Original Article

Journal of Investigative Dermatology (1998) 111, 1034–1039; doi:10.1046/j.1523-1747.1998.00411.x

Anti-Tyrosinase-Related Protein-2 Immune Response in Vitiligo Patients and Melanoma Patients Receiving Active-Specific Immunotherapy

Tetsuro Okamoto, Reiko F Irie*, Shigeyuki Fujii, Sharon K S Huang, Anne J Nizze†, Donald L Morton† and Dave S B Hoon

  1. Department of Molecular Oncology Santa Monica, California, U.S.A
  2. *Department of Biotechnology Santa Monica, California, U.S.A
  3. †John Wayne Cancer Clinic, John Wayne Cancer Institute, Saint John's Health Center, Santa Monica, California, U.S.A

Correspondence: Dr Dave S. B. Hoon, Department of Molecular Oncology, John Wayne Cancer Institute, 2200 Santa Monica Blvd, Santa Monica, CA 90404

Received 22 April 1998; Revised 18 June 1998; Accepted 29 July 1998.

Top

Abstract

Several melanosome glycoproteins have been shown to be antigenic in humans. Correlation of antigen-specific immune responses in patients with the autoimmune disease vitiligo, therapy-induced hypopigmentation, and cutaneous melanoma has not been well studied. We examined antibody responses to a melanocyte autoantigen, tyrosinase-related protein-2 (TRP-2), as it is highly expressed in cutaneous melanoma and melanocytes. TRP-2 recombinant protein was synthesized for western blot and affinity anti-TRP-2 enzyme-linked immunosorbent assay. We demonstrated that patients with malignant melanoma, vitiligo, and active-specific immunotherapy-induced depigmentation had significant anti-TRP-2 IgG titers. The highest level of anti-TRP-2 IgG response was found in vitiligo patients. Induction and enhancement of anti-TRP-2 IgG responses were observed in melanoma patients treated with a polyvalent melanoma cell vaccine containing TRP-2. Active-specific immunotherapy could induce and/or augment the TRP-2 IgG antibody titers. Melanoma patients who developed hypopigmentation and had improved survival after polyvalent melanoma cell vaccine had significantly augmented anti-TRP-2 antibody responses compared with patients with poor prognosis. This study demonstrates that TRP-2 autoantigen is immunogenic in humans. TRP-2 antibody responses provide a linkage between autoimmune responses by vitiligo patients and melanoma patients responding to immunotherapy who have induced hypopigmentation.

Keywords:

autoantibodies, autoimmunity, human-clinical studies, tumor immunity

Abbreviations:

MAA, melanoma-associated antigen; MAH, melanoma-associated hypopigmentation; PMCV, polyvalent melanoma cell vaccine; TRP-1,2, tyrosinase-related protein 1 and 2

The melanogenesis biochemical pathway in melanocytes consists of a number of enzymes and substrates necessary to synthesize eumelanin. The major known enzymes in melanogenesis are tyrosinase, TRP-1 (5,6-dihidroxyindole-5-carboxylic acid oxidase), and TRP-2 (DOPAchrome tautomerase) (del Marmol & Beermann 1996). These proteins are also classified as melanocyte-specific differentiation antigens due to their primary restriction to melanocytes (Naftzger et al. 1996). There are other melanogenesis differentiation antigens such as gp100 and MART-1; however, their role in the melanogenesis pathway is unknown (Adema et al. 1994;Chen et al. 1996). Because melanogenesis proteins are consistently expressed in normal melanocytes and are highly expressed in melanoma, they have recently received increased attention as detection markers and potential targets of active-specific immunotherapy (Spagnoli et al. 1995;Zhai et al. 1996;Sarantou et al. 1997). There are two major diseases, vitiligo and cutaneous melanoma, in which the disease etiology evolves around melanocytes. Cutaneous melanoma is a malignant tumor derived from transformed melanocytes. Despite improvements in diagnosis and treatment for melanoma, patients who have distant metastasis have poor prognosis (Barth & Morton 1996). Vitiligo is a skin disorder of depigmented patches resulting from loss of melanocytes in the lesion and is thought to be due to autoimmunity against melanocytes (Naughton et al. 1983;Norris et al. 1988;Song et al. 1994). Autoimmunity is not clearly implicated in vitiligo; however, there is evidence implicating that this response is associated with known autoimmune diseases. Vitiligo can be psychologically debilitating to patients as a consequence of its cosmetic effect.

Activation and suppression of melanogenesis antigen-specific immune responses in vitiligo and melanoma patients, respectively, have been strongly suggested to be related to their disease etiologies (Berd et al. 1996). Melanoma patients receiving active-specific immunotherapy such as vaccines have been shown to have elevated antigen-specific immune responses that associate with more favorable prognosis (Jones et al. 1981;Livingston et al. 1994). Enhancement and induction of both antigen-specific antibody and cell-mediated immunity to melanoma-associated antigen (MAA) have been demonstrated in vaccinated patients (Morton et al. 1992;Hayashi et al. 1993;Hoon et al. 1995a). Vitiligo-like depigmentation called melanoma-associated hypopigmentation (MAH) has been reported in melanoma patients responding to immunotherapy (Mattes et al. 1983;Richards et al. 1992;Rosenberg 1997). The onset of MAH can also occur in patients having received therapy other than immunotherapy or in the absence of therapy. The mechanism of concomitant depigmentation is thought to involve immune responses against autoantigens found in melanomas and melanocytes. Both vitiligo and melanoma patients have been found to have antibodies to melanoma cells in their sera (Naughton et al. 1983). Appearance of MAH in melanoma patients who respond to treatment has been linked to a better prognosis (Mattes et al. 1983;Richards et al. 1992); however, linkage between immune responses to autoantigen in both diseases has not been well characterized.

Tyrosinase is one of the melanoma autoantigens and the presence of anti-tyrosinase antibody in sera of both melanoma and vitiligo patients has been demonstrated (Merimsky et al. 1996). Another well-characterized melanoma autoantigen is TRP-1 (Naftzger et al. 1996). Immune responses to TRP-1 have been demonstrated in melanoma patients (Thomson et al. 1985). Anti-TRP-1 responses in a murine melanoma model system have been linked to melanoma regression and autoimmunity (depigmentation) (Takechi et al. 1996).

TRP-2 is a relatively newly discovered melanogenesis synthesis related enzyme and has recently been characterized and cloned (Bouchard et al. 1994). TRP-2 is DOPAchrome tautomerase involved in the conversion of DOPAchrome to 5,6-dihydroxyindole-5-carboxylic acid (Bouchard et al. 1994). TRP-2 (peptide) was recently shown to be recognized by CTL from melanoma patients in vitro (Wang et al. 1996); however, no studies to our knowledge have assessed TRP-2 immunogenicity in vivo nor immune responses in melanoma and vitiligo patients.

In this study, TRP-2 recombinant protein was synthesized for assessment of anti-TRP-2 antibody responses in melanoma and vitiligo patients' sera. Recombinant TRP-2 protein allows detection of multiple antibody responses to multiple antigenic epitopes that are not defined. Melanoma patients receiving therapy with polyvalent melanoma cell vaccine (PMCV) expressing TRP-2 were evaluated for anti-TRP-2 antibody responses. The PMCV is a whole cell vaccine that has been previously characterized to induce significant antibody and T cell responses to multiple MAA, and has been shown to improve melanoma patient prognosis in a phase II clinical trial (Morton et al. 1992).

In this study, we demonstrated anti-TRP-2 antibody (IgG) responses in melanoma and vitiligo patients. Melanoma patients who developed MAH after PMCV treatment had significantly higher anti-TRP-2 antibody responses. The study demonstrates the significance of anti-TRP-2 immune responses in melanoma patients with possible linkage to autoimmunity in vitiligo patients. Most importantly the study demonstrates that the autoantigen TRP-2 is immunogenic in humans and related to improved prognosis.

Top

MATERIALS AND METHODS

Patients and sera

All melanoma patients (American Joint Committee on Cancer stage III) were clinically and pathologically diagnosed. After the surgical removal of tumor(s), melanoma patients were started with PMCV as previously described (Morton et al. 1992). Briefly, patients received PMCV consisting of three well-defined irradiated allogeneic melanoma cell lines (8 times 106 cells per injection) three times biweekly and then monthly for a year, followed by every 3–6 mo thereafter. PMCV was given intradermally at multiple sites in the axillary and inguinal region. BCG (Tice strain, 3 times 106 CFU) was given with PMCV in the first two treatments. Follow-up clinical and laboratory evaluations were repeated monthly, which included a chest X-ray every 2–3 mo and if necessary MRI/PET and/or CT scans every 6–12 mo. Melanoma patients who developed hypopigmentation (MAH) during PMCV treatment were diagnosed at the John Wayne Cancer Clinic. Blood was collected from melanoma, MAH, and vitiligo patients. Normal blood was collected from volunteers who were age- and sex-matched with melanoma and vitiligo patients. Sera were processed, aliquoted, and cryopreserved at –80°C until used as previously described (Hoon et al. 1995a). All sera were coded and tested in a blind fashion by the individuals performing the affinity anti-TRP-2 enzyme-linked immunosorbent assay (ELISA) analysis.

Cell lines

Melanoma cell line, M12, and PMCV lines 1, 2, and 3 were separately grown in RPMI 1640 with 10% fetal calf serum (heat inactivated) plus antibiotics (Hoon et al. 1993). A primary normal melanocyte line was obtained from Clonetics (San Diego, CA) and cultured in melanocyte growth medium according to the manufacturer's instruction. The breast cancer cell line, BT-20 (ATCC HTB19) was obtained form ATCC (Rockvillle, MD) and grown in Eagle's modified essential medium with 10% fetal calf serum (heat inactivated). When the cell lines attained 75% confluency, total RNA was isolated, purified, and prepared for reverse transcriptase polymerase chain reaction (RT-PCR) analysis as previously described (Doi et al. 1996).

Expression of recombinant TRP-2 protein

Recombinant human TRP-2 protein was genetically engineered into a plasmid vector and expressed in E. coli. Briefly, total RNA was isolated from the melanoma cell line M12, and cDNA was made by reverse transcription as previously described (Hoon et al. 1995b). Then the cDNA of TRP-2 (1356 bp,Bouchard et al. 1994) was amplified by PCR, and restriction enzyme sites were incorporated for cloning. The PCR product obtained was cloned into the expression vector pGEX/HIS for expression of glutathione S-transferase (GST) fusion protein and a 6x-HIS affinity tag. The cloned cDNA sequence of TRP-2 was verified by nucleic acid sequencing. The pGEX/HIS vector was reengineered from the pGEX-2T vector (Pharmacia, Piscataway, NJ) by inserting into the Sma-I site a DNA fragment for expression of a 6x-HIS affinity tag.

Production of His-tagged recombinant TRP-2 protein

The vector construct was transformed into E. coli BL21 strain and selected on ampicillin-containing plates. One clone, confirmed by DNA sequencing, was induced by isopropyl beta-D-thiogalactoside (IPTG) and purified from bacterial lysates by affinity chromatography using Glutathione Sepharose 4B (Pharmacia, Piscataway, NJ). The GST fusion recombinant TRP-2 protein was bound to the affinity column, cleaved in the column by thrombin digestion, and then eluted. Protein was further purified by gel electrophoresis and Ni2+-NTA affinity chromatography. This affinity purification was repeated twice. The yield of affinity-purified recombinant TRP-2 protein ranged from 4 to 5 mg per liter of bacterial culture. Protein concentration was measured using a BCA Protein Assay Kit based on the bicinchoninic acid procedure (Pierce, Rockford, IL).

Western blotting analysis

Purified recombinant TRP-2 was analyzed by western blotting using Ni2+-NTA conjugate (Qiagen, Chatsworth, CA) to verify the presence of 6x-HIS tag, and an anti-GST antibody conjugate (Pharmacia) was used to confirm the absence of GST tag. The respective controls were used in assessment of the proteins in western blotting.

Western blotting was performed as previously described (Okamoto et al. 1997). Briefly, 5 mug of purified TRP-2 was run on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel and blotted to nitrocellulose membrane. Patients' sera were diluted to 1:50 or 1:100 and incubated for 1 h using a multiscreen apparatus (BioRad, Hercules, CA). This was followed by incubation with 1:1000 dilution of alkaline phosphatase-conjugated anti-human IgG (gamma-chain specific) or anti-human IgM (mu-chain specific) (Boehringer, Indianapolis, IN), and then developed by using 5-bromo-4-chloro-3-indolyl-1-phosphate and nitro blue tetrazolium (Promega, Madison, WI) according to the manufacturer's procedure. As a control, recombinant HOJ-1 protein, a 15 kDa tumor-associated protein with a 6x-HIS tag, was expressed in E. coli, induced, isolated, and purified in a similar manner to TRP-2 (Hoon et al. to be published). No IgG or IgM antibody response was detected against HOJ-1 protein by western blotting analysis using anti-TRP-2 positive sera and at the same protein concentration of TRP-2. Anti-TRP-2 positive patients' sera were also assessed for anti-MAGE-3 antibodies using recombinant MAGE-3 protein (Okamoto et al. 1997). MAGE-3 recombinant protein with a 6x-HIS tag was purified in a similar manner. Anti-MAGE-3 IgG and IgM antibodies were found infrequently and only at very low antibody dilutions (data not shown). These analyses indicated that the anti-TRP-2 responses were not directed towards the 6x-HIS tag or potential bacterial contaminating molecules.

Affinity ELISA for TRP-2 antibody analysis

Affinity ELISA was performed with purified recombinant TRP-2 protein containing a 6x-HIS tag on the N-terminus as reported previously (Okamoto et al. 1997). Recombinant protein was incubated overnight at room temperature in Ni2+ chelate-coated ELISA microwell plates (Xenopore, Hawthorne, NJ) according to the manufacturer's instructions. After blocking with specific buffer, sera were added in 2-fold dilutions in phosphate-buffered saline from 1:40 to 1:1280 and incubated for 2 h at room temperature in duplicate. Goat anti-human IgG (gamma-chain specific) or IgM (mu-chain specific) horseradish peroxidase-conjugate were used to detect antibodies, respectively. ELISA was developed and read using a Molecular Device ELISA Reader and analyzed with the instruments' software (San Francisco, CA). Antibody titers were defined as the highest serum dilution yielding an absorbance reading greater than the mean plus 2 SD of healthy donor volunteer background absorbance. Respective positive controls (positive serum samples) and negative controls (no serum, antihuman IgG, or IgM conjugate alone) were carried out. Assays were performed at least twice for verification of results.

RT-PCR analysis

The RT-PCR assay was carried out as previously described (Sarantou et al. 1997). One microgram of RNA was reverse transcribed using oligo (dT)15 primers. Specific pairs of TRP-2 primers (sense, 5'-GAGGTGCGAGCCGACACAAG-3'; anti-sense, 3'-CGGTGCCAGGTAACAAATGC-5') were used in the PCR reaction at 95°C for 5 min for 1 cycle; 95°C for 1 min; 60°C for 1 min; 72°C for 1 min for 35 cycles; and 72°C for 10 min. Specific PCR cDNA product was assessed by using 2% agarose gel electrophoresis and ethidium bromide. The predicted size of the amplified gene for TRP-2 was 476 bp. Positive and negative controls were incorporated for the RT-PCR assays. beta-actin RT-PCR was carried out on all samples, as a control for the presence of intact mRNA.

Statistical analysis

Affinity ELISA data were analyzed by the Wilcoxon rank sum test to compare the difference between diseases and by a generalized McNemar test to compare before and after PMCV treatment data (Fleiss 1981). p values < 0.05 (two-sided) were considered significant.

Top

RESULTS

The purified recombinant TRP-2 protein had an estimated Mr of 55–57 kDa as analyzed by gel electrophoresis (sodium dodecyl sulfate-polyacrylamide gel electrophoresis). This is similar to the predicted molecular mass of the protein:translation product (estimated at 52 kDa) of the native protein with 452 amino acids. Analysis of protein size by gel electrophoresis is usually slightly larger compared with the predicted size due to protein charge, amino acid composition, and migration ability in the gel. Purified recombinant TRP-2 protein allowed us to evaluate anti-TRP-2 antibody responses of patients' sera using western blot and affinity ELISA.

A representative western blot of anti-TRP-2 antibody response is shown in Figure 1. Anti-TRP-2 IgG antibodies were not detected in sera from 21 healthy donors at 1:50 dilution; however, seven of 15 melanoma, seven of 12 MAH, and 10 of 15 vitiligo patients were shown to have anti-TRP-2 IgG in their sera. Similar analyses were performed for detection of anti-TRP-2 IgM. No anti-TRP-2 IgM antibody was detected in 15 melanoma and 15 vitiligo patients by western blot analysis at 1:50 dilution. Therefore, in the study we focused on analysis of anti-TRP-2 IgG responses.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Western blot analysis for the detection of anti-TRP-2 IgG. Western blotting was performed with purified recombinant TRP-2 protein and patients' sera at a dilution of 1:100 with phosphate-buffered saline. Representative examples of healthy donors, melanoma patients, and vitiligo patients are shown. Lanes 1–4, healthy donor volunteers;lanes 5–8, melanoma patients;lanes 9–12, melanoma patients with MAH after PMCV treatment;lanes 13–16, vitiligo patients.

Full figure and legend (33K)

Specificity of the anti-TRP-2 response against recombinant TRP-2 protein was verified using control recombinant proteins with 6x-HIS tag (HOJ-1 and MAGE-3) that were prepared in an identical manner. Anti-TRP-2 positive sera did not have any significant reactivity to the control recombinant proteins, therefore indicating that the potential reactivity to 6x-HIS or bacterial contaminants was not present. To determine whether the anti-TRP-2 antibody response could recognize melanoma native TRP-2 and not an artifact response to recombinant TRP-2 conformation, we performed several specific adsorption studies. High titer anti-TRP-2 patient sera was adsorbed with melanoma tissue containing abundant TRP-2 protein and compared with adsorption with a TRP-2 negative tissue colon carcinoma tumor biopsy. The latter does not express TRP-2 as assessed by mRNA expression using RT-PCR. Adsorption with a highly melanotic melanoma tumor specimen significantly reduced (>70%) anti-TRP-2 antibody reactivity (anti-TRP-2 affinity ELISA), whereas absorption with control colon tissue did not significantly affect anti-TRP-2 antibody reactivity. This suggested that the anti-TRP-2 IgG antibodies in patients' sera can recognize and be adsorbed out with native TRP-2.

Analysis of anti-TRP-2 antibody in melanoma, MAH, and vitiligo patients

To assess anti-TRP-2 IgG response in a large number of patients, we developed an affinity anti-TRP-2 ELISA. As shown in Table 1, 98% (34 of 35) of healthy donors were negative at 1:40 serum dilution. Anti-TRP-2 IgG responses in 40% of melanoma, 55% of MAH, and 67% of vitiligo patients' sera were positive, and significantly higher than in normal sera (p = 0.0001, <0.0001, and <0.0001, respectively). Healthy donor volunteers were age- and sex-matched with vitiligo and melanoma patients. No significant trend was determined based on these parameters in vitiligo or melanoma patients. The data also indicated that TRP-2 protein is antigenic in humans with vitiligo or malignant melanoma. Vitiligo patients' anti-TRP-2 IgG titers were significantly higher than in melanoma patients (p = 0.007). The highest titer of anti-TRP-2 antibody response observed was 1:640 in four vitiligo patients. No significant difference was found between melanoma patients with MAH and vitiligo (p = 0.151), or melanoma patients with or without MAH (p = 0.138). The highest anti-TRP-2 IgG titer observed in melanoma patients with or without MAH was 1:320. These data demonstrated that melanoma patients with MAH have a similar antibody response against TRP-2 protein, as do vitiligo patients.


TRP-2 mRNA expression in the PMCV cell lines

PMCV cell lines were assessed for expression of TRP-2 by RT-PCR using a specific primer set for the TRP-2 gene. A normal melanocyte primary culture line mRNA, used as a positive control, was shown to express TRP-2 by RT-PCR analysis Figure 2. All PMCV cell lines also produced a strong PCR cDNA product of 476 bp, and the negative control nonmelanogeneic cell line BT-20 did not. Previously, we have demonstrated that TRP-2 mRNA is expressed in >80% of melanomas and in 50% of hypopigmented melanomas (Sarantou et al. 1997). These data show that all PMCV cell lines express abundant TRP-2 mRNA.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Representative analysis of RT-PCR analysis of TRP-2 expression by melanoma cell lines.Lane 1, normal melanocyte cell line;lane 2, PMCV line 1;lane 3, PMCV line 2;lane 4, PMCV line 3;lane 5, negative control, breast cancer cell line BT-20. TRP-2 PCR cDNA product was 476 bp. Control beta- actin RT-PCR was positive for all samples (201 bp).

Full figure and legend (24K)

Anti-TRP-2 antibody responses in PMCV-treated melanoma patients

In comparisons among antigen-specific IgG antibody induction after PMCV treatment, we have consistently observed that an initial peak of antibody response usually occurs around 6–16 wk after several PMCV treatments and then usually plateaus (Morton et al. 1992). Patients entering the PMCV protocol were clinically free of disease. To assess antibody responses in melanoma patients, we used the following criteria: patients entered in the study were American Joint Committee on Cancer stage III and had had tumors surgically removed to become a clinical status of no evidence of disease (NED). The study criteria for patients were the following clinical qualifications: NED at the start of PMCV, NED after 16 wk (four PMCV treatments), and no evidence of disease recurrence in clinical follow-up at week 16 of treatment. In addition, patients were categorized into the following two groups: group A, patients who survived longer than 5 y; group B, patients who died within 2 y due to recurrence of metastatic melanoma.

As shown in Figure 3(a) 43% (nine of 21) of the melanoma patients in group A had anti-TRP-2 IgG antibody pre-PMCV treatment. At 16 wk post-PMCV treatment, 52% (11 of 21) of the patients had anti-TRP-2 IgG. The anti-TRP IgG level 16 wk after PMCV treatment was significantly (p = 0.008) higher than pre-PMCV treatment. In group B melanoma patients Figure 3(b), 38% (eight of 21) had anti-TRP-2 IgG antibodies at the start of PMCV, and after 16 wk, 43% (nine of 21) had positive titers. In 29% of patients, antibody levels were elevated by PMCV, but three patients' antibody titers had dropped. The comparison of pre- and post-PMCV in group B was not significant (p = 0.254). There were no significant differences in distribution of anti-TRP-2 antibody levels between groups before PMCV treatment (p = 0.754). The data show that PMCV treatment can augment anti-TRP-2 IgG antibody response in some melanoma patients, and the response is different between the patients with good outcome and the patients with poor outcome. Not all melanoma patients showed induction or enhancement of TRP-2 IgG antibody after PMCV treatment. Overall, in both poor and good outcome groups (n = 42 patients), 15 (36%) had elevated titers of antibody. In the PMCV group with poor prognosis, three patients had a drop in antibody titer of one dilution Figure 3(b). In general the anti-TRP-2 antibody response was elevated one titer dilution in patients who had an antibody response to the PMCV.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Comparison of anti-TRP-2 antibody titers before and after PMCV treatment in melanoma patients. Individual patients' titers of anti-TRP-2 IgG were assessed before (day 0) and after (16 wk) PMCV treatment. Before and after treatment results of individuals are connected by a line to demonstrate changes of titer in a patient: (a) good prognosis group patients receiving PMCV; (b) poor prognosis group patients receiving PMCV; and (c) induced MAH in melanoma patients receiving PMCV.

Full figure and legend (33K)

Anti-TRP-2 IgG response in PMCV-treated patients with MAH

As shown in Figure 3(c), melanoma patients who developed MAH during PMCV treatment were assessed for anti-TRP-2 IgG levels by affinity ELISA. Blood was collected from 25 patients at the start of PMCV and 16 wk after treatment. In 48% (12 of 25) of patients, anti-TRP-2 IgG was detected before PMCV treatment. After 16 wk PMCV treatment, 56% (14 of 25) of patients had anti-TRP-2 IgG. Six of 12 anti-TRP-2 positive MAH patients at pre-PMCV treatment developed higher levels of antibody after PMCV treatment. Two of the patients had a drop in anti-TRP-2 antibody titers. These two patients had an antibody titer of 1:40 and 1:320 that dropped to 0 and 1:160, respectively, representing a one antibody titer dilution drop. Four of six anti-TRP-2 antibody positive MAH patients after PMCV showed no change in antibody titers. Three of the MAH patients were negative at pre-PMCV analysis and had induced anti-TRP-2 antibody at titers of 1:320, 1:320, and 1:80, respectively, after PMCV treatment. The overall TRP-2 antibody titers in MAH patients were significantly (p = 0.03) elevated (nine of 12 patients) after PMCV treatment. Patients (n = 25) have been observed to develop MAH at the specific intradermal site (n = 1) of PMCV injection, and in the areas adjacent of the PMCV injection sites (axillary, groin) after several weeks follow-up. One patient was observed to develop MAH at the PMCV injection site as well as in the legs adjacent to the injection site Figure 4a. In general MAH is often seen at the nearby lymphatic drainage area of the PMCV injection site. A representative example of a patient developing well-defined MAH after multiple PMCV treatment is shown in Figure 4(b); however, the anatomical sites of MAH induction after PMCV treatment can vary with no predicted specific site localization. The induction of MAH is strongly indicative of the PMCV ability to induce autoimmune responses that can affect melanocytes in the skin nearby the injection site.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Representative photographs induced MAH of two different patients after PMCV treatment. (a) Representative example of intradermal site of PMCV injection. Multiple PMCV injection sites at groin region all show MAH after several weeks. (b) Representative example of a patient with induced MAH on both legs after more than five PMCV treatments.

Full figure and legend (146K)

Top

DISCUSSION

This study demonstrates the immunogenicity of TRP-2 protein in humans, and observations of anti-TRP-2 antibodies in melanoma and vitiligo patients. The study also demonstrates that immune responses to TRP-2 in melanoma patients can be enhanced by active-specific immunotherapy with PMCV expressing TRP-2. Data analysis suggests that anti-TRP-2 IgG responses may be involved in the pathogenic mechanism of vitiligo. The augmentation of anti-TRP-2 antibody in melanoma patients with MAH and favorable survival outcomes after active-specific immunotherapy suggests common autoantigen-specific immune responses in melanoma and vitiligo.

The presence of TRP-2 antibody in the sera of both melanoma and vitiligo patients suggests that similar immune response(s) can be inducible in both diseases by the same antigen. There is a number of MAA that have been shown to be immunogenic in humans (Hoon & Irie 1997). Although T cell responses to MAA have received considerable attention, there are limited studies linking antigen-specific T cell responses consistently to significant improved survival of melanoma patients beyond predicted natural disease course. On the contrary, antibodies to MAA have been shown to be linked to favorable prognosis (Jones et al. 1981;Livingston et al. 1994). The melanosomal glycoproteins are one of the most abundant and consistently expressed autoantigens by melanoma cells and melanocytes. Recently, these autoantigens have received considerable attention as potential target antigens for antigen-specific immunotherapy.

Investigators have been interested in the phenomenon of depigmented skin lesions in melanoma patients treated with immunotherapy (Richards et al. 1992;Fishman et al. 1997;Rosenberg 1997). This is a potential indication that significant "effector" autoimmunity has been generated towards melanocyte autoantigens. Although at this time the pathologic significance of the antigen-specific autoimmunity is still unclear. Induced MAH can be a potential indicator of the effectiveness of the immunotherapy (Richards et al. 1992;Rosenberg 1997). There are a limited number of studies documenting consistent MAH in patients clinically responding to active-specific immunotherapy (Richards et al. 1992;Merimsky et al. 1996;Fishman et al. 1997;Rosenberg 1997). In our study some melanoma patients had anti-TRP-2 antibody before treatment, indicating that melanoma patients can generate anti-TRP-2 antibody during the natural course of disease. Only one healthy donor was shown to have anti-TRP-2 antibody response. This suggests that in general the level of natural antibody response to TRP-2 is very low in healthy individuals. The anti-TRP-2 responses in melanoma and vitiligo patients were IgG, suggesting that helper T cell function is also involved. Although the role of anti-TRP-2 IgG in the pathogenesis of vitiligo, MAH, or melanoma is not known, T helper cells may be directly or indirectly involved. T helper cells when antigen-activated release specific cytokines or directly interact with B cells and help in IgG production. The anti-TRP-2 IgG response may be an indicator of the immune response associated with the pathogenesis of melanocyte/melanoma destruction and continual release of TRP-2, as opposed to a direct effector mechanism. The mechanisms involved in immune tolerance breakdown to self-antigens in adults are not well understood. Vitiligo can be divided into active and inactive (stable) disease, whereby the former will have more autoantibodies present in the sera (Harnig et al. 1991). In melanoma patients tumor cell death would relate to above normal autoantigen release and potential induction of autoantibodies.

The anti-TRP-2 antibody responses were the highest in vitiligo patients, MAH melanoma patients treated with PMCV, and melanoma patients treated with PMCV with no MAH, respectively. Interestingly, the patients with MAH had similar levels of antibody responses as in the vitiligo patients. Previous studies have shown that anti-melanoma cell antibodies (Naughton et al. 1983) and anti-tryosinase antibodies (Song et al. 1994;Merimsky et al. 1996;Fishman et al. 1997) were present in vitiligo patients. Our study demonstrates that another melanoma/melanocyte autoantigen TRP-2 immune response may be related to autoimmunity associated with vitiligo. TRP-1 has been indicated to be also involved in autoimmunity associated with vitiligo in a murine model but has not been demonstrated in humans (Song et al. 1994). Tyrosinase, TRP-1, and TRP-2 are associated with melanosomes in melanocytes/melanoma cells and are considered as cytoplasmic autoantigens (Orlow et al. 1993). Recently, TRP-1 has been demonstrated to be expressed on the cell surface of melanocytes (Takechi et al. 1996). TRP-2 remains to be determined if it is also on the cell surface of melanoma cells. The role of autoimmune responses to these antigens in the etiology of vitiligo and MAH is not well understood.

There may be a de novo mechanism by which antibodies can recognize intracellular autoantigens directly via cell surface or cytoplasmic penetration, and participate in cell destruction. Antibodies have been shown to be taken up by nonhemopoietic cells and react with nuclear antigens in vivo (Isenberg et al. 1997). Antibody uptake is a potential mechanism whereby anti-TRP-2 antibodies may play a role in regulating or destroying melanocyte/melanoma cell growth and function. The induction of the autoimmune response to these autoantigens may be a secondary effect after the destruction of melanocytes/melanoma cells by some other physiologic event such as cytotoxic T cell lysis of melanocytes/melanoma cells. There are also suggestions that cells destroyed by apoptosis have limited release of intracellular proteins, whereas destruction of cells by immune attack tends to release more intracellular debris and is more likely to induce an immune response (Griffith & Ferguson 1997). The presence of the anti-TRP-2 antibody may represent an epiphenomenon of an ongoing T cell response against melanocyte/melanoma antigens. The presence of antibody responses may represent a breakdown of tolerance to antigen due to excess antigen release or a change in antigen presentation by the host. The destruction of melanocytes/melanoma cells may trigger both antibody and T cell responses, whereby the role of the former in the etiology and pathology of the disease is not known. The antibody may have an immunosurveillance function towards this intracellular antigen in that it is responsible for clearing out this shed autoantigen to prevent induction of autoimmunity.

Not all melanoma patients showed elevated levels of anti-TRP-2. Five of the patients treated with PMCV had a drop of one antibody titer dilution. Three of these patients were in the poor prognosis PMCV group. The drop in antibody titer could be related to antigen-specific tolerance or suppression. The poor prognosis patient group had recurrence of disease, whereby excess of tumor antigen may lead to immune suppression or tolerance. Alternatively, tumor antigen may be shed in these patients and antibodies binding to these shed antigens form immune complexes. The formation of immune complexes in active disease with shed antigens often reduces antibody titers. No drop in antibody titers was observed in the good prognosis PMCV group, suggesting the PMCV was not inducing antigen-specific suppression.

Interestingly, PMCV was able to enhance anti-TRP-2 antibody responses in 24 of 67 patients (36%), and was more prevalent in patients with improved survival. Immune recognition of PMCV in patients can be via direct recognition of MAA presented by HLA molecules on the PMCV cell surface, or degradation of melanoma cells and presentation of proteins by antigen-presenting cells to T cells (Hoon et al. 1998). A peptide sequence of TRP-2 has been shown to be recognized by cytotoxic T cells in a HLA-restricted manner (Wang et al. 1996). Recent investigations have also found that a murine TRP-2 peptide can be used to induce CTL activity (Bloom et al. 1997). Although these results suggest that a TRP-2 peptide is antigenic in humans in vitro, the immunogenicity of TRP-2 in patients was not demonstrated. Other melanosomal-related glycoproteins such as MART-1, gp100, tyrosinase, and TRP-1 have been shown to be antigenic in humans (Hoon & Irie 1997).

TRP-2 may be a good target antigen for active-specific immunotherapies. The detection of antibodies to TRP-2 in melanoma patients with induced MAH and in vitiligo may provide important clues on the relationship of the two disease etiologies. It is interesting that autoantibodies to melanocyte autoantigens are present in active vitiligo; however, all areas of the body are not affected. Further studies are needed to identify the etiology related to destruction and protection of melanocytes during autoimmunity. The protective mechanism may be of similar physiologic relevance to how melanomas can escape tumor destruction in the presence of strong antiautoantigen immunity.

Top

References

  1. Adema, GJ, de Boer, AJ, Vogel, AM, Leonen, Wam, Figdor, CG: Molecular characterization of the melanocyte lineage-specific antigen gp100. J Biol Chem 1994 269:20126–20133,
  2. Barth, A, Morton, DL: The role of adjuvant therapy in melanoma management. Cancer 1996 75:726–734,
  3. Berd, D, Mastrangelo, MJ, Lattime, E, Sato, T, Maguire, HC: Melanoma and vitiligo: immunology's Grecian urn. Cancer Immunol Immunother 1996 42:263–267,  | Article | PubMed | ISI | ChemPort |
  4. Bloom, MB, Perry-Larry, D, Robbins, PF, Li, Y, El-Gamil, M, Rosenberg, SA, Yang, JC: Identification of tyrosinase-related protein 2 as a tumor rejection antigen for the B16 melanoma. J Exp Med 1997 185:453–459,  | Article | PubMed | ISI | ChemPort |
  5. Bouchard, B, del Marmol, B, Jackson, IJ, Cherif, D, Dubertret, L: Molecular characterization of a human tyrosinase-related protein-2 cDNA patterns of expression in melanocytic cells. Eur J Biochem 1994 219:127–134,  | Article | PubMed | ISI | ChemPort |
  6. Chen, YT, Stockert, E, Jungbluth, A, Tsang, S, Coplan, KA, Scanlan, MJ, Old, LJ: Serological analysis of Melan-A (MART-1), a melanocyte-specific protein homogeneously expressed in human melanomas. Proc Natl Acad Sci USA 1996 93:5915–5919,  | Article | PubMed | ChemPort |
  7. Doi, F, Chi, Ddj, Charuworn, BB, Conrad, AJ, Russell, J, Morton, DL, Hoon, DSB:Detection of beta;-human chorionic gonadotropin mRNA as a marker for cutaneous malignant melanoma. Int J Cancer 1996 65:454–459,
  8. Fishman, P, Merimski, O, Baharav, E, Shoenfeld, Y: Autoantibodies to tyrosinase, the bridge between melanoma and vitiligo. Cancer 1997 79:1461–1464,  | Article | PubMed | ISI | ChemPort |
  9. Fleiss, J: The analysis of data from matched samples In: Statistical Methods for Rates and Proportions 1981 New York, John Wiley, pp. 112–132,
  10. Griffith, TS, Ferguson, TA: The role of Fas-L induced apoptosis in immune privilege. Immunol Today 1997 18:240–244,  | Article | PubMed | ISI | ChemPort |
  11. Harnig, R, Cui, J, Bystryn, JC: Relation between the incidence and level of pigment cell antibodies and disease activity in vitiligo. J Invest Dermatol 1991 97:1078–1080,
  12. Hayashi, Y, Hoon, DSB, Foshag, LJ, Park, MS, Terasaki, PI, Morton, DL:A preclinical model to assess the antigenicity of an HLA-A2 melanoma cell vaccine. Cancer 1993 72:750–759,
  13. Hoon, Dsb, Irie, RF:Current status of human melanoma vaccines. Biodrug 1997 7:66–84,
  14. Hoon, Dsb, Okamoto, T, Wang, H-J, et al: Is the survival of melanoma patients receiving polyvalent melanoma cell vaccine linked to the human leukocyte antigen phenotype of patients? J Clin Oncol 1998 16:1430–1437,
  15. Hoon, Dsb, Wang, Y, Dale, PS, et al: Detection of occult melanoma cells in blood with a multiple-marker polymerase chain assay. J Clin Oncol 1995b 13:2109–2116,  | PubMed | ISI | ChemPort |
  16. Hoon, Dsb, Wang, Y, Sze, L, et al: Molecular cloning of a human monoclonal antibody reactive to ganglioside GM3 antigen on human cancers. Cancer Res 1993 53:5244–5250,  | PubMed | ISI | ChemPort |
  17. Hoon, Dsb, Yuzuki, D, Hayashida, M, Morton, DL Melanoma patients immunized with melanoma cell vaccine induce antibody response to recombinant MAGE-1 antigen. J Immunol 1995a 154:730–737,  | PubMed | ISI | ChemPort |
  18. Isenberg, DA, Rahman, MA, Ravirajan, CT, Kalsi, JK: Anti-DNA antibodies: from gene usage to crystal structures. Immunol Today 1997 18:149–153,
  19. Jones, PC, Sze, LL, Liu, PY, Morton, DL, Irie, RF: Prolonged survival for melanoma patients with elevated IgM antibody to oncofetal antigen. J Natl Cancer Inst 1981 66:249–254,
  20. Livingston, PO, Wong, Gyc, Adluri, S, et al: Improved survival in stage III melanoma patients with GM2 antibodies: a randomized trial of adjuvant vaccination with GM2 ganglioside. J Clin Oncol 1994 12:1036–1044,  | PubMed | ISI | ChemPort |
  21. del Marmol, V, Beermann, F: Tyrosinase related proteins in mammalian pigmentation. FEBS Lett 1996 381:165–168,  | Article | PubMed | ISI | ChemPort |
  22. Mattes, MJ, Thomson, TM, Old, LJ, Lloyd, KO: A pigmentation-associated, differentiation antigen of human melanoma defined by a precipitation antibody in human serum. Int J Cancer 1983 32:717–721,
  23. Merimsky, O, Baharav, E, Shoenfeld, Y, Chaitchik, S, Tsigelman, R, Cohen-Aloro, D, Fishman, P: Anti-tyrosinase antibodies in malignant melanoma. Cancer Immunol Immunother 1996 42:297–302,
  24. Morton, DL, Foshag, LJ, Hoon, Dsb, et al: Prolongation of survival in metastatic melanoma after active specific immunotherapy with a new polyvalent melanoma vaccine. Ann Surg 1992 216:463–482,  | PubMed | ISI | ChemPort |
  25. Naftzger, C, Takechi, Y, Kohda, H, Hara, I, Vijayasaradhi, S, Houghton, AN: Immune response to a differentiation antigen induced by altered antigen: a study of tumor rejection and autoimmunity. Proc Natl Acad Sci USA 1996 93:14809–14814,  | Article | PubMed | ChemPort |
  26. Naughton, GK, Eisinger, M, Bystryn, JC: Antibodies to normal human melanocytes in vitiligo. J Exp Med 1983 158:246–251,  | Article | PubMed | ISI | ChemPort |
  27. Norris, DA, Kissinger, RM, Naughton, GM, Bystryn, JC: Evidence for Immunologic mechanisms in human vitiligo: patients' sera induce damage to human melanocytes in vitro by complement-mediated damage and antibody-dependent cellular cytotoxicity. J Invest Dermatol 1988 90:783–789,  | Article | PubMed | ISI | ChemPort |
  28. Okamoto, T, Kaneda, Y, Yuzuki, D, Huang, Sks, Chi, Ddj, Hoon, DsB: Induction of antibody response to human tumor antigens by gene therapy using a fusigenic viral liposome vaccine. Gene Ther 1997 4:969–976,  | Article | PubMed |
  29. Orlow, SJ, Boissy, RE, Moran, DJ, Pifko-Hirst, S: Subcellular distribution of tyrosinase and tyrosinase-related protein-1: implications for melanosomal biogenesis. J Invest Dermatol 1993 100:55–64,  | Article | PubMed | ISI | ChemPort |
  30. Richards, JM, Mehta, N, Ramming, K, Skosey, P: Sequential chemoimmunotherapy in the treatment of metastatic melanoma. J Clin Oncol 1992 10:1338–1343,  | PubMed | ISI | ChemPort |
  31. Rosenberg, SA Cancer vaccines based on the identification of genes encoding cancer regression antigens. Immunology Today 1997 18:175–182,
  32. Sarantou, T, Chi, Ddj, Garrison, DA, Conrad, AJ, Schmid, P, Morton, DL, Hoon, DsB: Melanoma-associated antigens as messenger RNA detection markers for melanoma. Cancer Res 1997 57:1371–1376,  | PubMed | ISI | ChemPort |
  33. Song, YH, Connor, E, Li, Y, Zorovich, B, Balducci, P, Maclaren, N: The role of tyrosinase in autoimmune vitiligo. Lancet 1994 344:1049–1052,  | Article | PubMed | ISI | ChemPort |
  34. Spagnoli, GC, Schaefer, C, Willimann, TE, et al: Peptide-specific CTL in tumor-infiltrating lymphocytes from metastatic melanomas expressing MART-1/Melan-A, gp100 and Tyrosinase genes: a study in an unselected group of HLA-A2.1-positive patients. Int J Cancer 1995 64:309–315,  | PubMed | ISI | ChemPort |
  35. Takechi, Y, Hara, I, Naftzger, C, Xu, Y, Houghton, AN: A melanosomal membrane protein is a cell surface target for melanoma therapy. Clin Cancer Res 1996 2:1837–1842,  | PubMed | ISI | ChemPort |
  36. Thomson, TM, Mattes, MJ, Roux, L, Old, LJ, Lloyd, KO: Pigmentation-associated glycoprotein of human melanomas and melanocytes: a definition with a monoclonal antibody. J Invest Dermatol 1985 85:169–174,
  37. Wang, RF, Appella, E, Kawakami, Y, Kang, X, Rosenberg, SA: Identification of TRP-2 as a human tumor antigen recognized by cytotoxic T lymphocytes. J Exp Med 1996 184:2207–2216,  | Article | PubMed | ISI | ChemPort |
  38. Zhai, Y, Yang, JC, Kawakami, Y, et al: Antigen-specific tumor vaccines. Development and characterization of recombinant adenovirus encoding MART-1 or gp100 for gene therapy. J Immunol 1996 156:700–710,  | PubMed | ISI | ChemPort |
Top

Acknowledgments

We thank the nursing staff at the John Wayne Cancer Clinic for their kind cooperation, V. Chiang for technical assistance, and Dr. Hejing Wang (UCLA School of Medicine, Department of Biostatistics) for her help in the biostatistical analysis. This study was supported by grants POI CA12582 and PO1 CA1038 from the National Cancer Institute NIH and by funding from the Roy E. Coats Research Foundation, Los Angeles.

Extra navigation

.
ADVERTISEMENT