Molecular Cloning of the CA125 Ovarian Cancer Antigen

CA125 is an ovarian cancer antigen that is the basis for a widely used serum assay for the monitoring of patients with ovarian cancer; however, detailed information on its biochemical and molecular nature is lacking. We now report the isolation of a long, but partial, cDNA that corresponds to the CA125 antigen. A rabbit polyclonal antibody produced to purified CA125 antigen was used to screen a λZAP cDNA library from OVCAR-3 cells inEscherichia coli. The longest insert from the 54 positive isolated clones had a 5797-base pair sequence containing a stop codon and a poly(A) sequence but no clear 5′ initiation sequence. The deduced amino acid sequence has many of the attributes of a mucin molecule and was designated CA125/MUC16 (gene MUC16). These features include a high serine, threonine, and proline content in an N-terminal region of nine partially conserved tandem repeats (156 amino acids each) and a C-terminal region non-tandem repeat sequence containing a possible transmembrane region and a potential tyrosine phosphorylation site. Northern blotting showed that the level of MUC16 mRNA correlated with the expression of CA125 in a panel of cell lines. The molecular cloning of the CA125 antigen will lead to a better understanding of its role in ovarian cancer. CA125 is an ovarian cancer antigen that is the basis for a widely used serum assay for the monitoring of patients with ovarian cancer; however, detailed information on its biochemical and molecular nature is lacking. We now report the isolation of a long, but partial, cDNA that corresponds to the CA125 antigen. A rabbit polyclonal antibody produced to purified CA125 antigen was used to screen a λZAP cDNA library from OVCAR-3 cells inEscherichia coli. The longest insert from the 54 positive isolated clones had a 5797-base pair sequence containing a stop codon and a poly(A) sequence but no clear 5′ initiation sequence. The deduced amino acid sequence has many of the attributes of a mucin molecule and was designated CA125/MUC16 (gene MUC16). These features include a high serine, threonine, and proline content in an N-terminal region of nine partially conserved tandem repeats (156 amino acids each) and a C-terminal region non-tandem repeat sequence containing a possible transmembrane region and a potential tyrosine phosphorylation site. Northern blotting showed that the level of MUC16 mRNA correlated with the expression of CA125 in a panel of cell lines. The molecular cloning of the CA125 antigen will lead to a better understanding of its role in ovarian cancer. monoclonal antibody polyacrylamide gel electrophoresis expressed sequence tag tandem repeat CA125 antigen is a serum marker that is used routinely in gynecologic practice to monitor patients with ovarian cancer. It is a mullerian duct differentiation antigen that is overexpressed in epithelial ovarian cancer cells and secreted into the blood, although its expression is not entirely confined to ovarian cancer. CA125 was first identified by Bast, Knapp, and colleagues (1Bast Jr., R.C. Feeney M. Lazarus H. Nadler L.M. Colvin R.C. Knapp R.C. J. Clin. Invest. 1981; 68: 1331-1337Crossref PubMed Scopus (1402) Google Scholar) in 1981 by a monoclonal antibody (OC125) that had been developed from mice immunized with an ovarian cancer cell line. These investigators subsequently developed a radioimmunoassay for the antigen and showed that serum CA125 levels are elevated in about 80% of patients with epithelial ovarian cancer but in less than 1% of healthy women (2Bast Jr., R.C. Klug T.L. St John E. Jenison E. Niloff J.M. Lazarus H. Berkowitz R.S. Leavitt T. Griffiths C.T. Parker L. Zurawski Jr., V.R. Knapp R.C. N. Engl. J. Med. 1983; 309: 883-887Crossref PubMed Scopus (1886) Google Scholar). Numerous studies since that time have confirmed the usefulness of CA125 levels in monitoring the progress of patients with epithelial ovarian cancer (3Bast Jr., R.C. Xu F.-J., Yu, Y.H. Barnhill S. Zhang Z. Mills G.B. Int. J. Biol. Markers. 1998; 13: 179-187Crossref PubMed Scopus (362) Google Scholar, 4Verheijen R.H. Von Mensdorff-Pouilly S. Van Kamp G.J. Kenemans P. Semin. Cancer Biol. 1999; 9: 117-124Crossref PubMed Scopus (71) Google Scholar, 5Menon U. Jacobs I.J. Curr. Opin. Obstet. Gynecol. 2000; 12: 39-42Crossref PubMed Scopus (102) Google Scholar, 6Meyer T. Rustin G.J. Br. J. Cancer. 2000; 82: 1535-1538Crossref PubMed Google Scholar). Most reports indicate that a rise in CA125 levels precedes clinical detection by about 3 months. During chemotherapy, changes in serum CA125 levels correlate with the course of the disease. CA125 is being used in our medical center (Memorial Sloan-Kettering Cancer Center) and elsewhere as a surrogate marker for clinical response in phase II trials of new drugs. On the other hand, CA125 is not useful in the initial diagnosis of epithelial ovarian cancer because of its elevation in a number of benign conditions (3Bast Jr., R.C. Xu F.-J., Yu, Y.H. Barnhill S. Zhang Z. Mills G.B. Int. J. Biol. Markers. 1998; 13: 179-187Crossref PubMed Scopus (362) Google Scholar, 7Meden H. Fattahi-Meibodi A. Int. J. Biol. Markers. 1998; 13: 231-237Crossref PubMed Scopus (89) Google Scholar). Despite this limitation, CA125 is considered to be one of the best available cancer serum markers. However, more information on its molecular nature is needed to fully explore its potential. Although CA125 antigen was first detected over 20 years ago, very little is known about its biochemistry and genetics. Most biochemical studies have concluded that CA125 is a high molecular mass glycoprotein, although estimates of its size range from 200 to 2000 kDa with smaller “subunits” being described by some investigators (8O'Brien T.J. Int. J. Biol. Markers. 1998; 13: 188-195Crossref PubMed Scopus (73) Google Scholar, 9Davis H.M. Zurawski V.R. Bast Jr., R.C. Klug T.L. Cancer Res. 1986; 46: 6143-6148PubMed Google Scholar, 10Matsuoka Y. Nakashima T. Endo K. Yoshida T. Kunimatsu M. Sakahara H. Koizumi M. Nakagawa T. Yamaguchi N. Torizuka K. Cancer Res. 1987; 47: 6335-6340PubMed Google Scholar, 11Nagata A. Hirota N. Sakai T. Fujimoto M. Komoda T. Tumour Biol. 1991; 12: 279-286Crossref PubMed Scopus (27) Google Scholar, 12de los Frailes M.T. Stark S. Jaeger W. Hoerauf A. Wildt L. Tumour Biol. 1993; 14: 18-29Crossref PubMed Scopus (21) Google Scholar, 13Kobayashi H. Ida W. Terao T. Kawashima Y. Am. J. Obstet. Gynecol. 1993; 169: 725-730Abstract Full Text PDF PubMed Scopus (14) Google Scholar). Most studies have shown that CA125 is a mucin-type molecule, but others have claimed that it is a typical glycoprotein with asparagine-linked sugar chains (14Zurawski Jr., V.R. Davis H.M. Finkler N.J. Harrison C.L. Bast Jr., R.C. Knapp R.C. Cancer Rev. 1988; 11–12: 102-118Google Scholar). Another study claimed that CA125 is a glycosylphosphoinositol-linked glycoprotein (11Nagata A. Hirota N. Sakai T. Fujimoto M. Komoda T. Tumour Biol. 1991; 12: 279-286Crossref PubMed Scopus (27) Google Scholar). Thus, no consensus emerged from these studies concerning the biochemical nature of this antigen. Recently, however, our studies have strongly indicated that CA125 is a typical mucin molecule with a high carbohydrate content and a preponderance of serine- and threonine-linked (O-linked) glycan chains (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar, 16Lloyd K.O. Yin B.W.T. Tumor Biol. 2001; 22: 77-82Crossref PubMed Scopus (37) Google Scholar). Possibly because of the mucinous nature of CA125 its peptide moiety has been very difficult to clone. The only published study on this topic (17Campbell I.G. Campbell I.G.,. Foulkes W.D. Senger G. Stamp G.W. Allan G. Boyers C. Jones K. Bast Jr., R.C. Solomon E. Hum. Mol. Gen. 1994; 3: 589-594Crossref PubMed Scopus (106) Google Scholar) described the isolation of a novel cDNA, later termed NBR-1 (18Chambers J.A. Solomon E. Genomics. 1996; 38: 305-313Crossref PubMed Scopus (24) Google Scholar), but this species does not seem to have any of the biochemical characteristics expected for CA125. Using a rabbit antiserum to purified CA125 we have now cloned, by expression cloning, a long partial cDNA sequence corresponding to a new mucin species (designated MUC16) that is a strong candidate for being the peptide core of the CA125 antigen. The NIH:OVCAR3 cell line was obtained from the American Type Culture Collection (Manassas, VA). Anti-CA125 antibody mAb1 OC125 was a generous gift from Dr. R. Bast, Jr. mAb VK-8, developed in our laboratory by immunization of mice with the human ovarian cancer cell line OVCAR-3, also identifies CA125 but reacts with a different epitope(s) than OC125 (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). Tumor cell lines were from the Sloan-Kettering Institute cell bank. CA125 was purified from the culture supernatant of NIH:OVCAR-3 cells in a simple two step procedure (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). Briefly, the cells were cultured as a monolayer in a synthetic medium (ITS, Life Technologies, Inc.) in RPMI medium containing 1% fetal bovine serum, and the culture medium was harvested every 7 days. Medium from 31 liters of supernatant medium was concentrated 10-fold and precipitated with perchloric acid (0.6 m final concentration). After centrifuging, the neutralized supernatant was passed through a column of normal mouse Ig-agarose (30 ml, 1.0 mg/ml) and then through a column of VK-8 mAb (80 ml, 2.0 mg/ml). The antibodies were linked to Actigel ALD gel according to the manufacturer's directions (Sterogene Bioseparations, Inc., Carlsbad, CA). The VK-8 column was washed at 4 °C with phosphate-buffered saline, then with 1 m NaCl in phosphate-buffered saline, and finally eluted with 3 mMgCl2. Fractions (6.0 ml) were collected and assayed for CA125 antigen by enzyme-linked immunosorbent assay with mAb VK-8 as described (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). Fractions from the MgCl2 eluate containing CA125 reactivity were pooled and used in subsequent studies. Analysis by SDS-PAGE and silver staining (Fig. 1) showed that the sample consisted of very high molecular weight components migrating in the stacking gel and in a region just below the gel interface; all of these species were reactive with mAb OC125 (data not shown). The sample also contained a lower molecular weight species originating from the fetal bovine serum used in the cell cultures. The amino acid content of the sample was determined as described previously (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). The CA125 sample was further purified by preparative SDS-PAGE, and the high molecular weight region of the gel indicated in Fig. 1 was excised. After homogenization in incomplete Freund's adjuvant the gel was used to immunize a rabbit (NZB white, female) by three subcutaneous injections, 1 week apart, in eight sites. Serum was obtained from the rabbit 10 days after the final immunization. An aliquot (3.0 ml) of the serum was absorbed with a pellet of melanoma cells (SK-MEL-28, -23, -30, and -33; 6.7 ml) that had been treated with 0.2% Nonidet P-40 and 0.1% protease inhibitor mixture (Sigma), and the absorbed serum was used to screen a cDNA library. A cDNA library was constructed from OVCAR-3 mRNA in the λZAP Express vector inEscherichia coli as described by the manufacturer (Stratagene, La Jolla, CA). The library contained 7.5 × 106 plaque-forming units. The library was plated onto 15 plates at ∼30,000 plaque-forming unit/150-mm plate, and plaques were transferred to nitrocellulose and screened with the absorbed rabbit antiserum (1:500). Positive plaques were identified using anti-rabbit Ig-horseradish peroxidase conjugate (Southern Biotechnology Assoc., Birmingham, AL) and 4-chloro-1-napthol reagent. After subcloning three times and retesting with antiserum, 54 positive clones remained. These clones contained inserts ranging from 1.5 to >4.0 kilobase pairs and were designated pBK-CMV-B1 to B54. The nucleotide sequence of the longest insert (B4) was determined using Big Dye terminators (PerkinElmer Life Sciences) and run on an ABI 3700 or ABI 377 DNA sequencer by the Cornell University BioResource Center, Ithaca, NY. Using the T3 primer and then a series of internal sequencing primers corresponding to less conserved regions of the gene, a 5797-base pair sequence was identified in B4. Partial sequencing of the other inserts demonstrated that the majority corresponded to different parts of the B4 sequence. mRNA was isolated from a panel of human tumor cell lines that had been serologically typed for CA125 expression, using an mRNA Isolation System kit (Invitrogen, Carlsbad, CA). mRNA samples (3 µg) were denatured with formaldehyde, separated by electrophoresis in 1.0% agarose, and transferred to nylon sheets (GeneScreen Plus, PerkinElmer Life Sciences). The blot was hybridized with a biotin-labeled probe from an insert containing three tandem repeat and the non-TR regions (B53) using a chemiluminescence procedure following the manufacturer's directions (Renaissance reagent, PerkinElmer Life Sciences). Tumor cell lines were assayed for CA125 expression with mAb OC125 and VK8 using a red cell rosetting method as described previously (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). Although most studies on the molecular cloning of mucins have utilized polyclonal antisera raised to the mucin in this study we used a rabbit antiserum the CA125 antigen. CA125 was purified by on an antibody column by conditions with a as described previously (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). The purified sample had an amino acid to that in other mucins and high CA125 × 106 of immunize the was further purified by SDS-PAGE, and gel containing high molecular weight CA125 antigen were used as the incomplete Freund's The antiserum was absorbed with a pellet of cancer cells after partially the cells in 0.2% Nonidet to of amino acid content of purified CA125 and deduced amino acid of CA125/MUC16 and its tandem repeat in a new The absorbed antiserum was used to screen a λZAP cDNA library from OVCAR-3 cells expressed in E. coli. positive clones were and 54 inserts were sequencing of the longest (B4) showed that it had nine partially conserved repeats of pairs and a sequencing with internal primers the of the sequence to include a stop a and a poly(A) region for a of pairs clear initiation sequence in a was detected at the that the sequence is The majority of the other inserts had from different parts of the B4 sequence. clones containing only were the no cDNA but human and and one mouse were these were to for B4. for the CA125 has been designated MUC16 by the of the B4 sequence with the of the human available from the on sequencing of this region is incomplete and of of more information the of further sequencing The nucleotide was into an amino acid sequence initiation at the of the in the The deduced amino acid sequence of amino acids (Fig. a mucin-type It had an amino acid that was high in and proline is very to that of the purified CA125 sample used in this study although the of these three amino acids is lower than in most other The sequence contained a region of nine tandem repeats of amino acids and a C-terminal region of amino of the nine repeats are but conserved and were (Fig. conserved the were The and are the but the regions have of and with is a of J. H. S. J. PubMed Scopus Google Scholar), acids and and acids These characteristics are typical of The high content of this however, is not in other features of include a sequence of amino acids the C-terminal a transmembrane and a acid region also a consensus tyrosine phosphorylation T. T. J.A. U. S. A. PubMed Scopus Google Scholar, J.A. T. J. Biol. Full Text PDF PubMed Google Numerous potential in the and non-TR regions (Fig. mRNA from a panel of 10 and cell lines was screened with a probe from of the cell lines positive and were The obtained is typical of that with other mucin These corresponded to the expression of CA125 antigen on the cell lines as determined by with antibodies to CA125 OC125 and The was by mRNA from OVCAR-3 the cell line from the CA125 was purified and the cDNA library was CA125 was by with at for 3 K.O. Yin B.W.T. P. H. Gen. 2000; Scopus Google Scholar). were obtained from a of the sample after SDS-PAGE and of the region to a nitrocellulose K.O. Yin B.W.T. P. H. Gen. 2000; Scopus Google Scholar). The was also with in were isolated by high and were of these corresponded to the and one to a sequence in the C-terminal region of the deduced MUC16 sequence acid from purified in CA125/MUC16 and other and other in a new on the following the is a strong candidate for the cDNA for the peptide core of the CA125 the CA125 antigen used in the study was isolated by on an monoclonal antibody column and was isolated from the purified CA125 sample corresponded to in the and MUC16 mRNA levels in a panel of cancer cell as determined by Northern correlated with the expression of CA125 in the cell as determined this biochemical studies that CA125 antigen is a mucin-type molecule (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). has been to of mucins are because of the size of mucin and not no 5′ initiation is in cDNA sequence. The sequence is to be at the because a stop a and a poly(A) have been identified (Fig. are difficult to because of and high of Most cloning have from cDNA with a polyclonal antiserum produced to the mucin in J. Cancer Rev. 1988; 11–12: Scholar, Jr., Am. Rev. 1991; Google Scholar, Rev. PubMed Scopus Google Scholar, E. C. S. C. A. A. E. 1997; PubMed Scopus Google Scholar, J. J. Biol. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar). human mucins have been or partially to and J. Cancer Rev. 1988; 11–12: Scholar, Jr., Am. Rev. 1991; Google Scholar, Rev. PubMed Scopus Google Scholar, E. C. S. C. A. A. E. 1997; PubMed Scopus Google Scholar, J. J. Biol. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar, Cancer Res. 1999; Scholar, M. J. Biol. 2001; Full Text Full Text PDF PubMed Scopus Google Scholar). this however, a polyclonal antiserum to the mucin was used to a cDNA corresponding to the peptide moiety of CA125/MUC16 antigen. have been because of the content of and in CA125/MUC16 in with most other The high of of the isolated as as the of a absorbed antiserum and the high expression of CA125 in the OVCAR-3 cell line used to the cDNA also have been in positive The deduced amino acid sequence of CA125/MUC16 other mucins in serine, threonine, and proline as amino However, its high content of is of The of tandem repeats is also typical of but the of the repeat (156 amino is with only tandem repeats E. Jr., J. Biol. 1993; Full Text PDF PubMed Google Scholar). have been identified with the repeat than the The amino acid in the are not although have conserved amino acids and and are in all or most of the amino acids in the are also and are the sequence but of and are in the These regions have or a that is in J. H. S. J. PubMed Scopus Google Scholar). sequence is also in the C-terminal non-TR Numerous potential is any amino acid are also in the two that are conserved in the It is however, that many of these are used as the content of glycan chains in purified CA125 is very (15Lloyd K.O. Yin B.W.T. Kudryashov V. Int. J. Cancer. 1997; 71: 842-850Crossref PubMed Scopus (93) Google Scholar). It is also to that the sequence and that are from the and this the of CA125 to K.O. Yin B.W.T. Tumor Biol. 2001; 22: 77-82Crossref PubMed Scopus (37) Google Scholar). for conserved in the the of in the deduced sequence. The of this is of the are in the tandem repeat and one is in the non-tandem repeat region acids were described as of with high levels of P. L. Scopus Google CA125/MUC16 this Recently, it has been that also regions to M. L. R. S. of the on Cancer Scholar). features of the non-TR region are is the of a acid of amino a have been in other mucins and The of the mucins that have been transmembrane regions and have regions with to J. J. Biol. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar). of this of mucins are secreted and that and epithelial CA125 is also secreted from ovarian and cell but the for its is be a in the C-terminal the N-terminal in L. M. J. J. Biol. Full Text PDF PubMed Google Scholar, M. L. J. S. J. Res. PubMed Scopus Google or are that the transmembrane sequencing of clones and the of (data not shown). The of in the non-TR sequence is a amino that a tyrosine phosphorylation sequence is conserved in the mouse that has with CA125/MUC16 at the C-terminal has tyrosine in its and at one of these is in S. A. 1994; PubMed Scopus Google Scholar, P. S. Cancer Res. Google Scholar). of the in in a a that is for to in in The phosphorylation in CA125/MUC16 was first in J. H. S. J. PubMed Scopus Google Scholar, T. T. J.A. U. S. A. PubMed Scopus Google Scholar). or not this tyrosine is in CA125 antigen is not Jr., T.J. Tumour Biol. 1997; PubMed Scopus Google Scholar) the of in CA125 from cells by and but concluded that the phosphorylation are on or however, the of CA125 is by through the is a known tyrosine Jr., T.J. Tumour Biol. 1997; PubMed Scopus Google Scholar). The that CA125/MUC16 is on tyrosine and is in further no are in some other mucins and were in The molecular cloning of CA125 antigen the to a better understanding of this its and its role in the of ovarian cancer. will be the of the epitope(s) by the monoclonal antibodies that CA125 K. Bast Jr., R.C. T.J. P. T. S. M. A. M. J. K. Sakahara H. Endo K. E. S. Kenemans P. J. Tumour Biol. 1996; PubMed Scopus Google Scholar). The of tandem repeats in the CA125/MUC16 is with the of a monoclonal antibody in for CA125 indicate that the antigen has (2Bast Jr., R.C. Klug T.L. St John E. Jenison E. Niloff J.M. Lazarus H. Berkowitz R.S. Leavitt T. Griffiths C.T. Parker L. Zurawski Jr., V.R. Knapp R.C. N. Engl. J. Med. 1983; 309: 883-887Crossref PubMed Scopus (1886) Google Scholar). studies lead to in the CA125 assay for the detection of ovarian cancer. We Dr. for on molecular cloning, for and for We also the at the Sloan-Kettering Institute H. and P. for amino acid Dr. (Memorial for CA125 and Dr. for the of