Mycobacterial catalase-peroxidase is a tissue antigen and target of the adaptive immune response in systemic sarcoidosis

Abstract

Sarcoidosis is a disease of unknown etiology characterized by noncaseating epithelioid granulomas, oligoclonal CD4(+) T cell infiltrates, and immune complex formation. To identify pathogenic antigens relevant to immune-mediated granulomatous inflammation in sarcoidosis, we used a limited proteomics approach to detect tissue antigens that were poorly soluble in neutral detergent and resistant to protease digestion, consistent with the known biochemical properties of granuloma-inducing sarcoidosis tissue extracts. Tissue antigens with these characteristics were detected with immunoglobulin (Ig)G or F(ab')(2) fragments from the sera of sarcoidosis patients in 9 of 12 (75%) sarcoidosis tissues (150-160, 80, or 60-64 kD) but only 3 of 22 (14%) control tissues (all 62-64 kD; P = 0.0006). Matrix-assisted laser desorption/ionization time of flight mass spectrometry identified Mycobacterium tuberculosis catalase-peroxidase (mKatG) as one of these tissue antigens. Protein immunoblotting using anti-mKatG monoclonal antibodies independently confirmed the presence of mKatG in 5 of 9 (55%) sarcoidosis tissues but in none of 14 control tissues (P = 0.0037). IgG antibodies to recombinant mKatG were detected in the sera of 12 of 25 (48%) sarcoidosis patients compared with 0 of 11 (0%) purified protein derivative (PPD)(-) (P = 0.0059) and 4 of 10 (40%) PPD(+) (P = 0.7233) control subjects, suggesting that remnant mycobacterial catalase-peroxidase is one target of the adaptive immune response driving granulomatous inflammation in sarcoidosis.

Figure 1.

Protease-resistant proteins in sarcoidosis tissues are targets of circulating IgG from patients with sarcoidosis. TX100-insoluble detergent extracts from sarcoidosis (S) and control (C) lymph nodes were treated with or without PK and subjected to denaturing SDS-PAGE and immunoblot analysis with purified IgG or F(ab′)₂ fragments from pooled control or sarcoidosis sera detected with anti–human Fcγ or Fab reagents, respectively. Four separate immunoblots (A–D) with samples corresponding to Table I are shown. Immunoblots in A and B were probed with control reagents, stripped, and then reprobed with sarcoidosis reagents. Numbers give molecular mass in kilodaltons (kD). Arrows demark antigenic bands in PK-treated sarcoidosis tissues when probed with sarcoidosis serologic reagents but not control reagents or secondary antibody alone.

Figure 2.

Sarcoidosis tissues extracted with sarkosyl contain poorly soluble protein antigens that bind to sarcoidosis IgG. (A) Protein fractions from sarcoidosis (S) or control (C) spleen. (B) Protein fractions from sarcoidosis lung. Protein fractions were analyzed by protein immunoblotting using purified IgG from sarcoidosis sera detected with anti–human Fcγ reagents. The initial homogenates (P22), intermediate supernatant fractions (S215, S1, and S2), or final cell pellets that were poorly soluble in sarkosyl (Pe) are shown. Pe were solubilized using 8 M urea plus β-mecaptoethanol and were precipitated before analysis by gel electrophoresis. Arrows point to antigenic bands seen in the Pe fractions of sarcoidosis but not control tissues. Arrowhead points to a 56-kD band detected with anti-Fcγ reagents alone in the Pe fraction from sarcoidosis lung (B).

Figure 3.

Immunoblot analysis confirms the presence of mKatG protein in sarcoidosis tissue. Shown are individual protein immunoblots of recombinant mKatG, sarkosyl-extracted sarcoidosis (S11), or control (C14) spleen extracts or TX100-insoluble sarcoidosis (S2) or control (C2) lymph node extracts using the IT57 anti-mKatG mAb. Arrowheads demark migration of corresponding mol wt. markers.

Figure 4.

Cellular localization of mkatG DNA by ISH with TSA. Shown are representative photomicrographs from biopsies of Mtb-infected tissue (A–C), normal controls (no pathology; D–F), Wegener granulomatosis (G–I), and sarcoidosis tissues (J–O). Tissue sections were hybridized with or without DIG-labeled probes, detected with anti-DIG F(ab′)₂ fragments conjugated with HRP, and the signal was amplified by incubation with biotinylated tyramine followed by streptavidin conjugated to HRP and a DAB chromogen solution. Mtb-infected tissue (A) and sarcoidosis samples (J and M) demonstrated focal collections of mkatG DNA that were not seen in biopsies from Wegener granulomatosis (G) nor in nongranulomatous control tissues (D). All tissues were negative using reverse mkatG and H. pylori 16S rRNA probes (C, inset, positive H. pylori control).

Figure 5.

Cellular localization of Mtb 16S rRNA DNA by ISH. Shown are representative photomicrographs from biopsies of Mtb-infected tissue (A and B), normal controls (no pathology; C and D), Wegener granulomatosis (E and F), and sarcoidosis tissues (G–J). Tissue sections were hybridized with or without DIG-labeled probes, detected with anti-DIG F(ab′)₂ fragments conjugated with alkaline phosphatase, and developed with Vector Red substrate. Mtb-infected tissue (A) and sarcoidosis samples (G and I) demonstrated focal collections of Mtb 16S rRNA DNA that were not seen in Wegener granulomatosis (E) nor in nongranulomatous control tissues (C). All tissues were negative using reverse Mtb 16S rRNA probes.

^{Figure 6.}

Circulating IgG from patients with sarcoidosis binds to recombinant mKatG and mHsp65 protein. Protein immunoblots with recombinant mKatG (A) or mHsp65 (B) were analyzed with IgG purified from the sera of healthy PPD⁻ (N), PPD⁺ (P) subjects or sarcoidosis (S) patients corresponding to Table IV. Arrowheads demark migration of mol wt markers (top to bottom: 188, 62, 49, and 38 kD). Arrows demark ∼80- and ∼60-kD antigenic bands reactive to mKatG or a 65-kD band reactive to mHsp65.