Mouse CCDC79 (TERB1) is a meiosis-specific telomere associated protein
© Daniel et al.; licensee BioMed Central Ltd. 2014
Received: 6 December 2013
Accepted: 14 May 2014
Published: 22 May 2014
Skip to main content
We’re sorry, something doesn't seem to be working properly. Please try refreshing the page. If that doesn't work, please contact us so we can address the problem.
© Daniel et al.; licensee BioMed Central Ltd. 2014
Received: 6 December 2013
Accepted: 14 May 2014
Published: 22 May 2014
Telomeres have crucial meiosis-specific roles in the orderly reduction of chromosome numbers and in ensuring the integrity of the genome during meiosis. One such role is the attachment of telomeres to trans-nuclear envelope protein complexes that connect telomeres to motor proteins in the cytoplasm. These trans-nuclear envelope connections between telomeres and cytoplasmic motor proteins permit the active movement of telomeres and chromosomes during the first meiotic prophase. Movements of chromosomes/telomeres facilitate the meiotic recombination process, and allow high fidelity pairing of homologous chromosomes. Pairing of homologous chromosomes is a prerequisite for their correct segregation during the first meiotic division. Although inner-nuclear envelope proteins, such as SUN1 and potentially SUN2, are known to bind and recruit meiotic telomeres, these proteins are not meiosis-specific, therefore cannot solely account for telomere-nuclear envelope attachment and/or for other meiosis-specific characteristics of telomeres in mammals.
We identify CCDC79, alternatively named TERB1, as a meiosis-specific protein that localizes to telomeres from leptotene to diplotene stages of the first meiotic prophase. CCDC79 and SUN1 associate with telomeres almost concurrently at the onset of prophase, indicating a possible role for CCDC79 in telomere-nuclear envelope interactions and/or telomere movements. Consistent with this scenario, CCDC79 is missing from most telomeres that fail to connect to SUN1 protein in spermatocytes lacking the meiosis-specific cohesin SMC1B. SMC1B-deficient spermatocytes display both reduced efficiency in telomere-nuclear envelope attachment and reduced stability of telomeres specifically during meiotic prophase. Importantly, CCDC79 associates with telomeres in SUN1-deficient spermatocytes, which strongly indicates that localization of CCDC79 to telomeres does not require telomere-nuclear envelope attachment.
CCDC79 is a meiosis-specific telomere associated protein. Based on our findings we propose that CCDC79 plays a role in meiosis-specific telomere functions. In particular, we favour the possibility that CCDC79 is involved in telomere-nuclear envelope attachment and/or the stabilization of meiotic telomeres. These conclusions are consistent with the findings of an independently initiated study that analysed CCDC79/TERB1 functions.
The production of haploid gametes from diploid germ cells during meiosis is fundamental for sexual reproduction. Haploid gametes are produced by a single round of premeiotic DNA replication followed by two rounds of chromosome segregation. Ploidy reduction depends on features of chromosome behaviour that are specific to the first meiotic division . One key feature is the pairing of homologous chromosomes (homologues) during the first meiotic prophase. High fidelity pairing of homologues requires chromosome movements, which are driven by the active movement of telomeres along the nuclear envelope (NE) in diverse taxa [2–5]. At the onset of meiotic prophase, telomeres associate with protein complexes that span the NE, and provide physical linkage between telomeres and cytoplasmic motor proteins. Hence, attachment of telomeres to the NE enables cytoplasmic motors to move telomeres along the NE [3, 6, 7].
Trans-NE protein complexes that connect telomeres to cytoplasmic motors contain SUN- and KASH-domain proteins, which are imbedded in the inner and the outer NE membrane, respectively . In various yeast species, telomeres are tethered to the SUN-domain inner NE protein through meiosis-specific protein complexes [5, 8, 9]. In fission yeast, the linkage of telomeres to the SUN-domain (Sad1) protein requires at least two meiosis-specific connector proteins, Bqt1 and 2, whose interaction with telomeres depends on the telomeric DNA repeat-binding protein Taz1 and spRap1 [8, 10, 11]. Although RAP1 is a constitutive component of mammalian telomeres and is the predicted mammalian orthologue of spRap1, it is not required for telomere-NE interaction in mice , Nevertheless, SUN-domain proteins, SUN1  and possibly SUN2 [14, 15], are required for tethering telomeres to the NE during meiosis in mice. These two proteins are present in the inner NE both in somatic and meiotic cells. Therefore, meiosis-specific modifications to constitutive telomere proteins, or additional meiosis-specific telomere components, must exist in order to establish telomere-NE attachments in mammalian meiocytes.
In addition to telomere-NE attachment, protection of chromosome ends during meiotic prophase requires meiosis-specific changes in telomere biology . The shelterin complex/telosome is employed in both somatic and meiotic cells to safeguard chromosome ends from DNA-damage response and enzymatic attacks, and to ensure maintenance of telomere length [17–19]. Although somatic and meiotic telomeres share known shelterin components, e.g. TRF1, TRF2 and RAP1, maintaining the stability of telomeres during meiosis is known to require an additional factor, possibly due to meiosis specific-features of recombination. The meiosis-specific cohesin SMC1B is required for both telomere stability and efficient telomere-NE attachment during the first meiotic prophase . However, the function of SMC1B in telomere biology is not well understood, and the molecular nature of meiosis-specific telomere modifications remains largely unexplored.
Here we report, in line with a recently published study , that the coiled-coil-domain containing protein 79 (CCDC79) is a meiosis specific telomere-associated protein in mice. By investigating the behaviour of CCDC79 in wild type (wt), SUN1-deficient and SMC1B-deficient meiocytes, we identify CCDC79 as a candidate for mediating meiotic telomere-NE interaction and for stabilising meiotic telomeres in mammals.
To identify uncharacterised proteins that are possibly involved in meiotic chromosome biology, we screened for mouse genes whose expression is upregulated in the developing gonads upon entry of germ cells into the first meiotic prophase in both sexes (our unpublished results). Ccdc79 was one of the identified genes.
Having established that Ccdc79 is upregulated in the gonads at the time of the first meiotic prophase, we tested if Ccdc79 was expressed in somatic tissues. We compared Ccdc79 expression in the testis to 17 somatic tissues by RT PCR (Figure 1C). Preferential expression of Ccdc79 was detected in the testis of adult mice, indicating that Ccdc79 is largely restricted to tissues that contain meiotic cells. We then asked if Ccdc79 is expressed specifically in meiotic germ cells. Using fluorescence-activated cell sorting to separate somatic cell populations and prophase stage oocytes from foetal ovaries , we measured Ccdc79 expression in the sorted cell populations by RT-PCR (Figure 1D). Ccdc79 expression was detected in oocytes and was not detected in somatic ovarian cells, indicating that Ccdc79 expression is restricted to meiotic cell types.
Given our finding that CCDC79 foci first appeared in leptotene, it was possible that CCDC79 association with telomeres coincided with the attachment of telomeres to the NE and with the formation of telomere attachment plates. To test this possibility we detected CCDC79, TRF1 and SUN1 in surface spreads of early-leptotene spermatocytes and quantified co-localization between the foci of these proteins (Figure 4B). We found that the vast majority (83.4%) of TRF1 foci, which constitutively marks telomeres, was associated with both anti-CCDC79 and anti-SUN1 signal. The second largest fraction (9.1%) of TRF1 foci co-localized with neither CCDC79 nor SUN1, and only very small fractions of foci showed co-localization with either CCDC79 (4.7%) or SUN1 (2.8%) foci. Thus, CCDC79 tends to associate with telomeres that also bind to SUN1 in both leptotene and pachytene, and CCDC79 is missing from the majority of telomeres that lack SUN1 in the early leptotene stage. This indicates that CCDC79 association with telomeres largely coincides with the initiation of telomere-NE attachment in wt spermatocytes.
Thus, there is a strong correlation between the localization of CCDC79 to telomeres and the attachment of telomeres to the NE in both wt and Smc1b-/- spermatocytes, consistent with the possibility that CCDC79 constitutes part of the protein complex that forms bridges between chromosome ends and the cytoplasmic cytoskeleton during meiotic prophase.
Telomeres perform meiosis-specific functions that are essential for ploidy reduction and maintenance of genome integrity during meiosis. To facilitate active chromosome movements and ensure high fidelity pairing of homologues during early meiotic prophase, telomeres must attach to trans-NE protein complexes, thereby establishing a link to cytoplasmic motor proteins [3, 16]. DNA double strand breaks are introduced into the genome and are required for homologue pairing and formation of inter-homologue crossovers. The breaks are subsequently repaired by recombination machinery that is substantially altered during meiosis, in comparison to mitosis . Chromosomal DNA ends resemble DNA double strand breaks, and one of the main roles of telomeres is to protect chromosome ends and prevent undesired recombination events. It is therefore expected that meiotic telomeres need to adapt to meiosis-specific characteristics of recombination . Consistent with this notion, maintenance of the structural integrity of telomeres during the first meiotic prophase requires the meiosis-specific cohesin SMC1B . Despite the importance of these meiosis-specific features of telomere biology, the molecular changes in telomeres that underpin telomere-NE attachment and maintenance of telomere integrity during meiosis have remained largely unexplored.
We identified CCDC79 as a protein that associates with telomeres specifically during the first meiotic prophase in mice. We found that CCDC79 and SUN1 association with telomeres tightly correlated with each other in both wt and SMC1B-deficient meiocytes. This indicates that CCDC79 association with telomeres coincides with telomere attachment to the SUN1 containing trans-NE protein complexes.
We further tested the relationship between CCDC79 and telomere-NE attachment by examining CCDC79 localization in the Sun1-/- spermatocytes, where telomeres attachment to the NE is severely reduced. The observation that CCDC79 localization to chromosome ends is unaffected in SUN1-deficient spermatocytes shows that CCDC79 localization to telomeres does not depend on SUN1 and telomere-NE attachment. This also indicates that CCDC79 is not part of the protein complex that extends from SUN1 toward the cytoplasmic motors through the NE. Rather, these observations are consistent with the possibility that CCDC79 is positioned between the telomeric DNA and SUN1 at sites of telomere-NE attachment.
In agreement with this scenario, CCDC79 contains a telobox-like Myb domain at its C terminus, a feature that is shared by shelterin complex components that bind to telomeric repeats [29, 30]. Thus, the domain structure of CCDC79 indicates a potential for CCDC79 to interact directly with telomeric repeats.
Our observations indicate a role for CCDC79 in meiotic telomere biology. In particular, the behaviour of CCDC79 in SMC1B-deficient meiocytes is informative. In the absence of the meiosis-specific SMC1B cohesin, meiocytes display a diverse set of telomere phenotypes during prophase. These include a failure to attach 13-20% of telomeres to the NE ([20, 31] and our observations), and structural abnormalities that affect a subset of telomeres in each individual meiocyte. Such structural abnormalities include shortening of telomeres, detachment of telomeres from the meiotic chromosome axes, telomere fusions, and the formation of stretched out telomeres [20, 31]. Importantly, failure of telomeres in NE attachment is correlated with the absence of CCDC79 from telomeres (this work) and with shortened telomere length , and is not correlated with other structural abnormalities of telomeres in Smc1b-/- meiocytes . Thus, our experiments identify CCDC79 as a protein that may contribute to both the stabilization of telomeres during meiosis and the formation of a “bridge” that tethers telomeres to SUN1 and the inner NE.
Consistent with these findings, an independently initiated study recently revealed key roles for CCDC79 (TERB1) in the recruitment of telomeres to the NE and in the recruitment of cohesins to telomeres to maintain the structural integrity of telomeres during meiosis . Given these CCDC79 functions, our observation that efficient CCDC79 recruitment to telomeres requires the meiotic cohesin SMC1B indicates synergy and mutual dependency between CCDC79 and meiotic cohesion functions at telomeres. At this point it remains unresolved if SMC1B and/or other cohesins have a direct role in CCDC79 recruitment to telomeres or if this role is more indirect and is exerted via stabilisation of telomere length and structure. Future experiments are required to address this question, as well as to dissect the functional interaction between meiotic cohesins, CCDC79 and telomeres, and also to provide a mechanistic understanding of the behaviour and functions of meiotic telomeres in mammals.
Our results suggest that the meiosis-specific protein CCDC79 and its functional interaction with meiotic cohesins are involved in the meiosis-specific behaviour of telomeres. Investigation of the molecular functions of CCDC79 will provide an opportunity to explore, at the molecular level, meiosis-specific aspects of mammalian telomere biology that have been inaccessible, despite their apparent importance in ensuring a high fidelity of meiotic recombination and the integrity of the genome in the germline.
For expression analysis wild type (wt) tissue was isolated from C57BL/6JOlaHsd mice. To stage embryonic development, the day of detection of a vaginal plug was marked as 0.5 days post coitum (dpc). Analysed null mutant mice strains (Smc1β -/-, Sun1 -/-), have been described previously [31, 32]. Experimental animals were compared with controls from the same litter (where possible) or from other litters of the same mating. All animals were bred and maintained under pathogen-free conditions according to animal welfare regulations provided by the animal ethical committee of the Technische Universität Dresden.
Anti-CCDC79 antibodies were raised against the C-terminal 103aa of CCDC79 (ENSEMBL Accession ENSMUSP00000067324). The corresponding cDNA fragment of Ccdc79 was sub cloned into the Escherichia coli expression vector pDEST17 (Cat#11803012, Invitrogen) and the protein was expressed in fusion with N-terminal 6xHis-tag and purified on Ni-Sepharose (Cat#17-5318-01, Amersham, GE Healthcare). Immunisation experiments were carried out at Harlan Laboratories immunisation department, Hillcrest UK. Two Hartley guinea pigs were treated using a modified version of the standard Harlan three-month guinea pig immunisation protocol. This included injection of 175 μg denatured recombinant protein in 100 μl of wet polyacrylamide gel slices at day 0, 14, 21, 49 and 77. At day 0 a small fraction of preimmune blood was taken as a control. Sera containing polyclonal antibodies against the injected protein were taken at day 84 (production bleed) and day 91 (final bleed). Polyclonal antibodies were affinity purified on antigen coupled Sepharose Beads (Cat# 17-0906-01, Amersham, GE Healthcare).
Total RNA was isolated from fresh adult mouse testis tissue or frozen embryonic gonads using the RNeasy Mini Kit (Qiagen). Mouse total RNA samples from different mouse somatic tissues were purchased via Ambion (Cat#7800) and Zyagen (Cat#MR-010). One or half a microgram of total RNA was reverse transcribed using Superscript III (Cat#18080-044, Invitrogen) and oligo dT (20) primers. In no-RT controls the reaction mixture contained water instead of reverse transcriptase. RT-PCR was performed using the Ccdc79- specific primers 5′-TGTGGTCTTTCCCCTTTCAG-3′ and 5′-AGGACCGAATCTCCTCCAGT-3′ and primers for the control genes Sycp3, Mvh, Xis t, S9, whose sequences were described previously . The cycling conditions were: 94°C 3 min; 94°C 30 s, 54°C 30 s, and 72°C 25 s for 30 cycles; and 72°C 7 min. The 17 mouse somatic tissues used in gene expression profiling (Figure 1A) were: liver, brain, thymus, heart, lung, spleen and kidney (acquired via Ambion) and mammary gland, pancreas, placenta, salivary gland, skeletal muscle, skin, small intestine, spinal cord, tongue and uterus (acquired via Zyagen). Fluorescence-activated cell sorting of cell populations from whole embryonic female genital ridges for gene expression analysis, RNA-isolation from these samples and associated RT-PCRs were performed as previously described .
Nuclear surface spreads of spermatocytes and oocytes were prepared as described previously  with slight modifications. In brief, dense cell suspensions were prepared in PBS by maceration and vigorous pipetting of the gonads. Cell suspensions were diluted 20× in 100 mM sucrose in 5 mM sodium borate buffer (pH 8.5). S-fix fixative (1% paraformaldehyde, 10 mM sodium borate buffer pH 9.2, 0.15% Triton X-100) was placed on glass slides and cell suspension was placed in small droplets on the surface of the fixative. Samples were incubated for two hours at room temperature in a humid chamber. Following fast drying under a hood, the slides were washed two times for one minute with 0.4% Agepon (AgfaPhoto) and another three times for one minute with water. Slides were used immediately or kept at 4°C in PBS pH 7.4 until IF staining. Nuclear surface spreads of Sun1-/- spermatocytes were prepared according to the previously published protocol .
Before immunostaining surface spreads, samples were blocked with blocking buffer (2% BSA (Cat# A2153, Sigma), in PBS pH 7.4) for 30 min. Primary antibodies diluted in blocking buffer were applied to samples for three hours or overnight at 37°C in a humid chamber. Slides were washed three times with PBS and incubated with secondary antibodies for 1 h, and finally mounted in Vectashield mounting medium with DAPI (Cat#H-1200, Linaris).
Primary antibodies used in this study were as follows: guinea pig anti-CCDC79_1 (1:200) and guinea pig anti-CCDC79_2 (1:200), monoclonal mouse anti-SYCP3 II52F10 (1:2, a gift from R. Jessberger) , rat anti-SYCP3 (Y. Watanabe), rabbit anti-SUN1 (1:1000, Cat#ab74758, Abcam), rabbit anti-TRF1 (Y. Watanabe), goat anti-TRF1 E-15 (1:500, Cat#sc-5475, Santa Cruz). Donkey secondary antibodies conjugated with DyLight405 and DyLight649 (Jackson ImmunoResearch Europe Ltd.) were used at a dilution of 1:300, donkey secondary antibodies conjugated with either Alexa Fluor 488 or 568 (Molecular Probes/Invitrogen) were used at a dilution of 1:600. Fluorescence was visualised with Zeiss Axiophot fluorescence microscope.
Staining patterns were assessed manually in 100 or more nuclear spreads, including at least 30 nuclei of each particular meiotic sub stage. Cells were initially identified and characterized based on the localization pattern of the axial element component SYCP3, before the appropriate additional wavelengths were captured to determine the pattern of the co-stained proteins. Two independent sets of nuclear spreads were examined from each mutant. Except for the Sun1-/-m utant, we examined nuclear spreads from at least three different animals. Images displayed in the figures are representative of the most predominant staining patterns. Presented images were processed for background correction and false coloured in overlays using Adobe Photoshop CS5. In nuclear surface spreads of spermatocytes and oocytes, where most of the detergent soluble cell material is removed, both antibodies raised against CCDC79 showed similar staining patterns.
To assess protein behaviour, foci specific to the proteins of interest were counted and compared in randomly selected nuclei of wt and mutant meiocytes of the indicated stage. Mutant and wt spreads were stained simultaneously using the same antibody mixes. Imaging of the cells for each experiment was carried out on the same day using the same microscope and camera settings. SYCP3/axis staining was used to select spermatocytes or oocytes of distinct stages. Telomeres were identified by TRF1-staining.
Myb-like domains of Mus musculus CCDC79 and Mus musculus c-MYB were determined via annotation according to the UniProt Protein knowledgebase (http://www.uniprot.org). Accession numbers of the proteins and compared amino acid ranges of the full length proteins are indicated in Figure 2. The Telobox consensus sequence was analysed as described previously . Protein sequence alignments were performed by using ClustalOmega.
Days post coitum
We are grateful to Anett Hientzsch for general lab support and Rolf Jessberger for sharing ideas, antibodies and the Smc1β -/-m ouse. We are also grateful to Michelle Stevense for revising and proofreading the manuscript. Katrin Daniel was supported by MeDDriveStart grant 60.303 (from TU Dresden) and DFG grants:TO 421/4-2 and 3–2, Daniel Tränkner was supported by DFG grants: TO 421/3-2, Attila Toth was supported by DFG grant: TO 421/5-1. Hiroaki Shibuya and Yoshinori Watanabe were supported by MEXT/JSPS KAKENHI.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.