This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stewart, T. J. Articles by Leggatt, G. R. Search for Related Content PubMed PubMed Citation Articles by Stewart, T. J. Articles by Leggatt, G. R.

Inhibition of Early Tumor Growth Requires J18-positive (Natural Killer T) Cells¹

Centre for Immunology and Cancer Research, University of Queensland, Princess Alexandra Hospital, Brisbane, Queensland, 4102 [T. J. S., G. J. P. F., I. H. F., G. R. L.], and Cancer Immunology Program, Peter MacCallum Cancer Institute, East Melbourne, Victoria, 8006 [M. J. S.], Australia

	ABSTRACT

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
The role of natural killer T (NKT) cells in the immune response to tumor cells has been largely unexplored. As a model of adoptive tumor immunotherapy, cells from the draining lymph nodes of mice immunized with a tumor-specific or irrelevant antigen were transferred to naïve recipients with established tumor. Inhibition of early tumor growth (day 4) required the transfer of both CD8⁺ and J18⁺ (NKT) cells from immunized animals without regard to immunogen. In contrast, CD8⁺ cells, but not J18⁺ cells, were necessary for the inhibition of late tumor growth (day 8). Thus, the developing tumor changes in sensitivity to NKT-mediated events and the role for NKT cells cannot be replaced by the presence of tumor-specific cells during early tumor growth. This suggests that recruitment/activation of J18⁺ NKT cells is an important consideration during the immune therapy of early stage tumors.

	Introduction

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
NKT³ cells are a lymphocytic population that coexpress cell surface markers of both the NK- and T-cell lineage. Most murine NKT cells are either CD4⁺CD8^- or CD4^-CD8^-and express a highly restricted TCR repertoire consisting of an invariant V14J18 TCR chain in association with a restricted Vß repertoire (1) . The TCR on NKT cells recognizes glycolipids presented by the CD1d molecule that is expressed by many different cell types. On stimulation through the TCR, NKT cells have the capacity to produce large amounts of proinflammatory cytokines rapidly, including IFN- and tumor necrosis factor, in addition to immunoregulatory cytokines including IL-4 and IL-10. NKT cell function and regulation is not well understood, but accumulating evidence suggests a role in the regulation and differentiation of adaptive immune responses (2) . The physiological role of NKT cells in tumor immunity may be multifaceted, with one study suggesting that NKT cell production of IL-13 suppresses the immune response to tumor cells, whereas others suggest a direct role for NKT cells in the control of methylcholanthrene-induced fibrosarcomas (3 , 4) . Our study addresses the role of adoptive immunotherapy, using mixed cell types, in the treatment of an established HPV16E7-expressing tumor, TC-1 (5) . Using this model, we have shown that different immune cells regulate tumor growth at different time points after TC-1 tumor implantation. Whereas transferred CD8⁺ cells are important regulators of tumor growth at all of the time points, transferred NKT cells are only necessary in the early stages of tumor development.

	Materials and Methods

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
Mice.
Female C57BL/6, C57BL/6J-Rag1^tm1Mom (Rag-1^-/-), and C.B-17-Igh-1^b-prkdc^scid (SCID) mice were obtained at 6–8 weeks of age from the Animal Resources Centre (Perth, Western Australia, Australia). B6.J18^-/- gene targeted mice have been described previously (6) . All of the mice were housed at the Princess Alexandra Hospital Biological Research Facility (Brisbane, Queensland, Australia). All of the mice entered experiments at 6–10 weeks of age, and were age and sex matched within experiments. The animal ethics committee of the University of Queensland approved all of the experiments.

Proteins and Adjuvant.
HPV16E7GST fusion protein was prepared as described by Fernando et al. (7) . Myo (Sigma, St. Louis, MO) was prepared at 10 mg/ml in distilled water. The saponin, Quil A (Spikoside; ISCOTEC A B, Lulea, Sweden), was used as an adjuvant in all of the immunizations.

Cell Lines and Culture.
TC-1, a tumorigenic, H-2^b cell line expressing the E6 and E7 proteins of HPV16 was obtained from Tzyy-Choou Wu (Ref. 5 ; John Hopkins University Medical School, Baltimore, MD). In preparation for implantation into mice, TC-1 cells were cultured to >80% confluence, harvested by trypsinization, and washed in PBS before injection. CTLs were generated from LNC and assayed as described previously (8) .

Tumor Assay.
Mice were challenged s.c. between the shoulder blades (neck scruff) with 5 x 10⁵ TC-1 tumor cells. One hundred percent of C57BL/6 mice challenged with this tumor dose developed tumors. Tumors were allowed to establish for defined periods before i.v. transfer of LNC. Except where indicated, mice were sacrificed at day 14 after tumor implantation, and the tumor excised and weighed. Each group consisted of at least 5 mice.

Cells for adoptive transfer were generated from mice immunized s.c. in both footpads by injection of either 50 µg HPV16E7GST or Myo and 10 µg of Quil A. After 4 days, the popliteal and periaortic lymph nodes were removed, and a single cell suspension was prepared by mechanical disruption. Dissociated cells, suspended in PBS, were then transferred into the tail vein (i.v.) of tumor-bearing mice.

Lymphocyte Depletions and Antibody Staining.
CD4 and CD8 Dynabeads (Dynal Biotech Pty. Ltd., Carlton, Victoria, Australia) were used for the depletion of CD4⁺ and/or CD8⁺ cells from the LNC suspension. Depletions were performed according to manufacturer’s instructions. The remaining cells were analyzed for depletion efficiency, before transfer, by analysis of antibody stained cells on a FACSCalibur (BD Biosciences, San Jose, CA). Sample depletion efficiency was calculated as: [1 -(percentage of stained cells in the depleted group/percentage of stained cells in an untreated group)] x 100.

Statistical Analysis.
Experimental groups were compared using the two-tailed, unpaired Student’s t test. Error bars represent the SE, and Ps < 0.05 were considered significant.

	Results and Discussion

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
Immunotherapy of TC-1 Tumor Established for 4 Days Requires CD8⁺ and J18⁺ Cells.
TC-1 is a rapidly growing, transplantable, murine tumor expressing the E7 protein from HPV type 16. A model for adoptive immunotherapy of TC-1 tumor was established using LNC from HPV16E7 immunized mice at day 4 after immunization as effector cells. LNC from mice immunized with either HPV16E7 (tumor antigen) or Myo protein (protein control) were capable of delaying tumor growth (Fig. 1A) , whereas the transfer of LNC from unimmunized mice had no apparent effect on tumor growth (Fig. 1A) . In additional studies (data not shown) transfer of LNC from mice given adjuvant alone (Quil A) was sufficient to reduce growth of the TC-1 tumor. Tumor growth was also reduced more effectively by E7 immune LNC than by Myo immune LNC when measured at multiple time points after cell transfer, although neither transfer prevented eventual tumor progression (Fig. 1B) . These observations suggested that immune targeting of tumor antigen improved adoptive therapy of TC-1, but was not necessary when cells were transferred at day 4 after tumor establishment.

View larger version (34K): [in this window] [in a new window] [Download PPT slide]   Fig. 1. Inhibition of tumor growth by donor LNC administered 4 days after tumor cell inoculation in the recipient is tumor antigen-nonspecific and ablated by removal of CD8+ or J18+ cells. A, 4 days after TC-1 tumor inoculation in a recipient mouse, LNCs from donor mice immunized with either HPV16E7 (E7) or Myo (MYO), or LNCs from untreated donor mice were transferred to the recipient. The s.c. tumor mass was excised and weighed at day 14. B, donor LNC from mice immunized with either E7 () or Myo () were transferred at day 4 (indicated by the arrow) into C57BL/6 recipient mice challenged with tumor on day 0. Tumor weights were recorded at various times after tumor inoculation. Tumor-bearing recipient mice without donor cell transfer are also shown at a single time point (). C and D, LNC from immunized donor mice were depleted of CD4+, CD8+, or CD4+ and CD8+ cells, or mock-depleted with control antibody (ISO) before transfer into C57BL/6 (C) or SCID (D) tumor-bearing recipient mice 4 days after tumor challenge. Fourteen days after tumor challenge, tumor masses were excised and weighed. CD8 and CD4 cell depletions were successful in removing >75% of CD8 cells and >95% of the CD4 cells as determined by flow cytometry. E, LNC from J18-/- (, ) or J18+/+ (, ) mice immunized with E7 protein were stimulated with E7 peptide and IL-2 for 5 days in vitro before assaying lytic activity. Cultured LNC were assayed for lytic activity, at different E:T ratios, using E7-expressing EL4 (,) or control EL4 (,) cell lines as targets. F, 4 days after tumor inoculation, tumor-bearing recipient mice received donor LNC from immunized J18+/+ or J18-/- mice. Tumors were excised and weighed at day 14. In all tumor experiments, data are representative of one of two similar experiments using 5 mice/group. Data are presented as mean; bars, ± SE. *, P < 0.05 for test group versus no transfer.

To investigate the role of NKT cells in immune LNC-mediated control of recently established tumor, J18^-/- mice, lacking V14 NKT cells, were immunized with E7 or Myo to generate LNC effectors. Comparable numbers of E7-specific CTLs (Fig. 1E) were generated in E7 immunized J18^-/- mice and J18^+/+ mice at day 4 after immunization. In addition, the J18^+/+ mice and J18^-/- mice had similar proportions of CD4⁺ cells (35.9% versus 32.8%, respectively) and CD8⁺ cells (27.3% versus 25.7%, respectively), as well as activated CD4⁺CD69⁺ cells (8.3% versus 6.1%, respectively) and CD8⁺CD69⁺ cells (7.4% versus 7.6%, respectively) in the draining LNC after immunization. However, immune LNC from J18^-/- mice, immunized with either E7 or Myo protein, failed to alter the growth of a 4-day established TC-1 tumor, in contrast with LNC from similarly immunized J18^+/+ mice (Fig. 1F) . Thus, delayed growth of a recently established tumor requires the transfer of CD8⁺ and J18⁺ cells, and T cells induced to a tumor antigen in the absence of J18 NKT cells fail to achieve tumor control.

One model compatible with our data would suggest that a CD8⁺J18⁺ cell was responsible for the delay in tumor outgrowth. This was unlikely given that J18⁺ NKT cells are almost exclusively CD4⁺, although populations of CD4^-/CD8^- J18⁺ cells may exist (9) . Consequently, it is more likely that CD8 and J18 proteins are expressed on separate cell types. Given the limited numbers of NKT cells in the donor LNC (based on numbers of NK1.1⁺ cells in the LNC transfers; data not shown), it is possible NKT cells play a role in the efficient priming of the antitumor response within the draining lymph node of the donor, before cell transfer. This role in immune priming does not apparently involve the generation of tumor antigen-specific CD8⁺ CTL numbers, as similar numbers of cytolytic T cells can be generated in J18^-/- mice and C57BL/6 mice (Fig. 1E) . Immune priming may also include other cell types and, to this end, NKT cells have been proposed to enhance the activity of NK cells and dendritic cells (10 , 11) . Alternatively, small numbers of transferred NKT cells may sustain the immune response or enhance homing of immune cells to the tumor site in the recipient. One novel finding was that CD8⁺ cells specific for tumor antigen were incapable of limiting tumor growth in the absence of J18⁺ cells, consistent with the idea that NKT cells coordinate the antitumor response at this early time point in tumor formation.

Consequently, we favor a model whereby NKT cells, CD8⁺ cells, and possibly additional cell types collaborate to prevent the outgrowth of day 4 tumor. Transferred NKT cells and/or CD8⁺ cells may initiate or enhance the activity of additional effector populations (i.e., NK cells or macrophages) present in either the donor or the recipient cell populations.

Immunotherapy of TC-1 Tumor Established for 8 Days Requires CD8⁺ Cells and Is Directed at the E7 Tumor Antigen.
To establish whether NKT and CD8⁺ cells also contribute to the control of more established tumors, we examined the ability of passively transferred lymphocytes to alter the growth of tumors established 8 days previously. In contrast to the immunotherapy of TC-1 tumors established for 4 days, LNC from mice immunized with Myo were unable to delay the growth of tumor when transferred at day 8 after tumor inoculation (Fig. 2, A and B) . However, LNC from mice immunized with the tumor antigen (E7) were capable of inducing regression of tumor mass at this time point (Fig. 2B) .

View larger version (28K): [in this window] [in a new window] [Download PPT slide]   Fig. 2. Inhibition of tumor growth by donor LNC administered 8 days after tumor cell inoculation in the recipient is tumor antigen-specific and involves CD8+ cells. A, C57BL/6 mice were inoculated with TC-1 tumor cells 8 days before transfer of donor LNC from mice immunized with either E7 or MYO. Tumor was excised and weighed on day 14. B, donor LNC from mice immunized with E7 () or MYO () were transferred 8 days (indicated by the arrow) after inoculation with tumor cells into C57BL/6 recipient mice. Tumor weights were taken at various times after tumor cell challenge. Tumor-bearing recipient mice without donor cell transfer are also shown (). C and D, LNC from E7-immunized mice were depleted as shown in Fig. 1C before transfer into C57BL/6 (C) or RAG-1-/- (D) recipient mice 8 days after tumor challenge. E, 8 days after tumor inoculation, mice received LNC from J18+/+ or J18-/- mice immunized previously with E7. Tumors were excised and weighed at day 14. In all tumor experiments, data are representative of one of two similar experiments using 5 mice/group. Data represent the mean; bars, ± SE. *, P < 0.05 for test group versus no transfer.

In contrast to the observation for 4-day established tumors, transfer of LNC from E7-immunized, J18^-/- mice reduced the growth of 8-day, established tumors, suggesting that an effector mechanism independent of NKT cells was operative with established day-8 tumors (Fig. 2E) .

Inhibition of growth of 8-day established tumor was exclusively mediated by tumor antigen-specific LNC, and J18⁺ cells were not necessary to induce or deliver effector function. The specificity of this antitumor response contrasts with that of 4-day established tumors, and suggests that conventional TCR ligation by peptide/MHC is an important feature of this response, particularly given the requirement for CD8⁺ cells. An additional antitumor mechanism is revealed by depleting transferred LNC of both CD4 and CD8 cells, suggesting that this mechanism is normally suppressed by T cells. Interestingly, this alternate antitumor effect is not apparent during early tumor growth (day 4 after tumor inoculation)

In this study, we have demonstrated that the control of growth of a transplantable tumor by immune effector cells involves more than antigen-specific CD8 CTL killing of tumor cells. Activated NKT cells are therapeutically effective early in the development of the tumor, whereas tumor antigen-specific cells are required at later time points. In circumstances whereby MHC class I restricted tumor antigens are lost or down-regulated, the demonstration of additional effector mechanisms provides alternative pathways for tumor immunotherapy.

	ACKNOWLEDGMENTS

 
We thank Barbara Murray (CICR) for HPV16E7GST-protein purification, David Wiseman (CICR) and Caron Maxim (CICR) for the care of animals, and Dr. Tzyy-Choou Wu (Johns Hopkins University Medical School, Baltimore, MD) for the gift of the TC-1 cell line. We also thank Dr. Jay Berzofsky (NIH, Bethesda, MD) and Dr. Masaki Terabe (NIH, Bethesda, MD) for critical reading of the manuscript, and Prof. Masaru Taniguchi (Chiba University, Chiba, Japan) for the J18^-/- mice.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Supported by funding from the University of Queensland.

² To whom requests for reprints should be addressed, at Centre for Immunology and Cancer Research, The University of Queensland, Research Wing, Building 1, Princess Alexandra Hospital, Woolloongabba, Queensland, Australia, 4102. Phone: 61-7-3240-5393; Fax: 61-7-3240-5946; E-mail: gleggatt{at}cicr.uq.edu.au

³ The abbreviations used are: NKT, natural killer T; HPV, human papillomavirus; LNC, lymph node cell; Myo, myoglobin; CTL, cytotoxic T lymphocyte; CICR, Centre for Immunology and Cancer Research; NK, natural killer; TCR, T-cell receptor; IL, interleukin.

Received 12/20/02. Accepted 4/25/03.

	REFERENCES

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 

1. Lantz O., Bendelac A. An invariant T cell receptor chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4–8- T cells in mice and humans. J. Exp. Med., 180: 1097-1106, 1994.[Abstract/Free Full Text]
2. Bendelac A., Rivera M. N., Park S. H., Roark J. H. Mouse CD1-specific NK1 T cells: development, specificity, and function. Annu. Rev. Immunol., 15: 535-562, 1997.[Medline]
3. Terabe M., Matsui S., Noben-Trauth N., Chen H., Watson C., Donaldson D. D., Carbone D. P., Paul W. E., Berzofsky J. A. NKT cell-mediated repression of tumor immunosurveillance by IL-13 and the IL-4R-STAT6 pathway. Nat. Immunol., 1: 515-520, 2000.[Medline]
4. Crowe N. Y., Smyth M. J., Godfrey D. I. A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. J. Exp. Med., 196: 119-127, 2002.[Abstract/Free Full Text]
5. Lin K. Y., Guarnieri F. G., Staveley-O’Carroll K. F., Levitsky H. I., August J. T., Pardoll D. M., Wu T. C. Treatment of established tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res., 56: 21-26, 1996.[Abstract/Free Full Text]
6. Cui J., Shin T., Kawano T., Sato H., Kondo E., Toura I., Kaneko Y., Koseki H., Kanno M., Taniguchi M. Requirement for V14 NKT cells in IL-12-mediated rejection of tumors. Science (Wash. DC), 278: 1623-1626, 1997.[Abstract/Free Full Text]
7. Fernando G. J., Murray B., Zhou J., Frazer I. H. Expression, purification and immunological characterization of the transforming protein E7, from cervical cancer-associated human papillomavirus type 16. Clin. Exp. Immunol., 115: 397-403, 1999.[Medline]
8. Leggatt G. R., Dunn L. A., De Kluyver R. L., Stewart T., Frazer I. H. Interferon- enhances cytotoxic T lymphocyte recognition of endogenous peptide in keratinocytes without lowering the requirement for surface peptide. Immunol. Cell Biol., 80: 415-424, 2002.[Medline]
9. Apostolou I., Cumano A., Gachelin G., Kourilsky P. Evidence for two subgroups of CD4-CD8- NKT cells with distinct TCR ß repertoires and differential distribution in lymphoid tissues. J. Immunol., 165: 2481-2490, 2000.[Abstract/Free Full Text]
10. Kitamura H., Iwakabe K., Yahata T., Nishimura S., Ohta A., Ohmi Y., Sato M., Takeda K., Okumura K., Van Kaer L., Kawano T., Taniguchi M., Nishimura T. The natural killer T (NKT) cell ligand -galactosylceramide demonstrates its immunopotentiating effect by inducing interleukin (IL)- 12 production by dendritic cells and IL-12 receptor expression on NKT cells. J. Exp. Med., 189: 1121-1128, 1999.[Abstract/Free Full Text]
11. Smyth M. J., Crowe N. Y., Godfrey D. I. NK cells and NKT cells collaborate in host protection from methylcholanthrene-induced fibrosarcoma. Int. Immunol., 13: 459-463, 2001.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. Ostrand-Rosenberg and P. Sinha Myeloid-Derived Suppressor Cells: Linking Inflammation and Cancer J. Immunol., April 15, 2009; 182(8): 4499 - 4506. [Abstract] [Full Text] [PDF]

				G. Bricard, V. Cesson, E. Devevre, H. Bouzourene, C. Barbey, N. Rufer, J. S. Im, P. M. Alves, O. Martinet, N. Halkic, et al. Enrichment of Human CD4+ V{alpha}24/V{beta}11 Invariant NKT Cells in Intrahepatic Malignant Tumors J. Immunol., April 15, 2009; 182(8): 5140 - 5151. [Abstract] [Full Text] [PDF]

				H.-J. Ko, J.-M. Lee, Y.-J. Kim, Y.-S. Kim, K.-A Lee, and C.-Y. Kang Immunosuppressive Myeloid-Derived Suppressor Cells Can Be Converted into Immunogenic APCs with the Help of Activated NKT Cells: An Alternative Cell-Based Antitumor Vaccine J. Immunol., February 15, 2009; 182(4): 1818 - 1828. [Abstract] [Full Text] [PDF]

				H. Wakao, R. Wakao, S. Sakata, K. Iwabuchi, A. Oda, and H. Fujita In vitro induction of natural killer T cells from embryonic stem cells prepared using somatic cell nuclear transfer FASEB J, July 1, 2008; 22(7): 2223 - 2231. [Abstract] [Full Text] [PDF]

				G. J. Renukaradhya, M. A. Khan, M. Vieira, W. Du, J. Gervay-Hague, and R. R. Brutkiewicz Type I NKT cells protect (and type II NKT cells suppress) the host's innate antitumor immune response to a B-cell lymphoma Blood, June 15, 2008; 111(12): 5637 - 5645. [Abstract] [Full Text] [PDF]

				J. A. Berzofsky and M. Terabe NKT Cells in Tumor Immunity: Opposing Subsets Define a New Immunoregulatory Axis J. Immunol., March 15, 2008; 180(6): 3627 - 3635. [Abstract] [Full Text] [PDF]

				E. Ambrosino, M. Terabe, R. C. Halder, J. Peng, S. Takaku, S. Miyake, T. Yamamura, V. Kumar, and J. A. Berzofsky Cross-Regulation between Type I and Type II NKT Cells in Regulating Tumor Immunity: A New Immunoregulatory Axis J. Immunol., October 15, 2007; 179(8): 5126 - 5136. [Abstract] [Full Text] [PDF]

				J. W. Molling, J. A.E. Langius, J. A. Langendijk, C. R. Leemans, H. J. Bontkes, H. J.J. van der Vliet, B. M. E. von Blomberg, R. J. Scheper, and A. J.M. van den Eertwegh Low Levels of Circulating Invariant Natural Killer T Cells Predict Poor Clinical Outcome in Patients With Head and Neck Squamous Cell Carcinoma J. Clin. Oncol., March 1, 2007; 25(7): 862 - 868. [Abstract] [Full Text] [PDF]

				M. Terabe, J. Swann, E. Ambrosino, P. Sinha, S. Takaku, Y. Hayakawa, D. I. Godfrey, S. Ostrand-Rosenberg, M. J. Smyth, and J. A. Berzofsky A nonclassical non-V{alpha}14J{alpha}18 CD1d-restricted (type II) NKT cell is sufficient for down-regulation of tumor immunosurveillance J. Exp. Med., December 19, 2005; 202(12): 1627 - 1633. [Abstract] [Full Text] [PDF]

				D. H. Chang, K. Osman, J. Connolly, A. Kukreja, J. Krasovsky, M. Pack, A. Hutchinson, M. Geller, N. Liu, R. Annable, et al. Sustained expansion of NKT cells and antigen-specific T cells after injection of {alpha}-galactosyl-ceramide loaded mature dendritic cells in cancer patients J. Exp. Med., May 2, 2005; 201(9): 1503 - 1517. [Abstract] [Full Text] [PDF]

				K. Matsumoto, G. R. Leggatt, J. Zhong, X. Liu, R. L. de Kluyver, T. Peters, G. J. P. Fernando, A. Liem, P. F. Lambert, and I. H. Frazer Impaired Antigen Presentation and Effectiveness of Combined Active/Passive Immunotherapy for Epithelial Tumors J Natl Cancer Inst, November 3, 2004; 96(21): 1611 - 1619. [Abstract] [Full Text] [PDF]

				J. E. Gumperz CD1d-restricted "NKT" cells and myeloid IL-12 production: an immunological crossroads leading to promotion or suppression of effective anti-tumor immune responses? J. Leukoc. Biol., August 1, 2004; 76(2): 307 - 313. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stewart, T. J. Articles by Leggatt, G. R. Search for Related Content PubMed PubMed Citation Articles by Stewart, T. J. Articles by Leggatt, G. R.