1. IntroductionPrions, proteins with the ability to adopt self-propagating conformations, have been found throughout the eukaryotic proteome [1]. These infectious agents play a role in the development of transmissible encephalopathies, such as kuru in humans, mad cow disease in cows, and scrapie in sheep [2]. Due to this association, prions have been assumed to play an obstructive role in cellular biology. However, in the baker’s yeast Saccharomyces cerevisiae, prions may enable advantageous responses to environmental stress [3,4,5,6].Over 20 unique prion proteins have been discovered in S. cerevisiae. Yeast prions can lead to positive cellular outcomes, such as improved cellular fitness and thicker cell walls [7,8]. Some yeast prion proteins include Sup35, Rnq1, Swi1, and Snt1, responsible for propagating the [PSI+], [PIN+], [SWI+], and [ESI+] prion states, respectively [9,10,11,12,13].Swi1 is a component of the SWI/SNF chromatin remodeling complex. Swi1 is able to bind DNA directly [14]. [SWI+], the prion state propagated by Swi1, abolishes flocculin gene expression [15]. Flocculins are proteins responsible for cell–cell and cell-surface adhesion, which allow for yeast to maintain facultative multicellularity [16]. Yeast multicellularity includes pseudohyphal growth of diploid cells [16]. The [SWI+] state results in a complete loss of pseudohyphal growth [15]. Yeast can control pseudohyphal growth depending on environmental conditions, and the presence of [SWI+] is hypothesized to be a major player in this switch [15].Rnq1 (Rich in Asparagine (N) and Glutamine (Q) 1) is prion protein responsible for the [PIN+] prion state [12]. Interestingly, Rnq1′s function in the non-prion state has yet to be determined. [PIN+] has been found to enhance the de novo formation of [PSI+], another yeast prion [12]. Null mutants show no phenotype [17]; however, its coexistence with other prions have alluded to Rnq1 playing a larger role in controlling overall gene expression [18].Some yeast prions, such as the [ESI+] state propagated by Snt1, have a deeper role in yeast biology [13]. Snt1 is a scaffold of an essential epigenetic modulator, the histone deacetylase (HDAC) Set3C [19]. [ESI+] leads to the formation of an activated chromatin state that can be inherited through generations [13]. The identification of Snt1 as a prion suggests that other yeast prions may also interact with the epigenetic landscape.Epigenetics refers to heritable changes in the phenotype of an organism that occur without changes in the underlying DNA sequence [20]. DNA is organized and compacted into chromatin. Chromatin structure can regulate gene expression. The basic unit of chromatin, termed a nucleosome, consists of DNA wrapped around a histone octamer core (containing two copies each of histone H2A, H2B, H3 and H4). One key mode of epigenetic regulation includes the post-translational modification (PTM) of histone proteins. The N-terminal tails of histones protrude out of the nucleosome and are modified with a multitude of chemical moieties (including methyl, acetyl, and phosphate groups) on various, but specific residues [20].Histone PTMs do not only directly impact the binding between DNA and histones, but also serve as binding platforms for other proteins. “Reader” proteins bind PTMs and lead to transcriptional regulation [21]. The location and type of modification determines whether transcription will be activated or silenced. For instance, acetylated lysine residues are “read” by bromodomains, which lead to gene activation [22]. Furthermore, there are histone PTMs “writers” which install these moieties while “erasers” remove them [21]. For example, lysine residues are acetylated by histone acetyltransferases (HATs) and deacetylated by histone deacetylases (HDACs) [23,24].Prion states are able to elicit phenotypic changes without changing an organism’s DNA sequence. As such, prions have been proposed as an alternative epigenetic mechanism [2]. Moreover, the presence of prions plays a broad role in gene expression [25,26]. These findings hint at a link between prions and other forms of epigenetic modification. Despite this, a comprehensive study of the interplay between yeast prions and histone modifications has not been carried out to date. Here, we establish an association between [PRION+] states and histone modifications, by showing that [SWI+] and [PIN+] each connect to distinct histone H3 PTM patterns. In particular, we find that [SWI+] is associated with decreases in the levels of H3K36me2 and H3K56ac compared to [swi−] yeast. On the other hand, decreases in H3K4me3, H3K36me2, H3K36me3 and H3K79me3 are linked to the [PIN+] state. Furthermore, curing of the [PRION+] state restored histone PTM levels to those found in the [prion−] state. These changes are distinct to those present in prion protein loss-of-function models. Our findings establish links between prion states and specific histone modifications. While the precise molecular mechanisms by which prions connect to the epigenome remain to be elucidated, establishing an interplay between [PRION+] states and the epigenetic landscape expands our knowledge of prion function. 3. DiscussionHere, we have characterized the epigenome connected to distinct [PRION+] states and revealed novel links between two yeast prions and the histone H3 PTM landscape (Figure 8). Surprisingly, we find very little overlap between the histone PTM alterations elicited in each [PRION+] state, suggesting these alterations are not linked to general protein aggregation pathways. Both [SWI+] and [PIN+] are associated with specific alterations in histone H3. These changes are reversible upon curing of the [PRION+] phenotype. While it is difficult to definitively establish causal relationships from our experiments, our prion curing results demonstrate a clear link between the prion state and histone PTM alterations. Additionally, total RNA levels were impacted in [PRION+] yeast. Disturbances in RNA levels were also reversed by curing of the [PRION+] state. It is important to note that we are unable to assign specific biological functions to these histone PTM changes as our experiments characterize genome-wide epigenetic changes and do not pinpoint the specific genes targeted by these alterations.We find that [SWI+] is associated with a decrease in H3K36me2 and H3K56ac, each associated with gene activation [29,52]. Swi1 is a subunit of the SWI/SNF chromatin remodeling complex [14]. Native Swi1 resides mainly in the nucleus [53]; however, in the [SWI+] state, Swi1-YFP mutants aggregate in the cytoplasm [54]. The lack of overlap between the histone PTM profiles associated with [SWI+] and Swi1 DAmP suggests that [SWI+] connects to the epigenome through a mechanism distinct from a mere loss of function. Mislocalization of Swi1 to the cytoplasm in the context of [SWI+] may also drag histone modifying enzymes along with it, thus influencing the histone PTM landscape. A hallmark of neurodegenerative disease is the cytoplasmic aggregation of prion-like proteins, which are typically diffuse in the nucleus in their native state [55]. For instance, aggregated FUS mutants associated with amyotrophic lateral sclerosis sequester the histone methyltransferase PRMT1 to the cytoplasm, leading to a decrease in the levels of asymmetric H4R3me2 [56]. Therefore, it is possible that the mislocalization of Swi1 in the [SWI+] state also leads to the sequestration of histone modifying enzymes, such as Rtt109 and Gcn5, the histone acetyltransferases which install H3K56ac. Microscopy experiments could illuminate interactions between histone modifying enzymes and [SWI+], alas we are unable to expediently carry out such experiments as antibodies for Rtt109 or Gcn5 are not commercially available. Alternatively, it is possible histone PTM disturbances arise from Swi1′s loss of function in [SWI+] [11]. While our results suggest a loss-of-function alone is not likely to be at play in the context of [SWI+], we cannot exclude a contribution from such a mechanism. Several interactions between mammalian SWI/SNF and histone modifying enzymes have been catalogued, such as the interaction between the Swi1 homologue ARID1A and the histone modifiers HDAC2 and the histone methyltransferase EZH2 [57]. Furthermore, the SWI/SNF-like chromatin remodeler SMARCAD1 associates with various PTM erasers such as HDAC1/2 and the histone methyltransferase G9a/GLP [58]. Hence, it is possible that loss of Swi1 function in [SWI+] disturbs the recruitment of various histone modifying enzymes and at least partially leads to histone PTM alterations.[PIN+] yeast display decreases in H3K4me3, H3K36me2, H3K36me3 and H3K79me3 levels, all marks implicated in gene activation. The reduction of di- and trimethylation of H3K36 in [PIN+] yeast underscores a potential role for Set2. Interaction with histone H4 is required for Set2 to install H3K36me2/me3, but it is not required for Set2 to install H3K36me1 [59]. Hence, it is possible that the interaction between Set2 and H4 is disrupted in the context of [PIN+]. Intriguingly, both H3K4me3 and H3K79me3 engage in positive histone crosstalk with H2BK123ub1 [47]. In yeast, H2BK123ub1 is installed by the ubiquitin ligase RAD6/BRE1. Loss of H2BK123ub1 by depletion of RAD6/BRE1 or mutation of the site causes severe loss of H3K4me3 and H3K79me3 [60,61]. Hence, it is possible that disruption of histone H2B ubiquitination is responsible for the selective decrease in H3K4me3 and H3K79me3, without disruption of other degrees of methylation on these sites.In contrast to [SWI+], ΔRnq1 cells overlapped with [PIN+] in direction of changes in H3K36me2/3 and H3K79me3. Overall, these results suggest that loss of function mechanisms might underlie [PIN+] connection to the epigenome, at least partially. Notably, a different genotype results in a different epigenetic profile for [PIN+]. Previous work reveals that [PSI+] elicits varied responses to external stresses dependent on yeast genotype [3,4,5,6]. Further investigation by co-immunoprecipitation of histone modifiers could reveal more information about the exact mechanisms linking [PIN+] to histone modifiers in different genetic backgrounds, but again, these experiments are complicated by the lack of commercially available antibodies against yeast histone modifiers. Importantly, as [PIN+] and ΔRnq1 yeast display different epigenetic landscapes compared to [pin−], our data suggests that Rnq1 might have a previously undescribed role in chromatin remodeling.In both cured [SWI+] and cured [PIN+] yeast, we find that RNA levels return to those seen in [prion−] yeast, suggesting that changes in global gene expression are linked to the prion state. We find that Swi1 DAmP yeast display RNA levels comparable to those observed in [SWI+] yeast. In contrast, the effects of [SWI+] on gene expression were previosuly found to be weaker than those of Δswi1 when compared to [swi−] controls in isogenic 1-4-1-1-D931 strains [62]. As for histone modifications, it is likely genetic background impacts gene expression and RNA levels. In the case of [PIN+], we find that [PIN+] and ΔRnq1 RNA levels diverge in both magnitude and direction, suggesting that the changes in overall gene expression in [PIN+] result from some mechanism distinct from Rnq1′s loss of function. Previous work in carried out in the context of Rnq1 overexpression found that several genes were differentially expressed in BY4741 [PIN+] and [pin−] yeast [63]. Out of the 23 upregulated genes, the majority corresponded to molecular chaperones and stress-related proteins, such as Hsp104, while many of the 27 downregulated genes were involved in cytokinesis, altogether suggesting that Rnq1 overexpression may cause defects in the cell cycle [63]. Further RNA-seq experiments can pinpoint specific genes differentially regulated in [PIN+] and ΔRnq1.Largely, discussion of prion function invokes loss of function phenotypes of the native protein. For instance, [PSI+] results in a loss of Sup35′s translation termination functionality [64]. Our findings highlight histone modifications as conduits for [PRION+] states to elicit phenotypic change. Furthermore, the lack of overlap between the histone PTM panoramas connected to [SWI+] and [PIN+] hint at specificity of biological relevance. Interplay with histone modifications can constitute an alternative, or perhaps complementary mechanism in [PRION+] function. On the basis of our results, chromatin immunoprecipitation experiments against H3K36me2 and H3K56ac in [swi−]/[SWI+] strains and H3K4me3, H3K36me2, H3K36me3 and K3K79me3 in [pin−]/[PIN+] strains would reveal which genes are impacted by these histone PTM changes and contribute to phenotype alterations. In the case of [PIN+], these genes might reveal valuable information about Rnq1′s function. Furthermore, future comprehensive characterization of modifications on other histone proteins–including histone H4 and H2B–will render a more complete picture of the [PRION+] epigenome. Overall, our results call attention to other potential mechanisms–separate from loss of function–by which fungal prions elicit phenotypic changes.