Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2007 Jan 16;104(4):1313–1318. doi: 10.1073/pnas.0607731104

Induction of inducible NO synthase in bystander human T cells increases allogeneic responses in the vasculature

Jonathan C Choy *,†, Yinong Wang *,‡, George Tellides *,‡, Jordan S Pober *,†,§,¶,‖
PMCID: PMC1783095  PMID: 17227851

Abstract

Inducible NO synthase (iNOS) in human T cells is implicated in the pathogenesis of graft arteriosclerosis. Here we analyze the regulation and role of iNOS in human peripheral blood T cells. Allogeneic endothelial cells (EC) or dermal fibroblasts induce iNOS mRNA and protein expression, as well as enzymatic activity in primary human CD8 T cells. Although human EC activate T cells through the presentation of alloantigen, iNOS induction is confined to nonactivated T cells and does not depend on MHC molecules or costimulators. iNOS induction does involve new transcription and depends on NF-κB. JAK signaling, initiated during T cell activation, inhibits iNOS expression. Even though iNOS is confined to bystander T cells, inhibition of iNOS activity reduces T cell proliferation in response to allogeneic EC, and addition of low levels of a NO donor rescues T cell responses. Similarly, iNOS is preferentially expressed by nonproliferating T cells within allografted arteries in vivo, and inhibition of iNOS activity reduces the number of activated T cells in these artery segments. These data identify a previously undescribed mechanism for enhanced activation of alloreactive T cells, namely stromal cell-mediated induction of iNOS in bystander T cells.

Keywords: endothelial cell graft, arteriosclerosis, NF-κβ, Janus Kinase

Inducible NO synthase (iNOS) produces the bioactive gas NO from the conversion of l-arginine to l-citrulline (1). iNOS plays a role in host defense. For example, mice with a disrupted iNOS gene show impaired clearance of certain bacterial pathogens. In addition, iNOS promoter polymorphisms and haplotype differences are associated with susceptibility to parasitic infections in humans (2, 3). Studies in iNOS-deficient mice have also implicated a role for this enzyme in pathological processes, including vascular diseases, transplant rejection, and degenerative neurological disorders (4–6). Furthermore, mouse iNOS functions as an immune regulator because disruption of functional iNOS leads to an enhanced T helper 1 response (2).

iNOS is expressed in a variety of mouse and rat cell types in response to many stimuli. However, the principal cell type expressing this enzyme in mice and rats is the cytokine-activated mononuclear phagocyte, and the most relevant physiological signals that induce iNOS in these cells are likely LPS and/or the combination of TNF, IL-1β, and IFNγ (1, 7). In contrast, the expression of iNOS in humans is highly restricted to relatively few cell types. Expression in human macrophages has been difficult to induce in cell culture, even in response to stimuli that robustly increase iNOS expression in mouse macrophages (8), although certain types of human macrophages can be induced to express iNOS in response to infection with Mycobacterium tuberculosis (9). These differences in human and mouse cell expression of iNOS illustrate the species-specific regulation of this enzyme, thereby underscoring the importance of investigating iNOS expression in human systems.

T cell expression of iNOS has been implicated in the pathogenesis of vascular diseases, which are the leading cause of mortality in developed countries. Specifically, iNOS has been detected in T cells (as well as macrophages) within human atherosclerotic lesions (10). Also, in an experimental chimeric human–SCID mouse model of graft arteriosclerosis (GA), which is the major cause of chronic solid organ allograft rejection, iNOS expression by human T cells infiltrating allogeneic human artery segments is causally linked to intimal thickening and vascular dysfunction (11). Although these data suggest that iNOS expression in T cells may be proinflammatory, the regulation and function of iNOS in human T cells are not known. The presence of iNOS within T cells in the arterial wall led us to examine whether endothelial cells (EC), which play an important role in regulating lymphocyte migration into and activation within vascular tissues, could contribute to iNOS induction. Here we report that EC (as well as other stromal cells) induce the expression of iNOS in primary human CD8 T cells. iNOS induction is limited to bystander T cells and does not depend on T cell receptor (TCR) stimulation. The induction of this enzyme depends on NF-κB and is inhibited by JAK signaling. Interestingly, iNOS in this allogeneically nonactivated bystander T cell population increases the number of proliferated T cells in response to allogeneic EC in vitro and in vivo. These data describe the TCR-independent induction of a previously un-described human T cell effector function mediated through stromal cell actions on bystander T cells.

Results

EC Induce iNOS Expression in Primary Human T Cells.

Because human EC are able to activate a variety of T cell responses (12), we hypothesized that EC provide signals that lead to iNOS induction in human T cells. Consistent with our hypothesis, coculture of CD8 T cells with allogeneic EC resulted in an increase in iNOS mRNA expression as compared with freshly isolated T cells (Fig. 1A). The expression of iNOS mRNA was also increased as compared with T cells cultured alone for 3 days (data not shown), but cell death occurs in many T cells cultured alone for this length of time and could have confounded analysis of RNA expression. To be certain that the iNOS mRNA was in the T cells and not in EC, we separated the EC and T cells present in the coculture and found that only the T cells contained iNOS transcript (Fig. 1A). Immunoblotting for VE-cadherin expression in the isolated T cells indicated that there was no contamination by EC [supporting information (SI) Fig. 7A]. CD4 T cells also increased iNOS mRNA levels when cocultured with allogeneic EC, but the response was less than that of CD8 T cells (data not shown), and we concentrated on CD8 T cell responses in the remainder of this study.

Fig. 1.

Fig. 1.

Open in a new tab

EC induce iNOS expression in human T cells. (A) RNA was isolated from primary human CD8 T cells immediately after isolation (0 days) or after collection from coculture with EC for 3 days. iNOS expression was measured with quantitative RT-PCR (Left). CD8 T cells were cocultured with EC for 3 days, T cells were purified with magnetic bead isolation, and RNA was isolated. RNA was also isolated from EC in the coculture, and iNOS mRNA was quantitated in the purified T cells or in the EC by quantitative RT-PCR (Right). (B) iNOS protein expression was quantitated by intracellular flow cytometry in CD8 T cells cultured alone or with EC for 1 and 3 days. Black line, IgG control; shaded area, iNOS staining. (C) The production of NO from iNOS in CD8 T cells was assessed by quantitating the amount of NO release that was inhibited with 1400W in EC, T cells alone, or T cells cocultured with EC for 3 days. Data are expressed as the percentage of untreated control (NO released by 1400W-treated cells/NO released by untreated cells) ± SD. All data are representative of at least three independent experiments.

In CD8 T cells cultured alone, there were a small number of cells that expressed iNOS protein, as detected by flow cytometry. Coculture of CD8 T cells with EC increased the percentage of T cells expressing iNOS. This EC-mediated induction of iNOS expression occurred as early as 1 day and was maintained after 3 days in coculture (Fig. 1B). In seven independent experiments using T cells and EC from different donors, there was a consistent and significant increase in the frequency of T cells expressing iNOS after coculture with EC as compared with T cells alone (40.1 ± 10.2% as compared with 19.7 ± 10.6%, respectively; P = 0.003). iNOS protein expression was also examined by Western blot, which confirmed an increase in the expression of iNOS in human T cells cocultured with EC as compared with T cells cultured alone (SI Fig. 7B). Fibroblasts were also able to induce expression of iNOS in T cells, although the frequency and level of iNOS induction appeared somewhat lower than that induced by EC (SI Fig. 7C).

The production of iNOS-derived NO from human T cells was then examined by quantitating 1400W-inhibitable nitrite production in the supernatant of T cell–EC cocultures. 1400W is a highly selective inhibitor of iNOS (13). Inhibition of iNOS did not affect the production of NO from EC and only minimally affected NO production from resting T cells. In contrast, in T cell–EC cocultures almost all NO production was inhibited with 1400W, indicating that iNOS expressed in T cells is enzymatically active (Fig. 1C).

iNOS Expression Is Confined to Bystander Naïve and Memory T Cells.

Human EC can act as semiprofessional antigen-presenting cells and mediate antigen-induced immune responses in blood vessels (12). The allogeneic response of CD8 T cells to human EC is mediated mainly by memory T cells and is characterized by T cell proliferation and the sustained expression of CD69 and HLA-DR (14). Dual-color flow cytometry indicated that all iNOS-expressing cells were CD8-positive. Unexpectedly, iNOS was similarly induced in naïve and memory T cells (SI Fig. 8). To examine the role of T cell activation in the induction of iNOS, we labeled T cells with carboxyfluorescein diacetate succinimidyl ester (CFSE) before coculture with allogeneic EC and isolated T cells that had or had not proliferated (CFSElow and CFSEhigh populations, respectively) after 10 days. iNOS expression was measured in these distinct populations by quantitative RT-PCR. Surprisingly, iNOS expression was confined exclusively to nonproliferating T cells and was not detected in T cells that had responded to allogeneic EC by proliferating (Fig. 2A). We also examined whether iNOS is expressed in allogeneically activated T cells by determining the expression of CD69 and HLA-DR on iNOS-expressing cells. After coculture of T cells with EC for 3 days, CD69- and HLA-DR-positive cells were predominantly iNOS-negative (Fig. 2B).

Fig. 2.

Fig. 2.

Open in a new tab

iNOS is expressed in allogeneically nonactivated human T cells. (A) Human T cells were labeled with CFSE before coculture with EC. After 10 days the proliferating (CFSElow) and nonproliferating (CFSEhigh) T cells were isolated by FACS sorting. iNOS expression in proliferating and nonproliferating T cells was measured by quantitative RT-PCR. Similar results were obtained in two independent experiments. (B) T cells were cocultured with allogeneic EC and the expression of CD69 and HLA-DR on cells not expressing or expressing iNOS examined by flow cytometry. (C) Neutralizing antibodies to MHC class I, LFA-3, ICAM, IFNγ, or TNF were added to T cell–EC cocultures. After 3 days, iNOS expression was assessed by flow cytometry. Unless indicated, all data are representative of at least three independent experiments.

Low-affinity or short-term engagement of the TCR by antigen or alloantigen may deliver a signal, yet not lead to proliferation or activation marker expression. To explore this possibility, we examined the effect of neutralizing antibodies to EC molecules that contribute to TCR-mediated activities on the induction of iNOS expression. Neither the addition of neutralizing antibodies to MHC class I nor the blockade of LFA-3 or ICAM-1 (two primary costimulator molecules used by EC) inhibited the up-regulation of iNOS expression in T cells (Fig. 2C), although these antibodies did block allogeneic activation of T cells (SI Fig. 9). Furthermore, although cytokines such as IFNγ and TNF are known to be key inducers of iNOS in other cell types, neutralization of these molecules also did not have any effect on iNOS expression in T cells (Fig. 2C). Inhibition of the T helper 2 cytokines IL-4 and IL-13, as well as TGFβ, similarly had no effect on iNOS expression (data not shown).

Because EC can modulate the function of leukocytes through soluble as well as cell-bound signals, iNOS expression was examined in T cells that were separated from EC with a transwell to prevent T cell–EC contact. The separation of T cell–EC contact did not affect the ability of EC to induce iNOS expression in T cells (SI Fig. 10A). Furthermore, addition of EC-conditioned medium to resting T cells was sufficient to induce iNOS expression (SI Fig. 10B). Inactivation of proteins in the conditioned medium by either heat treatment or proteinase K digestion before addition to CD8 T cells prevented the induction of iNOS expression in human T cells (SI Fig. 10B). Combined, the above results indicate that iNOS expression is confined to T cells that have not been activated by allogeneic signals and that EC basally release one or more soluble proteins that induce their expression in T cells.

iNOS Expression Is Induced by NF-κB-Mediated Gene Transcription.

To determine whether EC signals cause human T cells to increase iNOS expression through increased transcription, human T cells were transfected with the human iNOS promoter attached to a luciferase reporter gene, and the promoter activity was measured with a luciferase assay. No luciferase activity was detected in T cells transfected with a control vector containing a luciferase gene without a promoter. After 24 h in coculture there was an increase in iNOS promoter activity in human T cells exposed to EC as compared with resting T cells, indicating that the induction of iNOS expression is regulated, at least in part, at the transcriptional level in these cells (Fig. 3A).

Fig. 3.

Fig. 3.

Open in a new tab

iNOS expression is induced by NF-κB-mediated transcription in human T cells. (A) Human CD8 T cells were transfected with a control promoterless luciferase reporter construct (pXP2), a human iNOS promoter–luciferase (iNOS-luc) reporter construct, or the iNOS–luciferase reporter construct containing a mutation in the NF-κB binding site at −5.8 kb. Transfected T cells were cultured alone or with EC for 24 h, and promoter activity was determined with a luciferase assay. (B) T cells were cultured alone or with EC for 24 h, and dual-color immunofluorescence was performed for CD8 (green) to identify T cell borders and for p65 (red). Arrows depict T cells in which p65 shows predominantly nuclear localization, and arrowheads depict T cells in which p65 shows cytoplasmic localization. The percentage of T cells with nuclear localization of p65 was quantitated. (C) T cells were cultured alone or with EC in the absence or presence of the NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC). iNOS expression was quantitated by flow cytometry. All data are representative of at least three independent experiments.

Previous reports have shown that iNOS expression in some human cell lines depends on NF-κB-mediated transcription (15). There was an increase in nuclear translocation of p65 in T cells cocultured with EC as compared with resting T cells (Fig. 3B), and pharmacological inhibition of NF-κB activation using pyrrolidine dithiocarbamate inhibited iNOS expression in T cells (Fig. 3C). The role of NF-κB in the induction of iNOS gene transcription was further assessed by using a mutated human iNOS promoter–luciferase construct. Inactivation of a key NF-κB binding site within the human iNOS promoter (located at −5.8 kb) by mutation inhibited iNOS promoter activity (Fig. 3A).

iNOS Expression Is Inhibited in Activated T Cells by a JAK Pathway.

Because iNOS expression is limited to T cells that are not activated in response to alloantigen and TCR signals are known to activate NF-κB, we examined the possibility that iNOS induction is blocked during T cell activation. To do so, we increased the number of activated human T cells in our culture system by adding phytohemagglutinin (PHA) to CD8 T cells cocultured with allogeneic EC. PHA cross-links the TCR on a high proportion of T cells, and EC provide costimulation through LFA-3, leading to polyclonal proliferation and cytokine secretion (16, 17). Furthermore, in contrast to polyclonal activation with anti-CD3 antibodies, PHA will work with accessory cells, like EC, that lack Fc receptors. Stimulation of T cells with EC and PHA did indeed inhibit iNOS mRNA (SI Fig. 11A) and protein expression after 24 h of stimulation (Fig. 4A). Activation of T cells with a high dose of IL-2 (100 units/ml), which bypasses TCR signals, also inhibited iNOS expression (Fig. 4A). Although stimulation of T cells with EC and PHA for 3 days led to an increase in the per-cell level of iNOS expression as compared with T cells stimulated with EC alone (data not shown), iNOS expression was restricted to nonproliferated T cells (SI Fig. 11B). This finding is consistent with our data using an allogeneic system and further supports the notion that iNOS expression in human T cells is inhibited during T cell activation.

Fig. 4.

Fig. 4.

Open in a new tab

iNOS expression is inhibited in human T cells after TCR stimulation by a JAK pathway. (A) T cells were cultured alone or with EC and in the absence or presence of PHA or IL-2 (100 units/ml). After 24 h iNOS protein expression was quantitated by flow cytometry. (B) A JAK inhibitor was added to T cells that were cultured alone, with EC, with PHA, or with EC and PHA. After 24 h iNOS expression was quantitated by flow cytometry. DMSO, vehicle control. All data are representative of at least three independent experiments.

Because JAK/STAT signaling pathways are activated during T cell activation in response to both TCR stimulation and IL-2, we examined the contribution of this pathway to the inhibition of iNOS expression in activated T cells. T cells were cocultured with EC in the absence or presence of PHA and treated with a JAK inhibitor. JAK inhibition prevented the reduction of iNOS expression in T cells activated by EC and PHA (Fig. 4B).

iNOS Increases the Number of Proliferated T Cells in Response to Allogeneic EC.

Because NO has immune-modulatory properties, we wondered whether iNOS expression in bystander T cells could affect antigen-mediated immune responses (18). CD8 T cells were CFSE-labeled before coculture with allogeneic EC for 10 days to assess T cell proliferation, and either PBS (vehicle control) or 1400W was added to cocultures to determine the effect of iNOS activity on T cell proliferation. Allogeneic EC induced the proliferation of human CD8 T cells, and inhibition of iNOS with 1400W reduced the number of proliferated T cells (Fig. 5A). In three independent experiments using different donors, inhibition of iNOS with 1400W reduced the number of proliferated T cells in response to allogeneic EC by 47.8 ± 3.0% as compared with PBS-treated controls (P < 0.001). To confirm that the 1400W effect was mediated through reduced NO production, we treated PBS or 1400W-treated cocultures with the NO donor DETA-NO (1 μM). Replacing NO in this manner completely prevented the 1400W-mediated reduction in the number of proliferated T cells (Fig. 5A). Addition of high concentrations of DETA-NO (100 μM) inhibited proliferation in all groups (SI Fig. 11C), consistent with the inhibitory effects of high concentrations of NO on T cell proliferation reported previously (18). Similar to its effects on T cell proliferation, inhibition of iNOS with 1400W also reduced IL-2 levels in the supernatant of T cell–EC cocultures (Fig. 5B). These results demonstrate that iNOS in bystander T cells increases the number of proliferated T cells in response to allogeneic EC.

Fig. 5.

Fig. 5.

Open in a new tab

iNOS increases the number of proliferated T cells in response to allogeneic EC. (A) CD8 T cells were CFSE-labeled before addition to allogeneic EC, and iNOS activity was inhibited with 1400W. The NO donor DETA-NO (1 μM) was added to T cell cultures, and the number of proliferated T cells was quantitated. (B) CD8 T cells were cocultured with allogeneic EC in the absence or presence of 1400W. After 3 days supernatant was removed and IL-2 levels were measured by ELISA. All data are representative of at least three independent experiments.

iNOS Is Expressed in Nonactivated T Cells in Vivo.

As noted in the Introduction, we first identified iNOS-expressing human T cells within the wall of human arteries undergoing rejection in our humanized mouse model of GA (11). In this model iNOS is expressed only by infiltrating T cells and is not detected within vascular cells (11). To determine whether iNOS-expressing human T cells in vivo have been activated by alloantigen, the expression of iNOS in proliferating T cells [proliferating cell nuclear antigen (PCNA)-positive] or CD69-positive T cells was examined by two-color immunohistochemistry on sections of human transplanted coronary artery segments harvested from SCID/beige mice 4 weeks after adoptive transfer of allogeneic human T cells. This model of vascular inflammation leads to antigen-mediated arterial changes similar to those observed in human GA (11). iNOS-positive cells were distinct from PCNA- and CD69-positive cells in these arteries (Fig. 6A). The percentage of iNOS-positive cells that were PCNA- or CD69-negative was significantly greater than the percentage of iNOS-positive cells that were also PCNA- or CD69-positive (Fig. 6B). To determine whether there was a preferential expression of iNOS in nonactivated T cells, the above data were compared with the percentage of total T cells within transplanted artery segments that were not proliferating and not activated by quantitating the number of CD3-positive cells that were PCNA- or CD69-negative, respectively. The percentage of iNOS-positive cells that were PCNA-negative was significantly greater than the percentage of total T cells that were PCNA-negative (88.0 ± 11.0% and 47.8 ± 6.3%, respectively; P < 0.015). Also, the percentage of iNOS-positive cells that were CD69-negative was also significantly greater than the percentage of total T cells that were CD69-negative (84.3 ± 15.2% and 27.1 ± 16.7%, respectively; P < 0.015). These results indicate that iNOS is preferentially expressed in nonproliferating T cells that have not been activated by alloantigen in vivo.

Fig. 6.

Fig. 6.

Open in a new tab

iNOS is expressed in bystander T cells and increases vascular inflammation in vivo. (A) Human coronary artery segments were transplanted into SCID/beige mice, which were subsequently reconstituted with allogeneic human T cells. Arteries were harvested after 4 weeks, and immunohistochemistry was performed for iNOS (red) and PCNA or CD69 (blue). (Magnification: ×400.) (Left) iNOS and PCNA. (Right) iNOS and CD69. (B) The percentage of iNOS-positive cells that were PCNA- or CD69-negative or -positive in human transplanted arterial segments was quantitated. Data are presented as the mean ± SD of three arteries from different donors. ∗, P < 0.015. (C) Human coronary artery segments were transplanted into SCID/beige mice, which were subsequently reconstituted with allogeneic human T cells. Mice were injected daily with either PBS or 1400W, and arteries were harvested after 4 weeks. Arterial cross-sections were stained for CD3 (red), and the total number of T cells in the intima was quantitated (∗, P < 0.002; n = 5 artery segments from three separate donors). (D) Arterial cross-sections were stained for IL-2 (red), and the number of IL-2-positive T cells in the intima was quantitated (∗, P < 0.025; n = 5 artery segments from three separate donors).

iNOS Increases the Number of Activated T Cells in Allogeneic Arteries in vivo.

Finally, to determine whether iNOS contributes to antigen-dependent T cell responses in the vasculature in vivo, artery segments were interposed into the aortae of SCID/beige mice, and the mice were reconstituted with allogeneic T cells and treated with PBS or 1400W. The arteries were harvested after 4 weeks, and cross-sections were stained for CD3 or IL-2. There was a significant reduction in the number of T cells and of IL-2-expressing T cells within the intima of arteries from mice treated with 1400W as compared with PBS controls (Fig. 6 C and D). These data are consistent with the hypothesis that bystander T cells expressing iNOS within the vessel wall enhance the responses of alloreactive T cells.

Discussion

iNOS, expressed in human T cells, contributes to the pathology of GA in a humanized mouse model (11). In the present study, we show that EC induce the expression of iNOS in primary human T cells through an increase in NF-κB-mediated gene transcription. iNOS expression appears greater in CD8 than in CD4 T cells, possibly because of the differential signaling pathways often used by these types of cells (19). Surprisingly, iNOS expression is limited to T cells that are not activated by alloantigen, an observation explained by our finding that iNOS expression is inhibited during T cell activation by a JAK signaling pathway. The induction of iNOS in these bystander T cells results in the generation of NO, which increases the number of proliferated T cells in response to allogeneic EC.

In the current work, we have focused our studies on the intracellular pathways that regulate iNOS expression within human T cells. The mouse and human iNOS promoters are very different. The NF-κB responsive sites in the mouse are located within −1.0 kb of the transcription start site, whereas the main NF-κB responsive sites in the human gene promoter are positioned between −8.3 and −4.7 kb of the promoter (15, 20, 21). We have shown that iNOS expression in human T cells is induced, at least in part, by NF-κB-mediated transcription and determined that EC-induced iNOS expression in human T cells depends on a NF-κB binding site at −5.8 kb in the human promoter. The soluble factor(s) that induces iNOS in human T cells remains an area of investigation. Serum also has some ability to induce iNOS in T cells, but the EC effect does not depend on serum (SI Fig. 10C). Preliminary studies have suggested that iNOS induction is not mediated by IL-6 or several chemokines known to be secreted by EC (IL-8, IP-10, or SDF-1) (data not shown). Biochemical studies will be needed to isolate and identify the iNOS-inducing factor.

A surprising finding of the current study is the restriction of iNOS expression to human T cells that have not been activated. Indeed, the inhibition of iNOS expression in T cells in response to EC and PHA, or IL-2, shows that iNOS expression is inhibited during T cell activation, either directly by TCR-mediated signaling [because this can induce STAT5 activation (22)] or indirectly through the secretion of cytokines. The almost exclusive localization of iNOS to nonactivated T cells using allogeneic systems in vitro and in vivo, combined with the finding that inhibition of iNOS reduces T cell activation in vitro and in vivo, imply a biologically significant role for this pathway in vascular pathology. We are uncertain of the biological reason behind iNOS suppression in activated T cells and speculate that this may represent a means of escaping the inhibitory effects of high levels of NO that might occur when the NO source and target are the same cell. In the mouse, iNOS can be detected in unstimulated peripheral blood T cells, and its expression is inhibited by stimulation of these cells with IL-7 (23). Our results are consistent with this notion, and we have determined that inhibition of iNOS during T cell activation occurs through a JAK enzyme. Activation of JAK/STAT pathways is crucial in inducing the expression of proinflammatory molecules. On the other hand, JAK/STAT signaling can inhibit the immune response under certain circumstances. STAT1, STAT3, and STAT5 have all been shown to inhibit the expression of certain NF-κB-controlled inflammatory genes (24–26). STAT1 also inhibits cytokine-induced iNOS expression in the human fibroblast cell line 2fTGH through the inhibition of NF-κB-mediated gene transcription (27). Taken together, these findings implicate a complex interaction between NF-κB and JAK/STAT pathways in the regulation of inflammatory genes that may depend largely on cell type and activation status.

iNOS expression in T cells plays a physiological role in T cell development in the mouse and a pathological role in GA in a humanized mouse model. Using iNOS-knockout mice, Vig et al. (28) demonstrated that iNOS expressed by T cells regulates the development of memory T cells. In an experimental system using human cells and tissues we have determined that iNOS is preferentially expressed in nonactivated T cells. Because inhibition of iNOS, which is expressed only in infiltrating T cells within the arterial wall in our chimeric human–SCID mouse model (11), reduces vascular rejection, our current results implicate a role for a unique T cell effector response that is not induced by antigen recognition in vascular pathology. Indeed, we have shown that NO production by iNOS in these bystander T cells increases the number of proliferated and activated T cells in response to allogeneic EC both in vitro and in vivo. Although high levels of NO can inhibit T cell responses, NO has also been shown to have immune-stimulatory properties. Niedbala et al. (18) reported that treatment of human T cells with low concentrations of NO can increase T cell proliferation. Also, NO production by endothelial NO synthase augments TCR signaling in human T cells by affecting organization of the immunological synapse (29). Finally, NO is known to increase the activation of ras in human T cells, thereby providing a possible mechanism by which low concentrations of NO can augment T cell responses (30). In summary, our data identify bystander T cells as an endogenous source of NO in the human immune system and show that production of NO by iNOS is a previously undescribed antigen-independent effector function of this cell population.

Materials and Methods

Cell Isolation and Culture.

Human umbilical vein EC, fibroblasts, and T cells were isolated and cocultured by using a T cell:EC ratio of 10:1 as described previously (14).

Quantitation of Nitrite Release.

Four hundred microliters of supernatant from cocultures of T cells and EC were collected after 3 days, the proteins were removed by ethanol precipitation, and the remaining constituents were reconstituted in 100 μl of RPMI medium 1640. Nitrite concentration was quantitated by using a NO-specific chemiluminescence analyzer (Sievers, Boulder, CO).

CFSE Labeling and Cell Sorting.

T cells were labeled with 1 μM CFSE (Molecular Probes, Carlsbad, CA) before coculture with EC as described in http://www.pnas.org/cgi/content/full/0607731104/DC1SI Materials and Methods and either FACS sorted or analyzed as described previously (14).

Luciferase Assay.

Plasmids containing the wild-type and mutated −7.2-kb region of the human iNOS promoter attached to a luciferase reporter gene (iNOS-luc) were obtained from David Geller (University of Pittsburgh, Pittsburgh, PA). Plasmids were transfected into unstimulated human T cells as per the manufacturer's protocol (Amaxa, Gaithersburg, MD). After 24 h in coculture luciferase activity was quantitated with a Bright-Glo Luciferase Assay system (Promega, Madison, WI). Light output was quantitated by using a luminometer (Berthold LB9501; Wallac, Gaithersburg, MD).

Human Artery Transplants.

Human arteries were transplanted into SCID/beige mice, reconstituted with human PBMC, and treated as described previously (11).

Statistics.

A paired t test was used to assess significant differences between the groups indicated. Significant differences are defined as having a P < 0.05.

Details.

For additional information see http://www.pnas.org/cgi/content/full/0607731104/DC1SI Materials and Methods.

Supplementary Material

Supporting Information
pnas_0607731104_index.html (9.9KB, html)

Acknowledgments

We thank Dr. David Geller for kindly providing the human iNOS promoter–luciferase reporter constructs and Dr. Sankar Ghosh for helpful discussions and comments. The research performed was funded by National Institutes of Health Grant P01 HL070295-06. J.C.C. is the recipient of generous support from a Canadian Institutes of Health Research postdoctoral fellowship.

Abbreviations

CFSE

carboxyfluorescein diacetate succinimidyl ester

EC

endothelial cells

GA

graft arteriosclerosis

iNOS

inducible NO synthase

PCNA

proliferating cell nuclear antigen

PHA

phytohemagglutinin

TCR

T cell receptor.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS direct submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0607731104/DC1.

References

  • 1.Lowenstein CJ, Glatt CS, Bredt DS, Snyder SH. Proc Natl Acad Sci USA. 1992;89:6711–6715. doi: 10.1073/pnas.89.15.6711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wei XQ, Charles IG, Smith A, Ure J, Feng GJ, Huang FP, Xu D, Muller W, Moncada S, Liew FY. Nature. 1995;375:408–411. doi: 10.1038/375408a0. [DOI] [PubMed] [Google Scholar]
  • 3.Cramer JP, Mockenhaupt FP, Ehrhardt S, Burkhardt J, Otchwemah RN, Dietz E, Gellert S, Bienzle U. Trop Med Int Health. 2004;9:1074–1080. doi: 10.1111/j.1365-3156.2004.01312.x. [DOI] [PubMed] [Google Scholar]
  • 4.Kuhlencordt PJ, Chen J, Han F, Astern J, Huang PL. Circulation. 2001;103:3099–3104. doi: 10.1161/01.cir.103.25.3099. [DOI] [PubMed] [Google Scholar]
  • 5.Koglin J, Glysing-Jensen T, Mudgett JS, Russell ME. Circulation. 1998;97:2059–2065. doi: 10.1161/01.cir.97.20.2059. [DOI] [PubMed] [Google Scholar]
  • 6.Nathan C, Calingasan N, Nezezon J, Ding A, Lucia MS, La Perle K, Fuortes M, Lin M, Ehrt S, Kwon NS, et al. J Exp Med. 2005;202:1163–1169. doi: 10.1084/jem.20051529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Stuehr DJ, Cho HJ, Kwon NS, Weise MF, Nathan CF. Proc Natl Acad Sci USA. 1991;88:7773–7777. doi: 10.1073/pnas.88.17.7773. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Schneemann M, Schoedon G, Hofer S, Blau N, Guerrero L, Schaffner A. J Infect Dis. 1993;167:1358–1363. doi: 10.1093/infdis/167.6.1358. [DOI] [PubMed] [Google Scholar]
  • 9.Dumarey CH, Labrousse V, Rastogi N, Vargaftig BB, Bachelet M. J Leukocyte Biol. 1994;56:36–40. doi: 10.1002/jlb.56.1.36. [DOI] [PubMed] [Google Scholar]
  • 10.Esaki T, Hayashi T, Asai Y, Kumar TN, Kano H, Muto E, Iguchi A. Heart Vessels Suppl. 1997;12:89–92. [PubMed] [Google Scholar]
  • 11.Koh KP, Wang Y, Yi T, Shiao SL, Lorber MI, Sessa WC, Tellides G, Pober JS. J Clin Invest. 2004;114:846–856. doi: 10.1172/JCI21767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Choi J, Enis DR, Koh KP, Shiao SL, Pober JS. Annu Rev Immunol. 2004;22:683–709. doi: 10.1146/annurev.immunol.22.012703.104639. [DOI] [PubMed] [Google Scholar]
  • 13.Garvey EP, Oplinger JA, Furfine ES, Kiff RJ, Laszlo F, Whittle BJ, Knowles RG. J Biol Chem. 1997;272:4959–4963. doi: 10.1074/jbc.272.8.4959. [DOI] [PubMed] [Google Scholar]
  • 14.Dengler TJ, Pober JS. J Immunol. 2000;164:5146–5155. doi: 10.4049/jimmunol.164.10.5146. [DOI] [PubMed] [Google Scholar]
  • 15.Taylor BS, de Vera ME, Ganster RW, Wang Q, Shapiro RA, Morris SM, Jr, Billiar TR, Geller DA. J Biol Chem. 1998;273:15148–15156. doi: 10.1074/jbc.273.24.15148. [DOI] [PubMed] [Google Scholar]
  • 16.Weiss A, Imboden J, Shoback D, Stobo J. Proc Natl Acad Sci USA. 1984;81:4169–4173. doi: 10.1073/pnas.81.13.4169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hughes CC, Savage CO, Pober JS. J Exp Med. 1990;171:1453–1467. doi: 10.1084/jem.171.5.1453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Niedbala W, Wei XQ, Campbell C, Thomson D, Komai-Koma M, Liew FY. Proc Natl Acad Sci USA. 2002;99:16186–16191. doi: 10.1073/pnas.252464599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ravichandran KS, Burakoff SJ. J Exp Med. 1994;179:727–732. doi: 10.1084/jem.179.2.727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Xie QW, Kashiwabara Y, Nathan C. J Biol Chem. 1994;269:4705–4708. [PubMed] [Google Scholar]
  • 21.Marks-Konczalik J, Chu SC, Moss J. J Biol Chem. 1998;273:22201–22208. doi: 10.1074/jbc.273.35.22201. [DOI] [PubMed] [Google Scholar]
  • 22.Welte T, Leitenberg D, Dittel BN, al-Ramadi BK, Xie B, Chin YE, Janeway CA, Jr, Bothwell AL, Bottomly K, Fu XY. Science. 1999;283:222–225. doi: 10.1126/science.283.5399.222. [DOI] [PubMed] [Google Scholar]
  • 23.Sade H, Sarin A. Cell Death Differ. 2004;11:416–423. doi: 10.1038/sj.cdd.4401347. [DOI] [PubMed] [Google Scholar]
  • 24.Hoentjen F, Sartor RB, Ozaki M, Jobin C. Blood. 2005;105:689–696. doi: 10.1182/blood-2004-04-1309. [DOI] [PubMed] [Google Scholar]
  • 25.Kramer OH, Baus D, Knauer SK, Stein S, Jager E, Stauber RH, Grez M, Pfitzner E, Heinzel T. Genes Dev. 2006;20:473–485. doi: 10.1101/gad.364306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Luo G, Yu-Lee L. Mol Endocrinol. 2000;14:114–123. doi: 10.1210/mend.14.1.0399. [DOI] [PubMed] [Google Scholar]
  • 27.Ganster RW, Taylor BS, Shao L, Geller DA. Proc Natl Acad Sci USA. 2001;98:8638–8643. doi: 10.1073/pnas.151239498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Vig M, Srivastava S, Kandpal U, Sade H, Lewis V, Sarin A, George A, Bal V, Durdik JM, Rath S. J Clin Invest. 2004;113:1734–1742. doi: 10.1172/JCI20225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ibiza S, Victor VM, Bosca I, Ortega A, Urzainqui A, O'Connor JE, Sanchez-Madrid F, Esplugues JV, Serrador JM. Immunity. 2006;24:753–765. doi: 10.1016/j.immuni.2006.04.006. [DOI] [PubMed] [Google Scholar]
  • 30.Lander HM, Ogiste JS, Pearce SF, Levi R, Novogrodsky A. J Biol Chem. 1995;270:7017–7020. doi: 10.1074/jbc.270.13.7017. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
pnas_0607731104_index.html (9.9KB, html)
pnas_0607731104_1.pdf (191.4KB, pdf)
pnas_0607731104_2.pdf (200.7KB, pdf)
pnas_0607731104_3.pdf (203.9KB, pdf)
pnas_0607731104_4.pdf (233.6KB, pdf)
pnas_0607731104_5.pdf (234.1KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES