Arthritis severity locus Cia4 is an early regulator of IL-6, IL-1β, and NF-κB activators' expression in pristane-induced arthritis

Abstract

Cia4 is a locus on rat chromosome 7 that regulates disease severity and joint damage in models of rheumatoid arthritis, including pristane-induced arthritis (PIA). To identify molecular processes regulated by Cia4, synovial tissues from MHC-identical DA (severe erosive) and DA.F344(Cia4) congenics (mild nonerosive) rats were collected at preclinical and recent onset stages following the induction of PIA and analyzed for gene expression levels. Il6 levels were significantly higher in DA compared with congenics on day 10 (135-fold) after PIA induction (preclinical stage) and remained increased on days 14 (47.7-fold) and 18 (29.41-fold). Il6 increased before Il1b suggesting that Il6 could be driving Il1b expression and early synovial inflammation; 187 genes had significantly different expression levels and included inflammatory mediators increased in DA such Slpi (10.94-fold), Ccl7 (5.17-fold), and Litaf (2.09-fold). Syk or NF-κB activating and interacting genes, including Cd74 Ccl21, were increased in DA; 59 genes implicated in cancer-related phenotypes were increased in DA. Genes involved in cell metabolism, transport across membranes, and tissue protection such as Dgat1, Dhcr7, and Slc1a1 were increased in DA.F344(Cia4) congenics; 21 genes differentially expressed or expressed in only one of the strains were located within the Cia4 interval and could be the gene accounting for the arthritis effect. In conclusion, the Cia4 interval contains at least one new arthritis gene that regulates early Il6, Il1b expression, and other inflammatory mediators. This gene regulates the expression of cancer genes that could mediate the development of synovial hyperplasia and invasion, and cartilage and bone destruction.

rheumatoid arthritis (ra) is a common and potentially debilitating form of erosive arthritis. Recent advances in the understanding of RA pathogenesis have led to the development of new and better treatments (20, 32, 44). Yet disease remission is rarely achieved (45), and more effective therapies are needed. The identification of genes implicated in the regulation of arthritis severity and articular damage has the potential to generate new and better targets for therapies aimed at preserving joint architecture and function and reducing the risk of developing joint deformities. However, little is known about those genes (27), and the large cohorts of RA patients used in genome-wide association studies for susceptibility were not designed to address disease severity and articular damage.

We have previously identified several disease severity and articular damage quantitative trait loci (QTL) in rat models of RA (5, 13, 17, 18, 36). Using a combination of positional cloning, functional studies, and transcriptome analyses of synovial cells and synovial tissues, we are beginning to understand the molecular processes regulating arthritis severity and joint damage in pristane- and collagen-induced arthritis (PIA and CIA) (3, 5, 21–23).

Cia4 is a 45.5 Mb QTL on rat chromosome 7 that regulates arthritis severity, cartilage and bone damage, synovial hyperplasia, and inflammation in both PIA and CIA (24, 36, 37). To gain new insight into pathogenic and protective cellular pathways regulated by the arthritis gene located within the Cia4 interval we used synovial tissues from MHC-identical DA.F344(Cia4) congenics (arthritis-protected) and from DA (arthritis-susceptible) rats in a microarray analysis of gene expression. Tissues were obtained at very early stages during the development of PIA to identify early effects of the Cia4 locus on gene expression. We identified a new role for the Cia4 locus in the expression of several genes central to RA pathogenesis and joint damage, such as cytokines Il6 and Il1b, chemokines, protease inhibitors, genes implicated in the activation of the Syk and NFκB pathways, and cancer-related genes.

MATERIALS AND METHODS

Animals

DA/BklArbNsi rats (DA) were originally purchased from Bantin & Kingman, maintained at Arthritis and Rheumatism Branch, National Institute of Arthritis and Musculoskeletal and Skin Disease, National Institutes of Health, and then transferred to the Feinstein Institute for Medical Research (FIMR, formerly North Shore-LIJ Research Institute). DA.F344(Cia4) congenic rats (Fig. 1) were generated based on genotype-guided breeding as previously described (36). Briefly, alleles from arthritis-resistant F344 at the Cia4 interval on chromosome 7 were introgressed into arthritis-susceptible DA rats through eight genotype-guided backcrosses, while excluding F344 alleles at other arthritis-regulatory loci such as Cia1 (MHC on chromosome 20), Cia3 (chromosome 4), and Cia5 (chromosome 10). Rats heterozygous only at Cia4 were then intercrossed to generate homozygous. Rats homozygous at the congenic interval were brother-sister mated for at least five generations before conducting arthritis and microarray experiments. We used 8- to 12-wk-old rats in the arthritis and microarray experiments. All experiments involving animals were reviewed and approved by the FIMR Institutional Animal Care and Use Committee.

Fig. 1.

Map of the Cia4 congenic interval on rat chromosome 7. Left column shows simple sequence length polymorphism (SSLP) markers and their position in the physical map of rat chromosome 7. Black bar (45.5 Mb) identifies the region homozygous for F344 alleles.

PIA

Rats were anesthetized and injected intradermally (id) with 150 μl of pristane (2,6,10,14-tetramethylpentadecane) (MP Bio, Solon, OH) divided into two injection sites at the base of the tail (day 0) (4, 42).

CIA

Purified bovine type II (BII) collagen (Chondrex, Redmond, WA) was dissolved in 0.1 N acetic acid (2 mg/ml) at 4°C and emulsified with incomplete Freund's adjuvant (IFA; Difco, Detroit, MI) to a final concentration of 1 mg/ml (12). Anesthetized rats were injected with 2 mg/kg BII-IFA id at the base of the tail (day 0) and received a booster injection of 100 μg 7 days later.

Arthritis Scoring

Arthritis severity was assessed with the arthritis severity index (ASI), an 80-point scoring system that measures disease severity over the course of 30 days postinduction of arthritis and correlates with the degree of synovial inflammation and with both cartilage and bone damage (4, 5).

For the histology, microarray, and quantitative PCR (qPCR) experiments, female DA and DA.F344(Cia4) rats were scored daily from day 12 postinduction of PIA. Paws and synovial tissues were collected 24 h after the onset of moderate ankle arthritis (at least one ankle with score ≥2), or on day 18 for rats without moderate ankle arthritis. For additional qPCR experiments, synovial tissue was also collected 7, 10, and 14 days postinduction of PIA.

Histology

At the end of the arthritis observation period, rats were euthanized, and the hind paw with the most swollen ankle was fixed in 10% formaldehyde. Paws were then decalcified in hydrochloric acid and 0.1 M EDTA (Cal-Ex, Fisher Scientific), and ankles embedded in paraffin. Slides were prepared and stained with hematoxylin and eosin and scored without knowledge of strain identity. We used a previously described comprehensive histological scoring system (4–5) where tibio-talar, talar-calcaneal, and midfoot joints were scored for the following parameters:

Synovial inflammation.

Five high-power magnification fields (HMF) were scored for the percentage of infiltrating mononuclear cells as follows: 0, absent; 1, mild (1–10%); 2, moderate (11–50%); 3, severe (51–100%). The mean of the five HMF was used for analyses.

Synovial hyperplasia.

0, absent; 1, mild (5–10 layers); 2, moderate (11–20 layers); 3, severe (>20 layers).

Extension of articular surface covered by pannus.

0, absent; 1, incipient; 2, partial; 3, extensive.

Synovial fibrosis.

0, absent; 1, mild (1–10%); 2, moderate (11–50%); 3, severe (51–100%).

Synovial vascularity (angiogenesis).

The mean number of vessels of five HMF.

Cartilage erosion.

Percentage of the cartilage surface that was eroded: 0, absent; 1, mild (1–10%); 2, moderate (11–50%); 3, severe (51–100%).

Bone erosion.

0, none; 1, erosion(s) observed only at HMF; 2, erosion(s) observed at low magnification; 3, severe transcortical erosion(s).

RNA Preparation and Microarray Experiments

Synovial tissue total RNA was extracted with the RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's instructions including a DNase treatment step. RNA was quantified and assessed for purity using the NanoDrop spectrophotometer (Rockland, DE). RNA integrity was verified with the BioAnalyzer 2100 (Agilent, Palo Alto, CA).

Microarray.

All reagents and procedures were previously optimized for use with the Illumina Whole-Genome Expression platform. Total RNA (200 ng) was amplified and biotinylated using the TotalPrep labeling kit (Ambion, Austin, TX). Samples were hybridized to the RatRef-12 Expression BeadChip (Illumina, San Diego, CA), which contains 22,522 probes covering 21,922 rat genes selected primarily from the NCBI RefSeq database (release 16). Hybridization was carried out in Illumina IntelliHyb chambers, followed by washing and staining with Cy3-streptavidin. The BeadChip was scanned on a high-resolution Illumina BeadArray reader with a two-channel 0.8 μm resolution confocal laser scanner.

Microarray analysis.

Fluorescence intensities were extracted using BeadStudio 2.0 (Illumina). To correct for systematic biases of nonbiological origin, background-subtracted intensities of all probes were normalized using the cubic spline algorithm followed by log₂-transformation. Probes consistently expressed in all arrays were used for analyses. Genes with a ≥1.5-fold difference in intensity between DA and DA.F344(Cia4) and a t-test P value ≤ 0.01 were considered significantly differentially expressed and selected for pathway detection analyses using IPA 5.5.1 (Ingenuity Systems, Redwood City, CA), public online databases (GeneCards, Oncomine, BioGps, Ensembl), and literature (PubMed) search. Probes with a detection difference between strains of 60% or more were considered preferentially expressed and were also included in the analyses.

qPCR Expression Studies

Differences in the expression of selected genes were validated with qPCR. The qPCR conditions used have been described elsewhere (21). Briefly, total RNA (200 ng) from each sample was reverse transcribed using Superscript III (Invitrogen). Primers and qPCR probes were designed to target the same exons as the corresponding Illumina RatRef-12 Expression BeadChip probes (Supplemental Table S1).¹ We used Universal ProbeLibrary (Roche, Indianapolis, IN) and Taqman (ABI, Applied Biosystems, Foster City, CA) probes labeled with FAM at the 5′-end and TAMRA at 3′-end. Reactions were prepared in duplicates with Eurogentec qPCR MasterMix (San Diego, CA), run on an ABI Prism 7700 thermocycler using SDS software version 1.9.1 (ABI). Ct (threshold cycle) values were adjusted for GAPDH in each sample (ΔCt). Fold differences were calculated with the 2^−ΔΔCt method (25). The same RNA samples used in the microarray experiments were used for qPCR validation.

Statistics

Differences in arthritis severity scores were calculated by the Mann-Whitney test. Differences in microarray log₂-transformed intensities, Ct values, and histology scores between DA and DA.F344(Cia4) were compared by the two-tailed t-test for two samples of unequal variance with a nominal cutoff P value of ≤0.01. Enrichment for differentially expressed genes located within the Cia4 interval was calculated with the χ² test.

RESULTS

Male and Female DA.F344(Cia4) Congenics Were Protected and Developed Significantly Milder Forms of CIA and PIA

DA.F344(Cia4) rats (Fig. 1) had a significantly milder form of PIA compared with DA rats, with a reduction in the median ASI of 80.95% in males (P = 0.006) and 93.10% in females (P ≤ 0.001, Fig. 2, A–C). DA.F344(Cia4) rats also had a significantly milder form of CIA, with a reduction in the median ASI of 60% in males (P = 0.007) and 87% in females (P = 0.003) compared with DA (Fig. 2, D–F). These results demonstrate the impact of the Cia4 locus on the regulation of disease severity in both sexes and in two different models of autoimmune erosive arthritis. The protective effect of F344 alleles at Cia4 was detected as early as day 14 and persisted throughout the 31-day observation period.

Fig. 2.

DA.F344(Cia4) congenics develop a significantly milder form of disease in pristane- (PIA) and collagen-induced arthritis (CIA). Male and female DA.F344(Cia4) congenic rats (○) had significantly lower arthritis severity scores both in PIA (A and B) and CIA (D and E) compared with DA rats (●). E: median arthritis severity index (ASI), which provides information of cumulative arthritis severity over time, was also significantly reduced in congenic males and females, both in PIA (C) and CIA (F). *P ≤ 0.01, Mann-Whitney test for difference between arthritis severity scores at each time point and between ASI.

Histologic analyses of joints collected at day 18 postinduction of PIA, shortly after the onset of arthritis, revealed mild overall changes in DA. Inflammatory cell infiltration was reduced in synovial tissues from DA.F344(Cia4) congenic rats (mean score ± SE = 0 ± 0) compared with DA (mean score = 0.52 ± 0.29; P = 0.1). No other histologic differences were detected likely due to the very early stage in disease (Fig. 3).

Fig. 3.

Histology section of joints from DA and DA.F344(Cia4) congenics. A: histologic section of a midfoot collected 18 days after the induction of pristane-induced arthritis shows pronounced synovial inflammatory infiltration in DA (×10 magnification), but not in DA.F344(Cia4) congenics (×20 magnification; hematoxylin and eosin staining).

Decreased Expression of Proinflammatory Cytokines in DA.F344(Cia4) Congenic Rats

We aimed to detect very early changes in synovial tissue gene expression regulated by the Cia4 locus. Tissues were collected before disease onset (day 10 and day 14 postpristane injection) and shortly after disease onset (day 18) and checked for the expression of Il6, Il1b, and Tnfa, three central mediators of disease pathogenesis in rodent models of arthritis and in RA, by qPCR (Fig. 4). We also examined the levels of these cytokines in synovial tissues from naïve DA rats (Supplemental Table S5, DA day 0), and determined that Il6 (32.67-fold) and Il1b (2.36-fold) expression were significantly increased even when compared with synovial tissues obtained from DA.F344(Cia4) congenics with PIA collected at day 10. These results demonstrate that a gene located within the Cia4 interval regulates the baseline levels of these cytokines even in naïve rats. At day 10, Il6 levels were 135-fold higher in synovial tissues of DA compared with DA.F344(Cia4) congenic rats (P = 0.008), and levels remained significantly increased in DA at day 14 (47.7-fold, P ≤ 0.001) and at day 18 (29.4-fold, P = 0.011).

Fig. 4.

Cia4 regulates the expression of key proinflammatory cytokines. Expression of IL-6, IL-1β, and TNF-α was measured by quantitative (q)PCR using cDNA from synovial tissue collected 10 (DA n = 8, Cia4 n = 6), 14 (DA n = 9, Cia4 n = 7), or 18 days (DA n = 5, Cia4 n = 8) postinduction of PIA. *P ≤ 0.05, #P ≤ 0.008 (t-test).

Synovial expression of Il1b was also significantly higher in DA rats at days 14 (4.21-fold, P ≤ 0.001) and 18 (2.97-fold, P = 0.045), but not at day 10 (1.37-fold, P = 0.66). The observation that Il6 levels increased before those of Il1b suggests that Il6 may be driving the Il1b expression and could be more directly regulated by the arthritis gene located within the Cia4 interval.

Synovial expression of Tnfa was initially reduced in DA by 0.45-fold (P = 0.043) at day 10 compared with DA.F344(Cia4) congenics, and increased modestly as disease progressed to 1.54-fold at day 18 (not significant with P > 0.05). The significance of this mild early increase of Tnfa levels in the protected strain is of unclear relevance.

Synovial Tissues From DA.F344(Cia4) Congenics With Recent-onset Arthritis Have a Distinct Gene Expression Signature

To evaluate the effect of Cia4 on the synovial transcriptome of recent-onset arthritis, we collected synovial tissues from six DA and six DA.F344(Cia4) female rats with PIA within 24 h of onset of definitive ankle arthritis (score ≥ 2 on a scale of 0–4). Only one DA.F344(Cia4) had ankle arthritis score of 2 by day 18, which is 4 days after the typical day of onset of PIA, and all six congenics were euthanized for tissue collection and RNA extraction at that point. All six DA and six DA.F344(Cia4) congenics were used in the microarray and qPCR validation experiments and analyses.

We found 6,928 genes to be expressed in synovial tissues of all rats. A total of 187 genes met the criteria for significance and were differentially expressed between DA and DA.F344(Cia4) synovial samples. Of the 125 genes with increased expression in DA, 33 are involved in inflammatory and immune responses; 62 genes are increased in DA.F344(Cia4) (Tables 1–3, Supplemental Table S2).

Table 1.

Genes with the most significant differences in expression between DA.F344(Cia4) and DA

Symbol	Name	RefSeq ID	Fold Difference	P Value*
Increased in DA [decreased in DA.F344(Cia4) congenics]
Slpi	secretory leukocyte peptidase inhibitor	NM_053372.1	10.94	0.003
Cd163	CD163 molecule	XM_001061914.1	5.85	0.001
Ccl7	chemokine (C-C motif) ligand 7	NM_001007612.1	5.17	0.004
Lst1	leukocyte specific transcript 1	NM_022634.1	4.97	0.008
Ifitm1	interferon induced transmembrane protein 1	XM_215117.4	3.94	0.001
Mt1a	metallothionein 1a	NM_138826.4	3.69	0.005
Spon1	spondin 1, extracellular matrix protein	NM_172067	3.67	0.003
Batf	basic leucine zipper transcription factor, ATF-like	XM_216745.3	3.62	0.004
Clec4a3	C-type lectin domain family 4, member a3	NM_001005891.1	3.58	0.005
Bcl2a1d	B-cell leukemia/lymphoma 2 related protein A1d	NM_133416.1	3.51	0.005
Increased in DA.F344(Cia4) congenics
Rpl30	ribosomal protein L30	NM_022699.2	92.76	0.00000005
Dhcr7	7-dehydrocholesterol reductase	NM_022389.2	3.01	0.005
Sgca	sarcoglycan, alpha (dystrophin-associated glycoprotein)	XM_001081302.1	2.90	0.005
Rrp7a	ribosomal RNA processing 7 homolog A (S. cerevisiae)	XM_001077728.1	2.47	0.00003
Satb1	SATB homeobox 1	NM_001012129.1	2.18	0.01
LOC679430	hypothetical protein LOC679430	XM_001056304.1	2.17	0.003
Slc25a29	solute carrier family 25, member 29	NM_001010958.1	2.15	0.008
Cirbp	cold inducible RNA binding protein	NM_031147.2	2.14	0.003
Mif4 gd	MIF4G domain containing	NM_001014122.1	2.09	0.002
Giyd2	GIY-YIG domain containing 2	NM_001009292.1	2.04	0.002
Dck2	Doublecortin-like kinase 2 (Dclk2), mRNA.	NM_001009691.2	1.99	0.0004

^*t-test.

Table 3.

Inflammatory and immune responses genes expressed in increased levels in DA and reduced levels in DA.F344(Cia4) congenics

Gene ID	Gene Name	RefSeq ID	Fold Difference	P Value*
Ada	adenosine deaminase	NM_130399.2	1.81	0.003
Ador2a	adenosine A2a receptor	NM_053294.3	1.52	0.009
Anpep	alanyl (membrane) aminopeptidase	NM_031012.1	1.55	0.008
Aprt	adenine phosphoribosyl transferase	NM_001013061.1	1.55	0.001
C1qa	complement component 1, q subcomponent, A chain	NM_001008515.1	1.69	0.002
Ccl21	chemokine (C-C motif) ligand 21	NM_001008513.1	1.78	0.004
Ccl7	chemokine (C-C motif) ligand 7	NM_001007612.1	5.17	0.004
Ccr1	chemokine (C-C motif) receptor 1	NM_020542.2	1.69	0.005
Cd163	CD163 molecule	XM_001061914.1	5.85	0.001
Cd74	Cd74 molecule, major histocompatibility complex, class II invariant chain	NM_013069.1	2.60	0.004
Cfp	complement factor properdin	XM_001056015.1	2.36	0.002
Coro1a	coronin, actin binding protein 1A	NM_130411.2	2.17	0.006
Cysltr1	cysteinyl leukotriene receptor 1	NM_053641.1	1.84	0.01
Fcer1 g	Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide	XM_001053627.1	2.09	0.004
Fgr	Gardner-Rasheed feline sarcoma viral (v-fgr) oncogene homolog	NM_024145.2	2.31	0.006
Hck	hemopoietic cell kinase	NM_013185.2	2.44	0.005
Hsp90aa1	heat shock protein 90, alpha (cytosolic), class A member 1	NM_175761.2	1.67	0.009
Il18bp	interleukin 18 binding protein	NM_053374.1	1.82	0.006
Itgb2	integrin, beta 2	NM_001037780	2.12	0.007
Mx1	myxovirus (influenza virus) resistance 1	NM_173096.2	3.15	0.005
Ncf2	neutrophil cytosolic factor 2	XM_344156.3	2.91	0.009
Plscr1	phospholipid scramblase 1	NM_057194.1	2.31	0.002
Pros1	protein S (alpha)	NM_031086.1	1.74	0.002
RT1-Ba	RT1 class II, locus Ba	NM_001008831.1	2.53	0.004
RT1-Da	RT1 class II, locus Da	NM_001008847.1	2.10	0.007
Slpi	secretory leukocyte peptidase inhibitor	NM_053372.1	10.95	0.003
Stat1	signal transducer and activator of transcription 1	NM_032612.2	1.60	0.005
Stat5b	signal transducer and activator of transcription 5b	NM_022380.1	1.78	0.009
Syk	spleen tyrosine kinase	NM_012758.1	2.02	0.007
Tfpi	tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor)	NM_017200.1	2.57	0.002
Timp1	TIMP metallopeptidase inhibitor 1	NM_053819.1	3.03	0.008
Tnfrsf12a	tumor necrosis factor receptor superfamily, member 12a	NM_181086.2	2.13	0.009
Tyrobp	Tyro protein tyrosine kinase binding protein	NM_212525.1	2.80	0.006

^*t-test.

Inflammatory mediators.

The most significantly increased expression in DA rats and therefore reciprocally reduced expression in DA.F344(Cia4) synovial tissues were the protease inhibitor Slpi (10.94-fold) and its receptor Plscr1 (2.31-fold) (35), the macrophage and monocyte scavenger receptor Cd163 (5.85-fold), the C-C chemokine Ccl7 (5.17-fold) and Lst1 (4.97-fold) (Tables 1–3, Fig. 5A). These genes have been previously found to be expressed in increased levels in synovial tissues or blood cells from either rodent models of arthritis or from patients with RA (1, 3, 8, 14, 30, 33, 39). These similarities with RA underscore the relevance of our observations and point to a potential role of the arthritis severity gene located in the Cia4 interval in the regulation of these events in human disease also. Cd163 is a macrophage-specific gene. Macrophage and other myeloid cells are also the predominant source of Ccl7 and Lst1, suggesting increased number or increased activation either of synovial macrophages or of infiltrating macrophages, neutrophils, and dendritic cells in DA. DA.F344(Cia4) synovial tissues also had reduced levels of the TNFα-inducing factor Litaf (2.09-fold), which regulates TNF-α and IL-6 levels and was recently shown to be an important mediator of arthritis severity and joint damage (28). Nine genes involved in respiratory burst were expressed in increased levels in DA (Table 2).

Fig. 5.

Network interactions among the differentially expressed genes. A: genes expressed in reduced levels in DA.F344(Cia4) congenics and reciprocally increased in DA and implicated in the activation or interaction with the Spleen tyrosine kinase (Syk) and NF-κB pathways. B: genes expressed in increased levels in DA.F344(Cia4) congenics and implicated in metabolism and transport across membranes and centered around the PDGF and insulin gene and pathways. Red, gene expression in increased levels in the respective strain; numbers below gene symbol: top, P value; bottom, fold-difference compared with DA. Red boxes highlight central pathways.

Table 2.

Selected genes with increased expression in DA synovial tissues and reduced levels in DA.F344(Cia4) congenics

Symbol	Name (description)	RefSeq ID	Fold	P Value**
Cell surface receptors
Adora2a‡	adenosine A2a receptor	NM_053294.3	1.52	0.009
Ccr1	chemokine (C-C motif) receptor 1	NM_020542.2	1.69	0.005
Cd163	Cd163 molecule	XM_001061914.1	5.85	0.001
Cd37	CD37 molecule	NM_017124.1	2.21	0.002
Cd74¶	Cd74 molecule, major histocompatibility complex, class II invariant chain, (Mif receptor)	NM_013069.1	2.60	0.004
Clec10a	C-type lectin domain family 10, member A	NM_022393.2	2.04	0.003
Clec4a1	C-type lectin domain family 4, member a1	NM_001005890.2	2.16	0.006
Clec4a3	C-type lectin domain family 4, member a3	NM_001005891.1	3.58	0.005
Cysltr1	cysteinyl leukotriene receptor 1	NM_053641.1	1.84	0.01
Fcer1 g‡	Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide	XM_001053627.1	2.09	0.004
Fcgr2b	Fc gamma receptor II beta (LOC498276)	NM_001135992	1.78	0.007
Itgb2‡	integrin, beta 2	NM_001037780	2.12	0.008
Plscr1	phospholipid scramblase 1 (receptor for Slpi; modulates mast cell responses)	NM_057194.1	2.31	0.002
RGD1561778	similar to dendritic cell-derived immunoglobulin(Ig)-like receptor 1, DIgR1	NM_001168284	2.74	0.004
RT1-Ba¶	RT1 class II, locus Ba	NM_001008831.1	2.53	0.005
RT1-Da¶	RT1 class II, locus Da	NM_001008847.1	2.10	0.007
RT1-M6-2	RT1 class I, locus M6, gene 2	NM_001008853.1	1.57	0.002
Scarb2	scavenger receptor class B, member 2	NM_054001.2	1.97	0.007
Slc39a8	solute carrier family 39 (zinc transporter), member 8	NM_001011952.1	1.84	0.005
Tnfrsf12a	tumor necrosis factor receptor superfamily, member 12a (TWEAK receptor)	NM_181086.2	2.12	0.009
Trpv4	transient receptor potential cation channel, subfamily V, member 4	NM_023970.1	2.01	0.004
Protease inhibitors
Slpi	secretory leukocyte peptidase inhibitor	NM_053372.1	10.94	0.003
Tfpi	tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor)	NM_017200.1	2.57	0.003
Timp1	TIMP metallopeptidase inhibitor 1	NM_053819.1	3.03	0.008
Proteases
Adamts1	ADAM metallopeptidase with thrombospondin type 1 motif, 1	NM_024400.1	2.72	0.004
Anpep‡	alanyl (membrane) aminopeptidase	NM_031012.1	1.55	0.008
Htra3	HtrA serine peptidase 3 (TGFb signaling antagonist)	XM_001058037.1	2.16	0.01
Lgmn	legumain	NM_022226.2	2.19	0.006
Oma1	OMA1 homolog, zinc metallopeptidase (S. cerevisiae)	XM_001063646.1	1.97	0.009
Signaling molecules
Fgr‡	Gardner-Rasheed feline sarcoma viral (v-fgr) oncogene homolog	NM_024145.2	2.31	0.006
Hck‡	hemopoietic cell kinase	NM_013185.2	2.44	0.005
Litaf#	lipopolysaccharide-induced TNF factor	XM_001078236.1	2.09	0.009
Map4k1	mitogen activated protein kinase kinase kinase kinase 1	XM_001074916.1	1.70	0.009
Rhog	ras homolog gene family, member G (rho G)	NM_001037195.1	1.53	0.006
Stat1#	signal transducer and activator of transcription 1	NM_032612.2	1.60	0.005
Stat5b#	signal transducer and activator of transcription 5B	NM_022380.1	1.78	0.009
Syk#‡	spleen tyrosine kinase	NM_012758.1	2.02	0.007
Tifa	TRAF-interacting protein with forkhead-associated domain	NM_001014044.1	2.03	0.003
Tnfaip8l2	tumor necrosis factor, alpha-induced protein 8-like 2 (TLR and TCR negative regulator)	NM_001014039.1	2.45	0.007
Tyrobp‡	Tyro protein tyrosine kinase binding protein	NM_212525.1	2.80	0.006
Chemokines, growth factors, and synthesis of reactive oxygen species
Ccl7	chemokine (C-C motif) ligand 7	NM_001007612.1	5.17	0.004
Ccl21	chemokine (C-C motif) ligand 21	NM_001008513.1	1.78	0.005
Fgf7	fibroblast growth factor 7	NM_022182.1	1.65	0.007
Ncf2‡	neutrophil cytosolic factor 2	XM_344156.3	2.91	0.009
Extracellular matrix, tissue repair, and antioxidant genes
Angptl4*	angiopoietin-like 4	NM_199115.2	2.99	0.006
Glrx2	glutaredoxin 2	NM_001013034.1	1.55	0.005
Gpx7	glutathione peroxidase 7	XM_001060175.1	2.10	0.008
Mt1a	metallothionein 1a	NM_138826.4	3.69	0.005
Ninj1§	ninjurin 1	NM_012867.1	1.56	0.005
Spon1§	spondin 1, extracellular matrix protein	NM_172067	3.67	0.003
Cytoskeleton
Coro1a	coronin, actin binding protein 1A	NM_130411.2	2.17	0.006
Gmfg	glia maturation factor, gamma	NM_181091.2	1.85	0.008
Lcp1#	lymphocyte cytosolic protein 1	NM_001012044.1	1.99	0.007
Parva	parvin, alpha	NM_020656.2	2.31	0.001
Pdpn*	podoplanin (influenza receptor, ERM binding and invasion/met correlation, lymphangiogenesi	NM_019358.1	1.82	0.009
Protein trafficking, folding, and degradation
Hsp90aa1	heat shock protein 90, alpha (cytosolic), class A member 1	NM_175761.2	1.67	0.009
Ubfd1	ubiquitin family domain containing 1	NM_001034911.1	2.21	0.009
Unknown function
Bcl2a1d	B-cell leukemia/lymphoma 2 related protein A1d	NM_133416.1	3.51	0.005
Ifi204¶	interferon activated gene 204	NM_001012029.1	2.98	0.001
Ifitm1¶	interferon induced transmembrane protein 1	XM_215117.4	3.94	0.001
Lst1¶	leukocyte specific transcript 1	NM_022634.1	4.97	0.008
Mx1¶	myxovirus (influenza virus) resistance 1	NM_173096.2	3.15	0.005
Rarres1	retinoic acid receptor responder (tazarotene induced) 1	NM_001014790.1	3.04	0.004

^*Gene associated with tumor development, growth, proliferation and/or invasion;

^#gene mediates cytokine expression;

^¶interferon-inducible gene;

^§nerve and/or axonal differentiation or growth;

^**t-test; &gene contained within the Cia4 interval;

^‡involved in respiratory burst. Boldface, fold-difference = gene among the 10 with most significant difference in expression.

There were 27 genes involved in the activation of the Spleen tyrosine kinase (Syk) or NFκB, or known to interact with these two genes and pathways expressed in increased levels in DA in significantly reduced levels in congenics, suggesting that the arthritis gene located in the Cia4 interval is involved in the regulation of these signaling pathways (Fig. 5A). These Syk or NFκB-interacting/activating genes and immune and inflammatory responses included Cd74 (2.60-fold), which is involved in antigen processing and is also a receptor for the proinflammatory cytokine Mif, MHC class I RT1-M6-2, MHC class II molecules RT1-Ba and RT1-Da (orthologs of human HLA-DQa1 and HLA-DRa), signal transduction molecules (Stat1, Stat5b, Syk, Tifa, Tyrobp), chemokines and chemokine receptors (Ccl7, Ccl21, Ccr1), adhesion molecules and c-type lectin receptors (Clec4a1, Clec4a3, Clec10a, Itgb2), molecules involved in innate immune responses (C1qa, Cfp, Fcer1g, Mca32, Ncf2), and additional myeloid receptors (Scarb2 and Fcgr2b) (Tables 1–3).

Rpl30 was the gene with the most significantly increased expression in DA.F344(Cia4) congenics (92-fold higher), compared with DA (Tables 1 and 4). The increased expression of Rpl30 in congenics was confirmed with qPCR; however, levels were at a less striking difference (1.57-fold, P = 0.038; Fig. 6).

Table 4.

Selected genes expressed in increased levels in DA.F344(Cia4) congenics, and reciprocally decreased in DA synovial tissues

Symbol	Name (description)	RefSeq ID	Fold Difference	P Value*
Most significantly increased
Rpl30#	ribosomal protein L30	NM_022699.2	92.76	0.00000005
Metabolism and nutrient transport
Acat1	acetyl-coenzyme A acetyltransferase 1	NM_017075.1	1.77	0.003
Dgat1#	diacylglycerol O-acyltransferase homolog 1 (mouse) (induced by Pparg)	NM_053437.1	1.80	0.007
Dhcr7	7-dehydrocholesterol reductase	NM_022389.2	3.01	0.004
Gstk1	glutathione S-transferase kappa 1	NM_181371.2	1.63	0.008
Gstm1	glutathione S-transferase mu 1	NM_017014.1	1.79	0.008
Nqo1	NAD(P)H dehydrogenase, quinone 1	NM_017000.2	1.68	0.006
Slc25a29	solute carrier family 25, (mitochondrial carrier, palmitoylcarnitine transporter) member 29	NM_001010958.1	2.15	0.008
Reduced levels associated with increased cancer risk
Sgca	sarcoglycan, alpha (dystrophin-associated glycoprotein)	XM_001081302.1	2.90	0.005
Others
Aco2#	aconitase 2, mitochondrial	NM_024398.2	1.63	0.003
Aurkc	aurora kinase C	XM_001067866.1	1.63	0.0001
C1qtnf7	C1q and tumor necrosis factor related protein 7	XM_001060132.1	1.84	0.005
Calcoco1	calcium binding and coiled coil domain 1	NM_139190.1	1.70	0.006
Cirbp	cold inducible RNA binding protein	NM_031147.2	2.14	0.003
Cuedc1	CUE domain containing 1	NM_001013971.1	1.76	0.001
Dclk2	doublecortin-like kinase 2	NM_001009691.2	1.99	0.0004
Flywch1	FLYWCH-type zinc finger 1	XM_001056224.1	2.00	0.004
Gabarapl2	GABA(A) receptor-associated protein like 2	NM_022706.2	1.65	0.01
Giyd2	GIY-YIG domain containing 2	NM_001009292.1	2.04	0.002
LOC679430	hypothetical protein LOC679430	XM_001056304.1	2.17	0.003
Mapk12#	mitogen-activated protein kinase 12	NM_021746.1	1.94	0.002
Mif4 gd	MIF4G domain containing	NM_001014122.1	2.09	0.002
Neurl2	neuralized homolog 2 (Drosophila)	XM_230848.4	1.76	0.003
Pdrg1	p53 and DNA damage regulated 1	NM_001014762.1	1.63	0.004
Ptms	parathymosin	XM_001062784.1	1.66	0.008
RGD1308059	similar to DNA segment, Chr 4, Brigham & Womens Genetics 0951 expressed	NM_001025022.1	1.76	0.006
Rrp7a#	ribosomal RNA processing 7 homolog A (S. cerevisiae)	XM_001077728.1	2.47	0.00003
Rtn1	reticulon 1	NM_053865.1	1.66	0.004
Slc1a1	solute carrier family 1 (neuronal/epithelial high affinity glutamate transporter, system Xag), member 1	NM_013032.2	1.58	0.0001

Boldface, fold-difference = gene among the 10 with most significant difference in expression.

^*t-test;

^#gene located within the Cia4 interval.

Fig. 6.

Validation of microarray results with qPCR. qPCR and microarray fold differences of genes expressed in increased levels in DA (n = 6, red circles) or in DA.F4344(Cia4) congenics' (n = 6, green circles) synovial tissues were significantly correlated (Pearson's correlation coefficient =0.916, P = 0.003). Exceptions were Cysltr1, Ptgs2, and Rrp7a. The microarray fold for Rpl30 was 92.76, and the qPCR fold-difference was 1.57 (qPCR P = 0.038, t-test).

The differential expression and fold-difference of a selected group of genes was confirmed with qPCR (Fig. 6). These included Cpa3, Cysltr1, Ifi204, P2rx5, Parva, Ptgs2, Rrp7a, Tprkb, and Tpsb2.

Genes involved in metabolism and tissue protection.

Other genes with increased expression ranging from 1.5- to 3-fold in DA.F344(Cia4) included those related to lipid and mitochondrial metabolism (Acat1, Dgat1, Dhcr7), the palmitoylcarnitine mitochondrial transporter Slc25a29, and the transmembrane glutamate and aspartate transporter Slc1a1 (required to prevent neurocytotoxicity induced by this neurotransmitter), and the inhibitor of oxidation Nqo1 (Table 4, Fig. 5B). Together, these genes suggest an increased expression of genes involved in metabolism and centered around insulin and Pdgf pathways as suggested by Ingenuity pathway analyses (Fig. 5B), and antioxidative genes in synovial tissues from DA.F344(Cia4) congenics.

Cancer, tumorigenesis, and genes involved in cell proliferation.

Fifty-nine of the genes expressed in increased levels in DA, including Cdk4, Fgf7, Itgb2, Rhog, Syk, Tryrobp, and others, have also been implicated in the regulation of cell proliferation, tumorigenesis, or cancer (Supplemental Table S3). These genes were expressed in reduced levels in DA.F344(Cia4) congenics, suggesting that a gene located within the congenic interval directly or indirectly regulates these genes and could be affecting the ability of the synovial tissue to proliferate (synovial hyperplasia) and invade and erode cartilage and bone.

Strain-specific gene expression.

Strain-specific synovial tissue gene expression could provide critical information about the identity and function of the arthritis gene located within the Cia4 interval. To quantify this difference in detection, we considered preferentially expressed genes those that were detected with a difference of at least 60% between DA and DA.F344(Cia4) (4 and 0, 5 and 0, 5 and 1, 6 and 0, 6 and 1, and 6 and 2) (Table 5). We found 130 genes to be preferentially expressed in DA and 58 genes in DA.F344(Cia4). Similarly to differentially expressed genes, 23 of the genes preferentially expressed in DA included cytokines and cytokine receptors (Il1rn, Il2ra, Il10, Il21r), signal transduction molecules implicated in inflammatory responses (Itk, Lck, Traf4), chemokines and chemokine receptors (Ccl20, Ccr2, Cx3cr1), members of the T cell receptor complex (Cd3d, Cd8b), adhesion molecules (Cd84, Sell), and genes encoding members of the innate immune response (Card11, Cfi, Nlrp3, Nlrc4, Ptger4, Tlr4). Sixteen genes implicated in NF-κB activation were also among the genes preferentially expressed in DA, further adding to the list of NF-κB-interacting/activating genes with significant difference between these two strains.

Table 5.

Selected genes preferentially expressed in the synovial tissues of DA or DA.F344(Cia4) rats

Symbol	Name	RefSeq ID	DA Pos*	Cia4 Pos*	P Value#	NF-κB§	Inflammation
PREFERENTIAL EXPRESSION IN DA
Mediators of inflammation
Ccl20	chemokine (C-C motif) ligand 20	NM_019233.1	5	1			+
Ccr2	chemokine (C-C motif) receptor 2	NM_021866.1	6	2			+
Cd3d	CD3 molecule delta polypeptide	NM_013169.1	5	1			+
Cd84	CD84 molecule	XM_001054923.1	5	0	0.01	+	+
Cd8b	CD8b molecule	NM_031539.1	4	0			+
Cx3cr1	chemokine (C-X3-C motif) receptor 1	NM_133534.1	4	0		+	+
Il10	interleukin 10	NM_012854.1	5	1		+	+
Il1rl1	interleukin 1 receptor-like 1	NM_013037.1	4	0		+	+
Il1rn	interleukin 1 receptor antagonist	XM_001071294.1	4	0		+	+
Il21r	interleukin 21 receptor	NM_001012469.1	6	1	0.01	+	+
Il2ra	interleukin 2 receptor, alpha	NM_013163.1	5	1		+	+
Itk	IL2-inducible T-cell kinase	XM_343880.2	4	0			+
Lck	lymphocyte-specific protein tyrosine kinase	XM_232763.4	6	2		+	+
Nlrc4	NLR family, CARD domain containing 4	XM_001065561.1	6	2		+	+
Nlrp3	NLR family, pyrin domain containing 3	XM_220513.4	5	1		+	+
Pla2 g7	phospholipase A2, group VII (platelet-activating factor acetylhydrolase, plasma)	NM_001009353.1	6	0	0.002		+
Ptger4	prostaglandin E receptor 4 (subtype EP4)	NM_032076.2	6	2			+
Sell	selectin L	NM_019177.1	5	1			+
Tlr4	toll-like receptor 4	NM_019178.1	5	1		+	+
Tnfsf14	tumor necrosis factor (ligand) superfamily, member 14	XM_001059278.1	5	0	0.01	+	+
Traf4	Tnf receptor associated factor 4	XM_220640.4	5	0	0.01	+	+
Solute carriers and ion channels
Slc25a32	solute carrier family 25, member 32	XM_235359.4	6	0	0.002
Slc41a2	solute carrier family 41, member 2	XM_343191.3	5	0	0.01
Slc7a14	solute carrier family 7 (cationic amino acid transporter, y+ system), member 14	XM_574912.1	4	0
Slc18a3	solute carrier family 18 (vesicular acetylcholine), member 3	NM_031663.2	4	0
Trpm1	transient receptor potential cation channel, subfamily M, member 1	NM_001037733.1	4	0
Cancer phenotypes and cell cycle regulation
Arhgap8	Rho GTPase activating protein 8	NM_001004242.1	6	0	0.002
Ccna2	cyclin A2	NM_053702.1	4	0
Ccne2	cyclin E2	XM_001064132.1	4	0
Ccnyl1	cyclin Y-like 1	XM_237211.4	5	1
Cdkn1c	cyclin-dependent kinase inhibitor 1C	NM_001033757.1	5	0	0.01
Erbb3	v-erb-b2 erythroblastic leukemia viral oncogene homolog 3 (avian)	NM_017218.2	4	0
Fut7	fucosyltransferase 7 (alpha (1,3) fucosyltransferase)	NM_199491.1	5	0	0.01
Gtse1	G-2 and S-phase expressed 1	XM_001078275.1	5	1
Igfbp2	insulin-like growth factor binding protein 2	NM_013122.1	5	1
Pak1	p21 protein (Cdc42/Rac)-activated kinase 1	NM_017198.1	6	2
Pak2	p21 protein (Cdc42/Rac)-activated kinase 2	NM_053306.2	5	1
Others
Apoc1	apolipoprotein C-I	NM_012824.1	5	0	0.01	+	+
Card11	caspase recruitment domain family, member 11	XM_001073551.1	6	1	0.01	+
Ccdc130	coiled-coil domain containing 130	NM_001037644.1	5	1
Ezh2	enhancer of zeste homolog 2 (Drosophila)	XM_001073768.1	5	0	0.01
Gfpt1	glutamine fructose-6-phosphate transaminase 1	NM_001005879.1	6	1	0.01
Gins1	GINS complex subunit 1 (Psf1 homolog)	XM_001057757.1	5	0	0.01
Gjc2	gap junction protein, gamma 2	XM_573100.2	6	1	0.01
Gpr182	G protein-coupled receptor 182 (adrenomedullin receptor)	NM_053302.2	6	2
Klhl6	kelch-like 6 (Drosophila)	XM_001062279.1	6	1	0.01
LOC24906	RoBo-1	NM_031537.1	5	0	0.01
Ly49 si1	immunoreceptor Ly49 si1	NM_001009497.1	6	0	0.002
Nup160	nucleoporin 160	XM_001077803.1	6	1	0.01
Olr1	oxidized low density lipoprotein (lectin-like) receptor 1	NM_133306.1	6	1	0.01	+	+
Penk	proenkephalin	NM_017139.1	5	0	0.01
Rgs18	regulator of G-protein signaling 18	XM_222692.4	5	1
Samsn1	SAM domain, SH3 domain and nuclear localization signals, 1	NM_130821.1	6	1	0.01
Spink2	serine peptidase inhibitor, Kazal type 2 (acrosin-trypsin inhibitor)	NM_001008870.1	5	1
St8 sia4	ST8 alpha-N-acetyl-neuraminide alpha-2,8-sialyltransferase 4	XM_001071501.1	6	1	0.01
Susd1	sushi domain containing 1	XM_001061697.1	6	0	0.002
Trh	thyrotropin releasing hormone	NM_013046.2	4	0
Ube2c	ubiquitin-conjugating enzyme E2C	XM_001070446.1	5	0	0.01
PREFERENTIAL EXPRESSION IN DA.F344(Cia4) CONGENICS
Aim1l	absent in melanoma 1-like	XM_216539.4	2	6
Ankrd9	ankyrin repeat domain 9	NM_001012112.1	0	4
Caskin1	CASK interacting protein 1	NM_080690.1	1	5
Catsperg1	cation channel, sperm-associated, gamma 1	XM_218462.4	1	5
Ddb2	damage specific DNA binding protein 2	XM_242065.3	2	6
Dnajc16	DnaJ (Hsp40) homolog, subfamily C, member 16	NM_001014194.1	2	6
Galntl2	UDP-N-acetyl-alpha-d-galactosamine:polypeptide N-acetylgalactosaminyltransferase-like 2	XM_001061731.1	1	5
Gnat2	guanine nucleotide binding protein, alpha transducing 2	XM_345271.3	1	6	0.01
Gpr75	G protein-coupled receptor 75	XM_573685.1	1	5
Gsta4	glutathione S-transferase alpha 4	XM_001059683.1	1	5
Hrasls5	HRAS-like suppressor family, member 5	XM_219546.3	0	4
Hspa4l	heat shock protein 4 like	XM_215549.4	1	5
Hydin	hydrocephalus inducing	XM_001077360.1	0	5	0.01
Lad1	ladinin 1	XM_001063663.1	2	6
LOC501110	similar to Glutathione S-transferase A1 (GTH1) (HA subunit 1) (GST-epsilon) (GSTA1-1)	NM_001024361.1	1	5
Mill1	MHC I like leukocyte 1	NM_001011931.2	1	5
Mlph	melanophilin	NM_001012135.1	1	5
Ntrk1	neurotrophic tyrosine kinase, receptor, type 1	NM_021589.1	1	6	0.01
Ostn	osteocrin	XM_579739.1	1	5
Pamr1	peptidase domain containing associated with muscle regeneration 1	XM_001071590.1	2	6
Pcdhb15	protocadherin beta 15	XM_001055818.1	1	6	0.01
Phldb1	pleckstrin homology-like domain, family B, member 1	XM_001067414.1	1	6	0.01
Plin5	perilipin 5	XM_001061544.1	1	6	0.01
Ptpn14	protein tyrosine phosphatase, nonreceptor type 14 (induces TGFb signaling)	XM_223062.4	2	6
Pvalb	parvalbumin	NM_022499.1	0	6	0.002
Rap1 gap	Rap1 GTPase-activating protein	XM_233609.4	1	6	0.01
Rnls	renalase, FAD-dependent amine oxidase	NM_001014167.1	2	6
Slc16a11	solute carrier family 16 (monocarboxylic acid transporters), member 11	XM_001079593.1	1	6	0.01
Slc39a5	solute carrier family 39 (zinc transporter), member 5	XM_343140.3	1	5
Slurp1	secreted Ly6/Plaur domain containing 1	XM_216966.3	1	6	0.01
Soat2	sterol O-acyltransferase 2	NM_153728.2	1	6	0.01
Spag1	sperm associated antigen 1	NM_001012116.1	2	6
Stap2	signal transducing adaptor family member 2	NM_001025026.1	1	5
Sytl3	synaptotagmin-like 3	XM_574310.2	2	6
Tmod2	tropomodulin 2	NM_031613.1	1	6	0.01

^*Boldface highlights differences of 5 or more between DA and congenics;

^#Fisher's exact test; blank cells, P values >0.05;

^§involved in NF-κB activation.

Five genes were expressed in all DA synovial tissues but in none of the tissues from DA.F344(Cia4) congenics (P = 0.002). These included Slc25a32, a mitochondrial folate carrier, the Rho GTPase activating protein Arhgap8, the phospholipase Pla2g7, the immunoreceptor Ly49si1 and Susd1 (Table 5).

The gene with most significant preferential expression in DA.F344(Cia4) congenics was Parvalbumin (Pvalb), a calcium-binding protein with properties and activity similar to calmodulin, but with unclear function in the synovial tissues or inflammatory cells. Pvalb was expressed in all synovial tissues from DA.F344(Cia4) congenics but in none of the DA tissues (P = 0.002) (Table 5).

Differentially expressed (or strain-specific genes) located within the Cia4 interval.

A total of 11 differentially expressed genes and 10 of the preferentially expressed genes mapped to the Cia4 region (Table 6). Genes with increased expression in DA.F344(Cia4) congenics were more likely to be located within the Cia4 interval, as opposed to those not differentially expressed (8.06 and 2.97%, respectively, P = 0.05, χ²).

Table 6.

Differentially and preferentially expressed genes located within the Cia4 interval

Symbol	Name	RefSeq ID	DA IDs	Cia4 IDs	Fold Increase	P Value*
Increased expression in DA
Cdk4	cyclin-dependent kinase 4	NM_053593.2	6	6	1.62	0.001
Chrac1	chromatin accessibility complex 1	XM_235400.4	6	6	1.96	0.00001
LOC257642	rRNA promoter binding protein	NM_147136	6	6	2.10	0.0004
RGD1565117	similar to 40S ribosomal protein S26	XM_001055593.1	6	6	1.56	0.008
Ttc35	tetratricopeptide repeat domain 35	XM_343234.3	6	6	1.50	0.000001
Utp23	UTP23, small subunit (SSU) processome component, homolog (yeast)	XM_216918.4	6	6	1.50	0.0006
Increased expression in DA.F344(Cia4)
Aco2	aconitase 2, mitochondrial	NM_024398.2	6	6	1.63	0.003
Dgat1	diacylglycerol O-acyltransferase homolog 1 (mouse)	NM_053437.1	6	6	1.80	0.007
Mapk12	mitogen-activated protein kinase 12	NM_021746.1	6	6	1.94	0.002
Rrp7a	ribosomal RNA processing 7 homolog A (S. cerevisiae)	XM_001077728.1	6	6	2.47	0.00003
Rpl30	ribosomal protein L30	NM_022699.2	6	6	92.76	0.00000005
Preferential expression in DA
Slc25a32	solute carrier family 25, member 32	XM_235359.4	6	0	NA	0.002
Arhgap8	Rho GTPase activating protein 8	NM_001004242.1	6	0	NA	0.002
Gpr182	G protein-coupled receptor 182	NM_053302.2	6	2	NA	0.06
Gtse1	G-2 and S-phase expressed 1	XM_001078275.1	5	1	NA	0.08
Zfat	zinc finger and AT hook domain containing	XM_343255.3	4	0	NA	0.06
Cyth4	cytohesin 4	XM_576307.2	4	0	NA	0.06
Preferential expression in DA.F344(Cia4)
Spag1	sperm associated antigen 1	NM_001012116.1	2	6	NA	0.06
Cpt1b	carnitine palmitoyltransferase 1b, muscle	NM_013200.1	2	6	NA	0.06
Slurp1	secreted Ly6/Plaur domain containing 1	XM_216966.3	1	6	NA	0.01
Pvalb	parvalbumin	NM_022499.1	0	6	NA	0.002

^*P values for quantitative comparisons between genes expressed by all sample were calculated with the t-test, with the Fisher's exact test for genes preferentially expressed in one of the strains. NA, not applicable.

DISCUSSION

Genetic factors have a major role in RA disease severity and joint damage (19), yet little is known about the specific identity of those genes (27). These arthritis severity and joint damage genes have the potential to become major targets for drug therapy and important prognostic biomarkers and we have focused our efforts on their discovery. Cia4 is a quantitative trait locus previously shown to regulate disease severity in three well-established models of RA, oil-induced arthritis, CIA, and PIA (36, 37). The human chromosomal regions syntenic to Cia4 are located in parts of chromosomes 8, 12, and 22 and are not known contain any human RA susceptibility or severity gene. However, RA severity and joint damage have not been systematically studied in a genome-wide manner, and therefore those human chromosomal regions simply have not been probed yet. In the present study we analyzed DA.F344(Cia4) congenics, which share most of their genome with DA rats (including the MHC), except for the Cia4 interval on chromosome 7. The simple presence of F344 alleles at the 45.5 Mb Cia4 interval was enough to significantly reduce arthritis severity on both sexes and to reduce joint inflammation even at a very early stage in the disease course. This observation underscores the major effect of the arthritis gene located within Cia4 on disease pathogenesis and severity.

Analyses of gene expression in synovial tissues of rats with PIA revealed a peak of 135-fold increase in mRNA levels of Il6 in DA at an early and preclinical stage, compared with DA.F344(Cia4) congenics. Il6 levels remained increased in all three time-points from preclinical to recent onset disease. Tnfa levels were not significantly different between these two strains, suggesting that Tnfa is not mediating the protective effect in DA.F344(Cia4) congenics. Il1b levels increased days after Il6 and at a lower magnitude, suggesting that Il6 was a major early driver of both arthritis and Il1b expression in synovial tissues. These results suggest that the arthritis severity gene located within the Cia4 interval is a new regulator of the Il6 expression via a yet unidentified mechanism.

Syk (7, 15, 43) and NF-κB (11, 29, 38, 41, 46) pathways are central to the pathogenesis and joint damage in RA and in rodent models of RA. Therefore, it was interesting that the list of genes expressed in increased levels or preferentially expressed in DA compared with DA.F344(Cia4) congenics included many genes that interact with these two pathways or are involved in their activation. While the number of Syk pathway genes detected in this study was not as high as that described in analyses of another congenic strain DA.F344(Cia5a) (2), our observations suggest a novel direct or indirect role for the arthritis gene located within the Cia4 interval in the regulation or activation of both Syk and NF-κB pathways, and it is conceivable that this may be the Cia4-mediated mechanism regulating the differential expression of Il6 in these two strains.

Slpi was the gene with the most significantly increased expression in DA and reciprocally reduced expression in congenics. The Slpi receptor Plscr1 (35) was also expressed in increased levels in DA. Slpi typically protects tissues from serine proteases but can also increase the expression of MMP-2 and tumor cell invasion (6). Slpi was previously found in increased levels in synovial tissues of rats with PIA (2) and in Streptococcal cell wall-induced arthritis (SCWIA) (39). Its exogenous administration reduced levels of TNF-α and joint damage in SCWIA (39). These observations suggest that the increased expression of Slpi and its receptor in DA synovial tissues might represent a failed or insufficient attempt at reducing inflammation and joint damage, perhaps due to insufficient protein synthesis. As a soluble protein, Slpi could be a potentially interesting candidate for therapeutic use in RA.

Several inflammatory mediators were also expressed in increased levels in synovial tissues from DA and reduced in DA.F344(Cia4) congenics. Those included genes preferentially expressed by myeloid cells such as Ccl7, Cd163, and Lst1. Ccl7 is expressed in increased levels in synovial tissue from RA (14) and in rodent models of arthritis (1, 3). It is a chemoattractant for macrophages, dendritic cells, and activated T cells, and two of its receptors, Ccr1 and Ccr2, were also expressed in increased levels in DA, supporting the potential relevance of this axis for cell migration into the synovial tissues at very early stages of arthritis development. Interestingly, Ccl7 has not yet been targeted for therapy in arthritis or other inflammatory diseases, and our results raise it as potentially important.

Ccl21 was also increased in DA compared with congenics. This gene is expressed in increased levels in RA synovial tissues and induces angiogenesis (34) and lymphoid aggregation (26). Polymorphisms in Ccl21 have also been associated with RA susceptibility (40) and with premature mortality in RA (9), underscoring the role of this gene in disease pathogenesis and outcome. Like Ccl7, Ccl21 has not yet been targeted for therapy in RA.

Levels of the TNF-α-inducing factor Litaf were significantly increased in DA compared with congenics. To our knowledge Litaf has not yet been studied in RA, but it was recently shown that Litaf knockout mice are significantly protected in collagen antibody-induced arthritis, developing reduced inflammation and articular damage (28), and suggesting that the increased expression we detected in early stages during disease course could in fact be involved in arthritis pathogenesis and joint damage.

We (16, 21) and others (10, 31) have previously described the increased expression of several genes implicated in cancer-related phenotypes such as cell invasion, proliferation, and metastasis in synovial tissues of arthritic joints. It is considered that those cancer-related genes favor increased synovial cell proliferation, synovial hyperplasia, increased synovial fibroblast invasion, and increased production of proteases that contribute to joint erosion and damage. In this study we identified yet another set of cancer-related genes, different from our previous reports in studies with the DA.F344(Cia5d) and DA.ACI(Cia10) congenics, suggesting that each arthritis severity gene/locus regulates a unique set of genes to cause synovial hyperplasia, synovial invasion, and joint damage, like phenocopies.

Rpl30 and other genes encoding ribosomal proteins, genes implicated in metabolism and transport across membranes, and genes implicated in tissue protection were expressed in increased levels in synovial tissues from DA.F344(Cia4) congenics, suggesting that preserving anabolic metabolism might have a protective role in arthritis or that, conversely, the pathogenic processes taking place in DA synovial tissues inhibit those processes favoring catabolism and increased oxidative and tissue damage.

The analyses of strain-specific gene expression revealed several additional unique and highly relevant differences. While those differences included proinflammatory and signaling mediators and additional cancer-related and cell cycle regulatory genes, these analyses also generated insight into the potential identity of the arthritis severity gene located within the Cia4 interval. Specifically, nine of the genes preferentially expressed in one strain and not in the other were located within the Cia4 interval, suggesting that they could be the actual disease-regulatory gene. Eleven of the differentially expressed genes also mapped to the Cia4 interval. We are currently sequencing these genes and looking for disease-causing variants at coding and noncoding regions, particularly at regulatory elements [5′-untranslated region (UTR) and 3′-UTR] that might explain the differences in expression. Once discovered, these new sequencing variants will be studied to determine their functional consequence and role in arthritis.

Lastly, we considered the possibility that at least part of the differential or preferential gene expression could be secondary to differences in synovial cellularity. We looked for genes specific for T cells (CD4+, CD8+), Tregs, NK cells, B cells, neutrophils, macrophages, dendritic cells, adipocytes, and endothelial cells. However, we did not detect an overrepresentation of cell-specific genes in either DA or DA.F344(Cia4) congenics, suggesting that the differences identified at this very early stage in the course of PIA were more likely due to increased or decreased cell activation/function and not due to obvious differences in cellularity.

In conclusion, we have determined that the arthritis gene located within the Cia4 locus is an early regulator of the expression of several key mediators of arthritis pathogenesis and joint damage. These new discoveries should facilitate the identification of the gene accounting by the Cia4 effect in arthritis and are anticipated to generate new targets for therapy and potential new prognostic biomarkers.

GRANTS

Funded by the National Institutes of Health Grants R01-AR-46213, R01-AR-052439, and R01-AI-54348 to P. S. Gulko.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: M.B. and P.S.G. performed experiments; M.B., T.L., and P.S.G. analyzed data; M.B. and P.S.G. prepared figures; M.B., T.L., and P.S.G. edited and revised manuscript; M.B., T.L., and P.S.G. approved final version of manuscript; T.L. and P.S.G. interpreted results of experiments; P.S.G. conception and design of research; P.S.G. drafted manuscript.