A human homologue of mouse Mater, a maternal ... - Semantic Scholar

Download PDF
11 downloads 0 Views 919KB Size Report
Comment
Sep 26, 2001 - RNA and Dr Robert Stillman for assistance in obtaining immature human oocytes. The NIH has filed a patent application concerning the subject ...
Human Reproduction Vol.17, No.4 pp. 903–911, 2002

A human homologue of mouse Mater, a maternal effect gene essential for early embryonic development Zhi-Bin Tong1, Carolyn A.Bondy, Jian Zhou and Lawrence M.Nelson Developmental Endocrinology Branch, NICHD, National Institutes of Health, Bethesda, MD 20892, USA 1To

whom correspondence should be addressed at: Building 10, Room 10N262, Developmental Endocrinology Branch, NICHD, NIH, 10 Center Drive, Bethesda, MD 20892-1862, USA. E-mail: [email protected]

BACKGROUND: Mater is a maternal effect gene required for early embryonic development in mice, and its protein serves as an autoantigen in a mouse model of autoimmune premature ovarian failure. METHODS: Human MATER cDNA was cloned by PCR techniques. The mRNA and protein were determined using hybridization and immunodetection respectively. The cDNA and protein sequences were analysed using bioinformatics software. RESULTS: Human MATER gene spans a ~63 kbp DNA at chromosome 19 and is composed of 15 exons and 14 introns. Expression of its mRNA (~4.2 kb) is restricted to the oocytes. Human MATER cDNA (3885 nt) shows an open reading frame (3600 nt) encoding a polypeptide chain composed of 1200 residues with a predicted molecular mass of 134 236 Da. MATER protein (~134 kDa) was detected in human oocytes. The human and mouse cDNA share 67% homology while their deduced polypeptide chains have 53% identity of amino acids. Also, their protein structures have a number of similar features. CONCLUSIONS: The human MATER and mouse Mater genes and proteins are conserved. Characterization of the human MATER and its protein provides a basis for investigating their clinical implications in autoimmune premature ovarian failure and infertility in women. Key words: embryo/infertility/MATER/oocyte/premature ovarian failure

Introduction MATER is a maternal oocyte protein essential for normal early embryonic development beyond the 2-cell stage in mice (Tong et al., 2000a). MATER was initially identified as an oocytespecific autoantigen associated with autoimmune oophoritis in mice (Tong and Nelson, 1999). The primary structure of mouse MATER, deduced from a cloned cDNA, is composed of 1111 amino acids with a molecular mass of 125 kDa. At the genomic level, Mater spans a ~32 kb DNA at the proximal end of the mouse chromosome 7 and contains 15 exons (Tong et al., 2000b). Mater is transcribed specifically in oocytes as oogenesis proceeds and the transcripts are degraded upon oocyte maturation and ovulation. As a maternal product, MATER accumulates during oogenesis and persists during embryogenesis. We have shown that Mater null female mice are sterile, yet they exhibit normal oogenesis, ovarian development, oocyte maturation, ovulation and fertilization (Tong et al., 2000a). The null males have normal fertility. Furthermore, development of the embryos from Mater null females exhibits arrest at the 2-cell stage, a critical stage for embryonic genome activation in mice. Thus, Mater is a maternal effect gene essential for early embryonic development in mice. It is known that maternal factors govern early embryonic development before embryonic genome activation takes place (Latham, 1999; Latham and Schultz, 2001) The maternal effect on early embryogenesis of Drosophila and Xenopus has been © European Society of Human Reproduction and Embryology

demonstrated by the identification of a number of maternal effect genes (Akam, 1987; Droin, 1992). However, the presence and importance of maternal effect genes in early mammalian development has been only inferred until the recent identification of maternal effect genes in mice, such as Mater (Tong et al., 2000a) and Hsf1 (Christians et al., 2000). In addition, a maternal effect in early mammalian embryogenesis has been demonstrated recently for some well-known proteins such as zona pellucida proteins (Rankin et al., 2000, 2001) and E-cadherin (Larue et al., 1994), although their effects are not restricted to embryonic development. To date, little is known about the mechanisms by which maternal products direct early development in mammals. In mice, oocytes are transcriptionally active during oogenesis. As oocytes undergo growth and maturation, maternal proteins accumulate within the oocytes (Schultz, 1993). After fertilization, activation of the embryonic genome is an essential event in early development (Flach et al., 1982; Conover et al., 1991; Bellier et al., 1997). In mice the zygotic genome activation begins in the late stage of 1-cell zygotes and becomes dominant at the 2-cell stage of embryos (Latham et al., 1991a,b; Bouniol et al., 1995; Nothias et al., 1995; Aoki et al., 1997). Metabolic inhibition of embryonic transcription does not block the first embryonic cleavage, but arrests development at the 2-cell stage in mice (Warner and Versteegh, 1974; Flach et al., 1982). This suggests that maternal products direct development before the embryo 903

Z.-B.Tong et al.

undergoes genome activation, while the products derived from embryonic genome are required for embryonic progression beyond the 2-cell stage. Although the molecular basis for this programmatic transition is poorly understood, clearly the maternal products are required to constitute appropriate conditions for activation of the embryonic genome. In this study, we characterize human MATER gene and its protein. MATER serves as an autoantigen in mouse autoimmune oophoritis (Tong and Nelson, 1999), a model for human autoimmune premature ovarian failure (Kalantaridou and Nelson, 1998). We are on a path to determine whether human MATER plays an antigenic role in the autoimmune pathogenesis of clinical premature ovarian failure. In addition, MATER as a maternal factor supports early embryonic development in mice, through which it determines female fertility (Tong et al., 2000a). This raises the possibility that a MATER gene mutation or MATER protein deficiency might cause infertility in some women with normal ovulatory function. Thus, characterization of human MATER gene and its protein provides a new determinant with which to investigate ovarian autoimmunity and unexplained infertility in women.

Materials and methods Cloning of the human MATER cDNA A database of the human htgs (High Through Genomic Sequence) was blasted with the mouse MATER cDNA sequence to search for human homologous sequences (http://www.ncbi.nlm.nih.gov). Based on DNA sequences of human chromosome 19 homologous to the mouse MATER cDNA, we designed oligonucleotides as primers for PCR reactions to amplify cDNA from a human ovarian cDNA library (Clontech, Inc., Palo Alto, CA, USA) and human ovarian cDNA (Invitrogen, Inc., Carlsbad, CA, USA; Origene Inc., Rockville, MD, USA). The PCR-cloned human cDNA was labelled with 32P as a probe to screen the human ovarian cDNA library, as reported previously (Tong and Nelson, 1999). Determination of human MATER locus and exon-intron map Using the sequence of cloned human MATER cDNA, one human genomic clone (GenBank accession: AC011470) was determined to carry an entire gene locus of human MATER. Nearly full-length human MATER cDNA sequence was blasted to the DNA sequence of this human genomic clone to determine the size and numbers of exons and introns. The AG–exon–GT rule was applied to determine a splicing site between exon and intron (Breathnach and Chambon, 1981). Northern and Southern hybridizations Human ovarian total RNA from a 19 year old woman was obtained under an IRB-approved protocol (McGill University, Montreal, Canada) and our use of it was approved by the NIH Office of Human Subjects Research. Ovarian RNA from a 26 year old woman was purchased from BioChain Inc. (Hayward, CA, USA). Separation of human ovarian total RNA (10 µg) was performed by 1% agarose/ formaldehyde gel electrophoresis. The RNA was transferred onto a nitrocellulose membrane to prepare a Northern blot. The PCR products were separated by 2% agarose gel electrophoresis and transferred onto a nylon membrane to prepare a Southern blot. Human MATER cDNA (~0.5 kbp) was labelled with α-[32P]dCTP (3000 Ci/mmol; ICN, Costa Mesa, CA, USA) using Ready-To-Go DNA dCTP Beads

904

(Pharmacia, Inc., Piscataway, NJ, USA). Both RNA and DNA blots were hybridized (68°C, 2 h) with 32P-labelled cDNA probe (~0.5 kbp) in QuikHyb solution (Stratagene, La Jolla, CA, USA). After washing the blots with a solution of 0.1⫻standard saline citrate/0.1% sodium dodecyl sulphate (SDS) at 65°C, the hybridization signals were detected by autoradiography. PCR Human ovarian cDNA (Invitrogen; Origene), human ovarian cDNA library (Clontech) and the Human Rapid-Scan Gene Expression Panel (Origene) were used as templates for PCR reactions per manufacturers’ instructions. Taq DNA polymerase (Life Technologies, Rockville, MD, USA) was used to amplify DNA products expected within 1.5 kbp at a condition of the PCR reaction (95°C, 5 min; 35 cycles of 95°C, 1 min; 60°C, 1 min; 72°C, 2 min; and 72°C, 10 min for further extention). Platinum PCR SuperMix (Life Technologies) was used for amplification of DNA products (2–5 kbp) with prolonging extension time at 72°C in every cycle, according to the product’s manual. The PCR products were electrophoresed on agarose gel containing 0.1% ethidium bromide and subcloned into a TA cloning vector (Invitrogen) for DNA sequencing. In-situ hybridizations Both [35S]UTP-labelled sense and antisense probes were synthesized by in-vitro transcription using human MATER cDNA as templates and T3/T7 RNA polymerase (Ambion, Inc., Austin, TX, USA). Short probes (~300 nt) of alkaline hydrolysis were hybridized at 60°C for 24 h with frozen sections (8 µm) of human ovary obtained from the National Disease Research Institute (NDRI; Philadelphia, PA, USA). After dipping in Kodak NTB-2 emulsion, the slides were exposed for 1 week on an X-ray film for developing in Kodak D-19 and Kodak Fixer. The slides were finally stained with haematoxylin and eosin, and examined by microscopy under both dark and light fields. Immunohistochemistry and immunoblotting After blocking with 10% normal goat sera in phosphate-buffered saline (PBS), frozen sections (8 µm) of human ovary (NDRI) were incubated at 25°C for 2 h with rabbit antisera (1:200) specific to mouse MATER peptide. The slides were washed with 0.05% Tween-20 in PBS, and incubated with a goat anti-rabbit IgG antibody conjugated with fluorescein isothiocyanate (1:300; Sigma Chemical Co., St Louis, MO, USA). After washing and applying with glycerol– glass cover, the slides were examined under fluorescent microscopy. Human immature oocytes were supplied for this study by a clinical IVF programme (Invitrogen; Origene) after obtaining consent and approval of the NIH Office of Human Subjects Research. For immunoblotting, human oocyte proteins were separated on 3–8% Tris–acetate SDS–polyacrylamide gel electrophoresis. After transferring onto a nitrocellulose membrane, 5% dry milk in PBS was used to block non-specific binding in the blot. Subsequently, the blot was incubated with rabbit anti-MATER peptide antibody (1:1000) for 2 h at 25°C. After the blot was washed, the primary antibody-bound proteins were detected by a goat anti-rabbit IgG antibody (1:1000) labelled with horse-radish peroxidase using an enhanced chemiluminescence system according to the manufacturer’s instructions (Amersham/Pharmacia, Inc., Piscataway, NJ, USA). DNA sequencing and computational analysis DNA was sequenced with a dRhodamine terminator cycle sequencing kit (PE Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. Sequencing reactions were conducted by PCR at 25 cycles of 95°C, 10 s; 50°C, 5 s; and 60°C, 4 min. Electrophoresis with 5% Long Ranger gels and sequence analysis

Human MATER

Figure 1. Human MATER mRNA and its oocyte-specific expression. (A) Northern hybridization: total human ovarian RNA were prepared from women at age 19 (lane 1) and 26 (lane 2) years. 10 µg of each RNA were separated by gel electrophoresis and transferred onto a nitrocellulose membrane. The blot was probed with 32P-labelled human MATER cDNA. RNA molecular sizes (kb) are indicated on the left of the panel. (B and C) RT–PCR: using the Human Rapid-Scan Gene Expression Panel (Origene Inc.) and specific primers, cDNA at ⫻100 concentration was used to amplify human β-actin (B), while cDNA at ⫻1000 concentration was used to amplify human MATER (C). The cDNA was synthesized from the RNA of the human brain (lane 1), heart (2), kidney (3), spleen (4), liver (5), colon (6), lung (7), small intestine (8), muscle (9), stomach (10), testis (11), placenta (12), salivary gland (13), thyroid (14), adrenal gland (15), pancreas (16), ovary (17), uterus (18), prostate (19), skin (20), blood leukocytes (21), bone marrow (22), fetal brain (23), fetal liver (24). Lanes 25 and 26 serve as a positive control (human ovarian cDNA; Invitrogen) and a negative control (without cDNA) respectively. PCR products were separated by 2% agarose gel electrophoresis and visualized by 0.1% ethidium bromide. Amplification of the PCR products of human β-actin (640 bp) and human MATER (280 bp) is indicated on the right. (D) Southern hybridization: a blot was prepared from the agarose gel separating the human MATER PCR product in (C), and probed with 32P-labelled human MATER cDNA. Hybridization signals were detected by autoradiography. Arrow indicates the position of PCR product hybridization of human MATER (280 bp) on the right. (E and F) In-situ hybridization: 35S-labelled antisense and sense probes were synthesized by in-vitro transcription using the human MATER cDNA as templates. The frozen human ovarian sections were hybridized with the antisense probe (E) and sense probe (F). Dark-field images are displayed, and location of the oocytes is indicated by an arrow. Scale bar ⫽ 100 µm. were carried out on the ABI Prism 377 DNA Sequencer. Computational analyses of DNA sequences and deduced peptide sequences were conducted using software available at the National Center for Biotechnology Information at NIH (http://www.ncbi.nlm.nih.gov), the EMBL-EBI (http://www.ensembl.org), the Sanger Center (http://www.sanger.ac.uk) and the ExPASy Molecular Biology Server of the Swiss Institute for Bioinformatics (http://www.expasy.ch).

Results Oocyte-specific expression of human MATER transcripts and protein Using human MATER cDNA as a probe, a Northern blot hybridization revealed the presence of a single ~4.2 kb MATER transcript in the human ovarian total RNA (Figure 1A). The human MATER transcripts appeared to be larger than the mouse MATER transcript (3.75 kb). The relatively greater abundance of MATER mRNA detected in the ovarian RNA from a 19 year old woman (lane 1), compared with the RNA from 26 year old woman (lane 2), may reflect the relatively greater number of oocytes present in ovaries from younger women (Adashi, 1999).

To determine whether expression of human MATER transcripts is ovarian-specific, we performed PCR reactions using cDNA that were derived from human RNA of 24 different tissues (Human Rapid-Scan Gene Expression Panel). As a control, shown in Figure 1B, the PCR products (640 bp) of human β-actin were amplified in all cDNA derived from 24 tissues and human ovary (lanes 1–25) using the human β-actin-specific primers (5⬘-GCATGGGTCAGAAGGAT-3⬘/ 3⬘-GTCCAGTAGTGGTAACC-5⬘). In contrast, using human MATER-specific primers (5⬘-TGACTTCTGATTGCTGTGAGG-3⬘/3⬘-CTACTGGCCATGACCACCTT-5⬘), the PCR products (280 bp) were only generated in the cDNA synthesized from the ovarian RNA (Figure 1C, lanes 17 and 25). Ovarianspecific expression of the human MATER transcripts was further evidenced by a Southern blot hybridization of the human MATER primer-amplified PCR reactions probed with radiolabelled MATER cDNA (Figure 1D). Furthermore, in-situ hybridization of human ovarian sections revealed the oocyte-specific expression of human MATER transcripts. As shown in Figure 1E, the hybridization signals using synthetic human MATER antisense probe were present within the 905

Z.-B.Tong et al.

Figure 2. Human MATER protein expression in the oocytes. (A and B) Immunohistofluorescence: frozen human ovarian sections were incubated with rabbit antisera (1:200) against the mouse MATER peptide. Fluorescein isothiocyanate-conjugated goat anti-rabbit IgG antibody was used as a second antibody to detect the primary antibody bound MATER protein. Tissue sections were visualized by fluorescence microscopy (A). Corresponding phase contrast is shown in (B). Arrows indicate the location of oocytes. Scale bar ⫽ 100 µm. (C) Immunoblotting: solubilized proteins of human oocytes (lane 1, 10 oocytes; lane 2, 15 oocytes) were separated by 3–8% Tris–acetate SDS–PAGE electrophoresis. A protein blot was incubated with rabbit anti-mouse MATER peptide antisera (1:1000). The enhanced chemiluminescence kit with horse-radish peroxidase-conjugated goat anti-rabbit IgG antibody was used to detect the blot-bound primary antibodies. Protein molecular masses (kDa) are indicated on the left. Human MATER protein is indicated by an arrow at ~134 kDa.

oocytes but not in the follicular cells. The control MATER sense probes did not hybridize with oocytes or other ovarian cells (Figure 1F). Thus, this oocyte-restricted expression of the human MATER gene appeared similar to the expression pattern of mouse Mater gene (Tong and Nelson, 1999). Expression of human MATER protein in oocytes was demonstrated by indirect immunofluorescence. Using a primary antibody of rabbit anti-mouse MATER peptide, the oocytes present in the human ovarian sections were visualized by FITC-conjugated second antibody (Figure 2A). No fluorescense was observed in the other somatic cells within ovarian sections. Normal rabbit sera did not stain the oocytes (data not shown). To determine the size of native MATER protein, human immature oocytes were used for immunoblotting. As shown in Figure 2C, human MATER protein (~134 kDa) was detected in the preparation of the human oocyte proteins using antimouse MATER antibody. It was noted that there were two close signals of MATER proteins present in the immunoblotting. This might reflect a post-translational modification of human MATER protein or a possibility of an alternative initiation site in the synthesis of the polypeptide chain. Isolation and characterization of human MATER cDNA The mouse MATER cDNA sequence was blasted to the htgs database to search for human DNA sequences homologous to the mouse MATER cDNA. A serial of discontinuous and short 906

DNA sequences homologous with the mouse MATER cDNA was initially found in the genomic clones (AC012107, AC024580, AC019238 and AC023887) of human chromosome 19. This finding held promise for locating the human MATER, since the human chromosome 19q13 is a syntenic region of the mouse Mater residing at the proximal end of the mouse chromosome 7 (Tong et al., 2000b). We designed human oligonucleotide primers and amplified the PCR products from human ovarian cDNA and its library. Using these PCR-cloned cDNA as the probes for in-situ hybridization, the human oocytes were confirmed to express the corresponding transcripts. DNA sequencing revealed that these PCR products are human MATER cDNA homologous to the mouse MATER cDNA. The PCR products were generated to span ~98% of the coding region from nt 76 to nt 3596. The 3⬘ non-coding regions were determined by sequencing cDNA clones isolated from the human ovarian cDNA library using the PCR-cloned cDNA as probe. The first 75 nt at the 5⬘ end were derived from a predicted cDNA of the genomic clone AC011470 (http://www.ensembl.org). The non-coding region present at the 5⬘ end of the MATER cDNA remains undetermined. Assuming the presence of a poly(A) tail (~150–200 nt) and the 5⬘ undetermined non-coding regions (~100–150 nt), assembly of our determined human MATER cDNA sequence, its poly(A) tail and 5⬘ undetermined non-coding region

Human MATER

in vertebrates (Kozak, 1991). The stop code (TGA) was followed by 282 nt of the 3⬘ non-coding sequence. A signal sequence (AATAAA) for polyadenylation was located just 18 nt upstream of the poly(A) tail. None of the known genes and proteins deposited in the GenBank was found to have significant homologies to the human MATER gene or MATER protein. However, the carboxyl-terminal sequence of the human MATER peptide had an ~32% homology with human and porcine ribonuclease inhibitors (457 residues). Thus, human MATER contains a structural domain of the ribonuclease inhibitor-like leucine-rich repeats at the carboxyl-terminus (see below), as found in mouse MATER (Kajava 1998; Tong et al., 2000b). The human and mouse MATER cDNA shared 67% homology while their deduced polypeptide chains had 53% identities of amino acids.

Figure 3. Nucleotide and amino acid sequences of human MATER. The nucleotides (3885 nt) from 5’ are displayed on the top line and their numbers are indicated on the right. The initiation code (ATG) and termination code (TGA) is underlined (solid) respectively. The polyadenylation signal (AATAAA) is underlined with a dashed line. The nucleotides at the 5’ end of the non-coding region remain to be determined. The deduced amino acid sequence (1200 residues) is under the nucleotide sequence and numbered on the right. Some structural motifs present in the protein sequence are shown in Figure 5B. These sequence data are submitted to the GenBank under an accession number AY054986.

corresponded in size with the MATER transcripts detected in the Northern hybridization (4.2 kb). This suggests that the cloned cDNA represents a full length of the coding region. As shown in Figure 3, the open reading frame (nt 1 to nt 3600) of the cloned cDNA (3885 nt) encodes a protein composed of 1200 amino acids (Figure 3). The entire polypeptide chain was predicted to have a pI 6.08 and a molecular mass of 134 235.76 Da, which corresponded to the size of the human oocyte protein recognized by the antisera specific to mouse MATER peptide. Expression of a recombinant protein further confirmed the cloned cDNA to encode the human MATER protein (data not shown). The initiator code (ATG) encoding the first methionine in the open reading frame (ORF) was in a context (ATCATGAAG), the consensus sequence recognized for initiation

Analysis of human MATER polypeptide chain A computational analysis was carried out to search for the structural features of human MATER protein. When the entire polypeptide chain of human MATER was compared with itself, the Dotplot analysis revealed the presence of repeat structural motifs at amino- and carboxyl-termini. These repeats were evidenced by a series of parallel but offset lines in these regions (Figure 4A), as also seen in the mouse MATER protein (Tong et al., 2000b). A comparison between the polypeptide chains of human and mouse MATER indicated the similarities and homologies of their structural motifs and compositions (Figure 4B). A hydropathy plot (Figure 5A) indicated that there were hydrophilic regions present at amino- (32–304 residues) and carboxyl-termini (1184–1200 residues) (Kyte and Doolittle, 1982). No structural motifs were found for either peptide signals of protein secretion or transmembrane domains. The peptide chain had some notable motifs as indicated in Figure 5B. A signal for nuclear localization was present at the residue 517–534 (NLS_BP), which was not detected in the mouse MATER. Because of the absence of the peptide domains for DNA binding, further experimental evidence is required to determine if the human MATER is translocated into nuclei. Human MATER protein had six leucine-rich repeats at the carboxyl-terminus and a PAAD_DAPIN domain (52–148 residues), both of which are recognized as the structural basis for protein–protein interactions (Kobe and Deisenhofer, 1995; Pawlowski et al., 2001; Staub et al., 2001). Their presence implies that human MATER probably interacts with other oocyte proteins to perform its function. In addition, human MATER protein had an aldo_ket_red domain (71–87) at the amino-terminus and an ATP-/GTP-binding domain (P-loop) at residue 286–293. The aldo_ket_red is a domain present in protein with oxidoreductase activity (Bohren et al., 1989), and the P-loop is the typical domain for binding to ATP/GTP in participating in an intracellular signal transduction (Saraste et al., 1990). The roles of these domains in the function of human MATER protein remain to be determined. Genomic organization of human MATER locus Using the sequence of the cloned human MATER cDNA to blast the human hgts database (http://www.ensembl.org), a 907

Z.-B.Tong et al.

Figure 4. Dotplot analysis of human and mouse MATER amino acid sequences. Human MATER amino acid sequence was compared with itself (A) and the mouse MATER amino acid sequence (B). The centre diagonal line from the left to the right represents the direct correspondence of the two sequences of the same protein from amino- to carboxyl-terminus. Identities are shown by the dots. The short diagonal lines parallel to the centre diagonal line are present at the carboxyl-terminus and a region close to the amino-terminus, reflecting repeated amino acid sequences at these regions. Display of the dots and short parallel lines delineate similarities of amino acid sequences between human MATER (1200 residues) and mouse MATER (1111 residues), reflecting homologies of the composition and structure of the MATER protein of the two species.

single genomic clone (AC011740) with a full DNA sequence was determined to harbour an entire gene locus of human MATER. To determine the number and size of the exons and introns, the sequence of the cloned cDNA was compared with the sequence of the AC011740 clone isolated from human chromosome 19. The 15 exons and 14 introns were identified to span a ~63 kbp DNA. The exons ranging in size from 56 to 1597 bp were intervened by the 14 introns ranging from 760 to 9327 bp (Table I). In addition, there were 282 nucleotides of the 3⬘ non-coding region within exon 15, and undetermined 908

Figure 5. Computational analysis of structural motifs in human MATER. (A) Kyte and Doolittle hydropathy of human MATER amino acid sequence. Hydrophilic repeats at the amino-terminus (32–304 amino acids) contain a short hydrophobic region (104–136 amino acids). At the carboxyl-terminus, there is a short hydrophilic region (1184–1200 amino acids). (B) Schematic representation of structural motifs of human MATER. The horizontal line from amino- to carboxyl-terminus represents the entire peptide chain of human MATER (1200 residues). The opened boxes represent structural motifs and their positions on the peptide chain. The protein contains a domain of PAAD_DAPIN at the amino-terminus and six domains of leucine-rich repeats (LRR) at the carboxylterminus. Both molecular domains are for protein–protein interactions. In addition, there are motifs with potentials for ATP/GTP-binding (P-loop), nuclear translocation (NLS_BP) and aldo/keto reductase activity (aldo_ket_red). Locations of these motifs are indicated in the numbers following the particular motifs.

nucleotides of the 5⬘ non-coding region within exon 1. The sequences switching between each exon and intron were confirmed to meet with the bordering consensus sequences (AG µ GT) for splice sites of the exon–intron (Breathnach and Chambon, 1981). Thus, human MATER locus spanned a ~63 kbp DNA and was composed of the 15 exons and 14 introns. Human MATER and mouse Mater loci had identical numbers of exons and introns, while the human locus was longer than the mouse locus (32 kbp). This mainly reflects the larger size of the human introns. The counterpart exons were comparable. Both had a large exon 7 spanning ~50% of the coding sequences, and the counterparts of exons 8–14 had an identical number of nucleotides. However, the first six exons (1–6) in the human locus ranged in size from 56 to 380 nt while their counterparts in the mouse locus had 48–78 nt. A larger exon 2 (380 nt) in the human MATER locus made the human MATER cDNA coding region (3600 nt) slightly longer than

Human MATER

Figure 6. Schematic representations of exon–intron maps of human MATER (A) and mouse Mater (B). The exon–intron map of human MATER (~63 kbp) was compared with the mouse Mater (~32 kbp) exon–intron map (Tong et al., 2000b). Both gene loci have 15 exons that are indicated by the Arabic numbers above each of the vertical bars representing the exons. The horizontal solid lines between exons represent introns, and the horizontal dashed lines at 5’ ends represent the undetermined 5’ ends of both gene loci. Scale bar (2 kbp) is shown under the map of human MATER.

Table I. Exon and intron and splice-junction sequences of human MATER Exon

Splice-junction sequences

No.

Position

Length (bp)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1–62 63–442 443–508 509–565 566–623 624–679 680–2276 2277–2447 2448–2615 2616–2786 2787–2957 2958–3128 3129–3299 3300–3470 3471–3885

⬎62 380 66 57 58 56 1597 171 168 171 171 171 171 171 415

Intron Exon

– TCTTTTAAAG TGTCTTCCAG TCTTTTGCAG TCTTTTGCAG TCTTTTGCAG CTGACTCCAG CCTGACATAG CCCATTGCAG GTCTCTGCAG CTTCTTGCAG TGCCCCACAG CCTCTTGCAG GACCCTGCAG GCCTCCCCAG

the mouse cDNA coding region (3333 nt). All of these structural similarities and variations reflect a genetic conservation and evolution between mouse Mater and human MATER loci. Discussion Human MATER and mouse Mater genes and proteins are conserved. Their cDNA share 67% homology in the nucleotide sequence and their deduced polypeptide chains have 53% identities of amino acids. The human and mouse MATER proteins share a number of similar features, such as leucinerich repeat at the carboxyl-terminus and an ATP-/GTP-binding domain (P-loop). These structural similarities between species suggest that MATER may have a similar function in the mouse and human. In mice, we have demonstrated that MATER plays a critical role in early development and that its absence causes female infertility (Tong et al., 2000a), and also that MATER serves as an oocyte-specific autoantigen associated with ovarian autoimmunity (Tong and Nelson, 1999). Thus, MATER may

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

GTCGACAGCC GTAAGCGAGA GTTAGAGGGG GTGAATGAAA GTGTGGAAAA GTGAGGAAAA GTGAGTACCC GTAAGGCTGC GTACGTCTCT GTGAGTGCCA GTGAGTCTGG GTGAGTTTCC GTAACTTCCT GTAAGTCGCC GTTGTCAAGT

No.

Length (bp)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

3928 4692 6982 3543 976 6487 4118 760 4313 2726 9327 3036 4606 2988

have potential clinical implications in autoimmune premature ovarian failure and infertility in women. Characterization of human MATER and its protein provides a basis for investigating these women’s diseases. The requirement of MATER for early embryonic development in mice is an example of how important the role of maternal products can be in the developmental programme of mammals (Tong et al., 2000a). The role of MATER appears to be specific to very early development. In mice, Mater null mutation does not affect oogenesis, folliculogenesis, oocyte maturation and ovulation, or fertilization. Phenotypic infertility in the null female mice lacking MATER is attributable to an arrested development at the 2-cell stage. The arrested embryos demonstrate a considerable decrease of embryonic gene transcription. The resultant 2-cell block appears similar to observations of mouse embryos treated with α-amanitin, in which the metabolic blockage of embryonic transcription results in an arrest of embryonic development specific to the 2-cell stage (Warner and Versteegh, 1974; Flach et al., 1982). MATER 909

Z.-B.Tong et al.

consistently remains within the embryonic cytoplasm during mouse development to the blastcyst stage. The exact function of MATER remains to be determined. The presence of an ATP-/GTP-binding domain in both human and mouse MATER proteins suggests their participation in an intracellular signal transduction (Saraste et al., 1990). Our preliminary data indicate that MATER may be truly involved in cellular signaling (Z.-B.Tong et al., unpublished observations). In addition to the carboxyl-terminal leucine-rich repeat domain similar to the mouse MATER (Tong et al., 2000b), human MATER has an amino-terminal PAAD_DAPIN domain (Pawlowski et al., 2001; Staub et al., 2001). Both domains are recognized as structural motifs for mediating protein–protein interactions (Kobe and Deisenhofer 1995; Pawlowski et al., 2001), raising a possibility that MATER may interact with other protein(s) within the embryos. Different from the mouse MATER, human MATER protein has an amino-terminal aldo_ket_red domain, a motif of proteins with oxidoreductase activity (Bohren et al., 1989), and a signal for nuclear localization. Further experiments are required to determine the significance of these structural motifs. Although little is known about maternal embryonic effect in humans, increasing evidence indicates that maternal factors present in human oocytes perform a critical role in biological events following fertilization (Barritt et al., 2001). Among infertile women seeking reproductive assistance in clinical IVF programmes, not all women reach a desirable outcome since the in-vitro fertilized oocytes of some women fail to undergo normal early embryonic development (Alikani et al., 1999; Kovacs et al., 2001). It is a possibility that the maternal factor(s) necessary for supporting early development is defective in the fertilized ova of some infertile women. Indeed, recent reports suggest that injection of normal oocyte proteins into the oocytes of some infertile women might restore normal early development and successful pregnancy (Cohen et al., 1997; Barritt et al., 2001). However, specific factors rescuing early development have not been identified. Mater null female mice are infertile because the absence of MATER protein in the fertilized ova leads to failure in embryonic development. We propose that a homozygous human MATER mutation or its protein deficiency may cause infertility in some women. To test this hypothesis, we plan to examine infertile women with a history of successful fertilization but yet failure in embryonic development for gene mutations of human MATER. If such a deficiency is found, normal embryonic development might be induced by injection of recombinant human MATER protein into the oocytes before fertilization. This might provide IVF programmes with a novel therapy to treat infertile women who have a homozygous MATER mutation and the resulting oocyte protein deficiency. Cloning and characterization of human MATER might also lead to insights regarding the pathogenesis of human autoimmune premature ovarian failure. This ovarian disease is a clinical syndrome characterized by amenorrhoea, infertility and menopausal symptoms attributable to hypoestrogenaemia and hypergonadotrophinaemia in women before age 40 (Kalantaridou and Nelson, 2000). However, specific ovarian antigenic targets involved in autoimmune premature ovarian 910

failure have yet to be identified (Hoek et al., 1997). There are striking similarities between human autoimmune premature ovarian failure and the autoimmune oophoritis in an animal model, which is generated by neonatal thymectomy in certain strains of mice (Taguchi et al., 1980; Taguchi and Nishizuka, 1980; Kalantaridou and Nelson, 2000). MATER was first identified as an oocyte-specific antigen associated with autoimmune oophoritis in this thymectomy mouse model (Tong and Nelson, 1999). A recent report has shown that the endogenous oocyte antigen may be essential for development of auotimmune ovarian disease in this animal model (Alard et al., 2001). We are developing an assay using human recombinant MATER protein to determine whether patients with premature ovarian failure have circulating anti-MATER antibodies. If the outcome is positive, such an assay would help physicians classify patients with premature ovarian failure, and thus provide a basis for developing therapy for the disease.

Acknowledgements We thank Dr David Wheeler for initial instruction of the human htgs database, Dr Jurrien Dean and Dr Jeffrey Trent for early discussions and Dr Jie Wang for technical assistance with in-situ hybridization. We are grateful to Dr Roger Gosden for providing human ovarian RNA and Dr Robert Stillman for assistance in obtaining immature human oocytes. The NIH has filed a patent application concerning the subject of this manuscript.

References Adashi, E.Y. (1999) The ovarian life cycle. In Yen, S.S.C. et al. (eds), Reproductive Endocrinology, 4th edn, W.B.Saunders, Philadelphia, pp. 153–190. Akam, M. (1987) The molecular basis for metameric pattern in the Drosophila embryo. Development, 101, 1–22. Alard, P., Thompson, C., Agersborg, S.S., Thatte, J., Setiady, Y., Samy, E. and Tung, K.S. (2001) Endogenous oocyte antigens are required for rapid induction and progression of autoimmune ovarian disease following day-3 thymectomy. J. Immunol., 166, 4363–4369. Alikani, M., Cohen, J., Tomkin, G., Garrisi, G.J., Mack, C. and Scott, R.T. (1999) Human embryo fragmentation in vitro and its implications for pregnancy and implantation. Fertil. Steril., 71, 836–842. Aoki, F., Worrad, D.M. and Schultz, R.M. (1997) Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev. Biol., 181, 296–307. Barritt, J., Willadsen, S., Brenner, C. and Cohen, J. (2001) Cytoplasmic transfer in assisted reproduction. Hum. Reprod. Update, 7, 428–435. Bellier, S., Chastant, S., Adenot, P., Vincent, M., Renard, J.P. and Bensaude, Q. (1997) Nuclear translocation and carboxyl-terminal domain phosphorylation of RNA polymerase delineate the two phases of zygotic gene activation in mammalian embryos. EMBO J., 16, 6250–6262. Bohren, K.M., Bullock, B., Wermuth, B. and Gabbay, K.H. (1989) The aldoketo reductase superfamily. cDNAs and deduced amino acid sequences of human aldehyde and aldose reductases. J. Biol. Chem., 264, 9547–9551. Bouniol, C., Nguyen, E. and Debey, P. (1995) Endogenous transcription occurs at the 1-cell stage in the mouse embryo. Exp. Cell Res., 218, 57–62. Breathnach, R. and Chambon, P. (1981) Organization and expression of eucaryotic split genes coding for proteins. Annu. Rev. Biochem., 50, 349–383. Christians, E., Davis, A.A., Thomas, S.D. and Benjamin, I.J. (2000) Maternal effect of Hsf1 on reproductive success. Nature, 407, 693–694. Cohen, J., Scott, R., Schimmel, T., Levron, J. and Willadsen, S. (1997) Birth of infant after transfer of anucleate donor oocyte cytoplasm into recipient eggs. Lancet, 350, 186–187. Conover, J.C., Temeles, G.L., Zimmermann, J.W., Burke, B. and Schultz, R.M. (1991) Stage-specific expression of a family of proteins that are major products of zygotic gene activation in the mouse embryo. Dev. Biol., 144, 392–404.

Human MATER Droin, A. (1992) The developmental mutants of Xenopus. Int. J. Dev. Biol., 36, 455–464. Flach, G., Johnson, M.H., Braude, P.R., Taylor, R.A. and Bolton, V.N. (1982) The transition from maternal to embryonic control in the 2-cell mouse embryo. EMBO J., 1, 681–686. Hoek, A., Schoemaker, J. and Drexhage, H.A. (1997) Premature ovarian failure and ovarian autoimmunity. Endocr. Rev., 18, 107–134. Kalantaridou, S.N. and Nelson, L.M. (1998) Autoimmune premature ovarian failure: of mice and women. J. Am. Women Assoc., 53, 10–20. Kalantaridou, S.N. and Nelson, L.M. (2000) Premature ovarian failure is not premature menopause. Ann. NY Acad. Sci., 900, 393–402. Kajava, A.V. (1998) Structural diversity of leucine-rich repeat proteins. J. Mol. Biol., 277, 519–527. Kobe, B. and Deisenhofer, J. (1995) Proteins with leucine-rich repeats. Curr. Opin. Struct. Biol., 5, 409–416. Kovacs, G.T., Maclachlan, V. and Brehny, S. (2001) What is the probability of conception for couple entering an IVF program? Aust. NZ J. Obstet. Gynaecol., 41, 207–209. Kozak, M. (1991) Structural features in eukaryotic mRNAs that modulate the initiation of translation. J. Biol. Chem., 266, 19876–19870. Kyte, J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol., 157, 105–132. Larue, L., Ohsugi, M., Hirchenhain, J. and Kemler, R. (1994) E-cadherin null mutant embryos fail to form a trophectoderm epithelium. Proc. Natl Acad. Sci. USA, 91, 8263–8267. Latham, K.E. (1999) Mechanisms and control of embryonic genome activation in mammalian embryos. Int. Rev. Cytol., 193, 71–124. Latham, K.E. and Schultz, R.M. (2001) Embryonic genome activation. Front. Biosci., 6, D748–759. Latham, K.E., Garrels, J.I., Chang, C. and Solter, D. (1991a) Quantitative analysis of protein synthesis in mouse embryos. I. Extensive reprogramming at the one- and two-cell stages. Development, 112, 921–932. Latham, K.E., Solter, D. and Schultz, R.M. (1991b) Activation of a two-cell stage-specific gene following transfer of heterologous nuclei into enucleated mouse embryos. Mol. Reprod. Dev., 30, 182–186.

Nothias, J.Y., Majumder, S., Kaneko, K.J. and DePamphilis, M.L. (1995) Regulation of gene expression at the beginning of mammalian development. J. Biol. Chem., 270, 22077–22080. Pawlowski, K., Pio, F., Chu, Z., Reed, J.C. and Godzik, A. (2001) PAAD—a new protein domain associated with apoptosis, cancer and autoimmune diseases. Trends Biochem. Sci., 26, 85–87. Rankin, T.L., Soyal, S. and Dean, J. (2000) The mouse zona pellucida: folliculogenesis, fertility and pre-implantation development. Mol. Cell. Endocrinol., 163, 21–25. Rankin, T.L., O’Brien, M., Lee, E., Wigglesworth, K., Eppig, J. and Dean, J. (2001) Defective zonae pellucidae in Zp2-null mice disrupt folliculogenesis, fertility and development. Development, 128, 1119–1126. Saraste, M., Sibbald, P.R. and Wittinghofer, A. (1990) The P-loop—a common motif in ATP- and GTP-binding proteins. Trends Biochem. Sci., 15, 430–434. Schultz, R.M. (1993) Regulation of zygotic gene activation in the mouse. Bioessay, 8, 531–538. Staub, E., Dahl, E. and Rosenthal, A. (2001) The DAPIN family: a novel domain links apoptotic and interferon response proteins. Trends Biochem. Sci., 26, 83–85. Taguchi, O. and Nishizuka, Y. (1980) Autoimmune oophoritis in the thymectomized mice: T cell requirement in adoptive cell transfer. Clin. Exp. Immunol., 42, 324–331. Taguchi, O., Nishizuka, Y., Sakakura, T. et al. (1980) Autoimmune oophoritis in thymectomized mice: detection of circulating antibodies against oocytes. Clin. Exp. Immunol., 40, 540–553. Tong, Z.B. and Nelson, L.M. (1999) A mouse gene encoding an oocyte antigen associated with autoimmune premature ovarian failure. Endocrinology, 140, 3720–3726. Tong, Z.B., Gold, L., Pfeifer, K.E., Dorward, H., Lee, E., Bondy, C.A., Dean, J. and Nelson, L.M. (2000a) Mater, a maternal effect gene required for early development in mice. Nat. Genet., 26, 267–268. Tong, Z.B., Nelson, L.M. and Dean, J. (2000b) Mater encodes a maternal protein in mice with a leucine-rich repeat domain homologous to porcine ribonuclease inhibitor. Mamm. Genome, 11, 281–287. Warner, C.M. and Versteegh, L.R. (1974) In vivo and in vitro effect of α-amanitin on preimplantation mouse embryo RNA polymerase. Nature, 248, 678–680. Submitted on September 26, 2001; accepted on November 27, 2001

911