Cyclin-dependent kinase inhibitor Cdkn2c deficiency promotes B1a cell expansion and autoimmunity in a mouse model of lupus

Abstract

The lupus-prone NZM2410 mice present an expanded B1a cell population that we have mapped to the Sle2c1 lupus susceptibility locus. The expression of Cdkn2c, a gene encoding for cyclin-dependent kinase inhibitor p18(Ink4c) and located within Sle2c1, is significantly lower in B6.Sle2c1 B cells than in B6 B cells. To test the hypothesis that the B1a cell expansion in B6.Sle2c1 mice was due to a defective p18 expression, we analyzed the B1a cell phenotypes of p18-deficient C57BL/6 mice. We found a dose-dependent negative correlation between the number of B1a cells and p18 expression in B cells, with p18-deficient mice showing an early expansion of the peritoneal B1a cell pool. p18 deficiency enhanced the homeostatic expansion of B1a cells but not of splenic conventional B cells, and the elevated number of B6.Sle2c1 B1a cells was normalized by cyclin D2 deficiency. These data demonstrated that p18 is a key regulator of the size of the B1a cell pool. B6.p18(-/-) mice produced significant amounts of anti-DNA IgM and IgG, indicating that p18 deficiency contributes to humoral autoimmunity. Finally, we have shown that Sle2c1 increases lpr-associated lymphadenopathy and T cell-mediated pathology. B6.p18(-/-).lpr mice showed a greater lymphadenopathy than B6.Sle2c1.lpr mice, but their renal pathology was intermediate between that of B6.lpr and B6.Sle2c1.lpr mice. This indicated that p18-deficiency synergizes, at least partially, with lpr-mediated pathology. These results show that Cdkn2c contributes to lupus susceptibility by regulating the size of the B1a cell compartment and hence their contribution to autoimmunity.

Figure 1

p18 deficiency increases the size of the Pc B1a cell compartment. The absolute number and percentage of B220^int CD5⁺ IgM⁺ B1a cells, the ratio of B220^int CD5⁺ IgM⁺ B1a / B220^hi CD5^- IgM⁺ B2 cells, and the percentage of IgM^hi CD5⁺ B220⁺ cells were compared between 2 month old (A) and 5-7 month old (B) B6, B6.Sle2c1 and B6.p18^-/- mice. Data were compared with the Dunn's Multiple Comparison Tests. For clarity, p ≤ 0.0001 significance levels between B6 and B6.p18^-/- values are not shown when the difference between B6 and B6.Sle2c1 was significant at p ≤ 0.001. Representative Facs plots showing the B1a (plain line) and B2 (broken line) gates (C) and the IgM^hi CD5⁺ population in B220⁺ gated cells (D) in each of the three strains. E. Correlation between Cdkn2c expression in splenic B cells and the percentage of Pc B1a cells. Each dot represents a single mouse. The P value indicates the significance of a Pearson's correlation test. F. Percentage of Pc B220^int CD5⁺ IgM⁺ B1a cells from 5 month old mice. Statistical significance is shown to a two-tailed t test. *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Figure 2

P18 deficiency results in the production of autoAbs. A, Serum total IgM, anti-ssDNA IgM, total IgG, and anti-dsDNA IgG in 7-12 months old B6, B6.Sle2c1 and B6.p18^-/- mice. B, Representative ANA IgG staining of Hep-2 cells incubated with serum from each of the three strains. The B6.Sle2c1 sample shows a specked pattern, and the B6.p18^-/- sample shows a nuclear pattern. C, ANA pattern distribution (C: cytoplasmic, S: speckled, N: nuclear) and staining intensity in pixels among the mice in each strain that showed a positive ANA stain. D. ELIPSOT analyses of total IgM production by Sp B cells (left) and anti-ssDNA IgM production by Pc (P) and Sp (S) B cells (right). E. IgM secretion by Pc B1a or Sp B cells purified from B6 and B6.p18^-/- mice, cultured for 5 d in medium alone, with LPS or PMA. Each dot represents a single mouse. For A, C and D, data were compared with Mann-Whitney tests between B6 and B6.Sle2c1 or B6.p18^-/- . In E, t tests were used to compare the two strains for each condition. *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001.

Figure 3

P18 deficiency results in the homeostatic expansion of Pc B1a cells. A, Absolute numbers of Pc CD5⁺ B220^lo IgM⁺ B1a cells before (D: donors) and after (R: recipients) transfer into B6.Rag1^-/- mice according to the donor cells’ strain of origin. Data were compared using paired t tests. B, Fold expansion of the Pc B cells 3 weeks after transfer according to the donor cells’ strain of origin. Expansion was calculated for total IgM⁺ cells, B1a: CD5⁺ B220^int IgM⁺ cells, B1b: CD5^- B220^int IgM⁺ cells, and B2: CD5^- B220^hi IgM⁺ cells. C, Serum IgM in the recipient B6.Rag1^-/- mice according to the donor cells’ strain of origin. For B and C, data were compared with t tests between B6 and B6.Sle2c1 or B6.p18^-/-. *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001.

Figure 4

Cyclin D2 deficiency abrogates Sle2c1-induced Pc B1a cell expansion. The absolute number and percentage of B220^int CD5⁺ IgM⁺ B1a cells, the ratio of B220^int CD5⁺ IgM⁺ B1a / B220^hi CD5^- IgM⁺ B2 cells, and the percentage of IgM^hi CD5⁺ B220⁺ cells were compared between 2 month old (A) and 5-7 month old (B) B6, B6.Sle2c1, B6.Sle2c1.D2^-/- and B6.D2^-/- mice. Data were compared with the Dunn's Multiple Comparison Tests. *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Figure 5

P18 deficiency synergizes with the lpr mutation to induce T cell hyperplasia and activation. A. Survival of B6.p18^-/-.lpr (N = 36) and B6.Sle2c1.lpr (N = 84) mice, which were sacrificed when their cervical lymph node mass reached 1.5 cM, displayed open sore dermatitis, or were found sick. B. Spleen and total lymph node weights at sacrifice. C. Percentage of CD3⁺ lymphocytes in the lymph nodes. Percentage of CD3⁺ T cells expressing B220 (D) and DN T cells (E). G. Expression of Foxp3 and CD44 on CD3⁺ CD4⁺ gated cells, with a graph showing the percentage of Foxp3⁺ CD4⁺ T cells. For D-F, a representative Facs plot is shown for each strain with the gates on interest, and values for the whole cohorts are shown on the right. Mice were between 4 and 6 month old. Data (B-G) were compared with the Dunn's Multiple Comparison Tests. *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Figure 6

P18 deficiency preferentially enhances the in vivo proliferation of lpr T cells. Five month old B6.lpr and B6.p18^-/-.lpr mice were exposed to BrdU for either 18 h (A) or 7 d (B-C). BrdU incorporation was measured in splenic (A-B) and lymph node (C) DN, CD4⁺ T cells or CD3^- B220⁺ B cells. Data were compared with t tests. *: p ≤ 0.05, **: p ≤ 0.01. BrdU incorporation IN the lymph nodes after 18 h exposure was low and not different between the two strains (data not shown).

Figure 7

B6.p18^-/-.lpr mice develop a level of renal pathology intermediate between B6.lpr and B6.Sle2c1.lpr. A, Representative histograms showing intracellular IL-17 in CD3⁺ CD4⁺ T cells from B6.p18^-/-.lpr (black) and B6.lpr (filled gray) spleens. The horizontal bar shows the gate used in panels B-D. Percentage of splenic CD3⁺ CD4⁺ IL-17⁺ (B), splenic CD3⁺ DN IL-17⁺ (C), and renal CD3⁺ DN IL-17⁺ (D) lymphocytes. E, Anti-dsDNA IgG present in the sera of the three strains of Fas-deficient mice. F, Percentages of dsDNA IgG⁺ mice in these cohorts. G, Distribution of GN scores between the three strains (N = 13-15 mice per strain), with either mesangial expansion (M) or proliferative global (Pg) scores on a 1-4 scale. The B6.p18^-/-.lpr and B6.Sle2c1.lpr distributions were compared with a chi-square test. H, Percentage of mice showing interstitial chronic inflammation (ICI). All mice were 4-5 month old. In F and H, data were compared with two-tailed Fisher exact tests. **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.