|Home | About | Journals | Submit | Contact Us | Français|
Among various surface molecules screened, CXCR4 was significantly up-regulated on monocytes, neutrophils, B-cell subsets, and plasma cells in multiple murine models of lupus with active nephritis, including B6.Sle1Yaa, BXSB, and MRL.lpr. TLR-mediated signaling and inflammatory cytokines accounted in part for this increase. Increased CXCR4 expression was associated with functional consequences, including increased migration and enhanced B-cell survival. Simultaneously, the ligand for CXCR4, CXCL12, was significantly upregulated in the nephritic kidneys. Treatment with a peptide antagonist of CXCR4 prolonged survival and reduced serum autoantibodies, splenomegaly, intra-renal leukocyte trafficking and end organ disease in a murine model of lupus. These findings underscore the pathogenic role of CXCR4/CXCL12 in lymphoproliferative lupus and lupus nephritis and highlight this axis as a promising therapeutic target in this disease.
CXCR4 (CD186) is a G-protein coupled receptor which serves many functions in the immune system. Studies of mice carrying a targeted gene disruption of CXCR4 have revealed its critical role in hematopoiesis, B-cell lymphopoiesis, myelopoiesis, germinal center organization, and maintenance of stem cell pools in the bone marrow (1–5). In addition, CXCR4 is a well-studied molecule in both HIV and cancer. Of note, CXCR4-tropism was found to be correlative with chronic immune activation and AIDS pathogenesis (6, 7), and studies in a variety of human cancers have shown the chemotaxis inducing potential of CXCR4 to be exploited during tumor cell metastasis (8). CXCR4 defects in humans lead to a syndrome characterized by warts, hypogammaglobulinemia, immunodeficiency, and myelokathexis, collectively termed the WHIM syndrome (9).
CXCR4 monogamously recognizes the CXC chemokine ligand 12 (CXCL12), also known as stromal cell-derived factor 1 (SDF-1), or pre-B-cell-growth-stimulating factor (PBSF). CXCL12 is known to be basally expressed by a variety of tissues including skin, heart, and kidney (10), and has been shown to be selectively upregulated in a wide range of tissues in response to damage, particularly the kidney (11). The upregulation of CXCL12 has been postulated to promote mobilization and recruitment of CD34+ progenitor cells to sites of damage as a mechanism of repair and repopulation. In the immune system, CXCL12 plays a major role in determining differential leukocyte output from the bone marrow. Low bone marrow levels of CXCL12 induced by inflammation promote granulocyte production and suppress B-cell generation (12).
We initially became interested in CXCR4 because we observed that it was dysregulated in expression in the BXSB model of spontaneous murine lupus (unpublished observation). Microarray screens of LPS-stimulated bone-marrow-derived macrophages revealed that CXCR4 was highly upregulated when compared to C57BL/6 (B6) controls. Given the known roles of CXCR4 in chemotaxis and B-cell lymphopoiesis (4, 5) and the reported roles of CXCL12 in other forms of renal damage, we postulated that increased CXCR4 expression may play a role in mediating autoreactivity and nephritis in this disease. In this work, we examine the expression of CXCR4 and its ligand in different murine models of lupus, the mechanistic basis for this increase, and the therapeutic potential of blocking this axis in this disease.
Male B6, B6.Sle1, B6.Yaa, B6.Sle1Yaa, BXSB, MRL.lpr, and female, B6.Sle1Sle2Sle3, were purchased from The Jackson Laboratories (Bar Harbor, ME) or bred in our colony at the University of Texas Southwestern Medical Center, and housed in the University of Texas Southwestern Medical Center Animal Resources Center’s specific pathogen-free facility. The care and use of laboratory animals in our facility conforms to the National Institutes of Health guidelines and institutional IACUC approved animal protocols.
Antibodies to the following mouse antigens were used for flow cytometry: CD21-FITC, CD23-PE, IgM-PerCPCy5.5, IgM-PeCy7, CD5-APC, B220-APCCy7, B220-FITC, CXCR4-PE, CXCR4-Biotin, GL7-FITC, CD138-PE, CD69-PE, CD69-PerCPCy5.5, Gr-1-APCCy7, CD3-FITC, CD4-APCCy7, CD86-PE, AA4.1-FITC, CD43-PE, CD45-FITC, CD45-PECy7, NK1.1-PECy7, AnnexinV-PE (BD Biosciences, Franklin Lakes, NJ); CD11b-APC, CD8-PECy7, CD44-APC, CD11c-PacificBlue, Ie/Ib-FITC, F4/80-PECy5, CD3-PacificBlue (eBioscience, San Diego, CA); CD19-PETexasRed, CD62L-PE Texas Red, PDCA-1-Biotin, Strepavidin-QDot655, Strepavidin-PE Texas Red, 7AAD (Invitrogen, Carlsbad, CA). Samples were Fc-blocked with the 2.4G2 antibody or 10% normal rabbit sera (Invitrogen, Carlsbad, CA). For spleen, bone marrow, and blood, at least 5 × 104 cells were acquired in the live gate, as defined by size and granularity. For kidney samples, at least 5 × 105 cells were acquired on the leukocyte gate, as defined by size and CD45 positivity. Samples were either acquired on an LSRII flow cytometer or FACSCalibur (BD Biosciences, Franklin Lakes, NJ), and analyzed using Flow Jo (Tree Star Technologies, San Carlos, CA). MFIs represent median fluorescent intensities.
Mice were perfused with 20 ml cold PBS to remove the blood (Invitrogen, Carlsbad, CA). One-half of a kidney was harvested for histological analyses. The remainder was mechanically disrupted and incubated with 1 mg/ml collagenase IV (Sigma-Aldrich, St. Louis, MO) and 1000 U/ml DNase (Roche, Basel, Switzerland) for 5 min in a 37°C water-bath, and then at 37°C for 30 min with shaking. Cells were furthered mechanically disrupted by passage through a 23-gauge needle and then subjected to hypotonic shock with potassium acetate lysis buffer, to remove residual red blood cells (Sigma-Aldrich, St. Louis, MO). Cells were centrifuged at 3000 rpm at 25°C through a 40% Percoll (Sigma-Aldrich, St. Louis, MO) gradient for 20 minutes. Leukocytes were recovered from the pellet and then enumerated by trypan blue exclusion and subjected to flow cytometric analysis.
Sterile splenocyte suspensions were plated in triplicate at 0.5 × 106 cells/ml in complete medium (RPMI1640 with L-glutamine, 10% FCS (HyClone, Logan, UT), HEPES (Sigma-Aldrich, St. Louis, MO), 1× Pen/Strep (Sigma-Aldrich, St. Louis, MO), beta-2 mercaptoethanol (Sigma-Aldrich, St. Louis, MO), L-glutamine (Sigma-Aldrich, St. Louis, MO), with varying concentrations of the following stimuli: recombinant CXCL12 (R & D Systems, Minneapolis, MN), with or without anti-IgM F(ab)’2, and anti-CD40 (Jackson ImmunoResearch, West Grove, PA). Cells were cultured at 37°C for 72 h and analyzed by flow cytometry.
Bone-marrow derived macrophages were derived from 6–8 week old B6 males, as described (31). Macrophages were plated at 5 × 105 cells/ml, and stimulated for 24 hours with TNF-α (0.05 ng/ml), IL-1-β (10 pg/ml), and/or IL-6 (0.1 ng/ml) (all from R & D Systems, Minneapolis, MN). For the blocking studies, macrophages were stimulated with 10 ng/ml LPS (Sigma-Aldrich, St. Louis, MO), and treated with anti-TNF-α (0.5 ug/ml), anti-IL-1- β (5 ug/ml), and/or anti-IL-6 (0.5 ug/ml) (all from R & D Systems, Minneapolis, MN). For flow cytometry analyses, at least 3 × 104 cells were acquired in the macrophage gate (i.e., cells which were CD11b+ F4/80+), and examined for CXCR4 expression.
6.5 mm Transwell plates with 5.0 µm pore polycarbonate membranes (Corning Inc., Corning, NY) were used. 600 µl of 50 ng/ml CXCL12 was placed in the lower chamber and pre-incubated at 37°C for 2 hours to equilibrate the membrane. 1.5 × 106 splenocytes were loaded into the upper well of the transwell plate. In some experiments, cells were resuspended in complete RPMI medium containing various doses of the CXCR4 inhibitor CTCE-9908 (Chemokine Therapeutics, Vancouver, Canada). Cells were allowed to migrate for 2 hours at 37°C. Cells in the lower chamber were counted using trypan blue exclusion and analyzed by flow cytometry.
Mice were perfused with 20 ml PBS, and single cell suspensions of the kidney were prepared by mechanical disruption and collagenase digestion. Cell suspensions (500 µl PBS per 2 kidneys) were centrifuged at 1500 rpm for 7 min, supernatant collected and further centrifuged at 14000 rpm for 10 min. Supernatant (termed “renal plasma”) was collected and used at various dilutions to quantitate CXCL12 concentration by ELISA (R & D Systems, Minneapolis, MN). OD450 was measured by an Elx800 Automated Microplate Reader (BioTek Instruments, Winooski, VT) and concentrations were extrapolated from a 4-point standard curve (R2 > 0.99), where the mean of experimental duplicates was used.
5-µm paraffin sections of the kidney were boiled in 10 mM citrate, quenched with 0.01% NaBH4, blocked with murine F(ab)’2 anti-IgG (Jackson ImmunoResearch, West Grove, PA), and stained with anti-mouse SDF-1/CXCL12 antibody (R & D Systems, Minneapolis, MN) at 4°C overnight. Sections were then washed and incubated with goat anti-mouse IgG conjugated to Cy3 (Jackson ImmunoResearch, West Grove, PA) for 1 h at room temperature. Sections were washed and then mounted with Vectashield (Vector Laboratories, Burlingame, CA), visualized and photographed with a Zeiss Axioplan 2 and digital camera (Carl Zeiss International, Germany), and analyzed using Axiovision (Carl Zeiss International, Germany).
The peptide antagonist CTCE-9908 (Chemokine Therapeutics, Vancouver, Canada) was resuspended in 5% dextrose water. Two cohorts of B6.Sle1Yaa male mice were assembled for the in vivo “prevention” or “treatment” studies. In “Cohort I”, comprising of 2-mo-old mice, 5 were injected with vehicle and 11 received CTCE-9908. In “Cohort II”, comprising of ANA-seropositive 4-mo-old mice, 5 were injected with vehicle and 8 were injected with CTCE-9908. In both study groups, 100 µl of CTCE-9908 at 50 mg/kg or vehicle placebo were injected intraperitoneally three times a week for the course of study (i.e., till 6 mo of age, or till death). All mice were monitored for serum autoantibodies, proteinuria, azotemia, and evidence of renal pathology, as detailed below.
Mice were bled before and at 1, 2, 3, and 4 mo after treatment with CTCE-9908 or placebo and the sera were stored at −20°C. ELISA detection of serum IgM and IgG autoantibodies to chromatin and dsDNA were performed as described (13). OD450 was measured using an Elx800 Automated Microplate Reader (BioTek Instruments, Winooski, VT) and the raw optical densities for anti-chromatin antibodies were converted to arbitrary normalized units using a six-point standard curve generated by an anti-nuclear mAb derived from a NZM2410 mouse (13).
24 hour urine samples were collected using metabolic cages for proteinuria analyses, which were assessed using the Coomassie Plus Protein Assay kit (Pierce, Rockford, IL) with BSA as a standard. OD630 was measured using an Elx800 Automated Microplate Reader (BioTek Instruments, Winooski, VT). Blood urea nitrogen (BUN) was assessed using the QuantiChrom Urea Assay kit (BioAssay Systems, Hayward, CA). The extent of glomerular and tubulo-interstitial disease, and the percentage of glomerular crescent formation were scored as detailed elsewhere (14).
For the in vivo disease prevention/treatment studies, the experimental mice were compared to the placebo group wherever sufficient placebo mice were still alive. For the 6-mo age comparisons, since most of the placebo controls were already dead, 5–6-mo-old untreated B6.Sle1Yaa mice were used as controls for the statistical analyses. Data were analyzed using InStat3 or GraphPad 4 (GraphPad, San Diego, CA). Where appropriate, one-way ANOVA with Dunnet or Bonferonni post hoc analysis, Welch t-test, Mann-Whitney U-test, or log-rank Mantel-Cox test were used. Error bars represent SEMs.
Examination of several murine lupus models, including BXSB, MRL.lpr, and B6.Sle1Yaa, revealed that CXCR4 was significantly upregulated on multiple cell types in 8–10 month old mice, regardless of their genetic composition, when compared to B6 controls (Figure 1 and Table 1). Similar differences were noted in a fourth lupus-prone strain, B6.Sle1Sle2Sle3 (Table 1). Cell types that exhibited the largest upregulation of CXCR4 compared to B6 in all murine models included cell subsets of both the myeloid and B-cell lineages. In particular, inflammatory monocytes, neutrophils, plasma cells and pre-plasma cells expressed the highest absolute levels of CXCR4, particularly in the lupus-prone strains (Figure 1, Table 1). Both in terms of the mean fluorescence intensities, as well as the percentage of cells expressing CXCR4, myeloid and B-cells from lupus mice exhibited about a two-fold increase in surface expression of CXCR4. Though these differences were most marked on splenic leukocytes, a similar pattern of expression difference was also noted on leukocytes isolated from lymph nodes, peripheral blood, and the bone marrow (Table 1 and data not shown). Although we detected an increase in CXCR4 expression on memory CD4+ T-cells, these results did not reach statistical significance.
CXCR4 is located on murine chromosome 1 at 130 MB, centromeric to the Sle1 interval (with Sle1b being located at 174 MB). To examine if CXCR4 hyper-expression was due to a genetic polymorphism or whether it arose as a consequence of disease, we adopted two approaches, using the B6.Sle1Yaa strain as a disease model. First, evaluation of young (2-mo-old) pre-disease B6.Sle1Yaa mice did not reveal a similar degree of CXCR4 hyper-expression, compared to young B6 controls (Figure 2A). Second, we examined monocongenic mice bearing the two main genetic elements that dictate lupus development in B6.Sle1Yaa mice—Sle1z, and Yaa (15–18). Interestingly, the upregulation of CXCR4 was present only in the bicongenic mice, and only in mice with active lupus, defined by the presence of IgG ANA (anti-chromatin) seropositivity and active proteinuria greater than 1 mg/24hr. (Figure 2). In contrast, the upregulation of CXCR4 was neither observed in B6.Sle1 nor in B6.Yaa mice (Figure 2). Furthermore, direct sequencing of the CXCR4 gene revealed no sequence polymorphisms between B6 and B6.Sle1Yaa (data not shown). Taken together, these data indicate that the upregulation of CXCR4 on B6.Sle1Yaa leukocytes is unlikely to be the direct consequence of any single genetic contribution responsible for disease (i.e., Sle1 or Yaa), but is instead likely to be the downstream consequence of the disease process.
We next explored the mechanistic basis of CXCR4 up-regulation in lupus. B-cell receptor ligation with or without CD40L ligation, or PMA were not able to up-regulate CXCR4 on B-cells; likewise, stimulation of T-cells in vitro with anti-CD3, anti-CD28, PMA, or ConA were also not able to up-regulate CXCR4 (data not shown). In contrast, TLR ligation and inflammatory cytokines appeared to be potent at up-regulating CXCR4 expression in vitro (Fig. 3A). This was partly dependent on inflammatory cytokine production since LPS induced CXCR4 up-regulation was partly dampened by antibodies to IL-1β, Il-6 and TNF-α (Fig. 3A). Interestingly, in all conditions in which IL-6 was neutralized, we observed a significant decrease in CXCR4 expression compared to LPS-treated macrophages, although both direct signaling through TLR4 as well as synergistic effects of both IL-1-β and TNF-α seem to also play a role in mediating CXCR4 upregulation. Likewise, when we stimulated macrophages directly with these recombinant cytokines, all conditions in which IL-6 was added led to significantly increased CXCR4 expression compared to unstimulated macrophages (Figure 3B). These observations are likely to be physiologically relevant since these cytokines were elevated in murine lupus, including the B6.Sle1Yaa strain (Fig. 3C–E) that shows prominent CXCR4 upregulation on various leukocyte subsets. Collectively, these studies indicate that TLR ligation and inflammatory cytokines (rather than antigen receptor stimulation) may be playing key roles in orchestrating the CXCR4 increase seen in lupus. On the other hand, neither stimulation of cells with the TLR7 agonist R837, the cytokine IFNα, nor a variety of TLR9 agonistic CpG ODNs were able to upregulate CXCR4 (data not shown).
We next examined the functional consequences of upregulation of CXCR4 in diseased mice. B6.Sle1Yaa splenocytes, which express more CXCR4, were found 3 times more cells migrated to positive CXCL12 gradients when compared to B6 controls (Figure 4A). Among the cell-types that expressed CXCR4, the migratory potential of neutrophils, monocytes, and B-cells differed most significantly between the strains, although all cell types from diseased mice which express higher CXCR4 levels migrated better to CXCL12, compared to the B6 controls (Figure 4, and data not shown). Also, when CXCL12 was added to both the top and bottom chambers, no significant migration was observed (data not plotted), excluding chemokinesis as a possible explanation for these data.
In addition to its role in chemotaxis, CXCL12 has also been shown to impact B-cell lymphopoiesis (19). Therefore, we assessed whether increased CXCR4 expression affected the responses of B cells to stimulation with CXCL12. We found that BCR-ligated B-cells from B6.Sle1Yaa spleens showed significantly better survival when exposed to CXCL12 (Figure 4B). We detected no differential B-cell responses to CXCL12 in terms of proliferation or the expression of activation markers assessed (data not shown). We also detected no difference in the proliferation or activation of B6.Sle1Yaa T-cells when exposed to CXCL12 (data not shown). These data indicate that the increased expression of CXCR4 on lupus leukocytes is associated with multiple functional consequences, the most profound of which include enhanced BCR-triggered B-cell survival and chemotaxis to positive CXCL12 gradients.
We next examined if the increased renal disease in lupus might be in part driven by heightened CXCR4/CXCL12 activity. We observed a robust increase in CXCL12 expression in the kidneys of B6.Sle1Yaa, but not B6 mice, both by immunohistochemistry (Figure 5A) and by ELISA (Figure 5B). We observed increased CXCL12 expression in both the glomeruli and the tubules. More CXCL12 was also expressed in the interstitium, and Bowman’s capsules of B6.Sle1Yaa kidneys, compared to B6 controls (Figure 5A). Similar increases were also observed in BXSB and MRL.lpr kidneys (data not shown). In addition to the increase in CXCL12, we also noted CXCR4+ cells to accumulate within B6.Sle1Yaa kidneys, when examined ex vivo (Figure 6). Collectively, these data suggest that the CXCR4/CXCL12 axis may be instrumental for leukocyte trafficking into the kidneys, and thus may play an important role in mediating renal pathology in lupus nephritis.
To explore if increased CXCR4/CXCL12 might be a good therapeutic target in lupus, we tested the CXCR4 peptide antagonist CTCE-9908, obtained from Chemokine Therapeutics Corp. Initial characterization indicated that CTCE-9908 was effective in blocking chemotaxis by B6.Sle1Yaa splenocytes in vitro, with the migration of all cell types being inhibited equally (Supplementary Figure S1). To test the therapeutic efficacy in vivo, B6.Sle1Yaa mice were subjected to two sets of placebo-control studies. First, the preventive efficacy of CXCR4 blockade was tested on a cohort of 2-month old B6.Sle1Yaa mice (Cohort I), which typically do not have anti-nuclear antibodies or nephritis. Second, the therapeutic efficacy of CXCR4 blockade was tested on 4-month old B6.Sle1Yaa mice (Cohort II), which by this age have already developed detectable impairment of renal function, high titers of ANAs and a large spectrum of immunological changes (18).
Blocking the CXCR4 axis from the age of 2 months significantly prolonged survival (by about two months), and reduced splenomegaly (P < 0.05), T-cell and B-cell activation (P < 0.05), autoantibody production (P < 0.001) and nephritis (P < 0.01) (Figure 7, and Table 2). In particular, tubulo-interstitial disease and crescent formation, rather than glomerulonephritis, were the pathological features that were most significantly reduced (P < 0.01; Figure 7). The observed change in splenomegaly was accompanied by significant reductions in peripheral T and B-cells, as well as various myeloid cells (Cohort I in Table 2). Importantly, similar disease amelioration and lifespan prolongation was also seen when treatment was started after disease onset (Figure 8, and data not shown). Although splenic cellularity was not reversed in the latter study (Cohort II in Table 2 and data not shown), the “treatment” regime significantly reduced renal disease, including tubulo-interstitial disease (P < 0.01), glomerulonephritis (P < 0.001) and glomerular crescent formation (P < 0.05) (Figure 8). The absolute numbers of leukocytes recruited into the kidneys, particularly various myeloid cell subsets, were significantly reduced by CXCR4 blockade, both in the preventative (P < 0.001) and treatment (P < 0.01) studies, (Table 2, Figure 9). Both the preventative and therapeutic treatments were not associated with any changes in body weight, peripheral red blood cell counts, hemoglobin levels or liver function tests (data not shown), alluding to the safety of the administered drug.
The data presented in this communication suggest an important role for CXCR4/CXCL12 in mediating lymphoproliferative lupus and lupus nephritis. It is clear that CXCR4 is hyper-expressed in all mouse models of lupus examined in this study. These are genetically diverse, including strains harboring lupus susceptibility loci originating from the NZB/NZW, BXSB, as well as the MRL.lpr genetic backgrounds This observation suggests that CXCR4 hyper-expression may be a generalized feature of lupus, independent of the underlying genetic basis. In all of the strains examined, the highest levels of CXCR4 were noted on myeloid cells (particularly on neutrophils and inflammatory macrophages) as well as terminally differentiated B-cells (notably on plasma cells and pre-plasmablasts), an observation that is consistent with the expression patterns reported in the literature (2–4, 19). Although we detected an increase in CXCR4 expression on memory CD4+ T-cells isolated from mice with lupus, these result did not reach statistical significance; this trend, however, is consistent with our previous studies using microarray analysis of splenic CD4+ T-cells which revealed significantly increased CXCR4 mRNA in B6.Sle1Yaa, compared to B6 (18).
Given the increased expression of CXCR4 on myeloid cells and terminally differentiated B-cells, it appears most likely that altered trafficking patterns of these two cell types may be contributing to disease in lupus. With respect to the myeloid cells, the enhanced trafficking of these cells to the kidneys may be a key contributor to the heightened nephritis seen in these mice. This notion is supported by the observation that kidneys from lupus-afflicted mice exhibit profound increases in the absolute numbers of CXCR4+ myeloid cell infiltrates, as well as the ligand for CXCR4, CXCL12 (Figures 5 and and6).6). This was bolstered by the finding that blocking CXCR4 reverses the infiltration of myeloid cells into the kidneys and accompanying renal inflammation (Figures 7–9, Table 2). Among the myeloid cell subtypes that are most likely responsible for the renal pathology in these disease models are neutrophils and inflammatory monocytes, based on their highest CXCR4 expression levels, and previous literature reports (20–21). Indeed, the role of intra-renal CXCL12 in recruiting inflammatory cells into the kidneys has been elegantly demonstrated in other experimental models of nephritis (11, 20).
The potential mechanisms through which CXCR4+ B-cells may be contributing to lupus are less obvious. Besides the enhanced chemotaxis towards CXCL12 (Figure 4A), the heightened CXCR4 levels on B-cells is also likely to confer prolonged survival advantage to these presumably autoreactive B-cells (Figure 4B). One can envision a scenario in which this could contribute to a breach in peripheral B-cell tolerance. The prolonged survival of autoreactive germinal center B-cells and plasma cells, as a consequence of increased CXCL12-triggered signaling could potentially lead to increased autoantibody levels (21–23). This model is consistent with the observation that blocking CXCR4/CXCL12 reduces autoantibody levels, particularly in the preventative study. Potentially, the increased CXCR4 levels on plasmablasts may facilitate increased trafficking of these cells to the kidneys and promote their survival within intrarenal niches; however, we did not find any increase in CXCR4+ plasma cells (or B-cells) within B6.Sle1Yaa kidneys (data not shown). Whether the CXCR4+ plasma cells in lupus might have homed to yet other (i.e., non-renal) niches is a question that warrants further study. Finally, since CXCR4 levels are tightly regulated in different subsets of germinal center B-cells (3, 5), a further possibility is that the heightened CXCR4 levels observed on lupus B-cells may lead to altered trafficking and affinity maturation patterns within the germinal centers—a possibility that needs to be formally evaluated.
Given the observation that the heightened expression of CXCR4 on myeloid and B-cells may have biological consequences that facilitate disease pathogenesis, a key question that arises is the potential therapeutic utility of CXCR4 blockade in lupus. Indeed, there has been an isolated report on the therapeutic benefit of targeting its ligand, CXCL12: Balabanian et al showed that treatment of New Zealand Black / New Zealand White (NZB/W) mice with anti-CXCL12 antibody ameliorated several lupus phenotypes, although in their study, the analyses were focused on the role of peritoneal B1a cells (24). Given that CXCR4 also plays an important role in AIDS and tumor metastases, there has been a flurry of research reports based on novel CXCR4 blocking agents (25–28). This includes CTCE-9908 from Chemokine Therapeutics Corp, which has proven effective in separate studies in osteosarcoma and prostate cancer (29, 30). Preliminary phase I/II data supporting the inhibitory potential of this agent in solid tumors was recently presented at the AACR-NCI-EORTC Molecular Targets and Cancer Therapeutics International Conference in October 2007 (Abstract A153). Given the demonstrated safety and reported efficacy of this agent in the completed and ongoing human clinical trials, this appeared to be a suitable drug choice for testing in lupus. Indeed, the early administration of the peptide antagonist of CXCR4 prevented all serological, cellular and clinical manifestations of lupus, indicating that all component lupus phenotypes (and the associated pathogenic events) are absolutely dependent upon CXCR4 expression, early in the disease process. This might relate to the critical requirement for CXCR4 hyper-expression on B-cells and myeloid cells, in order to initiate autoantibody production and renal disease.
In contrast, when CXCR4 blockade is instituted late in disease, well after the onset of proteinuria and elevated serum autoantibodies, most of the cellular changes associated with lupus in the B6.Sle1Yaa model were recalcitrant to treatment. However, late-phase therapy was still able to curtail the progression of renal disease, with attendant prolongation of lifespan in these mice. These observations have important ramifications. First, they suggest that the dominant cause of death in these mice is nephritis, rather than splenomegaly (or associated hematological abnormalities). Second, they indicate that renal disease in lupus can be divorced from systemic cellular and serological changes—in other words, mice with severe splenomegaly and high autoantibody levels can live a “normal” lifespan if the renal disease is therapeutically controlled. Third, in both the preventative and treatment studies, CXCR4 blockade dampened tubulo-interstitial disease and crescent formation (rather than glomerulonephritis), 2 histopathological phenotypes associated with poor prognosis in lupus nephritis. Finally, these studies raise hope that instituting CXCR4 blockade in patients with active lupus nephritis might also be therapeutically effective.
We would like to acknowledge Drs. Borna Mehrad, Srividya Subramanian, Alice Chan, Yuyang Fu, Laurie Davis and Zoran Kurepa for technical assistance and helpful discussions.
1This work was supported by NIH PO1 AI 039824, the Arthritis Foundation, the O’Brien Kidney Research Center, P30 DK079328, and an NIH training grant support to AW.