| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Obstetrics and Gynecology,2 Keio University School of Medicine, Shinjuku-ku, Tokyo 160-0082, Japan
Department of Obstetrics and Gynecology,3 Chiba University School of Medicine, Chuo-ku, Chiba 260-8670, Japan
Division of Reproductive Sciences,4 Oregon National Primate Research Center, Oregon Health and Sciences University, Beaverton, Oregon 97006
The Jackson Laboratory,5 Bar Harbor, Maine 04609
Center for Animal Transgenesis and Germ Cell Research,6 University of Pennsylvania, School of Veterinary Medicine, Kennett Square, Pennsylvania 19348-1692
Division of Urology,7 Department of Surgery, University of Utah, School of Medicine, Salt Lake City, Utah 84132
Department of Gynecological Endocrinology,8 University of Medical Sciences in Poznan, 60-535 Poznan, Poland
Developmental Endocrinology Branch,9 National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892
National Cancer Institute,10 NIH, Bethesda, Maryland 20892
Department of Biology,11 University of Pennsylvania, Philadelphia, Pennsylvania 19104-6018
Division of Reproductive Sciences,12 Huntsman Cancer Institute, University of Utah Health Sciences Center, Salt Lake City, Utah 84112
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
5301 base pair) H1foo gene maps to chromosome 6 and that the corresponding primary H1foo transcript gives rise to two distinct, alternatively spliced mRNA species (H1foo
and H1fooß). The expression of the oocytic H1FOO transcript and protein proved temporally coupled to the recruitment of resting primordial follicles into a developing primary follicular cohort and thus to the critical transition marking the onset of oocytic growth. The corresponding potential protein isoforms (H1FOO and H1FOOß), both nuclear localization sequence-endowed but export consensus sequence-free and possessing a significant net positive charge, localized primarily to perinucleolar heterochromatin in the oocytic germinal vesicle. Further investigation will be required to define the functional role of the H1FOO protein in the ordering of the chromatin of early mammalian development as well as its potential role in defining the primordial-to-primary follicle transition.
meiosis, oocyte development
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
More recently, and unexpectedly, we discovered two transcriptional variants of the H1foo gene. To further clarify this phenomenon, the chromosomal assignment, the allelic representation, and the genomic organization of the H1foo gene were studied. The genesis, postnatal, and perifertilizational ontogeny, as well as the nuclear and subnuclear localization of the two alternatively spliced isoforms were also evaluated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female C57BL/6J mice, purchased from Jackson Laboratories (Bar Harbor, ME), 19 days of age upon arrival, were initially quarantined for 3 days at the University of Utah Animal Resources Center. The latter adheres to the guidelines outlined by The Animal Welfare Act and to Institutional Animal Care and Use Committee protocols.
Oocytic and early embryonic H1foo transcripts in the C57BL/6J and 129/SvJ mouse strains Reverse transcription-polymerase chain reaction (RT-PCR) Analysis. Oocytes and embryos were collected as described [1]. The corresponding total RNA (isolated with Qiagen's RNA easy Kit Qiagen Inc., Valencia, CA) was subjected to RT-PCR by using forward primer-691-710-5'-TCAAACGCAGTGGGAGCAGG-3' and reverse primer 791-770-5'-GGAAGCAACGCTATTCTGGACC-3' to amplify the 3' region of exon 4 and the 5' region of exon 5.
Cloning and sequencing of the mouse H1foo gene A mouse bacterial artificial chromosome (BAC) genomic library (RPCI-22, Research Genetics, Huntsville, AL) was screened with the H1foo cDNA. Several BAC clones were identified, amplified, purified, and sequenced.
In situ hybridization Ovaries were snap frozen on dry ice, stored at –70°C, serially sectioned (10-µm thickness) at –15°C and mounted onto poly-L-lysine coated slides for in situ hybridization as previously described [5].
Generation of green fluorescent protein (GFP)-H1foo fusion constructs
Complementary DNAs corresponding to the two H1foo varieties ( and ß) were prepared by RT-PCR of total mouse ovarian RNA. The consequent cDNAs were subcloned into a human cytomegalovirus-driven promotor NT-GFP-TOPO vector (Invitrogen Corp., Carlsbad, CA) replete with a bovine growth hormone (BGH) polyadenylation sequence.
Subcellular localization of GFP-H1FOO and H1FOOß fusion proteins: studies in the germinal vesicle-stage oocyte
Given that fully grown germinal vesicle (GV)-stage oocytes are transcriptionally inert [6], oocytic expression of exogenous templates requires the injection of translatable mRNA templates. GFP-H1foo and GFP-H1fooß mRNAs, generated by in vitro transcription and polyadenylation (Ambion, Inc., Austin, TX), were injected into oocytes 45–48 h after harvesting. An additional incubation period of 18–20 h at 37°C followed in the presence of 10 µM milrinone (a potent inhibitor of meiotic maturation). Intact living oocytes were imaged on a temperature-controlled stage with a Zeiss 510 Laser scanning confocal microscope as previously described [7, 8].
Generation of H1FOO-directed antibodies
The first H1FOO polyclonal serum (#1), used previously [1], was raised against a 30-mer peptide corresponding to residues 217–246 of the H1FOO ORF (proximal segment of the C-tail). The latter peptide is common to the H1FOO and H1FOOß proteins, save the last three amino acids, which are H1FOO unique (Fig. 3). Thus, it is likely that antibody #1 is not isoform specific. The second H1FOO polyclonal serum (#2), raised against a 15-mer peptide corresponding to residues 283–297 of the H1FOO ORF (distant segment of the C-tail), is H1FOO isoform specific. The third H1FOO polyclonal serum (#3), raised against a 16-mer peptide corresponding to residues 21– 36 of the proximal N-tail, is not isoform specific. All rabbit antimouse antibodies were raised and affinity-purified by Bethyl Laboratories Inc. (Montgomery, TX).
|
Detection of oocytic H1FOO protein(s) Oocytes (132–1600) were extracted, frozen at –80°C and subjected to Western blot analysis as previously described [1]. Immunofluorescent detection of H1FOO in whole mounted oocytes was performed as previously described [1]. Immunohistochemical detection of H1FOO made use of ovaries from 2-wk-old mice. Fixed, paraffin-embedded sections were processed as previously described [1] with antibody #3.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In silico mining of the public mouse genome databases as well as fluorescence in situ hybridization (FISH) and Southern blot analysis (not shown) mapped a single copy of the mouse H1foo gene to chromosome 6/band 6E3 at a position marking 75% of the distance from the heterochromatic-euchromatic boundary to the telomere.
The Single CopyH1fooGene Is Alternatively Transcribed: Two Distinct Transcripts
In the course of sequence-verifying PCR amplicons of eight putative (randomly picked) H1foo clones from mouse (C57BL/6J) cDNA clones prepared by reverse-transcription of ovarian total RNA, two sequences, varying by a GTAG within the H1foo C-tail, were identified (nucleotides 772– 775) (Fig. 1A). To further evaluate the apparent existence of two distinct ovarian H1foo transcripts, total RNA of GV-stage oocytes of C57BL/6J mice was reverse transcribed, amplified by PCR with H1foo-specific primers, and resolved by PAGE. As shown (Fig. 1, B and C), two distinct (100 and 104 base pairs [bp]) H1foo amplicons were detected. Control reactions without reverse transcriptase or cDNA template did not yield PCR products (not shown). Although the two H1foo transcripts were detected throughout oogenesis and early preimplantation embryogenesis, the longer (GTAG-replete) H1foo transcript (designated H1fooß), proved less abundant than its shorter (GTAG-less) counterpart (H1foo). This latter phenomenon was particularly apparent at the zygote stage, wherein only limited, albeit measurable representation could be documented for the H1fooß isoform. To evaluate the possibility that the two H1foo transcripts represent strain-dependent genomic polymorphism of the exon 4-intron 4 junction, tail genomic DNA from C57BL/6J and 129/SvJ mice was PCR amplified. Single amplicons were clearly detected for both strains of mice (Fig. 1C), thereby ruling out genomic, strain-dependent polymorphism.
|
Genomic Organization of theH1fooGene: The Mechanism Underlying the Alternative Splicing of H1foo RNA
To better understand the mechanisms responsible for generating the H1foo and H1fooß transcripts, the structure of the H1foo gene was characterized. The mouse H1FOO gene comprised five exons and four introns (Fig. 2A) spanning 5301 bp, a finding confirmed by mining the public mouse genome database. Importantly, the H1foo mouse gene featured a triple AGGT repeat in and around the junction of exon 4 with intron 4. Given the role of the AGGT motif as a general consensus splicing signal at the 5' end of the intron (reviewed, [9]), it is suggested that the two H1foo transcripts result from alternative splicing of exon 4 of the H1foo transcript to the downstream exon 5 by way of the first or second AGGT repeat, respectively (Fig. 2B). Specifically, the H1foo transcript appears to make use of the first AGGT splice signal thereby splicing the 3' end of exon 4 to the 5' end of exon 5. This splicing event requires scission between the two Gs of the first AGGT splice signal. H1fooß, in turn, is likely the result of a splicing event involving the second AGGT. The third AGGT repeat is apparently not used. To determine if the alternative splicing of H1foo transcripts may transpire in species other than the mouse, a comparable analysis of the National Center for Biotechnology Information (NCBI) genomic databases of the corresponding rat, human, and chimpanzee orthologues was undertaken. As shown (Fig. 2C), the rat H1foo gene also features a triple AGGT repeat at the junction of exon 4 with intron 4. In contrast, only two splice sites were noted in the comparable region of the human and chimpanzee genomes. The chimpanzee genome contains a GC instead of the second GT found in the other three species, a known variant of the 5' splicing spite (reviewed, [10]]). It is unknown at this time if alternative transcripts actually exist in species other than the mouse.
|
TheH1fooGene Is Alternatively Transcribed: Two Deduced Proteins
The incorporation of the GTAG insert into the H1fooß transcript gives rise to a translational frame shift, thereby introducing codons corresponding to G, P, and E followed by a premature stop codon and thus premature chain termination of the ORF (Fig. 3) at 246 residues (as compared with the 304 residues previously reported for H1FOO). H1FOOß is thus marked by partial truncation of the C-tail. These data suggest that two RNA species, one with and one without a 4-nt (GTAG) insert, may encode two H1FOO protein isoforms.
Net positive charge of H1FOO and H1FOOß
An analysis of the H1FOO protein features 44 lysine residues for a total representation of 14.5% (Fig. 3). The majority of the lysine residues (33) localize to the C-tail, wherein they comprise 18.6% of the total complement of amino acids. Ten additional lysine residues are situated in the globular domain (11.4%). Only one lysine residue (2.6%) resides in the N-tail. Whereas lysine may be the dominant basic amino acid of H1FOO, it is by no means the sole basic residue. Indeed, H1FOO contains, in addition, 19 arginine residues and 6 histidine residues. Thus, the total complement of basic residues in H1FOO is 69, or 22.7%, of the total amino acid complement. Accordingly, the net positive charge of the C-tail, i.e., (R + K + H) – (D + E) is 32, that of the globular domain 16, and that of the N-tail 2, for an overall net positive charge of 50 for H1FOO. Correspondingly, H1FOOß revealed a net positive charge of 43 due to the loss of a total of 11 basic residues and 4 acidic residues as a consequence of the C-tail truncation (Fig. 3).
Structural analysis of the H1FOO protein
Given the likely nuclear residence of H1FOO, a putative bipartite nuclear localization sequence (NLS) was sought and identified in amino acids 154–170 of the proximal C-tail of H1FOO (http://psort.nibb.ac.jp). The amino acid sequence in question, KKDQVGKATMEKGQKRR (amino acids 154–170) (Fig. 3), fulfills the criteria [11] of having two proximal basic residues (K x 2), a 10-amino acid linker, and a 5-amino acid stretch in which at least three of five residues are basic (KRR). Five additional putative monopartite NLS sites (seven residues each) were also noted at residues 123, 125, 183, 289, and 291. In that no nuclear export consensus sequence (NES; L-X[2–3]-F/I/L/V/M-X[2–3]-L-X-L/I or LX[1–3]LX[2–3]LXL) could be uncovered [12] for H1FOO and in that no immunoreactive (Fig. 6E) or GFP-tagged (Fig. 6, F–K) H1FOO has been detected in the oocytic cytoplasm, H1FOO is likely a stable nuclear resident. Because the very terminal portion of the C-tail of H1FOO is generally absent in the generic somatic H1, the term nonsomatic extension was coined to characterize this segment of the C-tail of H1FOO beyond the somatic-like basic terminal triplet (Fig. 3). Additional domain analysis failed to detect other major structural motifs.
|
The H1FOO isoforms are expressed in the mouse oocyte
To evaluate the expression profile of the projected H1FOO protein isoform during oocyte growth, use was made of the H1FOO-specific antiserum (#2, Fig. 3). Western blot analysis of mouse oocytes revealed, on Days 5 and 18 of life (Fig. 4), a single immunoreactive band. No signal corresponding to the H1FOO protein was noted in Day 2 oocytes despite the use of a total of 1600 oocytes. In contrast, only 132 oocytes were required to provide a meaningful signal on Day 18. Analysis with antibody #1, which recognizes both isoforms, revealed two bands of 37 and 42 kDa 2 days after birth, increasing with intensity to 18 days (Fig. 4).
|
Postnatal ontogeny of the H1foo transcript and the protein it encodes Evaluation of the follicular ontogeny of H1foo transcripts revealed primary and secondary ovarian follicles as well as preovulatory follicles to encase H1foo-expressing oocytes (Fig. 5, C–H). H1foo transcripts were less abundant but also noted in the oocytes of primordial follicles (Fig. 5, A and B). Parallel assessment of the follicular ontogeny of H1FOO immunoreactivity (Fig. 5, I–L; antibody #3, Fig. 3), confirmed the preceding Western blot observations, the signal being limited to primary, secondary preantral and secondary antral follicles.
|
Subnuclear localization of endogenous immunoreactive H1FOO: studies in the GV-stage oocyte
H1FOO immunoreactivity (antibody #1, Fig. 3) colocalized with 4',6'-diamidino-2-phenylindole (DAPI)-positive perinucleolar nuclear substrate in fully grown GV-stage oocytes in both the nonsurrounded nucleolus (NSN; Fig. 6, A and C) and surrounded nucleolus (SN; Fig. 6, B and D) configuration. The latter terminology refers to the progressive condensation of GV chromatin around the nucleolus, a phenomenon associated with, but not necessarily causally related to, the acquisition of meiotic competence [13]. A component of H1FOO-associated nuclear chromatin localized to nonperinucleolar loci displaying apparent continuity with the nuclear membrane (Fig. 6E). Comparable observations were made on the intraoocytic injection of in vitro-transcribed/ polyadenylated/capped mRNAs encoding for H1FOO and H1FOOß-GFP fusion proteins (Fig. 6, F–K).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It is not surprising that earlier, low-resolution Northern blot gels [1] failed to detect the alternative H1fooß transcript. In all likelihood, the generation of the H1foo and H1fooß transcripts results from the splicing of the primary H1foo transcript such that two alternative 3' ends of exon 4 are spliced onto the 5' end of exon 5 (Fig. 2, A and B). It is highly probable that the choice of splice site alternates at random and that the resultant two transcripts are thus equally represented. It is unknown if there is any biological significance to the apparent decrease of the H1fooß transcript in the zygote (Fig. 1B).
In an earlier report [1], Western blot analysis of the H1FOO protein in GV-intact oocytes of C57BL/6J mice disclosed a major 42-kDa band as well as a minor 37-kDa band when using isoform-common antibody #1. The precise nature of the two distinct immunoreactive H1FOO species was not clear at the time. The present observation of alternative mRNA splicing raises the possibility that the two previously reported H1FOO immunoreactive bands [1] may reflect two alternatively encoded proteins, i.e., a foreshortened amino acid sequence of 246 residues (H1FOOß) and a longer 304 residue H1FOO protein (Fig. 3). Studies employing the H1FOO-specific antibody #2 (Fig. 4) provide direct documentation of the expression of the H1FOO protein isoform in oocytes. The likelihood that the H1FOOß protein isoform is also expressed in mouse oocytes can only be inferred from the presence of a lower molecular weight immunoreactive band in oocytes probed with the isoform-common antibody #1 (Fig. 4) [1]. The generation of an H1FOOß-specific antiserum for the purpose of documenting the existence of H1FOOß, although desirable, is probably not feasible given the limited differences in the pretruncation sequence of the two putative H1FOO isoforms. Accordingly, alternative approaches (e.g., sequential immunoprecipitation and/or proteomics) may be required to document the in vivo existence of two distinct H1FOO protein isoforms.
Assuming in vivo translation of H1foo and H1fooß transcripts, might one expect functionally distinct proteins? For one, partial loss of the lysine-rich C-tail (a complement of six lysine, three arginine, and two histidine residues) would likely change the theoretical pI of H1FOO, a change that might alter its interaction with DNA (Fig. 3). Indeed, the net positive charge of H1FOOß, 43, proved lower than that of H1FOO, 50, thereby raising the prospect of functional differences. Consideration must also be given to the possibility that H1FOOß might act in a dominant negative inhibitory manner given the retention of the globular DNA-binding domain. Gain of a deleterious function is also a possibility. No data are available at this time to support or refute the above hypotheses.
Given the nucleosomal role of H1s, nuclear localization must be assured. In that context, H1FOO features five putative monopartite NLS sites (seven residues each). A bipartite NLS (Fig. 3), the most common NLS variety, was also apparent within amino acids 154–170 (KKDQVGKATMEKGQKRR) of the proximal C-tail (http://psort.nibb.ac.jp). Inevitably, predictions of an NLS would have to be verified by deletion analysis. Given the basic nature of H1FOO, it is surprising that more potential NLS motifs (characteristically basic in nature) were not uncovered. It is thus conceivable that other NLS motifs may exist, i.e., noncanonical NLS sequences that may also be recognized by other members of the NLS receptor transport superfamily. Either way, the very fact that a GFP-H1FOO fusion protein (estimated at over 60 kDa), likely too large to enter the nucleus by NLS-independent diffusion, localizes to the nucleus, suggests that H1FOO contains NLS functionality that could be defined by deletion analysis.
As shown, the two Gfp-H1foo fusion constructs ( and ß), localized exclusively to the GV of mouse oocytes (Fig. 6, F–K). Sub-GV localization of GFP-tagged H1FOO (and of immunofluorescently labeled H1FOO; Fig. 6, A and E) revealed predominant perinucleolar presence as well as some limited peripheral presence at the nuclear membrane and along the putative centromeric heterochromatin. These observations suggest that the C-tail truncation of H1FOOß (residues 247–304, inclusive of 11 basic residues) does not preclude nuclear localization, thereby ruling out the potential relevance of the putative monopartite NLS motifs at residues 289 and 291. Finally, because no nuclear envelope exists in MII oocytes and because none is reformed until the pronuclear embryo stage, H1FOO could interact with nuclear material during this developmental window in an NLS-independent manner. In that no NES could be uncovered for H1FOO (consensus sequence: L-X[2–3]-F/I/L/V/ M-X[2–3]-L-X-L/I or LX[1–3]LX[2–3]LXL) and in that no immunoreactive (Fig. 6, A–E) or GFP-tagged (Fig. 6, F– K) H1FOO is detected in the cytoplasm, it is highly probable that H1FOO is a stable nuclear resident.
To date, genes that may partake in the primordial-to-primary follicular transition remain largely elusive. Among somatic genes, the kit ligand, Amh (anti-Mullerian hormone), and the activins have been implicated [18, 19]. Among germ cell genes, however, no viable candidates can be pointed out at this time. In that the expression of the H1foo gene is first detected in the prophase I-arrested oocytes of primordial follicles (Fig. 5, A and B) and in that a marked increase in H1foo transcripts is detected for oocytes of primary follicles (Fig. 5, C and D), it is highly probable that the expression of the H1foo gene is tightly and temporally coupled to the process of primordial follicle recruitment and, by extension, to the onset of oocytic growth. It is of interest to note that previous work implicated this very same developmental transition in the disappearance of somatic H1s [20]. It is thus possible that the transition from somatic to pre-embryonic H1s is completed at a point in time when the resting primordial follicle is recruited into the growing primary follicular pool.
The significance of H1FOO to the chromatin of early development is strongly suggested by the recognition that nature saw fit to link the pre-embryonic developmental program to H1FOO rather than to one of seven other known mammalian H1s. The likely extraordinary significance of H1FOO also draws on its possible role in the programming of the maternal germline and its potential involvement in the reordering of the male pronucleus. Further, H1FOO is a likely downstream target of the oocytic master gene Fig (unpublished), one of only a few known oocyte-specific proteins of which ablation gave rise to a profound reproductive phenotype [21].
More recently, Wilbrand and Olsen [22], as well as Muller et al. [23], have reported the discovery of H1M, a pre-embryonic H1 of the zebrafish. Importantly, H1M marks the primordial germ cells during gastrulation as well as migration for up to 24 h postfertilization. As such, these observations implicate H1M in the ordering of the chromatin of the primordial germ cell, thereby suggesting that pre-embryonic H1s may play a critical role not only during oocytic growth/maturation and zygotic progression but also potentially during primordial germ cell formation and migration. Moreover, these observations may open up the possibility that pre-embryonic H1s may partake in the regulation of early meiosis. The recent observation that an H1 linker histone is capable of suppressing homologous recombination in the yeast paradigm may be commensurate with the meaningful role in the first component of the meiotic process [24]. The significance of this apparent biphasic temporal pattern of expression for H1FOO is unclear at this time. Comparable findings have yet to be reported for the mouse H1FOO paradigm.
Although the precise role of H1s, in general, and of pre-embryonic H1s, in particular, remains uncertain, explosive developments in the chromatin field have reaffirmed the critical role of H1s as ligands, stabilizers, and gate keepers of the nucleosomal assemblage [25, 26]. How and to what degree H1s partake in the ordering of pre-embryonic chromatin constitutes a substantial knowledge gap. In this context, H1FOO may well constitute a critical component of the genetic and epigenetic programs of the chromatin of early mammalian development.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received: 28 May 2004.
First decision: 25 June 2004.
Accepted: 27 August 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Maeda, S. Sato, N. Hattori, S. Tanaka, S. Yagi, and K. Shiota DNA Hypomethylation Circuit of the Mouse Oocyte-Specific Histone H1foo Gene in Female Germ Cell Lineage Biol Reprod, May 1, 2008; 78(5): 816 - 821. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V.J. Hall, D. Compton, P. Stojkovic, M. Nesbitt, M. Herbert, A. Murdoch, and M. Stojkovic Developmental competence of human in vitro aged oocytes as host cells for nuclear transfer Hum. Reprod., January 1, 2007; 22(1): 52 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. J. Racki and J. D. Richter CPEB controls oocyte growth and follicle development in the mouse Development, November 15, 2006; 133(22): 4527 - 4537. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
