| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
* Microbiology/Immunology and
Otolaryngology-Head and Neck Surgery, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Papillary thyroid cardcinoma (PTC) is a relatively rare cancer, with an incidence of 3 of 100,000 (5); however, it is the most prevalent endocrine tumor, comprising nearly 80% of all thyroid malignancies (1). Nearly 60% of differentiated thyroid carcinomas of the epithelium, including PTC, is characterized by the frequent expression of one or more members of a family of unique oncogenic fusion proteins resulting from chromosomal inversions or translocations (6, 7). In a majority of PTC, the tyrosine kinase domain (TKD) of the c-RET or tropomyosin receptor kinase proto-oncogenes and the amino-terminal portion of one of several widely expressed genes are juxtaposed to form a hybrid protein with ectopic constitutive tyrosine kinase activity in the cytosol of thyroid epithelial cells (3). The family of 10 fusion genes associated with c-RET rearrangement is collectively known as RET/PTC. One member of this fusion gene family, RET/PTC3 (RP3), is formed following a paracentric inversion involving c-RET and RFG/ARA70 (8, 9). Of note, RP3 can be found in thyroid tissue from patients with the chronic thyroid inflammatory disease, Hashimoto’s thyroiditis (10, 11). Likewise, somatic rearrangement of c-RET has been observed in irradiated human thyroid tissue grafts in SCID mice, supporting the notion that ionizing radiation is an etiologic agent for thyroid cancer (12). Tumor progression to advanced stages (poorly differentiated or undifferentiated thyroid cancer) in humans and in animal models does not correlate with persistent RET/PTC expression (4, 13, 14), a finding inconsistent with other models of multistep carcinogenesis (15).
In human cancers, mutant proteins involved in cellular transformation harbor structural abnormalities, including rearranged coding regions, internal deletions, point mutations, or viral proteins resulting from insertional mutagenesis. Altered gene products encoding novel fusion oncoproteins may represent candidate tumor-specific Ags capable of eliciting immune recognition and may be ideal targets for tumor immunotherapy because their expression is required for transforming activity. Although published data suggest a role for fusion gene products in human tumor-specific immune responses (16, 17, 18, 19), there is no clear understanding of whether fusion proteins represent immunogenic epitopes enabling tumor rejection.
The relationship between lymphocytic infiltrates and thyroid cancer, particularly PTC, in humans is well documented (20, 21). The observation that lymphocytic infiltration of the thyroid coexists with cancer suggests that Ags expressed by tumor cells are responsible for inducing an antitumor immune response (22). We hypothesized that reactivity to transformed thyroid cells or occult tumors may lead to oncoprotein-reactive lymphocyte accumulation in the thyroid. In this study, we have investigated the immunogenic capability of the RP3 fusion protein when presented in the presence of a strong adjuvant or when expressed in preneoplastic thyroid tissue or invasive thyroid tumor grafts to define its role as an antigenic nonself protein and thyroid-specific tumor Ag. Although the RP3 protein represents the fusion of two widely expressed self proteins to which immunological tolerance is most likely induced, we find that fusion confers a unique immunogenic capability upon the RP3 protein. The results of these studies may help to define novel strategies for the immunotherapeutic targeting of oncoproteins at early stages of neoplastic progression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RP3 transgenic mice (3209 line) have been previously described (23) and have subsequently been backcrossed for >10 generations to endotoxin-resistant C3H/HeJ mice using polymorphism selection. Homozygous RP3 mice were crossed with C57BL/6-Tpr53 mice and double heterozygotes crossed to generate RP3p53-/- mice (4). Progeny were genotyped using RP3- and p53-specific primers in a standard DNA amplification reaction (PCR), as previously described (4). Six- to 8-wk-old C57BL/6J-Prkdcscid/SzJ (C57BL/6 SCID) mice (The Jackson Laboratory, Bar Harbor, ME) were used for the growth of tumor tissue and to determine specimen viability. Female C3H/HeJ and male B6C3F1/J, 6–8 wk old, were obtained from The Jackson Laboratory.
Expression plasmid construction
Human (h) or mouse (m) RP3, mouse RetTKD, or control coding sequence T cell locus 1 (Tcl1) were cloned into the pET-29a+ protein expression vector (Novagen, Madison, WI) adjacent to a COOH-terminal hexahistidine tag to facilitate protein purification, and a selectable kanamycin marker to aid in positively selecting transfectants. The hRP3 gene sequence was amplified from the construct previously described for the generation of RP3 transgenic mice (23). The amino-terminal gene segment of mouse Rfg/Ara70 (AF159461) and mouse RetTKD (AF209436) were amplified using endonuclease restriction site-linked gene-specific primers (SalI-linked Rfg/Ara70 forward primer, GTC/GAC/GTC/GAC/ATG/AAC/ACA/TCC/CTG/GAA/CAG/AGT/GG; BamHI-linked Rfg/Ara70 reverse primer, CGG/GAT/CCT/CCT/GAC/TAT/TCT/CTG/AAG/TCT/GTT/TTG/G; BamHI-linked Ret forward primer, CTG/GAT/CCA/AAG/TGG/GAA/TTT/CCT/CGG/AAG/AAC; and NotI-linked Ret reverse primer, GTC/GCG/GCC/GCT/TAG/CTA/TCA/AAT/GTG/TCC/ATT/AAT/TTT/GC). Partial and complete BamHI digestion of Rfg/Ara70 and c-RET genes, respectively, was required for proper ligation of full-length RP3. All constructed plasmids were verified by sequence analysis.
Recombinant proteins and peptides
To generate recombinant protein, plasmid constructs were introduced into BLR (DE3) pLysS-competent cells (Novagen, Madison, WI), and protein expression was induced in kanamycin-resistant cells using isopropyl B-D-thiogalactopyranoside at a final concentration of 1 mM for 4 h at 37°C. Cells were lysed in denaturing buffer (8 M urea, 100 mM NaH2PO4, 10 mM Tris-Cl, pH 8.0), cellular debris were removed by ultracentrifugation at 45,000 rpm for 30 min at 4°C, and hexahistidine affinity-tagged proteins first bound to a Ni-NTA agarose column (Qiagen, Valencia, CA) were then eluted under denaturing conditions using an imidazole gradient. A second round of column chromatography was performed again using a Ni-NTA agarose column to ensure high protein purity. Recombinant proteins were dialyzed against PBS, 0.22 µm filtered, and analyzed for the presence of the 68-kDa rhRP3, the 75-kDa rmRP3, the 45-kDa rmRetTKD, or the 14-kDa rTcl1 protein (negative control) on SDS-PAGE by Coomassie blue staining and, in parallel, by Western blot analysis using polyclonal rabbit anti-RET (C-19) and (C-20) (Santa Cruz Biotechnology, Santa Cruz, CA) or anti-Tcl1 Abs, as previously described (23). Proteins were verified to be endotoxin free using the endotoxin neutralizing protein from Limuluspolyphemus (Sigma-Aldrich, St. Louis, MO). An RP3 peptide (KQTLENSQEDPKWEF) specific for the fusion breakpoint region was synthesized by the Thomas Jefferson University Peptide Synthesis Facility. This sequence was chosen based on its high MHC-predicted binding coefficient, as measured by an established algorithm (http://syfpeithi.bmiheidelberg.com/scripts/MHCServer.dll/home.htm). Peptide scores >20 were predicted to have significant binding potential.
Viral preparation
Recombinant vaccinia virus was propagated, as previously described (24). The RP3 construct was cloned into a modified pSC11 plasmid for expression from the vaccinia P7.5 promoter. RP3 gene sequence was introduced into the vaccinia genome by homologous recombination in CV-1 cells, and recombinant viruses were subsequently triple plaque purified in 143B cells in the presence of 5 mg/ml 5-bromo-2'-deoxyuridine (Boehringer Mannheim, Indianapolis, IN). Viral stocks were expanded and titered using 143B HuTK- cells.
Immunization protocols
Female C3H/HeJ mice, RP3 transgenic mice, or nontransgenic littermates were s.c. immunized in the hind flank with 10 µg rhRP3 or rmRP3, 5 µg rmRetTKD, or 10 µg rTCL (control) protein emulsified in CFA (Sigma-Aldrich). This injection was followed 1–2 wk later by a second injection of the same material into the same mice (10 µg of rRP3, 5 µg rmRetTKD, or 10 µg rmTcl1 control protein); however, this injection used protein emulsified in IFA (Sigma-Aldrich). For specified experiments, additional controls included injections of PBS in CFA, followed by PBS in IFA (nonimmune mice). One to two weeks following final immunization, lymph node cells (LNC; mandibular, superficial cervical, axillary, lateral axillary, superficial inguinal, mesenteric, and popliteal nodes), spleen cells, and serum were harvested for analysis. For tumor transplant immunization, anesthetized male 6- to 8-wk-old B6C3F1/J or C57BL/6 SCID control mice received 3-mm3 RP3p53-/- thyroid tumor fragments. Mice were monitored each week for signs of tumor rejection (including reduction in tumor size or stabilization of tumor growth). After 2 mo, LNC and serum were harvested from mice that had rejected tumor fragments.
Lymphoproliferation assays
Single cell suspensions of LNC and/or spleen cells from rRP3-, tumor-, or nonimmunized mice were prepared in RPMI 1640 medium (Invitrogen, Carlsbad, CA) containing 25 µM 2-ME, 200 U/ml of penicillin, 200 U/ml of streptomycin, 10 mM L-glutamine, and 10% FBS. For T cell proliferation assays, cells were cultured in 96-well plates (2 x 105 cells/well) to which recombinant proteins, RP3-specific peptide, or bacterial control eluates were added at a previously determined optimal concentration of 1 µg/ml. Plates were cultured in a 5% CO2 humidified incubator for 48 h at 37°C and incubated with 1 µCi of [3H]thymidine/well for additional 18 h. Wells were harvested onto nylon filters, and radioactivity was quantified using scintillant in a gamma counter (Beckman Instruments, Palo Alto, CA). Results are expressed in cpm (mean of quadruple wells).
Enzyme-linked immunosorbent assay
Antisera were collected from rhRP3-immune, RP3p53-/- tumor-immune, and nonimmunized mice by collecting blood from the retroorbital sinus, followed by clotting and centrifugation at 14,000 rpm for 5 min. Serum was removed and stored at 4°C. In ELISA, 250 ng of rhRP3 protein was adsorbed overnight to each well of a 96-well microtiter plate at 4°C. Serial dilutions of sera were added to protein-coated wells and incubated at room temperature for 2 h. Plates were washed and incubated with biotin-conjugated horse anti-mouse Ab (1:1000; Vector Laboratories, Burlingame, CA), followed by streptavidin-alkaline phosphatase (1:4000; Southern Biotechnology Associates, Birmingham, AL) and subsequently p-nitrophenyl phosphate solution. Product detection was performed at 405 nm.
Immunoprecipitation assay
A 1/10 dilution of RP3 immune or nonimmune mouse antisera in immunoprecipitation buffer (2% Triton X-100, 5 mM Tris-HCl (pH 7.5), 10 mM NaCl) was bound to protein A-Sepharose beads (Amersham Pharmacia Biotech, Piscataway, NJ) overnight with 10 µg of rhRP3, and washed twice with immunoprecipitation buffer and twice with 10 mM Tris-HCl, pH 7.5. Equivalent amounts of rhRP3 protein in 1x SDS loading buffer was resolved using multiple lanes of a 8% SDS-PAGE gel and transferred to Hybond ECL nitrocellulose membrane (Amersham Pharmacia Biotech) for 100 V/h. Membranes were blocked for 1 h in 5% milk-TBST (20 mM Tris-HCl (pH 7.6), 135 mM NaCl, and 1% Tween 20) and cut into individual strips corresponding to the lanes on the original SDS-PAGE gel. Each strip was incubated with a 1/1000 dilution of goat anti-RET (C-19) polyclonal Ab (Santa Cruz Biotechnology) in 5% milk-TBST. After incubation, membranes were treated with HRP-labeled anti-goat IgG (ECL reagent kit; Amersham Pharmacia Biotech) for 30 min at room temperature, according to the manufacturer’s protocol. Substrate was added from the ECL reagent kit (Amersham Pharmacia Biotech) and exposed to x-ray film for 1–5 min.
Histological analysis
Thyroid tissues were removed, fixed in 10% Formalin for 
24 h,
and desiccated. Following desiccation, tissue samples were embedded in
liquid paraffin and cooled. Paraffin-embedded tissue was cut into
6-µM sections and placed on silanized slides (Fisher Scientific,
Pittsburgh, PA). After deparaffinizing, the sections were rehydrated
through decreasing concentrations of xylene and alcohol and microwaved
for 15 min in 100 mM citrate buffer (pH 6.0). Slides were then stained
with H&E, dehydrated, fixed, and mounted.
Statistical analysis
Grouped data were compared and analyzed for statistical significance using an independent samples t test (Analyze-it, version 1.65, program add-in for Microsoft Excel; Analyze-It Software, Leeds, U.K.). Groups were considered different when calculated p values were less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RP3 is a somatically activated, unique protein that consists of two fused self proteins that may encode immunogenic determinants after intracellular Ag processing. To determine the immunogenicity of the RP3 fusion protein, naive C3H/HeJ female mice were immunized with small amounts (10 µg) of purified rhRP3 protein in CFA. Two weeks following immunization, LNC were harvested from rhRP3-immunized and mock-immunized mice and cultured with rhRP3 or control proteins to assess T cell proliferative responses in vitro. LNC from naive mice immunized with rhRP3 proliferated in response to rhRP3, but not to an unrelated control protein purified using an identical purification methodology (Fig. 1A). Flow cytometric analysis identified the predominant proliferating cell population as CD4+ (not shown). Consistent with these data, cell proliferation was significantly inhibited using anti-class II I-Ak or I-Ek Ab, but not with an isotype control antiserum (Fig. 1B). Because the novel amino acid sequence of the hRP3 fusion region represents a unique nonself amino acid sequence, a synthetic 15-mer peptide corresponding to this region and selected based on its MHC-binding potential was used for in vitro stimulation. Interestingly, lymphocytes from rhRP3-immunized mice responded only mildly to this synthetic peptide, suggesting that sequences other than the unique fusion region may constitute the natural immunodominant epitopes of RP3 (Fig. 1A). In contrast to rhRP3-immunized mice, LNC from control C3H/HeJ female mice failed to proliferate significantly above background in response to either rhRP3 protein or to the fusion region peptide (not shown).
|
LNC proliferation and cross-reactivity following immunization with mRP3 protein
Because rhRP3 protein immunization of mice induced reactivity to whole RP3 fusion protein, but not fusion region peptide, the possibility existed that the observed RP3 reactivity was a product of a xenogenic response. Although the amino acid sequence of hRP3 and its respective counterparts expressed in mouse are 92% homologous, the immunogenic properties of hRP3 protein in mice may result from the remaining 8% amino acid differences. To address this concern, a mRP3 fusion gene was engineered because such fusions are not known to exist in mouse thyroid tissue. Using specific PCR primers, the amino-terminal portion of the mouse Rfg/Ara70 gene and the TKD of the c-RET proto-oncogene were amplified from adult C3H/HeJ spleen and salivary gland cDNAs, respectively. Subsequently, the two products were juxtaposed and cloned, in frame, into a bacterial expression vector. Because the c-RET tyrosine kinase domain (RetTKD) of both mouse and human forms of RP3 contain the highest scoring-predicted immunogenic epitopes, the gene sequence identical with the nonmutated RetTKD segment of the mRP3 fusion gene was also cloned using the same vector system. rmRP3 and rmRetTKD proteins were produced and purified, and their expression was confirmed by Western blot analysis using a human-mouse cross-reactive RetTKD-specific Ab (Fig. 2A).
|
To dissect the components of RP3 immunogenicity, we engineered a self protein that is identical with the c-RET portion of the mRP3 molecule (rmRetTKD). This region also contains a predicted immunogenic self epitope, but lacks the unique fusion region segment of RP3. We next compared the immunogenicity of the RP3 fusion protein to RetTKD. Interestingly, rmRP3-immune LNC were capable of recognizing the rmRetTKD protein, as measured by lymphocyte proliferation. In addition, proliferation of both rmRP3 and rhRP3-immune LNC occurred in rmRetTKD-stimulated cultures, although to a lesser extent than when these cultures were stimulated with whole mRP3 or hRP3 protein (Fig. 2B). Neither immune nor nonimmune lymphocytes proliferated well to control recombinant protein. These data indicated that an immunogenic self epitope(s) is encoded within the TKD portion of the RP3 molecule and that the unique breakpoint-specific amino acid sequence may not be essential for anti-RP3 immunity.
Tolerance to the RetTKD self protein
The above experiments demonstrate the immunogenicity of the mouse RetTKD when presented as a portion of the fused RP3 protein. Because the RetTKD is a self protein to which mice should be tolerant, we investigated whether immunization under conditions that would distinguish self from nonself proteins by injecting mice with this nonfused domain of mRP3 is capable of evoking an immune response. Naive mice were immunized with either whole rmRP3 or rmRetTKD protein in adjuvant; LNC were harvested and cultured; and in vitro proliferation assays performed. As observed previously, LNC from rmRP3 fusion protein-immunized mice responded strongly to both rmRP3 and rhRP3 stimulation and to a lesser extent to rmRetTKD protein (Figs. 2 and 3). In contrast, LNC from rmRetTKD-immunized mice failed to proliferate in response to stimulation with rmRP3, or the immunizing protein, rmRetTKD (Fig. 3). LNC from neither immunized mice proliferated when stimulated with a protein control. Given these results, we predicted that only when the RP3 fusion protein was synthesized as a self protein would tolerance be observed.
|
RP3 is the product of two fused self proteins to which T cell tolerance has most likely been induced; however, the above experiments demonstrate LNC and serological reactivity to the novel fusion protein occurs in immunized mice. Unlike wild-type mice, RP3 transgenic mice express the hRP3 fusion gene as a self protein exclusively in the thyroid epithelium (23) by virtue of a bovine thyroglobulin promoter that is known to be active at an early stage of glandular development. To determine whether thyroid-specific expression of RP3 confers T cell tolerance to whole RP3 protein, adult transgenic mice (>4 mo of age) were immunized with rhRP3 protein, and 2 wk later LNC were harvested. Compared with nontransgenic mice immunized with rhRP3 protein, T cells from immunized RP3 transgenic mice failed to proliferate in response to rhRP3 protein (Fig. 4). LNC from both rhRP3-immunized RP3 and nontransgenic mice failed to proliferate to an unrelated control protein. Although RP3 mice are reported to develop mild leukocytic infiltrates in their thyroid (25, 26), thyroid glands resected from immunized and nonimmunized RP3 transgenic mice showed no difference in the degress of infiltration, as determined by histological analysis (not shown).
|
Because injection of purified rRP3 indicated that exogenous exposure of the fusion protein can evoke a strong T cell proliferative response, we next assessed whether RP3-expressing thyroid tissue is also capable of evoking responses in naive mice following organ transplantation. To measure acute immune responses against RP3-expressing preneoplastic thyroid in vivo, thyroid lobes harvested from 4-mo-old RP3 transgenic mice were s.c. transplanted into haplocompatible B6C3F1 and SCID control mice. Thyroid tissue from donor mice at 4 mo of age is composed primarily of small RP3-expressing follicles containing primary and secondary hyperplasia (23). Three weeks posttransplant (p.t.), 10 of 10 thyroid grafts remained intact (Fig. 5) and functional, as determined by thyroglobulin expression and follicular morphology (not shown). These transplants were free of immune cell infiltration in both B6C3F1 recipients and immunodeficient SCID recipient mice, as determined by histopathological analysis.
|
To measure the ability to induce antithyroid immunity mediated by anti-RP3 responses, naive haplocompatible and SCID mice were s.c. transplanted with a single RP3-expressing transgenic thyroid lobe and monitored for graft rejection for 5 mo. All grafts in recipient mice formed stable and measurable masses (10 of 10), suggesting a lack of complete tissue rejection and resolution (Fig. 5). To determine whether peripheral T cells are capable of recognizing the graft when stimulated under inflammatory conditions, recipient mice were infected i.v. with 107 PFU recombinant vaccinia virus encoding RP3, 107 PFU wild-type vaccinia virus, or PBS at 5 mo p.t. Mice were monitored weekly for rejection of established transplants. After 2 mo, minimal reductions in graft size were noted from recombinant virus-infected mice (-7.79 +/- 2.21%) compared with wild-type virus-infected mice (+3.17 +/- 3.45%). To determine whether thyroid transplants were rejected and to measure the degree of leukocyte involvement following immunization, the transplanted RP3 thyroid lobes were resected, embedded, sectioned, and analyzed histopathologically. Examination of transplanted RP3 thyroid lobes from RP3 vaccinia virus-infected mice (Fig. 6) revealed thyroiditis with infiltration of neutrophils, monocytes, and lymphocytes in 25–50% of the transplant (Fig. 6C), whereas transplants from PBS and wild-type vaccinia virus-infected mice remained free of leukocytic infiltration (Fig. 6, A and B). Extensive lymphocytic infiltration of the follicles caused significant organ disruption, although noninfiltrated regions of the thyroid transplant remained normal in morphology (Fig. 6C). Despite widespread thyroiditis, complete destruction and resolution of RP3 thyroid grafts were not observed following single immunization with either recombinant or wild-type virus. The sequential order of thyroid transplant and immunization was not critical because similar results were observed in naive mice vaccinated with recombinant vaccinia virus 2 wk before a transgenic thyroid transplants (not shown).
|
Because thyroids from 4-mo-old RP3 transgenic mice are slow growing, rarely metastasize, and remain encapsulated (23), it was possible that thyroid organ transplants, in contrast to vaccinia virus-infected tissues, were incapable of eliciting a sufficiently strong inflammatory signal for the initiation of an appreciable graft rejection response. Consequently, to derive a strong inflammatory signal from thyroid grafts, we used tumor fragments expressing RP3 protein. RP3 mice that have been crossed to p53 null mutants (RP3p53-/-) develop large, fast-growing, poorly differentiated thyroid carcinomas (4). Although RP3 expression is eventually lost in these advanced tumors, thyroid tumors from 6-mo-old RP3p53-/- mice still express uniformly detectable levels of RP3, as previously demonstrated (4). To address the possibility that the inflammatory properties of malignant growth may enhance the induction of antithyroid responses in vivo, 3-mm3 thyroid tumor fragments from 6-mo-old RP3p53-/- mice were transplanted s.c. into haplocompatible recipient mice and monitored weekly for tumor outgrowth. Whereas 100% of the RP3p53-/- thyroid tumor fragments transplanted into SCID mice grew rapidly, only 23% of tumor fragments transplanted into B6C3F1 recipients grew progressively, as assessed at 3 wk p.t. (Fig. 5). Histopathological analysis was performed on the remaining thyroid transplants resected from B6C3F1 recipients at 1 mo p.t. Compared with tumors grown in SCID mice, tumor resected from B6C3F1 mice was infiltrated with numerous polymorphonuclear neutrophils and lymphocytes (Fig. 7).
|
Leukocyte involvement suggested Ag-specific T cell responses may be elicited during RP3 tumor rejection. To investigate this, 2 mo following tumor transplant, B6C3F1 recipient mice were sacrificed and autopsied to verify the complete rejection of RP3p53-/- thyroid tumor. Complete rejection was confirmed by analysis of the transplant site following specimen embedding, fixation, and H&E staining. Furthermore, if RP3 is a dominant tumor Ag in tumor-rejected recipients, a robust lymphoproliferative response should be detectable by standard proliferation assay. To measure T cell responses to RP3 in these mice, regional and distal lymph nodes were harvested from tumor-rejecting mice, cultured, and assayed in vitro for a proliferative response to serial dilutions of rhRP3 protein. RP3p53-/- tumor-sensitized LNC proliferated strongly to rhRP3 protein, but not to unrelated control protein stimulation (Fig. 8). Because RP3 is composed of two widely expressed self proteins, RFG/ARA70 and c-RET, the only unique portion of RP3 in the tumor tissue is the fusion breakpoint region. Therefore, to measure fusion region-specific responses, cultured tumor-rejecting LNC were stimulated with the fusion region-specific peptide in vitro. Similar to rRP3-immune lymphocytes, LNC from tumor-rejecting mice did not proliferate to the fusion region peptide, suggesting that natural processing of this epitope from tumors does not evoke a thyroid-specific immune response. LNC from nonimmunized B6C3F1 mice failed to respond to rRP3, an RP3 fusion region peptide, or an unrelated protein (not shown). To measure the humoral response to RP3 during thyroid tumor rejection, serum was collected from tumor-rejecting mice and assayed for Ag specificity by ELISA. In contrast to the serological reactivity detected in rRP3-immunized mice, Ab responses specific to rhRP3 protein were not detected in serum samples from RP3p53-/- tumor-rejecting B6C3F1 mice, suggesting that Ab reactivity is not required for tumor rejection (Fig. 1C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Results obtained from proliferation assays demonstrate the specific T cell immunogenicity of RP3 fusion protein when presented by APCs. We predicted that immunity generated against a protein derived from the fusion of two widely expressed self proteins, such as RFG/ARA70 and c-RET, would be directed against the unique fusion region segment of the chimeric molecule because tolerance to the flanking regions would likely occur following T cell development. On the contrary, we found little reactivity to the fusion breakpoint peptide predicted to bind I-Ek, suggesting that a nonbreakpoint region of the fusion protein encoded the dominant immunogenic epitope(s). Interestingly, robust T cell responses to the mRetTKD were observed following RP3 immunization consistent with the location of an immunogenic epitope in this region, as predicted by MHC-binding algorithms. Tolerance in normal mice was observed, however, when the self protein parts of RP3 were used as immunogens. These data indicate that an autoimmune response against proteins may occur by a process that increases or alters epitope production following the fusion of two unrelated proteins.
However, how might the formation of a fusion protein lead to such changes in Ag processing? Presumably, folding of the c-RET kinase domain is not altered by fusion to RFG/ARA70, given the preservation of its activity. Thus, one might expect similar handling of this domain by the Ag-processing machinery in the context of both c-RET and RP3, and therefore tolerance to any c-RET-derived epitopes generated by processing of RP3. There are several potential explanations for why this is not the case. One factor to consider is the change in subcellular location, from the inner leaflet of the plasma membrane (c-RET) to the cytosol (RP3). This may substantially alter the Ag-processing pathways that are available. For example, both endosomal/lysosomal proteases and the cytosolically located proteasome have been implicated in the processing of cytosolic Ags for class II-restricted presentation (29). Perhaps c-RET is targeted principally to the endocytic compartment, whereas RP3 is targeted to the proteasome, with the epitopes that are generated in each case being distinct. This scenario presumes endogenous presentation of both proteins (presentation by the cell that expresses the protein), but it may also be that most presentation of c-RET and RP3 requires transfer of the proteins to professional APCs (cross-priming). In this case, both proteins would most likely be presented through the classic endocytic pathway, a possibility that is currently being tested. Even so, variable processing of the carboxyl-terminal domain of RP3 is still a strong possibility, as it has been shown that binding of Ab to an Ag, analogous to fusion of the carboxyl-terminal region of c-RET with the amino-terminal region of RFG/ARA70, significantly alters the efficiency with which particular epitopes are generated (30). Finally, it should be noted that lysosomal proteases cathepsin B, D, and L are up-regulated in thyroid cancers and in thyroid tissue from patients with autoimmune disease (31, 32). Thus, it may be a change in the processing machinery rather than a change in the substrate that is responsible for generation of the transformation-associated epitope. Whichever of these mechanisms proves correct, the formation of RP3 predisposes thyroid epithelium to overcome tolerance through the recognition of self epitopes. Future studies will help to resolve the processing mechanism specific for the RP3 fusion protein.
Even though the fusion of two self proteins is an important feature of the immunogenicity of RP3, the malignant state of the RP3-expressing tissue determined whether thyroid-specific immunity could be induced. For example, although direct immunization led to RP3-specific immunity, premalignant thyroidal expression of RP3 alone was not sufficient for the unprimed induction of immunity against thyroid tissue grafts expressing RP3. Although the immunogenic properties of thyroid grafts are different from other organs such as skin (33), the lack of immune response to RP3-expressing thyroid grafts may be due to insufficient proinflammatory mediator production at the time of priming. Indeed, even strongly immunogenic viral Ags have a tolerizing effect when expressed in skin grafts without coincident inflammation and are immunogenic only following a localized inflammatory reaction (34, 35). Furthermore, tissue grafts encoding the E7 tumor Ag of human papillomavirus type 16 are rejected more readily when Ag priming occurs in the presence of an LPS-induced inflammatory reaction (36). Consistent with this idea, we find high level expression of a number of key proinflammatory mediators by developed thyroid carcinomas, but fewer and lower amounts of these mediators in preneoplastic thyroid lesions of RP3 transgenic mice.4 Early neoplastic lesions may thus represent an immunological crossroad between tolerance to self and immunity to tumor, with direction being determined by the extent and amount of cytokines produced from stressed or transformed cells. In addition to these inflammatory mediators, the state of newly grafted tissue, tumor fragments vs single thyroid lobes excised en bloc, may also influence the effector response because increased cellular apoptosis and necrosis may facilitate Ag uptake and cross-presentation by professional APCs, resulting in thyroid tissue rejection. Even though, in some cases, antitumor Ab may facilitate Ag processing for tumor rejection, humoral responses were not detected in RP3 tumor-rejecting mice, a finding not inconsistent with other tumor systems (37). Even though these data support a T cell-mediated rejection mechanism, they do not explain the reasons for the lack of humoral responses. In addition, thyroid cells expressing RP3 secrete a number of inflammatory cytokines (J. Russell, M. Santoro, and J. L. Rothstein, manuscript in preparation) that may bias immune responses toward T cell immunity; however, future studies will be necessary to understand the strength of this response in vivo.
Previous data have shown frequent expression of RP3 fusion gene at very early stages of cancer (38) and in thyroid tissue of patients with thyroid autoimmune disease, suggesting that this expression may underlie the coexistence of neoplasia (10, 11) and autoimmunity. Studying the early stages of RP3 expression in mouse thyroid tissue will help to evaluate the inflammatory conditions within the organ observed at very early stages following thyroid transformation, but before carcinoma and/or autoimmune disease occurs. This is important because the time frame between cancer initiation and detection can be many years, over which time the immunological landscape of the transformed thyroid may change. This is more apparent when considering the studies presented in this work, which may provide the basis for understanding the widespread effects of how oncogene-altered epithelial cells may evoke immune responses. RP3 expression in thyroid tissue may provide both a strong antigenic and chemotactic stimulus for monocytes and lymphocytes to infiltrate oncogene-expressing tissue in patients with disease-prone MHC backgrounds. Accordingly, this suggests that to evoke an effective immune response against oncogene-expressing tissue, an alteration in the target tissue may be necessary. This change appears following the transition of thyroid tissue from normal to neoplastic, possibly mimicking the proinflammatory effects brought on by immunizing the host with adjuvants or with highly immunogenic viruses. Thus, the expression of RP3 in thyroid tissue of patients with neoplastic thyroid disease supports a model of autoimmunity whereby organ immunogenicity is a consequence of oncogenic transformation, resulting in the development of chronic inflammation.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jay L. Rothstein, Departments of Microbiology/Immunology and Otolaryngology-Head and Neck Surgery, Kimmel Cancer Center, Thomas Jefferson University, 233 South 10th Street, BLSB 909, Philadelphia, PA 19107. E-mail address: rothstei{at}lac.jci.tju.edu
3 Abbreviations used in this paper: RET/PTC, rearranged during transfection/papillary thyroid carcinoma; PTC, papillary thyroid carcinoma; h, human; LNC, lymph node cell; m, mouse; p.t., posttransplant; RP3, RET/PTC3; Tcl1, T cell locus 1; TKD, tyrosine kinase domain.
4 J. P. Russell, R. M. Melillo, M.D. Castellone, M. Santoro, and J. L. Rothstein. The thyroid tumor antigen and tyrosine kinase oncoprotein, RET/PTC3, induces the secretion of proinflammatory cytokines. Submitted for publication.
Received for publication July 23, 2002. Accepted for publication November 12, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 fusion in oncogene human thyroid carcinoma. Science 289:1357.This article has been cited by other articles:
|
|
 |
|
  R. Flavin, G. Jackl, S. Finn, P. Smyth, M. Ring, E. O'Regan, S. Cahill, K. Unger, K. Denning, Jinghuan Li, et al. RET/PTC Rearrangement Occurring in Primary Peritoneal Carcinoma International Journal of Surgical Pathology, June 1, 2009; 17(3): 187 - 197. [Abstract] [PDF]
|
|
|
|
 |
|
 J. S. Pufnock and J. L. Rothstein Oncoprotein Signaling Mediates Tumor-Specific Inflammation and Enhances Tumor Progression J. Immunol., May 1, 2009; 182(9): 5498 - 5506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Burniat, L. Jin, V. Detours, N. Driessens, J.-C. Goffard, M. Santoro, J. Rothstein, J. E. Dumont, F. Miot, and B. Corvilain Gene Expression in RET/PTC3 and E7 Transgenic Mouse Thyroids: RET/PTC3 But Not E7 Tumors Are Partial and Transient Models of Human Papillary Thyroid Cancers Endocrinology, October 1, 2008; 149(10): 5107 - 5117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Mologni, E. Sala, S. Cazzaniga, R. Rostagno, T. Kuoni, M. Puttini, J. Bain, L. Cleris, S. Redaelli, B. Riva, et al. Inhibition of RET tyrosine kinase by SU5416. J. Mol. Endocrinol., October 1, 2006; 37(2): 199 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Rhoden, K. Unger, G. Salvatore, Y. Yilmaz, V. Vovk, G. Chiappetta, M. B. Qumsiyeh, J. L. Rothstein, A. Fusco, M. Santoro, et al. RET/Papillary Thyroid Cancer Rearrangement in Nonneoplastic Thyrocytes: Follicular Cells of Hashimoto's Thyroiditis Share Low-Level Recombination Events with a Subset of Papillary Carcinoma J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2414 - 2423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Puxeddu, J A Knauf, M A Sartor, N Mitsutake, E P Smith, M Medvedovic, C R Tomlinson, S Moretti, and J A Fagin RET/PTC-induced gene expression in thyroid PCCL3 cells reveals early activation of genes involved in regulation of the immune response Endocr. Relat. Cancer, June 1, 2005; 12(2): 319 - 334. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Hwang, D. W. Kim, J. H. Hwang, H. S. Jung, J. M. Suh, Y. J. Park, H. K. Chung, J. H. Song, K. C. Park, S. H. Park, et al. Regulation of Signal Transducer and Activator of Transcription 1 (STAT1) and STAT1-Dependent Genes by RET/PTC (Rearranged in Transformation/Papillary Thyroid Carcinoma) Oncogenic Tyrosine Kinases Mol. Endocrinol., November 1, 2004; 18(11): 2672 - 2684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Russell, J. B. Engiles, and J. L. Rothstein Proinflammatory Mediators and Genetic Background in Oncogene Mediated Tumor Progression J. Immunol., April 1, 2004; 172(7): 4059 - 4067. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or nonimmunized (
) mice failed to produce
specific Ab reactivity. Data are representative of three independent
experiments. Statistical significance was determined by an independent
samples t test (Analyze-it).
; 10 of 10) or C57BL/6 SCID
(
