A Tlr7 translocation accelerates systemicautoimmunity in murine lupusSrividya Subramanian*, Katalin Tus*, Quan-Zhen Li*, Andrew Wang*, Xiang-Hong Tian*, Jinchun Zhou*,Chaoying Liang*, Guy Bartov†, Lisa D. McDaniel†, Xin J. Zhou†, Roger A. Schultz†, and Edward K. Wakeland*‡

*Center for Immunology and †Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX 75235

Communicated by Diane Mathis, Harvard Medical School, Boston, MA, May 11, 2006 (received for review April 18, 2006)

The y-linked autoimmune accelerating (yaa) locus is a potentautoimmune disease allele. Transcription profiling of yaa-bearingB cells revealed the overexpression of a cluster of X-linked genesthat included Tlr7. FISH analysis demonstrated the translocation ofthis segment onto the yaa chromosome. The resulting overexpres-sion of Tlr7 increased in vitro responses to Toll-like receptor (TLR)7 signaling in all yaa-bearing males. B6.yaa mice are not overtlyautoimmune, but the addition of Sle1, which contains the auto-immune-predisposing Slam�Cd2 haplotype, causes the develop-ment of fatal lupus with numerous immunological aberrations.B6.Sle1yaa CD4 T cells develop the molecular signature for TFH cellsand also show expression changes in numerous cytokines andchemokines. Disease development and all component autoimmunephenotypes were inhibited by Sles1, a potent suppressor locus.Sles1 had no effect on yaa-enhanced TLR7 signaling in vitro, andthese data place Sles1 downstream from the lesion in innateimmune responses mediated by TLR7, suggesting that Sles1 mod-ulates the activation of adaptive immunity in response to innateimmune signaling.

Sle1 � Sles1 � Toll-like receptor 7 � yaa

Susceptibility to systemic lupus erythematosus is mediated bycomplex genetic and environmental interactions (1, 2). Be-cause of this complexity, we and others have used congenicdissection of lupus-prone mouse strains to characterize theunderlying immune abnormalities responsible for lupus suscep-tibility (3–11). These analyses have provided important insightsinto the component phenotypes associated with individual dis-ease alleles and allowed an assessment of genetic interactionsthat mediate severe lupus. In addition, congenic dissection canbe a key step in the identification of causative alleles, asdemonstrated by the association of polymorphisms in theSLAM�CD2 gene cluster with Sle1b, the most potent lupussusceptibility locus in Sle1 (12).

The BXSB lupus model is a recombinant inbred mouse strainproduced from an intercross of C57BL�6J (B6) and SB�Le (13,14). BXSB male mice develop a severe form of murine lupusbecause of interactions between the y-linked autoimmune accel-erator (yaa) locus on the Y chromosome and other autoimmunedisease alleles in the BXSB genome (reviewed in ref. 15). yaa isthe most potent disease allele detected in the BXSB strain (14).Previous studies found that yaa dysregulated the B cell lineageand that CD4 T cells were necessary for disease but need notcarry the yaa mutation (16–18). Introgression of yaa onto thenonautoimmune C57BL�6J (B6) background revealed that yaacould produce disease only when combined with other autoim-mune-promoting genes (3, 8, 19–21). In our B6 congenic systemwe found that the bicongenic B6.Sle1yaa strain developed severedisease, indicative of strong epistatic interactions between thesetwo susceptibility loci (8).

We previously described four suppressive modifier loci, Sles1–4,which suppressed autoimmunity in NZW mice (22). Sles1 was themost potent of these loci, and we recently fine-mapped it into an�956-kb region at the proximal end of the murine MHC. Geneticcomplementation studies indicated that the suppressive effects of

Sles1 were not mediated by MHC class II molecules and that miceexpressing Sles1 mounted normal immune responses to exogenousstimulation, indicating that this suppressive modifier did not broadlyinhibit the immune system (23).

Here we present data demonstrating that the causative geneticlesion for yaa involves a translocation of the Toll-like receptor(TLR) Tlr7 from the X chromosome onto the yaa Y chromo-some. This translocation causes a 2-fold overexpression of Tlr7in male mice bearing yaa, which is sufficient to dysregulateTLR7-mediated activation of innate immune responses. Wefurther demonstrate that the introduction of Sle1 synergizes withthis regulatory lesion in innate signaling to cause profounddysregulation of the adaptive immune system. Finally, the in-troduction of Sles1 onto the B6.Sle1yaa lupus model(B6.Sle1yaaSles1) eliminated pathogenic autoimmunity and as-sociated phenotypes without suppressing the hyperresponsiveproperties of the TLR7 dysregulation caused by yaa.

ResultsGenetic Interactions Among yaa, Sle1, and Sles1 Modulate SystemicAutoimmunity. To characterize the genetic interactions among theseloci, we compared the immunologic phenotypes in a congenic seriesconsisting of B6.yaa, B6.Sle1, B6.Sle1yaa, and B6.Sle1yaaSles1. Asshown in Fig. 1A, B6.Sle1yaa mice had significant cumulativemortality by 9 months, with all surviving mice exhibiting severekidney pathology characterized by hyalanized end-stage disease inmost kidney glomeruli (Fig. 5, which is published as supportinginformation on the PNAS web site). In contrast, B6.yaa, B6.Sle1,and B6.Sle1yaaSles1 mice had normal lifespans, with little or noevidence of kidney disease by 9 months (Figs. 1A and 5) (5). Theproduction of serum IgM and IgG anti-chromatin, anti-dsDNA,and anti-GBM (glomerular basement membrane) autoAbs was alsoimpacted by these genetic interactions. Relative to B6, B6.yaa micehad increased levels of IgM autoAbs by 4–6 months, and B6.Sle1mice produced high IgG titers against chromatin. Significant levelsof autoAbs were detectable in B6.Sle1yaa by 6–8 weeks, and,consistent with the incidence of severe glomerulonephritis, thetiters dramatically increased by 4–6 months (Fig. 5). The autoan-tigen microarray profiling of these sera (shown in Fig. 1B) revealedthat B6.Sle1yaa mice at 4–6 months preferentially produced hightiters of IgG autoAbs against dsDNA and kidney glomerularantigens, which strongly correlated with severe kidney pathology(24). These data indicate that yaa and Sle1 dictate independentcontributions to the development of autoAbs, and combining thetwo loci amplified these phenotypes, resulting in accelerated pro-duction of disease-associated IgG autoAbs.

yaa Contains a Translocation of the X-Linked Tlr7 Gene. Modulationsin global gene transcription ensuing from the epistatic interactions

Conflict of interest statement: No conflicts declared.

Abbreviations: TLR, Toll-like receptor; BAC, bacterial artificial chromosome; TFH, follicularT helper.

‡To whom correspondence should be addressed. E-mail: edward.wakeland@utsouthwestern.edu.

© 2006 by The National Academy of Sciences of the USA

9970–9975 � PNAS � June 27, 2006 � vol. 103 � no. 26 www.pnas.org�cgi�doi�10.1073�pnas.0603912103

Dow

nloa

ded

by g

uest

on

June

4, 2

021

among these susceptibility loci were measured with the SentrixMouse-6 (48,000) array (Illumina). This analysis compared expres-sion profiles from three biological replicas of sorted splenic CD19�and CD4� cells from B6, B6.yaa, B6.Sle1yaa, and B6.Sle1yaaSles1mice at 4–6 months of age. A Boolean comparison identified 18genes that were differentially expressed in yaa-bearing B cells at acorrected significance of P � 0.01 compared with B6 (Fig. 6, whichis published as supporting information on the PNAS web site). Fourof these genes (Trappc2, Rab9, Tlr7, and Prps2) are located in an�1-Mb segment, just proximal to the pseudoautosomal region atthe telomeric end of the X chromosome (Fig. 2A Upper) (www.ensembl.org�Mus�musculus) (25). The expression of these fourgenes was 2-fold greater in CD19 B cells from yaa strains comparedwith B6 (Fig. 2A Upper). The up-regulation of Tlr7 was verified byreal-time quantitative PCR (Fig. 2B). Tlr8, which is located justdistal to Tlr7, was not detectably expressed in the B or CD4 T cellsfrom any of these mice (data not shown).

Based on these data, we hypothesized that yaa contained atranslocation of this segment of distal X to the Y chromosome,resulting in aberrant overexpression of genes that normallyundergo X inactivation in females. An X-to-Y translocation wasconfirmed by using two-color FISH with bacterial artificialchromosome (BAC) probes specific for the Y chromosome(RP24-208N6; pink) and the X chromosome region containingTlr7 and Prps2 (RP23-92P6; green). As shown in Fig. 2C,hybridization of metaphase chromosome spreads identified Tlr7and Prps2 (labeled in green) on the X chromosomes of both B6and B6.yaa mice. However, an additional copy of Tlr7 and Prps2was found on the Y chromosome only in B6.yaa mice (Fig. 2CRight). This additional copy on the Y chromosome caused a2-fold increase in Tlr7 gene expression in purified B cells frommales in comparison to littermate females from yaa strains,consistent with the circumvention of normal X inactivationmechanisms by genes in the translocated segment (Fig. 6). DNAsequencing of both Tlr7 and Rab9 detected no polymorphisms

within the coding regions of these genes between B6 and B6.yaamales, indicating that the translocated copies on yaa encodeproteins that are identical to those on the X chromosome (datanot shown). In addition, no changes in the tissue specificity forTlr7 have been detected (data not shown), indicating that theprimary consequence of this translocation is the 2-fold increasein gene transcription.

Fig. 1. Genetic interactions among yaa, Sle1, and Sles1 modulate systemicautoimmunity. (A) Cohorts of male mice of the various yaa strains indicatedwere aged to 9 months to assess for spontaneous mortality (n � 15–23). (B)Sera were obtained from 4- to 6-month-old mice and assayed for the produc-tion IgG autoAbs of different specificities by using protein antigen arrays (24).

Fig. 2. yaa contains a translocation of the X-linked Tlr7 gene. (A) Expression-level fold changes of the indicated X chromosome genes in B cells from 4- to6-month-old mice of the different yaa strains relative to B6 controls, asdetermined by using Illumina BeadChip arrays (Upper). (Lower) Positions ofthe changed genes (bold) on the F5 band of mouse chromosome X accordingto www.ensembl.org. (B) Quantitative real-time PCR analyses of Tlr7 expres-sion in purified B cells from different strains, confirming up-regulation in Tlr7expression in yaa-bearing mice relative to B6. Error bars represent SEM. (C)Two-color FISH analyses were performed by using FITC-labeled RP23-92P6 BACDNA (green) and Spectrum Orange-labeled RP24-208N6 BAC DNA (pink).Hybridization was done to metaphase chromosome spreads of splenic lym-phocyte cultures from 7-week-old B6 and B6.yaa mice. Hybridization of re-spective probes is indicated: arrows, RP23-92P6 (green); arrowheads, RP24-208N6 (pink).

Subramanian et al. PNAS � June 27, 2006 � vol. 103 � no. 26 � 9971

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

June

4, 2

021

Functional Changes Associated with the Tlr7 Translocation in yaa Mice.Recent work has documented the roles of TLR7 and TLR9,which are nucleic acid-binding members of the innate immunesystem family of TLRs, in the molecular interactions that lead toa breach in tolerance to nuclear antigens in different systems(26–33). TLR7 recognizes viral ssRNA (34–36), thus making itan excellent candidate for involvement in autoimmunity toRNA-containing antigens. Furthermore, TLR7 is expressed onB and myeloid lineage cells, but not in CD4 T cells, consistentwith the documented functional properties of yaa (16–18). Giventhese data, Tlr7 emerged as a strong candidate for the autoim-mune phenotypes mediated by yaa.

The functional effects of a 2-fold increase in the transcription ofTLR7 were tested in vitro by using the synthetic TLR7 agonist R837.Splenocytes from 6- to 8-week-old male mice were cultured for 24and 48 h with various concentrations of R837 and assayed fordifferences in activation markers, proliferation, and IL12p40 secre-tion after stimulation. As shown in Fig. 3A, splenocytes from B6.yaamice responded significantly more strongly to stimulation withR837 than B6 controls. Analogous experiments assessing responsesof these strains to stimulation through TLR4 (LPS) and TLR9(CpG ODN 1826) indicated that yaa enhances the proliferativeresponses of B cells to TLR4, but not TLR9. In this regard,stimulation by R837 in strains expressing yaa was inhibited bychloroquine, consistent with the maintenance of the specific ex-pression of TLR7 in endocytic vesicles (Fig. 7, which is published assupporting information on the PNAS web site) (37). Taken to-gether, these data indicate that the enhanced stimulation with theTLR7 agonist was not due to aberrant subcellular localization orgeneralized aberrancies in the stimulation properties of receptorslocated within such vesicles.

As discussed above, B6.yaa mice develop a minimal autoim-mune phenotype involving the production of low titers of IgMautoAbs (Fig. 5). However, as shown in the heat map in Fig. 3B,both the B6.yaa and B6.Sle1yaa strains produced IgM autoAbsthat recognized RNA-associated autoantigens, including Ro andLa, as well as a variety of small nuclear ribonucleoproteins(SnRNPs). Whereas B6.yaa produced only IgM autoAbs againstthese antigens, B6.Sle1yaa mice transitioned to the production ofIgG (Fig. 1). In contrast, sera from autoimmune B6.Sle1 mice didnot preferentially recognize this subset of antigens. These resultssupport the hypothesis that dsyregulated TLR7 signaling may bedirectly involved in mediating the breach in B cell immunetolerance to RNA-containing antigens in yaa-bearing mice.

Finally, the in vivo phenotypes of B6.yaa mice were concentratedin the B cell compartment, as evidenced by significant decreases inthe percentages of both T2 (B220�CD23�CD21hiIgMhi) and MZ(B220�CD23�CD21�IgM�) B cells in B6.yaa mice relative to B6(Fig. 8, which is published as supporting information on the PNASweb site) (38). Although previous studies reported the absence ofmonocytosis and other associated immunological abnormalities inB6.yaa mice (20, 39), we found mild myeloid and B lymphocytedysregulations compared with B6 (Fig. 8). Overall, B6.yaa micehave subtle immune phenotypes, indicating that the deleteriousconsequences of the up-regulation of Tlr7 are predominantly mod-ulated by immunoregulatory mechanisms present in B6 mice.

Interactions of Sle1 with yaa Drives Severe Autoimmune Disease. Theaddition of Sle1 to B6.yaa causes a transition to severe autoim-munity accompanied by the development of a variety of immunesystem aberrations, including extensive splenomegaly and atremendous increase in splenic cellularity. To assess the originsof this dysregulation, the cell lineage distributions in thymic,bone marrow, and splenic populations in the entire congenicpanel were analyzed at 6–8 weeks and at 4–6 months (Tables 1and 2, which are published as supporting information on thePNAS web site). This analysis revealed an extensive expansion ofall immune cell lineages, detectable by 6–8 weeks of age and

dramatically increasing by 4–6 months. Most notably, a dramaticincrease in the splenic CD11b� population was observed, with astriking expansion in the monocyte and neutrophil populations.These phenotypes are consistent with many of the immunologicdysregulations described in the parental BXSB.yaa strain (15).

A comparison of the immune cell populations of B6.yaa andB6.Sle1yaa revealed dramatic increases in the activation of the T celllineage as a major consequence of the genetic interaction betweenthese two disease loci. T lymphocytes from B6.Sle1yaa mice showedearly and progressive activation, such that by 4–6 months �50% ofthe CD3 T cells from B6.Sle1yaa mice were CD69� (Table 2). BothCD4 and CD8 T cells exhibited this chronic activation, along witha marked differentiation toward a CD62L�CD44hi memory phe-notype (Table 2). Fig. 4A presents a short list of genes with

Fig. 3. Functional changes associated with the Tlr7 translocation in yaa. (A)Sterile splenocyte suspensions from 6- to 8-week-old mice were plated inquadruplicate at 0.5 � 106 cells per ml in complete RPMI medium 1640 withvarying concentrations of the synthetic TLR7 agonist R837 (Imiquimod). Du-plicate plates were set up, and cells were cultured at 37°C for 24 and 48 h.Either plates were pulsed with [3H]thymidine for the last 9 h and [3H]thymi-dine incorporation was measured by �-scintillation counting to assay for cellproliferation, or cells and supernatants were collected for flow-cytometric andELISA analyses. In all bar charts error bars represent SEM (n � 3–6). *, P � 0.05;**, P � 0.01; ***, P � 0.001 (versus B6 by ANOVA). (B) Sera obtained from 4-to 6-month-old mice were assayed by using protein antigen arrays and auto-Abs specific for RNA-containing antigens examined using cluster analyses (24).Color coding of individual samples: black, mean of all samples; increasinggreen, increasingly below mean; increasing red, increasingly above mean.

9972 � www.pnas.org�cgi�doi�10.1073�pnas.0603912103 Subramanian et al.

Dow

nloa

ded

by g

uest

on

June

4, 2

021

expression profiles specifically associated with B6.Sle1yaa CD4 Tcells among the congenic series. Most notably, expression of Il21,Il4, Ifn�, and many other chemokine and cytokine genes werestrongly up-regulated at this point in disease development.B6.Sle1yaa T cells at this stage of disease also had a replicativesenescent phenotype, which is commonly a consequence of thechronic activation associated with severe autoimmunity in lupus-prone mouse strains (Fig. 9, which is published as supportinginformation on the PNAS web site) (8, 40–42).

As shown in Fig. 4B, this expanded population of CD4�effectors exhibited a molecular signature characteristic of thefollicular T helper (TFH) cell population [up-regulated Icos,Pdcd1, Blr1 (Cxcr5), Ccl5 (Rantes), Il21, Cd200, Cd84, Tiam1,and neuropilin], which was confirmed by flow cytometry (Fig.4C). Examination of younger mice revealed that the dysregulatedexpression of these molecules occurs early in the course ofdisease, most notably for ICOS and PD1, and can be detected by6 weeks of age in B6.Sle1yaa splenic CD4 T cells and by 9 weeksin the peripheral blood (data not shown). There was no evidencefor a TFH molecular signature in CD4 cells from B6.yaa. Thus,the expansion of the TFH population seen in B6.Sle1yaa mice isassociated with the development of severe pathology and mayresult from the specific interactions between Sle1 and yaa.

Sles1 Inhibits Disease Development in B6.Sle1yaa Mice. As shown inFigs. 1, 3, and 4 and in the supporting information, B6.Sle1yaaSles1

mice exhibited no signs of pathogenic autoimmunity and had ex vivophenotypes completely analogous to B6. These results demon-strated that Sles1 fully suppresses the autoimmune phenotypes andimmune cell dysregulations mediated by both Sle1and yaa. Asshown in Fig. 3A, Sles1 did not impact the increased responsivenessto R837 that was associated with the TLR7 up-regulation mediatedby yaa. Similarly, Sles1 did not affect the inherent expressiondifferences of the SLAM�CD2 family members mediated by theautoimmune Slam�Cd2 haplotype of Sle1b. These results indicatethat Sles1 operates downstream from both of these primary geneticlesions and operates to reestablish normal tolerance and immuneregulation within an immune system that has dysregulations ingenes involved in both innate and adaptive responses.

DiscussionThese studies demonstrate that a translocation between the Xand Y chromosomes is the genetic lesion underlying the abilityof the BXSB-derived yaa locus to mediate systemic autoimmu-nity in specific genomic contexts. This translocation results in2-fold overexpression of the translocated genes, which includesthe ssRNA-recognizing TLR7 member of the TLR family ofreceptors (34–36). This dysregulation caused increased in vitroresponses to a TLR7 agonist in yaa-bearing strains and poten-tiated the preferential recognition of RNA-associated autoan-tigens in the B6.yaa mouse. Taken together, these data providecompelling evidence that the overexpression of Tlr7 is causal for

Fig. 4. TFH cell signature and cytokine dysregulations observed in B6.Sle1yaa T cells. (A and B) Heat map of gene-expression profiles of different cytokines,chemokines, and their receptors (A) and TFH cell genes (B) in sorted CD4 cells from 4- to 6-month-old mice, as determined by using Illumina BeadChip arrays. (C–E)ICOS (C), PD1 (D), and CXCR5 (E) expression on splenic CD4 subpopulations from the mice at 4–6 months. (Left) Representative histograms of ICOS, PD1, and CXCR5expression on CD4� spleen cells. Light gray, B6; blue, B6.yaa; dark gray, B6.Sle1yaa; brown, B6.Sle1yaaSles1. (Right) Cumulative ICOS, PD1, and CXCR5expression on different CD62L versus CD44 splenic CD4 subpopulations cells at 4–6 months (n � 8 per genotype). Error bars represent SEM. *, P � 0.05; **, P �0.01; ***, P � 0.001 (versus B6 by ANOVA).

Subramanian et al. PNAS � June 27, 2006 � vol. 103 � no. 26 � 9973

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

June

4, 2

021

the autoimmune phenotypes associated with yaa, although con-tributions from other genes within the translocated genomicsegment cannot be ruled out without further studies. In thisregard, murine TLR8 does not appear to recognize ssRNA or thesynthetic agonist R837, which contrasts with the properties ofhuman TLR8 (43, 44).

These analyses provide important insights into how synergisticinteractions between the innate and adaptive immune responsesmay be a key element in genetic susceptibility to autoimmunity.B6.yaa mice are not autoimmune, despite the enhanced signalingmediated by the translocated copy of Tlr7. B6.Sle1 mice develop abenign autoimmunity characterized by the production of IgGanti-chromatin autoAbs. When these two loci are combined, theyinteract to drive a profound and chronic activation of both innateand adaptive immune responses, culminating in fatal glomerulo-nephritis associated with the production of high-titered pathogenicIgG autoAbs. The immune systems of these mice are dramaticallytransformed as a consequence of such unregulated immune acti-vation, which can be detected as early as 6 weeks of age, well beforethe development of significant autoimmune pathology.

The CD4 T cell lineage in B6.Sle1yaa is profoundly dysregulated,exhibiting an early and progressive activation that leads to adramatic increase in IFN�-secreting cells and ultimately a pheno-type resembling that of chronic activation-induced ‘‘replicativesenescence’’ (Fig. 9). Although B6.Sle1 mice also show similaraberrations, the introgression of yaa profoundly amplifies both thekinetics and magnitude of these phenotypes. Because CD4 cells donot express TLR7 (37), this activation is likely impacted by thedysregulation of TLR7-bearing populations in combination withthe inherent B and T lymphocyte defects caused by Sle1. In thisregard, the yaa-mediated increase in the secretion of IL12p40 bysplenocytes stimulated in vitro with R837 is consistent with thedramatic expansion of the IFN�-secreting CD4 T cells. The precisenature of these cellular interactions remains to be elucidated;however, it is reasonable to predict that a TLR7-mediated dysregu-lation of the myeloid lineage is responsible for the increasedproduction of IL12p40 in yaa-bearing mouse strains.

The molecular signature of TFH cells found in B6.Sle1yaa miceis intriguing given the recent association of this phenotype withsevere disease in the san roque lupus model (45, 46). Interest-ingly, key characteristics of TFH cells include the increasedexpression of members of the SLAM�CD2 gene family. Werecently associated extensive polymorphisms of the SLAM�CD2gene cluster with Sle1b, which is the most potent autoimmunelocus identified within Sle1 (12, 47). Furthermore, Sle1b wasrecently identified as the predominant locus responsible forinteractions between yaa and Sle1 (19). The Slam�Cd2 genes,which encode adhesion�costimulatory molecules, participate incellular interactions between both innate and adaptive immunecells, like yaa, thus making them ideal candidates for modulating‘‘crosstalk’’ between these cells. Consequently, the autoimmune-prone Slam�Cd2 haplotype expressed by Sle1, coupled with theyaa-associated enhancement in TLR7 expression, may be the keyto the early chronic activation of the immune system thatprecedes disease development.

Sles1 remains the most intriguing element delineated in thiscongenic dissection of systemic autoimmunity. Our findingsindicate that this modifier fully suppresses the development ofchronic immune activation and autoimmune pathologies inB6.Sle1yaa mice. Because Sles1 neither suppresses the enhancedTLR7 stimulatory capacity nor impacts the polymorphic expres-sion of the Slam�Cd2 gene cluster, and yet fully abrogates lupusin this model, we postulate that Sles1 acts downstream from theinitiating trigger mediated by the interactions between these twoalleles. Importantly, B6.Sle1yaaSles1 mice respond appropriatelyin vitro to immunogenic stimuli. Thus, this modifier reestablishesa balance between the innate and adaptive immune systemswithout suppressing normal functions of the immune system. In

this regard, B6.Sle1yaaSles1 mice do not express the mild auto-immune phenotypes associated with B6.yaa, indicating that thissuppressive modifier can regulate the enhanced TLR7 signalingmediated by yaa more effectively than the B6 genome. Conse-quently, the pathway modulated by Sles1 is an outstanding targetfor therapeutic intervention in human lupus.

Materials and MethodsMice. Mice were housed in the University of Texas SouthwesternMedical Center Animal Resources Center’s specific pathogen-freefacility with the approval of the Institutional Animal Care and UseCommittee of the University of Texas Southwestern MedicalCenter. Original breeders for all strains were obtained from TheJackson Laboratory. The B6.Sle1yaa line was derived by breedingfemale B6.Sle1 mice to male B6.yaa and subsequent intercrossing ofprogeny (8). The introgression of the NZW-derived Sles1 intervalwas performed in a similar manner. The primers used to select forand identify the intervals have been previously reported (48).

Serology. Autoantigen arrays were performed on serum samplesin the University of Texas Southwestern Medical Center Mi-croarray Core, as recently described (24).

Illumina BeadChip Gene Expression Analyses. CD19� and CD4� Band T cells were sorted from the spleens of three mice pergenotype using the MoFlo Flow Cytometer (DakoCytomation)and antibodies against CD19, CD4, CD8, CD11b, and B220 (BDPharmingen�eBiosciences). Purity achieved was �95%. RNAwas isolated from the sorted cells and hybridized to IlluminaSentrix Mouse 6 BeadChip arrays.

BAC FISH Hybridizations. Slides of metaphase spreads were pre-pared from stimulated splenocyte cultures as described previ-ously (49). BAC clones (BACPAC Resources, Oakland, CA)were cultured, and BAC DNA was isolated by using thePhasePrep BAC DNA kit (Sigma). DNA was labeled by using thePrime-It Fluor Fluorescence Labeling Kit (Stratagene) andSpectrum Orange (Vysis, Downers Grove, IL) according to themanufacturers’ instructions, precipitated, resuspended in LSI�WCP hybridization buffer (Vysis), and hybridized to slidesovernight at 37°C. Washed and dehydrated slides were mountedby using Vectashield Mounting Medium with DAPI (VectorLaboratories) and visualized with an AX70 light microscope(Olympus, Melville, NY) using ISIS 5.0 FISH imaging software(MetaSystems, Altlussheim, Germany).

In Vitro TLR Splenocyte Stimulation Assays. Sterile splenocyte sus-pensions were plated in quadruplicate at 0.5 � 106 cells permilliliter in complete RPMI medium 1640 with varying concen-trations of the following stimuli: anti-IgM (Jackson Immuno-Research), LPS (Sigma), CpG ODN 1826 (Coley Pharmaceuti-cals, Ottawa, ON, Canada) and Imiquimod-R837 (InvivoGen,San Diego). In some assays, cells were preincubated with 2�g�ml chloroquine (InvivoGen) for 30 min at 37°C before thestimuli addition. Plates were set up in duplicate, and cells werecultured at 37°C for 24 and 48 h. Either plates were pulsed with[3H]thymidine for the last 9 h and [3H]thymidine incorporationwas measured to assay for cell proliferation or cells and super-natants were collected for flow cytometric and ELISA analyses.

Cytokine ELISAs. Supernatant and serum concentrations of thecytokines IL12p40 and IL6 were determined by using OptEIAELISA kits (BD Biosciences) according to the manufacturer’sinstructions.

Statistical Analysis. Data were analyzed by using INSTAT3 (Graph-Pad, San Diego). Unless otherwise indicated, error bars repre-sent SEMs.

9974 � www.pnas.org�cgi�doi�10.1073�pnas.0603912103 Subramanian et al.

Dow

nloa

ded

by g

uest

on

June

4, 2

021

We are grateful to Dr. Jose Casco for excellent management of the mousecolony and to Angie Mobley for cell sorting. We thank the members of theE.K.W. laboratory for discussions, critical comments, and technical advice,

in particular Alice Chan and Miwako Yamamoto. These studies weresupported by grants from the National Institutes of Health, the LupusResearch Institute, and the Alliance for Lupus Research (to E.K.W.).

1. Wakeland, E. K., Liu, K., Graham, R. R. & Behrens, T. W. (2001) Immunity15, 397–408.

2. Wandstrat, A. E. & Wakeland, E. K. (2001) Nat. Immunol. 2, 802–809.3. Hudgins, C. C., Steinberg, R. T., Klinman, D. M., Reeves, M. J. & Steinberg,

A. D. (1985) J. Immunol. 134, 3849–3854.4. Liu, K., Liang, C., Liang, Z., Tus, K. & Wakeland, E. K. (2005) J. Immunol. 174,

1630–1637.5. Mohan, C., Alas, E., Morel, L., Yang, P. & Wakeland, E. K. (1998) J. Clin.

Invest. 101, 1362–1372.6. Mohan, C., Yu, Y., Morel, L., Yang, P. & Wakeland, E. K. (1999) J. Immunol.

162, 6492–6502.7. Morel, L., Mohan, C., Croker, B. P., Tian, X.-H. & Wakeland, E. K. (1997)

J. Immunol. 158, 6019–6028.8. Morel, L., Croker, B. P., Blenman, K. R., Mohan, C., Huang, G., Gilkeson, G.

& Wakeland, E. K. (2000) Proc. Natl. Acad. Sci. USA 97, 6670–6675.9. Rozzo, S. J., Allard, J. D., Choubey, D., Vyse, T. J., Izui, S., Peltz, G. & Kotzin,

B. L. (2001) Immunity 15, 435–443.10. Vidal, S., Kono, D. H. & Theofilopoulos, A. N. (1998) J. Clin. Invest. 101, 696–702.11. Zhang, D., Fujio, K., Jiang, Y., Zhao, J., Tada, N., Sudo, K., Tsurui, H.,

Nakamura, K., Yamamoto, K., Nishimura, H., et al. (2004) Proc. Natl. Acad.Sci. USA 101, 13838–13843.

12. Wandstrat, A. E., Nguyen, C., Limaye, N., Chan, A. Y., Subramanian, S., Tian,X. H., Yim, Y. S., Pertsemlidis, A., Garner, H. R., Jr., Morel, L., et al. (2004)Immunity 21, 769–780.

13. Andrews, B. S., Eisenberg, R. A., Theofilopoulos, A. N., Izui, S., Wilson, C. B.,McConahey, P. J., Murphy, E. D., Roths, J. B. & Dixon, F. J. (1978) J. Exp. Med.148, 1198–1215.

14. Murphy, E. D. & Roths, J. B. (1979) Arthritis Rheum. 22, 1188–1194.15. Izui, S. (1990) Autoimmunity 6, 113–129.16. Fossati, L., Sobel, E. S., Iwamoto, M., Cohen, P. L., Eisenberg, R. A. & Izui,

S. (1995) Eur. J. Immunol. 25, 3412–3417.17. Lawson, B. R., Koundouris, S. I., Barnhouse, M., Dummer, W., Baccala, R.,

Kono, D. H. & Theofilopoulos, A. N. (2001) J. Immunol. 167, 2354–2360.18. Merino, R., Fossati, L., Lacour, M. & Izui, S. (1991) J. Exp. Med. 174,

1023–1029.19. Croker, B. P., Gilkeson, G. & Morel, L. (2003) Genes Immun. 4, 575–585.20. Izui, S., Higaki, M., Morrow, D. & Merino, R. (1988) Eur. J. Immunol. 18,

911–915.21. Kikuchi, S., Fossati-Jimack, L., Moll, T., Amano, H., Amano, E., Ida, A.,

Ibnou-Zekri, N., Laporte, C., Santiago-Raber, M. L., Rozzo, S. J.. et al. (2005)J. Immunol. 174, 1111–1117.

22. Morel, L., Tian, X.-H., Croker, B. P. & Wakeland, E. K. (1999) Immunity 11,131–139.

23. Subramanian, S. & Wakeland, E. K. (2005) Novartis Found. Symp. 267, 76–88.24. Li, Q. Z., Xie, C., Wu, T., Mackay, M., Aranow, C., Putterman, C. & Mohan,

C. (2005) J. Clin. Invest. 115, 3428–3439.25. Perry, J., Palmer, S., Gabriel, A. & Ashworth, A. (2001) Genome Res. 11,

1826–1832.26. Barrat, F. J., Meeker, T., Gregorio, J., Chan, J. H., Uematsu, S., Akira, S.,

Chang, B., Duramad, O. & Coffman, R. L. (2005) J. Exp. Med. 202, 1131–1139.

27. Boule, M. W., Broughton, C., Mackay, F., Akira, S., Marshak-Rothstein, A. &Rifkin, I. R. (2004) J. Exp. Med. 199, 1631–1640.

28. Krieg, A. M. (2002) Nat. Immunol. 3, 423–424.29. Vollmer, J., Tluk, S., Schmitz, C., Hamm, S., Jurk, M., Forsbach, A., Akira, S.,

Kelly, K. M., Reeves, W. H., Bauer, S., et al. (2005) J. Exp. Med. 202, 1575–1585.30. Lau, C. M., Broughton, C., Tabor, A. S., Akira, S., Flavell, R. A., Mamula, M. J.,

Christensen, S. R., Shlomchik, M. J., Viglianti, G. A., Rifkin, I. R., et al. (2005)J. Exp. Med. 202, 1171–1177.

31. Leadbetter, E. A., Rifkin, I. R., Hohlbaum, A. M., Beaudette, B. C., Shlomchik,M. J. & Marshak-Rothstein, A. (2002) Nature 416, 603–607.

32. Marshak-Rothstein, A., Busconi, L., Rifkin, I. R. & Viglianti, G. A. (2004)Rheum. Dis. Clin. North Am. 30, 559–574.

33. Viglianti, G. A., Lau, C. M., Hanley, T. M., Miko, B. A., Shlomchik, M. J. &Marshak-Rothstein, A. (2003) Immunity 19, 837–847.

34. Diebold, S. S., Kaisho, T., Hemmi, H., Akira, S. & Reis e Sousa, C. (2004)Science 303, 1529–1531.

35. Heil, F., Hemmi, H., Hochrein, H., Ampenberger, F., Kirschning, C., Akira, S.,Lipford, G., Wagner, H. & Bauer, S. (2004) Science 303, 1526–1529.

36. Lund, J. M., Alexopoulou, L., Sato, A., Karow, M., Adams, N. C., Gale, N. W.,Iwasaki, A. & Flavell, R. A. (2004) Proc. Natl. Acad. Sci. USA 101, 5598–5603.

37. Sioud, M. (2006) Trends Mol. Med. 12, 167–176.38. Amano, H., Amano, E., Moll, T., Marinkovic, D., Ibnou-Zekri, N., Martinez-

Soria, E., Semac, I., Wirth, T., Nitschke, L. & Izui, S. (2003) J. Immunol. 170,2293–2301.

39. Amano, H., Amano, E., Santiago-Raber, M. L., Moll, T., Martinez-Soria, E.,Fossati-Jimack, L., Iwamoto, M., Rozzo, S. J., Kotzin, B. L. & Izui, S. (2005)Arthritis Rheum. 52, 2790–2798.

40. Chu, E. B., Ernst, D. N., Hobbs, M. V. & Weigle, W. O. (1994) J. Immunol. 152,4129–4138.

41. Mohan, C., Morel, L., Yang, P., Watanabe, H., Croker, B. P., Gilkeson, G. S.& Wakeland, E. K. (1999) J. Clin. Invest. 103, 1685–1695.

42. Shi, X., Xie, C., Kreska, D., Richardson, J. A. & Mohan, C. (2002) J. Exp. Med.196, 281–292.

43. Hemmi, H., Kaisho, T., Takeuchi, O., Sato, S., Sanjo, H., Hoshino, K., Horiuchi,T., Tomizawa, H., Takeda, K. & Akira, S. (2002) Nat. Immunol. 3, 196–200.

44. Jurk, M., Heil, F., Vollmer, J., Schetter, C., Krieg, A. M., Wagner, H., Lipford,G. & Bauer, S. (2002) Nat. Immunol. 3, 499.

45. Vinuesa, C. G., Tangye, S. G., Moser, B. & Mackay, C. R. (2005) Nat. Rev.Immunol. 5, 853–865.

46. Vinuesa, C. G., Cook, M. C., Angelucci, C., Athanasopoulos, V., Rui, L., Hill,K. M., Yu, D., Domaschenz, H., Whittle, B., Lambe, T., et al. (2005) Nature 435,452–458.

47. Morel, L., Blenman, K. R., Croker, B. P. & Wakeland, E. K. (2001) Proc. Natl.Acad. Sci. USA 98, 1787–1792.

48. Subramanian, S., Yim, Y. S., Liu, K., Tus, K., Zhou, X. J. & Wakeland, E. K.(2005) J. Immunol. 175, 1062–1072.

49. McDaniel, L. D., Chester, N., Watson, M., Borowsky, A. D., Leder, P. &Schultz, R. A. (2003) DNA Repair (Amsterdam) 2, 1387–1404.

Subramanian et al. PNAS � June 27, 2006 � vol. 103 � no. 26 � 9975

IMM

UN

OLO

GY

Dow

nloa

ded

by g

uest

on

June

4, 2

021