B cells from periodontal disease patients express surface Toll-like receptor 4

Abstract

Chronic systemic inflammation links periodontal disease (PD) to increased incidence of cardiovascular disease. Activation of TLRs, particularly TLR4, promotes chronic inflammation in PD by stimulating myeloid cells. B cells from healthy individuals are generally refractory to TLR4 agonists as a result of low surface TLR4 expression. Unexpectedly, a significantly increased percentage of gingival and peripheral blood B cells from patients with PD expressed surface TLR4. Surface expression correlated with an active TLR4 promoter that mimicked the TLR4 promoter in neutrophils. B cells from PD patients were surface myeloid differentiation protein 2-positive and also packaged the enhancer of a proinflammatory cytokine, IL-1 beta, into an active structure, demonstrating that these cells harbor key characteristics of proinflammatory cell types. Furthermore, B cells lacked activating signatures of a natural IL-1 beta inhibitor, IL-1 receptor antagonist. Surprisingly, despite multiple signatures of proinflammatory cells, freshly isolated B cells from PD patients had decreased expression of TLR pathway genes compared with B cells from healthy individuals. Decreases in inflammatory gene expression were even more dramatic in B cells stimulated with a TLR4 ligand from a periodontal pathogen, Porphyromonas gingivalis LPS 1690. In contrast, B cell TLR4 was not activated by the prototypic TLR4 ligand Escherichia coli LPS. These findings raise the unexpected possibility that TLR4 engagement modulates B cell activation in PD patients.

Fig. 1.

TLR4 is up-regulated on B cells from patients with PD. Fresh whole blood or isolated B cells from tonsillitis tonsils, appendicitis appendices, or PD gingiva were analyzed by flow cytometry for surface TLR4 expression. (A) Representative plots from one blood sample from a PD patient or healthy donor or gingival B cells from a PD patient as indicated. Boxes indicate the percentage of B cells that expresses TLR4 and were drawn using isotype control as background (right panel for gingival tissue, and data not shown for whole blood). (B) Composite data showing percent TLR4+ B cells in the peripheral blood of healthy (left; n=19) or PD (right; n=17) donors. The two cohorts have significantly different percentages of TLR4+ B cells in circulation (*, P=0.0074, by nonparametric Mann Whitney U test). (C) Composite data showing percentage of TLR4+ B cells in tonsillitis (n=10), appendicitis (n=5), and PD gingival tissue (n=7). Healthy gingival tissue was not available for analysis. The difference in TLR4 expression on B cells from PD tissues versus tonsil or appendix is highly significant (*, P=0.001, by the Kruskal-Wallis test). Percentage of TLR4+ B cells is similar between tonsil and appendix (P>0.05).

Fig. 2.

B cells from PD patients have molecular signatures of TLR4 activation. (A) TLR4 expression on CD3+ T cells from a representative PD patient of the three patients analyzed. (B) A representative plot of purified B cells used for biochemical analyses of the TLR4 gene in panels C and D. B cells were routinely devoid of significant contamination with monocytes, as shown, and CD3+ T cells (not shown). FACS plot is representative of multiple, independent B cell purifications. In no case were B cells less than 98% pure. (C) The TLR4 promoter of purified monocytes or B cells from healthy or PD patients was treated with the indicated units of MNase prior to analysis by CHART-PCR. Slightly lower accessibility in B cells from PD patients versus healthy donors is unlikely to be biologically significant compared with the complete inaccessibility of the promoter in human embryonic kidney (HEK) 293 cells and similar accessibility in monocytes. Results show averages and range of three independent determinations. (D) Analysis of TLR4/transcription factor binding in B cells from PD patients by ChIP using antibodies recognizing proteins as indicated. We conservatively designate twofold enrichment as an indication of association. his, Histidine. *, Significant difference (P<0.05 by Student’s t-test). (E) Analysis of TLR4/transcription factor binding in neutrophils from healthy donors by ChIP using antibodies recognizing proteins as indicated. (D) Average and se of four to eight independent determinations. (E) Average and range of two independent determinations.

Fig. 3.

B cells from PD patients have additional signatures of cells that modulate inflammation. (A) MD-2, the protein that bridges TLR4 with ligand [25], is detectable on the surface of TLR4+ B cells, as analyzed by flow cytometry. Virtually all B cells in PD patients that were TLR4+ were also MD-2+. Shown is a representative of three independent analyses. (B) ChIP assays show transcription factor association with the IL-1β enhancer in B cells from the indicated donors (x-axis). IRF-4 does not associate with an unrelated region of the IL-1β gene that lacks a consensus-binding sequence (not shown). *, Differences in IRF-4/enhancer association between B cells from PD versus healthy donors are highlighted (P<0.05 by Student’s t-test). (C) Transcription factor association with the IL-1β promoter in B cells from PD patients. Bars are averages and se of greater than or equal to three independent determinations for B and C. (D) IL-1β promoter accessibility to MNase (2 U) in cell types indicated above each bar. Shown are average and range of greater than or equal to two independent determinations. (E and F) Association between the transcription factor PU.1 and the natural competitive inhibitor of IL-1β, IL-ra, in (E) B cells from PD patients or (F) neutrophils from healthy donors. Shown are averages and ranges of two to three independent determinations from the same ChIP samples amplified in B and C and Figure 2, D and E.

Fig. 4.

B cells from PD patients constitutively up- and down-regulate genes in TLR pathways. (A) Superarray plate analysis comparing unstimulated B cells from three PD patients and three healthy donors (for a total of nine comparisons). Shown are genes up-regulated (black bars), unchanged (gray bars), or down-regulated (white bar) in B cells isolated from PD peripheral blood, purified and then rested for 1 h on ice prior to analysis. (B) Gene expression in B cells from PD patients versus healthy donors incubated ex vivo in media alone for 24 h. Shown is the average and sd of values obtained from two donor samples from each cohort (for a total of four comparisons). *, Statistical significance (P<0.05) by the Student’s t-test. Additional genes that approached significant differences (P<0.08) were IRF-1 and cJun (A) and MyD88 and NF-κB-1 (B). IKK-beta, IκB kinase; AGMX1, Bruton’s tyrosine kinase.

Fig. 5.

Cellular responses to TLR4 ligands, as measured by changes in gene expression and NF-κB pathway activation downstream of TLR. (A and B) Western blots quantifying IκB-α levels in B cells from PD patients (A) or healthy donors (B), treated as indicated. E. coli LPS was added at 1 μg/ml. The proteosome inhibitor MG132 blocks IκB-α degradation in activated cells. Percentage of TLR4+ B cells was 39.3% or 15.1% for samples from PD or healthy donors, respectively. Caspase-1 (Casp-1) was the loading control (lower panels). Results are representative of three independent analyses. (C and D) TLR signaling Superarray analysis of B cells from PD patients (white bars) or healthy donors (black bars) incubated with the TLR4-activating P. gingivalis LPS 1690 [31] (C, 10 μg/ml for 24 h) versus media alone or P. gingivalis LPS 1690 versus a combination of P. gingivalis LPS 1690 + P. gingivalis LPS 1435/1449 [31] (D, 10 μg/ml each for 24 h). Genes are arranged as in Figure 4A for comparison. Average and range of results from two donor samples are shown. Negative values indicate that mRNA levels in P. gingivalis LPS 1690-treated cells are lower than mRNA levels in cells treated with (C) media or (D) P. gingivalis LPS 1690 + P. gingivalis LPS 1435/1449. Genes whose expression was variably changed by the TLR4 ligand are not shown. A twofold difference was defined as a change in gene expression in these assays, as indicated by the dotted lines. (E) Monocyte response to P. gingivalis LPS 1690. Mono-Mac-6 human monocytes were stimulated with 10 ng/ml E. coli LPS or 10 μg/ml P. gingivalis LPS 1690 ± P. gingivalis LPS 1435/1449 for 4 h as indicated by the key. Monocytes were then analyzed for IL-1β mRNA levels by quantitative PCR. Shown are average and sd of three independent determinations.