TRAP1 chaperone protein mutations and autoinflammation

PBMCs isolated from patient 4 (see main text) and 3 healthy controls (HC) were stained for 15 min with 5 μM MitoSox or 10 μM DHE and analysed immediately by flow cytometry. Dotted lines in (A, B, C, D) denote HC (one representative experiment); red lines denote patient 4. (A, B) Mitochondrial oxide in monocytes (A) showed no significant difference, but (B) was increased in lymphocytes from patient 4 (***P < 0.001, two-tailed Z-test). (C, D) Oxidative stress detected by H2DCFDA was elevated in both patient monocytes (C) and lymphocytes (D). Fig 2E–H: 500,000 THP1 cells were seeded into a 24-well plate with 100 nM PMA. After 24 h, the medium was replaced and lipofectamine and siRNA complexes were added. After a further 24 h, the medium was replaced with RPMI with 2% FCS for another 24 h. (E, F, G, H) Cells were then trypsinized and either stained for 15 min with 5 μM MitoSox (E), 10 μM DHE (F), or 10 μM H2DCFDA (H) and analysed immediately by flow cytometry; or the cells were resuspended in antioxidant buffer, spin probe CMH was then added, and superoxide production was measured 10 times over 10 min via electron spin resonance spectroscopy (G). Cytoplasmic superoxide was significantly increased in MEFV knockdown (kd) cells compared with scrambled control. Mitochondrial superoxide, superoxide production, and oxidative stress levels were significantly increased in the TRAP1 kd cells. (I) THP1 cells were transfected with full-length C-terminal DDK-tagged TRAP1 in a pCMV6-Entry vector, with either wild-type TRAP1 sequence or with the same point mutation inducing the p.R128H protein change or the empty vector alone. 500,000 THP1 cells were seeded into a 24-well plate in RPMI with 10% FCS and 100 nM PMA overnight. Relevant wells were treated with 5 μM CCCP for 6 h. The cells were then stained with 5 μM MitoSox, trypsinized, and analysed immediately by flow cytometry. At baseline, MitoSox fluorescence was slightly but significantly decreased in only the WT-TRAP1 overexpressing cells compared with the vector control. Upon treatment with CCCP, mitochondrial superoxide increased in vector control and R128H mutant TRAP1, but to a significantly lesser extent in the WT-TRAP1 cells. kd, knockdown; WT, wild-type.

(A) XBP1 mRNA is spliced during the unfolded protein response under ER stress. XBP1 splicing was evaluated in patient 4 with qPCR and normalised to the average of three HCs (wild-type WT). There was no significant difference in the level of unspliced XBP1 (uXBP1), but the amount of spliced XBP1 (sXBP1) was significantly increased (P < 0.001, two-tailed Z-test) and thus the ratio of sXBP1/uXBP1 was also increased (P < 0.001, two-tailed Z-test). (B) PBMCs from three HCs and patient 4 were seeded onto polylysine coverslips and incubated with mitotracker deep red to label the mitochondria (grey). These were fixed with methanol and labelled with DAPI (blue), anti-TRAP1 antibodies (green), and anti-calnexin antibodies (red) to stain the ER. Scale bar = 2 μm. TRAP1 showed localization to the ER in WT cells. In patient 4, the mitochondria showed increased association with the ER and TRAP1 and less structural definition.

(A) PBMCs were isolated from TRAPS patients with different mutations and from three HCs (wild-type, WT), fixed, and labelled with the nuclear stain DAPI (blue), anti-TNFR1 antibodies (red), and anti-TRAP1 antibodies (green). Scale bar = 5 μm. (B) The degree of red-green co-localization was quantified in at least three images for each patient and from three HCs (WT). *P < 0.05 **P < 0.01 (unpaired two-tailed t tests).