Microarray Analysis of Cytokine and Chemokine Gene Expression after Prednisolone Treatment in Murine Experimental Autoimmune Uveoretinitis

Abstract

purpose. The purpose of this study was to investigate changes in gene expression of cytokines and chemokines and their receptors after systemic prednisolone treatment in experimental autoimmune uveoretinitis (EAU).

methods. EAU mice received one intravenous injection of 7.5 mg/kg prednisolone (PDS) sodium phosphate at the peak of inflammation. EAU mice treated with only solvent served as the control. Total RNA was extracted from the whole eyes 1, 2, and 3 days after treatment. Gene expression analysis was conducted with a cDNA microarray, which contains 117 individual transcripts encoding the genes of the cytokines and chemokines and their receptors (29 cytokines and 34 cytokine receptors; 33 chemokines and 21 chemokine receptors). Comparisons of expression between PDS-treated and placebo-treated EAU mice at each time point were performed. The genes were sorted into clusters based on expression profiles, to clarify the gene regulation pattern after treatment.

results. Forty-seven genes had a significant decrease in expression 1 day after treatment, 10 genes on day 2, and 46 genes on day 3. Ten genes were upregulated on day 1, but no gene was upregulated thereafter. Hierarchical cluster analysis of the microarray demonstrated that gene expression changes in EAU after treatment with PDS showed four patterns: flat, mountain, steep downhill, and less steep downhill.

conclusions. The implications of the clusters after treatment with PDS remain unclear. The results showed that hierarchical cluster analysis based on comprehensive gene expression profiles may provide a powerful tool for identifying genes not previously associated with the therapeutic targets in ocular inflammation.

Experimental autoimmune uveoretinitis (EAU) is an organ-specific, T-cell (CD4⁺)-mediated disease that is induced in genetically susceptible murine strains by immunization with retinal proteins (e.g., interphotoreceptor retinoid-binding proteins [IRBPs] of tyrosinase family proteins) or by the adoptive transfer of T-cells specific for these antigens into naïve syngenetic recipients. ^{1 2 3 4 5} EAU is characterized by granuloma formations in the retina, infiltration of polynuclear lymphocytes and macrophages into the retina, vasculitis, and destruction of photoreceptor cells. ^{6 7 8} EAU serves as an animal model for ocular inflammatory diseases with a putative autoimmune etiology (e.g., ocular sarcoidosis, Behçet’s disease, or the Vogt-Koyanagi-Harada syndrome). ^{4 7}  

Administration of systemic corticosteroids is a standard therapy for severe panuveitis. The pharmacologic effects of corticosteroids are based on several mechanisms. The accumulation of neutrophils and macrophages in inflamed tissue decreases after corticosteroid administration. ⁹ Corticosteroids also modulate cytokine expression and induce apoptosis of T-cells in a dose-dependent manner. ^{10 11 12 13} Recently, the production of interferon (IFN)-γ, one of the best known Th1 cytokines, was reported to decrease after corticosteroid treatment in an EAU rat model. ¹⁴  

In previous studies, it was reported that domination of Th1-type immune responses plays an important role in the pathogenesis of EAU. ^{15 16 17 18} The expression of several chemokines and cytokines was identified by quantitative polymerase chain reaction (PCR) and immunohistochemistry in an EAU mice model. ¹⁷ The roles of some cytokines and chemokines—for example, interleukin (IL)-2, IFN-γ, tumor necrosis factor (TNF)-α, chemokine (C-X-C motif) ligand 9 (CXCL9)/monokine induced by IFN-γ (Mig), CXCL10/IP-10 (IFN-γ-inducible protein of 10 kDa), and chemokine (C-X-C motif) receptor 3 (CXCR3)—have been reported in the development of EAU. ^{19 20 21} In addition, the Th2-type cytokines IL-4 and IL-10 play a role in remission of the disease. ^{20 22 23} However, the regulatory impact of corticosteroids on inflammatory genes in the EAU model is still a matter for speculation. 

We examined comprehensive gene expression analysis of cytokines and chemokines and their receptors after corticosteroid treatment in an EAU murine model. The results provide new information about which inflammatory genes play key roles in the remission of EAU after corticosteroid treatment. 

Materials and Methods

Mice

Ninety C57BL/6 mice were purchased from Japan SLC, Inc. (Shizuoka, Japan), housed in pathogen-free conditions in our animal facility at Osaka University, and used at 8 to 10 weeks of age. The animals were treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Peptides and Adjuvants

Human IRBP peptides corresponding to IRBP_1-20 (GPTHLFQPSLVLDMAKVLLD) was selected as an uveitogenic peptide, as described previously. ^{8 24} The peptide was purchased from Gene Net Co., Ltd. (Fukuoka, Japan). Complete Freund’s adjuvant (CFA) and Mycobacterium tuberculosis H37Ra were purchased from Sigma-Aldrich (St. Louis, MO) and DIFCO (Detroit, MO), respectively. Bordetella pertussis inactive bacterial suspension was purchased from Wako Pure Chemical Industries (Tokyo, Japan). IRBP peptides were dissolved in phosphate-buffered saline at 5 mg/mL. 

Induction and Clinical Assessment of EAU

C57BL/6 mice were inoculated subcutaneously in the right hind footpad, right thigh, and base of the tail with 100 μg IRBP peptide in emulsion with CFA supplemented with 6.0 mg/mL M. tuberculosis (1:1, vol/vol). Inactivated suspension of B. pertussis (10¹⁰ organisms/mouse) in a volume of 100 μL was injected intraperitoneally as additional adjuvants. Beginning 10 days after immunization, slit lamp examinations were performed daily. A peak inflammatory response was observed on day 16 after the IRBP peptide injection. The EAU clinical scores were graded from 0 to 4, as previously described. ^{24 25} We analyzed mice with grade 2 EAU throughout the study. 

Administration of Prednisolone Sodium Phosphate

The EAU mice at the peak of inflammation were treated with one intravenous injection of 7.5 mg/kg of body weight prednisolone (PDS) sodium phosphate (LKT Laboratories, Inc., St. Paul, MN) dissolved in sterile 0.9% NaCl solution. ^{26 27} This dose is similar to that used clinically to treat uveitis, such as occurs in the Vogt-Koyanagi-Harada syndrome. ²⁸  

Tissue Sampling from Mice

For the examination of gene expression in EAU, 10 mice were killed, and the eyeballs were collected on day 16. Mice injected with only adjuvant served as control animals. For the analysis of the effect of PDS, 10 mice were killed, and the eyeballs were collected on days 1, 2, and 3 after administration of PDS. EAU mice that received only sterile 0.9% NaCl solutions served as the control and the eyeballs were collected on days 1, 2, and 3 after administration of solvent. 

Total RNA Isolation

The eyes of 10 mice were pooled as one sample at each time point, to minimize variation among the animals. ^{29 30} The globes were removed from each mouse and prepared immediately. The pooled eyes were homogenized (model HG30 homogenizer; Hitachi Koki Co., Ltd., Tokyo, Japan) in RNA extraction reagent (TRIzol; Invitrogen Inc., Carlsbad, CA). Total RNA was isolated according to the manufacturer’s instructions. The cetyltrimethylammonium bromide method was used to remove polysaccharide contamination. ^{29 31} Any DNA contamination was removed by incubation with DNase I. Final products yielded a 260/280-nm ratio of 1.9 to 2.0 and a 230/260-nm ratio less than 0.5. The quality was checked via gel electrophoresis by visualization of intact 28S and 18S ribosomal RNA bands. The quantity was determined based on 260-nm absorbance. 

DNA Chips

A description of mouse cytokine and chemokine chips (Kakengeneqs Co., Ltd., Chiba, Japan) is available at http://www.geocities.jp/nohashida2005/genes.pdf. Individual transcripts (n = 117) were included on the DNA chip that contained most cytokines and chemokines and their receptors (29 cytokines, 34 cytokine receptors; 33 chemokines, 21 chemokine receptors). Two housekeeping genes (GAPDH and β-actin) and three unrelated murine reagents (tobacco chloroplast DNA, part of a cloning vector, and a solvent that suspends the PCR products) were spotted onto the chip as internal positive and negative controls, respectively. This cDNA chip covers almost all genes of the cytokines and chemokines and their receptors. ²⁹  

cDNA Preparation and Array Hybridization

Total RNA samples (100 μg) were converted to double-strand cDNA by using a custom kit (LabelStar Array kit; Qiagen, Valencia, CA) and labeled with cyanine 3 (Cy3)-conjugated dUTP. Reference total RNA was labeled with cyanine 5 (Cy5)-conjugated dUTP (PerkinElmer, Boston, MA), according to the manufacturer’s protocol. Each RNA sample was labeled in triplicate with Cy3 and Cy5. The labeled cDNAs were mixed and hybridized simultaneously to the cDNA microarray chip. Expression patterns in the EAU mice were compared with those of control mice. Expression patterns after PDS treatment in the treated mice were compared with those in the day-matched control mice. Array hybridization was performed according to the manufacturer’s instructions (Kakengeneqs Co., Ltd.), except that the hybridization and washing temperature was 65°C. 

Microarray Quantification

The fluorescent images of hybridized microarrays were primarily obtained with an array scanner (model 428; Affymetrix, Santa Clara, CA), with independent laser excitation at 532- and 635-nm wavelengths for the Cy3 and -5 labels, respectively. Each chip was scanned twice. Raw fluorescence intensity data were used to calculate signal intensities of the spots (DNASIS Array; Hitachi Software Engineering Co., Ltd., Tokyo, Japan). The software provides absolute analysis of the data from each chip, including normalization of identified signal intensities, background calculation, and comparative analysis between any two chips. Data were normalized by the median of the intensity of the housekeeping gene. Triplicate data for one RNA sample were averaged for each gene. Transcripts not found in two of three experiments in either the control and experimental groups were not analyzed further. Normalized and averaged fluorescence ratios of genes from the experimental mice were compared with those from the control animals. Data were used to calculate the percentage increase or decrease of change. A twofold change in gene expression was used as the cutoff. Three independent experiments were performed to compare the experimental and control data. 

Quantitative RT-PCR

We performed quantitative RT-PCR of the selected genes to validate the microarray data. Commercially available primers and probe sets of the genes were prepared for the analysis. IL21r, TNFa, and CXCL1 were selected for the experiment involving developing EAU, and CXCL4, CXCL5, CXCR6, and IL16 were selected for that after corticosteroid treatment. Two genes (GAPDH, 18s rRNA) were selected as the positive control, to monitor PCR reactions. 

Total RNA (500 ng) was reverse transcribed with random hexamer primers (GE Healthcare, Piscataway, NJ) to synthesize double-strand cDNA (SuperScript II; Invitrogen Corp.), according to the manufacturer’s instructions. A low-density array (Applied Biosystems, Inc. [ABI], Foster City, CA) containing a preset assay-on-demand mix of primers and FAM dye-labeled minor groove-binding probes (TaqMan; ABI) for target genes were used for real-time PCR. Low-density arrays are research tools for profiling gene expression using the comparative C _t method of relative quantification. ^{32 33} 18s rRNA were used as a calibrator in relative quantification calculations. Gene-specific PCR products were measured continuously by a sequence detection system (Prism 7900HT; ABI), according to the manufacturer’s protocols. The cycling parameters were incubation for 2 minutes at 50°C and 10 minutes at 94.5°C, 40 PCR cycles at 97°C for 15 seconds each, and annealing-extension at 59.7°C for 1 minute. Each sample was run in quadruplicate. 

Statistical Analysis and Clustering Method

The microarray data were hierarchically clustered with average linkage clustering ³⁴ and visualized (DNASIS Stat software; Hitachi Software Engineering Co.). The Silhouette index and Dunn index in the clustering analysis were used to determine the number of clusters. Comparison of microarray analysis and quantitative RT-PCR experiments at respective time points was performed by t-test. ³⁵ P < 0.05 was considered statistically significant. 

Results

Reliability of the Microarray Technique

Figure 1Ais a scatterplot of Cy3- versus Cy5-labeled intensity. Pair-wise comparisons of gene expression levels in three independent experiments displayed a correlation coefficient of R ²= 0.995, indicating reliability. The reproducibility of the microarray technique was determined by comparing the cluster image of each analysis of independent amplification from the same RNA source. Figure 1Bshows the cluster images from six independent experiments. In three independent experiments, total RNA derived from the eyes of the EAU and control mice was labeled with Cy3 and Cy5, respectively. In another three experiments, total RNA derived from the eyes of EAU and control mice was labeled with Cy5 and Cy3, respectively. Chip-to-chip normalization was performed by comparing the hybridization intensities of control probe sets spotted onto each microarray chip. Results of triplicate analysis of each sample were consistent. 

Microarray Data Results of Expression Changes in Developing EAU

Most analyzed genes were upregulated in EAU (Table 1) . Gene expression was considered consistently changed only if the average multiple of change between the control and experimental pools was higher than two in at least two hybridizations in three independent experiments. Under these conditions, of the 117 genes on the chips, 91 genes were significantly upregulated (more than twofold) during EAU; 17 genes had no significant changes. We could not evaluate nine genes because their expression was undetectable in the control animals. However, these nine genes were certainly upregulated during the development of EAU, although we could not calculate the extent to which this occurred. 

Most upregulated genes were Th1-type cytokines and chemokines and their receptors; however, some Th2-type genes were among those significantly upregulated. Notably, cytokine and chemokine receptors tended to be preferentially upregulated in developing EAU (8 of the top 10 genes). 

Microarray Results of Expression Changes after Administration of PDS

Table 2shows the functional categories and the changes (x-fold) in expression. The expression of 47 genes decreased significantly on day 1 after treatment, that of 10 genes on day 2, and that of 46 genes on day 3. Ten genes were upregulated on day 1, but none thereafter. 

Table 3shows the 10 upregulated genes on day 1 and the top 10 downregulated genes at each time point. The top 10 downregulated genes differed greatly at each time point. The clinical score did not change significantly throughout the observation period (data not shown). 

Pattern of Clustering of Gene Expression after Treatment with PDS

Figure 2Bshows the cluster image of the different gene expression profiles. The number of clusters was determined by calculating the scores of the Silhouette index and Dunn index. The genes were clustered into four groups according to the highest score on each index (Silhouette index, −0.12; Dunn index, 0.51; Fig. 2A ). 

Cluster 1 contained 28 genes with no change in expression (approximately onefold) on day 1 or 2, or both, after treatment. On day 3, some of them showed significantly decreased expression, but most of them did not. Therefore, the expression pattern of this cluster could be called flat (Fig. 2B) . A representative gene in cluster 1 was IL13ra2, the expression of which increased approximately 14-fold during EAU, but no significant change in expression was observed after administration of PDS. 

Cluster 2 contained most of the examined genes (n = 67) with constantly decreased expression at all time points after treatment. Because the degree of downregulation was weak on day 2, the shape of this cluster resembled a mountain (Fig. 2B) . The representative genes in cluster 2 were CCR5; IL2ra; CCL11/eotaxin; and CCL5/regulated on activation, normally T-expressed, and presumably secreted (RANTES). 

Cluster 3 contained five genes (IL7r, XCL1/lymphotactin, CXCR6/BONZO, CCR2, and lymphotoxin/TNFb) with significantly upregulated expression (>2.5-fold) on day 1 but dramatically decreased expression on days 2 and 3. This cluster had a steep downhill pattern. 

Cluster 4 contained 17 genes, with slightly increased expression (twofold or less) on day 1 and gradually decreased expression on days 2 and 3 (less steep downhill pattern). Figure 3shows the detailed information on each cluster. 

Quantitative RT-PCR Results

Because quantitative differences in analysis of relative mRNA levels are often observed when the Northern blot, array hybridization, and real-time PCR methods ³⁶ were used, we confirmed the exact quantitative changes during development of EAU and after treatment with PDS in selected genes by RT-PCR. Figure 4shows the multiple changes in these genes in microarray analysis and quantitative RT-PCR experiments. The expression ratios of the genes correlated well between the microarray analysis and quantitative RT-PCR experiment, indicating experimental reliability. 

Discussion

To the best of our knowledge, this article is the first to report a comprehensive expression analysis of inflammatory genes after corticosteroid administration in an EAU murine model. The results demonstrate that administration of a corticosteroid suppressed gene expression of many cytokines and chemokines and their receptors, the expression of which was upregulated at the inflammatory peak in the EAU model. This study also showed that gene expression profiles after corticosteroid treatment differed depending on the observation time. One day after administration of PDS, 47 genes were significantly downregulated; however, only 10 genes were downregulated on day 2, and then 46 genes on day 3. In addition, the top-10 downregulated genes differed remarkably at each time point. Therefore, to clarify the gene regulation pattern after treatment with PDS, we estimated these results in a hierarchical clustering format, which is believed to be a powerful method to analyze the multiple and sequential changes in microarray data. We also found that the genes were clustered into four groups by pattern: flat, mountain, steep downhill, and less steep downhill patterns. 

In cluster 1, which contained 28 genes, the gene expression change remained almost constant (flat pattern), and the degree of change was unremarkable (twofold or less). GAPDH and β-actin were in this cluster, suggesting that these genes may not be the main target of PDS. Of interest, IL2 was also in this cluster. The mRNA expression of IL2 reaches a peak during the active phase of EAU and declines parallel to disease resolution. ²⁰ We also observed a significant increase in gene expression of this cytokine at the inflammatory peak; however, the expression pattern did not change dramatically after treatment with PDS. Several studies also showed that corticosteroids failed to inhibit IL-2 production significantly. ^{37 38 39} However, cyclosporine, a representative T-cell immunosuppressant, downregulates IL-2 gene expression. ^{40 41 42} Recent investigations have shown the effectiveness of cyclosporine in treating uveitis refractory to corticosteroid therapy. ^{43 44 45} The difference between PDS and cyclosporine in regulating IL-2 may explain those recent clinical experiences. 

Among 117 cytokines and chemokines and their receptors, more than half (n = 67) were in cluster 2, which had a mountain pattern. The number of significantly downregulated genes was fewer only on day 2. Considering all evidence, cluster 2 could represent the main pattern of gene expression changes after PDS treatment. This cluster contained IFNg, IL-6, TGFb, IL2r, CXCL9, and RANTES/CCL5, all well-known participants in ocular inflammation. ^{18 19 20 21 46 47 48 49} Kodama et al. ¹⁴ recently showed that the level of intraocular IFN-γ in EAU rats decreased after dexamethasone was injected into the anterior chamber, which agrees with our data. The reason downregulation of gene expression by PDS occurred on day 2 to a lesser extent than at the other time points remains unclear. One possibility is that there is a temporary physiologic compensatory mechanism during PDS treatment. 

A noteworthy finding was the existence of cluster 3. Five genes, IL7r, XCL1/lymphotactin, CXCR6/BONZO, CCR2, and lymphotoxin A/TNFb, were significantly upregulated by PDS on day 1 but sequentially downregulated on days 2 and 3 (steep downhill pattern). IL-7 plays a role in stimulating the proliferation of murine pre-B cells. ⁵⁰ Investigators have shown that B cells develop in the bone marrow from progenitor cells. ⁵¹ Differentiation of progenitor cells requires signaling through IL-7r, mediated by the growth-signaling factor, consisting of IL-7 and 30-kDa protein cofactor. ⁵² The gene of CXCR6/BONZO, the receptor for CXCL16, is important in the trafficking of effector T-cells mediating type-1 inflammation. ⁵³  

XCL1/lymphotactin is chemotactic for CD4⁺ and CD8⁺ T-cells but not for monocytes and induces a rise in intracellular calcium in peripheral blood lymphocytes. ^{54 55} These genes are not associated with EAU development and/or the effect of corticosteroid treatment on ocular inflammation. 

Other representative genes in cluster 3 were CCR2 and lymphotoxin A/TNFb. CCR2, a major receptor for CCL2/monocyte chemotactic protein (MCP)-1, was present at high levels in infiltrating lymphocytes in an experimental autoimmune encephalomyelitis (EAE)/EAU rat model. ^{56 57 58} An earlier study has shown that CCL2/MCP-1 enhances the Th1 immune response in an EAE model. ⁵⁸ Keino et al. ²¹ also reported that upregulated CCR2 is involved in recruiting inflammatory cells, such as Th1-type T-cells, into the eyes of EAU mice. ²¹ Lymphotoxin A, a member of the TNF family, plays important roles in lymphocyte activation, immune regulation, and lymphoid tissue development. ^{59 60 61} It is secreted by lymphocytes and implicated in inflammatory diseases such as Behçet’s disease. ⁶² The level of lymphotoxin A in patients with Behçet’s disease is higher than that in healthy control subjects. ⁶² Considering those studies, cluster 3 contains genes that play important roles in lymphocyte behavior. Further investigation of this cluster may lead to a new therapeutic approach to the control of ocular inflammation. 

The expression pattern of cluster 4 was similar to that of cluster 3, but the extent of up- and downregulation was less than in cluster 3 (less steep downhill). The five genes in this cluster were significantly upregulated on day 1, but the upregulation was <2.5-fold. The representative genes in this cluster were CCR8 and CXCL11/interferon-inducible T-cell α-chemoattractant (ITAC). CCR8, expressed in natural killer cells, T lymphocytes, and monocytes, increases strongly in the presence of inflammation. ^{63 64} Previous results have indicated that CCR8 and CXCL11/I-TAC are potent regulators of endothelial cell function and act as angiogenic molecules such as fibroblast growth factor 2 and vascular endothelial growth factor. ⁶⁵ The formation of focal subretinal neovascularization is an ocular manifestation of EAU ⁶⁶ and it is well known that corticosteroids suppress angiogenesis. ^{67 68} Therefore, we can speculate that corticosteroids control the angiogenesis through CXCL11 and CCR8. 

Several cytokines and chemokines have been identified as upregulated genes by quantitative PCR in development of EAU. ¹⁷ Whereas the pathogenesis of EAU has been related to an elevation in the amount of Th1-type cytokines during the acute phase, ¹⁸ Th2-type cytokines also were reported to increase simultaneously in EAU. ¹⁷ Our comprehensive gene expression analysis in EAU development confirmed these previous studies. Our results also revealed that most genes of cytokines/chemokines and their receptors (91/117 genes) were significantly upregulated at the peak of inflammation. In the top 10 upregulated genes especially, we found eight cytokine/chemokine receptors, suggesting that cytokine/chemokine receptors may play a pivotal role in the pathogenicity of EAU. 

Studies have indicated that corticosteroids decrease accumulation of activated T-cells, neutrophils, and macrophages in inflamed tissue. ⁹ Investigators have also demonstrated that the chemokines and cytokines produced by both infiltrated cells and ocular resident cells play a role in development of EAU. ¹⁷ However, we could not show whether the expression changes after treatment with PDS represent changes in infiltration of lymphocytes, resident inflamed ocular cells, or both, and thus further studies are needed. In addition, as previously reported, the clinical score did not change significantly throughout the observation period. ²⁷ Therefore, we could not correlate the gene expression changes observed in this study with the improvements in the clinical scores in EAU mice. 

In summary, we performed hierarchical cluster analysis based on comprehensive gene expression profiles of the cytokines and chemokines and receptors after corticosteroid treatment in a murine EAU model. The genes clustered into four patterns. Although the implications of each cluster remain unclear, these findings may open a new avenue to identify new therapeutic targets in ocular inflammation. 

 Supported by Grant in Aid for Scientific Research C-15591852 from the Japanese Ministry of Education, Culture, Sports, Science, and Technology.

 Submitted for publication March 18, 2005; revised June 15, 2005; accepted September 13, 2005.

 Disclosure: N. Hashida, None; N. Ohguro, None; K. Nakai, None; M. Kobashi-Hashida, None; S. Hashimoto, None; K. Matsushima, None; Y. Tano, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Reliability of the microarray technique. (A) Scatterplot of Cy3 versus Cy5. The Cy3 and Cy5 intensities were log₂ transformed and are shown on the horizontal axis and vertical axis, respectively. The intensities of Cy3 and Cy5 displayed a strong linear relationship (R ² = 0.995). Red and blue lines: twofold and fourfold up- or downregulation, respectively. This result represents three independent experiments. (B) Comparison of cluster image of each experiment. Total RNA from EAU mice and normal mice was labeled with Cy3 and Cy5, respectively (lanes 1–3). In another experiment, total RNA from EAU mice and normal mice was labeled with Cy5 and Cy3, respectively (lanes 3–6). Cluster images of each experiment are simultaneously juxtaposed. The clustering pattern of the genes in each experiment shows the correlated pattern.

Table 1.

 

Changes in Gene Transcription in EAU Mice

Gene	GenBank	Change (x-fold)
Interleukin 13 receptor, alpha 2 (IL 13ra2)	NM_008356	14.49 ± 2.75
Interleukin 7 receptor (IL 7r)	M29697	14.30 ± 4.31
Chemokine (C-X-C motif) ligand 12 (CXCL 12)	D43804	8.24 ± 0.44
Transforming growth factor, beta receptor 1 (TGFbR1)	D25540	6.30 ± 0.15
Chemokine (C-C motif) receptor 5 (CCR5)	U47036	4.92 ± 1.61
Interleukin 2 receptor, alpha chain (IL2ra)	M30856	4.54 ± 0.12
Chemokine (C-X-C motif) ligand 15 (CXCL 15)	NM_011339	4.45 ± 0.47
Colony-stimulating factor 2 receptor, alpha, low-affinity (granulocyte-macrophage) (CSF2ra)	M85078	4.41 ± 0.24
Chemokine (C motif) XC receptor 1 (XCR1)	NM_011798	4.38 ± 0.50
Interleukin 11 receptor, alpha chain 1 (IL 11ra)	U14412	4.35 ± 0.63
Transforming growth factor, beta 2 (TGFb2)	X57413	4.30 ± 0.21
Chemokine (C-C motif) receptor 7 (CCR7)	NM_007719	4.27 ± 1.38
Interleukin 12 receptor, beta 2 (IL 12rb2)	U64199	4.22 ± 0.52
Chemokine (C-X-C motif) ligand 1 (CXCL 1)	J04596	4.13 ± 1.10
Chemokine (C-C motif) receptor 1-like 1 (CCR1l1)	NM_007718	4.04 ± 0.89
Chemokine (C-C motif) ligand 11 (CCL 11/eotaxin)	U40672	4.03 ± 0.45
Chemokine (C-X-C motif) receptor 4 (CXCR4)	X99582	4.00 ± 0.76
Chemokine (C-C motif) ligand 25 (CCL 25)	AJ249480	3.95 ± 0.31
Tumor necrosis factor (TNF)	X02611	3.85 ± 0.31
Interleukin 21 receptor (IL 21r)	AF279436	3.81 ± 0.21
Chemokine (C-C motif) receptor-like 2 (CCRl2)	AF030185	3.74 ± 0.36
Interleukin 10 receptor, beta (IL 10rb)	NM_008349	3.71 ± 0.19
Interleukin 15 receptor, alpha chain (IL 15ra)	NM_008358	3.67 ± 0.14
Interleukin 13 (IL 13)	NM_008355	3.64 ± 0.19
Interleukin 1 receptor, type II (IL 1r2)	X59769	3.60 ± 0.41
Interleukin 2 receptor, gamma chain (IL 2rg)	X75337	3.57 ± 0.15
Interleukin 1 receptor antagonist (IL 1rn)	M64404	3.45 ± 0.20
Interleukin 6 signal transducer (IL 6st)	M83336	3.43 ± 0.35
Interleukin 9 receptor (IL 9r)	M84746	3.31 ± 0.91
Transforming growth factor, beta receptor III (TGFbr 3)	AF039601	3.31 ± 0.25
Interleukin 5 receptor, alpha (IL 5ra)	D90205	3.27 ± 0.36
Colony-stimulating factor 3 receptor (granulocyte) (CSF3r)	M58288	3.26 ± 0.18
Lymphotoxin A (LTA)	NM_010735	3.20 ± 0.37
Colony-stimulating factor 2 receptor, beta 2, low-affinity (granulocyte-macrophage) (CSF2rb2)	M29855	3.17 ± 0.25
Chemokine (C-C motif) ligand 7 (CCL7)	Z12297	3.16 ± 1.03
Colony stimulating factor 2 (granulocyte-macrophage) (CSF2)	X03019	3.14 ± 0.13
Chemokine (C-X-C motif) ligand 5 (CXCL 5)	U27267	3.11 ± 0.23
Interleukin 1 receptor, type 1 (IL 1r1)	M20658	3.11 ± 0.26
Chemokine (C-C motif) ligand 17 (CCL 17)	AJ242587	3.10 ± 0.24
IL-13 receptor alpha chain (IL 13ra1)	S80963	3.10 ± 0.32
Chemokine (C-C motif) ligand 1 (CCL 1)	M17957	3.09 ± 0.10
Interleukin 18 receptor accessory protein (IL 18rap)	AF077347	3.05 ± 0.58
Chemokine (C-C motif) ligand 9 (CCL 9)	U15209	3.04 ± 0.17
Chemokine (C-C motif) ligand 12 (CCL 12)	U50712	3.02 ± 0.36
Chemokine (C-C motif) ligand 8 (CCL 8)	AB023418	3.00 ± 0.72
Interleukin 1 beta (IL 1b)	M15131	2.98 ± 0.23
Chemokine (C-X-C) receptor 2 (CXCR2/IL 8Rb)	D17630	2.97 ± 0.16
Chemokine (C-C) receptor 8 (CCR8)	NM_007720	2.94 ± 0.18
Interleukin 2 receptor, beta chain (IL 2rb)	NM_008368	2.93 ± 0.17
Chemokine (C-C) receptor 3 (CCR3)	NM_009914	2.90 ± 0.14
Chemokine (C-X-C) receptor 3 (CXCR3)	AF045146	2.90 ± 0.34
Interleukin 5 (IL 5)	NM_010558	2.82 ± 0.65
Interferon (alpha and beta) receptor (IFNar 1)	M89641	2.79 ± 0.15
Interleukin 1 alpha (IL 1a)	X01450	2.78 ± 0.58
Interleukin 4 receptor, alpha (IL 4ra)	M27959	2.77 ± 0.14
Interleukin 17 receptor (IL 17r)	U31993	2.73 ± 0.45
Chemokine (C-C motif) ligand 22 (CCL 22)	AJ238238	2.69 ± 0.24
Chemokine orphan receptor 1 (CMKor 1)	AF000236	2.66 ± 0.10
Chemokine (C-C motif) ligand 21n (leucine) (CCL2la)	U88322	2.65 ± 0.13
Chemokine (C-X3-C motif) ligand 1 (CX3CL 1)	NM_009142	2.65 ± 0.16
Colony-stimulating factor 2 receptor, beta 1, low-affinity (granulocyte-macrophage) (CSF2rb1)	M34397	2.64 ± 0.12
Chemokine (C-X-C motif) ligand 16 (CXCL 16)	AF277001	2.61 ± 0.28
Interleukin 3 receptor, alpha chain (IL3ra)	X64534	2.59 ± 0.17
Transforming growth factor, beta 1 (TGFb1)	AJ009862	2.58 ± 0.11
Interleukin 10 (IL 10)	NM_010548	2.54 ± 0.41
Chemokine (C motif) ligand 1 (XCL 1)	D43769	2.53 ± 0.19
Chemokine (C-C motif) receptor 6 (CCR6)	NM_009835	2.51 ± 0.21
Interleukin 6 receptor, alpha (IL6ra)	X51975	2.50 ± 0.37
Chemokine (C-C motif) ligand 24 (CCL24)	AF244367	2.50 ± 0.15
Interleukin 11 (IL11)	U03421	2.49 ± 0.17
Chemokine (C-C) receptor 2 (CCR2)	NM_009915	2.45 ± 0.82
Chemokine (C-X-C motif) ligand 11 (CXCL 11)	AF179872	2.45 ± 0.18
Interleukin 2 (IL2)	X01772	2.39 ± 0.12
Interferon gamma (IFNg)	K00083	2.35 ± 0.14
Chemokine (C-C) receptor 10 (CCR10)	NM_009913	2.34 ± 0.09
Transforming growth factor, beta 3 (TGFb3)	M32745	2.31 ± 0.12
Chemokine (C-C motif) ligand 20 (CCL20)	AB015136	2.28 ± 0.47
Chemokine (C-X3-C) receptor 1 (CX3CR1)	AF102269	2.26 ± 0.22
Chemokine (C-C motif) ligand 3 (CCL3)	NM_011337	2.22 ± 0.07
Chemokine (C-C) receptor 1 (CCR1)	NM_009912	2.21 ± 0.26
Chemokine (C-X-C motif) ligand 10 (CXCL 10)	M86829	2.20 ± 0.10
Interleukin 7 (IL 7)	NM_008371	2.17 ± 0.83
Interleukin 12 receptor, beta 1 (IL 12rb1)	U23922	2.15 ± 0.13
Chemokine (C-C motif) ligand 2 (CCL 2)	NM_011333	2.13 ± 0.41
Duffy blood group (Dfy)	AF016584	2.11 ± 0.12
Leukemia inhibitory factor (Lif)	NM_008501	2.10 ± 0.13
Chemokine (C-C motif) ligand 6 (CCL6)	NM_009139	2.08 ± 0.27
Interleukin 12a (IL 12a)	M86672	2.05 ± 0.09
Colony stimulating factor 3 (granulocyte) (CSF3)	NM_009971	2.03 ± 0.11
Tumor necrosis factor (ligand) superfamily, member 5 (TNFsf5)	NM_011616	2.02 ± 0.30
Chemokine (C-X-C motif) ligand 9 (CXCL 9)	NM_008599	2.00 ± 0.12
Chemokine (C-C) receptor 9 (CCR9)	NM_007721	1.96 ± 0.16
Chemokine (C-C motif) ligand 4 (CCL 4)	NM_013652	1.83 ± 0.25
Interleukin 18	NM_008360	1.83 ± 0.14
Interleukin 16	NM_010551	1.82 ± 0.15
Interleukin 18 binding protein (IL 18bp)	NM_010531	1.79 ± 0.15
Chemokine (C-C motif) ligand 19 (CCL 19)	AF059208	1.75 ± 0.11
Chemokine (C-C) receptor 4 (CCR4)	NM_009916	1.68 ± 0.20
Chemokine (C-X-C motif) ligand 4 (CXCL4)	AB017491	1.63 ± 0.55
Chemokine (C-X-C motif) ligand 14 (CXCL 14)	AF144754	1.56 ± 0.13
Interleukin 4 (IL4)	M25892	1.55 ± 0.22
Interferon beta, fibroblast (IFNb)	NM_010510	1.50 ± 0.50
Interleukin 10 receptor, alpha (IL 10ra)	L12120	1.46 ± 0.22
Chemokine (C-X-C motif) ligand 2 (CXCL 2)	X53798	1.39 ± 0.12
Interleukin 15 (IL 15)	U14332	1.38 ± 0.48
Interferon gamma receptor (IFNgr)	M28233	1.37 ± 0.16
Chemokine (C-C motif) ligand 5 (CCL5)	M77747	1.22 ± 0.15
Chemokine (C-X-C) receptor 5 (CXCR5)	NM_007551	0.83 ± 0.12
Chemokine (C-X-C) receptor 6 (CXCR6)	AF305709	NA
Interleukin 1 receptor accessory protein (IL 1rap)	X85999	NA
Interleukin 12b (IL 12b)	M86671	NA
Interleukin 17 (IL 17)	U43088	NA
Interleukin 9 (IL 9)	NM_008373	NA
Interleukin 6 (IL 6)	X54542	NA
Chemokine (C-C motif) ligand 27 (CCL27)	NM_011336	NA
Chemokine (C-X-C motif) ligand 13 (CXCL 13)	AF044196	NA
Transforming growth factor, beta receptor II (TGFbr2)	D32072	NA

 Change represents the average of three possible pair-wise ratios of EAU mice divided by normal control mice control ± SD. NA, Data unavailable in microarray.

Table 2.

 

Gene Expression Changes after Prednisolone Treatment

Gene	GenBank	Day 1	Day 2	Day 3
Cytokines and Cytokine Receptors
Interleukin 1 alpha (Il1a)	X01450	1.759	1.181	0.465
Interleukin 1 beta (Il1b)	M15131	0.344	0.746	0.418
Interleukin 2 (Il2)	X01772	0.812	0.479	0.678
Interleukin 4 (Il4)	M25892	0.851	0.608	0.569
Interleukin 5 (Il5)	NM_010558	0.474	1.041	0.498
Interleukin 6 (Il6)	X54542	0.741	0.551	0.676
Interleukin 7 (Il7)	NM_008371	0.257	0.438	0.605
Interleukin 9 (Il9)	NM_008373	0.359	0.755	0.830
Interleukin 10 (Il10)	NM_010548	1.624	0.846	0.667
Interleukin 11 (Il11)	U03421	0.303	0.624	0.567
Interleukin 12a (Il12a)	M86672	0.802	0.784	0.522
Interleukin 12b (Il12b)	M86671	0.656	0.517	0.579
Interleukin 13 (Il13)	NM_008355	0.956	0.704	0.371
Interleukin 15 (Il15)	U14332	0.382	0.604	0.741
Interleukin 16 (Il16)	NM_010551	1.759	1.226	0.646
Interleukin 17 (Il17)	U43088	0.397	0.425	0.818
Interleukin 18 (Il18)	NM_008360	2.105	0.541	0.647
Interleukin 1 receptor, type I (Il1r1)	M20658	0.583	0.554	0.504
Interleukin 1 receptor, type II (Il1r2)	X59769	1.278	0.743	0.508
Interleukin 1 receptor, accessory protein (Il1rap)	X85999	0.258	0.581	0.688
Interleukin 1 receptor antagonist (Il1rn)	M64404	0.604	0.805	0.374
Interleukin 2 receptor, alpha chain (Il2ra)	M30856	0.419	0.949	0.376
Interleukin 2 receptor, beta chain (Il2rb)	NM_008368	0.369	0.835	0.489
Interleukin 2 receptor, gamma chain (Il2rg)	X75337	1.519	0.647	0.421
Interleukin 3 receptor, alpha chain (Il3ra)	X64534	1.295	0.947	0.539
Interleukin 4 receptor, alpha (Il4ra)	M27959	2.190	1.079	0.483
Interleukin 5 receptor, alpha (Il5ra)	D90205	0.395	0.806	0.467
Interleukin 6 receptor, alpha (Il6ra)	X51975	0.288	0.824	0.576
Interleukin 6 signal transducer (Il6st)	M83336	0.289	0.692	0.604
Interleukin 7 receptor, (Il7r)	M29697	3.180	1.415	0.508
Interleukin 9 receptor (Il9r)	M84746	2.003	1.422	0.552
Interleukin 10 receptor, alpha (Il10ra)	L12120	0.330	0.851	0.477
Interleukin 10 receptor, beta (Il10rb)	NM_008349	1.725	0.602	0.403
Interleukin 11 receptor, alpha chain 1 (Il11ra1)	U14412	0.413	0.427	0.575
Interleukin 12 receptor, beta 1 (Il12rb1)	U23922	0.236	0.740	0.507
Interleukin 12 receptor, beta 2 (Il12rb2)	U64199	0.742	0.872	0.463
Interleukin 13 receptor, alpha 1 (Il13ra1)	S80963	0.300	0.568	0.531
Interleukin 13 receptor, alpha 2 (Il13ra2)	NM_008356	0.791	0.869	0.434
Interleukin 15 receptor, alpha chain (Il15ra)	NM_008358	0.522	0.748	0.382
Interleukin 17 receptor (Il17r)	U31993	0.579	0.634	0.472
Interleukin 18 receptor accessory protein (Il18rap)	AF077347	0.212	0.428	0.549
Interleukin 21 receptor (Il21r)	AF279436	1.375	0.757	0.369
Transforming growth factor, beta 1 (Tgfb1)	AJ009862	0.250	0.618	0.465
Transforming growth factor, beta 2 (Tgfb2)	X57413	0.354	0.558	0.420
Transforming growth factor, beta 3 (Tgfb3)	M32745	0.306	0.589	0.553
Transforming growth factor, beta receptor I (Tgfbr1)	D25540	0.451	1.250	0.386
Transforming growth factor, beta receptor II (Tgfbr2)	D32072	0.911	1.157	0.597
Transforming growth factor, beta receptor III (Tgfbr3)	AF039601	0.204	0.538	0.500
Colony stimulating factor 2 (granulocyto-macrophage) (Csf2)	X03019	0.337	0.745	0.352
Colony-stimulating factor 3 (granulocyte) (Csf3)	NM_009971	0.591	0.551	0.579
Colony-stimulating factor 2 receptor, alpha, low-affinity (granulocyte-macrophage) (Caf2ra)	M85078	0.310	0.883	0.361
Colony stimulating factor 2 receptor, beta 1, low-affinity (granulocyte-macrophage) (Caf2rb1)	M34397	0.430	0.846	0.465
Colony-stimulating factor 2 receptor, beta 2, low-affinity (granulocyte-macrophage) (Caf2rb2)	M29855	1.082	0.942	0.415
Colony-stimulating factor 3 receptor (granulocyte) (Csf3r)	M58288	0.339	0.714	0.403
Interferon (alpha and beta) receptor 1 (Ifnar1)	M89641	0.322	0.674	0.504
Interferon beta 1, fibroblast (Ifnb1)	NM_010510	0.451	0.880	0.490
Interferon gamma (Ifng)	K00083	0.368	0.535	0.622
Interferon gamma receptor (Ifngr)	M28233	0.633	0.619	0.522
Interferon gamma inducing factor binding protein (lgilbp)	NM_010531	0.426	0.794	0.469
Leukemia inhibitory factor (Lif)	NM_008501	0.876	0.463	0.636
Lymphotoxin A (Lta)	NM_010735	2.899	0.742	0.537
Tumor necrosis factor (Tnf)	X02611	0.332	0.794	0.403
Tumor necrosis factor (ligand) superfamily, member 5 (Tnfsf5) (CD40 ligand)	NM_011616	0.429	0.711	0.544

Chemokines and Chemokine Receptors
Chemokine (C motif) ligand 1 (Xcl1)	D43769	2.568	1.030	0.482
Chemokine (C-C motif) ligand 1 (Ccl1)	M17957	0.379	0.661	0.447
Chemokine (C-C motif) ligand 2 (Ccl2)	NM_011333	1.155	0.806	0.559
Chemokine (C-C motif) ligand 3 (Ccl3)	NM_011337	2.079	1.383	0.615
Chemokine (C-C motif) ligand 4 (Ccl4)	NM_013652	0.909	0.704	0.526
Chemokine (C-C motif) ligand 5 (Ccl5)	M77747	0.368	0.750	0.327
Chemokine (C-C motif) ligand 6 (Ccl6)	NM_009139	0.354	0.803	0.582
Chemokine (C-C motif) ligand 7 (Ccl7)	Z12297	0.357	0.631	0.531
Chemokine (C-C motif) ligand 8 (Ccl8)	AB023418	1.442	0.966	0.531
Chemokine (C-C motif) ligand 9 (Ccl9)	U15209	0.657	0.533	0.459
Chemokine (C-C motif) ligand 11 (Ccl11)	U40672	0.645	0.440	0.503
Chemokine (C-C motif) ligand 12 (Ccl12)	U50712	1.095	0.776	0.579
Chemokine (C-C motif) ligand 17 (Ccl17)	AJ242587	0.915	0.785	0.518
Chemokine (C-C motif) ligand 19 (Ccl19)	AF059208	0.633	0.650	0.664
Chemokine (C-C motif) ligand 20 (Ccl20)	AB015136	0.980	0.843	0.652
Chemokine (C-C motif) ligand 21 (Ccl21)	U88322	1.144	0.521	0.548
Chemokine (C-C motif) ligand 22 (Ccl22)	AJ238238	0.402	0.801	0.553
Chemokine (C-C motif) ligand 24 (Ccl24)	AF244367	1.191	0.626	0.601
Chemokine (C-C motif) ligand 25 (Ccl25)	AJ249480	0.411	0.666	0.535
Chemokine (C-C motif) ligand 27 (Ccl27)	NM_011336	0.578	0.821	0.772
Chemokine (C-X-C motif) ligand 1 (Cxcl1)	J04596	1.812	1.294	0.397
Chemokine (C-X-C motif) ligand 2 (Cxcl2)	X53798	0.447	0.557	0.703
Chemokine (C-X-C motif) ligand 4 (Cxcl4)	AB017491	0.501	0.527	0.602
Chemokine (C-X-C motif) ligand 5 (Cxcl5)	U27267	0.340	0.652	0.476
Chemokine (C-X-C motif) ligand 9 (Cxcl9)	NM_008599	0.576	0.567	0.323
Chemokine (C-X-C motif) ligand 10 (Cxcl10)	M86829	0.828	0.501	0.285
Chemokine (C-X-C motif) ligand 11 (Cxcl11)	AF179872	2.059	0.565	0.614
Chemokine (C-X-C motif) ligand 12 (Cxcl12)	D43804	0.438	0.666	0.362
Chemokine (C-X-C motif) ligand 13 (Cxcl13)	AF044196	0.569	1.058	0.699
Chemokine (C-X-C motif) ligand 14 (Cxcl14)	AF144754	0.443	0.682	0.784
Chemokine (C-X-C motif) ligand 15 (Cxcl15)	NM_011339	0.433	0.519	0.400
Chemokine (C-X-C motif) ligand 16 (Cxcl16)	AF277001	0.650	0.599	0.465
Chemokine (C-X3-C motif) ligand 1 (Cx3cl1)	NM_009142	0.878	0.491	0.528
Chemokine (C-C motif) receptor 1 (Ccr1)	NM_009912	0.478	0.641	0.616
Chemokine (C-C motif) receptor 2 (Ccr2)	NM_009915	2.546	0.813	0.556
Chemokine (C-C motif) receptor 3 (Ccr3)	NM_009914	0.373	0.616	0.752
Chemokine (C-C motif) receptor 4 (Ccr4)	NM_009916	1.017	0.716	0.560
Chemokine (C-C motif) receptor 5 (Ccr5)	U47036	0.338	0.847	0.470
Chemokine (C-C motif) receptor 6 (Ccr6)	NM_009835	0.665	0.570	0.675
Chemokine (C-C motif) receptor 7 (Ccr7)	NM_007719	0.776	0.568	0.765
Chemokine (C-C motif) receptor 8 (Ccr8)	NM_007720	2.273	0.548	0.555
Chemokine (C-C motif) receptor 9 (Ccr9)	NM_007721	1.773	0.595	0.565
Chemokine (C-C motif) receptor 10 (Ccr10)	NM_009913	1.132	0.367	0.838
Chemokine (C-C motif) receptor-like 1 (Ccrl1)	NM_007718	0.267	0.440	0.528
Chemokine (C-C motif) receptor-like 2 (Ccrl2)	AF030185	0.567	0.523	0.474
Chemokine (C motif) receptor 1 (Xcr 1)	NM_011798	0.390	0.833	0.516
Interleukin 8 receptor, beta (Il8rb) (Cxcr2)	D17630	1.206	0.809	0.428
Chemokine (C-X-C motif) receptor 3 (Cxcr3)	AF045146	1.064	0.886	0.529
Chemokine (C-X-C motif) receptor 4 (Cxcr4)	X99582	1.176	0.834	0.532
Chemokine (C-X-C motif) receptor 5 (Cxcr5)	NM_007551	0.944	0.922	0.726
Chemokine (C-X-C motif) receptor 6 (Cxcr6)	AF305709	2.701	1.242	0.750
Chemokine (C-X3-C) receptor 1 (Cx3cr1)	AF102269	0.327	0.756	0.563
Duffy blood group (Dfy)	AF016584	0.236	0.717	0.656
Chemokine orphan receptor 1 (Cmkor1)	AF000236	1.549	1.008	0.526
Number of suppressed genes (> 2-fold)		47	10	46

 Data are x-fold changes and represent the average of four possible pair-wise ratios of PDS-treated EAU mice divided by placebo-treated control EAU mice.

Table 3.

 

Top 10 Up- and Downregulated Genes

Gene Description	Change (x-fold)
Top 10 upregulated genes in 1 Day
Interleukin 7 receptor (Il7r)	3.18
Lymphotoxin A (Lta)	2.89
Chemokine (C-X-C motif) receptor 6 (Cxcr6)	2.70
Chemokine (C motif) ligand 1 (Xcl1)	2.56
Chemokine (C-C motif) receptor 2 (Ccr2)	2.54
Chemokine (C-C motif) receptor 8 (Ccr8)	2.27
Interleukin 4 receptor, alpha (Il4ra)	2.19
Interleukin 18 (Il18)	2.10
Chemokine (C-C motif) ligand 3 (Ccl3)	2.07
Chemokine (C-X-C motif) ligand 11 (Cxcl11)	2.05
Top 10 downregulated genes in each day
Day 1
Transforming growth factor, beta receptor III (Tgfbr2)	0.20
Interleukin 18 receptor accessory protein (Il18rap)	0.21
Duffy blood group (Dfy)	0.23
Interleukin 12 receptor, beta 1 (Il12rb1)	0.23
Transforming growth factor, beta 1 (Tgfb1)	0.24
Interleukin 7 (Il7)	0.25
Interleukin 1 receptor accessory protein (Il1rap)	0.26
Chemokine (C-C motif) receptor-like 1 (Ccrl1)	0.27
Interleukin 6 receptor, alpha (Il6ra)	0.28
Interleukin 6 signal transducer (Il6st)	0.28
Day 2
Chemokine (C-C motif) receptor 10 (Ccr10)	0.36
Interleukin 17 (Il17)	0.42
Interleukin 11 receptor, alpha chain 1 (Il11ra1)	0.42
Interleukin 18 receptor accessory protein (Il18rap)	0.42
Interleukin 7 (Il7)	0.43
Chemokine (C-C motif) receptor-like 1 (Ccrl1)	0.44
Chemokine (C-C motif) ligand 11 (Ccl11)	0.44
Leukemia inhibitory factor (Lif)	0.46
Interleukin 2 (Il2)	0.47
Chemokine (C-X3-C) receptor 1 (Cx3cr1)	0.49
Day 3
Chemokine (C-X-C motif) ligand 10 (Cxcl10)	0.28
Chemokine (C-X-C motif) ligand 9 (Cxcl9)	0.32
Chemokine (C-C motif) ligand 5 (Ccl5)	0.32
Colony-stimulating factor 2 (granulocyte-macrophage) (Csf2)	0.35
Colony-stimulating factor 2 receptor, alpha, low-affinity (granulocyte-macrophage) (Csf2ra)	0.36
Chemokine (C-X-C motif) ligand 12 (Cxcl12)	0.36
Interleukin 21 receptor (Il21r)	0.36
Interleukin 13 (Il13)	0.37
Interleukin 1 receptor antagonist (Il1rn)	0.37
Chemokine (C-X3-C motif) ligand 1 (Cx3cl1)	0.38

Figure 2.

 

(A) The graphs of the Silhouette index and Dunn index display the valid clusters. The highest score for each index helps to determine the numbers of clusters. (Silhouette index, −0.12; Dunn index, 0.51). (B) By microarray analysis, cluster images show the different classes of gene expression profiles of 117 genes and the expression changes after PDS treatment. The subsets of genes are hierarchically clustered. The expression pattern of each gene in this set is displayed as a horizontal strip. For each gene, the ratio of the RNA levels in the eyes from the EAU mice at the indicated days after PDS treatment to the levels in the eyes of saline-treated mice is represented by a color, according to the scale at the bottom. The graphs show the representative expression profiles of the genes in the corresponding clusters 1 through 4.

Figure 3.

 

The expression patterns and detailed functional categories of inflammatory genes in each cluster. The subset of 117 genes encoding cytokines and chemokines and their receptors is clustered hierarchically into groups based on the similarity of the expression profiles. For each gene, the ratio of the RNA levels in the eyes of EAU mice at the indicated days after PDS treatment to its levels in the eyes of saline-treated mice is represented by a color, according to the scale at the bottom.

Figure 4.

 

Comparison of expression changes of microarray analysis ( ) and quantitative RT-PCR (□) experiments. The same RNA sources were used for both experiments. (A) IL21r, TNFa, and CXCL1 are the representative genes with upregulated patterns in the eyes of EAU mice. For microarray analysis, samples were analyzed in triplicate. In the RT-PCR experiments, x-fold change was determined by averaging the changes obtained for the four replicates. Comparisons were performed by t-test. The change was considered significant at P < 0.05 (*P < 0.01) (B) A representative gene in each cluster was selected and the sequential expression changes analyzed after PDS treatment: CXCL4 (cluster 1), CXCL5 (cluster 2), CXCR6 (cluster 3), and IL16 (cluster 4). Comparisons between quantitative RT-PCR data and microarray data were performed by t-test. Change was considered significant at P < 0.05 (*P < 0.01 by t-test).