Zinc A2-Glycoprotein: A Multidisciplinary Protein

Zinc A2-glycoprotein (ZAG) is a protein of interest because of its ability to play many important functions in the human body, including fertilization and lipid mobilization. After the discovery of this molecule, during the last 5 decades, various studies have been documented on its structure and functions, but still, it is considered as a protein with an unknown function. Its expression is regulated by glucocorticoids. Due to its high sequence homology with lipid-mobilizing factor and high expression in cancer cachexia, it is considered as a novel adipokine. On the other hand, structural organization and fold is similar to MHC class I antigen-presenting molecule; hence, ZAG may have a role in the expression of the immune response. The function of ZAG under physiologic and cancerous conditions remains mysterious but is considered as a tumor biomarker for various carcinomas. There are several unrelated functions that are attributed to ZAG, such as RNase activity, regulation of melanin production, hindering tumor proliferation, and transport of nephritic by-products. This article deals with the discussion of the major aspects of ZAG from its gene structure to function and metabolism. (Mol Cancer Res 2008;6(6):892–906) Introduction Zinc a2-glycoprotein (ZAG) is a 40-kDa single-chain polypeptide (1), which is secreted in various body fluids (2). ZAG is known to stimulate lipolysis in murine epididymal adipocytes through stimulation of adenylate cyclase in a GTPdependent process (3) via binding through h3-adrenoreceptor (4). It is involved preferentially in depletion of fatty acids from adipose tissues, subsequently named as lipid-mobilizing factor (5). This factor, which is highly expressed in cancer cachexia (6), is characterized by the extensive reduction of fat in the human body. The protein assay (7) and mRNA expression (8) in the mammary tumor have shown that there is a relation between the ZAG levels at histologic grade of the breast cancer tumors. Moreover, many studies suggested that ZAG is also a potential serum marker of prostate cancer that may be elevated early in tumor growth (9). High degree of similarity between ZAG and class I MHC molecules has been evaluated both at sequence and structural levels (10-12). The crystal structure of ZAG consists of a large groove analogous to class I MHC peptide-binding grooves. The structure and environment of groove reflect its role in immunoregulation and in lipid catabolism (11). The alteration in residues of the peptide-binding groove clearly showed uniqueness of ZAG among MHC class I– like proteins (13, 14). These observations indicate that ZAG might be able to bind with different peptides, antigens, and ligands. Hence, it was suggested that function of ZAG has diverged from the peptide presentation and T-cell interaction functions of class I MHC molecules (15). There are various reports on different aspects of ZAG. The works, however, have never been reviewed. Here, we have compiled for the first time a detailed analysis of all the relevant information for ZAG that may be helpful in understanding of functional importance of ZAG in the body system. Site of Expression The ZAG was first reported in human serum and subsequently purified (1). The presence of ZAG in human seminal fluid was reported in 6-fold molar excess compared with human serum (16, 17), which suggested as a key element for fertilization without any experimental evidences. The serum ZAG is synthesized by the genes of liver. The seminal ZAG, however, is of prostatic origin (9, 16, 17). The presence of ZAG in normal human body fluids and kidney extract has been described (18). Jirka and Blanicky (19) have reported three isoforms of ZAG through immunoelectrophoresis and reported that concentration of each human serum ZAG isoform increases from its lowest value in the fetal and early newborn period to the highest ones in children and adults. Through immunohistochemical analysis, Mazoujian (20) studied normal skin and 41 benign sweat gland tumors and found that ZAG was expressed predominantly in tumors of apocrine differentiation. It was also, however, expressed in some tumors of eccrine differentiation (21). Abundant proteins expressed in human saliva are ZAG with some other proteins determined by proteomic analysis and mass fingerprinting. Recently, the gene expression of ZAG in normal human epidermal and buccal epithelia was reported (22). ZAG is also produced by adipocytes, where the mRNA of ZAG was detected by reverse transcription-PCR in the mouse Received 12/10/07; revised 1/17/08; accepted 1/23/08. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Requests for reprints: Faizan Ahmad, Centre for Interdisciplinary Research in Basic Sciences, Jamia Millia Islamia, New Delhi 110025, India. Phone: 91-1126983409; Fax: 91-11-26983409. E-mail: [email protected] Copyright D 2008 American Association for Cancer Research. doi:10.1158/1541-7786.MCR-07-2195 Mol Cancer Res 2008;6(6). June 2008 892 on June 16, 2017. © 2008 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from white adipose tissue and in the interscapular brown fat. Finally, ZAG is synthesized by epithelial cells of prostate gland and liver, secreted into various body fluids such as serum (1), semen (23), sweat (24), saliva (24), cerebrospinal fluid (24), milk (24), urine (25), and amniotic fluid (26). The concentration of ZAG has been reported to increase dramatically in carcinomas (3). Therefore, it is also considered as a good biomarker for prostate (27), breast (28), oral (22), and epidermal (29) carcinomas. Gene Structure and Regulation The gene for ZAG, assigned to the chromosome 7q22.1 through fluorescent hybridization karyotyping, comprised four exons and three introns (30). The first exon (exon 1) is for the region from the cap site to the 6th amino acid (Gly), the second exon (exon 2) is for the region from the 6th to the 93rd amino acid (Gly), the third exon (exon 3) is for the region from the 93rd to the 185th amino acid (Asp), and the fourth exon (exon 4) is for the region from the 185th amino acid to the end of the mRNA (31). The exon 4 also contains the entire 3¶-untranslated region, including a hexanucleotide AATAAA that probably represents the polyadenylation signal of the gene (32). The serum ZAG shows complete nucleotide sequence homology with that from the prostate, which includes the signal peptide (29). The full gene of ZAG was reported by Freije et al. (8). The gene sequence of ZAG includes an open reading frame encoding an 18-amino acid long hydrophobic signal peptide and the 278 residues of the mature protein. A comparison of the amino acid sequence deduced from the nucleotide sequence with that determined for the protein isolated from the serum through chemical methods (10) reveals some minor differences. There are two substitutions (i.e., Gln and Glu) present at positions 65 and 222 of the mature polypeptide chain, respectively, instead of the Glu and Gln residues at these positions as determined by the protein sequencing. In addition, the nucleotide sequencing of ZAG (8, 32) detected the insertion of an Ile-Phe pair between residues located at positions 75 and 76 of the previous sequence determined by chemical methods (10). The nucleotide sequence analysis of ZAG gene reported that there are some intervening sequences (32), and the length of intron is larger due to the presence of an unusually high density of Alu repetitive sequences within them. A total of nine Alu sequences were identified. The five are present in the first intron, and the remaining four in the second intron. This reflects that repetitive sequences are clearly overrepresented in the ZAG gene. Eight of these Alu sequences are oriented in the opposite direction of the ZAG, whereas the remaining one (Alu-6) is oriented in the same direction (32). In addition, two MER sequences belonging to subfamilies 12 and 14 (33) and one MIR element (34) were found in the first and second introns of the gene, respectively. The gene expression of ZAG is predominantly regulated by androgens and progestins (35, 36). Glucocorticoids are also responsible for the increased ZAG expression in adipose tissue (37). Russell and Tisdale (38) examined its 5¶-flanking region that could affect the transcription of the gene. They proposed that 5¶-flanking region of the gene containing several consensus sequences could be relevant in the transcription of the ZAG gene. Finally, they suggested that ZAG expression is likely to be mediated by the interaction of several transcription factors acting synergistically on different cisacting elements. In addition, the lipolytic action of dexamethasone was attenuated by anti-ZAG antibody, suggesting that the induction of lipolysis was mediated through an increase in ZAG expression. It was further proposed that expression of ZAG is mediated through the h3-adrenoreceptor present on the ZAG gene (35). ZAG as a Biomarker Although the exact mechanism by which ZAG actively participates in tumor proliferation is not known, a large body of data exists in favor of the expression of ZAG with respect to the stages of tumor (39-42). ZAG is designated as a potential biomarker of different types of carcinomas (6, 7, 27, 35, 38, 43-46). ZAG is synthesized in the prostate itself, and its high concentrations in prostatic tissue and prostatic secretion should facilitate its action in prostate and in other tissues (9). Furthermore, the increased concentration of ZAG in semen is directly linked with the prostate pathophysiology. The immunohistochemical analysis revealed an elevated concentration of ZAG in prostatic adenocarcinoma (47). In other analysis, it was observed that the majority of prostate cancer cells tested (i.e., 35 of 48 cells) have reacted with anti-ZAG antibodies (44). Moreover, high-grade tumors expressed a minimal ZAG than the moderate-grade tumors (44). Recently, the gene expression of ZAG in prostate cancer was analyzed using cDNA microarrays (48). Interestingly, ZAG seemed to be one of the genes differentially expressed in prostate cancer samples. Furthermore, they evaluated ZAG expression through immunohistochemistry using a set of 232 tumors and found strong staining for ZAG, which was associated with a decreased risk of recurrence. The clinical and pathologic multivariate analyses were done by Hull et al. (49). They found four independent variables, pathologic stage, Gleason grade in the prostatectomy specimen, and surgical margin status. More recently, the expression of ZAG in a wide range of prostate cancer samples was determined for its role as a putative biomarker for prostate cancer progression using a semiautomated microscope system (50). There is a marked direct relation between ZAG expression and the tumor stage in prostate cancer. The ZAG expression has also been used to predict clinical recurrence and metastatic tumor progression. Low ZAG expression is associated with the clinical recurrence (27). We have also identified the elevated expression of ZAG in prostate cancer seminal fluid compared with the normal sample through two-dimensional gel electrophoresis and matrix-assisted laser desorption/ionizationtime-of-flight mass spectrometry (51). Apart from prostate cancer, ZAG is also considered as a potential biomarker for breast carcinoma due to its elevated concentration in breast tissues compared with that in the normal, benign, and malignant breast specimens (24). Many studies have confirmed that ZAG is a reliable immunohistochemical marker of apocrine cell differentiation of mammary epithelial tissues (7, 8, 35, 36, 52-56). Sanchez et al. (28) have shown that many proteins are expressed in breast cancer cell line, and ZAG is one of them. Interestingly, there is a higher ZAG level in the well-differentiated tumors than that in the moderately or poorly differentiated tumors. It was proposed that A Review on Structure and Function of ZAG Mol Cancer Res 2008;6(6). June 2008 893 on June 16, 2017. © 2008 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from ZAG may be considered as a biochemical marker of differentiation in breast cancer (7). Lopez-Otin and Diamandis (57) reviewed five common markers of prostate and breast cancer and ZAG is one of them. ZAG is normally expressed in liver, but an increased expression of the gene has been reported in both benign and malignant breast tissues (53). This may raise the plasma concentration above the threshold required to produce lipolysis. The high-level expression of ZAG in cancer cachexia has been evaluated for many years (4-6, 37, 38, 45, 58). The ability of ZAG to stimulate lipolysis and cyclic AMP (cAMP) formation was compared with the effect of lipid mobilization factor (LMF) isolated from the cancer patient’s urine. A direct relationship has been established in the expression of ZAG in cancer cachexia and their concentration in urine, and hence, subsequently, ZAG is also termed as a biomarker for cancer cachexia. Apart from the above-mentioned carcinomas, ZAG is also considered as a potential tool for the investigation of other tumors. Recently, the biomarkers for liver fibrosis in hepatitis C patients were determined by using two-dimensional gel electrophoresis and mass spectrometry. The most prominent differences were observed when serum samples from cirrhotic patients were compared with those from a healthy control serum. Inter-a-trypsin inhibitor heavy chain H4 fragments, a1 antichymotrypsin, apolipoprotein L1, prealbumin, albumin, paraoxonase/arylesterase 1, and ZAG were decreased in cirrhotic serum, suggesting the role of ZAG there (39). Irmak et al. (41) used the same techniques followed by immunoblotting and immunohistochemistry and they identified two proteins, orosomucoid and human ZAG, which were increased in the urine samples of patients with bladder cancer in comparison with the urine samples of healthy volunteers. The gene expression of ZAG from histopathologically graded oral squamous cell carcinomas was compared with that in the perilesional normal. It was observed that ZAG levels are (a) higher in the controls than those in the tumors and (b) higher in well-differentiated tumors than those in the poorly differentiated tumors. These findings led to the conclusion that ZAG can also be considered as a marker of the oral epithelial maturation (59). Mazoujian (60) studied the immunohistochemical localization of ZAG in the normal skin and in 41 benign sweat gland tumors, and they found that ZAG was expressed predominantly in tumors of apocrine differentiation. It was also, however, expressed in some tumors of eccrine differentiation (21, 60). Recently, Abdul-Rahman et al. (43) for the first time determined the presence of ZAG in gynecologic cancer. They determined the higher expression of ZAG in serum of patients with squamous cell cervical carcinoma and cervical adenocarcinoma when compared with normal. In another experiments, Garcı́a-Ramı́rez et al. (61) did the proteomic analysis of the human vitreous fluid by differential gel electrophoresis for an accurate quantitative comparison between patients with the proliferative diabetic retinopathy and nondiabetic human. They found that ZAG as a new potential candidate was involved in the pathogenesis of proliferative diabetic retinopathy. In other experiment, Jain et al. (25) have shown that ZAG is among the additional proteins in urine samples of microalbuminuriapositive diabetes patients. These proteins can be used as markers for specific and accurate clinical analyses of diabetic nephropathy. The peptide fingerprint of urine protein glomerular disease was analyzed, and they found that ZAG is a biomarker for glomerular disease (40). In conclusion, ZAG not only is a biomarker for prostate and breast cancer but also covers the most commonly occurring cancer. ZAG and Cancer The seminal plasma ZAG is synthesized by prostate epithelial cells and secreted into seminal fluid (62), and it constitutes 30% of the proteins present in the seminal fluid (18). Hale et al. (44) determined the expression of ZAG in prostate tumors and found that ZAG is produced by 73% of prostate cancers. Moreover, ZAG production is associated with tumor differentiation status. It is decreased or absent in more poorly differentiated tumors. They have also shown that this observation is similar to that of breast cancer, where the loss of ZAG production was associated with the lack of tumor differentiation. It is interesting to note that the expression of ZAG in prostate cancer is similar to that of other relevant prostatic proteins (57), such as prostate-specific antigen and prostatic acid phosphatase, which are significantly decreased in prostatic tumors (63-65). Moreover, immunohistochemical studies assured the partial loss of ZAG expression in prostatic tissue after malignant transformation (47). Recently, Lapointe et al. (48) determined the gene expression profiling of prostate cancer from 62 primary prostate tumors, as well as 41 normal prostate specimens and 9 lymph node metastases, using cDNA microarrays containing 26,000 genes. They suggested that prostate tumors can be usefully classified according to their gene expression patterns, and these tumor subtypes may provide a basis for improved prognostication and treatment stratification. ZAG is a reliable immunohistochemical marker of apocrine cell differentiation in human breast (24) and standard prognostic factors in women with early breast cancer. Women with more advanced breast cancer had higher serum ZAG levels than those with the early stage of the disease. A comparative analysis in mammary tissues from women with different diseases revealed enhanced expression of ZAG gene in benign breast lesions and a variable expression level in breast cancers (7, 8). ZAG was expressed predominantly in tumors of apocrine and eccrine differentiations (60). ZAG is considered as a marker of oral epithelia (59). In a patient with cancer cachexia and prostate cancer together, ZAG production by prostate cancer can lead to systemically elevated serum ZAG levels that may be useful diagnostically (44). Clinically, cachexia manifests with excessive weight loss in the setting of ongoing disease, usually with disproportionate muscle wasting (66). Many pieces of evidence suggest that cytokines play a central role in the pathogenesis of cachexia (67). The role of ZAG in cancer cachexia was determined by Bing et al. (6). They showed that ZAG is produced by both white and brown adipose tissues, and their mRNA and protein levels are markedly increased in adipose tissue of mice with cancer cachexia. Moreover, they suggested that ZAG may be a unique protein factor in the local modulation of lipid metabolism and contribute particularly to the substantial reduction of adipose in cancer cachexia. The ability of ZAG to induce uncoupling protein expression has been determined, and it has been found that ZAG directly influences the Hassan et al. Mol Cancer Res 2008;6(6). June 2008 894 on June 16, 2017. © 2008 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from expression of uncoupling proteins, which may play an important role in the lipid use during cancer cachexia (58). High sequence homology of ZAG with LMF further suggests the potential role of ZAG in the lipid mobilization (3, 45, 68). Despite these small chemical differences between ZAG and LMF, the biological activity of ZAG seems to be very similar to that of LMF, and pharmacologic and biochemical evidences indicate that both molecules induce lipolysis in vitro by a cAMP-mediated system through interaction with a h3-adrenoreceptor (69). Russell et al. (4) used the freeze/thawing methods to determine the effect of ZAG on lipolysis in vivo . The alteration in the secondary structure due to freeze thaw alters the conformation of ZAG, which is important for the interaction of h3-adrenoreceptor and may require ZAG for biological activity. Recently, Rolli et al. (12) used the ZAG-deficient mice to elucidate the pathophysiology of ZAG. They ablated both ZAG alleles in the mouse genome through standard genetargeting technology and found that ZAG knockout mice were overweight with respect to wild-type littermates. That is, ZAG overexpression should lead to weight loss and eventually cachexia. ZAG may be related to the development of superficial bladder cancer and to its switch to an invasive phenotype (41). In conclusion, ZAG is a clinically important protein, which is directly involved in various types of tumor proliferation. Sequence Analysis The first full polypeptide sequence of ZAG was determined by Araki et al. (10) through biochemical analysis followed by cDNA sequence determination of this protein from different sources by many groups (32, 70, 71). The studies suggested that all the ZAG from different sources has same polypeptide sequence. Some differences, however, occur at the posttranslational level. ZAG is synthesized as a 295-amino acid-long immature polypeptide, which, after deletion of NH2-terminal 17-amino acid-long signal polypeptide, is converted into mature ZAG. Interestingly, the 18th residue (i.e., first of the mature ZAG) of the immature ZAG is glutamine, which is cyclized to form pyroglutamine. Hence, the serum ZAG has blocked NH2 terminus. On the other hand, with the deletion of four to six NH2-terminal residues of the mature ZAG, seminal ZAG has free NH2 terminus. Recently, our group has purified a novel ZAG, which is complexed with the prolactin-inducible protein in the human seminal fluid and has blocked NH2 terminus due to the presence of pyroglutamine. The polypeptide structure of the mature ZAG comprised 278 amino acid residues. The amino acid composition, which has a molecular mass of 31,889 Da, is as follows: Asp = 18, Asn = 11, Thr = 9, Ser = l8, Glu = 21, Gln = 20, Pro = l7, Gly = l6, Ala = 18, Cys = 4, Val = 21, Met = 3, 11e = 9, Leu = 19, Tyr = 18, Phe = 9, Lys = 20, His = 7, Ag = 12, and Trp = 8. The presence of high tryptophan and tyrosine residues accounts for the high extinction coefficient (E 1cm) of 18.0. The 278-amino acid-long polypeptide folds into three domains; each is composed of f90 amino acid residues, named as a1, a2, and a3 domains. The mature ZAG contains four putative glycosylation sites (Asn-Xaa-Ser/Thr) at Asn, Asn, Asn, and Asn. Of the four (except Asn), three Asn residues carry glucosamines. Interestingly, Asn, which is located on a3 domain, is the bulkiest, sialylated (90%) on the mannose at 2,6-arm. The three glycans possessed the typical N-linked biantennary structure, being substituted with N-acetylneuraminic acid in a 2,6-linkage and lacking fucose. The molecular mass of the carbohydrate moiety of ZAG is 6,644 Da and that of the native glycoprotein is 38,478 Da (10). The amino acid sequence of ZAG contains four half-cystine residues. The position of cysteine and disulfide pairing determined through both biochemical and structural analysis reveals the bond formation in between residues 101 and 164 and second one between residues 203 and 258. Interestingly, the loops formed by the disulfide bonds are similar in length, 64 and 56 residues, respectively. ZAG share high sequence homology with antigen-presenting molecules such as MHC-I, which will be discussed in the next section. The sequence comparison of the human ZAG with other mammalian ZAG molecules, which is available from genome projects, is illustrated in Table 1. The sequence alignment in the table clearly indicates regions of proteins that are highly conserved across the mammals and those that are divergent. Interestingly, the amino acids of a1-a2 domains are highly conserved, strongly indicative of the functional importance of the groove in the biological functions of ZAG that are evident from the crystal structure and ligand-binding mutational studies (13, 72). On the other hand, the a3 domain of ZAG is less conserved indicative of species specificity and evolutionary constrains. The four cysteine residues, which are involved in disulfide pairing, are conserved in all mammalian ZAG. Interestingly, of four putative glycosylation sites, two (Asn and Asn) are totally conserved in all the mammalian ZAG. Fold and Stability Secondary structure of ZAG, predicted from its sequence, revealed the presence of 23% a-helix, 27% h-sheet, and 22% h-turns (10). The polypeptide folds into three domains with almost equal number of amino acid residues. Both the a1 and a2 domains of ZAG have disulfide bonds in homologous positions to MHC-I, responsible for the antigen binding in case of MHC-I molecules. ZAG is considered as a member of immunoglobulin gene superfamily based on its amino acid sequence and domain structure (10). Interestingly, ZAG does not require peptides and h2-microglobulin (h2M) for proper folding as MHC-I. The structure of classic MHC-I molecules is simply the assembly of the heavy chain with h2M to form noncovalent heterodimer loaded with peptides that are usually 8 to 9 amino acids in length. These peptides were generated by cleavage of proteins in the cytosol. These peptides are actively transported by a heterodimeric complex, known as transporter associated with antigen presentation, into the endoplasmic reticulum, where assembly with class I molecules takes place (73). The trimeric complex is then transported through the Golgi apparatus to the cell surface. The crucial role of h2M in the proper folding and function of MHC-I was determined using cell lines and knockout mice devoid of h2M expression (74-77). These experiments led to conclude that h2M (a) stabilizes cell surface expression of class I heavy chains, (b) 4 M.I. Hassan et al., unpublished result. A Review on Structure and Function of ZAG Mol Cancer Res 2008;6(6). June 2008 895 on June 16, 2017. © 2008 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from facilitates the binding of purified class I molecules to antigenic peptides both on cells and plastic surfaces, and (iii) generates additional high-affinity peptide-binding sites in preparations of soluble purified class I molecules (74-76). The far-UV CD spectrum of the native ZAG measured at pH 7.4 and 25jC (77) shows an intense negative peak at 218 nm. The best fit for the spectrum between 210 and 250 nm was found with a linear combination of 3% a-helix, 60% h-sheet, and 1% h-turns. The remaining 36% represented unordered structure (random coil), which agrees with the earlier determined secondary structure from amino acid sequence (10). In the near-UV range (250-320 nm), positive CD peak was observed at 265 nm, most probably corresponding to the disulfide chromophores (77). Furthermore, a positive peak at 295 nm is typical of tryptophan residues. The reversible thermal transition curve was determined by observing changes in the CD signal at 218 nm, which disappeared gradually on heating the protein sample in the temperature range 25jC to 85jC. The midpoint of denaturation (Tm) estimated from this curve was 66jC. The classic MHC-I heavy chain is thermally less stable in the absence of either peptide (78-82) or h2M (83), whereas ZAG exists as an isolated class I MHC-like heavy chain without bound peptides. The melting curve of ZAG shows a Tm of 65jC, which is comparable with previously obtained Tms for TABLE 1. Multiple Sequence Alignment of Human ZAG with Other Mammalian ZAGs 1 10 20 30 40 50 j j j j j j Human (100) QENQDGRYSLTYIYTGLSKHVEDVPAFQALGSLNDLQFFRYNSKDRKSQPMGLWRQVEGM 60 Monkey (92) QETQDGRYSLTYIYTGLSKLVEDIPQFQALGSLNDLQFFRYNSKDRKSQPVGLWRQVEGM 60 Bovine (66) ---QAGNYSLSFLYTGLSKPREGFPSFQAVAYLNDQPFFHYNSEGRRAEPLAPWSQVEGM 57 Rat (58) QET--GSYSLIFLYTGLSRPSKGLPRFQATAFLNDQAFFHYNSNSGKAEPVEPWSHVEGM 58 Mouse (57) QET--GSYSLTFLYTGLSRPSKGFPRFQATAFLNDQAFFHYNSNSGKAEPVGPWSQVEGM 58 Dog (60) QETQGGPYSLSFFYTGLSRPSDGFPSFQATAYLNDQDFFHYDSETGKAIPRYPWSQMEGI 60 61 70 80 90 100 110 | | | | | | Human EDWKQDSQLQKAREDIFMETLKDIVEYYNDSNGSHVLQGRFGCEIENNRSSGAFWKYYYD 120 Monkey EDWKQDSQLQKAREDIFMETLNDIVEYYNDSNGSHNLHGRFGCEIENNRSSGSFWKYYYD 120 BOVINE EDWEKESALQRAREDIFMETLSDIMDYYKDREGSHTFQGAFGCELRNNESSGAFWGYAYD 117 RAT EDWEKESQLQRAREEIFLVTLKDIMDYYEDSTGSHTFQGMFGCEITNNRSSGAVWRYAYD 118 MOUSE EDWEKESQLQRAREEIFLVTLKDIMDYYKDTTGSHTFQGMFGCEITNNRSSGAVWRYAYD 118 DOG EDWEKESKLQKAREDIFMVTLKDIMEYYKDKEGSHTFQGMFGCELQNNKNSGAFWRYAYD 120 121 130 140 150 160 170 | | | | | | Human GKDYIEFNKEIPAWVPFDPAAQITKQKWEAEPVYVQRAKAYLEEECPATLRKYLKYSKNI 180 Monkey GKDYIEFNKEIPAWVALDPAAQNTKQKWEAEPVYVQRAKAYLEEECPETLRKYLKYSKNI 180 Bovine GQDFIKFDKEIPAWVPLDPAAQNTKRKWEAEAVYVQRAKAYLEEECPGMLRRYLPYSRTH 177 Rat GEDFIEFNKEIPAWIPLDPAAANTKLKWEAEKVYVQRAKAYLEEECPTMLKKYLTYSRSH 178 Mouse GEDFIEFNKEIPAWIPLDPAAANTKLKWEAEKVYVQRAKAYLEEECPEMLKRYLNYSRSH 178 Dog GRNFIEFNKEIPAWVPQDPAALNTKKKWEAEEVYVQRAKAYLEEECPVMLQRYLEYGKTY 180 181 190 200 210 220 230 | | | | | | Human LDRQDPPSVVVTSHQAPGEKKKLKCLAYDFYPGKIDVHWTRAGEVQEPELRGDVLHNGNG 240 Monkey LDRQDPPSVVVTSHQAPGEKKKLKCLAYDFYPGKIDVHWTRDGKVQEPELRGEVLHNENG 240 Bovine LDRQESPSVSVTGHAAPGHKRTLKCLAYDFYPRSIGLHWTRAGDAQEAESGGDVLPSGNG 237 Rat LDRTDPPTVKITSRVAPGRNRIFRCLAYDFYPQRISLHWNQASKKLASEPERGVFPNGNG 238 Mouse LDRIDPPTVTITSRVIPGGNRIFKCLAYGFYPQRISLHWNKANKKLAFEPERGVFPNGNG 238 Dog LDRQEPPSVSITSHGTPEGIQTLKCWVSGFYPQEIDLHWIQADDTQETKSGGALLPSGNN 240 241 250 260 270 | | | | Human TYQSWVVVAVPPQDTAPYSCHVQHSSLAQPLVVPWEAS-------------278 Monkey TYQSWVVVAVPPQDTAPYSCYVQHSSLAQPLVVPGEAR-------------278 Bovine TYQSWVVVGVPPEDQAPYSCHVEHRSLTRPLTVPWDPRQQAE---------279 Rat TYLSWMEVEVPPQNRDPFVCHIEHKGLSQSLSVQWDEKSKV----------279 Mouse TYLSWAEVEVSPQDIDPFFCLIDHRGFSQSLSVQWDRTRKVKDENNVVAQPQ 290 Dog TYQAWVVMSASPQDLA-----------------------------------256 NOTE: Swiss-Prot entry for Homo sapiens (human), P25311;Macaca mulatta (monkey), XP_001098813; Bos taurus (bovine), Q3ZCH5; Rattus norvegicus (rat), Q63678; Mus musculus (mouse), Q64726; and Canis familiaris (dog), Q4GX49. The conserved cysteine residues involved in disulfide formation are shaded in black. The residues shaded in light gray are involved in ligand binding. Asn residues shaded in dark gray are glycosylation sites. The sequence identities with human ZAG are written in parenthesis. The secondary structure elements are shown below the sequence as loop in black line, h strand as in arrow, and a-helix as a cylinder. Hassan et al. Mol Cancer Res 2008;6(6). June 2008 896 on June 16, 2017. © 2008 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from peptide-filled class I MHC molecules: H-2K, 57jC (78); HLAA2, 66jC (84). The Tm (65jC) obtained from the ZAG melting curve is substantially higher than that (43-45jC) of an empty class I MHC molecule; H-2K complexed with human or murine h2M (78, 85). Interestingly, the newly discovered complex of ZAG with prolactin-inducible protein by our group has shown dramatic increase in Tm, which is 75jC. 4 ZAG shows high stability in the absence of peptide and h2M, which has been explained after solving the crystal structure of serum ZAG (11). The crystal structure of ZAG reveals that a network of hydrogen bonds is formed between a3 and a1-a2, whereas only single interaction has been made in h2M-binding class I proteins. In addition, the extra hydrogen bonds between loop connecting h4 strand (residues 51-54) and the helical region of the ZAG a1 domain platform a3 (236-241) are closer in ZAG compared with their MHC-I counterpart. Moreover, the burial of larger surface area 970 Å in ZAG, which in class I molecules is 660 Å, leads to constraints in the position of the ZAG a3 domain relative to a1-a2 (11).