The JI Acurri Cytometers
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, J.
Right arrow Articles by Herlyn, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, J.
Right arrow Articles by Herlyn, D.
The Journal of Immunology, 2003, 171: 2922-2929.
Copyright © 2003 by The American Association of Immunologists

Recombinant CD63/ME491/Neuroglandular/NKI/C-3 Antigen Inhibits Growth of Established Tumors in Transgenic Mice 1

Jian Li2,*, Weiping Li2,*, Shaohong Liang*, Dewei Cai*, Marie Paule Kieny{ddagger}, Lutz Jacob*, Alban Linnenbach*, Jan W. Abramczuk*, Hans Bender*, Katrin Sproesser*, Rolf Swoboda*, Rajasekharan Somasundaram*, DuPont Guerry{dagger} and Dorothee Herlyn3,*

* The Wistar Institute, and {dagger} Pigmented Lesion Clinic, University of Pennsylvania, Philadelphia, PA 19104; and {ddagger} 2 Unité 544, Institut National de la Santé et de la Recherche Médicale, Strasbourg, France


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Attempts to vaccinate against tumors can be hindered by the induction of immunological tolerance to the target Ag as a result of Ag expression on normal tissues. In this study, we find that transgenic mice expressing the melanoma-associated Ag CD63/ME491/neuroglandular/NKI/C-3 on their normal tissues do, in fact, exhibit immunological tolerance to the Ag, recapitulating the conditions in cancer patients. In these mice, growth of murine melanoma cells expressing the Ag after gene transfer was inhibited by immunization with Ag-expressing recombinant vaccinia virus combined with IL-2, but not by immunization with the protein alone, anti-idiotypic Abs, or irradiated tumor cells. The effect of the recombinant virus was demonstrated both for nonestablished and established tumors. Infiltration with both CD4+ and CD8+ T lymphocytes was significantly more extensive in tumors from experimental mice than in tumors from control mice. MHC class I-positive, but not class I-negative, tumors were inhibited by the vaccine, suggesting that MHC class I-restricted T lymphocytes play a role in the antitumor effects. Abs did not appear to be involved in the vaccine effects. CD63 was immunogenic in 2 of 13 melanoma patients, pointing to the potential of this Ag, combined with IL-2, as a vaccine for melanoma patients.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Tumor-associated Ags are usually expressed both on the surface and in the cytoplasm of tumor cells, rendering these Ags potential targets for both B and T cells. In contrast, most tumor-specific Ags, which are usually the protein product of mutated normal genes, are intracytoplasmic and thus preferential targets for T cells only (1, 2, 3). MART-1, gp-100, tyrosinase, HER-2/neu, and Ep-CAM are naturally occurring tumor-associated self-Ags and are recognized by cancer patients’ T cells (4, 5, 6, 7, 8, 9, 10). Self-Ags that are targets for Abs have also been described (11, 12, 13, 14).

Because most tumor-associated Ags are also expressed on normal tissues, animal models of vaccinations against these Ags must take into account the immunological tolerance of the host as a result of normal tissue expression. Animal models of vaccines intended for use in humans will likely need to be developed for each Ag considered for targeting by cancer immunotherapy.

Transgenic (Tg) 4 mice expressing human tumor-associated Ags on their normal tissues, in a distribution similar to that in humans, are excellent models for the study of tumor-associated Ag vaccines. Mice transgenically expressing human carcinoembryonic Ag (CEA) (15, 16, 17), epithelial mucin-1 (18, 19, 20, 21, 22, 23), prostate-specific Ag (24), SV40 T Ag (25), or HER-2/neu (10, 26) have served as model systems for active specific immunotherapy against syngeneic tumors expressing the human Ag after gene transfer. However, only a few of these studies have demonstrated immunological tolerance to the transgene in the mice (15, 16, 17). Furthermore, although those vaccines were effective in the prophylactic setting, none has been shown to induce regression of established tumors in Tg mice. Recently, regression of established tumors in CEA Tg mice vaccinated with recombinant vaccinia virus (VV) and/or fowlpox vectors expressing CEA and costimulatory molecules has been reported (27, 28). These vaccines hold much promise in conferring protective immunity to cancer patients who usually present with measurable tumor burden at the time of treatment. Several mechanisms have been suggested to underlie the antitumor vaccine effects, such as cytotoxic Abs and proliferative and cytolytic T lymphocytes (27, 28).

In the present study, we used the CD63/neuroglandular/NKI/C-3 Ag (29, 30, 31, 32, 33), referred to hereafter as CD63 Ag, as a target for active specific immunotherapy in Ag Tg mice. The Ag is a lysosomal protein that is expressed on the surface of melanoma cells, but not normal melanocytes (29, 30). It is also expressed on colon and prostate carcinomas (34) and on various normal cells (29, 30). CD63 is immunogenic in a small fraction of melanoma patients (35, 36), raising the possibility of patient response to booster vaccinations with the Ag. We demonstrate in this study that the CD63 expressed in rVV and administered together with IL-2 can inhibit established melanoma growth in CD63-Tg mice immunologically tolerant to the Ag.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patients

Thirteen patients with metastatic melanoma were studied. Lymphocytes were obtained by Ficoll-Hypaque (Amersham Biosciences, Uppsala, Sweden) centrifugation of heparinized peripheral blood.

Mice

Eight- to 10-wk-old female non-Tg (wild-type (wt)) C57BL/6 mice (H-2Kb, Thy-1.2) were obtained from Taconic (Germantown, NY). Mice Tg for the CD63 were generated using standard methods. Briefly, ~75 copies of the CD63 genomic fragment under the control of the human CD63 promoter were introduced into one-cell stage embryos. The human CD63 promoter has been proven to function in B16 melanoma cells (see Results). Following DNA injection, embryos were transferred into oviducts of pseudopregnant Swiss Webster female mice. Offspring were analyzed at 4 wk of age for the presence of CD63 sequences by Southern blot analysis of DNA isolated from tail specimens or by PCR. For PCR, the CD63-specific 630-bp genomic fragment was isolated from genomic DNA using sense (5'-GCCTGTGCAGTGGGACTGATT-3) and antisense (5'-ATCTGCCTGCATCCTGTCCA-3') primers. Tg mice were designated CD63-Tg mice.

Cell lines

Cells of clone B78H1 of B16 melanoma cells (C57BL/6 mouse origin) were grown in a humidified 5% CO2 atmosphere in Eagle’s MEM containing 10% heat-inactivated FCS (Life Technologies, Grand Island, NY). WM35 and WM164 human melanoma cells expressing CD63 (see Results) were grown in Leibovitz L-15 medium supplemented with 10% FCS.

Antibodies

mAb B8-24-3 (American Type Culture Collection, Manassas, VA) is directed against MHC H-2Kb. mAb ME491 reacts with human primary, and to a lesser extent, metastatic melanoma tissues (29, 30), and also with various normal tissues (Table I).


View this table:
[in this window]
[in a new window]
  		Table I. Distribution of CD63 in human and CD63 Tg-mouse tissues

 
Generation of murine melanoma transfectants

Murine melanoma cell line B78H1 (MHC class I- and II-negative) was cotransfected with plasmids carrying the human CD63 and neomycin resistance genes. The R33 genomic clone for CD63 (37) was digested with HindIII, and a 9.4-kb DNA fragment containing promoter and coding sequences for the CD63 gene was subcloned into the pBluescript SK vector. Plasmid subclone pR33HB1 was used for gene transfer into B78H1 cells by standard calcium-phosphate precipitation method. Cells (7.5 x 104 in each of 20-mm dishes) were incubated overnight with 20 µg of CD63 recombinant DNA and 0.15 µg of pSV2-neo DNA. After an additional 48-h incubation in normal medium, transfectants were selected with the antibiotic G418 (1 mg/ml; Life Technologies). Colonies expressing the CD63 were isolated following rosetting of mAb 491-coated tumor cells with SRBC precoated with mouse anti-SRBC Ab and rabbit anti-mouse IgG Ab (38). More than 400 G418-resistant colonies were obtained, most of which reacted with mAb ME491. Of those colonies, five (491-6, -13, -17, -18, and -20) were cloned, and one (clone 20) was selected for further study.

Because CD63-20 cells do not express MHC class I Ags, they were cotransfected with the H-2Kb gene. Two plasmids carrying the hygromycin resistance gene or the H-2Kb gene, respectively, were kindly provided by Dr. L. Pease (Mayo Clinic, Rochester, MN).

Transfectants were selected in medium containing 70 µg/ml hygromycin, and positive cells were isolated using the rosetting assay with mAb B8-24-3 as described above for the isolation of CD63 transfectants. The transfectants were designated CD63-20Kb.

Western blot analysis

Proteins were extracted from Tg mouse tissues with solubilizing buffer (0.5% Nonidet P-40, 140 mM NaCl, 10 mM NaF, 10 mM Tris, 5 mM EDTA, aprotinin (100 IU/ml kallikrein), and 1 mM PMSF (pH 7.5)). Reactivity of Abs with the proteins in this extract and immunoaffinity-purified CD63 (positive control; prepared as described in Ref.39) was analyzed by Western blotting under nonreducing conditions using a 10% polyacrylamide gel. Blots were incubated for 2 h at room temperature with various Ab preparations at 1 µg/ml in PBS containing 0.1% Tween 20 and then incubated (2 h at room temperature) with 125I-labeled goat anti-mouse F(ab')2 (200,000 cpm/ml).

Mixed hemadsorption assay (MHA)

Sera of immunized mice were tested for binding to tumor cells as described (38). Briefly, cells attached to flat-bottom microtiter wells were incubated with various serum dilutions followed by the addition of indicator SRBC preincubated sequentially with mouse anti-SRBC IgG and goat anti-mouse IgG. The number of rosetted tumor cells was determined by counting under the light microscope.

Scatchard and flow cytometry analyses

To determine the maximal number of Ab-binding sites on murine melanoma transfectants, cells were incubated with 125I-labeled mAb ME491 or a mixture of labeled and unlabeled mAb, as described in detail (40). Radioactivity bound to tumor cells was determined in a gamma-scintillation counter. Maximal number of binding sites per cell was determined as described by Scatchard (41). Binding of mAbs ME491 and B-824-3 to tumor cells was determined by flow cytometry. Cells were incubated with 10 µg/ml mAb or normal mouse IgG and washed, and fluoresceinated goat anti-mouse F(ab')2 Ab was added. Ab binding was determined in the cytofluorograph.

ELISA

Binding of serum Abs from melanoma patients and immunized mice to immunoaffinity-purified Ag (prepared as described in Ref.39) immobilized onto microtiter plates was determined using HRP-labeled goat anti-mouse (or human) F(ab')2 for detection of Ab binding and o-phenylenediamine as substrate. OD405 of the wells was determined.

Splenocyte proliferation assay

Splenocytes from immunized mice were prepared and incubated (4 x 105 cells in 0.2 ml/well) with affinity-purified CD63 protein or Con A at various concentrations for 3 days as described (42). Proliferation of lymphocytes was assayed using a 16-h [3H]thymidine pulse. Stimulation index was calculated as follows: stimulation index = cpm in splenocytes with stimulant/cpm in splenocytes without stimulant.

Generation of VV CD63 (VVTG6120) and wt VV (VV wt)

CD63 cDNA was cloned as an EcoRI fragment into M13TG131 (43). A SmaI site as well as an A in position -3 vs the initiation ATG were introduced by localized mutagenesis upstream of the CD63 coding sequence. CD63 cDNA was excised as a SmaI fragment (newly introduced SmaI in 5'-SmaI site in the polylinker of M13H131 at the 3' end) and cloned downstream of the P7.5k VV promoter in pTG194 (VV transfer vector) into a SmaI restriction site to generate pTG6120. pTG194 is identical with plasmid pTG168 poly (44), except that it lacks the polylinker sequence. Double homologous recombination in chicken embryo fibroblasts using VV Copenhagen ts7 strain, VV wt purified DNA, and pTG6120 resulted in the generation of VV CD63 (VVTG6120). VV wt is the nonrecombinant VV strain Copenhagen.

Immunization and tumor challenge of mice

To determine whether CD63-Tg mice develop a humoral immune response to CD63, Tg and wt mice (three to four mice per group) were immunized s.c. four times at 2-wk intervals with 10 µg of CD63 protein purified on mAb ME491 immunoaffinity columns as described (39) or with BSA control protein. Sera were obtained 10 days after the last immunization. To study lymphoproliferative immune responses to the CD63 protein, mice (four to five mice per group) were immunized s.c. two times at a 2-wk interval with 10 µg of CD63 protein in CFA (first injection) or IFA (second injection). Ten days after the last immunization, mice were sacrificed, and splenocytes were obtained for use in lymphocyte proliferation assays.

For tumor protection studies, wt or Tg mice were immunized with the following: 1) 300 or 5 µg of alum-precipitated CD63 protein or BSA four times at 2-wk intervals, with or without cyclophosphamide (0.3 mg/mouse); 2) 5 x 107 irradiated CD63-13 or B78H1 tumor cells in CFA (first injection) or IFA (subsequent three injections at 2-wk intervals); or 3) 100 µg of anti-idiotypic Ab (Ab2) HP2 4D7 or normal mouse IgG coupled to keyhole limpet hemocyanin in CFA (first immunization) or IFA (subsequent three immunizations at 2-wk intervals). Fourteen days after the last immunization, all mice were challenged with 5 x 105 CD63-20 tumor cells. In another set of protection studies, mice were immunized with VV CD63 or VV wt (107 PFU/mouse by tail scarification) on days 1, 8, and 33 with or without IL-2. On the first and second day of each vaccination, 105 IU of IL-2 in 0.3 ml of PBS were administered i.p. twice daily. On day 48, mice were challenged s.c. with 5 x 106 CD63-20Kb or CD63-20 cells. To determine the effect of specific and control virus on the growth of established tumors, mice were first injected s.c. with tumor cells, followed 13 days later (when the tumor was vascularized as determined in histological sections) by vaccination with specific or control virus together with IL-2.

In all vaccinated mice, tumor areas (length x width, in square millimeters) were recorded at 5-day intervals.

Production of Ab2-producing hybridoma cells

Hybridoma cells producing Ab2 were generated by fusing splenocytes from BALB/c mice immunized with mAb ME491 to nonsecretory murine myeloma cells 6/5/3 as described (45). The Ab2 inhibited binding of mAb ME491 to melanoma cells (not shown), thus mimicking CD63.

Analysis of tumor-infiltrating leukocytes (TIL)

TIL were analyzed for phenotypic markers in mice (eight per group) immunized prophylactically with VV wt or VV CD63, both with IL-2, as described in Materials and Methods. When s.c. tumors of the VV CD63-vaccinated group were ~1 cm in maximal diameter (days 37–59 after tumor challenge), mice were sacrificed, and the entire tumor mass was removed and weighed. Simultaneously, the largest tumors were removed from mice in the control group. Thus, the largest size tumors were analyzed for TIL at one time point in both groups.

Single cells were prepared from 0.5–1 g of fresh tumor tissue (entire tumor mass) and suspended in RPMI 1640 medium. To detect mouse CD3, CD4, CD8, CD45, CD19, Mac-1, and NK1.1, 1 x 105 leukocytes were incubated with 0.5 µg/ml FITC- or PE-labeled goat Abs directed to these markers (BD PharMingen, San Diego, CA) in PBS for 60 min at 4°C. Cells were washed in PBS/0.1% BSA and analyzed by flow cytometry. Results are expressed as percentage of leukocytes with a particular marker among the total number of leukocytes infiltrating the tumor mass.

Statistical analysis

Differences between experimental and control groups in CD63-20Kb and CD63-20 tumor growth were compared using paired Student’s t test. For TIL analysis, values obtained from paired control and experimental mice at a particular time point were pooled, and differences in the percentage of infiltrating lymphocytes of a particular phenotype were compared using paired Student’s t test. Differences at p < 0.05 were considered statistically significant. Survival rates of control and experimental groups were compared using Kaplan-Meier statistics.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
CD63 tissue distribution in Tg mice vs humans

Comparison of CD63 expression in humans, as determined immunohistochemically (29, 30), with that in Tg mice, as determined by Western blot analysis, revealed a similar distribution, except for the presence of the Ag in intestine and the absence in thymus of Tg mice (Table I).

CD63 Ag immunogenicity in melanoma patients

Two of 13 melanoma patients produced serum Abs specific for the immunoaffinity-purified CD63 Ag as determined in ELISA. These Abs did not bind to BSA. However, the same Ag preparation was unable to specifically (vs BSA control) stimulate three patients’ lymphocytes in culture (not shown).

Human CD63 expression by transfectants

Three CD63 Ag-transfected clones expressed low, intermediate, and high levels of CD63 (clones 6, 20, and 13 with 7.8 x 105, 1.7 x 106, and 4.3 x 106 antigenic sites per cell, respectively, by Scatchard analysis) after transfection of the parental B78H1 melanoma cells. Clone 20 with intermediate Ag expression was used for all in vivo studies, and was cotransfected with MHC class I Ag (CD63-20Kb). CD63-20Kb cells expressed both the CD63 and MHC-Kb, the CD63-20 clone expressed only CD63, and the parental B78H1 cells expressed neither Ag (Fig. 1). The density of CD63 on human melanoma cell line WM164 was 1.6 x 104 sites/cell, and >90% of the cells expressed the Ag (not shown).



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 1. Expression of CD63 and MHC Ags by transfected B78H1 murine melanoma cells. Nontransfected parental B78H1 and transfected CD63-20 and CD63-20Kb cells were evaluated for CD63 and MHC class I Ag expression by FACS analysis using mAbs ME491 and B8-24-3, respectively.

 
CD63-Tg mice show humoral, but not cellular, immunological tolerance to CD63

MHA analysis of Tg and wt mouse sera obtained 10 days after the fourth immunization s.c. with 10 µg of purified CD63 or BSA control protein revealed significant binding to CD63-20 cells for CD63 Ag-immunized wt mice, but no binding for immunized Tg mice (Fig. 2A). Binding of the sera from experimental mice to CD63-20 cells was specific, because no serum binding to the parental CD63 Ag-negative B78H1 cells was detected (not shown).



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 2. B cell, but not T cell, tolerance of CD63-Tg mice to immunization with CD63 protein. A, CD63-Tg ({blacktriangleup}) and wt (•) mice (three and four mice per group, respectively) were immunized with 10 µg of purified CD63 protein (solid line) or BSA control protein (dashed line) four times at 2-wk intervals. Sera were obtained 10 days after the last immunization and tested for binding to CD63-20 cells in MHA. B, Induction of a lymphoproliferative immune response to the CD63 protein ({blacksquare}) or BSA ({square}) was evaluated in Tg mice (four and five mice per group, respectively), immunized s.c. twice at a 2-wk interval with 10 µg of CD63 protein in CFA (first injection) or IFA (second injection). Ten days after the last immunization, mice were sacrificed, and splenocytes were obtained for lymphocyte proliferation ([3H]thymidine incorporation) assay. Error bars represent the SD of three to four mice per group.

 
Wt and Tg mice immunized with CD63 protein without adjuvant showed no lymphocyte-proliferative responses (not shown), so that further immunizations were conducted using the protein in CFA and IFA. [3H]Thymidine incorporation assay of splenocytes from wt and CD63-Tg mice immunized s.c. two times with 10 µg of Ag in CFA or IFA at 2-wk intervals and stimulated 10 days after the last immunization with the Ag or BSA revealed significant and specific proliferation after stimulation with CD63 protein compared with BSA control in Tg mice, whereas BSA-immunized mice showed no significant lymphoproliferative responses (Fig. 2B). Similar lymphoproliferative responses were seen in wt mice immunized with CD63 protein in adjuvant (not shown).

Thus, Tg mice showed B cell, but not T cell, tolerance to CD63, compared with wt mice, which showed neither B nor T cell tolerance.

Antitumor effects of various forms of CD63 in wt and Tg mice

Immunoaffinity-purified, alum-precipitated protein Ag (with and without cyclophosphamide), and keyhole limpet hemocyanin-coupled Ab2 or irradiated tumor cells, both in CFA/IFA, were evaluated for their capacity to induce protective immunity against CD63-20 (H-2Kb-negative) tumor cells in wt mice, which were expected to respond to immunizations more effectively than Tg mice. None of the three immunogens protected wt mice against challenge with tumor cells (Table II), although all three preparations induced Abs binding to tumor cells (not shown). Thus, Abs are unable to control tumor growth in our system. Based on previous studies demonstrating that rVV can induce CTL responses (45, 46), we tested VV CD63 for tumor-protective effects in wt mice in a prophylactic immunization setting; the recombinant virus had no protective effect against challenge with CD63-20Kb cells (Table II). Combinations of recombinant viral vaccines and IL-2 often have shown antitumor effects superior to the viral vaccines alone (9, 47). Tg mice immunized with VV CD63 plus IL-2 showed significant (p < 0.01) tumor protection against subsequent challenge with CD63-20Kb cells (Table II, Fig. 3A) and significantly (p < 0.01) enhanced survival of the mice (Fig. 3B) as compared with control mice injected with VV wt and IL-2. Thus, six of eight mice in the VV CD63 plus IL-2 group survived beyond 70 days, whereas only two of eight mice in the control group survived.


View this table:
[in this window]
[in a new window]
  		Table II. Tumor protective activity of the human CD63 in wt and CD63 Tg mice

 


View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 3. Inhibition of nonestablished tumor growth (A) and enhancement of survival (B) by VV CD63 plus IL-2 in CD63-Tg mice. A, Eight mice were injected with VV CD63 (107 PFU/mouse) and IL-2 (105 IU/mouse, i.p.) ({blacksquare}) as described in Materials and Methods. Similarly, eight control mice were immunized with VV wt and IL-2 ({blacktriangleup}). All mice were challenged s.c. with 5 x 106 CD63-20Kb cells. Tumor areas, recorded at 5-day intervals, were significantly (p < 0.01; paired t test) smaller in mice immunized with VV CD63 plus IL-2. Ratios indicate number of tumor-bearing mice to total number of mice. B, The survival of mice immunized and challenged with tumor cells as described in A was evaluated. The survival rate was significantly higher (p < 0.01; Kaplan-Meier) in mice immunized with VV CD63 plus IL-2 (solid line) than in VV wt- plus IL-2-immunized mice (dashed line). Error bars represent SD of eight mice per group.

 
VV CD63 plus IL-2 did not inhibit growth of MHC class I-negative CD63-20 cells, although immunized mice produced Abs binding significantly and specifically to the Ag-positive tumor cells (not shown). Thus, tumor growth inhibition appears to be dependent on MHC class I-restricted T lymphocytes.

Analysis of Tg mice given therapeutic vaccinations with VV CD63 plus IL-2 revealed significant growth inhibition (p < 0.05) of 13-day-old established (vascularized) CD63-20Kb tumors (Fig. 4A) and significantly (p < 0.05) enhanced survival (Fig. 4B) as compared with control mice injected with IL-2 or VV wt plus IL-2.



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 4. Inhibition of established tumor growth (A) and enhancement of survival (B) by VV CD63 plus IL-2 in CD63-Tg mice. A, CD63-20Kb cells (5 x 106/mouse) were injected into the flank of eight Tg mice. Thirteen days later (), when tumors were vascularized as determined histopathologically (not shown), experimental mice were immunized with VV CD63 and IL-2 (•) as described in Materials and Methods. Similarly, eight control mice were injected with VV wt and IL-2 ({blacktriangleup}) or IL-2 only ({blacksquare}). Tumor areas, recorded at 5-day intervals were significantly smaller (p < 0.05; paired t test) in mice immunized with VV CD63 plus IL-2 than in the control groups. Ratios indicate number of animals with tumors to total number tested. B, Survival of mice immunized and challenged with tumor cells as described in A was evaluated. The survival rate was significantly higher (p < 0.05; Kaplan-Meier) in mice immunized with VV CD63 plus IL-2 (solid line) than in the IL-2- (dashed line) or VV wt- plus IL-2 (dotted line)-immunized mice. Error bars represent SD of eight mice per group.

 
Characteristics of TIL

To determine which leukocyte type might be responsible for the antitumor effects observed with VV CD63 plus IL-2 in Tg mice, cell surface Ag expression on TIL from VV CD63- plus IL-2-treated mice in the third experiment demonstrating significant tumor regression was examined by FACS analysis. The VV CD63- plus IL-2-treated regressor mice had significantly (p < 0.01) higher numbers of TIL-expressing T cell markers (CD45+, CD3+, CD4+, and CD8+) as compared with the TIL of the VV plus IL-2 wt group (progressor mice) (Fig. 5). There was also a significant increase in the CD19+ B cell infiltration of tumors of regressor mice, although the absolute number of infiltrating B cells was rather low. There was no difference between regressor and progressor mice in the tumor infiltration with Mac-1+ and NK1.1+ cells.



View larger version (12K):
[in this window]
[in a new window]
  		FIGURE 5. Characteristics of TIL in Tg mice. Eight mice per group were immunized with VV CD63 ({blacksquare}) or VV wt ({square}), both with IL-2, and challenged with CD63-20Kb cells as described in Fig. 3. Tumors were collected between days 37 and 59 after challenge, when tumors in the VV CD63-vaccinated groups were ~1 cm in maximal diameter. Simultaneously, tumors were collected from paired mice with the largest tumors in the VV wt control group. Cell suspensions from 0.5 to 1 g fresh tumor tissue were prepared, and cell surface markers were determined by FACS analysis. Results are expressed as percentage of positive TIL relative to total number of TIL. *, Significantly different (p < 0.01) from control. Error bars represent SD of eight mice per group.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present data demonstrate that mice transgenically expressing a human tumor-associated Ag in a distribution similar to that in humans, can be effectively immunized against established tumors expressing the Ag after gene transfer, consistent with the findings of Grosenbach et al. (27) and Greiner et al. (28). When immunized with native CD63 protein in the absence of adjuvant, Tg mice showed humoral immunological tolerance as compared with wt mice. However, neither Tg nor wt mice showed proliferative lymphocyte responses to this type of immunization. Such responses were obtained in both types of mice after immunization with the Ag in adjuvant (CFA/IFA). Thus, Tg mice exhibited humoral but not cellular immunological tolerance as compared with wt mice. Split tolerance also has been reported in other Tg mouse studies (19), although it should be noted that the outcome of immunological tolerance studies might depend on the form of Ag and the type of adjuvant used (48). In CD63 Tg mice, CD63 was not expressed in the thymus so that Ag-reactive lymphocytes most likely were not clonally deleted, a notion consistent with the demonstrated induction of protective antitumor effects by vaccination against the Ag. However, in humans, low-level expression of the Ag in the thymus (29, 30) may lead to various degrees of clonal deletion of Ag-specific lymphocytes and, in turn, a low immune response rate to the Ag in melanoma patients (Refs.35 and 36 ; the present study).

Histopathological analyses of Ag-expressing tissues from VV CD63-immunized mice revealed no toxicity when examined 3–4 mo after vaccination (all tissues listed in Table I except for stomach and liver were examined) (not shown), although it is not known whether the Ag is expressed on normal cells in a form recognizable by lymphocytes. Remarkable selectivity of destructive immunity for tumor tissues as opposed to normal tissues has also been reported in other Tg mouse systems (18, 19, 49). Normal tissues may be more resistant to injury than tumor tissues (50, 51, 52), possibly due to factors such as lower MHC Ag and/or costimulatory molecule expression by normal tissues. In contrast, Tg mice expressing lymphocytic choriomeningitis viral Ag on their normal tissues, such as pancreas, developed diabetes after infection with the virus (53).

Abs elicited by irradiated tumor cells, native protein, or Ab2, each administered in alum or CFA/IFA, or elicited by VV CD63 were unable to protect mice against subsequent challenge with CD63 Ag-positive/MHC class I-negative tumor cells. VV CD63 in combination with IL-2 was able to inhibit class I-positive but not class I-negative tumor cells, suggesting that MHC class I-restricted T lymphocytes play an important role in tumor control in our system. Importantly, tumors from VV CD63- plus IL-2-immunized mice showed significantly higher infiltration with T lymphocytes (similar proportions of CD4+ and CD8+ lymphocytes) than tumors from control mice. Thus, both CD4+ and CD8+ T lymphocytes may be important effectors against CD63-20Kb cells. A role for B cells seems less likely, because minimal numbers of B cells were found in the TIL, and MHC class I-negative tumors were not inhibited by VV CD63 despite the presence of Abs in the immunized mice. Similar to our study, both CD8+ and CD4+ T cells were instrumental in the inhibition of tumor growth in vaccinated HER-2/neu Tg or CEA Tg/MIN mice (10, 28). Several other studies in which vaccine effects were demonstrated in tumor-bearing Tg mice (16, 17, 20, 21, 22, 49) did not address the mechanisms of these effects.

In some studies, tumor-associated Ag vaccines were found to be ineffective in preventing tumor growth in Tg mice, indicating profound immunological tolerance of the mice to the vaccines; however, adoptive transfer of in vitro-stimulated lymphocytes from the tolerant mice to Tg mice did result in protection from tumor challenge (17, 18, 25, 49). Those studies suggest that CD4+ and/or CD8+ T cells mediated the antitumor effects in vivo.

RVV was an effective vaccine in Tg mice bearing established tumors when administered in combination with IL-2, as demonstrated in three independent experiments. We (9) and others (21, 54) have shown that cytokines can boost the development of immunity to weakly immunogenic self-proteins. VV CEA and/or fowlpox CEA was effective in both prophylactic and therapeutic vaccinations in Tg mice (27, 28), and VV CEA has induced cellular immune responses in cancer patients (55). VV has a wide host range and is capable of accepting large inserts of a foreign gene. Copresentation of a weakly immunogenic self-Ag with highly immunogenic VV proteins can boost the immune response to the inserted gene product. Following infection of host cells with rVV, the foreign gene product is usually expressed in association with the host cell’s MHC class I Ag, which induces a CTL response.

Mice Tg for melanoma-associated CD63 expressed on both tumor and normal tissues in a distribution similar to that in humans represent a clinically relevant animal model, because these mice are immunologically tolerant to the Ag. This tolerance can be broken by rVV expressing CD63 and IL-2, resulting in the regression of established tumors. Our study provides a rationale for trials based on tumor-associated Ags that are also expressed by normal tissues in cancer patients. VV CD63 appears to be a promising candidate vaccine for melanoma patients.


   Acknowledgments
 
We thank Dr. Larry Pease (Mayo Clinic) for providing the MHC class I H-2Kb plasmid, and the Commonwealth Universal Research Enhancement Program (Pennsylvania Department of Health).


   Footnotes
 
1 This work was supported by Grants CA10815 and CA25874 from the National Institutes of Health, and a grant from the Commonwealth Universal Research Enhancement Program (Pennsylvania Department of Health). Back

2 J.L. and W.L. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Dorothee Herlyn, The Wistar Institute, 3601 Spruce Street, Philadelphia, PA 19104. E-mail address: dherlyn{at}wistar.upenn.edu Back

4 Abbreviations used in this paper: Tg, transgenic; CEA, carcinoembryonic Ag; VV, vaccinia virus; wt, wild-type; MHA, mixed hemadsorption assay; TIL, tumor-infiltrating leukocyte. Back

Received for publication April 21, 2003. Accepted for publication July 15, 2003.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Forrester, K., C. Almoguera, K. Han, W. E. Grizzle, M. Perucho. 1987. Detection of high incidence of K-ras oncogenes during human colon tumorigenesis. Nature 327:298.[Medline]
  2. Lane, D. E.. 1992. Cancer: p53, guardian of the genome. Nature 358:15.[Medline]
  3. Serrano, M., G. J. Hannon, D. Beach. 1993. A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4. Nature 366:704.[Medline]
  4. Hellstrom, I., K. E. Hellstrom. 1998. T cell immunity to tumor antigens. Crit. Rev. Immunol. 18:1.[Medline]
  5. Jaffee, E. M., D. M. Pardoll. 1996. Murine tumor antigens: is it worth the search?. Curr. Opin. Immunol. 8:622.[Medline]
  6. Kawakami, Y., S. A. Rosenberg. 1997. Human tumor antigens recognized by T-cells. Immunol. Res. 16:313.[Medline]
  7. Van den Eynde, B. J., T. Boon. 1997. Tumor antigens recognized by T lymphocytes. Int. J. Clin. Lab. Res. 27:81.[Medline]
  8. Van den Eynde, B. J., P. Van der Brugen. 1997. T cell defined tumor antigens. Curr. Opin. Immunol. 9:684.[Medline]
  9. Basak, S., S. Eck, R. Gutzmer, A. J. Smith, B. Birebent, E. Purev, L. Staib, R. Somasundaram, J. Zaloudik, W. Li, et al 2000. Colorectal cancer vaccines: anti-idiotypic antibody, recombinant protein, and viral vector. Ann. NY Acad. Sci. 910:237.[Abstract/Free Full Text]
  10. Reilly, R. T., M. B. C. Gottlieb, A. M. Ercolini, J.-P. H. Machiels, C. E. Kane, F. I. Okoye, W. J. Muller, K. H. Dixon, E. M. Jaffee. 2000. HER-2/neu is a tumor rejection target in tolerized HER-2/neu transgenic mice. Cancer Res. 60:3569.[Abstract/Free Full Text]
  11. Herlyn, D., M. Wettendorff, E. Schmoll, D. Iliopoulos, I. Schedel, U. Dreikhausen, R. Raab, A. H. Ross, H. Jaksche, M. Scriba, H. Koprowski. 1987. Anti-idiotype immunization in cancer patients: modulation of the immune response. Proc. Natl. Acad. Sci. USA 84:8055.[Abstract/Free Full Text]
  12. Nakamura, H., Y. Hinoda, N. Nakagawa, Y. Makiguchi, F. Itoh, T. Endo, K. Imai. 1998. Detection of circulating anti-MUC1 mucin core protein antibodies in patients with colorectal cancer. J. Gastroenterol. 33:354.[Medline]
  13. Stockert, E., E. Jäger, Y.-T. Chen, M. J. Scanlan, I. Gout, J. Karbach, M. Arand, A. Knuth, L. J. Old. 1998. A survey of the humoral immune response of cancer patients to a panel of human tumor antigens. J. Exp. Med. 187:1349.[Abstract/Free Full Text]
  14. Mosolits, S., U. Harmenberg, U. Ruden, L. Ohman, B. Nilsson, B. Wahren, J. Fagerberg, H. Mellstedt. 1999. Autoantibodies against the tumour-associated antigen GA733-2 in patients with colorectal carcinoma. Cancer Immunol. Immunother. 47:315.[Medline]
  15. Clarke, P., J. Mann, J. F. Simpson, K. Rickard-Dickson, F. J. Primus. 1998. Mice transgenic for human carcinoembryonic antigen as a model for immunotherapy. Cancer Res. 58:1469.[Abstract/Free Full Text]
  16. Kass, E., J. Schlom, J. Thompson, F. Guadagni, P. Graziano, J. W. Greiner. 1999. Induction of protective host immunity to carcinoembryonic antigen (CEA), a self-antigen in CEA transgenic mice, by immunizing with a recombinant vaccinia-CEA virus. Cancer Res. 59:676.[Abstract/Free Full Text]
  17. Mizobata, S., K. Tompkins, J. F. Simpson, Y. Shyr, F. J. Primus. 2000. Induction of cytotoxic T cells and their antitumor activity in mice transgenic for carcinoembryonic antigen. Cancer Immunol. Immunother. 49:285.[Medline]
  18. Tempero, R. M., M. L. VanLith, K. Morikane, G. J. Rowse, S. J. Gendler, M. A. Hollingsworth. 1998. CD4+ lymphocytes provide MUC1-specific tumor immunity in vivo that is undetectable in vitro and is absent in MUC1 transgenic mice. J. Immunol. 161:5500.[Abstract/Free Full Text]
  19. Acres, B., V. Apostolopoulos, J.-M. Balloul, D. Wreschner, P.-X. Xing, D. Ali-Hadji, N. Bizouarne, M. P. Kieny, I. F. C. McKenzie. 2000. MUC1-specific immune responses in human MUC1 transgenic mice immunized with various human MUC1 vaccines. Cancer Immunol. Immunother. 48:588.[Medline]
  20. Carr-Brendel, V., D. Markovic, K. Ferrer, M. Smith, J. Taylor-Papadimitriou, E. P. Cohen. 2000. Immunity to murine breast cancer cells modified to express MUC-1, a human breast cancer antigen, in transgenic mice tolerant to human MUC-1. Cancer Res. 60:2435.[Abstract/Free Full Text]
  21. Koido, S., M. Kashiwaba, D. Chen, S. Gendler, D. Kufe, J. Gong. 2000. Induction of antitumor immunity by vaccination of dendritic cells transfected with MUC1 RNA. J. Immunol. 165:5713.[Abstract/Free Full Text]
  22. Wilkinson, R. W., E. L. Ross, A. E. Lee-MacAry, R. Laylor, J. Burchell, J. Taylor-Papadimitriou, D. Snary. 2000. A transgenic mouse model for tumour immunotherapy: induction of an anti-idiotype response to human MUC1. Br. J. Cancer 83:1202.[Medline]
  23. Soares, M. M., V. Mehta, O. J. Finn. 2001. Three different vaccines based on the 140-amino acid Muc1 peptide with seven tandemly repeated tumor-specific epitopes elicit distinct immune effector mechanisms in wild-type versus Muc1 transgenic mice with different potential for tumor rejection. J. Immunol. 166:6555.[Abstract/Free Full Text]
  24. Wei, C., R. A. Willis, B. R. Tilton, R. J. Looney, E. M. Lord, R. Barth, J. G. Frelinger. 1997. Tissue-specific expression of the human prostate-specific antigen gene in transgenic mice: implications for tolerance and immunotherapy. Proc. Natl. Acad. Sci. USA 94:6369.[Abstract/Free Full Text]
  25. Romieu, R., M. Baratin, M. Kayibanda, V. Lacabanne, M. Ziol, J.-G. Guillet, M. Viguier. 1998. Passive but not active CD8+ T cell-based immunotherapy interferes with liver tumor progression in a transgenic mouse model. J. Immunol. 161:5133.[Abstract/Free Full Text]
  26. Cefai, D., B. W. Morrison, A. Sckell, L. Favre, M. Balli, M. Leunig, C. D. Gimmi. 1999. Targeting Her-2/neu for active-specific immunotherapy in a mouse model of spontaneous breast cancer. Int. J. Cancer 83:393.[Medline]
  27. Grosenbach, D. W., J. C. Barrientos, J. Schlom, J. W. Hodge. 2001. Synergy of vaccine strategies to amplify antigen-specific immune responses and antitumor effects. Cancer Res. 61:4497.[Abstract/Free Full Text]
  28. Greiner, J. W., H. Zeytin, M. R. Anver, J. Schlom. 2002. Vaccine-based therapy directed against carcinoembryonic antigen demonstrates antitumor activity on spontaneous intestinal tumors in the absence of autoimmunity. Cancer Res. 62:6944.[Abstract/Free Full Text]
  29. Atkinson, B., C. S. Ernst, B. F. D. Ghrist, M. Herlyn, M. Blaszczyk, A. H. Ross, D. Herlyn, Z. Steplewski, H. Koprowski. 1984. Identification of melanoma-associated antigens using fixed tissue screening of antibodies. Cancer Res. 44:2577.[Abstract/Free Full Text]
  30. Atkinson, B., C. S. Ernst, B. F. D. Ghrist, A. H. Ross, W. H. Clark. 1985. Monoclonal antibody to a highly glycosylated protein reacts in fixed tissue with melanoma and other tumors. Hybridoma 4:243.[Medline]
  31. Vennegoor, C., J. Calafat, P. Hageman, F. van Buttenen, H. Janssen, A. Kolk, P. Rümke. 1985. Biochemical characterization and cellular localization of a formalin-resistant melanoma-associated antigen reacting with monoclonal antibody NK1/C-3. Int. J. Cancer 35:287.[Medline]
  32. Sikora, L. K. J., A. Pinto, D. J. Demetric, W. T. Dixon, S. J. Urbanski, W. Temple, L. M. Jerry. 1987. Characterization of a novel neuroglandular antigen (NGA) expressed on abnormal human melanocytes. Int. J. Cancer 39:138.[Medline]
  33. Metzelaar, M. J., P. L. Wungaard, P. J. Peters, J. J. Sixma, H. K. Nieuwenhuis, H. C. Clevers. 1991. CD63 antigen: a novel lysosomal membrane glycoprotein, cloned by a screening procedure for intracellular antigens in eukaryotic cells. J. Biol. Chem. 266:3239.[Abstract/Free Full Text]
  34. Ernst, C. S., J.-W. Shen, S. Litwin, M. Herlyn, H. Koprowski, H. F. Sears. 1986. Multiparameter evaluation of the expression in situ of normal and tumor-associated antigens in human colorectal carcinoma. J. Natl. Cancer Inst. 77:387.
  35. Reisfeld, R. A.. 1993. Report of the international consensus conference on human melanoma antigens. Melanoma Res. 3:209.[Medline]
  36. Smith, M., R. Bleijs, K. Radford, P. Hersey. 1997. Immunogenicity of CD63 in a patient with melanoma. Melanoma Res. 7:S163.
  37. Hotta, H., A. H. Ross, K. Huebner, M. Isobe, S. Wendeborn, M. V. Chao, R. P. Ricciardi, Y. Tsujimoto, C. M. Croce, H. Koprowski. 1988. Molecular cloning and characterization of an antigen associated with early stages of melanoma tumor progression. Cancer Res. 48:2955.[Abstract/Free Full Text]
  38. Herlyn, M., W. H. Clark, Jr, M. J. Mastrangelo, D. P. Guerry IV, D. E. Elder, D. LaRossa, R. Hamilton, E. Bondi, R. Tuthill, Z. Steplewski, H. Koprowski. 1980. Specific immunoreactivity of monoclonal anti-melanoma antibodies. Cancer Res. 40:3602.[Abstract/Free Full Text]
  39. Demetrick, D. J., D. Herlyn, M. Tretiak, D. Creasey, H. Clevers, L. A. Donoso, C. J. Vennegoor, W. T. Dixon, L. M. Jerry. 1992. The ME491 melanoma-associated glycoprotein family: antigenic identification of the ME491, NKI/C-3, NGA and CD63 proteins. J. Natl. Cancer Inst. 84:422.[Abstract/Free Full Text]
  40. Herlyn, D., J. Powe, A. H. Ross, M. Herlyn, H. Koprowski. 1985. Inhibition of human tumor growth by IgG2a monoclonal antibodies correlates with antibody density on tumor cells. J. Immunol. 134:1300.[Abstract]
  41. Scatchard, G.. 1949. The attraction of proteins for small molecules and ions. Ann. NY Acad. Sci. 51:660.
  42. Li, W., K. Berencsi, S. Basak, R. Somasundaram, R. P. Ricciardi, E. Gonczol, J. Zaloudik, A. Linnenbach, H. Maruyama, P. Miniou, D. Herlyn. 1997. Human colorectal cancer (CRC) antigen CO17-1A/GA733 encoded by recombinant adenovirus inhibits growth of established CRC cells in mice. J. Immunol. 159:763.[Abstract]
  43. Kieny, M. P., R. Lathe, J. P. Lecocq. 1983. New versatile cloning and sequencing vectors based on bacteriophage M13. Gene 26:91.[Medline]
  44. Kieny, M. P., G. Rautmann, D. Schmitt, K. Dott, S. Wain-Hobson, M. Girard, S. Chamaret, A. Laurent, L. Montagnier, J. P. Lecocq. 1986. AIDS virus env protein expressed from a recombinant vaccinia virus. Biotechnology 4:790.
  45. Herlyn, D., R. Somasundaram, J. Zaloudik, W. Li, L. Jacob, D. Harris, M.-P. Kieny, R. Ricciardi, E. Gonczol, H. F. Sears, M. J. Mastrangelo. 1995. Cloned antigens and anti-idiotypes. Hybridoma 14:159.[Medline]
  46. Zaloudik, J., W. Li, L. Jacob, M. P. Kieny, R. Somasundaram, B. Acres, H. Song, T. Zhang, J. Li, D. Herlyn. 2002. Inhibition of tumor growth by recombinant vaccinia virus expressing GA733/CO17-1A/EpCAM/KSA/KS1-4 antigen in mice. Cancer Gene Ther. 9:382.[Medline]
  47. Bronte, V., K. Tsung, J. B. Rao, P. W. Chen, M. Wang, S. A. Rosenberg, N. P. Restifo. 1995. IL-2 enhances the function of recombinant poxvirus-based vaccines in the treatment of established pulmonary metastases. J. Immunol. 154:5282.[Abstract]
  48. Apostolopoulos, V., G. Pietersz, B. Loveland, M. Sandrin, I. McKenzie. 1995. Oxidative/reductive conjugation of mannan to antigen selects for T1 or T2 immune responses. Proc. Natl. Acad. Sci. USA 92:10128.[Abstract/Free Full Text]
  49. Öhlén, C., M. Kalos, D. J. Hong, A. C. Shur, P. D. Greenberg. 2001. Expression of a tolerizing tumor antigen in peripheral tissue does not preclude recovery of high-affinity CD8+ T cells or CTL immunotherapy of tumors expressing the antigen. J. Immunol. 166:2863.[Abstract/Free Full Text]
  50. Speiser, D. E., R. Miranda, A. Zakarian, M. F. Bachmann, K. McKall-Faienza, B. Odermatt, D. Hanahan, R. M. Zinkernagel, P. S. Ohashi. 1997. Self antigens expressed by solid tumors do not efficiently stimulate native or activated T cells: implications for immunotherapy. J. Exp. Med. 186:645.[Abstract/Free Full Text]
  51. Vierboom, M. P., H. W. Nijman, R. Offringa, E. I. van-der-Voort, T. van-Hall, L. van-den-Broek, G. J. Fleuren, P. Kenemans, W. M. Kast, C. J. Melief. 1997. Tumor eradication by wild-type p53-specific cytotoxic T lymphocytes. J. Exp. Med. 186:695.[Abstract/Free Full Text]
  52. Morgan, D., H. Kreuwel, S. Fleck, H. Levitsky, D. Pardoll, L. Sherman. 1998. Activation of low avidity CTL specific for a self epitope results in tumor rejection but not autoimmunity. J. Immunol. 160:643.[Abstract/Free Full Text]
  53. Ohashi, P. S., S. Oehen, K. Buerki, H. Pircher, C. T. Ohashi, B. Odermatt, B. Malissen, R. M. Zinkernagel, H. Hengartner. 1991. Ablation of "tolerance" and induction of diabetes by virus infection in vitro in antigen transgenic mice. Cell 65:305.[Medline]
  54. Lees, C. J., V. Apostolopoulos, B. A. Acres, C. S. Ong, V. Popovski, I. F. C. McKenzie. 2000. The effect of T1 and T2 cytokines on the cytotoxic T cell response to mammalian-MUC1. Cancer Immunol. Immunother. 48:644.[Medline]
  55. Tsang, K. Y., S. Zaremba, C. A. Nieroda, M. Z. Zhu, J. M. Hamilton, J. Schlom. 1995. Generation of human cytotoxic T cells specific for human carcinoembryonic antigen epitopes from patients immunized with recombinant vaccinia-CEA vaccine. J. Natl. Cancer Inst. 87:949.[Free Full Text]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
S. Maia, W. N. Haining, S. Ansen, Z. Xia, S. A. Armstrong, N. P. Seth, P. Ghia, M. L. den Boer, R. Pieters, S. E. Sallan, et al.
Gene Expression Profiling Identifies BAX-{delta} as a Novel Tumor Antigen in Acute Lymphoblastic Leukemia
Cancer Res., November 1, 2005; 65(21): 10050 - 10058.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Li, J.
Right arrow Articles by Herlyn, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Li, J.
Right arrow Articles by Herlyn, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS