Interruption of the CXCL13/CXCR5 Chemokine Axis Enhances Plasma IgM Levels and Attenuates Atherosclerosis Development

Atherosclerosis is a chronic inflammatory disease typified by the subendothelial accumulation of lipids and inflammatory cells.[1] With disease progression, lesions can develop into fibroatheromatous lesions, prone to rupture causing thrombus formation, which is the main cause of cardiovascular complications like myocardial infarction and stroke.[2] Chemokines play an important role in the development and progression of atherosclerosis.[3] In the past decade, multiple other functions, beyond the classical chemotaxis, have been reported like cell survival and angiogenesis, and platelet activation.[4] [5] [6] [7]

CXCL13 is a homeostatic chemokine, highly expressed by stromal tissue and follicular dendritic cells, where it is instrumental in the guidance of lymphocytes to secondary lymphoid organs.[8] [9] [10] Its only known receptor, CXCR5, is also implicated in lymphocyte homing.[10] The CXCL13-CXCR5 chemokine axis has been associated with various inflammatory conditions.[9] [11] [12] Surprisingly, we could show that CXCL13, although being a homeostatic chemokine, is prominently expressed in human atherosclerotic lesions and is increased in plasma of patients with carotid atherosclerosis.[13] Together, these data infer a potential causative role for the CXCL13-CXCR5 chemokine axis in atherosclerosis.

In this study, we aimed to address the impact of defective hematopoietic CXCL13-CXCR5 chemokine signaling on atherosclerosis formation in mice. Hereto, we generated LDLr^−/− chimeras (female, 12 weeks) with hematopoietic deficiency of either Cxcl13 or Cxcr5 by bone marrow transplantation. Mice transplanted with C57Bl/6J bone marrow (wild-type [WT]) were used as controls. All animal work was approved by the regulatory authority of Maastricht University and performed in compliance with the Dutch government guidelines. After bone marrow reconstitution, mice were placed on a Western type diet (WD; 0.25% cholesterol) for 4 weeks, after which a semiconstrictive carotid artery collar was placed bilaterally to accelerate atherogenesis[14] and mice were left for an additional 8 weeks on WD ([Fig. 1A]). The right common carotid artery was paraformaldehyde fixed and paraffin-sectioned (4 μm; at 100 μm intervals), after which they were stained by hematoxylin/eosin and analyzed for lesion size and composition (Macrophages: Mac-3, BD-Pharmingen; Collagen content: Sirius Red staining).

Fig. 1 The proatherogenic role of the CXCL13-CXCR5 dyad. (A) Female, 12-week-old, LDLr^−/− mice were irradiated with 9 Gy and subsequently transplanted with bone marrow cells from C57Bl/6J (wild-type [WT]), CXCR5^−/−, or CXCL13^−/− mice. After 6 weeks' recovery with antibiotic treatment, mice were fed a Western type diet in total 12 weeks. After 4 weeks, semiconstrictive collars were placed. (B) Quantification of hematoxylin and eosin stained carotid arteries (n = 11–13). (C) Quantification of lesional MAC-3⁺ cell counts, expressed as percentage of total plaque cells (DAPI stained) (n = 11–13). (D) Plaque collagen content (Sirius Red staining), expressed as percentage of total plaque area (n = 11–13). (E) Necrotic core area (defined as anucleated area) as percentage of total plaque area (n = 11–13). (F) Quantification of hematoxylin and eosin stained carotid arteries (n = 11). (G) Quantification of lesional MAC-3⁺ cell counts, expressed as percentage of total plaque cells (DAPI stained) (n = 11). (H) Plaque collagen content (Sirius Red staining), expressed as percentage of total plaque area (n = 11). (I) Necrotic core area (defined as anucleated area) as percentage of total plaque area (n = 11). (J) Blood flow cytometry analysis. B1a were identified as CD19⁺IgM⁺B220^lowCD5⁺CD23⁻, B1b as CD19⁺IgM⁺B220^lowCD5⁻CD23⁻, B2 as CD19⁺B220^high (n = 12). (K) Lymph node flow cytometry analysis. B1a were identified as CD19⁺IgM⁺B220^lowCD5⁺CD23⁻, B1b as CD19⁺IgM⁺B220^lowCD5⁻CD23⁻, follicular [Foll.] B2 as CD19⁺IgM^lowB220⁺CD5⁻CD23⁺ (n = 6). (L) Bone marrow flow cytometry analysis. PrePro were identified as CD11b⁻B220^highCD24⁻CD43⁺, Pro were identified as CD11b⁻B220^highCD24⁺CD43⁺, immature were identified as CD11b⁻B220^highCD24⁺CD43⁻CD127^low (n = 12). (M) Spleen flow cytometry analysis. B1a were identified as CD19⁺CD138⁻CD21⁻ CD23⁻B220^lowIgM⁺CD5⁺, B1b as CD19⁺CD138⁻CD21⁻ CD23⁻B220^lowIgM⁺CD5⁺, follicular [Foll.] B2 as CD19⁺CD138⁻IgM^lowB220⁺CD5⁻CD23⁺, marginal zone [MZ] B2 as CD19⁺CD138⁻IgM^lowB220⁺CD5⁻CD23⁻CD21⁺ (n = 12). (N) Plasma total-immunoglobulin M (IgM) levels (n = 12). (O) Plasma oxidized low-density lipoprotein (oxLDL)-IgM levels (n = 12). Data represent mean ± standard error of the mean (SEM). D'Agostino–Pearson omnibus normality test was applied; *p < 0.05; **p < 0.01; ***p < 0.001, as analyzed by Student's t-test with Welsh correction or Mann–Whitney test, as appropriate. (P) Regional Manhattan plots showing the association between variants in the CXCL13 locus (Chr4:78382907–78582988) and ischemic stroke (gray circles) in the CADISP cohort (n = 9,814). The genomic regions are displayed along the x-axis with known genes aligned below. p < 1.4 × 10⁻⁴ after Bonferroni correction for variant counts (dashed line) was considered statistically significant.

Hematopoietic deficiency of Cxcr5 did neither affect body weight (20.0 g ± 0.4 vs. 20.4 g ± 0.9 for WT mice) nor plasma cholesterol levels (1067 pg/mL ± 67.8 vs. 1085 pg/mL ± 89.6 for WT mice). However, atherosclerotic lesion development in the carotid artery was significantly reduced in hematopoietic Cxcr5 ^−/− mice, compared with WT controls ([Fig. 1B]). While plaque macrophage content did not differ ([Fig. 1C]), lesional collagen content was decreased ([Fig. 1D]) as was necrotic area ([Fig. 1E]). These data clearly point to a causal role of hematopoietic CXCR5 in atherosclerosis development.

As CXCL13 is expressed both by stromal cells and various leukocyte subsets, we addressed the question whether the reduced atherogenic response in Cxcr5 ^−/− mice was driven by hematopoietic or by stromal CXCL13. Hematopoietic Cxcl13 deficient mice showed no difference in body weight (21.3 g ± 0.7 vs. 20.4 g ± 0.9 for WT mice) and plasma cholesterol levels (895.2 pg/mL ± 73.7 vs. 1085 pg/mL ± 89.6 for WT mice). Similar to hematopoietic Cxcr5 ^−/− deficiency, that of Cxcl13 also resulted in a significantly reduced lesion burden ([Fig. 1F]). Both lesional macrophage and collagen content was unchanged in hematopoietic Cxcl13 deficient mice ([Fig. 1G, H]). Similar to Cxcr5 ^−/− chimeras, hematopoietic Cxcl13 deficiency resulted in a striking decrease in necrotic core size ([Fig. 1I]).

Interestingly, B1 cells were significantly enriched in blood of both Cxcr5 ^−/− and Cxcl13 ^−/− mice, which was, for Cxcl13 ^−/− mice, accompanied by an additional significant increase of B2 cells ([Fig. 1J]). This aligned with a reduced B1 contents in the lymph nodes ([Fig. 1K]), while B1 and B2 contents in the spleen ([Fig. 1M]) were not affected by interruption of the CXCL13-CXCR5 chemokine axis (data not shown); nor were B cell progenitors in bone marrow affected by either of the deficiencies ([Fig. 1L]). B1 and B2 were reported to exert divergent, even opposite, effects on atherosclerosis, at which B1 cells were shown to attenuate atherosclerosis, while B2 cells appear to aggravate disease.[15] Although the exact role of B1b cells in atherosclerosis has for long been unclear, it was recently shown that they are, similarly to B1a cells, atheroprotective. The exact mechanism for their protective activity seems to differ from that of B1a cells,[16] though for both subsets it is mainly mediated by the production and release of immunoglobulin M (IgM).[17] [18] In support of this notion, levels of total ([Fig. 1N]; ProcartaPlex Immunassays, Invitrogen) as well as atheroprotective oxidized low-density lipoprotein-specific IgM ([Fig. 1O]) in serum were indeed sharply elevated in both Cxcr5 ^−/− and Cxcl13 ^−/− mice compared with WT mice.

These results clearly demonstrate that the homeostatic CXCL13-CXCR5 chemokine axis plays a causal role in atherosclerosis development, most likely by modulation of B1 cell distribution patterns to influence atheroprotective IgM production and secretion by these cells. Hematopoietic Cxcr5 ^−/− and Cxcl13 ^−/− also impacted on lesion composition, with a striking decrease in necrotic core content, potentially reflecting the delayed lesion development in Cxcl13/Cxcr5 deficient mice, carrying less advanced lesions. As our model reflects early stages of lesion development, we were unable to explore whether the antiapoptotic properties of CXCL13 could potentially result in beneficial plaque stabilizing effects within unstable lesions.[13] Though this aspect regarding unstable lesions remains to be elucidated, the observed effects of CXCL13 deficiency on early stage atherosclerotic lesions align very well to the observations made by Kyaw et al,[17] demonstrating that splenectomy reduced peritoneal B1a content and plasma IgM levels, leading to increased atherosclerotic lesion size and necrotic core expansion.

To further substantiate the relevance of the CXCL13-CXCR5 axis to human atherosclerosis, on top of already published associations,[13] we examined the association of genetic variants at the CXCR5 and CXCL13 loci with atherosclerosis-related clinical outcomes. Applying expressive quantitative trait loci analysis in the Genotype-Tissue Expression data set,[19] we identified the G-allele at rs77564610 which is associated with higher CXCR5 expression in whole blood (normalized effect size, +0.530; p = 1.9 × 10⁻⁶) and with increased risk of (subsequent) myocardial infarction (odds ratio [OR], 2.85; p = 1.1 × 10⁻³) in the cohort of the UK Biobank (n = 452,264).[20] Moreover, we scrutinized common variants at the CXCR5 locus (±50 Kb) in the UK Biobank cohort and identified rs187248852 (minor allele frequency [MAF] = 0.4%) and rs73575424 (MAF = 10.6%) to be associated with ischemic stroke (OR, 4.87; p = 8.9 × 10⁻⁶) and myocardial infarction (OR, 1.10; p = 5.1 × 10⁻⁵), respectively. Finally, analysis performed on the CADISP cohort (n = 9,814)[21] [22] yielded suggestive evidence of association with stroke for multiple single-nucleotide polymorphisms in the CXCL13 locus (p < 1.4 × 10⁻⁴), although most of them are intron variants ([Fig. 1P]). Together with evidence showing higher CXCL13 in patients with atherosclerosis,[13] our analyses imply the involvement of CXCR5/CXCL13 axis in human atherosclerosis.

In conclusion, we are the first to describe a proatherogenic role of the CXCL13-CXCR5 dyad in atherosclerosis, an effect that is likely attributable to its impact on B1 cells and their IgM production. However, while we did not observe any difference in splenic B2 cell levels (Follicular/Marginal zone), a role of these cells, via T cell interaction,[23] cannot be fully excluded at this point. Identifying this chemokine axis as novel regulator of atherosclerosis development, this study creates new opportunities for chemokine targeting therapy in atherosclerosis.

• References

• 1 Weber C, Zernecke A, Libby P. The multifaceted contributions of leukocyte subsets to atherosclerosis: lessons from mouse models. Nat Rev Immunol 2008; 8 (10) 802-815
• 2 Weber C, Badimon L, Mach F, van der Vorst EPC. Therapeutic strategies for atherosclerosis and atherothrombosis: past, present and future. Thromb Haemost 2017; 117 (07) 1258-1264
• 3 van der Vorst EP, Döring Y, Weber C. Chemokines and their receptors in atherosclerosis. J Mol Med (Berl) 2015; 93 (09) 963-971
• 4 Gunn MD, Ngo VN, Ansel KM, Ekland EH, Cyster JG, Williams LT. A B-cell-homing chemokine made in lymphoid follicles activates Burkitt's lymphoma receptor-1. Nature 1998; 391 (6669): 799-803
• 5 Luther SA, Bidgol A, Hargreaves DC. , et al. Differing activities of homeostatic chemokines CCL19, CCL21, and CXCL12 in lymphocyte and dendritic cell recruitment and lymphoid neogenesis. J Immunol 2002; 169 (01) 424-433
• 6 Meier R, Mühlethaler-Mottet A, Flahaut M. , et al. The chemokine receptor CXCR4 strongly promotes neuroblastoma primary tumour and metastatic growth, but not invasion. PLoS One 2007; 2 (10) e1016
• 7 Salcedo R, Wasserman K, Young HA. , et al. Vascular endothelial growth factor and basic fibroblast growth factor induce expression of CXCR4 on human endothelial cells: In vivo neovascularization induced by stromal-derived factor-1alpha. Am J Pathol 1999; 154 (04) 1125-1135
• 8 Ansel KM, Harris RB, Cyster JG. CXCL13 is required for B1 cell homing, natural antibody production, and body cavity immunity. Immunity 2002; 16 (01) 67-76
• 9 Bürkle A, Niedermeier M, Schmitt-Gräff A, Wierda WG, Keating MJ, Burger JA. Overexpression of the CXCR5 chemokine receptor, and its ligand, CXCL13 in B-cell chronic lymphocytic leukemia. Blood 2007; 110 (09) 3316-3325
• 10 Förster R, Mattis AE, Kremmer E, Wolf E, Brem G, Lipp M. A putative chemokine receptor, BLR1, directs B cell migration to defined lymphoid organs and specific anatomic compartments of the spleen. Cell 1996; 87 (06) 1037-1047
• 11 Bagaeva LV, Rao P, Powers JM, Segal BM. CXC chemokine ligand 13 plays a role in experimental autoimmune encephalomyelitis. J Immunol 2006; 176 (12) 7676-7685
• 12 Khader SA, Rangel-Moreno J, Fountain JJ. , et al. In a murine tuberculosis model, the absence of homeostatic chemokines delays granuloma formation and protective immunity. J Immunol 2009; 183 (12) 8004-8014
• 13 Smedbakken LM, Halvorsen B, Daissormont I. , et al. Increased levels of the homeostatic chemokine CXCL13 in human atherosclerosis - potential role in plaque stabilization. Atherosclerosis 2012; 224 (01) 266-273
• 14 von der Thüsen JH, van Berkel TJ, Biessen EA. Induction of rapid atherogenesis by perivascular carotid collar placement in apolipoprotein E-deficient and low-density lipoprotein receptor-deficient mice. Circulation 2001; 103 (08) 1164-1170
• 15 Tsiantoulas D, Sage AP, Mallat Z, Binder CJ. Targeting B cells in atherosclerosis: closing the gap from bench to bedside. Arterioscler Thromb Vasc Biol 2015; 35 (02) 296-302
• 16 Srikakulapu P, McNamara CA. B cells and atherosclerosis. Am J Physiol Heart Circ Physiol 2017; 312 (05) H1060-H1067
• 17 Kyaw T, Tay C, Krishnamurthi S. , et al. B1a B lymphocytes are atheroprotective by secreting natural IgM that increases IgM deposits and reduces necrotic cores in atherosclerotic lesions. Circ Res 2011; 109 (08) 830-840
• 18 Rosenfeld SM, Perry HM, Gonen A. , et al. B-1b cells secrete atheroprotective IgM and attenuate atherosclerosis. Circ Res 2015; 117 (03) e28-e39
• 19 Consortium GT. ; GTEx Consortium. Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science 2015; 348 (6235): 648-660
• 20 Canela-Xandri O, Rawlik K, Tenesa A. An atlas of genetic associations in UK Biobank. Nat Genet 2018; 50 (11) 1593-1599
• 21 CDKP. Cerebrovascular Disease Knowledge Portal, NINDS grant #1R24NS09ƒ2983. Available at: http://cerebrovascularportal.org/ . Accessed 27 May, 2019
• 22 Debette S, Kamatani Y, Metso TM. , et al; International Stroke Genetics Consortium; CADISP Group. Common variation in PHACTR1 is associated with susceptibility to cervical artery dissection. Nat Genet 2015; 47 (01) 78-83
• 23 Nus M, Sage AP, Lu Y. , et al. Marginal zone B cells control the response of follicular helper T cells to a high-cholesterol diet. Nat Med 2017; 23 (05) 601-610