Biol Reprod Email Content Delivery
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.
Agricola
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.
Biology of Reproduction 64, 90-99 (2001)
© 2001 Society for the Study of Reproduction, Inc.

Regular Article

Pregnancy-Specific Glycoprotein (PSG) in Baboon (Papio hamadryas): Family Size, Domain Structure, and Prediction of a Functional Region in Primate PSGs1

Guang-Qian Zhoua, and Sten Hammarström2,a

a Department of Immunology, Umeå University, SE-90185 Umeå, Sweden

ABSTRACT

Pregnancy-specific glycoprotein (PSG) constitutes a major component of serum of pregnant women and appears to be essential for a successful pregnancy. Its function is, however, still unknown. Because of the evolutionary divergence between human and rodent PSG, functional studies may require a primate animal model. We have characterized PSG transcripts in a baboon placenta cDNA library and analyzed baboon genomic DNA. The main PSG isoform had the domain structure N-A1-B2-C similar to the human type IIa isoform. The type I isoform (N-A1-A2-B2-C) was also expressed. Fifteen similar PSG genes were identified of which at least nine were simultaneously expressed in third trimester baboon placenta. Thus, the baboon PSG family was as complex as that of humans. Recombinant baboon PSG (isoform IIa) had a molecular weight of 38 kDa and reacted with antibodies against human PSG. Comparative analysis of 43 N-domain amino acid sequences of PSG from four species and nine primate carcinoembryonic antigen subgroup N domain sequences identified a number of residues in the GFCC'C'' ß-sheet and FG loop that are probable candidates for PSG binding to its putative ligand.

pregnancy, reproductive immunology

INTRODUCTION

Primates and rodents synthesize a group of placental glycoproteins termed pregnancy-specific glycoprotein (PSG), (also known as Schwangerschafts Proteine 1 [SP1]) [1–4]. In humans, PSG is produced by the syncytiotrophoblast starting as early as about the time of implantation [5]. Pregnancy-specific glycoprotein is released into the maternal circulation and increases in concentration as pregnancy proceeds, reaching a concentration of 200–400 µg/ml at term [6]. Similarly, baboon, rhesus, and cynomolgus monkeys have been shown to synthesize PSG or a PSG-like protein in the placenta and to release it into the circulation of the pregnant female in a gestation-course-dependent way [3, 7–9]. In rodents, PSG is produced by the secondary giant cell-spongiotrophoblast, i.e., the tissue corresponding to the syncytiotrophoblast [2, 4]. Although high levels of PSG are only seen in the syncytiotrophoblast, PSG cannot be considered to be entirely pregnancy specific because PSG mRNA has been demonstrated in fetal liver, salivary gland, testis, and myeloid cells [10, 11].

Pregnancy-specific glycoprotein seems to be necessary for normal pregnancy. Thus, injection of antibodies specific for human PSG into pregnant monkeys was shown to induce abortion [12]. Moreover, in humans, lower serum PSG levels have been correlated with fetal growth retardation, fetal stress, and fetal loss [13]. It has been suggested that PSG may function as an inhibitor of cell-matrix interactions because most human PSGs (8 of 11) contain the tripeptide motif, arginine-glycine-asparic acid (RGD), at an exposed position in the N-terminal domain [14]. However, in mouse and rat PSG there is no RGD motif in the N-terminal domain [15]. Pregnancy-specific glycoprotein has also been suggested to have an immunomodulatory function [16–18] and an embryotrophic activity [19], and to enhance platelet and white blood cell counts after bone marrow transplantation [20].

In humans, there are 11 similar PSG genes (>90% sequence identity at the nucleotide level), clustered within a 700-kilobase (kb) region on chromosome 19q13.2 [21, 22]. In mouse and rat 15 and 4 different PSG genes, respectively, have been identified [15, 23]. The PSG subgroup, the carcinoembryonic antigen (CEA) subgroup and the CEACAM-ps6–11 subgroup constitute the CEA family [24, 25], which in turn belongs to the immunoglobulin superfamily (IgSF) [26]. Human PSGs are made up of one Ig variable-like domain (N), followed by one to three Ig constant type 2-like domains of two different types (A and B), and a short hydrophilic tail (C). Common domain arrangements are type I (L-N-A1-A2-B2-C), type IIa (L-N-A1-B2-C), and type IIb (L-N-A2-B2-C) [27–29], where L represents the signal peptide that is cleaved off in the mature protein. Mature PSGs from human placenta have molecular weights of 72, 64, and 54 kDa [24] and contain approximately 30% carbohydrate [30]. The rodent PSGs are made up of multiple N-domains followed by a single A domain. The following arrangements have been seen: (L-N1-N2-N3-A), (L-N1-N3-A), and (L-N1-N2-N3-N4-N5-A) [15].

As a step to obtain a relevant animal model for functional studies of human PSG, we have characterized PSG transcripts in a baboon placenta cDNA library and L/N exons of baboon PSG genes and expressed PSG cDNAs as proteins in vitro.

MATERIALS AND METHODS

Screening of cDNA Library

A baboon (Papio hamadryas) placenta cDNA library constructed in Lambda ZAP II vector was obtained commercially (Stratagene, La Jolla, CA). The library contained cDNA from placenta of a wild-caught baboon that was 7–8 yr old and in the last trimester of pregnancy (140–160 days). The library was amplified in the Escherichia coli XL-1 blue MFR' strain, and the plaques were transferred to nylon filters (Boehringer Mannheim, Mannheim, Germany). A human PSG1d cDNA fragment containing L-B2 exons was labeled with digoxigenin (Boehringer Mannheim) and was then used as a probe to screen the library. Prehybridization and hybridization were carried out in 5x SSC (1x SSC = 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0), 0.1% N-lauroylsarcosine, 0.02% SDS, and 1% blocking reagent (Boehringer Mannheim), at 60°C for 18 h. Final washes were performed in 0.5x SSC containing 0.2% SDS at 60°C for 30 min. Positive plaques were picked, mixed, and subjected to secondary and tertiary screenings. After the third screeing in vivo, single-clone excision of the 12 pBluescript phagemids containing positive inserts was performed and the resulting pBluescript phagemids were used for subsequent sequencing analysis.

Polymerase Chain Reaction Amplification of Baboon PSG L/N Exons and PSG cDNA Pools

A consensus primer pair (#1 and #2) for baboon PSGL/N exon was constructed on the basis of sequence information from isolated baboon PSG cDNA clones and human PSG L/N exons [21], corresponding to positions 14–34 and 350–370 of the exon, respectively. The sequences of these primers were 5'-CTTCTGGAAYCYGCCCACCA-3' and 5'-GAKATAAGGTGWAAYGTMCA-3', respectively. A primer pair (#3 and #4) derived from the human CEA subgroup molecules [31] was also used to amplify baboon PSG L/N sequences. These primers were 5'-CGCGAATTCTGGAACCYRCCMAMCASTGC-3' and 5'-GTGCTGCAGCNYNGAACTGKCYRGTTRCTTCTTC-3', corresponding to the sequence 15–37 and 348–373 of L/N exons of CEA subgroup genes, respectively. A primer (#5) with the sequence 5'-CCCTCCYCAGCCCCTYCCTGCAC-3', derived from L exon (position 9–31) of human PSG and a primer (#6) with the sequence 5'-GCTGAGTTAYGARCAGAGCAA-3', from position 195–215 of human PSG B2 exons, were also utilized. In the sequence formulas, ambiguous bases are marked Y for T or C; K for G or T; W for A or T; M for A or C; R for A or G; S for C or G; and N for A or C or T or G.

For amplification of PSG L/N exon sequences from the baboon genome, total genomic DNA (100 ng per 50 µl reaction) isolated from a baboon lymphoblastoid cell line, 26CB1 (European Collection of Cell Culture, UK), and primer pairs #1 + #2 and #3 + #4 (0.2 mM of each primer), respectively, were used. To amplify L-L/N exons from the placenta library a phage suspension (5 µl in 50 µl reaction mixture) of the baboon placenta cDNA library (titer = 1 x 109 plaque-forming units [pfu]) was used as template in combination with primer pairs #5 + #2 or #5 + #4. Other components in the polymerase chain reaction (PCR) were 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 100 mM of each dNTP, 0.001% gelatin, and 2.5 units of Thermus aquaticus (Taq) DNA polymerase (MBI, Vilnius, Lithuania). The samples were subjected to 34 cycles of PCR amplification at 95°C for 1 min, 51°C for 1 min, and 72°C for 1 min, except in the first cycle (95°C for 2 min), and the last cycle (72°C for 5 min), using a programmable thermal cycler (MJ Research, Inc., Watertown, MA).

To determine the relative frequency of alternatively spliced PSGs, phage suspensions of human and baboon placenta cDNA libraries and individual PSG-positive clones from the first cDNA library screening, respectively, and primer pair #5 + #6 were used for PCR amplification. Except for the extension step for each cycle that was 72°C for 2 min instead of 1 min, all other parameters were the same as those mentioned above. The amplified DNA fragments were analyzed by electrophoresis in 1% agarose gel and visualized by ethidium bromide staining.

Subcloning and Sequencing

The PCR products of baboon PSG L/N and L-L/N exons were inserted into the plasmid pBluescript II SK(+) (Strategene). Five to 10 individual recombinant clones containing the inserts of correct size from each amplification were sequenced using the dideoxy-chain-termination method with thermosequence fluorescent-labeled primer cycle sequencing kit (7-deaza dGTP) (Amersham Pharmacia Biotech, Solna, Sweden) and Cyt-5 labeled T7 and pUC/M13 reverse primers. For sequencing of isolated cDNA clones, the specific internal primers and AutoRead sequencing kit (Amersham Pharmacia Biotech) were also utilized.

Sequence Alignment and Phylogenetic Analysis

The Genetic Computer Group (GCG) software package version 10 was used to compare nucleotide and amino acid sequences. Nucleotides corresponding to the primers were excluded. The nucleic acid and deduced amino acid sequences were aligned using the PileUp program of the GCG package for progressive and pairwise alignments. Based on the alignment, the sequences of human PSG L/N exons, baboon PSG L/N exons, rodent PSG L/N exons, and a few human CEA subgroup L/N exons were used to make a distance matrix. For this purpose the minimum evolutionary distance in the PAUPsearch program was utilized. The trees were constructed using the neighbor-joining algorithm. Human CEA subgroup N-domain sequences were used as outgroups to make a phylogenetic tree of primate PSG N-domains. When phylogenetic analysis included CEA-related sequences from different species, human and mouse CD2 and human P0 IgV sequences were used as outgroups. To estimate the confidence level of the branching topology in the phylogenetic trees, bootstrap analysis was performed. The values give the frequency (%) by which a particular branching point was obtained (500 replications). A value of >95 is considered reliable.

Expression of Recombinant Baboon PSGs and Immunodetection

To express baboon PSG58 and PSG70 cDNAs, primers located in the PSG cDNA L/N and B2 exons were used. They were 5'-GGGCAAGTCACGATTGAAGCC-3' and 5'-CTCAATGATGATGATGATGATGATGTGTCATGGATT-3', respectively. Restriction endonuclease sites (underlined) for BglII and EcoRI were introduced in the primers to facilitate subsequent cloning to the DES vector, pMT/BiP/V5-His (Invitrogen, Carlsbad, CA). This vector makes the recombinant protein secretory. Using this cloning strategy, the mature recombinant proteins contain two extra amino acid residues (arginine and serine) at the N-terminal and six histidine residues at the C-terminal. The expression constructs were confirmed by sequencing and introduced into Drosophila Schneider (S2) cells using the standard calcium phosphate precipitation method. The supernatants of transient transfectants were collected 4 days post-transfection. Ten microliters of the supernatant were mixed with 10 µl of 2x SDS-PAGE sample buffer and subjected to electrophoresis. The proteins were then transferred to a polyvinylidene fluoride (PVDF) nylon membrane and probed with polyclonal rabbit anti-human PSG antibodies (Dako, Glostrup, Denmark) followed by swine anti-rabbit Ig antibody-horseradish peroxidase conjugate. The ECL Western blotting analysis system (Amersham, Buckinghamshire, UK) was utilized for the final detection.

RESULTS

Isolation and Characterization of Baboon PSG cDNA Clones

A baboon placenta cDNA library was screened with a probe corresponding to the L/N to B2 region of human PSG1d cDNA. One hundred fifty positive clones were obtained. Twelve of these were partially sequenced and were found to represent five different sequences similar to human PSG (Fig. 1). One (B12) was almost of full length, while the other four were truncated. Clone B12 coded for a PSG with the domain formula (L-N-A1-B2-C) (Fig. 1). Three additional clones (B10, B105, B98) coded for PSGs with the same domain formula while the fifth clone (B102) most likely coded for a PSG with the formula (L-N-A1-A2-B2-C). Clone B98 and B102 were alternatively spliced forms of the same PSG gene. The predicted N-domain sequences of the baboon PSG clones were found to be two amino acids shorter than the N-domain of human PSGs. None of the four baboon PSG species contained an RGD motif. When baboon and human PSG sequences were compared at the nucleotide level, the highest degree of identity was 83%, 90%, 86%, 87%, and 83% for the L, L/N, A1, A2, and B2 exons, respectively. Two different types of short, hydrophilic C-terminal tails were identified (Fig. 1A). The tail encoded by clones B10, B12, B102, and B98 was different from all known human C-terminal tail sequences [22, 29], while the one encoded by clone B105 was similar (66%) to the tail of human PSG3-4C5 and PSG5–3C5 [32, 33]. Baboon PSGs contained several potential glycosylation sites in the N, A1, and A2 domains (Fig. 1). The mature proteins should therefore be glycosylated in analogy with human PSGs. Interestingly, the B2 domain lacked any glycosylation sites.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 1. A) Multiple alignment of deduced amino acid sequences of five baboon PSG cDNA clones and of human PSG1-4C4 (PSG1d). Domain borders are indicated by arrows. Potential glycosylation sites are boxed. B) Schematic display of domain structure of the five baboon PSG cDNA clones. Glycosylation sites are indicated as lollypops

To determine which of the different splice variants of PSG were preferentially expressed in baboon placenta the cDNA library was amplified by PCR with primers flanking the L and B2 exons (Fig. 2). For comparison, a human placental cDNA library was also analyzed. The dominating splice form in baboon placenta was the type II splice variant (L-L/N-A1/A2-B2-C) (Fig. 2). However, type I, III, and IV forms were also detected. In human placenta, types I and IV splice variants were more strongly expressed. Two faint bands (~1900 base pairs [bp] and ~300 bp) were nonspecifically amplified products as confirmed by sequencing (Fig. 2). A predominance of type II transcripts was also found when phage suspensions of isolated PSG-positive clones were used as template in PCR; 40 out of 55 baboon PSG clones yielded a band corresponding to type II transcripts (data not shown).



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 2. Demonstration of type I, II, III, and IV PSG transcripts in baboon and human placenta cDNA libraries. The location of the PCR primers in the L and B2 exons and the expected sizes of the transcripts are indicated

Size of Baboon PSG Gene Family and PSG Genes Expressed in Placenta

To obtain an estimate of the number and identity of PSG genes in the baboon genome and of PSG genes expressed in baboon placenta, we took advantage of the fact that all human CEA-PSG genes contain a single L/N exon [21] and that PSG transcripts lacking the L/N exon are rare [28]. The baboon PSG L/N exon fragments from total genomic DNA and the L-L/N exon fragments from the placental cDNA library were amplified. From each source 50 and 40 clones were sequenced. The results are summarized in Table 1 and Figure 3. All genes except one allele of PSG69 were in the correct reading frame. Altogether 15 different baboon PSG L/N exon sequences were identified, 14 in genomic DNA and 9 in the cDNA library (8 of which were also seen among the genomic sequences and 1, PSG70, only in the cDNA library). The PSG transcripts were present in different frequencies in the placental cDNA library; PSG58 was found in nine of 40 cDNA clones and, therefore, seems to be the main PSG species expressed in baboon term placenta (Table 1). The baboon L/N exon sequences were very similar to each other (~90% at the nucleotide level) and showed 83–91% identity to human PSG genes and less than 65% identity to rodent PSGs.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Number and frequency of different baboon PSG N-domain exons isolated from total genomic DNA and a term placental cDNA library by PCR



View larger version (99K):
[in this window]
[in a new window]
  		FIG. 3. Alignment of all known PSG N-domain amino acid sequences. For rodent PSGs the N1 domain sequence was used. Potential glycosylation sites are boxed. The RGD sequence is underlined. ß-strands are indicated (A to G) and amino acid residues characteristic for the Ig-fold are shown as filled and open squares. Sequence comparisons were made in two ways. The upper line shows the comparison between baboon and human and the lower line the comparison between all four species. A filled circle denotes that >=90% of the residues at that position were identical and an asterisk denotes that >=60% of the residues at that position belonged to the same conservative group of residues as defined in parentheses (G, A, S), (I, L, M, V), (F, H, W, Y), (D, E), (H, K, R), (S, T), and (N, Q). The DNA sequences reported in this paper have been deposited in the GenBank database. Accession numbers are: PSG56-AF226643; PSG57-AF226644; PSG58 (B10)-AF226638; PSG59-3C (B98)-AF22639; PSG59-4C (B102)-AF226641; PSG60-AF226645; PSG61-AF226646; PSG62-AF226647; PSG63-AF226648; PSG64-AF226649; PSG65-AF226650; PSG66-AF226651; PSG67-AF226652; PSG68-AF226653; PSG69 (B105)-AF226642; PSG70 (B12)-AF226640.

Expression of Recombinant Baboon PSGs

Baboon PSG58 (B10) and PSG70 (B12) cDNA clones were used for expression studies. The molecular weight of the two PSGs was calculated to 38 400 Da, assuming that two of two NXS/NXT sites (B10) and two of three NXS/NXT sites (B12) in the PSGs are substituted by a complex type of carbohydrate chain of a molecular weight of 2800 Da [34]. Figure 4 shows an immunoblot of supernatants from Schneider cells transfected with baboon PSG58 and PSG70 constructs, respectively, using polyclonal rabbit anti-human PSG antibodies for detection. A prominent band with a molecular weight of 38 kDa was seen for both constructs. The value is in excellent agreement with the expected molecular weight.



View larger version (56K):
[in this window]
[in a new window]
  		FIG. 4. Immunoblotting with rabbit anti-human PSG antiserum of two recombinant baboon PSGs expressed in Drosophila Schneider cells, electrophoresed on SDS-PAGE, and transferred to PVDF nylon membrane. The molecular weights of three standard proteins are indicated

Phylogenetic Analysis of PSG N-Domains

Figure 5A shows a dendrogram of the deduced peptide sequences of all human and baboon PSG L/N exons. Using any human CEA subfamily member as functional outgroup, all N-domains of baboon PSGs were clustered together in one group and all N-domains of human PSGs in another. The bootstrap value for the branching point between the two species was 100, thus the branching is highly reliable (Fig. 5A). Some N-domain sequences within each species were clustered together. However, only for PSG56 and PSG57 and for PSG58 and PSG59, respectively, were the bootstrap values high enough to consider the branching points almost reliable. The degree of homology between human and baboon PSG N-domains was 73–89% at the protein level.



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 5. Phylogenetic trees showing the relationships between N-domain peptide sequences of baboon, human, rat, and mouse PSG and CEA subgroup members. A) Relationship between human and baboon PSG N-domains. B) Relationship between N-domains of representative members of the CEA and the PSG subgroups in four different species. Bootstrapping analysis (500 replications) assessed the reliability of the branching pattern. Numbers are bootstrap percentage. Only values greater than 50% are indicated

If members of the CEA subgroup from primates and rodents were included in the phylogenetic analysis, and either human or mouse CD2 or human P0 protein was used as an outgroup, the primate PSGs were clustered together with primate CEA subfamily molecules while rodent PSGs were excluded (Fig. 5B). Again, baboon PSGs were most closely related to human PSGs, while N-domain sequences of baboon CEA subgroup members were subgrouped together with the corresponding human CEA subgroup sequences. Very similar phylogenetic trees were obtained using the nucleotide sequences (data not shown).

Multiple Alignment of Primate and Rodent PSG N-Domains

Including the baboon PSG N-domains identified in this study, altogether 43 different PSG N-domain sequences in four species are now known. A multiple alignment is shown in Figure 3. In the figure the ß-strands of the N-domain of the human CEA family are indicated [34, 35]. Assuming that the PSGs have the same function in all four species and that the ubiquitous N-terminal domain is directly involved in this function, it should be possible to make some predictions about the location of the surface that interacts with a hypothetical PSG ligand. Eight positions in the PSG N-domain may be glycosylated. The glycosylation sites were located in the A-strand (in a few baboon PSGs only), in the loop between the B- and C-strands (in primate PSGs and one mouse PSG), in the C-, C'-, C''-strands or loops between them (three positions in some rodent PSGs only), in the loops between the D- and E-strands and between the E- and F-strands (in practically all PSGs), and in the beginning of the G-strand (rodent PSGs only). This leaves residues in the C-, C'-, and F-strands as well as in the CC', C'C'', and FG loops free to interact with the hypothetical PSG ligand.

Arginine 64 and Asp82 were conserved in all species probably forming an intramolecular salt bridge [10]. Similarly, a number of amino acid residues essential for the characteristic Ig folding pattern were also conserved in all PSG N-domains from all four species. They were Val17, Leu19, Trp33, Gly63, Leu73, Ile75, Asp82, Gly84, and Tyr86 (Fig. 3). The degree of sequence variability is shown as invariant residues (circles), similar residues (asterisks), and variant residues (unmarked) (Fig. 3). The upper line in Figure 3 shows sequence variability for primates only and the lower line for all four species. Two regions, positions 37 to 60 and positions 88 to 108, corresponding to the C'- and C''-strands plus the CC' and C'C'' loops and the G-strand plus the FG loop, respectively, displayed the most pronounced sequence variability. However, within these regions there were several relatively well-conserved residues.

A number of amino acid residues that are shared by human and baboon but are replaced by other residues in rodents were also identified. These were Ser12, Lys15, Glu16, Gln26, Tyr31, Thr46, Tyr61, Val67, Ala71, Leu74, Ser85, and Ile90. Tyrosine 44 in primate PSGs is deleted in rodent PSGs. These amino acid residues are distributed among most family members in a lineage-specific way and suggest a process of sequence homogenization during evolution.

Because the phylogenetic analysis using N and N1 (for rodents) domain amino acid sequences showed that rodent PSGs are less related to human PSGs than primate CEA subgroup members (Fig. 5B) and that putative carbohydrate binding sites had a rather different distribution in rodent PSG N-domains compared to primate PSG N-domains (Fig. 3), we focused further comparisons on CEA family members from primates. It is generally considered that the PSGs have a different physiological function than CEA or CEACAM1 [25, 36]. Therefore, direct sequence comparisons between PSG and CEA subfamily members in primates might indicate possible contact amino acids in a putative PSG binding site. Indeed, when consensus sequences of primate PSG N-domains were compared with N-domain sequences of expressed CEA subgroup molecules, several amino acid residues were readily identified that were conserved in primate PSGs (present in >90% of the sequences), while all or almost all primate CEA subgroup molecules contained an amino acid residue with different properties at the corresponding position (Fig. 6). They were Ile32, Gln37, Tyr 44, Thr46, Asp51, Ile55, Gly94, Asp95, Thr97, and Gly99. It is interesting to note that all these residues were located in the C-, C'-, and C''-strands or in the CC', C'C'', and FG loops.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6. Alignment of N-domain consensus sequences for human and baboon PSGs with the N-domain sequences of known human and baboon CEA subgroup sequences. Primate PSG N-domain residues characteristic for PSG as compared to CEA subgroup members are underlined

DISCUSSION

Nonhuman primates have previously been shown to express PSG and PSG mRNA as revealed by immunological cross-reactivity with anti-human PSG antisera [3, 7–9] and by cross-hybridization with a human PSG cDNA probe [37].

In this study we have characterized baboon PSGs and compared them with human PSGs. We found that 1) the degree of sequence homology was very high and that PSG from the two species had the same types of domain arrangements. However, in baboon the type II domain arrangement dominated while in humans the type I domain arrangement dominated. This finding suggests that the A domain is of less importance from a functional point of view. 2) Both the baboon and the human PSG gene families were large with >=15 and 11 members, respectively. 3) Similar to human and rodent PSGs [28, 38, 39], many baboon PSG genes and alternative splice forms of them were coordinately expressed in the placenta. 4) Recombinant baboon PSGs were shown to contain epitopes recognized by antibodies against human PSGs. 5) Finally, in contrast to all other CEA family molecules but in accordance with human and rodent PSGs, the C-terminal IgC-like domain did not contain any glycosylation sites [15]. The peptide backbone of this PSG domain may either play a role in the formation of dimers (oligomers) or may in some way be involved in ligand interaction. We conclude that baboon PSG is sufficiently similar to human PSG for the baboon to constitute an adequate animal model for functional studies.

The size of recombinant baboon PSG from both cDNA clones was approximately 38 kDa, indicating that the two baboon PSGs were glycosylated to the same degree in Drosophila Schneider (S2) cells even though the B12 clone contained a third potential site for N-glycosylation. Perhaps the glycosylation sites at Asn77 and Asn81 in B12 are too close together for both to be glycosylated in the same molecule. In previous studies [7, 24, 37], size heterogeneity of human and baboon PSGs from placenta was observed. Although the simultaneous expression of type I to IV splice forms contributes significantly to size heterogeneity, differential glycosylation of the PSG polypeptide may also contribute. In addition, dimerization of at least some PSGs may occur [30].

The CEA gene family in humans is thought to have arisen by gene duplication from an ancestral CEACAM1-like gene. The genes in the PSG and CEA subgroups have evolved separately after an initial gene duplication event some 108 million yr ago or about the time of human-rodent divergence [15, 40, 41]. Phylogenetic analysis using N-domain sequences from human and baboon PSGs showed that PSG members within a species were more closely related to each other than to PSG members from the other species. The presence of lineage-specific residues and deletions, the lack of apparent counterparts between baboon and human PSG genes, and the difference in number of PSG genes in the two species support the view that the CEA-PSG gene family is still under the process of evolution by gene duplication and/or gene conversion, leading to sequence homogenization of each family member of a species. This conforms to the general pattern of evolution of multigene families. As can be seen from the comparative sequences alignment (Fig. 3), however, the impact of the evolutionary processes is not random (see below).

The IgV-like N-domain has been shown to be the most important domain functionally in a number of IgSF molecules, including CD2, CD58, CD48, CD22, CD80, CTLA-4, and CEACAM1 (BGP), CEACAM5 (CEA), and CEACAM6 (NCA) [25, 36, 42]. In most cases the N-terminal domain alone contained the ligand binding site, even though this may not always be the case [25, 36, 42–45]. There are good reasons to believe that the IgV-like N-terminal domain of the PSGs is the functionally most important. Although the crystal structure of the IgV-like N domain of a CEA family molecule has not yet been determined, molecular modeling based on x-ray crystallographic data from CD2, CD4, and Bence-Jones protein REI suggests that the domain is made up of two stacked pairs of antiparallel ß-sheets, A, B, E, D and G, F, C, C', C'', respectively, forming a ß-barrel [29, 34, 35, 44, 45]. Interestingly, the epitopes of four monoclonal antibodies directed against the N-domain of CEA that were able to inhibit CEA-mediated cell adhesion were located in a GFCC' ß-sheet [46]. Furthermore, all amino acid residues identified to be important for Neisseria meningitidis and Neisseria gonorrhoeae opacity-associated (Opa) protein binding were located on the GFCC' face of the human CEA subgroup molecules [35, 47]. The GFCC'C'' ß-sheet in the CEA subfamily molecules is free of oligosaccharide side chains and most amino acid residues in the peptide chain are accessible to the solvent [34].

A comparative alignment of the 43 PSG N-domain sequences from primates and rodents (Fig. 3) revealed that amino acid residues in the C-, C'-, and F-strands as well as the CC', C'C'', and FG loops are free to interact with the hypothetical ligand. If primate PSG N-domain sequences are considered separately, the carbohydrate-free area of the N-domain can be extended to the entire GFCC'C'' face. It should be noted, however, that these conclusions are based on the assumption that PSG is not forming dimers in solution, at least not via this face of the domain [48]. A number of amino acid residues on the GFCC'C'' face were well conserved particularly when primate and rodent sequences were considered separately. Furthermore, when primate PSG N-domain sequences were compared with primate CEA subgroup sequences (Fig. 6), several residues were identified that were well conserved among the 26 primate PSG sequences, while in the CEA subgroup sequences (n = 9) there were amino acid residues belonging to a different group of amino acids. Among 14 relatively well-conserved residues, 10 residues (Ile32, Gln37, Tyr 44, Thr46, Asp51, Ile55, Gly94, Asp95, Thr97, and Gly99) were the most conserved. Possibly, some or all of these are interacting with the putative ligand. Obviously, some amino acid residues that are conserved in the entire primate CEA family located on the GF CC'C'' face may also be involved in ligand binding such as Tyr34, Lys35, Gly36, Asp40, Ile45, Tyr48, and Lys92. That the corresponding amino acid residues at these positions in human CEA subfamily molecules are solvent exposed and able to interact with Opa proteins have been demonstrated in two recent studies [35, 47].

Five of 15 baboon PSG N-domains and 8 of 11 human PSGs [14] contained the RGD motif at positions 93–95 in the N-domain, while mouse and rat PSGs lack this motif [15]. We have tested the hypothesis that human PSGs could act as inhibitors of integrin-matrix interactions. However, our experiments have not provided evidence that PSGs containing the RGD motif interact with integrins [Zhou and Hammarström, unpublished]. Moreover, a PSG11-derived peptide that contains the RGD sequence and was shown to bind to the promonocyte cell line THP-1 did not reveal a binding protein of the size of an integrin [17]. Thus, although G94 and D95 may be involved in the interaction with a PSG ligand it is probably not a classical RGD-dependent integrin-matrix interaction.

Finally, as commented upon by Holness and Simmons [43], in spite of the very conserved patterns of domain folding and domain organization seen in members of the IgSF, there is a surprising diversity in their mode of interaction. Binding sites can be localized to single domains or involve faces encompassing several domains. Whichever is the case for primate PSGs remains to be elucidated. To understand the function of primate PSG is the primary objective for future studies.

FOOTNOTES

First decision: 14 April 2000.

1 This work was supported by grants from the Swedish Cancer Society (0706-B97-25XAC) and the Swedish Medical Research Council (K2000-71X-09945-09C). Back

2 Correspondence. FAX: 46 90 7852250; sten.hammarstrom{at}climi.umu.se Back

Accepted: August 25, 2000.

Received: March 9, 2000.

REFERENCES

  1. Tatarinov YS, Masyukevich VN. Immunological identification of a new beta1-globulin in the blood serum of pregnant women. Bull Exp Biol Med 1970; 69:66–68.
  2. Chemnitz J, Hau J, Svendsen P, Folkersen J, Westergaard JG, Christensen BC. Immunohistochemical demonstration of human and murine pregnancy-associated serum proteins in maternal and placental tissue. Bibl Anat 1982; 22:87–92.
  3. Braunstein GD, Asch RH. Pregnancy-specific beta 1-glycoprotein concentrations throughout pregnancy in the rhesus monkey. J Clin Endocrinol Metab 1986; 62:1264–1270.[Abstract/Free Full Text]
  4. Rebstock S, Lucas K, Weiss M, Thompson J, Zimmermann W. Spatiotemporal expression of pregnancy-specific glycoprotein gene rnCGM1 in rat placenta. Dev Dyn 1993; 198:171–181.[Medline]
  5. Grudzinskas JG, Lenton EA, Gordon YB, Kelso IM, Jeffrey D, Sobowale O, Chard T. Circulating levels of pregnancy-specific beta1-glycoprotein in early pregnancy. Br J Obstet Gynaecol 1977; 84:740–742.[Medline]
  6. Lin TM, Halbert SD, Kiefer D. Characterization of four pregnancy-associated plasma proteins. Am J Obstet Gynecol 1974; 118:223–236.[Medline]
  7. Bohn H, Sedlacek H. Comparative studies on placenta specific proteins in man and subhuman primates (author's translation). Arch Gynakol 1975; 220:105–121.[CrossRef][Medline]
  8. Stevens VC, Bohn H, Powell JE. Serum levels of a placental protein during gestation in the baboon. Am J Obstet Gynecol 1976; 124:51–54.[Medline]
  9. Germain G, Mignot T, Bischof P, Mouren M, Cedard L. Plasma levels of SP1-like immunoreactive material (SP1) and progesterone throughout pregnancy and following RU 486-induced early abortion in the cynomolgus monkey (Macaca fascicularis). Hum Reprod 1989; 4:975–980.[Abstract/Free Full Text]
  10. Thompson JA, Grunert F, Zimmermann W. Carcinoembryonic antigen gene family: molecular biology and clinical perspectives. J Clin Lab Anal 1991; 5:344–366.[Medline]
  11. Hammarström S, Khan W, Teglund S, Hammarström M-L, Ramos T, Baranov V, Yeung MMW, Frängsmyr L. The carcinoembryonic antigen family. In: van Regenmortel MHV (ed.), Structure of Antigen, vol. 2. Boca Raton, FL: CRC Press; 1993: 341–376.
  12. Bohn H, Weinmann E. Antifertility effect of an active immunization of monkeys with human pregnancy-specific beta 1-glycoprotein (SP1) (author's translation). Arch Gynaekol 1976; 221:305–312.[CrossRef][Medline]
  13. Tamsen L, Johansson SGO, Axelsson O. Pregnancy-specific beta1-glycoprotein (SP1) in serum from women with pregnancies complicated by intrauterine growth retardation. J Perinat Med 1983; 11:19–24.[Medline]
  14. Khan WN, Teglund S, Bremer K, Hammarström S. The pregnancy-specific glycoprotein family of the immunoglobulin superfamily: identification of new members and estimation of family size. Genomics 1992; 12:780–787.[CrossRef][Medline]
  15. Zimmermann W. The nature and expression of the rodent CEA families: evolutionary considerations. In: Stanners CP (ed.), Cell Adhesion and Communication Mediated by the CEA Family: Basic and Clinical Perspectives. Amsterdam: Harwood Academic Publisher; 1998: 31–55.
  16. Majumdar S, Bapna BC, Mapa MK, Gupta AN, Devi PK, Subrahmanyam D. Pregnancy specific proteins: suppression of in vitro blastogenic response to mitogen by these proteins. Int J Fertil 1982; 27:66–69.[Medline]
  17. Rutherfurd KJ, Chou TY, Mansfield BC. A motif in PSG11s mediates binding to a receptor on the surface of the promonocyte cell line THP-1. Mol Endocrinol 1995; 9:1297–1305.[Abstract/Free Full Text]
  18. Arnold LL, Doherty TM, Flor AW, Simon JA, Chou JY, Chan WY, Mansfield BC. Pregnancy-specific glycoprotein gene expression in recurrent aborters: a potential correlation to interleukin-10 expression. Am J Reprod Immunol 1999; 41:174–182.
  19. Wu SM, Arnold LL, Rone J, Trivadi M, Chan WY. Effect of pregnancy-specific ß1-glycoprotein on the development of preimplantation embryo. Proc Soc Exp Biol Med 1998; 220:169–177.[CrossRef][Medline]
  20. Blomberg LA, Cohn ML, Cahill RA, Chan WY. Effect of human pregnancy specific ß1 glycoprotein on blood cell regeneration after bone marrow transplantation. Proc Soc Exp Biol Med 1998; 217:212–218.[CrossRef][Medline]
  21. Teglund S, Olsen A, Khan WN, Frängsmyr L, Hammarström S. The pregnancy-specific glycoprotein (PSG) gene cluster on human chromosome 19: fine structure of the 11 PSG genes and identification of 6 new genes forming a third subgroup within the carcinoembryonic antigen (CEA) family. Genomics 1994; 23:669–684.[CrossRef][Medline]
  22. Thompson J, Koumari R, Wagner K, Barnert S, Schleussner C, Schrewe H, Zimmermann W, Müller G, Schempp W, Zaninetta D, Ammaturo D, Hardman N. The human pregnancy-specific glycoprotein genes are tightly linked on the long arm of chromosome 19 and are coordinately expressed. Biochem Biophys Res Commun 1990; 167:848–859.[CrossRef][Medline]
  23. Beauchemin N, Draber P, Dveksler G, Gold P, Gray-Qwen S, Grunert F, Hammarström S, Holmes KV, Karlsson A, Kuroki M, Lin SH, Lucka L, Najjar SM, Neumaier M, Öbrink B, Shively JE, Skubitz KM, Stanners CP, Thomas P, Thompson JA, Virji M, von Kleist S, Wagener C, Watt S, Zimmermann W. Redefined nomenclature for members of the carcinoembryonic antigen family. Exp Cell Res 1999; 252:243–249.[CrossRef][Medline]
  24. Watanabe S, Chou JY. Isolation and characterization of complementary DNAs encoding human pregnancy-specific beta1-glycoprotein. J Biol Chem 1988; 263:2049–2054.[Abstract/Free Full Text]
  25. Hammarström S. The carcinoembryonic antigen (CEA) family: structure, suggested functions and expression in normal and malignant tissues. Semin Cancer Biol 1999; 9:67–81.[CrossRef][Medline]
  26. Paxton RJ, Mooser G, Pande H, Lee TD, Shively JE. Sequence analysis of carcinoembryonic antigen: identification of glycosylation sites and homology with the immunoglobulin supergene family. Proc Natl Acad Sci U S A 1987; 84:920–924.[Abstract/Free Full Text]
  27. Chou JY, Plouzek CA. Pregnancy-specific-glycoprotein. Semin Reprod Endocrinol 1992; 10:116–126.[CrossRef]
  28. Teglund S, Zhou GQ, Hammarström S. Characterization of cDNA encoding novel pregnancy-specific glycoprotein variants. Biochem Biophys Res Commun 1995; 211:656–664.[CrossRef][Medline]
  29. McLenachan PA, Lockhart PJ, Faber HR, Mansfield BC. Evolutionary analysis of the multigene pregnancy-specific beta 1-glycoprotein family: separation of historical and nonhistorical signals. J Mol Evol 1996; 42:273–280.[CrossRef][Medline]
  30. Osborn JC, Rosen SW, Nilsson B, Calvert I, Bohn H. Physiochemical studies of pregnancy-specific beta1-specific glycoprotein: unusual ultracentrifugal and circular dichroic properties. Biochemistry 1982; 21:5523–5528.[CrossRef][Medline]
  31. Khan WN, Frängsmyr L, Teglund S, Israelsson A, Bremer K, Hammarström S. Identification of three new genes and estimation of the size of the carcinoembryonic antigen family. Genomics 1992; 14:384–390.[CrossRef][Medline]
  32. Zoubir F, Khan WN, Hammarström S. Carcinoembryonic antigen gene family members in submandibular salivary gland: demonstration of pregnancy-specific glycoproteins by cDNA cloning. Biochem Biophys Res Commun 1990; 69:203–216.
  33. Khan WN, Hammarström S. Carcinoembryonic antigen family: molecular cloning of cDNA for a PSG/FL-NCA glycoprotein with a novel domain arrangement. Biochem Biophys Res Commun 1989; 161:525–535.[CrossRef][Medline]
  34. Boehm MK, Mayans MO, Thornton JD, Begent RHJ, Keep PA, Perkins SJ. Extended glycoprotein structure of the seven domains in human carcinoembryonic antigen by X-ray and neutron solution scattering and an automated curve fitting procedure: implication for cellular adhesion. J Mol Biol 1996; 259:718–736.[CrossRef][Medline]
  35. Virji M, Evans D, Hadfield A, Grunert F, Teixeira AM, Watt SM. Critical determinants of host receptor targeting by Neisseria meningitidis and Neisseria gonorrhoeae: identification of Opa adhesiotopes on the N-domain of CD66 molecules. Mol Microbiol 1999; 34:538–551.[CrossRef][Medline]
  36. Öbrink B. CEA adhesion molecules: multifunctional proteins with signal-regulatory properties. Curr Opin Cell Biol 1997; 9:616–626.[CrossRef][Medline]
  37. Tease LA, Fazleabas AT, Chan WY. Characterization of baboon pregnancy-specific beta 1-glycoprotein. Biol Reprod 1989; 41:1113–1121.[Abstract]
  38. Streydio C, Vassart G. Expression of human pregnancy specific beta 1 glycoprotein (PSG) genes during placental development. Biochem Biophys Res Commun 1990; 166:1265–1273.[CrossRef][Medline]
  39. Kromer B, Finkenzeller D, Wessels J, Dveksler G, Thompson J, Zimmermann W. Coordinate expression of splice variants of the murine pregnancy-specific glycoprotein (PSG) gene family during placental development. Eur J Biochem 1996; 242:280–287.[Medline]
  40. Rudert F, Zimmermann W, Thompson J. Intra- and interspecies analysis of the carcinoembyonic antigen (CEA) gene family reveal independent evolution in primates and rodents. J Mol Evol 1989; 29:126–134.[CrossRef][Medline]
  41. Frängsmyr L, Israelsson A, Teglund S, Matsunaga T, Hammarström S. Evolution of the carcinoembryonic antigen family. Structures of cgm9, cgm11 and pregnancy-specific glycoprotein promoters. Tumor Biol 2000; 21:63–81.[CrossRef]
  42. Davis S, Ikemizu S, Wild M, van der Merwe PA. CD2 and the nature of protein interactions mediating cell-cell recognition. Immunol Rev 1998; 163:217–236.[CrossRef][Medline]
  43. Holness CL, Simmons DL. Structural motifs for recognition and adhesion in members of the immunoglobulin superfamily. J Cell Sci 1994; 107:2065–2070.[Medline]
  44. Bates PA, Luo J, Stern MJ. A predicted three-dimensional structure for the carcinoembryonic antigen (CEA). FEBS Lett 1992; 301:207–214.[CrossRef][Medline]
  45. Jones EY, Harlos K, Bottomley MJ, Robinson RC, Driscoll PC, Edwards RM, Clements JM, Dudgeon TJ, Stuart DI. Crystal structure of an integrin-binding fragment of vascular cell adhesion molecule-1 at 1.8 Å resolution. Nature 1995; 373:539–544.[CrossRef][Medline]
  46. Nasu T, Fukuda Y, Hashino J, Nagahira K, Kawashima H, Noguchi C, Oikawa S, Nakanishi T. Epitope mapping of monoclonal antibodies against N-domain of carcinoembryonic antigen (CEA). Immunol Lett 1999; 67:57–62.[CrossRef][Medline]
  47. Bos, MP, Hogan D, Belland RJ. Homologue scanning mutagenesis reveals CD66 receptor residues required for neisserial Opa protein binding. J Exp Med 1999; 190:331–340.[Abstract/Free Full Text]
  48. Lin SH, Cheng HW, Earley K, Luo WP, Chou J. Demonstration of adhesion activity of the soluble Ig-domain protein C-CAM4 by attachment to the plasma membrane. Biochem Biophys Res Commun 1998; 245:472–477.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 ReproductionHome page
F. Wynne, M. Ball, A. S McLellan, P. Dockery, W. Zimmermann, and T. Moore
Mouse pregnancy-specific glycoproteins: tissue-specific expression and evidence of association with maternal vasculature.
Reproduction, April 1, 2006; 131(4): 721 - 732.
[Abstract] [Full Text] [PDF]

Home page
 J. Leukoc. Biol.Home page
C. T. Ha, R. Waterhouse, J. Wessells, J. A. Wu, and G. S. Dveksler
Binding of pregnancy-specific glycoprotein 17 to CD9 on macrophages induces secretion of IL-10, IL-6, PGE2, and TGF-{beta}1
J. Leukoc. Biol., June 1, 2005; 77(6): 948 - 957.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.
Agricola
Right arrow Articles by Zhou, G.-Q.
Right arrow Articles by Hammarström, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS