Identification of Makorin 1 as a novel SEREX antigen of esophageal squamous cell carcinoma

This study was approved by the Ethic Committee of Chiba University, Graduate School of Medicine. Collection of tissues and sera was agreed upon by patients and healthy donors, after provision of written, informed consent. Recombinant DNA work was done with official permission and in accordance with the rules of the government of Japan. The human esophageal SCC cell line, T.Tn, was established in the Department of Clinical Molecular Biology, Chiba University, Graduate School of Medicine [22, 23]. Total RNA was prepared from T.Tn cells by the acid guanidinium thiocyanate-phenol-chloroform method [24] and purified to poly(A)⁺RNA using Oligotex-dT₃₀ (Super) mRNA Purification Kit (Takara Biochemicals, Kyoto, Japan) according to the manufacturer's instructions. cDNA was ligated into EcoRI-XhoI site of the λZAP II phage. The original library size was 1.8 × 10⁶.

Sera were obtained from 73 esophageal cancer patients with various stages of disease progression, as well as from 16 breast, 16 gastric, and 16 colorectal cancer patients prior to treatment. The patients with esophageal carcinoma consisted of 66 men (82%) and 7 women (18%), with a median age of 65 years (range 39 to 86 years). No treatment-related death was observed in these patients. They were pathologically classified according to the TNM/UICC classification [25] using resected specimens. The patients included 14 with stage I cancer, 21 with stage II, 27 with stage III, and 11 with stage IV cancer (Table 1). All patients underwent R0 resection with extended lymphadenectomy. After surgery, they were followed until the end of March 2006 or their deaths with clinical examinations and imaging studies on a regular basis. The median follow-up for survivors was 30 months. Each sample was centrifuged at 3,000 g for 5 min and then frozen at -80°C until use. Repeated thawing and freezing of samples was avoided.

Serum CEA and CYFRA21-1 concentrations were measured with Enzymun-TEST CEA and Enzymun-TEST CYFRA21-1 (Boehringer Mannheim, Mannheim, Germany), respectively. SCC-Ag levels were measured in serum with an immunoradiometric Assay kit (Dynabott, Tokyo, Japan). The cutoff values for serum CEA, CYFRA21-1, and SCC-Ag levels were 4.6 ng/ml, 2.57 ng/ml and 1.5 ng/ml, respectively, in accordance with the manufacturer's instructions. The specificity at these cut-off values is 95%.

Esophageal carcinoma antigens were screened by SEREX as previously published by Sahin et al. [8]. Escherichia coli XL1-Blue MRF' was infected with λZAP II phage which contained a cDNA library, and the expression of cDNA was induced by blotting on nitrocellulose membranes (NitroBind, Osmonics Inc., Minnetonka, MN), which had been pretreated with 10 mM isopropyl β-D-thiogalactoside (IPTG, Wako, Kyoto, Japan) for 30 min. The membranes were then washed three times with TBS-T [20 mM Tris-HCl (pH 7.5), 0.15 M NaCl and 0.05% Tween-20], and blocking was performed by treatment with 1% protease-free bovine serum albumin (Wako) in TBS-T for 1 h. The membranes were exposed in 1:2000-diluted serum for 1 h. After washing with TBS-T three times, the membranes were treated with 1:5000-diluted alkaline phosphatase-conjugated Fc fragment-specific goat anti-human IgG and making it react for 1 h. Positive reactions were detected by incubation in color development solution [100 mM Tris-HCl (pH 9.5), 100 mM NaCl and 5 mM MgCl₂] containing 0.3 mg/ml of nitroblue tetrazolium chloride (Wako) and 0.15 mg/ml of 5-bromo-4-chloro-3-indolyl-phosphate (Wako). Positive clones were recloned twice to obtain monoclonality and retested for serum reactivity.

Monoclonalized phage cDNA clones were converted to pBluescript phagemid by in vivo excision using the ExAssist helper phage (Stratagene, La Jolla, CA). Plasmid DNA was obtained from Escherichia coli SOLR strain transformed by the phagemid. The cDNA inserts were sequenced by the dideoxy chain termination method using the DNA Sequencing kit BigDyeTM Terminator (Applied Biosystems, Foster City, CA) and ABI PRISM 3700 DNA Analyzer (Applied Biosystems). Sequences were analyzed for homology with public databases of known genes and proteins using NCBI-BLAST.

Escherichia coli JM109 cells which contained cDNA clones recombined in pBluescript II were cultured with or without 0.1 mM IPTG for 2.5 h. Cells were then washed with phsphate-buffered saline and lysed by incubation at 100°C for 3 min in SDS sample buffer [26]. Escherichia coli lysate was then subjected to SDS-PAGE followed by western blotting using sera of 73 patients with esophageal cancer, 14 with breast cancer, 9 with gastric cancer, and 16 with colorectal cancer.

The chi-square test or Fisher exact test (2-sided) was performed. The survival probabilities were calculated using the product-limit method of Kaplan and Meier, considering all deaths. Survival differences between groups were determined using the log-rank test. Fisher's exact probability test was used to determine the significance of the differences between two groups. All statistical analyses were carried out using the Stat View 5.0 program for Windows (SAS Institute Inc., Cary, NC). P values at the 0.05 level were considered statistically significant.

Total cellular RNA was isolated from the T.Tn esophageal SCC cell line using AquaPure RNA Isolation Kit (Bio-Rad, Hercules, CA), or from the tumor tissues using FastPure RNA Kit (Takara Bio, Otsu, Japan). RNA of normal esophageal keratinocytes was purchased from Cybrdi (Frederick, MD). Reverse transcription was performed with an oligo(dT)₂₀ primer using the ThermoScript RT-PCR System (Invitrogen). The presence of MKRN1 transcripts was examined via PCR amplification using the following primers: sense, 5'-AACGGGTGCCTCAGTGTCCTT-3'; antisense, 5'-CTCCTGATGTTGTGGCTGTTG-3'. β-actin-specific PCR primers (sense, 5'-ACCACAGCTGAGAGGGAAATC-3'; antisense, 5'-AGCACTGTGTTGGCATAGAGG-3') were used as a loading control. PCR was performed using KOD-Plus-DNA polymerase (Toyobo, Osaka, Japan) as follows: an initial denaturation step at 94°C for 3 min, followed by 33 cycles for MKRN1 or 25 cycles for β-actin of denaturation at 94°C for 15 sec, annealing at 60°C for 30 sec, and extension at 68°C for 30 sec, with a final extension of 5 min at 68°C.

Activated Ha-ras-transformed NIH3T3 mouse fibroblasts (clone F25) (ras-NIH3T3) [27, 28] were cultured in Dulbecco's modified Eagle's medium (Nissui Pharmaceutical, Tokyo, Japan) supplemented with 5% heat-inactivated bovine serum. Eukaryotic expression vector pME18S-FL3 containing human MKRN1 cDNA was purchased from Toyobo (Osaka, Japan). ras-NIH3T3 cells (5 × 10⁴ cells) were co-transfected with MKRN1 cDNA and neo gene using Lipofectamine reagent (Invitrogen) as described [28, 29]. After culture in the presence of G418 (400 μg/ml) for two weeks, surviving colonies were isolated.

Two peptides consisting of amino acids between 104 and 118 (CRYEHSKPLKQEEAT) and of amino acids between 399 and 412 (EPQRQKVGTSSRYR) from the MKRN1 protein were synthesized and called MKRN1-K1 and K2, respectively. The peptide was coupled to keyhole limpet haemocyamine that had been activated with N-succinimidyl-3-maleinimido-proprionate. To obtain antibodies, rabbits (bastards) were immunized by subcutaneous injections of 300 μg of coupled peptide emulsified in Freund's complete adjuvant. The immunization was repeated twice at 2-week intervals with the same amount of coupled peptide in Freund's incomplete adjuvant. The titer was tested after each bleed by enzyme-linked immunosorbent assay (ELISA).

Peptide-specific antibodies were enriched by using the pretested rabbit serum columns (Amersham Biosciences, Piscataway, NJ) according to the manufacturer's instructions. In brief, 6 mg of the MKRN1 peptide were coupled to the NHS-activated Sepharose column. Six ml of the rabbit serum were diluted 1:10 with 10 mM Tris-HCl (pH 7.5) and the diluted serum was applied twice to the equilibrated column. Antibodies were eluted with 100 mM glycine buffer (pH 2.5). Fractions were immediately neutralized with 1 M Tris-HCl (pH 8.0). The final concentration of the purified serum was 1:1.83.

Extracts of MKRN1-transfected ras-NIH3T3 cells were immunoprecipitated with anti-ubiquitin antibody (Santa Cruz) followed by SDS-polyacrylamide gel electrophoresis as described previously [30]. The gel was stained with silver (Invitrogen) and the gel pieces containing 80- and 82-kDa proteins were washed with water and dehydrated with acetonitrile. After removing acetonitrile, the gel piece was dried in a vacuum centrifuge, supplied with 10 mM dithiothreitol in 100 mM ammonium bicarbonate, and incubated for 1 h at 56°C. The solution was replaced by the same volume of 55 mM iodoacetamide in 100 mM ammonium bicarbonate and incubated for 45 min at room temperature in the dark. The gel piece was washed with 100 mM ammonium bicarbonate, dehydrated with acetonitrile, rehydrated with 100 mM ammonium bicarbonate, dehydrated again with acetonitrile, and dried in a vacuum centrifuge following elimination of acetonitrile. The dried gel piece was immersed in trypsin solution containing 10 μg/ml trypsin in 50 mM ammonium bicarbonate and 5 mM calcium chloride, incubated for 45 min at 4°C, and further at 37°C for 12 h. Peptides were extracted using sequential steps of 20 mM ammonium bicarbonate, followed by 5% formic acid in 50% acetonitrile. The extract was dried in a vacuum centrifuge, supplied with 0.1% trifluoroacetic acid and desalted using ZipTip C18 (Millipore) according to the manufacturer's protocol. The peptides were eluted with 0.1% trifluoroacetic acid in 50% acetonitrile and applied onto MALDI plate.

Mass spectrometry of peptides was performed using MALDI-TOF MS AXIMA-CFR (Shimadzu Corporation, Tsukuba, Japan). Protein identification was performed using Mascot search (Matrix Science, http://​www.​matrixscience.​com/​).

A total of approximately 2 × 10⁷ pfu from the T.Tn cDNA library were screened using serum samples derived from 21 patients with esophageal SCC. One of the isolated clones, 15J1-3, showed strong homology with the makorin1 (MKRN1; NCBI accession number: NM_013446) gene. MKRN1 appeared to be of much interest because its expression is much higher in the brain and testis than other tissues [31]. A recent report revealed that MKRN1 is involved in the differentiation of various tissues including osteoblasts, eye epithelial cells and kidney [32].

The existence of antibodies against the MKRN1 gene product was examined with 73 sera of patients with esophageal SCC and 43 sera of healthy donors by Western blot analysis. In order to confirm that the antigen reacting with serum antibodies derives from the insert cDNA, protein extracts of IPTG-treated Escherichia coli bearing MKRN1 cDNA expression plasmids or the control empty plasmids were compared with those of non-treated bacteria. The 50-kDa band which appeared only in MKRN1-expressing bacteria in response to IPTG was judged as positive. Figure 1 shows the representative results of s-MKRN1-Abs-negative serum of a healthy donor, #1, and s-MKRN1-Abs-positive serum of a patient, #15. For this 50-kDa band, 18 of 73 (25%) sera of patients with esophageal SCC were sero-positive (data not shown). In contrast, all 43 healthy donors were sero-negative. Similarly, positive bands were detected in the sera of two of 14 (14%) breast cancer patients, none of the 9 (0%) gastric cancer patients and six of the 16 (38%) colon cancer patients (data not shown).

We then examined the relationship between the presence of s-MKRN1-Abs and the clinicopathological features of the patients (Table 1). As for histopathological grading, well-differentiated tumors (G1) were associated with the presence of s-MKRN1-Abs significantly compared to poorly differentiated tumors (G3), and vice versa (P = 0.008). Nodal involvement was more frequently observed in the patients with s-MKRN1-Abs than in those without s-MKRN1-Abs, yet the difference was not statistically significant (P = 0.194). There was no correlation between the presence of s-MKRN1-Abs and other features such as age, gender, location, tumor sizes, depth of tumor, distant metastasis and stages. Kaplan-Meier plotting showed no significant correlation between s- MKRN1-Abs and survival (Figure 2).

When the patients were divided into two groups according to the presence of s-MKRN1-Abs, there was no significant correlation between the presence of s-MKRN1-Abs and the positivity of conventional serum tumor markers (Table 1). Consequently, the sensitivity of serum markers to detect esophageal SCC was improved by using s-MKRN1-Abs in combination with conventional markers. For example, the positivity increased from 36% of SCC-Ag alone to 51 % by combination of s-MKRN1-Abs and SCC-Ag.

The expression levels of MKRN1 mRNA in esophageal tissue specimens of SCC or normal cells were examined by RT-PCR analysis. T.Tn esophageal SCC cells, as a positive control, showed higher MKRN1 mRNA levels than normal esophageal keratinocytes (Figure 3). MKRN1 mRNA levels in carcinoma tissues were higher than those in the normal counterparts of patient numbers 2, 6, 7, 8 and 10. Thus, this elevated expression may account for the development of s-MKRN1-Abs.

To examine the biological role of MKRN1, human MKRN1 cDNA was recombined in pcDNA3-Myc, His and transfected into ras-NIH3T3 cells as described [28]. The protein (Figure 4a) and mRNA (Figure 4b) of transduced MKRN1 were observed in a stable transfected clone, Fmkrn-24, but not in the parental ras-NIH3T3 cells. The phase morphology of Fmkrn-24 was markedly different from the parental cells; i.e. enlargement of cell bodies with flat and extended cell periphery (Figure 4c). Because E3 ubiquitin-protein ligase activity was suspected to MKRN1 [31, 33], ubiquitinated proteins were examined by Western blotting using anti-ubiquitin antibody. Two ubiquitinated proteins of 80 and 82 kDa were clearly observed in two MKRN1 transfectants but not in the parental ras-NIH3T3 cells (Figure 4a). These two proteins in Fmkrn-24 cells were immunoprecipitated with anti-ubiquitin antibody, separated by SDS-polyacrylamide gel electrophoresis and analyzed by mass spectrometry. Six tryptic polypeptides were completely matched to the sequence of mouse L-FILIP (filamin A interacting protein 1) (Figure 5). This suggests that the ubiquitination of L-FILIP was induced by MKRN1.