Molecular Human Reproduction, Vol. 9, No. 5, 279-290,
May 2003
© 2003 European Society of Human Reproduction and Embryology
Article |
A novel serine protease of the mammalian HtrA family is up-regulated in mouse uterus coinciding with placentation
Submitted on December 18, 2002; accepted on February 4, 2003
1 Prince Henry’s Institute of Medical Research, P.O.Box 5152, 246 Clayton Road, Clayton, Victoria 3168, Australia 2 Current address: Department of Obstetrics and Gynecology, Mie University School of Medicine, Mie, Japan
3 To whom correspondence should be addressed. e-mail: guiying.nie{at}med.monash.edu.au
 |
ABSTRACT |
|---|
 Top ABSTRACT IntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
This paper characterizes a novel gene, previously identified as uniquely regulated at implantation in mouse uterus. We cloned its full mRNA sequence encoding a serine protease possessing an IGF-binding domain and named it pregnancy-related serine protease (PRSP). PRSP is structurally similar to mammalian HtrA1 (56% amino acid similarity). Northern analysis revealed that the expression of PRSP mRNA was low before pregnancy, but it was increased at implantation and markedly up-regulated post-implantation. In-situ hybridization localized low levels of mRNA expression to the epithelium and stroma during very early pregnancy, but high expression to the decidual cells on day 8.5, primarily at the mesometrial pole where the placenta was forming. By day 10.5, PRSP mRNA was detected in the placenta. We also cloned an alternatively spliced PRSP mRNA that is expressed at a very low level. We located PRSP gene on chromosome 5 and established its intron/exon structure, which unambiguously explains how the two mRNA variants are produced through alternative splicing. Based on PRSP protein domain structure and its unique expression during pregnancy, we propose that PRSP plays an important role in the formation/function of the placenta.
Key words: HtrA/implantation/mouse uterus/placentation/pregnancy-related serine protease
|
Introduction |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Embryo implantation is a critical factor limiting successful pregnancy. It is a complex process involving active interaction between the blastocyst and the uterus. The uterus must transform itself from a non-receptive to a receptive state primarily through the coordinated effects of the ovarian hormones. In the mouse, on day 4.5 of pregnancy (vaginal plug = day 0), the embryos are distributed evenly along the uterine horns and are starting to attach to the endometrium. At this time, the uterus undergoes considerable morphological changes in association with cell proliferation and differentiation, leading to the acquisition of a receptive state (Finn and McLaren, 1967; Abrahamsohn and Zorn, 1993). Uterine remodelling at this time is marked by an increase in vascular permeability at implantation sites (Psychoyos, 1973). It is likely that the proliferation and differentiation of endometrial cells at this time are associated with up-/down-regulation of a number of unique genes. To search for such genes, we utilized the technique of RNA differential display (DDPCR) and compared the mRNA expression pattern of implantation and interimplantation sites on day 4.5 of pregnancy (Nie et al., 2000a). This initial study identified 17 candidate genes, 10 of which were confirmed to show differential expression between the two sites (Nie et al., 2000a). Subsequent molecular characterization of these confirmed genes found that one represented by band 10 on the DDPCR gel (Nie et al., 2000a) was novel. The full-length mRNA of this gene encodes a putative serine protease, and the expression of this gene changed during pregnancy. We have termed this novel gene pregnancy-related serine protease (PRSP).
PRSP protein was most homologous to the mammalian HtrA proteins. The main function of HtrA family proteins is to control cell fate via regulated protein metabolism (reviewed by Clausen et al., 2002). HtrA was first identified in E. coli as a periplasmic protein required for high temperature tolerance (Lipinska et al., 1988; Strauch et al., 1989). It has been suggested to act as a serine protease to cleave toxic denatured proteins (Lipinska et al., 1990). Recently two mammalian homologues of E. coli HtrA (40% identical) have been described: HtrA1 and HtrA2. Human HtrA1 was initially identified as being repressed in SV40-transformed fibroblasts (Zumbrunn and Trueb, 1996). The same protein was later identified in osteoarthritic cartilage (Hu et al., 1998). HtrA2 (also called OMI) was identified as a binding protein to the mammalian inhibitor of apoptosis homologue A (Verhagen et al., 2002) or mix2-interacting protein (Faccio et al., 2000). The exact in-vivo function of HtrA1 is still to be identified. Recently, a number of studies have established that HtrA2 plays an important role in programmed cell death (Suzuki et al., 2001; Martins et al., 2002; van Loo et al., 2002; Verhagen et al., 2002).
The present investigation reports the identification, detailed molecular cloning and initial characterization of PRSP mRNA and protein. It also describes the expression pattern and cellular localization of PRSP mRNA in the mouse, particularly in the uterus during early–mid-pregnancy.
|
Materials and methods |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Animals and tissue preparation
Swiss outbred mice were housed and handled according to the Monash University animal ethics guidelines on the care and use of laboratory animals. Adult female mice (6–8 weeks old) were mated with fertile males of the same strain to produce pregnant mice. The morning of finding a vaginal plug was designated as day 0 of pregnancy. Uterine tissues were collected from non-pregnant mice and pregnant mice on days 3.5–10.5 (n = 4 per time point). A selection of other mouse organs was also collected from non-pregnant mice. Tissues were snap-frozen in liquid nitrogen for extracting RNA and DNA, and also fixed in 4% buffered formalin (pH 7.6) for in-situ hybridization and immunohistochemistry.
For non-pregnant and day 3.5 pregnant mice, the entire uterus was collected. On day 4.5 of pregnancy, implantation sites were visualized by i.v. injections of a Chicago Blue dye (Sigma Chemical Co., St. Louis, MO, USA) solution (1% in saline, 0.1 ml/mouse) into the tail vein 5 min before killing the animals. From day 5.5 onwards, implantation sites were visualized without dye injection. On days 4.5–6.5, implantation sites were separated from interimplantation sites and both sites were collected without further dissection. From day 7.5 onwards, only implantation sites were collected and further dissected (Hogan et al., 1994) to separate the maternal and fetal tissues whenever possible. Briefly, on days 7.5–8.5, the decidua and the enclosed embryo in the implantation unit were sampled together and separated from the uterus under a dissecting microscope. On day 9.5 the decidua, that by now contains the developing placenta, was separated from the embryo and the uterus. On day 10.5, the clearly formed placenta was separated from the embryo and uterus.
Extraction of cDNA from DDPCR gel, reamplification and sub-cloning
To establish the identity of the gene discovered by the previous DDPCR analysis (Nie et al., 2000a), the band representing this gene was cut from the DDPCR gel, cDNA extracted, reamplified, cloned into pGEM-T vector and sequenced as described previously (Nie et al., 2000c).
Northern analysis
Total RNA was isolated and Northern analysis was performed as previously published (Nie et al., 2000a;b). In brief, RNA (15 µg) was denatured with dimethylsulphoxide (DMSO) and glyoxal and analysed following electrophoresis through a 1.2% agarose gel and transferred to positively charged nylon membranes (Hybond-N+; Amersham). Radioactively labelled cDNA probes were generated by random primer labelling of 25 ng cDNA with [32P]deoxy-CTP (50 µCi/reaction). Hybridization was at 42°C overnight. To determine lane-to-lane loading variation, each blot was also probed with a mouse cDNA probe for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or 18S ribosomal RNA.
cDNA library screening, 5' and 3' RACE (rapid amplification of cDNA ends)
A 
gt11 cDNA library prepared from mouse uterus (Clontech, USA) was screened as previously described (Nie et al., 2000c). The positive phase DNA was isolated by the Lambda mini kit (Qiagen Pty Ltd, Australia), and the cDNA insert was released by digestion with Bsi WI restriction enzyme. The insert was then sequentially treated with Klenow and Taq DNA polymerase, sub-cloned into pGEM-T easy vector (Promega, USA) according to the manufacturer’s instructions and sequenced.
Poly-(A)+ mRNA was isolated from total RNA with the PolyATract mRNA Isolation System (Promega) and 5' and 3' RACE were performed using the 5'/3' RACE kit (Roche, Australia) and Expand High Fidelity PCR system (Roche). 1 µg mRNA from day 4.5 interimplantation sites was used for the 5' RACE, and 1 µg mRNA from day 4.5 interimplantation sites, day 8.5 decidua and day 10.5 placenta were used for the 3' RACE. One oligo d(T)-anchor primer and one anchor primer (Table I) were used for both RACE analyses. In addition, three nested lower primers 5'RACE-L1, L2 and L3 were used for the 5' RACE, and one upper primer mP10-U was used for the 3' RACE (Table I).
|
RT–PCR and T/A cloning
RT–PCR was carried out as previously described (Nie et al., 2000b). In brief, 1 µg DNA-free RNA was reverse-transcribed with random hexanucleotide primers and AMV reverse transcriptase (Boehringer Mannheim, Australia) and PCR performed in a total volume of 40 µl with 1–1.5 µl of the RT reaction. The high fidelity Taq polymerase (Boehringer Mannheim) was used for cloning open reading frames (ORF) and normal taq DNA polymerase (Boehringer Mannheim) was used for cloning probes for Northern analysis. PCR products were analysed on 1.5% agarose gel and bands of interest were cut out, purified with Qiaquick gel extraction kit (Qiagen) and cloned into a pGEM-T easy vector (Promega). All primers used for RT–PCR cloning are listed in Table I. A 785 bp fragment (mP10–785), representing both the long and short mRNA forms, was amplified with primer pair L-ORF-U and 5' RACE-L1 from day 4.5 interimplantation sites. Long and short form specific cDNA probes, long-specific P (475 bp) and short-specific P (476 bp), were amplified with primer pair longP-U and longP-L, shortP-U and shortP-L respectively from day 8.5 decidua. The long form ORF was cloned with primer L-ORF-U and two nested lower primers L-ORF-L1 and L-ORF-L2 from day 4.5 interimplantation sites. The short form ORF was cloned with primer S-ORF-U and S-ORF-L from day 4.5 interimplantation sites, day 8.5 decidua and day 10.5 placenta.
In-situ hybridization and immunohistochemistry
Sense and anti-sense digoxigenin (DIG)-labelled RNA probes against a cDNA fragment of 344 bp (nt 821–1164, sequence a, Figure 2) were generated using the DIG RNA Labeling kit (Boehringer Mannheim), and the concentrations determined according to the manufacturer’s instructions. Sections (5 µm) of formalin-fixed paraffin-embedded tissues were subjected to in-situ hybridization as previously described (Nie et al., 2000b). Immunohistochemistry for desmin (goat polyclonal antibody; Santa Cruz, USA) was carried out as previously published (Nie et al., 2000a).
|
Gene analysis by Southern blot analysis and by bioinformatics
For Southern blot analysis, genomic DNA was isolated from non-pregnant uterus and kidney using the DNeasy Tissue Kit (Qiagen) and digested (10 µg for each enzyme) at 37°C for 14 h with an excess of one each of the following four restriction endonucleases: Taq I, Hind III, Eco R1 and Bam H1. The DNA was then fractionated on 0.8% agarose gel, transferred to positively charged nylon membranes (Hybond-N; Amersham) using the standard Southern blotting procedure (Sambrook et al., 1989) and probed with radioactively labelled cDNA as described for the Northern analysis.
To search the genomic structure of PRSP gene and its chromosomal location in the mouse, initially the mouse genome sequence in the NCBI Database was searched against the mRNA sequence(s), however, the DNA sequence available at the time in the region where PRSP is located was not complete. Hence the genome sequence available in the Celera Databases was compared with the mRNA sequence(s) using the BLAST program on the Celera server. The gene whose sequence shows perfect alignment to the mRNA was further analysed, and a possible intron–exon structure was then determined using the alignment result.
|
Results |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Cloning of cDNA obtained from DDPCR and further confirmation by Northern blotting
In our previous study we identified a gene represented by band 10 on the DDPCR gel that was expressed more highly in interimplantation than implantation sites (Nie et al., 2000a). This differential expression pattern was confirmed by Northern analysis using the cDNA extracted from band 10 (Nie et al., 2000a).
To establish the identity of this gene, the cDNA extract of band 10 was cloned. Clone 10.9, when used as a probe on a Northern blot, detected a clear single transcript of 2.8 kb with a higher level of expression in interimplantation than implantation sites on day 5.5 of pregnancy (Figure 1A). This differential expression pattern is identical to that displayed by band 10 on the DDPCR gel and to that detected by the crude cDNA extract of band 10 (Nie et al., 2000a), confirming that clone 10.9 represents the original band 10.
|
Sequence analysis of clone 10.9 and comparison with the Genbank database
The nucleotide sequence of clone 10.9 (Figure 1B) contains 359 nt. Importantly, it contains the unique sequence TCTGTGCTGG at its 5' end and the reverse complementary sequence of T12MG at its 3' end (underlined in Figure 1B). This is important because band 10 resulted from DDPCR amplification of day 4.5 interimplantation site RNA with the specific 5' primer TCTGTGCTGG (OPA-14) and 3' primer of T12MG (Nie et al., 2000a). The presence of these primer sequences in clone 10.9 unambiguously verifies that this cDNA was amplified specifically during the DDPCR exercise.
When the GenBank database was searched against clone 10.9 (early 1998), only two short expressed sequence tags (EST) AA839689 (393 nt, from pregnant mouse uterus) and AA823108 (328 nt, from mammary gland) showed 95% similarity. Longer mRNA sequences that showed high homology were not found, indicating that clone 10.9 represented a novel gene whose full length mRNA sequence was yet to be cloned.
Cloning the full length cDNA sequence
To obtain the full length mRNA sequence represented by clone 10.9, a gt11 cDNA library derived from mouse uterus was screened using the 359 bp cDNA of clone 10.9 as a probe. A number of positive clones were identified and three were sub-cloned and fully sequenced. Two of these were identical and contained 1675 nt (represented by nt 772–2446 of sequence a, Figure 2), the other was much shorter and contained only 980 nt (represented by nt 1467–2446 of sequence a, Figure 2). The shorter one was identical to the last 980 nt of the longer ones, indicating that these sequences represented different lengths of the same mRNA. The 3' ends of all these library-derived clones were 98% identical to clone 10.9. However, none of these sequences contained the translation start codon.
To acquire more nt towards the 5' end, three nested reverse primers 5' RACE-L1, L2 and L3 (Table I) were designed based on the 1675 nt sequence and 5' RACE was performed on RNA from day 4.5 interimplantation sites. The longest sequence obtained from the RACE revealed an additional 771 nt to the 5' end of the 1675 nt sequence (represented by nt 1–771 of sequence a, Figure 2). Compiling the sequences obtained from the 5' RACE and the library screening resulted in a long sequence of 2446 nt, designated sequence ‘a’ (Figure 2). This sequence contains a clear open reading frame (ORF) of 1377 nt with the start codon ATG at nt 127–129 and stop codon TGA at nt 1504–1506. Outside the ORF is a 5' untranslated G/C-rich (72% G/C) region of 126 nt and a long stretch of 3' untranslated region of 940 nt (Figure 2). To further confirm the accuracy of this ORF, given that it was compiled from two separate sequences, nested RT–PCR was performed using RNA from day 4.5 interimplantation sites and primers flanking the ORF: L-ORF-U, L-ORF-L1 and L-ORF-L2 (Table I). A single band of the expected size was obtained. When cloned and sequenced, it matched exactly the ORF of sequence a (Figure 2), confirming that this 2446 nt sequence represents the full length mRNA sequence of clone 10.9.
The above ORF can be translated into an amino acid (aa) sequence of 459 residues (sequence A, Figure 2). This protein is predicted to have a molecular mass of 49 kDa and an isoelectric point of 7.08. The N-terminal end of the sequence contains a long stretch of hydrophobic amino acids that may represent a signal peptide (aa 1–23). A search of the NCBI protein structure database showed that this protein contains a further four domains (Figure 3): (i) an insulin-like growth factor (IGF) binding domain (aa 34–85), (ii) a kazal-type S protease inhibitor domain (aa 95–132), (iii) a trypsin protease domain (aa 182–347) and (iv) a PDZ domain (aa 390–446).
|
Comparison of sequences a and A (Figure 2) with all entries in the GenBank and Swissprot database (early 1999) revealed that the most homologous entries were mammalian HtrA1. At the cDNA level, there is 63% identity with the mouse (Accession: AF172994) and 65% with the human (Accession: D87258 and Y07921) HtrA1 cDNA sequences. At the protein level, there is 56% identity with the mouse (Accession: AAD49422) and 58% with the human (Accession: BAA13322 and CAA69226) HtrA1 proteins (Figure 4). The HtrA1 proteins were reported to be homologous to the HtrA/Do proteases from bacteria (Zumbrunn and Trueb, 1996; Hu et al., 1998). These HtrA proteins belong to a family of serine proteases that possess the amino acid sequence motif of GNSGGAL (in bacteria) or GNSGGPL (in mammals) in their active sites. In addition, these proteins display a motif of TNAHV residues in the vicinity of GNSGGPL and the catalytic triad His, Asp and Ser for their catalytic activity (Zumbrunn and Trueb, 1996). Interestingly, all of these motifs are present on the protein shown in sequence A (Figure 2 and Figure 4): the serine protease active site sequence GNSGGPL is at aa 309–315, the TNAHV residues are at aa 194–198, and the catalytic triad His (aa 197), Asp (aa 233) and Ser (aa 311). It can be concluded that we have cloned a novel mRNA encoding a functional serine protease that possesses an IGF-binding motif and is related to pregnancy. We have tentatively named this protein pregnancy-related serine protease (PRSP).
|
Expression of PRSP mRNA in the uterus during implantation and placentation
Northern analysis was carried out to determine the expression pattern of this gene in the uterus during the peri-implantation period. A cDNA fragment of 785 bp (nt 76–860 of sequence a, Figure 2) was used as a probe. Total RNA from the uterus of non-pregnant mice (estrus) and pregnant mice during implantation was analysed (Figure 5A). Relatively low expression of a single band of 2.8 kb was detected in non-pregnant mice, a marginally higher level was seen on day 3.5 and a small increase was detected around the time of implantation (days 4.5–6.5). Notably, the increased level of mRNA expression was detected initially in interimplantation sites (day 4.5–5.5), although increased expression was detected in both sites at day 6.5 (Figure 5A).
|
Interestingly, a marked up-regulation of PRSP mRNA was detected post-implantation (Figure 5B). Clear up-regulation was first detected on day 7.5, with further increases on days 8.5 and 9.5. On day 7.5 and 8.5, the decidua/embryo could be easily dissected out of the uterus in one piece but further separation was difficult. At these time points, high expression was detected in the decidua/embryo. On day 9.5, the embryo could be dissected from the uterus and the decidua (which now contains the developing placenta). At this time, high expression was detected in the decidua/developing placenta, with relatively lower expression in the embryo, and very low expression in the remainder of the uterus. On day 10.5, expression was mainly in the placenta (now clearly formed).
Organ distribution of PRSP mRNA
The organ distribution of PRSP mRNA was investigated by Northern analysis (Figure 6). When equal amounts of total RNA were compared, the highest level of expression was in day 10.5 placenta. Apart from the uterus, among the 12 organs examined, testis, ovary and heart showed moderate expression while muscle and lung showed low expression. No expression was detected in any other organs.
|
Cloning of an alternatively transcribed PRSP mRNA variant
Only a single band at 2.8 kb was detected on most Northern blots in this study. However, an additional low intensity band at 2.2 kb was sometimes observed (Figure 6, day 10.5 placenta). This 2.2 kb band could represent an additional low abundance transcript of PRSP, which might be due to alternative splicing at the 3' end, because the 5' RACE experiments consistently detected only a single mRNA species at the 5' end. Therefore, 3' RACE analysis was performed with a forward primer (mP10-U) located in the middle (nt 1096) of sequence A (boxed, Figure 2) and used previously for sequencing the library-derived clones. The kit used for the 5' RACE was also used for 3' RACE and two clear bands of
800 bp and 1.3 kb were obtained in all three samples tested, the 1.3 kb product being the most abundant (Figure 7). This provided strong evidence that PRSP is alternatively transcribed, and that the splicing point is after nt 1096 on sequence a. Cloning and sequencing of both bands (1.3 kb and 800 bp) demonstrated that the sequence of the 1.3 kb band was 100% identical to the 3' end of sequence a, whereas the 800 bp band (underlined in sequence b, Figure 2) was somewhat different. It contains 100 nt at its 5' end that can be aligned perfectly to nt 1096–1195 of sequence a (Figure 2), but the rest of the sequence is unique (italicized in sequence b, Figure 2). This indicates that this 800 bp sequence represents the 3' end of an alternatively transcribed mRNA variant, and that this mRNA is likely to be identical to sequence a until nt 1196. To verify these findings, RT–PCR cloning was performed with a forward primer (S-ORF-U, Table I) located at the extreme 5' end of sequence a and a lower primer (S-ORF-L, Table I) located on the unique sequence of the 800 bp band. A single band of the expected size of 1.5 kb was obtained and this cDNA was cloned. Its sequence shows identical nt to sequence a until nt 1196; it then aligns perfectly to the unique sequence of the 800 bp band. Thus the full sequence of this alternatively spliced mRNA variant can be illustrated as that of sequence b, Figure 2. It is 1897 nt long and shows a clear ORF with the start codon at nt 127–129 and the stop codon at nt 1216–1218 (boxed on sequence b, Figure 2). The deduced protein of this ORF contains 363 aa (sequence B, Figure 2). It has a predicted molecular mass of 38.5 kDa and an isoelectric point of 7.73. Structurally, at the domain level, the protein of sequence B is identical to that of sequence A, except that it completely lacks the PDZ domain (Figure 3).
|
Sequence b is 549 nt shorter than sequence a, the corresponding protein sequence B is 96 aa shorter than sequence A (Figure 2). Thus sequence a/A will be referred to as the long form mRNA/protein, and sequence b/B as the short form mRNA/protein. Figure 8 schematically illustrates the relationship between the long and short mRNA variants, the long and short isoform proteins, and between the mRNA and proteins. This illustration also demonstrates why only the long form mRNA was obtained by the library screening using the DDPCR-derived clone 10.9, which represents only the 3' end of the long form. The mP10–785 probe used on previous Northern blots (Figures 5A, B and 6) detects both mRNA forms.
|
Northern blot analysis to confirm the expression of the short form PRSP mRNA
Following cloning of the short form mRNA sequence of PRSP, both long and short mRNA specific probes were designed and the expression of the two mRNA was examined by Northern analysis. The expression pattern of the long form is identical to that of the 2.8 kb band detected previously by probe mP10–785 (data not shown). A single 2.2 kb band was detected using the short form specific probe (Figure 9), confirming the identity of the 2.2 kb band detected by probe mP10–785. This mRNA was detectable only at very low levels in tissues where the long form is highly expressed (such as day 10.5 placenta). Relative expression of the two mRNA forms in day 4.5 interimplantation and day 10.5 placental tissues is shown in Figure 9.
|
Localization of PRSP mRNA in the uterus during pregnancy
The cell types that express PRSP mRNA in the uterus were determined by in-situ hybridization using riboprobes specific for PRSP. In the non-pregnant uterus, a very low level of expression was detected in the glandular and luminal epithelium, and in sub-luminal stroma (Figure 0
A). As predicted from the previous Northern analysis (Figure 5), this expression was slightly increased on day 4.5 and 5.5 of pregnancy (data not shown). Much stronger expression was detected from day 6.5 onwards at the implantation site (Figure 0B and D). On day 6.5, staining was detected in the uterine stroma, both mesometrially and anti-mesometrially (Figure 0B). Positive staining was also detected in the embryo and ectoplacental cone (Figure 0B). When a serial section was immunostained with the decidual marker desmin, the pattern did not correlate exclusively with the expression pattern of PRSP mRNA (Figure 0C). Not all anti-mesometrial desmin-positive cells stained for PRSP mRNA; likewise, not all mesometrial PRSP mRNA-positive cells were positive for desmin. The intensity of uterine staining for PRSP mRNA was further increased after day 6.5 and a very high level of expression was detected on day 8.5. At this time, a strong signal was detected in the decidual basalis at the mesometrial pole, on the same section, staining was also seen anti-mesometrially in the decidual capsularis (Figure 0D). Interestingly, this pattern mirrored the expression of desmin on a serial section (Figure 0F). At this time, a signal was also seen in the embryo. The specificity of the probe was demonstrated using a sense probe on a serial section (Figure 0E). When day 10.5 implantation sites (which included the newly fully formed placenta) were examined, mRNA was localized in both the placenta and the metrial triangle region (data not shown).
|
Genomic characterization of PRSP gene by Southern analysis and bioinformatics
Southern analysis was performed on the total genomic DNA isolated from the uterus and the kidney and digested with one each of four restriction enzymes: Taq I, Hind III, Eco RI and Bam HI. Similar results were obtained for both organs and the result for the uterus is shown in Figure 1. For all four digestions, the pattern indicated that there is only a single copy of the PRSP gene in the genome.
|
The genomic structure of PRSP was analysed by comparing its mRNA sequences with the mouse genome sequence available in the Celera Databases. A perfect match was found on a gene designated as mCG7096 located on chromosome 5. The alignment of mRNA sequences a and b to gene mCG7096 is summarized in Table II. Alignment of sequence a partitioned the long form mRNA into nine distinctive segments (segments 1–6, 8–10, Table II), each aligned perfectly to a unique region on the gene (Table II). Alignment of sequence b partitioned the short mRNA into seven clear segments (segments 1–7, Table II); each likewise aligned perfectly to a unique region on the gene (Table II). These alignment data clearly indicate that each unique mRNA segment represents an individual exon. Thus, according to the specific location of each mRNA segment and the distance between each segment on the gene, the exon–intron structure of this gene can be predicted as that shown in Figure 2. This gene consists of
10 exons, exon 10 being the largest and 5 being the smallest exon 1 contains the ATG start codon, and both exon 7 and 10 contain the TGA stop codon. This structure also clearly explains how the exons are alternatively spliced to produce the long and short form mRNA. The short mRNA (thus the short protein) is produced by utilizing exons 1–7, whereas the long mRNA (thus the long protein) is obtained by utilizing all but exon 7 (Figure 2).
|
|
|
Discussion |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
This study characterized a novel gene (PRSP) previously identified by DDPCR to be differentially expressed between implantation and interimplantation sites in the mouse uterus. We cloned the full length mRNA sequences of two alternatively spliced variants of PRSP (long and short forms). Using a bioinformatics approach, we identified the chromosomal location of the PRSP gene in the mouse and revealed its genomic structure. This structure explicitly demonstrates how the two mRNA variants are produced through alternative splicing. The genomic analysis further corroborates the accuracy of the two mRNA sequences as they could be perfectly aligned with specific regions on the gene sequence. This, in turn, verified the accuracy of the two protein sequences that could be deduced from these mRNA. It is of note that the structure of the PRSP gene established by this study is not accurately predicted by Celera’s computational annotation. PRSP is represented by gene mCG7096 (Celera databases). However, mCG7096 was predicted to transcribe an mRNA sequence of only 1846 nt (mCT6193, Celera databases); this predicted mRNA was found to correspond to only a partial sequence of the long form of PRSP mRNA. In addition, exon 7 identified in this study is completely missed by prediction and alternative splicing is not projected. This highlights the necessity and importance of experimentally verifying mRNA sequences/gene structures generated by automated computational analysis.
PRSP displays an interesting expression pattern in the mouse. It is not widely expressed, but high levels of expression were detected by Northern analysis in the heart, ovary, testis, pregnant uterus and day 10.5 placenta. Thus, apart from the heart, it is predominantly in reproductive organs with highest expression in day 10.5 placenta. In the uterus, PRSP is up-regulated during pregnancy; in particular, there is a marked increase in mRNA expression post-implantation in the mesometrial decidual cells on day 8.5 during placental development. By the time the placenta is fully formed (day 10.5), high expression of PRSP is detected in the placenta. There is a developing body of evidence supporting the notion that uterine decidual cells are critical for successful placentation (reviewed by Aplin, 2000). In particular, several gene deletions have been reported in which impaired decidualization is associated with failure of placentation and pregnancy. One of these is the inactivation of IL-11 R
(Bilinski et al., 1998; Robb et al., 1998); in these mice the extent of decidualization is much reduced and the mesometrial decidua is totally absent. As a result, the trophoblast giant cells expand to the area that normally contains the mesometrial decidua and the mice cannot form normal placentas. Further evidence supporting the important role of maternal decidua during placental development came from a recent study in which IGF binding protein 1 (IGFBP-1) was over-expressed in the maternal decidua with resultant impairment of placental development (Crossey et al., 2002). It is therefore anticipated that the expression of PRSP in the decidua may be important for placental development, although functional studies are required to identify its exact role. It is also interesting that in all 12 organs examined, the long form of PRSP is predominant over the short form, which is detected at only very low levels. Thus, at least in the 12 tissues examined in the mouse, including the uterus and placenta, the function of PRSP is probably conveyed mainly by the long isoform.
The two forms of PRSP mRNA are predicted to encode two isoforms of PRSP protein (long and short). The long isoform contains four clear domains: an IGF binding domain, a Kazal serine protease inhibitor domain, a serine protease domain and a PDZ domain. The short isoform is identical to the long one except that it completely lacks the PDZ domain. This protein domain structure suggests that PRSP protein is a serine protease that may also be capable of binding insulin-like growth factors. The Kazal inhibitor domain indicates that PRSP may be self-regulating, but it may also have the capacity to regulate other serine proteases. The PDZ domains are found in diverse proteins and believed to mediate protein–protein interactions (Woods and Bryant, 1991; Itoh et al., 1993). The presence or absence of a PDZ domain may modulate the substrate specificity of PRSP serine protease activity. Thus the two isoforms of PRSP protein may cleave different protein substrates.
To establish the function of PRSP, nucleotide and protein sequence databases (mainly GenBank and Swissprot) were periodically searched. By the time we were characterizing PRSP gene structure, an mRNA sequence of 2525 nt (Accession AY037300, version AY037300.1) cloned from mouse appeared in the GenBank (May 2001). It was predicted to encode a protein of 460 aa called toll-associated serine protease (TASP). This mRNA sequence was very similar to PRSP mRNA (99% identical) except 8 nt. However, the predicted protein of this sequence was not identical to PRSP: it contained 460 residues whereas PRSP contained 459 residues; in addition, it showed 12 residues (in the region of aa 265–279) totally different from PRSP (data not shown). The precise tissue origin of TASP mRNA is not clear in the GenBank, and the detailed cloning and characterization of TASP has not been reported. During preparation of this manuscript, a mouse mRNA sequence of 1718 nt (Accession XM_124505, version XM_124505.1) predicted by automated computational analysis (NCBI annotation project) appeared in GenBank (May 2002). Although this sequence is much shorter than PRSP mRNA sequences, it is 99% identical to the long form PRSP mRNA, and predicted to encode a protein of 459 residues that is 100% identical to the long form of PRSP protein. This is only a predicted sequence and the function of its translation product was not investigated. To date, no mRNA entry similar to the short form PRSP mRNA is present in the databases.
Previously characterized proteins of most similarity with PRSP are the mammalian HtrA1 proteins. HtrA was initially identified in E. coli. To date, two mammalian members (HtrA1 and HtrA2) of this family have been described. Mammalian HtrA1 is structurally similar to the bacterial HtrA at the C-terminal where the catalytic residues and a single PDZ domain are located, but their N termini differ completely (Zumbrunn and Trueb, 1996; Hu et al., 1998). Biochemical analysis of recombinant HtrA1 showed that it exhibits endoproteolytic activity, and that depends on the presence of the putative active site Serine 328; HtrA1 also forms a stable complex with 1-antitrypsin (Hu et al., 1998). However, its exact in-vivo function is not yet established in any tissue. HtrA2 also has a conserved serine protease domain and C-terminal PDZ domain that are found in all HtrA proteins, but its N-terminus is distinct from that of either the bacterial HtrA or mammalian HtrA1. Compared with HtrA1, the N-terminus of HtrA2 does not contain IGF-binding and Kazal domains, but instead contains a regulatory region called RD domain (Faccio et al., 2000). In addition, HtrA2 was described as an intracellular protein with a predominantly nuclear localization whereas HtrA1 is a secreted protein (Gray et al., 2000). HtrA2 was reported to have proteolytic activity, and to play an essential role during apoptosis (Suzuki et al., 2001; Martins et al., 2002; van Loo et al., 2002; Verhagen et al., 2002).
Based on protein structures, it can be envisaged that PRSP represents another member of the mammalian HtrA protein family. Due to its striking structural similarity to HtrA1 rather than HtrA2, PRSP can also be predicted to have a function(s) similar to that of HtrA1. However, the function of HtrA1 is not yet known. To date, no HtrA-related proteins have been examined in reproductive tissues/processes.
Given the unique expression of PRSP during pregnancy, the presence of an IGF binding domain is probably of importance. During pregnancy, IGF are important for growth of fetal and maternal tissues. Ablation of either the IGF-I or IGF-II gene reduced fetal size to 60% that of normal littermates (Warburton and Powell-Braxton, 1995). When both genes were deleted, pup size was reduced to 30% (Liu et al., 1993; Warburton and Powell-Braxton, 1995). It was speculated that these smaller fetuses resulted from impaired placental function. This was recently confirmed when the expression of IGF-II was abolished exclusively in mouse placenta (Constancia et al., 2002). Thus IGF-II is a major modulator of placental and fetal growth. Furthermore, it is well established that IGFBP regulate IGF bioavailability (Clemmons, 1998). The stringent regulation of the IGF/IGFBP system during placentation has been clearly demonstrated by over-expression of IGFBP-1 in the decidua in transgenic mice (Crossey et al., 2002), resulting in reduced IGF bioavailability in the decidual microenvironment. Development of the placenta is impaired in these mice which have much reduced trophoblast invasion. Another level of regulation of this system is through the proteolytic cleavage of IGFBP and/or the IGF–IGFBP complexes leading to increased levels of bio-active IGF (Ferry et al., 1999).
Given that PRSP is predicted to be a serine protease possessing an IGF binding domain and highly expressed in the decidual cells during placentation, we propose that it acts as a protease regulating the IGF–IGFBP system during the establishment of pregnancy and placental development/function. Future investigations will examine the expression of PRSP protein in the uterus from implantation to term pregnancy, and will establish its precise function in the establishment of pregnancy.
|
Acknowledgements |
|---|
We are grateful to Sue Panckridge and Samantha Park for assistance with the preparation of this manuscript. The study was made possible by funding from the Rockefeller Foundation Contraceptive 21 Program, the Wellcome Trust (grant 52666), the NH&MRC of Australia (grant 143798 to L.A.S. and 983212 and 198705 to J.K.F.) and Rockefeller/World Health Organization Initiative on Implantation.
|
REFERENCES |
|---|
|
TopABSTRACTIntroductionMaterials and methodsResultsDiscussionREFERENCES |
|---|
Abrahamsohn, P.A. and Zorn, T.M.T. (1993) Implantation and decidualization in rodents. J. Exp. Zool., 266, 603–628.[CrossRef][Web of Science][Medline]
Aplin, J. (2000) Maternal influences on placental development. Semin. Cell Dev. Biol., 11, 115–125.[CrossRef][Web of Science][Medline]
Bilinski, P., Roopenian, D. and Gossler, A. (1998) Maternal IL-11Ralpha function is required for normal decidua and fetoplacental development in mice. Genes Dev., 12, 2234–2243.
Clausen, T., Southan, C. and Ehrmann, M. (2002) The HtrA family of proteases. Implications for protein composition and cell fate. Mol. Cell, 10, 443–455.[CrossRef][Web of Science][Medline]
Clemmons, D.R. (1998) Role of insulin-like growth factor binding proteins in controlling IGF actions. Mol. Cell. Endocrinol., 140, 19–24.[CrossRef][Web of Science][Medline]
Constancia, M., Hemberger, M., Hughes, J. et al. (2002) Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature, 417, 945–948.[CrossRef][Medline]
Crossey, P.A., Pillai, C.C. and Miell, J.P. (2002) Altered placental development and intrauterine growth restriction in IGF binding protein-1 transgenic mice. J. Clin. Invest., 110, 411–418.[CrossRef][Web of Science][Medline]
Faccio, L., Fusco, C., Chen, A. et al. (2000) Characterisation of a novel human serine protease that has extensive homology to bacterial heat shock endoprotease HtrA and is regulated by kidney ischemia. J. Biol. Chem., 275, 2581–2588.
Ferry, J.R.J., Cerri, R.W. and Cohen, P. (1999) Insulin-like growth factor binding proteins: new proteins, new functions. Horm. Res., 51, 53–67.[CrossRef][Web of Science][Medline]
Finn, C.A. and McLaren, A. (1967) A study of the early stages of implantation in mice. J. Reprod. Fertil., 13, 259–267.
Gray, C.W., Ward, R.V., Karran, E. et al. (2000) Characterization of human HtrA2, a novel serine protease involved in the mammalian cellular stress response. Eur. J. Biochem., 267, 5699–5710.[Web of Science][Medline]
Hogan, B., Beddington, R., Costantini, F. and Lacy, E. (1994) Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York.
Hu, S.-I., Carozza, M., Klein, M. et al. (1998) Human HtrA, an evolutionarily conserved serine protease identified as a differentially expressed gene product in osteoarthritic cartilage. J. Biol. Chem., 273, 34406–34412.
Itoh, M., Nagafuchi, A., Yonemura, S. et al. (1993) The 220-kD protein colocalizing with cadherins in non-epithelial cells is identical to ZO-1, a tight junction-associated protein in epithelial cells: cDNA cloning and immunoelectron microscopy. J. Cell Biol., 121, 491–502.
Lawrence, J.B., Oxvig, C., Overgaard, M.T. et al. (1999) The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc. Natl Acad. Sci. USA, 96, 3149–3153.
Lipinska, B., Sharma, S. and Georgopoulos, C. (1988) Sequence analysis and regulation of the htrA gene of Escherichia coli: a sigma 32-independent mechanism of heat-inducible transcription. Nucleic Acids Res., 16, 10053–10067.
Lipinska, B., Zylicz, M. and Georgopoulos, C. (1990) The HtrA (DegP) protein, essential for Escherichia coli survival at high temperatures, is an endopeptidase. J. Bacteriol., 172, 1791–1797.
Liu, J.P., Baker, J., Perkins, A.S. et al. (1993) Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell, 75, 59–72.[Web of Science][Medline]
Martins, L.M., Iaccarino, I., Tenev, T. et al. (2002) The serine protease Omi/HtrA2 regulates apoptosis by binding XIAP through a reaper-like motif. J. Biol. Chem., 277, 439–444.
Nie, G.-Y., Li, Y., Hampton, A.L. et al. (2000a) Identification of monoclonal nonspecific suppressor factor beta (MNSFbeta) as one of the genes differentially expressed at implantation sites compared to interimplantation sites in the mouse uterus. Mol. Reprod. Dev., 55, 351–363.[CrossRef][Web of Science][Medline]
Nie, G.-Y., Li, Y., Wang, J. et al. (2000b) Complex regulation of calcium-binding protein D9k (Calbindin-D9k) in the mouse uterus during early pregnancy and at the site of embryo implantation. Biol. Reprod., 62, 27–36.
Nie, G.Y., Li, Y., Batten, L., Griffiths, B., Wang, J., Findlay, J.K. and Salamonsen, L.A. (2000c) Uterine expression of alternatively spliced mRNAs of mouse splicing factor SC35 during early pregnancy. Mol. Hum. Reprod., 6, 1131–1139.
Psychoyos, A. (1973) Endocrine control of egg implantation. In Greep, R.O., Astwood, E.G. and Geiger, S.R. (eds), Handbook of Physiology, vol. II, Female Reprodctive System, part 2. American Physiological Society, Washington, pp. 187–215.
Robb, L., Li, R., Hartley, L. et al. (1998) Infertility in female mice lacking the receptor for interleukin 11 is due to a defective uterine response to implantation. Nat. Med., 4, 303–308.[CrossRef][Web of Science][Medline]
Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York.
Strauch, K.L., Johnson, K. and Beckwith, J. (1989) Characterization of degP, a gene required for proteolysis in the cell envelope and essential for growth of Escherichia coli at high temperature. J. Bacteriol., 171, 2689–2696.
Suzuki, Y., Imai, Y., Nakayama, H. et al. (2001) A serine protease, HtrA2, is released from the mitochondria and interacts with XIAP, inducing cell death. Mol. Cell, 8, 613–621.[CrossRef][Web of Science][Medline]
van Loo, G., van Gurp, M., Depuydt, B. et al. (2002) The serine protease Omi/HtrA2 is released from mitochondria during apoptosis. Omi interacts with caspase-inhibitor XIAP and induces enhanced caspase activity. Cell Death Differ., 9, 20–26.[CrossRef][Web of Science][Medline]
Verhagen, A.M., Silke, J., Ekert, P.G. et al. (2002) HtrA2 promotes cell death through its serine protease activity and its ability to antagonize inhibitor of apoptosis proteins. J. Biol. Chem., 277, 445–454.
Warburton, C. and Powell-Braxton, L. (1995) Mouse models of IGF-I deficiency generated by gene targeting. Receptor, 5, 35–41.[Web of Science][Medline]
Woods, D.F. and Bryant, P.J. (1991) The discs-large tumor suppressor gene of Drosophila encodes a guanylate kinase homolog localized at septate junctions. Cell, 66, 451–464.[CrossRef][Web of Science][Medline]
Zumbrunn, J. and Trueb, B. (1996) Primary structure of a putative serine protease specific for IGF-binding proteins. FEBS Lett., 398, 187–192.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. Penna, H. Du, R. Ferriani, and H. S. Taylor Calpain5 expression is decreased in endometriosis and regulated by HOXA10 in human endometrial cells Mol. Hum. Reprod., October 1, 2008; 14(10): 613 - 618. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kashiwagi, C. M. DiGirolamo, Y. Kanda, Y. Niikura, C. T. Esmon, T. R. Hansen, T. Shioda, and J. K. Pru The Postimplantation Embryo Differentially Regulates Endometrial Gene Expression and Decidualization Endocrinology, September 1, 2007; 148(9): 4173 - 4184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Nie, Y. Li, K. Hale, H. Okada, U. Manuelpillai, E. M. Wallace, and L. A. Salamonsen Serine Peptidase HTRA3 Is Closely Associated with Human Placental Development and Is Elevated in Pregnancy Serum Biol Reprod, February 1, 2006; 74(2): 366 - 374. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Nie, Y. Li, M. Wang, Y. X. Liu, J. K. Findlay, and L. A. Salamonsen Inhibiting Uterine PC6 Blocks Embryo Implantation: An Obligatory Role for a Proprotein Convertase in Fertility Biol Reprod, April 1, 2005; 72(4): 1029 - 1036. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. De Luca, M. De Falco, L. De Luca, R. Penta, V. Shridhar, F. Baldi, M. Campioni, M. G. Paggi, and A. Baldi Pattern of Expression of HtrA1 During Mouse Development J. Histochem. Cytochem., December 1, 2004; 52(12): 1609 - 1617. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. De Luca, M. De Falco, V. Fedele, L. Cobellis, A. Mastrogiacomo, V. Laforgia, I. L. Tuduce, M. Campioni, D. Giraldi, M. G. Paggi, et al. The Serine Protease HtrA1 Is Upregulated in the Human Placenta During Pregnancy J. Histochem. Cytochem., July 1, 2004; 52(7): 885 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. De Luca, M. De Falco, A. Severino, M. Campioni, D. Santini, F. Baldi, M. G. Paggi, and A. Baldi Distribution of the Serine Protease HtrA1 in Normal Human Tissues J. Histochem. Cytochem., October 1, 2003; 51(10): 1279 - 1284. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||