluor 488-conjugated streptavidin. (A) and (B), Negative control: Cells transfected with pcDNA3.1 vector. (Scale bars, 20 μm.) (C) Schematic representation of the biotin-based proximity labeling approach that identified nucleus-targeted effectors of microsporidia. (D) Immunoblot detection of biotinylated proteins in host cell nucleus purified by streptavidin beads. M, molecular mass marker lane. (E) Mass spectrometry revealed that both EnP1 and EnP2 were identified. Both proteins exhibit characteristics of potential effectors targeted to the nucleus, possessing both the SP and the nuclear localization signal (NLS). The diagram illustrated the probability for Signal and NLS predictors. EnP1 harbors two NLS sequences, with probability of 0.56 and 0.82, respectively. (F) and (G) Immunoblot analysis of EnP1 and EnP2 heterologous expression in HEK293T cells. (H) IFA of the EnP1 and EnP2 heterologous expression in HEK293T cells. (Scale bars, 20 μm.)

To search for effectors, the host cells expressing 3HA-TurboID-NLS were infected with microsporidia, and the biotinylated proteins in host cell nucleus were purified using streptavidin-conjugated beads, and identified using mass spectrometry (MS) (Fig. 1 C and D). The MS results revealed a high abundance of EnP1 and EnP2, which are predicted to have SP and NLS, suggesting that they could be microsporidian nucleus-targeted effectors (Fig. 1E and SI Appendix, Table S3). Proteins in Fig. 1E including EnP1 and EnP2 were therefore expressed in HEK293T cells (Fig. 1 F and G) and their subcellular localization was analyzed (Fig. 1H and SI Appendix, Fig. S1C). In contrast to EnP2, which was found in both the host cytosol and nucleus, the expression of EnP1 was solely found in the host nucleus and it was selected for characterization as a microsporidian nucleus-targeted effector.

EnP1 Trafficking into the Host Nucleus Is Mediated by NLS and Promotes Microsporidian Proliferation during Infection.

According to the sequence analysis, EnP1 contains a SP and two NLSs (NLS^156–221 and NLS^312–327) (Fig. 2A). Based on the multiple-sequence alignment (MSA) of EnP1 homologs, it was observed these signal sequences were partially conserved in Encephalitozoon spp. (SI Appendix, Fig. S3E). To further confirm that EnP1 could be secreted into the host cell nucleus, the secretory function of the EnP1 SP was evaluated using the yeast signal sequence trap system (YSST) (24, 25). Both the pSUC2-EnP1 and positive control pSUC2-Ps87 enabled the invertase-deficient yeast strain YKT12 to grow on the YPRAA medium (Fig. 2B). Both yeast strains catalyzed the conversion of 2,3,5-triphenyl tetrazolium chloride to insoluble red-colored 1,3,5-triphenylformazan (SI Appendix, Fig. S2A). These results confirm the presence of a functional secretory SP in EnP1.

Fig. 2.

EnP1 is secreted into the host nucleus during infection and promoted microsporidian proliferation. (A) EnP1 amino acid sequence analysis. SP sequences are highlighted in red and NLS sequences were highlighted in yellow (NLS^156–221) and green (NLS^312–327). (B) YSST assay of EnP1 SP. CMD-W plates were used to select yeast strain YTK12 carrying the pSUC2 vector. YPRAA plates were used to indicate invertase secretion. (C) Immunoblot of secreted EnP1 in cytosol protein extractions (CPE) and nuclear protein extractions (NPE) of infected cells using EnP1 mPcAb. (D) IFA demonstrating localization of EnP1 in infected cells using EnP1 mPcAb. N, host cell nucleus; PV, parasitophorous vacuole. (Scale bars, 5 μm.) (E) Immunoblot of EnP1ΔNLS^156–221 (lane 1), EnP1ΔNLS^312–327 (lane 2), and EnP1ΔNLS (lane 3) in CPE and NPE respectively. (F) IFA of the localization of the exogenously expressed EnP1ΔNLS^156–221, EnP1ΔNLS^312–327, and EnP1ΔNLS in HEK293T cells stained with αFlag antibody. (Scale bars, 20 μm.) (G) Parasite load of microsporidia in HEK293T cells expressing EnP1 and EnP1ΔNLS at 48 hpi. Pathogen copy number was determined by quantitative PCR. (H and I) The effect of EnP1 and EnP1ΔNLS expression in HFF cells on Eh infection was assessed by counting PVs number on each slide (H) and the total area of PVs within the infected cells (I) at 48 hpi. *P < 0.05, ***P < 0.001, ns = not significant. Tubulin and H2B were used as the control of CPE and NPE respectively.

An αEnP1 mouse polyclonal antibody (mPcAb) was produced by immunizing mice with purified recombinant EnP1 protein (rEnP1). EnP1 mPcAb recognized a reactive band in E. hellem (Eh) spore lysates with a molecular weight (MW) of ~35 kDa, which is similar to the predicted EnP1 MW (37 kDa) (SI Appendix, Fig. S2B). With the EnP1 mPcAb, we could demonstrate by IFA and IB the localization of EnP1 in the nucleus of host cells infected by Eh (Fig. 2 C and D, SI Appendix, Fig. S2C). As the infection time progressed, the signal of EnP1 in the nucleus exhibited a rough upward trend (SI Appendix, Fig. S2D). To confirm the MS data, EnP1 was also detected by IB using the EnP1 mPcAb from the samples purified by streptavidin-conjugated beads from infected host cells that expressed 3HA-TurboID-NLS (SI Appendix, Fig. S2E). Overall, these data support that EnP1 is a secreted protein that is transported into the host cell nucleus in microsporidia infected cells.

To validate the NLS function of EnP1, EnP1ΔNLS^156–221, EnP1ΔNLS^312–327, or EnP1ΔNLS (lacked both NLS sequences) were expressed in HEK293T cells (SI Appendix, Fig. S2F) and the subcellular localization of EnP1 mutants was evaluated by IB of nuclear-cytosol fractionation and IFA. As shown in Fig. 2 E and F and SI Appendix, S2G, EnP1ΔNLS^156–221, or EnP1ΔNLS^312–327 could still localize to the host cell nucleus, however, EnP1ΔNLS no longer localized to the host cell nucleus and was instead found in host cell cytosol. These data indicate that EnP1 targets the host nucleus via an NLS-mediated pathway.

EnP1 stable expression cells did not affect the kinetics of microsporidian infection at 6 h postinfection (hpi) (SI Appendix, Fig. S2H), but did increase the number of organisms seen at 48 hpi (SI Appendix, Fig. S2I). Furthermore, we quantified the infection rate and the proliferation of microsporidia at 48 hpi by calculating parasitophorous vacuole (PV) numbers and area size as reported previously (26). The microsporidian infection rate was not significantly affected by EnP1 expression (SI Appendix, Fig. S2J), but a significant increase in microsporidian proliferation within host cells was observed (SI Appendix, Fig. S2K). These data suggest that EnP1 enhances the proliferation but not primary infection of host cells by microsporidia. When expressed at levels comparable to EnP1 (SI Appendix, Fig. S2L), EnP1ΔNLS lost the ability to promote microsporidian proliferation (Fig. 2G), and the phenomenon was further visualized by comparing the number and area of PVs with the matched group (Fig. 2 H and I). These findings suggest that microsporidia secreting EnP1 into the host cell nucleus is dependent on NLS and functions to enhance their proliferation within the host cell.

EnP1 Interacts with Host Histone H2B.

We aimed to unveil host regulatory mechanisms of EnP1 by elucidating its interaction with host nucleus proteins. For this purpose, we conducted proteomic analysis on a coimmunoprecipitation (Co-IP) of rEnP1-HA to identify EnP1 interacting proteins within the host nucleus. Silver staining of the rEnP1 Co-IP complexes revealed extra bands in the test lane compared to control (Fig. 3A). LC-MS/MS data identified several host cell proteins that might be the potential targets of EnP1 (SI Appendix, Fig. S3A). Among those proteins, the most likely candidate was H2B (UniProt: O60814) due to its location in the host cell nucleus and that it was enriched in LC-MS/MS data.

Fig. 3.

EnP1 exhibits specific binding affinity toward H2B, particularly with H2B1K. (A) Silver staining of immunoprecipitation (IP) samples precipitated by αHA magnetic beads from EnP1-HA expressing host cells. Negative control: Cells transfected with an empty vector. Arrowheads indicate the bands of interest. (B) IFA colocalization of EnP1 and H2B in the nucleus of HEK293T cells which were cotransfected with pcDNA3.1::EnP1-Flag and pcDNA3.1-3HA::H2B. αFlag (green) and αHA (red) mAb were used according to the procedures outlined in the Materials and Methods section. The merged image demonstrates the colocalization of EnP1 and H2B in the cell nucleus. (Scale bars, 20 μm.) The boxed region represents magnified images. (Scale bars, 5 μm) (C) Immunoblot of EnP1 in samples precipitated by H2B mAb from cells expressing EnP1. (D) His-tag pull-down demonstrating detection of H2B from the NPE of cells expressing H2B-HA using EnP1-His conjugated Ni-NTA agarose beads. αHA mAb was used to detect H2B-HA. Negative control: BSA-conjugated Ni-NTA agarose beads. (E) Far-WB of the interaction between EnP1 and H2B. Purified rH2B-Flag-His was subjected to SDS-PAGE, transferred to a PVDF membrane (lane 1), and then purified rEnP1-HA-His was incubated with the PVDF membrane, and the EnP1 that bound to H2B was detected by an αHA antibody (lane 2). Negative control: BSA (lane 3). (F) Immunoblot of EnP1 or EnP1S157A in IP samples precipitated by αHA magnetic beads from host cells expressing Flag-EnP1 and HA-H2B or Flag-EnP1S157A and HA-H2B. Immunoblot with αHA (H2B) and αFlag antibodies (EnP1 or EnP1S157A). (G) Quantitative analysis of the interaction between EnP1 and H2B family using Far-WB and recombinant proteins (expressed in E. coli). The purified rH2B variants (rH2B1A, rH2B1B, rH2B1D, rH2B1O, and rH2B1K) fused with Flag tag were loaded on the SDS-page gel and transferred to a PVDF membrane. The purified rEnP1-HA-His was then incubated with the PVDF membrane and the EnP1 that bound to H2B was detected by an αHA antibody. Negative control: BSA. H2B variants were detected by αFlag antibody.

In Fig. 3B, exogenously expressed rEnP1 and rH2B were observed to colocalize within the nucleus of HEK293T cells using IFA, indicating a potential interaction of them. To further confirm this interaction, Co-IP assay was conducted to assess the binding of exogenous EnP1 and host H2B. The result demonstrated that H2B, as well as EnP1, could be coprecipitated simultaneously (Fig. 3C). In addition, a pull-down assay confirmed rEnP1 could pull down rH2B (Fig. 3D). Moreover, a Far-Western Blot (Far-WB) employing H2B as a probe confirmed a robust interaction between H2B and EnP1 (Fig. 3E). These findings substantiate the direct interaction of H2B with EnP1 within the host cell nucleus.

AlphaFold (27) was utilized to predict the structure and interaction sites of EnP1 and H2B (SI Appendix, Fig. S3B). This simulation indicated that EnP1 interacts with H2B primarily through its N-terminal end, where EnP1S157 and EnP1P156 interact with H2BR73 via hydrogen bonding interactions. Interestingly, the interaction capacity of EnP1S157A mutant with H2B was notably reduced compared to wild-type EnP1 (Fig. 3F). Furthermore, a more pronounced decrease in interaction was observed in the EnP1P156AS157A double mutant (SI Appendix, Fig. S3D). These results strongly suggest that residues EnP1S157 and EnP1P156 are critical for its interaction with H2B. Remarkably, these residues are highly conserved in Encephalitozoon spp., as indicated by the MSA of EnP1 and its homologs (SI Appendix, Fig. S3E). In addition, EnP1M1, L4, and I9 are predicted to interact with H2BY41, V45, and V49 via hydrophobic interactions (SI Appendix, Fig. S3B).

Mammalian H2B exhibits numerous specific variants with only slight differences in their amino acid sequences, including H2B1A, H2B1B, H2B1D, H2B1O, and H2B1K, all of which are frequently detected in MS data on EnP1. These variants were heterogenous expressed in Escherichia coli (SI Appendix, Fig. S3C) and their ability to interact with EnP1 was quantified using Far-WB. Among these variants, the interaction of H2B1K with EnP1 was more robust than the other H2B variants (Fig. 3G), which was in accordance with the results obtained from the MS that showed H2B1K was the most highly enriched variant in EnP1 pulldown assays. Furthermore, the binding efficiency of H2B variants with EnP1 positively correlated with their sequence similarity to H2B1K (Fig. 3G and SI Appendix, Fig. S3C).

EnP1 Promotes Microsporidian Proliferation by Restraining H2Bub through the Inhibition of RNF20 Operation.

H2B is one of the core histones of the nucleosome, which is the fundamental unit of chromatin (28). H2Bub orchestrated by upstream ubiquitin ligases RNF20/40 and downstream deubiquitinating enzymes USP22 can activate or repress gene transcription (29, 30). H2BK120 serves as the specific monoubiquitination site for RNF20/RNF40, and the K120R mutant of H2B (H2BK120R) is incapable of undergoing ubiquitination (31). Pathogens can modulate the host immune defense responses by manipulating host H2Bub levels during infection, thus favoring their colonization (5). Given the demonstrated interaction between EnP1 and H2B, along with its role in promoting microsporidian proliferation, we postulated that microsporidia might epigenetically modulate host cells by manipulating H2Bub through the nuclear entry of EnP1. To validate this hypothesis, we initially assessed host cells H2Bub with and without EnP1 expression or Eh infection. We utilized an internal reference antibody against H3 and a protein-specific modification site antibody targeting H2Bub. These investigations revealed a significant decrease of H2Bub in cells expressing EnP1 compared to the negative control (Fig. 4 A and B). Following this, we observed a significant decrease in the dimethylation and trimethylation levels of H3K4 (H3K4me2 and H3K4me3) upon EnP1 overexpression (SI Appendix, Fig. S4 A and B), aligning with the theoretical premise that H2Bub serves as a prerequisite for H3K4 methylation (32). Similarly, Eh infection also reduced H2Bub (Fig. 4 C and D), H3K4me2 (SI Appendix, Fig. S4C), and H3K4me3 (SI Appendix, Fig. S4D), compared to uninfected cells. Meanwhile, the H2B level did not change with EnP1 overexpression or Eh infection (SI Appendix, Fig. S4 E and F). These findings indicated that the decrease in H2Bub resulting from microsporidian infection is attributed to EnP1 secretion, and is not a consequence of diminished H2B levels. The decrease of H2Bub could alter gene expression and may be etiologic in the increased growth of microsporidia from the nuclear localization of EnP1.

Fig. 4.

EnP1 promotes microsporidian proliferation by suppressing H2B monoubiquitination. (A) H2Bub detection in host cell nucleus following EnP1 expression. Negative control: Cells expressing GFP. (B) Relative band intensities quantification using ImageJ based on three independent experiments, including the one shown in (A). (C) H2Bub in host cell nucleus after Eh infection. Negative control: Uninfected cells. (D) Relative band intensities quantification using ImageJ based on three independent experiments, including the one shown in (C). (E) Immunoblot of H2Bub in siRNF20 treated host cells. Negative control: Untreated cells and control siRNA. (F) Immunoblot of H2Bub in RNF20 (RNF20OE) expressed cells. (G) Immunoblot of H2Bub in the nucleus of host cells expressing H2BK120R. Negative control: Cells expressing wild-type H2B (H2BWT). (H–J) Effect of RNF20 knockdown (H), RNF20 overexpression (RNF20OE) (I), or H2BK120R expression (J) in cells on microsporidian proliferation within host cells. Pathogen copy number was determined by quantitative PCR. (K) Immunoblot of H2Bub in host cells expressing EnP1 and GFP with or without RNF20 overexpression. (L) Immunoblot of RNF20 in IP samples precipitated by αFlag magnetic beads from host cells expressing Flag-EnP1 and HA-RNF20. (M) Far-WB of the binding of H2BK120R and EnP1. Negative control: H2BWT. (N) Immunoblot of RNF20 in IP sample precipitated by αH2B antibodies from host cells expressing both Flag-EnP1 and HA-RNF20. Negative control: IP sample precipitated from cells expressing RNF20-HA. (O) Relative band intensities quantification using ImageJ based on three independent experiments, including the one shown in (N). (P) Schematic representation of EnP1 regulating host H2Bub. Immunoblot with αHA (RNF20) and αFlag (EnP1) antibodies in (L) and (N). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

To further explore the impact of H2Bub on microsporidian proliferation, we used RNF20 siRNA to suppress RNF20 expression in host cells (SI Appendix, Fig. S4 G and H), this resulted in a reduction in H2Bub (Fig. 4E). Microsporidian infection and proliferation were significantly increased in the siRNF20 group as compared to control (Fig. 4H and SI Appendix, Fig. S4 K and L). Overexpression of RNF20 (SI Appendix, Fig. S4I) in host cells increased H2Bub (Fig. 4F) and this was associated with significant inhibition of both microsporidian infection and proliferation in these cells (Fig. 4I and SI Appendix, Fig. S4 M and N). Furthermore, in addition to the change in RNF20 expression, previous reports have indicated that H2BK120R overexpression can downregulate endogenous H2Bub (33). Therefore, we inhibited H2Bub by overexpressing H2BK120R (SI Appendix, Fig. S4J and Fig. 4G), and microsporidian proliferation was significantly increased (Fig. 4J and SI Appendix, Fig. S4 O and P). Overall, these data indicate that elevated levels of host H2Bub adversely affect the microsporidian proliferation and there is a close association between the H2Bub and microsporidian proliferation in host cells.

To validate the mechanism of EnP1 reducing H2Bub, we examined the effect of RNF20 on H2Bub in host cells expressing EnP1. As demonstrated in Fig. 4K, EnP1 reduced endogenous H2Bub in these cells without overexpressing RNF20. However, RNF20 overexpression in host cells could not restore the consequence of EnP1 on the H2Bub level (Fig. 4K), indicating the ability of RNF20 to modulate H2Bub was limited, probably due to the interaction of EnP1 with RNF20 or H2B. To test this hypothesis, the interaction of EnP1 and RNF20 was evaluated by Co-IP, and these data demonstrated no direct interaction between EnP1 and RNF20 (Fig. 4L and SI Appendix, Fig. S4Q). When RNF20-HA and EnP1-Flag were coexpressed in HEK293T cells and subjected to IP using αH2B antibody, the presence of EnP1 significantly inhibited the interaction between RNF20 and H2B (Fig. 4 N and O). Cells that only expressed RNF20-HA were used as a negative control. Simultaneously, our observations revealed that the presence of EnP1 influences the regular expression of RNF20 (SI Appendix, Fig. S4R). These findings imply that EnP1 has the capability to disrupt RNF20’s functionality by impeding its access to H2Bub and suppressing RNF20 expression.

An H2B prokaryotic expression mutant H2BK120R was constructed, and the binding capacity of wild-type H2B and H2BK120R with EnP1 was evaluated using Far-WB. The H2BK120R had a markedly reduced interaction with EnP1 than did wild-type H2B, suggesting that the 120 K residue is one of the active binding sites for EnP1 binding to H2B (Fig. 4M). These data suggest that the effect of EnP1 in reducing H2Bub is probably not due to the direct interaction of EnP1 with RNF20 but is a consequence of competition in their respective binding to H2B, which reduces ubiquitination efficiency (Fig. 4P).

EnP1 Promotes Microsporidian Proliferation by Suppressing Host Ferroptosis.

Subsequently, we investigated the molecular mechanism that drives microsporidian proliferation by EnP1 in host cells by identifying EnP1-modulated genes in the host. Using RNA-seq on host cells stably expressing EnP1, a total of 1,320 differentially expressed genes (Log₂ Fold change >1; P < 0.05; 1,140 upregulated and 180 downregulated) were identified (Fig. 5A). KEGG pathway analysis indicated there was a significant enrichment of genes involved in ferroptosis, suggesting EnP1 could dramatically affect ferroptosis in host cells (Fig. 5B). Further analysis showed that the expression of SLC7A11, a key regulator of ferroptosis, was significantly upregulated after EnP1 expression (Fig. 5C); e.g. the EnP1 group exhibited significantly higher Transcripts Per Million (TPM) values of SLC7A11 gene compared to control (SI Appendix, Fig. S5A). SLC7A11, vital for intracellular glutathione levels and redox balance via cystine/glutamate transport, notably impacts cellular ferroptosis and engages in nutrient metabolism, cancer progression, oxidative stress, and inflammation pathways (33, 34). As SLC7A11 is critically involved in negatively regulating host ferroptosis (7), the SLC7A11 transcription and protein expression were investigated in host cells expressing EnP1 and infected with Eh. As shown in Fig. 5 D and F, the SLC7A11 transcription level was significantly increased by EnP1 expression and microsporidian infection. Likewise, the SLC7A11 protein levels also significantly increased by these conditions (Fig. 5 E and G and SI Appendix, Fig. S5 B and C). Moreover, the temporal relationship observed between EnP1 secretion by microsporidia during infection and the expression of SLC7A11 (Fig. 5H) unveiled a direct association between the levels of EnP1 (SI Appendix, Fig. S5D) and its stimulatory influence on SLC7A11 (SI Appendix, Fig. S5E), which confirmed the positive regulation of SLC7A11 gene by EnP1. Knockdown of SLC7A11 (SI Appendix, Fig. S5F and Fig. 5I) inhibited microsporidian proliferation (Fig. 5J and SI Appendix, Fig. S5 G and H). These data suggest that SLC7A11 is a key host response regulatory gene targeted by microsporidia during infection (e.g. microsporidian EnP1 induces SLC7A11 expression altering host cell ferroptosis).

Fig. 5.

EnP1 suppresses host ferroptosis by enhancing the expression of SLC7A11. (A) The heat map of clustering between the EnP1 expression group and control group demonstrates that the samples are well clustered and significantly different from each other. Upregulated and downregulated genes are colored in red and blue, respectively. (B) Differential gene KEGG enrichment analysis focused on pathways related to the expression of EnP1. (C) Expression difference volcano plot, which demonstrates that SLC7A11 was significantly up-regulated in the EnP1 group. (D) qRT-PCR analysis of the change in SLC7A11 expression in host cells expressing EnP1. Negative control: Cells expressing GFP. (E) Immunoblot of SLC7A11 expression in host cells expressing either EnP1 or GFP. (F) qRT-PCR analysis of the SLC7A11 expression in host cells infected with Eh. Negative control: Noninfected cells. (G) Immunoblot analysis of SLC7A11 expression in host cells postinfection with Eh compared to noninfected cells. (H) Immunoblot of the relative protein levels of EnP1 and SLC7A11 at different time points (0, 24, 48, 72 hpi) during host cell infection with Eh. (I) Immunoblot of SLC7A11 in host cells transfected with siSLC7A11. Negative controls: Untreated cells and control siRNA. (J) Effect of SLC7A11 knockdown on the proliferation of microsporidia in host cells expressing EnP1 or GFP. Pathogen copy number was determined by quantitative PCR. (K) Host cells expressing EnP1 or GFP were treated with increasing concentrations of erastin and their cell viability was measured using CCK-8. (L) Host cells expressing EnP1 or GFP were treated with increasing concentrations of erastin and the LDH cytotoxicity was assayed. (M) Measurement of the ratio of GSH to GSSG in host cells expressing EnP1 and control host cells after induction with 20 μM and 25 μM erastin. (N) Quantification of the number of microsporidia within host cells after treatment with 10 μM Fer-1. Pathogen copy number was determined by quantitative PCR. (O) Quantification of the number of microsporidia within host cells after treatment with 20 μM erastin. The pathogen copy number was determined by quantitative PCR. (P) Schematic representation of EnP1 suppressing host ferroptosis by enhancing the expression of SLC7A11. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Elevating SLC7A11 expression can increase cystine uptake and decrease sensitivity to ferroptosis (35). Hence, we examined whether EnP1-induced SLC7A11 upregulation could alleviate HEK293T susceptibility to erastin-induced ferroptosis. Erastin induces oxidative stress, leading to mitochondrial dysfunction, membrane compromise, and decreased GSH levels, which were assessed using the CCK-8, LDH release assays, and the GSH/GSSG ratio (36–38). EnP1 expression in host cells significantly decreased cell death (Fig. 5K), the level of LDH (Fig. 5L), and elevated the GSH/GSSG ratio (Fig. 5M) induced by erastin, suggesting that EnP1 expression in host cells diminished their ferroptosis susceptibility. To further prove that EnP1-mediated protection against ferroptosis relies on SLC7A11, the SLC7A11 was knocked down in EnP1 stable expression cells, followed by evaluation of erastin-induced ferroptosis in host cells. As shown in SI Appendix, Fig. S5 I–K, the resistance of EnP1 stable cell lines to ferroptosis markedly declined, and a reduced microsporidian proliferation was also observed (SI Appendix, Fig. S5L). These findings firmly establish the SLC7A11-dependent nature of EnP1’s impact on ferroptosis.

To further evaluate the role of ferroptosis on microsporidian proliferation, host cell ferroptosis was either inhibited or activated by Fer-1 or erastin, which are respectively the inhibitor and activator of ferroptosis (39). Cells treated with Fer-1 or erastin were then infected by microsporidia. There was a significant increase in parasite proliferation when host cell ferroptosis was inhibited by Fer-1 (Fig. 5N and SI Appendix, Fig. S5 M and N). Conversely, treating the cell with erastin, which activated ferroptosis, decreased microsporidian proliferation in these treated cells (Fig. 5O and SI Appendix, Fig. S5 O and P). Taken together, these data are consistent with EnP1 inhibiting host cell ferroptosis by positively regulating SLC7A11 gene expression, resulting in increased microsporidian proliferation in their host cells (Fig. 5P).

EnP1 Suppresses p53-Mediated Ferroptosis by Inhibiting H2B Monoubiquitination.

A study in tumor cells demonstrated that tumor suppressor p53 can sensitize cells to ferroptosis by repressing the expression of SLC7A11 (Fig. 6A) (40). As SLC7A11 expression was found to be upregulated in cells expressing EnP1, we posited that EnP1 might enhance SLC7A11 expression by negatively modulating p53. To examine this hypothesis, we evaluated p53 expression using qRT-PCR and immunoblot in cells expressing EnP1 and those infected with Eh. Both the transcription and protein expression of p53 were significantly reduced (Fig. 6 B and C and SI Appendix, Fig. S6 A–D), indicating that p53 was likely a key effector in regulating SLC7A11 in cells expressing EnP1.

Fig. 6.

EnP1 negatively regulates p53 expression by inhibiting H2Bub. (A) Schematic representation of the p53 tumor suppressor enhancing ferroptosis by transcriptional suppression of SLC7A11 in tumor cells (40). (B) Immunoblot of p53 protein expression in cells expressing EnP1 or GFP. (C) Immunoblot of p53 protein expression in Eh cells and uninfected cells. (D) Detection of TP53 transcription level in host cells transfected with siRNF20 and siNC. (E) Immunoblot of p53 protein expression in host cells transfected with siRNF20 and siNC. (F) Detection of TP53 transcription level in cells overexpressing RNF20 (RNF20OE) and untreated cells. (G) Immunoblot of p53 protein expression in cells overexpressing RNF20 (RNF20OE) and untreated cells. (H) Detection of SLC7A11 transcription level in host cells transfected with sip53. (I) Immunoblot of SLC7A11 protein expression in host cells transfected with sip53. Negative controls of (H) and (I): Untreated cells and control siRNA. (J) Detection of SLC7A11 transcription level in cells overexpressing p53 (p53OE) and untreated cells. (K) Immunoblot of SLC7A11 protein expression in cells overexpressing p53 (p53OE) and untreated cells. (L) Cell viability of host cells expressing GFP, EnP1, and EnP1 with p53OE was measured using CCK-8 after induction with 20 μM and 25 μM erastin. (M) The LDH cytotoxicity of host cells expressing GFP, EnP1, and EnP1 with p53OE was assayed after induction with 20 μM and 25 μM erastin. (N) Measurement of the ratio of GSH to GSSG in host cells expressing GFP, EnP1, and EnP1 with p53OE after induction with 20 μM and 25 μM erastin. (O) Effect of p53 knockdown on microsporidian proliferation within these host cells. Pathogen copy number was determined by quantitative PCR. (P) Effect of p53 overexpression (p53OE) on microsporidian proliferation within these host cells. Pathogen copy number was determined by quantitative PCR. (Q) Schematic representation of EnP1 regulating host ferroptosis. **P < 0.01, ***P < 0.001, ****P < 0.0001.

It has been reported that inhibition of H2Bub downregulates p53 expression in tumor cells (41). Hence, EnP1 might downregulate p53 expression by inhibiting H2Bub. Our conjecture was confirmed through the reduction of H2Bub via RNF20 knockdown, subsequently leading to the downregulation of p53 expression (Fig. 6 D and E and SI Appendix, Fig. S6E). Oppositely, in cells overexpressing RNF20, p53 expression was significantly increased when H2Bub was increased (Fig. 6 F and G and SI Appendix, Fig. S6F). Similarly, when H2Bub was reduced by the expression of H2BK120R, the expression of p53 was also inhibited (SI Appendix, Fig. S6 G–I). The data suggest that EnP1 exerts a negative regulatory influence on p53 expression through the inhibition of H2Bub, a phenomenon that can be largely counteracted by the overexpression of RNF20 (SI Appendix, Fig. S6J).

Subsequently, an in-depth exploration was undertaken to investigate the p53 influence on the downstream regulation of SLC7A11 and microsporidian proliferation. When cells were transfected with sip53, which reduced p53 expression (SI Appendix, Fig. S6 K and L), both SLC7A11 transcription (Fig. 6H) and protein expression (Fig. 6I) were significantly elevated. Consequently, the cells exhibited heightened resistance to ferroptosis (SI Appendix, Fig. S6 N–P). Conversely, when p53 was overexpressed (SI Appendix, Fig. S6M), it inhibited SLC7A11 transcription (Fig. 6J) and protein expression (Fig. 6K). Consequently, these cells became more susceptible to ferroptosis (SI Appendix, Fig. S6 Q–S). Additionally, we noted that the enhancing influence of EnP1 on SLC7A11 could be counteracted by the overexpression of p53 (SI Appendix, Fig. S6T). P53 overexpression consequently reduced the host cells’ susceptibility to ferroptosis, as confirmed by examining various indicators associated with ferroptosis (Fig. 6 L–N). Of greater significance, the reduction in p53 expression led to heightened susceptibility of host cells to ferroptosis, thereby promoting the proliferation of microsporidia (Fig. 6O and SI Appendix, Fig. S6 U and V), whereas p53 overexpression had the opposite effect (Fig. 6P and SI Appendix, Fig. S6 W and X). Overall, these results indicate that EnP1 could inhibit p53 expression by suppressing H2Bub, leading to an increase in SLC7A11 expression, which results in an inhibition of host cell ferroptosis (Fig. 6Q).

Discussion

Limited research exists on how microsporidia manipulate their host organisms to establish a conducive internal environment for their survival, including the functions of their secreted proteins. This study presents compelling evidence that microsporidia (E. hellem) can regulate ferroptosis in host cells by secreting EnP1, which targets the host cell nucleus, disrupting H2Bub and triggering an increase in the ferroptosis regulator, SLC7A11. This inhibition of ferroptosis promotes microsporidian proliferation within host cells. Our investigation into EnP1, H2Bub, and ferroptosis converges on their collective impact on microsporidian proliferation and elucidation of a sophisticated regulatory network encompassing “EnP1-H2B-H2Bub-p53-SLC7A11-ferroptosis”.

Infections provoke varied host responses, including innate immune activation, inflammation, and cell death (42). Pathogens can induce or hinder regulated cell death forms like apoptosis, necrosis, or autophagy, aiding their proliferation or evading host immune surveillance (43). Ferroptosis, a recently characterized form of cell death reliant on iron and resulting from disruptions in the repair of lipid peroxidation, plays a pivotal role in a wide spectrum of physiological processes and pathological conditions (44, 45). Its link to diverse diseases underscores the significance of studying cellular ferroptosis modulation for understanding and managing related disorders (46). While extensively researched in cancer, inflammation, neurology, and development, the role of ferroptosis in infection pathogenesis remains poorly understood (43).

Microsporidia, with highly reduced genomes, depend on host cells for energy and essential resources. They modulate host responses by suppressing immunity (47), halting cell cycle development (48), and regulating apoptosis (49). Microsporidia inhibit caspase-3 cleavage, downregulate p53 transcription (50), enhance expression of apoptosis inhibitors like iap-2 and BIRC5 (51, 52), and downregulate Apaf-1 and cytochrome C expression to prevent apoptotic bodies formation, thereby inhibiting apoptosis and ROS production (53). Therefore, microsporidian infection generally leads to a decrease in the sensitivity of host cells to apoptosis, preventing apoptosis to facilitate their multiplication (54, 55). This paper introduces the following finding: microsporidia’s regulatory influence on ferroptosis, an alternative programmed cell death mode in hosts. Specifically, we observed that EnP1, a microsporidian-secreted protein, inhibits SLC7A11, crucial for cell protection against oxidative stress-induced ferroptosis. Erastin, inducing ferroptosis via SLC7A11 targeting, significantly reduces microsporidian proliferation. This finding hints at therapeutic potential for erastin and similar compounds against microsporidiosis.

Ferroptosis, an iron-dependent form of nonapoptotic cell death, is closely tied to host cell iron homeostasis. Pathogens like HBV alter intracellular free iron levels by manipulating transferrin receptor TfR1, inhibiting ferroptosis, and promoting conditions like liver fibrosis (56). Microsporidian infection can induce transferrin expression in host cells, resulting in decreased iron content within host cells and fostering an environment conducive to its own proliferation (57). In our previous work, we established that polar tube protein 4 (PTP4) interacts with host TfR1 during microsporidian infection (12). As TfR1 regulates cellular iron uptake and metabolism (58), the interaction between microsporidia components, such as PTP4, and TfR1, raises the possibility that these interactions could also interfere with intracellular free iron levels, thereby manipulating host ferroptosis to the advantage of the parasite. The details of this mechanism warrant further investigation.

Pathogens employ effector secretion to colonize hosts, interacting with surface receptors, cytoplasm, and even nuclei, modulating host responses (59, 60). In Toxoplasma gondii infection, effectors like ROP16, GRA16, GRA24, etc. are secreted into the host cytoplasm and many of these penetrate into the host nucleus. These effectors can function as epigenetic regulators and transcription factors, effectively reshaping host gene expression, which can counteract the host innate immune response and influence the host cell cycle (61, 62). Within microsporidia, Nosema bombycis can secrete NbRBL28, an effector possessing both SP and NLS sequence, targeting the host nucleus and altering the cell cycle (63). In addition, at least four secretory effectors targeting the nucleus of Caenorhabditis elegans have been identified in Nematocida parisii, influencing host transcription and nuclear organization (17). With EnP1-specific antibodies, we successfully identified the presence of endogenous EnP1 within the host nucleus during Eh infection. This finding provides conclusive evidence that EnP1 is secreted into host cells during microsporidian infection, possibly in the immature stages of microsporidian before spore wall formation, and enters the host cell nucleus aided by NLS. These findings suggest that functional diversification of structural proteins in intracellular parasites, particularly microsporidian with highly reduced genomes, is critical for their parasitism. Further investigation revealed that EnP1 interacts with histone H2B within the host nucleus. This interaction leads to the inhibition of RNF20’s ability to perform H2B monoubiquitination modifications.

It’s noteworthy that H2Bub crucially influences histone H3 methylation at lysine positions 4 and 79 (64, 65). While we’ve examined alterations in H3K4 dimethylation and trimethylation due to EnP1’s interference with H2Bub, its impact on other H3 methylation modifications remains unclear. This raises the intriguing question of whether EnP1’s actions have broader implications for other histone modifications and the regulation of downstream target genes and pathways. Further research is imperative to grasp EnP1’s full epigenetic influence and implications. This study is an important initial step for advancing our comprehension of the intricate mechanisms governing microsporidia–host interactions and broader pathogen–host dynamics.

Our findings provide clear evidence that microsporidian infection exerts a suppressive effect on p53 expression, which had been previously noted in studies on apoptosis, p53 transcription, and microsporidia infection of host cells (50). It is well established that activated p53 regulates genes crucial for DNA repair, cell cycle arrest, apoptosis, and senescence (66). Recent reports have described that p53 is involved in the regulation of ferroptosis by transcriptionally repressing the expression of SLC7A11 (40, 67). This implies that microsporidian-induced p53 inhibition impacts host regulation, including ferroptosis, apoptosis, and the cell cycle. Some studies have shown that SLC7A11 is a downstream target gene regulated by H2Bub modification, and a reduction in H2Bub reduces SLC7A11 expression (67). Within the complex regulatory network established by the pathogen within the host, factors like p53 and other potential regulators exert a more substantial influence on the regulation of SLC7A11 than the impact of H2Bub on its expression. Exploring these additional regulatory factors more deeply could provide valuable insights into their contributions to this intricate network of regulation. Furthermore, our transcriptomic data reveal upregulated SAT1 expression associated with ferroptosis. Studies by Ou et al. have demonstrated that SAT1 upregulation induces lipid peroxidation and ferroptosis under oxidative stress (68). Intriguingly, EnP1 expression inhibits host ferroptosis, suggesting SLC7A11’s elevated expression triggered by EnP1 dominantly regulates ferroptosis, affirming the complexity of host gene regulation by pathogens and their secreted proteins.

Fig. 7 presents a model illustrating how Encephalitozoon microsporidia may modulate the host cell to promote their replication. During the proliferative phase of microsporidia within host cells, the effector EnP1 is secreted and targets the host cell nucleus, where it interacts with host cell histone H2B. This interaction and the inhibitory effect of EnP1 on RNF20 disrupt H2Bub status, consequently influencing the expression of p53. As a result, the inhibitory effect of p53 on the downstream target gene SLC7A11 is attenuated, thereby bolstering the host cell’s ability to withstand ferroptosis during microsporidian infection. Ultimately, this favorable condition facilitates the replication of microsporidia in its host cell and prevents host cell death allowing the transmission of microsporidia to the next set of host cells.

Fig. 7.

The model of EnP1 modules host ferroptosis. Schematic diagram of the mechanism by which the microsporidia effector protein EnP1 enters the nucleus to regulate host cell functions and promote the proliferation of microsporidia.

Materials and Methods

The Scientific Investigation Board of the Medical School of Shandong University approved all animal studies, ensuring ethical compliance and oversight. Summaries of the key techniques and protocols are presented below. Comprehensive descriptions of the animals, cells and parasite strains, plasmids, materials, instruments, and biological assays involved are available in SI Appendix.

YSST.

The predicted EnP1 SP region was inserted into the pSUC2T7M13ORI vector. After transforming YTK12 cells, the transformants were screened on CMD-W medium. Positive control strain pSUC2-Ps87, negative control strain pSUC2-Mg87, and verified positive yeast strains were cultured in YPDA medium to an OD600 of 0.3. These cultures were then plated on CMD-W and YPRAA media and incubated for 2 to 4 d. Secretory properties were evaluated using TTC solution.

IFA.

The plasmid was transfected into HEK-293 T cells, which were then fixed and treated with Triton X-100 for permeabilization. The cells were blocked with TBST containing 5% BSA and subsequently incubated with the primary antibody followed by the secondary antibody. DAPI was added to stain the nucleus. Finally, the cells were sealed with an anti-fluorescence quencher and imaged using confocal microscopy.

Cell Viability and GSH/GSSG Ratio Assays.

Cell viability was evaluated with the CCK-8 assay (Topscience). LDH release, indicative of cell membrane damage, was assessed using the LDH Cytotoxicity Test Kit (Beyotime). GSH and GSSG were measured using the GSH/GSSG Assay Kit (Beyotime), and their ratios were analyzed statistically. All experimental procedures strictly adhered to the manufacturer’s instructions.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

The raw RNA sequencing data reported in this paper have been deposited in GEO Repository (GSE247835) (69). All other data are included in the manuscript and/or supporting information.