HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services 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 Karnbach, C. Articles by Nakamura, M. C. Search for Related Content PubMed PubMed Citation Articles by Karnbach, C. Articles by Nakamura, M. C.

Immune Rejection of a Large Sarcoma Following Cyclophosphamide and IL-12 Treatment Requires Both NK and NK T Cells and Is Associated with the Induction of a Novel NK T Cell Population¹

Department of Medicine, University of California, San Francisco, CA 94143; and Veterans Administration Medical Center, San Francisco, CA 94121

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Combined immunotherapy with cyclophosphamide (Cy) and IL-12, but not IL-12 alone, stimulates eradication of a large established solid tumor (20 mm), MCA207, a methylcholanthrene-induced murine sarcoma. In these studies we demonstrate that NK1.1⁺ cells and CD1d-dependent NK T cells each play important yet distinct roles in regression of a large tumor in response to Cy and IL-12, and we define a novel NK T cell subset, selectively increased by this treatment. Mice depleted of NK1.1⁺ cells demonstrated more rapid initial tumor growth and prolonged tumor regression following treatment, but tumors were eventually eradicated. In contrast, initial tumor regression following therapy was unimpaired in CD1d^-/- mice, which are deficient in most NK T cells, but tumors recurred. No tumor regression occurred following Cy and IL-12 therapy in CD1d^-/- mice that were depleted of NK1.1⁺ cells. We found that Cy and IL-12 induced the selective increase in liver and spleen lymphocytes of a unique NK T subpopulation (DX5⁺NK1.1^-CD3⁺). These cells were not induced by treatment in CD1d^-/- mice. Our studies demonstrate a contribution of both NK and NK T cells to the Cy- and IL-12-stimulated anti-tumor response. We describe the selective induction of a distinct NK T cell subset by Cy and IL-12 therapy, not seen following IL-12 therapy alone, which we suggest may contribute to the successful anti-tumor response induced by this immunotherapeutic regimen.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Administration of IL-12 generates potent anti-metastatic and antitumor activity against certain solid tumors in mice, although complete IL-12-dependent tumor eradication can generally be obtained only when tumors are small (1, 2, 3, 4, 5, 6, 7). The addition of the alkylating agent cyclophosphamide (Cy)⁵ to IL-12 immunotherapy enhances tumor rejection and can enable the complete eradication of a large (20-mm diameter), established murine methylcholanthrene (MCA)-induced sarcoma (MCA207) in C57BL/6 mice (8). While initially described as an NK stimulatory factor, the anti-tumor effects of IL-12 are multiple and include activation of NK, NK T, and/or T cells in addition to immune cell-independent anti-angiogenic effects (9, 10, 11). The mechanisms by which Cy may augment the effects of IL-12 and enable immune rejection of a larger tumor burden are not established.

Previous studies have demonstrated important roles for NK and/or NK T cells in IL-12-mediated regression of certain tumors (1, 4, 7, 12, 13, 14), with the relative roles dependent on both tumor model and dose of IL-12 (15, 16). NK cells, having lower IL-12R expression than NK T cells, require correspondingly higher IL-12 doses to be fully activated (17). Cui et al. (12) were the first to show a definitive role for NK T cells in IL-12-induced tumor rejection, using J281^-/- mice that lack V14⁺ NK T cells. These studies, however, examined models of microdisease, where treatment was commenced within the first day following tumor injection. In cases of large established tumors or in models of highly metastatic disease, IL-12 alone is often insufficient to induce complete tumor rejection. In these cases, combination therapies using IL-12 together with other agents such as IL-2 (18), IL-15 (19), IL-18 (20), TNF- (21), inducing protein-10 (22), or Cy (8) have proved considerably more effective.

Cy is an alkylating agent that can prevent cell division by cross-linking DNA strands leading to apoptosis of certain tumors (23). Treatment with Cy alone, however, has minimal effects on tumor growth in the MCA207 sarcoma model, suggesting that direct anti-tumor effects are insufficient to explain its synergism with IL-12 (8). In addition to its cytotoxic effect on tumors, Cy is known to have multiple dose-dependent effects on the immune system, which can include both activation and suppression. Thus, Cy can stimulate delayed-type hypersensitivity, and it has been proposed that Cy may act as a facilitator of a Th1 anti-tumor response stimulated by IL-12 (8, 24). Such immunopotentiation has been commonly noted with low dose Cy. Cy has been shown to down-regulate TGF-, IL-10, and NO levels in the spleen (25) and to preferentially delete (or to inactivate) an immunosuppressive Thy1.2⁺ (24) or CD4⁺ cell population in the spleen (26). Thus, a possible action for Cy may be the deletion (or the alteration of the cytokine profile) of a suppressive immune cell population, facilitating the anti-tumor effect of IL-12-activated cells. Alternatively, Cy has also been shown to induce tumor expression of chemotactic factors that attract activated NK cells (27), suggesting that Cy could have a direct action on the recruitment of cytotoxic cells to the tumor site.

In this study we investigated the role of NK cells and NK T cells in tumor growth and regression of large (20-mm) established MCA207 tumors in response to therapy with Cy and IL-12. We administered Cy and IL-12 for treatment of established MCA207 tumors in C57BL/6 mice depleted of NK1.1⁺ cells, in mice deficient in NK T cells due to a targeted deletion of CD1d (CD1d^-/-), and in CD1d^-/- mice depleted of NK1.1⁺ cells. Our results demonstrate that both NK and NK T cells are important in Cy- and IL-12-mediated MCA207 tumor regression and that they play distinct roles in the anti-tumor response. NK cells promote rapid regression of MCA207 sarcomas in response to Cy and IL-12, but NK T cells are required for complete tumor eradication. In the absence of both NK and NK T cells, Cy and IL-12 fail to induce tumor regression, demonstrating that the anti-tumor effect is dependent on the combination of NK and NK T cells. The effects of Cy and IL-12 on tumor regression show an interesting correlation with an increase in a unique subpopulation of NK T cells not previously described; Cy and IL-12 treatment led to a marked and selective increase in liver and splenic NK T cells with the phenotype DX5⁺NK1.1^-CD3⁺ (D-NK T cells). This increase did not occur when cells were treated with IL-12 alone, demonstrating that the increase in D-NK T cells is a unique feature of the combined immunotherapy with Cy and IL-12. The D-NK T population also failed to increase following Cy and IL-12 therapy in CD1d^-/- mice, in which tumor eradication does not occur. Our studies demonstrate a contribution of both NK and NK T cells in the Cy- and IL-12-stimulated anti-tumor response. We tentatively suggest that selective induction of the unique D-NK T cell subset by Cy and IL-12 may contribute to the successful eradication of a large solid tumor by this therapy.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice

Female C57BL/6 mice were purchased from Simonsen Laboratories (Gilroy, CA) at 6–8 wk of age. CD1d^-/- mice on a C57BL/6 background were bred from homozygous breeding pairs that were the generous gift of A. Bendelac (Princeton University, Princeton, NJ) (28). Mice had been backcrossed onto the C57BL/6 background for at least seven generations. All animals were housed in the approved San Francisco Veterans Affairs Medical Center animal care facilities according to institutional guidelines.

Tumor model

MCA207, a 3-methylcholanthrene-induced fibrosarcoma cell line from C57BL/6 mice, was provided by Dr. J. A. Norton (Veterans Affairs Medical Center, San Francisco, CA) and has been previously described (5). All mice were between 12 and 14 wk of age at the time of tumor cell inoculation. MCA207 tumor cells were cultured in vitro in RPMI supplemented with 10% FBS, 2-ME, glutamine, and penicillin/streptomycin. MCA207 tumor cells (1 x 10⁶) in 0.2 ml saline were injected s.c. in the left flank of mice. Tumor diameter was measured twice a week with a caliper. As previously described by Tsung et al. (8), when tumor size reached 20 mm in diameter, animals were treated with cyclophosphamide (100 mg/kg i.p. in 0.2 ml saline, day 0) and IL-12 (500 ng i.p. in 0.2 ml saline, days 4, 6, and 8). Recombinant murine IL-12 (supplied by Genetics Institute, Andover, MA) and cyclophosphamide (Mead-Johnson, Princeton, NJ) were reconstituted in sterile water and diluted in saline before use. Tumor size was assessed twice weekly following treatment. Complete rejection was defined as lack of visible tumors following treatment for at least 4 mo. Animals with tumors >30 mm in size following therapy were euthanized. At least seven mice per treatment group were examined throughout the study.

Depletion of NK1.1⁺ cells

Anti-NK1.1 (PK136; American Type Culture Collection, Manassas, VA) and isotype-matched control Ab (anti-rat gp42, 3G7) (29) were partially purified by ammonium sulfate precipitation from ascites obtained from hybridoma-bearing mice. Mice were treated with mAb (200 µg i.p. in 0.2 ml saline) twice weekly for 2 wk, then weekly until death or sacrifice of the animals (30). Depletion of NK1.1⁺ cells was confirmed by examination of NK1.1 and DX5 expression, and examination of CD1d tetramer binding cells by FACS of spleen and liver lymphocytes. Cells were also stained with species-specific donkey anti-mouse Ig to confirm that NK1.1 Ag was not masked by the depleting Ab. Levels of NK cells and NK1.1⁺ NK T cells in spleen and liver lymphocytes were negligible after the first 2 wk of mAb therapy and remained negligible at least 1 wk following Ab injection (data not shown). MCA207 tumors were implanted after 2 wk of mAb therapy.

Isolation of spleen and liver lymphocytes

Spleen and liver lymphocytes from individual mice were isolated and analyzed by standard methods. Briefly, single-cell suspensions were prepared by passage through stainless steel mesh, and RBCs were removed by hypotonic lysis. Spleen lymphocytes were used in this form, while liver lymphocytes were further separated by density centrifugation over Percoll (31).

Antibodies and flow cytometry

All directly conjugated Abs were purchased from PharMingen (San Diego, CA). CD1d tetramers were provided by M. Kronenberg (La Jolla Institute for Allergy and Immunology, La Jolla, CA). Isolated spleen and liver lymphocytes were preincubated with 2.4G2 supernatant to block nonspecific binding to Fc receptors. Cells were stained using mAb concentrations of 0.5–1 µg/10⁶ cells and were analyzed on either a FACScan or a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA).

Statistical analysis

Differences between treatment groups in tumor size at different time points and comparison of lymphocyte subpopulations in different treatment groups were analyzed using one-way ANOVA with Tukey pairwise comparisons using the InStat software package (Stillwater, MN).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
NK1.1⁺ cells control early tumor growth and rapid tumor regression, but are not required for the rejection of MCA207 following CY and IL-12 treatment

In the MCA207 large sarcoma model, complete immune-mediated tumor regression occurs only when mice are treated with both Cy and IL-12 (8). To determine the effector cells involved in this rejection, we first examined whether NK1.1⁺ cells might be important by examining tumor rejection in C57BL/6 mice depleted of NK1.1⁺ cells by weekly i.p. injection of anti-NK1.1 Ab. Previous studies have demonstrated the efficacy of this treatment regimen, and lymphocyte staining of depleted mice confirmed that virtually no NK1.1⁺ cells were present (30) (data not shown). Following 2 wk of anti-NK1.1 treatment, MCA207 tumors were implanted in animals depleted of NK1.1⁺ cells and wild-type mice by s.c. injection of 1 x 10⁶ MCA207 cells in the left flank. Tumor growth was monitored twice a week.

As previously shown (5, 8), MCA207 tumors grew rapidly in untreated C57BL/6 mice, reaching a diameter of 20 mm after 20 days. As shown in Fig. 1A, tumors in mice treated with isotype-matched control Ab (3G7) showed the same growth kinetics as untreated mice. Tumors in NK1.1-depleted mice, however, grew significantly faster and reached a diameter of 20 mm after only 14 days (p < 0.001; Fig. 1A).

View larger version (14K): [in this window] [in a new window]   FIGURE 1. Growth and regression of MCA207 sarcomas in C57BL/6 mice. MCA207 cells (1 x 106) were injected s.c. into B6 mice (), NK1.1-depleted B6 mice (•), or isotype control (3G7) mice (). Tumor diameter was measured before treatment (A) and following treatment with Cy and IL-12 (B). Points are mean diameters from at least seven mice, and error bars represent the SEM.

Thus, depletion of NK1.1⁺ cells led to rapid MCA207 tumor growth and delayed tumor regression following treatment, indicating that NK1.1⁺ cells limit tumor expansion during the early phase of tumor growth and are important for rapid tumor regression following Cy and IL-12 therapy. Interestingly, the depletion of NK1.1⁺ cells did not prevent complete regression of MCA207 following treatment, indicating that while NK1.1⁺ cells influence tumor growth before and after Cy and IL-12 therapy, they are not required for complete tumor rejection.

Complete rejection of MCA207 following combined immunotherapy with Cy and IL-12 requires CD1d-dependent NK T cells

NK1.1 depletion removes both NK cells and NK T cells. To determine whether NK or NK T cells might be the important effector cells in the process of MCA207 regression following immunotherapy, we examined tumor regression in C57BL/6 mice with a targeted deletion of CD1d. These mice do not express CD1d and do not develop CD1d-dependent NK T cells, but they have normal numbers of NK cells and conventional T cells and retain certain CD1d-independent NK T cells (32).

In contrast to NK1.1-depleted mice, when MCA207 tumors were implanted into CD1d^-/- mice, initial tumor growth was identical with tumor growth in wild-type C57BL/6 mice (Figs. 1A and 2A). On day 21 tumor size reached 20 mm in diameter, and treatment with Cy⁺IL-12 was initiated. In C57BL/6 wt mice and mice deficient in CD1d-dependent NK T cell (CD1d^-/- mice) initial tumor regression following CY and IL-12 treatment was similar and followed the same time course (Figs. 1B and 2B). Surprisingly, despite apparently complete rejection of MCA207 tumors in CD1d^-/- mice, tumor regrowth occurred 1 wk later (Figs. 2B and 3). Thus, mice deficient in CD1d-dependent NK T cells were unable to completely reject MCA207 tumors following CY and IL-12 therapy.

View larger version (15K): [in this window] [in a new window]   FIGURE 2. Growth and regression of MCA207 sarcomas in CD1d-/- mice. MCA207 cells (1 x 106) were injected s.c. into CD1d-/- mice () or NK1.1-depleted CD1d-/- mice (). Tumor diameter was measured before treatment (A) and following treatment with, Cy, and IL-12 (B). Points are mean diameters from at least seven mice, and error bars represent the SEM.

Lack of both CD1d-dependent NK T cells and NK1.1⁺ cells abrogates tumor regression in response to Cy and IL-12

Since the tumor recurrence seen in CD1d-deficient mice was not seen in NK1.1-depleted mice, this suggested that an NK1.1^- CD1d-dependent lymphocyte population might play a role in the complete tumor rejection. We therefore determined the combined effect of CD1d-dependent NK1.1^- T cells and NK1.1⁺ cells upon tumor growth by depleting CD1d-deficient mice of NK1.1⁺ cells.

As seen with wild-type C57BL/6, MCA207 tumors in NK1.1-depleted CD1d^-/- mice grew significantly faster than those in nondepleted CD1d^-/- mice (p < 0.001; Fig. 2A). Following treatment, MCA207 tumors in NK1.1-depleted CD1d^-/- mice showed a brief growth stabilization phase, followed by rapid tumor growth (Fig. 2B). No tumor regression was observed, and animals of this group had to be sacrificed on day 20 following initiation of therapy because of excessive tumor burden. Thus, NK1.1⁺ cells and CD1d-dependent NK T cells play important, but separate, roles in the complete regression of MCA207 following immunotherapy, and the lack of both lymphocyte populations completely abrogates the response to Cy and IL-12. This suggests that the anti-tumor effect of Cy and IL-12 is dependent on the combination of NK and NK T cells.

Cy and IL-12 treatment of C57BL/6 mice induces a marked increase in a distinct NK T cell subpopulation of DX5⁺NK1.1^-CD3⁺ lymphocytes

Our findings demonstrated that both NK cells and CD1d-dependent NK T cells play important, but separate, roles in eradication of solid MCA207 tumors following Cy and IL-12 therapy. To further understand the effect of treatment with Cy and IL-12 on NK and NK T cells, we treated wild-type C57BL/6 mice with Cy and IL-12 and used three-color flow cytometry to examine the surface marker expression on freshly isolated spleen and liver lymphocytes at different time points.

As shown in Fig. 3, following Cy and IL-12 therapy, a rapid decrease in NK1.1⁺ cell populations was observed 1 day following the first IL-12 dose (day 5), which is similar to previously reported observations following IL-12 treatment alone (33). This decrease was even more noticeable the day following the final IL-12 dose (day 9), but by day 11 most of the NK1.1⁺ cell populations had recovered toward normal levels. Remarkably, treatment with Cy plus IL-12 caused a significant rise in a novel T cell population expressing an NK cell marker, DX5, but not NK1.1. By day 11 of treatment, this cell subset was markedly increased in both spleen and liver (Fig. 3, A and B). This population is distinct from the classical NK T cell population, not only in the lack of NK1.1 expression, but also in levels of CD3, which are higher than the intermediate levels seen on classical NK T (NK1.1⁺CD3⁺) cells, although not as high as the levels seen on conventional T cells (Fig. 4). To distinguish these cells from classical NK1.1⁺CD3⁺ NK T cells we have termed them D-NK T cells.

View larger version (19K): [in this window] [in a new window]   FIGURE 3. Alterations in lymphocyte populations induced by Cy and IL-12. B6 mice were treated with Cy (day 0) and IL-12 (days 4, 6, and 8). Mice were sacrificed on days 5, 9, and 11, and spleen (A) and liver (B and C) lymphocytes were analyzed by FACS for the indicated populations. See Fig. 5 for representative FACS plots and gates. Bars represent the means of three mice, and error bars represent the SEM. Asterisks denote statistically significant differences from control with a confidence interval of 99%.

View larger version (43K): [in this window] [in a new window]   FIGURE 4. Cy and IL-12 induce a novel DX5+NK1.1-CD3+ lymphocyte subset. CD3+ liver lymphocytes were analyzed for DX5 and NK1.1 expression without treatment (A) or after Cy and IL-12 treatment (B; day 11). CD3 expression by DX5+ cells (C) is higher than the intermediate CD3 expression by NK1.1 cells (D).

D-NK T cells express NK cell markers and show some CD1d reactivity

To further characterize the D-NK T population, we conducted four-color flow cytometry on liver lymphocytes to determine the expression of other NK cell and activation markers. As shown in Table I, in concordance with classical NK T cells, and in contrast to conventional T cells, D-NK T cells express other NK cell receptors: inhibitory Ly49 family receptors and the NK-activating coreceptor 2B4. Like classical NK T cells, D-NK T cells all express CD44 and some express CD25, which contrasts sharply with the low expression on conventional T cells. Additionally, the pattern of CD4 and CD8 expression is similar between classical NK T cells and D-NK T cells and differs from the pattern shown by conventional T cells. As previously described (34), classical liver NK T cells are generally CD69⁺ CD62 ligand^-. Conventional T cells show an inverse pattern, with low CD69 expression and high CD62 ligand expression, and the pattern of expression of these two markers by D-NK T cells lies between these two extremes. The most marked difference between the classical NK T population and the D-NK T lymphocytes lies in their reactivity to -galactosyl ceramide (-GalCer)-loaded CD1d tetramers. As expected, the majority of classical NK T cells stain positively with those CD1d tetramers. In contrast, only 10% of the D-NK T cells appear to react with -GalCer-loaded CD1d tetramers. Fifteen percent of the conventional liver T cell population also reacts with -GalCer-loaded CD1d tetramers, which is in broad agreement with published data showing that a significant proportion of CD1d-reactive T cells are NK1.1^- (35, 36).

Thus, D-NK T cells are clearly distinct from both conventional T cells and classical NK T cells in their marker expression and in their specificity for -GalCer-loaded CD1d. Furthermore, following Cy and IL-12 therapy, in addition to the increase in the D-NK T cell population in both liver and spleen, the D-NK T phenotype alters to suggest increased activation and increased reactivity toward -GalCer-loaded CD1d.

The Cy and IL-12-mediated increase in D-NK T lymphocytes is impaired in mice deficient in CD1d

Because the administration of Cy and IL-12 dramatically expanded D-NK T cells in wild-type C57BL/6 mice, we next examined the requirement for CD1d in the induction of D-NK T cells by Cy and IL-12. Basal levels of these cells in CD1d^-/- mice were similar to those in wild-type C57BL/6 mice (data not shown), suggesting that the development of this population is not strictly dependent on CD1d. This is in agreement with the data in Table I showing that only 10% of these cells are reactive with -GalCer-loaded CD1d tetramers. In NK1.1-depleted mice, the D-NK T population comprised a greater proportion of basal lymphocytes due to the lack of NK1.1⁺ cells in these mice.

In C57BL/6 mice and NK1.1-depleted C57BL/6 mice, treatment with Cy and IL-12 led to a 6-fold increase in liver D-NK T lymphocytes (Fig. 5). In CD1d^-/- mice, however, only a 2.5-fold increase in D-NK T lymphocytes was observed, while in NK1.1-depleted CD1d^-/- mice this population barely increased. Thus, the expansion of D-NK T lymphocytes following Cy and IL-12 correlates with observed rejection of MCA207 sarcomas. D-NK T cells are significantly expanded in wild-type C57BL/6 mice following Cy and IL-12 immunotherapy, and tumors are rejected. The D-NK T cells show impaired expansion following immunotherapy in CD1d^-/- mice, which fail to completely clear tumor cells, and show little or no expansion in NK1.1-depleted CD1d^-/- mice, in which no tumor regression occurs. These data coupled with the necessity of Cy for both tumor rejection and induction of this population lead us to suggest that the increase in D-NK T cells may be important in the clinical anti-tumor response following Cy and IL-12 therapy.

View larger version (28K): [in this window] [in a new window]   FIGURE 5. Induction of DX5+NK1.1-CD3+ lymphocytes is impaired in CD1d-/- mice. Wild-type C57BL/6 mice or CD1d-/- mice, with or without NK1.1 depletion, were treated with Cy and IL-12 or left untreated. Induction of DX5+NK1.1-CD3+ lymphocytes by Cy and IL-12 was determined. Bars represent data from six mice.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In addition to defining distinct roles for NK and NK T cells in tumor rejection, we describe a novel NK T cell population with the phenotype DX5⁺NK1.1^-CD3⁺ (D-NK T), whose appearance is stimulated by combination immunotherapy with Cy and IL-12 and whose induction correlates with a successful anti-tumor response. Our results suggest that the combined action of several effector cell types is required for efficient tumor eradication of a large established tumor, and that these effector cells each have distinct roles.

The MCA207 model used in our studies can be cured with IL-12 alone when tumor size is below approximately 8 mm (5). Studies using IL-12-transfected MCA207 demonstrated a role for NK cells as well as CD4⁺ and CD8⁺ cells in suppression of tumor growth stimulated by local expression of IL-12 by the tumor itself (38). At the larger 20-mm tumor size examined here, however, combination immunotherapy with Cy is essential for any effect of IL-12 treatment. Therefore, the effector cells involved in tumor rejection as well as the mechanisms of tumor rejection might be somewhat distinct from those determined for therapy with IL-12 alone. Consistent with this, we have recently demonstrated that Cy- and IL-12-induced tumor rejection is intact in perforin-deficient mice (data not shown), which contrasts with previous studies that have demonstrated a perforin dependence for IL-12-mediated tumor rejection (13, 39). In the present study mice lacking CD1d-dependent NK T cells and depleted of NK cells were completely resistant to therapy with Cy and IL-12. Thus, in this large tumor model, mainstream T cells are not sufficient to reject tumors in the absence of NK and NK T cells, and combination therapy with Cy and IL-12 appears to induce a qualitatively distinct immune response compared with that induced by IL-12 alone.

The use of NK1.1 to deplete NK and NK T cells is not without controversy. Although most groups have shown that this treatment depletes both NK and NK T cells, some have suggested that NK1.1 depletes NK cells, but not NK T cells (40). For NK cells, which mostly stain with both DX5 and NK1.1, the depletion was confirmed by the lack of DX5⁺ cells. For NK T cells, the depletion was confirmed by lack of CD1d tetramer-positive cells, ruling out the possibility that NK1.1 treatment simply led to a down-regulation of NK1.1 on the surface of DX5^- NK T cells (data not shown). Although our studies do not examine the lack of NK cells alone, mAb depletion of NK1.1⁺ cells, but not deficiency of CD1d-dependent NK T cells, led to rapid tumor growth and reduced rate of tumor rejection following treatment. This suggests that the effect of NK1.1 depletion is primarily due to the loss of NK cells. NK1.1 may also be expressed on a proportion of T cells in vivo following viral infection or in vitro following cytokine stimulation (41, 42). Therefore, an alternate possibility is that depletion of NK1.1-positive cells may include some Ag-specific T cells generated in response to the tumor. However, previous studies have demonstrated that the development of cytotoxic T cells is unimpaired by the depletion of NK1.1⁺ cells (30). Additionally, since the difference in tumor growth in NK1.1-depleted animals is apparent in the first few days after tumor inoculation, the generation of an Ag-specific response at this early stage is unlikely. We cannot, however, completely rule out the possibility that NK1.1 depletion reduces Ag-specific T cells during the later tumor regression phase, and this may contribute to the reduced rate of tumor regression in NK1.1-depleted animals. It should also be noted that Cy- and IL-12-mediated tumor rejection does not lead to an effective long term Ag-specific memory response in this model, since rechallenge with the same tumor 6 mo following initial tumor resolution leads to tumor growth (data not shown).

Previous descriptions of NK T cells have described them as NK1.1⁺ CD3^int T cells, most of which demonstrate restricted TCR expression and also express other NK cell markers including DX5 (31). In our studies we found that very low numbers of T cells expressed DX5, and less than half of these cells coexpressed NK1.1. In contrast, after treatment with CY and IL-12, a DX5⁺NK1.1^- population became selectively increased in the liver. We have termed these cells D-NK T cells. Combination therapy with Cy and IL-12 is required for tumor rejection, and IL-12 alone has no effect on tumor growth in this tumor model. The observation that the increase in D-NK T cells is also dependent on the combination therapy, while IL-12 alone has no effect, suggests that D-NK T cells may play a role in the tumor rejection induced by Cy and IL-12.

Classification of D-NK T cells as an NK T cell population is supported by their CD3^intCD44⁺ phenotype and by their coexpression of a number of NK cell markers. This population is, however, distinct from the classic NK T cell population, in that it lacks NK1.1 expression, the CD3 expression is somewhat higher, and it shows reduced reactivity with -GalCer-loaded CD1d tetramers (Table I and Fig. 4). Low levels of D-NK T cells seen in wild-type mice were also present in CD1d^-/- mice (data not shown), indicating that the majority of these cells can develop in the absence of CD1d. The increase in this population seen following Cy and IL-12 treatment, however, appears to be dependent on either CD1d expression or CD1d-dependent NK T cells, since the same increase is not seen in CD1d^-/- mice. Notably, there is an increase in the proportion of -GalCer-loaded CD1d tetramer-reactive D-NK T cells from 10 to 30% of the D-NK T cells in C57BL/6 mice following treatment. Although this difference does not account fully for the increase in population size, it is possible that the remainder of the difference is due to D-NK T cells that recognize other ligands in the context of CD1d. Alternatively, classical NK T cells may be required to facilitate the increase in D-NK T cells following Cy and IL-12 treatment. NK1.1 expression on NK1.1⁺ cells has been reported to be down-regulated following in vitro T cell activation through the TCR (43), although DX5 expression was not simultaneously examined. It remains possible that some change in the D-NK T cell population is due to down-regulation of NK1.1 on classical NK T cells and up-regulation of DX5. However, the differences observed in the expression of other receptors within these populations make this less likely. DX5 has also recently been shown to appear on a portion of virus-induced Ag-specific T cells (41). The D-NK T cells described here were observed following treatment with Cy and IL-12 in the absence of tumor or other Ags. Additionally, they express many of the markers of classical NK T and NK cells, and following treatment a large portion of them react with the nonclassical MHC molecule, CD1d, confirming that they are not merely activated conventional T cells.

Although an increase in the D-NK T cell population was observed in the spleen, it was not as prominent as that observed in the liver. Previous studies have shown that liver NK T cells may have anti-tumor activity, while splenic NK T cells lack the same activity (4). Furthermore, liver and spleen NK T cell populations appear to be distinct in other aspects as well, including expression of CD4/CD8, activation marker expression, TCR repertoire, and CD1d reactivity (31, 34). It has recently been demonstrated that liver NK T cells are recruited from the thymus by NK cells, and this recruitment is probably mediated by LFA-1 expression on the NK cells (44). It is possible that NK cells at the tumor site may be important in recruiting NK T cells locally as well. NK T cells have also been demonstrated to stimulate NK cells (28); thus, reciprocal interactions between these cell types in both activation and localization may be of importance.

Following surgical resection of tumors, one of the major problems is the presence of minimal residual lesions that lead to tumor recurrence. In our studies with CD1d^-/- mice, the tumor appeared to have completely regressed. Clearly, however, minimal residual lesions remained and allowed tumor regrowth. The uniformity of tumor recurrence in the CD1d^-/- mice suggests that a specialized lymphocyte population is required for the removal of the final remnants of the tumor, and that this population is absent or insufficiently activated in the CD1d^-/- mice. The most prominent difference in lymphocyte subsets that we could define between wild-type and CD1d^-/- mice following Cy and IL-12 treatment was an impaired increase in the D-NK T cell population in CD1d^-/- mice. Possibly these cells are necessary for the complete eradication of remnant tumor cells, which might otherwise be able to avoid immune detection. It is of interest to note that in NK1.1-depleted mice, no tumor regression is seen until after day 11 of treatment, the time point at which the D-NK T cell population increases. This suggests that this population may also be particularly important for tumor rejection in the absence of NK cells. Alternatively, this D-NK T population may be a regulatory cell, necessary for the activation of another effector cell type or of APC. Recent studies have implicated NK T cells as both positive and negative regulators of anti-tumor immunity (12, 45), and their exact role is likely to be dependent on the specific NK T cell subpopulation involved and the specific tumor model. Notably, our phenotypic analysis of the D-NK T cell population demonstrates some heterogeneity of expression of activation markers, regulatory NK cell receptors, and -GalCer-loaded tetramer binding, which may indicate that this population itself is composed of further subsets.

The identification of effector cell populations that can effect or activate rejection of residual tumor lesions would be of obvious importance. The unique increase in the D-NK T cell population following combined therapy with Cy and IL-12 suggests that the immunopotentiating effect of Cy is not solely due to amplification of the response stimulated by IL-12 alone. Further studies will be important to elucidate the contribution of D-NK T cells as effector or regulatory cells in the anti-tumor response stimulated by Cy and IL-12 therapy.

	Acknowledgments

 
IL-12 was kindly supplied by Genetics Institute (Andover, MA). We thank W. E. Seaman and J. C. Ryan for review of the manuscript, A. Bendelac for CD1d^-/- breeding pairs of mice, and J. A. Norton for MCA207 cells. Thanks also to Stephane Sidobre, Olga Naidenko, and Mitchell Kronenberg for supplying us with the CD1d tetramers.

	Footnotes

 
¹ This work was supported by the Department of Veterans Affairs. C.K. is supported by the Mildred Scheel Cancer Foundation. M.C.N. is the recipient of a Veterans Affairs Career Development Award and is supported by National Institutes of Health Grant K11AR01927, the Arthritis Foundation, and the American Cancer Society.

² Current address: Universitätsklinikum Essen, Hufelandstrasse 55, 45122 Essen, Germany.

³ C.K. and M.R.D. contributed equally to this work.

⁴ Address correspondence and reprint requests to Dr. Mary C. Nakamura, Immunology Section 111-R, Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, CA 94121. E-mail address: marynak{at}itsa.ucsf.edu

⁵ Abbreviations used in this paper: Cy, cyclophosphamide; D-NKT cells, DX5⁺NK1.1^-CD3⁺ lymphocytes; MCA, 3-methylcholanthrene; -GalCer, -galactosyl ceramide.

Received for publication December 11, 2000. Accepted for publication June 29, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Brunda, M. J., L. Luistro, R. R. Warrier, R. B. Wright, B. R. Hubbard, M. Murphy, S. F. Wolf, M. K. Gately. 1993. Antitumor and antimetastatic activity of interleukin 12 against murine tumors. J. Exp. Med. 178:1223.[Abstract/Free Full Text]
2. Nastala, C. L., H. D. Edington, T. G. McKinney, H. Tahara, M. A. Nalesnik, M. J. Brunda, M. K. Gately, S. F. Wolf, R. D. Schreiber, W. J. Storkus. 1994. Recombinant IL-12 administration induces tumor regression in association with IFN- production. J. Immunol. 153:1697.[Abstract]
3. Zou, J. P., N. Yamamoto, T. Fujii, H. Takenaka, M. Kobayashi, S. H. Herrmann, S. F. Wolf, H. Fujiwara, T. Hamaoka. 1995. Systemic administration of rIL-12 induces complete tumor regression and protective immunity: response is correlated with a striking reversal of suppressed IFN- production by anti-tumor T cells. Int. Immunol. 7:1135.[Abstract/Free Full Text]
4. Takeda, K., S. Seki, K. Ogasawara, R. Anzai, W. Hashimoto, K. Sugiura, M. Takahashi, M. Satoh, K. Kumagai. 1996. Liver NK1.1⁺ CD4⁺ T cells activated by IL-12 as a major effector in inhibition of experimental tumor metastasis. J. Immunol. 156:3366.[Abstract]
5. Tsung, K., J. B. Meko, G. R. Peplinski, Y. L. Tsung, J. A. Norton. 1997. IL-12 induces T helper 1-directed antitumor response. J. Immunol. 158:3359.[Abstract]
6. Popovic, D., K. M. El-Shami, E. Vadai, M. Feldman, E. Tzehoval, L. Eisenbach. 1998. Antimetastatic vaccination against Lewis lung carcinoma with autologous tumor cells modified to express murine interleukin 12. Clin. Exp. Metastasis 16:623.[Medline]
7. Schultz, J., J. Pavlovic, B. Strack, M. Nawrath, K. Moelling. 1999. Long-lasting anti-metastatic efficiency of interleukin 12-encoding plasmid DNA. Hum. Gene Ther. 10:407.[Medline]
8. Tsung, K., J. B. Meko, Y. L. Tsung, G. R. Peplinski, J. A. Norton. 1998. Immune response against large tumors eradicated by treatment with cyclophosphamide and IL-12. J. Immunol. 160:1369.[Abstract/Free Full Text]
9. Trinchieri, G., P. Scott. 1995. Interleukin-12: a proinflammatory cytokine with immunoregulatory functions. Res. Immunol. 146:423.[Medline]
10. Duda, D. G., M. Sunamura, L. Lozonschi, T. Kodama, S. Egawa, G. Matsumoto, H. Shimamura, K. Shibuya, K. Takeda, S. Matsuno. 2000. Direct in vitro evidence and in vivo analysis of the antiangiogenesis effects of interleukin 12. Cancer Res. 60:1111.[Abstract/Free Full Text]
11. Coughlin, C. M., K. E. Salhany, M. S. Gee, D. C. LaTemple, S. Kotenko, X. Ma, G. Gri, M. Wysocka, J. E. Kim, L. Liu, et al 1998. Tumor cell responses to IFN affect tumorigenicity and response to IL-12 therapy and antiangiogenesis. Immunity 9:25.[Medline]
12. Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M. Kanno, M. Taniguchi. 1997. Requirement for V14 NK T cells in IL-12-mediated rejection of tumors. Science 278:1623.[Abstract/Free Full Text]
13. Kodama, T., K. Takeda, O. Shimozato, Y. Hayakawa, M. Atsuta, K. Kobayashi, M. Ito, H. Yagita, K. Okumura. 1999. Perforin-dependent NK cell cytotoxicity is sufficient for anti-metastatic effect of IL-12. Eur. J. Immunol. 29:1390.[Medline]
14. Hashimoto, W., K. Takeda, R. Anzai, K. Ogasawara, H. Sakihara, K. Sugiura, S. Seki, K. Kumagai. 1995. Cytotoxic NK1.1 Ag⁺ T cells with intermediate TCR induced in the liver of mice by IL-12. J. Immunol. 154:4333.[Abstract]
15. Takeda, K., Y. Hayakawa, M. Atsuta, S. Hong, L. Van Kaer, K. Kobayashi, M. Ito, H. Yagita, K. Okumura. 2000. Relative contribution of NK and NKT cells to the anti-metastatic activities of IL-12. Int. Immunol. 12:909.[Abstract/Free Full Text]
16. Smyth, M. J., M. Taniguchi, S. E. Street. 2000. The anti-tumor activity of IL-12: mechanisms of innate immunity that are model and dose dependent. J. Immunol. 165:2665.[Abstract/Free Full Text]
17. Kawamura, T., K. Takeda, S. K. Mendiratta, H. Kawamura, K. Van, H. Yagita, T. Abo, K. Okumura. 1998. Critical role of NK1⁺ T cells in IL-12-induced immune responses in vivo. J. Immunol. 160:16.[Abstract/Free Full Text]
18. Emtage, P. C., Y. Wan, M. Hitt, F. L. Graham, W. J. Muller, A. Zlotnik, J. Gauldie. 1999. Adenoviral vectors expressing lymphotactin and interleukin 2 or lymphotactin and interleukin 12 synergize to facilitate tumor regression in murine breast cancer models. Hum. Gene Ther. 10:697.[Medline]
19. Lasek, W., J. Golab, W. Maslinski, T. Switaj, E. Z. Balkowiec, T. Stoklosa, A. Giermasz, M. Malejczyk, M. Jakobisiak. 1999. Subtherapeutic doses of interleukin-15 augment the antitumor effect of interleukin-12 in a B16F10 melanoma model in mice. Eur. Cytokine Netw. 10:345.[Medline]
20. Nagai, H., I. Hara, T. Horikawa, M. Fujii, M. Kurimoto, S. Kamidono, M. Ichihashi. 2000. Antitumor effects on mouse melanoma elicited by local secretion of interleukin-12 and their enhancement by treatment with interleukin-18. Cancer Invest. 18:206.[Medline]
21. Taniguchi, F., H. Yamagishi, H. Fujiwara, Y. Ueda, N. Fuji, T. Yoshimura, T. Oka. 1998. Systemic administration of rIL-12 synergistically enhances the therapeutic effect of a TNF gene-transduced cancer vaccine. Gene Ther. 5:1677.[Medline]
22. Narvaiza, I., G. Mazzolini, M. Barajas, M. Duarte, M. Zaratiegui, C. Qian, I. Melero, J. Prieto. 2000. Intratumoral coinjection of two adenoviruses, one encoding the chemokine IFN--inducible protein-10 and another encoding IL-12, results in marked antitumoral synergy. J. Immunol. 164:3112.[Abstract/Free Full Text]
23. Allison, A. C.. 2000. Immunosuppressive drugs: the first 50 years and a glance forward. Immunopharmacology 47:63.[Medline]
24. Kaufmann, S. H., H. Hahn, T. Diamantstein. 1980. Relative susceptibilities of T cell subsets involved in delayed-type hypersensitivity to sheep red blood cells to the in vitro action of 4-hydroperoxycyclophosphamide. J. Immunol. 125:1104.[Abstract]
25. Matar, P., V. R. Rozados, A. D. Gonzalez, D. G. Dlugovitzky, R. D. Bonfil, O. G. Scharovsky. 2000. Mechanism of antimetastatic immunopotentiation by low-dose cyclophosphamide. Eur. J. Cancer 36:1060.
26. Awwad, M., R. J. North. 1988. Cyclophosphamide (Cy)-facilitated adoptive immunotherapy of a Cy-resistant tumour: evidence that Cy permits the expression of adoptive T-cell mediated immunity by removing suppressor T cells rather than by reducing tumour burden. Immunology 65:87.[Medline]
27. Kuramitsu, Y., M. Nishibe, M. Kobayashi, Y. Togashi, L. Yuan, M. Takizawa, F. Okada, M. Hosokawa. 1996. Transforming growth factor 1 (TGF-1) produced in tumour tissue after chemotherapy acts as a lymphokine-activated killer attractant. Br. J. Cancer 74:274.[Medline]
28. Carnaud, C., D. Lee, O. Donnars, S. H. Park, A. Beavis, Y. Koezuka, A. Bendelac. 1999. Cutting edge: cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J. Immunol. 163:4647.[Abstract/Free Full Text]
29. Seaman, W. E., E. C. Niemi, M. R. Stark, R. D. Goldfien, A. S. Pollock, J. B. Imboden. 1991. Molecular cloning of gp42, a cell-surface molecule that is selectively induced on rat natural killer cells by interleukin 2: glycolipid membrane anchoring and capacity for transmembrane signaling. J. Exp. Med. 173:251.[Abstract/Free Full Text]
30. Seaman, W. E., M. Sleisenger, E. Eriksson, G. C. Koo. 1987. Depletion of natural killer cells in mice by monoclonal antibody to NK-1.1: reduction in host defense against malignancy without loss of cellular or humoral immunity. J. Immunol. 138:4539.[Abstract]
31. Hammond, K. J., S. B. Pelikan, N. Y. Crowe, E. Randle-Barrett, T. Nakayama, M. Taniguchi, M. J. Smyth, I. R. van Driel, R. Scollay, A. G. Baxter, et al 1999. NKT cells are phenotypically and functionally diverse. Eur. J. Immunol. 29:3768.[Medline]
32. Chen, Y. H., N. M. Chiu, M. Mandal, N. Wang, C. R. Wang. 1997. Impaired NK1⁺ T cell development and early IL-4 production in CD1-deficient mice. Immunity 6:459.[Medline]
33. Eberl, G., H. R. MacDonald. 1998. Rapid death and regeneration of NKT cells in anti-CD3- or IL-12-treated mice: a major role for bone marrow in NKT cell homeostasis. Immunity 9:345.[Medline]
34. Eberl, G., R. Lees, S. T. Smiley, M. Taniguchi, M. J. Grusby, H. R. MacDonald. 1999. Tissue-specific segregation of CD1d-dependent and CD1d-independent NK T cells. J. Immunol. 162:6410.[Abstract/Free Full Text]
35. Benlagha, K., A. Weiss, A. Beavis, L. Teyton, A. Bendelac. 2000. In vivo identification of glycolipid antigen-specific T cells using fluorescent CD1d tetramers. J. Exp. Med. 191:1895.[Abstract/Free Full Text]
36. Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi, C. R. Wang, Y. Koezuka, M. Kronenberg. 2000. Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192:741.[Abstract/Free Full Text]
37. Smyth, M. J., K. Y. Thia, S. E. Street, E. Cretney, J. A. Trapani, M. Taniguchi, T. Kawano, S. B. Pelikan, N. Y. Crowe, D. I. Godfrey. 2000. Differential tumor surveillance by natural killer (NK) and NKT cells. J. Exp. Med. 191:661.[Abstract/Free Full Text]
38. Tahara, H., L. Zitvogel, W. J. Storkus, H. J. Zeh, T. G. McKinney, R. D. Schreiber, U. Gubler, P. D. Robbins, M. T. Lotze. 1995. Effective eradication of established murine tumors with IL-12 gene therapy using a polycistronic retroviral vector. J. Immunol. 154:6466.[Abstract]
39. Hashimoto, W., T. Osaki, H. Okamura, P. D. Robbins, M. Kurimoto, S. Nagata, M. T. Lotze, H. Tahara. 1999. Differential antitumor effects of administration of recombinant IL-18 or recombinant IL-12 are mediated primarily by Fas-Fas ligand- and perforin-induced tumor apoptosis, respectively. J. Immunol. 163:583.[Abstract/Free Full Text]
40. Asea, A., J. Stein-Streilein. 1998. Signalling through NK1.1 triggers NK cells to die but induces NK T cells to produce interleukin-4. Immunology 93:296.[Medline]
41. Slifka, M. K., R. R. Pagarigan, J. L. Whitton. 2000. NK markers are expressed on a high percentage of virus-specific CD8⁺ and CD4⁺ T cells. J. Immunol. 164:2009.[Abstract/Free Full Text]
42. Assarsson, E., T. Kambayashi, J. K. Sandberg, S. Hong, M. Taniguchi, L. Van Kaer, H. G. Ljunggren, B. J. Chambers. 2000. CD8⁺ T cells rapidly acquire NK1.1 and NK cell-associated molecules upon stimulation in vitro and in vivo. J. Immunol. 165:3673.[Abstract/Free Full Text]
43. Chen, H., H. Huang, W. E. Paul. 1997. NK1.1⁺ CD4⁺ T cells lose NK1.1 expression upon in vitro activation. J. Immunol. 158:5112.[Abstract]
44. Miyamoto, M., M. Emoto, V. Brinkmann, N. van Rooijen, R. Schmits, E. Kita, S. H. Kaufmann. 2000. Cutting edge: contribution of NK cells to the homing of thymic CD4⁺NKT cells to the liver. J. Immunol. 165:1729.[Abstract/Free Full Text]
45. Moodycliffe, A. M., D. Nghiem, G. Clydesdale, S. E. Ullrich. 2000. Immune suppression and skin cancer development: regulation by NKT cells. Nat. Immunol. 1:521.[Medline]

This article has been cited by other articles:

				D. G. Pellicci, K. J. L. Hammond, J. Coquet, K. Kyparissoudis, A. G. Brooks, K. Kedzierska, R. Keating, S. Turner, S. Berzins, M. J. Smyth, et al. DX5/CD49b-Positive T Cells Are Not Synonymous with CD1d-Dependent NKT Cells J. Immunol., October 1, 2005; 175(7): 4416 - 4425. [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]

				K. Adachi, H. Tsutsui, E. Seki, H. Nakano, K. Takeda, K. Okumura, L. Van Kaer, and K. Nakanishi Contribution of CD1d-unrestricted hepatic DX5+ NKT cells to liver injury in Plasmodium berghei-parasitized erythrocyte-injected mice Int. Immunol., June 1, 2004; 16(6): 787 - 798. [Abstract] [Full Text] [PDF]

				M. L. Salem, A. N. Kadima, Y. Zhou, C. L. Nguyen, M. P. Rubinstein, M. Demcheva, J. N. Vournakis, D. J. Cole, and W. E. Gillanders Paracrine Release of IL-12 Stimulates IFN-{gamma} Production and Dramatically Enhances the Antigen-Specific T Cell Response after Vaccination with a Novel Peptide-Based Cancer Vaccine J. Immunol., May 1, 2004; 172(9): 5159 - 5167. [Abstract] [Full Text] [PDF]

				Y. Yang, A. Ueno, M. Bao, Z. Wang, J. S. Im, S. Porcelli, and J.-W. Yoon Control of NKT Cell Differentiation by Tissue-Specific Microenvironments J. Immunol., December 1, 2003; 171(11): 5913 - 5920. [Abstract] [Full Text] [PDF]

				S.-H. Park, T. Kyin, A. Bendelac, and C. Carnaud The Contribution of NKT Cells, NK Cells, and Other {gamma}-Chain-Dependent Non-T Non-B Cells to IL-12-Mediated Rejection of Tumors J. Immunol., February 1, 2003; 170(3): 1197 - 1201. [Abstract] [Full Text] [PDF]

				J. G. Segal, N. C. Lee, Y. L. Tsung, J. A. Norton, and K. Tsung The Role of IFN-{gamma} in Rejection of Established Tumors by IL-12 : Source of Production and Target Cancer Res., August 15, 2002; 62(16): 4696 - 4703. [Abstract] [Full Text] [PDF]

				N. Y. Crowe, M. J. Smyth, and D. I. Godfrey A Critical Role for Natural Killer T Cells in Immunosurveillance of Methylcholanthrene-induced Sarcomas J. Exp. Med., July 1, 2002; 196(1): 119 - 127. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services 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 Karnbach, C. Articles by Nakamura, M. C. Search for Related Content PubMed