Each graph is representative of 3 independent experiments. C. Cells transfected with AHR containing a valine point-mutation show similar ED50 to Cos-1 cells with AHRb isoform. Cos-1 cells were transfected with an AHR containing the same point mutation (valine for alanine) thought to be responsible for the low affinity of the AHRd isoform compared to AHRb, and compared to the wildtype AHR response. These cells also harbor a luciferase gene next to the DRE. C. Cos-1 cells were exposed to TCDD. D. Cos-1 cells were exposed to SU5416. The graphs represent normalized data from 0 response to 100% response. They are representative of 2 independent experiments. cells, we also analyzed the upregulation of CD39, which is an ectoenzyme that degrades ATP to AMP and is strongly associated with Tregs that can suppress ATP-related effects and pathogenic Th17 cells. As can be seen in figure 5G, SU5416 upregulated CD39 in the FoxP3+ T cells, a finding that has recently been reported with TCDD [28]. Finally, as literature is emerging that the ability of AHR ligands to enhance T-cell differentiation may be dependent as much on surrounding conditions and inflammatory milieu as on the ligand tested, we assessed the ability of SU5416 to enhance Th17 differentiation in Th17 conditions. ?Naive T cells were placed in culture with IL-6 and TGF-b, and harvested after 3 days of culture. Figure S4 shows that at low doses SU5416 caused a small increase in IL-17 protein by ELISA in the supernatant. At higher doses we did not see this effect.
Discussion
SU5416 was specifically designed as an inhibitor to VEGF-R2 [29], with the hope that it would join the armamentarium of antiangiogenesis drugs used to combat malignancy. It initially showed promise in preclinical trials, was used in phase I and phase II studies of hematologic and solid cancers [30,31,32], and ultimately did make it to phase III trials for metastatic colorectal cancer.While the drug was well tolerated, its ability to significantly reduce the growth of cancer over standard regimens was unimpressive, and it remains an experimental drug. While its clinical role remains unclear, SU5416 continues to be utilized in laboratory studies to confirm the importance of VEGF in various mechanistic studies, including cell trafficking, organ rejection, and autoimmunity [33,34,35]. The data presented in this manuscript demonstrate that SU5416 is a strong ligand of the AHR. The unique finding that SU5416 binds the high- and low-affinity polymorphisms of the AHR similarly was rather surprising to us, and will require further attention and characterization. The mouse AHR can arise from an allele that encodes a receptor with high binding affinity for ligand (denoted Ahrb allele) or with low binding affinity for ligand (denoted as the Ahrd allele). The AHRd is known to have approximately one-fifteenth to one-twentieth the binding affinity to TCDD as the AHRb [36], and this low affinity polymorphism resembles the isoform found in humans [37,38,39,40]. C57BL/6 mice harbor the high-affinity AHRb receptor, and this strain has been utilized for much of the initial characterization of TCDD and other environmental toxicants [3].
In our search for relevant ligands of the AHR, we decided to focus on those that hadFigure 5. SU5416 upregulates CYP1A1. A. Spleens were harvested from mice and processed in the standard fashion. Cells were incubated for 4 hours in culture with titrating doses of SU5416 as indicated, and afterwards mRNA was measured for CYP1A1. B. SU5416 upregulates CYP1B1 and IDO. Same assay as A, but mRNA for CYP1B1 and IDO were measured. C. Upregulation of CYP1A1 mRNA is dependent on the AHR. Splenocytes from C57BL/6J or AHR2/2 mice were exposed to media, IFN-c 100 ng/ml, or SU5416 500 nM for 4 hours. mRNA was then harvested and assayed for CYP1A1. nd represents “not detected”. *** – p#0.001. D. IDO upregulation by SU5416 is AHR-dependent. Same assay as in C, but IDO mRNA was assessed. * – p#0.05. * – p#0.01. E. SU5416 induces IDO in the pDC/T cell coculture to a greater extent than TCDD. pDC/T cell coculture was utilized as described previously [25]. Culture was performed for 5 days with SU5416 500 nM, TCDD 10 nM, FICZ 100 nM, or control, at which point mRNA was harvested and measured for IDO. * – p#0.05. *** – p#0.001. F. SU5416 induces FoxP3 in the pDC/T cell coculture to a greater extent than TCDD. Same ?assay as in E, but mRNA was assayed for FoxP3. * – p#0.05. ** – p#0.01. G. SU5416 enhances FoxP3 expression and CD39 on naive T cells in the ?presence of TGF-b. Naive T cells were placed in culture with DMSO (1:46104 dilution), TGF-b (2 ng/ml), or TGF-b (2 ng/ml) and SU5416 (250 nM), harvested on day 3, and analyzed by flow cytometry. All of the above figures are representative of 3 independent experiments.
significant potency in the AHRd isoform, as these ligands would have more clinical relevance in humans. We inadvertently identified that SU5416 had similar binding characteristics with both polymorphisms at doses that are similar to what were used in humans in Phase I trials with SU5416 [30], as seen in the titration in figure 3D. This is an unusual characteristic that has rarely been exhibited by any of the known ligands of the AHR [41]. The importance of this is due to the following: First, the information is clinically significant given that humans harbor an AHR isoform that more similarly represents the AHRd. Second, its structure will serve as a model in our search for endogenous ligands of the AHR. It makes sense that a true endogenous ligand would activate both polymorphisms of the AHR similarly, given that mice (and humans) that harbor the low affinity polymorphism do not exhibit the patent ductus venosus found in AHR nulls and hypomorphs. This is further supported by the ability of SU5416 to close the DV in AHR hypomorphs (a requirement of the true endogenous ligand) [12]. To this point we have been unable to model the binding sites of these polymorphisms by crystallography, but the finding that SU5416 can bind both of these similarly may help us in these efforts.
