In carp, seasonal acclimatization involves a reprogramming of molecular and cellular functions, which begin by sensing environmental clues (temperature and photoperiod) that occur during an annual cycle. This process entails a fine modulation of a vast array of genes in order to provide allostasis. At a cellular level, one of the most dramatic differences corresponds to the reorganization of the nucleolar components observed during seasonal adaptation, when rRNA transcription and processing are clearly affected [45, 46]. In this context, we previously reported that epigenetic mechanisms are clearly involved in the phenotypical plasticity that is achieved by the carp. In particular, the hypermethylated state of ribosomal gene promoter in winter-acclimatized carp, as well as histone H2A replacement by variant subtypes of macroH2A, is dependent on the seasonal transcriptional activity of ribosomal genes, corroborating that changes in the architecture of the chromatin are reversible and dynamically respond to environmental changes [32, 33].
In the present study, the expression of H2A.Z was detected in all tissues studied. However, we observed that the total content of H2A.Z transcripts can change in the liver and testis in a seasonally-dependent manner. The same difference was found in western blot experiments, when liver H2A.Z content tended to be higher during the winter than summer. On the other hand, in immunodetection assays from pituitary cells, protein content of H2A.Z was significantly higher in winter-acclimatized fish. Nevertheless, this result is not correlated with mRNA levels. It is important to notice that in the present study, polyclonal antibodies were used for immunodetection and these antibodies cannot differentiate among different H2A.Z subtypes because they are directed to the extremely conserved C-terminal region of the protein; thus, limiting us to relate mRNA levels with protein levels of different H2A.Z subtypes.
We report for the first time four distinct subtypes of histone H2A.Z in the same organism. One of the deduced amino acid sequences from carp shares a 100% identity with the H2A.Z.2.1 of humans , whereas the other two (H2A.Z.3.1 and H2A.Z.3.2) do not correspond to any H2A.Z described until now. Between them, they share a 99.2% identity, differing only in Gly11 to Ser. In carp, we could not identify the existence of histone H2A.Z.1, but we do not discard its existence in other tissues different to those studied in this work.
Unexpectedly, we identified a new subtype of the histone variant H2A.Z (which was designated as H2A.Z.7) that is similar to H2A.Z.3.2, but possesses seven additional residues, more than other H2A.Z variants. The extra codons in H2A.Z.7 mRNA are in the same location as an intron in human H2A.Z, most likely indicating that H2A.Z.7 is a splicing (or mis-splicing) variant of H2A.Z.3.2. These seven additional residues are located in the region that corresponds to α-helix 2 , which could deviate H2A.Z C-terminal docking domain, most likely affecting the interaction surface with (H3-H4)2 tetramer. Clearly, structural studies should be carried out to determine the functional implications of these additional residues. However, considering we have no indication that the H2A.Z.7 protein is translated, it could also be possible that this corresponds to a pseudogene.
H2A.Z.7 presents a tissue-specific expression, which was detected only in the brain (summer and winter) and pituitary tissues (winter). This is not the first time that an unusual length of H2A.Z has been described. In human tissues and cell lines, a shorter version named H2A.Z.2.2 was found, and whose incorporation leads to severely unstable nucleosomes due to it containing a characteristic-docking domain .
Compared to mammals, the presence of this particular and species-specific histone H2A.Z.7 could be due to the three whole genome duplications (WGDs) that teleost have experienced. In particular, carp are believed to have had another round of genome duplications (4R) and became an evolutionarily recent tetraploid fish . This suggests that the presence of H2A.Z.7, and the other forms of carp H2A.Z, could be a consequence of WGD as a selective and important mechanism for eukaryote genome evolution in response to biotic and abiotic stress in carp.
The clustering pattern of carp H2A.Z subtypes in the phylogenetic tree is consistent with the differentiations and diversifications of teleost fish. So far, H2A.Z.7 from carp has an intense evolutionary diversification between other H2A.Z from mammals and, even more, within teleost fish. The presence of more than one subtype of H2A.Z is not a whim of random genetic drift, but different forms may have acquired new or complementary functions . Future experiments must show if the new subtypes described in this work perform similar or different roles in chromatin.
The incorporation of histone H2A.Z within chromatin is crucial for proper gene regulation [17, 22]. Genome-wide studies in several organisms show that H2A.Z is found throughout the genome; however, H2A.Z deposition is not random on chromatin, but rather specific and highly regulated [35, 49]. In yeast, H2A.Z is distributed in active and inactive RNA polymerase II promoters [35, 38, 39], whereas in higher eukaryotes it is mostly in active genes . Interestingly, we report for the first time the seasonally-dependent enrichment of H2A.Z throughout the ribosomal cistron, concomitant with differential modulation of rDNA activity using a natural model of environmental plasticity, such as the carp fish. We observed that the levels of H2A.Z enrichment were higher in the ribosomal cistron during summer, a season in which rRNA transcription is highly active . This particular evidence confirms that the correlation between H2A.Z enrichment with the ribosomal transcriptional activity is not only associated to RNA polymerase I-related genes, since we detected that the enrichment of H2A.Z agrees with seasonally active gene promoters that are transcribed by RNA polymerase II.
In higher eukaryotes, p400 and SRCAP (SWI2/SNF2-related CBP activator protein) complexes are ATP-dependent chromatin remodelers specifically exchanging canonical H2A–H2B with H2A.Z–H2B dimers within the nucleosomes [51, 52]. In carp, both glycogen and lipid content in hepatocytes from acclimatized carp differ significantly between seasons . Taking into account that ribosome production is a major biosynthetic and energy-consuming activity of eukaryotic cells, which adapts rapidly to changes in intracellular energy status , it is plausible to speculate that variations of cellular AMP/ATP ratios could modulate H2A.Z–H2B incorporation by p400 and SRCAP complexes in an ATP-availability manner. Certainly, this speculation must be demonstrated experimentally.
Additionally, in order to corroborate if H2A.Z enrichment is consistent with a chromatin state, we evaluated H2A.Z colocalization with specific epigenetic markers of active (H4K12ac)  or inactive (H3K9me3)  gene expression. Our results showed that H2A.Z and the euchromatin marker (H4K12ac) co-localize along the ribosomal cistron but not in a seasonally-dependent manner. In contrast, we showed a significant colocalization between H3K9me3 and H2A.Z on rDNA cistron during winter, a season in which rRNA transcription and processing are highly inactive. Then, when we compared the colocalization of H2A.Z with these epigenetic markers on RNA polymerase II genes, they were consistent with the active or inactive gene states. These results agree with the controversial role of H2A.Z in the transcription of RNA polymerase II genes .
Histone tails are subjected to PTMs that alter the higher-order folding of chromatin. In particular, H2A.Z has been reported to be acetylated, sumoylated, and ubiquitylated. The H2A.Zub marker is involved in the maintenance or formation of heterochromatin because it was found predominantly on the inactive X chromosome . In this study, we investigated if H2A.Z ubiquitylation in the ribosomal cistron can contribute to regulate carp seasonal rRNA transcription. First, we evaluated transcriptional expression of the most relevant ubiquitylation factors. We found a greater gene expression of RING1b and Bmi1 in summer than in winter, whereas USP10 showed no change between seasons. These findings are consistent with the relevant enrichment of H2A.Zub on the ribosomal cistron of summer-acclimatized carp, when the activity of rRNA transcription is significantly higher. Unlike the uniform pattern of H2A.Z on the ribosomal cistron during summer, H2A.Zub enrichment in the same season was found mainly on the IGS and CP regions. These two locations of the rDNA gene might play a crucial role in rRNA transcription since H2A ubiquitylation (and consequently H2A.Zub) can serve as a docking site for recruiting other factors, or preventing chromatin access to transcriptional regulators . On the other hand, at least in hepatocytes isolated from carp, the two promoters of RNA polymerase II studied did not show significant enrichment of H2A.Zub in any season. Nevertheless, similar findings have been reported in Saccharomyces cerevisiae, where H2BK123ub has negative consequences on the assembly of RNA polymerase II at promoters and stimulates transcription elongation . Finally, it would certainly be desirable to obtain antibodies specific for carp H2A.Zub to unambiguously corroborate the ubiquitylation state of carp H2A.Z. Additionally, wide-genome screening of H2A.Zub should be performed in order to better understand the contribution of this PTM in transcriptional regulation during carp seasonal acclimatization.