A Defect in COPI-Mediated Transport of STING Causes Immune 2 Dysregulation in COPA Syndrome

bioRxiv preprint doi: https://doi.org/10.1101/2020.05.20.106500; this version posted May 20, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 Title: A defect in COPI-mediated transport of STING causes immune 2 dysregulation in COPA syndrome

3 4 Authors: Zimu Deng1†, Zhenlu Chong1†, Christopher S. Law1, Kojiro Mukai2, Frances O. Ho1, 5 Tereza Martinu3, Bradley J. Backes1, Walter L. Eckalbar1, Tomohiko Taguchi2, Anthony K. 6 Shum1,4,*

7 Affiliations: 8 9 1Department of Medicine, University of California San Francisco, San Francisco, CA.

10 2Laboratory of Organelle Pathophysiology, Department of Integrative Life Sciences, Graduate

11 School of Life Sciences, Tohoku University, Sendai, Japan

12 3Toronto Lung Transplant Program, University Health Network, University of Toronto, Toronto,

13 ON, Canada

14 4Cardiovascular Research Institute, University of California San Francisco, San Francisco, CA.

15 *Corresponding author: [email protected]

16 † equal contribution

17 Pathogenic COPA variants cause a Mendelian syndrome of immune dysregulation

18 with elevated type I interferon signaling1,2. COPA is a subunit of coat protein complex I

19 (COPI) that mediates Golgi to ER transport3. Missense mutations that disrupt the COPA

20 WD40 domain impair binding and sorting of proteins targeted for retrieval to the ER but

21 how this causes disease remains unknown1,4. Given the importance of COPA in Golgi-ER

22 transport, we speculated that type I interferon signaling in COPA syndrome involves

23 missorting of STING. Here we show that a defect in COPI transport due to mutant COPA

24 causes ligand-independent activation of STING. Furthermore, SURF4 is an adapter

25 molecule that facilitates COPA-mediated retrieval of STING at the Golgi. Activated STING

26 stimulates type I interferon driven inflammation in CopaE241K/+ mice that is rescued in

27 STING-deficient animals. Our results demonstrate that COPA maintains immune

28 homeostasis by regulating STING transport at the Golgi. In addtion, activated STING

29 contributes to immune dysregulation in COPA syndrome and may be a new molecular target

30 in treating the disease.

31 COPA syndrome is a genetic disorder of immune dysregulation caused by missense

32 mutations that disrupt the WD40 domain of COPA1. COPA is a subunit of coat protein complex I

33 (COPI) that mediates retrograde movement of proteins from the Golgi apparatus to the

34 endoplasmic reticulum (ER)3. Prior studies have shown that alterations to the COPA WD40

35 domain lead to impaired binding and sorting of proteins bearing a Carboxyl-terminal dilysine motif

36 as well as a defect in retrograde Golgi to ER transport1,4. To date, the molecular mechanisms of

37 COPA syndrome remain unknown including whether missorted proteins are critical for initiating

38 the disease.

39 A clue to the pathogenesis of COPA syndrome recently arose with the observation that

40 type I interferon signaling appears to be highly dysregulated in the disease2. This led us to

41 investigate whether COPA syndrome shares features with any of the well-described Mendelian

42 interferonopathy disorders5. COPA syndrome manifests similarly to the type I interferonopathy

43 STING-associated vasculopathy with onset in infancy (SAVI). Both diseases present at early age

44 with interstitial lung disease, activation of type I interferon stimulated genes (ISG) and evidence

45 of capillaritis6,7. STING is an ER-localized transmembrane protein involved in innate immune

46 responses to cytosolic nucleic acids. After binding cyclic dinucleotides STING becomes activated

47 as it translocates to the ER-Golgi intermediate compartment (ERGIC) and Golgi. At the

48 ERGIC/Golgi, STING forms multimers and activates the kinase TBK1 which subsequently

49 phosphorylates the transcription factor IRF3 to induce expression of type I interferons and other

50 cytokines8. In SAVI, gain of function mutations cause STING to aberrantly exit the ER and traffic

51 to the Golgi and become activated9. Prior work has suggested that COPI may be involved in

52 STING transport at the Golgi, but this is not well established and the molecular interactions

53 between COPI and STING remain unknown8,10. Because COPA plays a critical role in mediating

54 Golgi to ER transport, we hypothesized that activation of type I interferon signaling in COPA

55 syndrome involves missorting of STING.

56 To examine this, we assessed lung fibroblasts from a COPA syndrome patient to determine

57 if there was evidence of STING activation. We measured mRNA transcript levels of several

58 interferon stimulated genes and found they were significantly elevated in comparison to healthy

59 control lung fibroblasts in the presence or absence of a STING agonist (Fig. 1a and Extended Data

60 Fig. 1a, b). Confocal microscopy of COPA syndrome fibroblasts revealed prominent co-

61 localization of STING with the ERGIC (Fig. 1b). Western blots of cellular protein lysates showed

62 an increase in STING multimerization (Fig. 1c) consistent with localization of STING at the

63 ERGIC/Golgi and also higher levels of phosphorylated TBK-1 (pTBK1) and phosphorylated

64 STING (pSTING) (Fig. 1d, Extended Data Fig. 1c), indicative of STING activation11. These data

65 suggest that the elevated type I ISGs observed in COPA syndrome patients may be caused by

66 spontaneous activation of STING.

67 To establish the specific role of mutant COPA in triggering STING pathway activation, we

68 transduced cells with retroviral vectors encoding EGFP-STING and then transfected them with

69 plasmids encoding wild-type or E241K mutant COPA. We performed confocal microscopy to

70 examine whether mutant COPA caused STING to localize on the ERGIC/Golgi, similar to what

71 we observed in patient fibroblasts. In cells expressing wild-type COPA, EGFP-STING was

72 normally distributed throughout the cytoplasm8 whereas in cells with E241K mutant COPA,

73 EGFP-STING co-localized with the Golgi marker GM130 (Fig. 2a). We next reconstituted

74 HEK293T cells (which lack endogenous STING) with retroviral vectors encoding EGFP-STING

75 and then transfected cells with wild-type or E241K mutant COPA. We found that even in the

76 absence of a STING agonist, cells demonstrated a significant increase in pTBK1 (Fig. 2b) and

77 higher levels of mRNA transcripts encoding IFNB1 and type I ISGs (Fig. 2c). Importantly, all of

78 these increases were abolished in HEK293T cells without EGFP-STING, indicating that mutant

79 COPA activates TBK1 and type I interferon signaling specifically through STING rather than other

80 innate immune pathways such as TRIF or MAVS (Fig 2b, c)12. Taken together, these data show

81 that mutant COPA causes ligand-independent activation of STING on the Golgi with upregulation

82 of type I interferon signaling.

83 We hypothesized that retention of STING on the Golgi might reflect a failure of STING to

84 be taken up into mutant COPA containing COPI complexes for transport to the ER. To evaluate

85 this, we analyzed the protein-protein interaction between COPA and STING. We performed co-

86 immunoprecipitation experiments and found that although we could pull-down STING with wild-

87 type COPA the amount of STING that co-immunoprecipitated with E241K mutant COPA was

88 substantially reduced (Fig. 2d). We previously showed that disease-causative COPA mutations

89 cause a defect in binding between the COPA WD40 domain and C-terminal dilysine motif (e.g.

90 KKxx and KxKxx) of proteins targeted for retrieval to the ER1. Because STING lacks a dilysine-

91 tag, we hypothesized that an adaptor protein mediates the interaction between STING and COPA.

92 A review of published studies uncovered seven STING interacting partners that contain a C-

93 terminal dilysine motif (Extended Table 1)13-16. Among these, SURF4 was the only protein shown

94 to cycle between the ER and Golgi via COPI3,17 and function as a cargo receptor18. Thus, we

95 speculated that SURF4 was a likely candidate for mediating an interaction between COPA and

96 STING. We confirmed through co-immunoprecipitation assays that SURF4 associates with

97 STING and COPA (Fig. 2e and Extended Data Fig. 2a) and then mutated the C-terminal dilysine

98 motif of SURF4 by replacing the lysines at positions -3, -4 and -5 from the C-terminus with serines

99 (SURF4-SSS). The amount of SURF4-SSS pulled down with COPA was significantly less in

100 comparison to wild-type SURF4, demonstrating the importance of the dilysine motif to the

101 SURF4-COPA interaction (Fig. 2e). Consistent with this, we found that the association between

102 SURF4 and mutant COPA was markedly reduced in comparison to wild-type COPA, reflecting a

103 defect in binding between the mutant COPA WD40 domain and SURF4 di-lysine tag (Fig. 2e).

104 Finally, as further evidence that SURF4 functions as an adapter molecule for STING and COPA,

105 loss of SURF4 led to a reduction in the amount of STING that co-immunoprecipitated with wild-

106 type COPA (Extended Data Fig. 2b) and also led to an increase in transcript levels of Ifnb1 and

107 type I ISGs (Extended Data Fig. 2 c). In aggregate, our data suggests that SURF4 functions as a

108 cargo receptor for STING and that mutant COPA is unable to bind SURF4 and incorporate STING

109 into COPI vesicles. This results in retention of STING on the Golgi where it becomes

110 spontaneously activated and triggers type I interferon signaling.

111 We next turned to a mouse model of COPA syndrome to understand how mutant COPA-

112 mediated STING activation causes immune dysregulation in vivo. We previously reported that

113 CopaE241K/+ mice which express one of the same disease causative mutations as patients

114 spontaneously develop activated cytokine-secreting T cells and T cell-mediated lung disease19.

115 Through bone marrow chimera and thymic transplant experiments, we showed that mutant COPA

116 within thymic epithelial cells perturbs thymocyte development and leads to a defect in immune

117 tolerance. Because STING is highly expressed in thymic tissue20, we wondered how missorting of

118 STING due to mutant COPA might contribute to the T cell phenotypes we observed in CopaE241K/+

119 mice.

120 To evaluate this, we first confirmed that CopaE241K/+ mice demonstrate retention of

121 activated STING on the Golgi with elevated type I interferon signaling, similar to what we

122 observed in patient cells and our transfection assays (Fig. 3a-c). We next performed bulk RNA

123 sequencing to identify gene expression programs that were altered in thymic epithelial cells. Genes

124 involved in response to viruses were highly enriched in the set of genes differentially expressed

125 between CopaE241K/+ and wild-type mice in medullary thymic epithelial cells (mTEC) (Extended

126 Data Fig. 3). Among the differentially expressed genes, we found significant upregulation of Ifnb

127 and several type I ISGs, consistent with STING activation (Fig. 4a). Thymocytes migrating

128 through the thymic epithelium were impacted by higher Ifnb levels because they exhibited an

129 elevated type I interferon signature (Fig. 4b) and higher levels of Qa2 (Fig. 4c), a cell surface

130 marker expressed in response to IFN-b21. We examined thymocyte populations using a staining

131 protocol that subsets increasingly mature single positive (SP) cells into semi-mature (SM), mature

132 1 (M1) and mature 2 (M2) populations (Fig. 4d). Prior studies found that type I interferons are

133 required during the transition of thymocytes from SM to M2 cells21. In CopaE241K/+ mice, we found

134 a significant increase in the proportion of M2 cells (Fig. 4d), suggesting that higher IFN-b levels

135 promoted an expansion of late stage thymocytes. To determine the role of STING in these changes,

136 we crossed CopaE241K/+ mice to STING-deficient Sting1gt/gt mice. We measured mRNA transcripts

137 of thymic epithelial cells for Ifnb1 and type I ISGs and found that all of the increases returned to

138 wild-type levels in CopaE241K/+xSting1gt/gt mice (Fig. 4e). In addition, loss of STING abrogated all

139 of the thymocyte phenotypes caused by mutant COPA including the elevated type I IFN signature,

140 increased Qa2 expression and higher proportion of M2 cells (Fig. 4c-e). Taken together, these data

141 indicate that activated STING in the thymus of CopaE241K/+ mice perturbs T cell development by

142 elevating type I interferons to promote late stage thymocyte maturation.

143 We next focused on peripheral lymphoid organs to examine whether mutant COPA-

144 mediated STING-activation contributed to systemic inflammation in CopaE241K/+ mice. We

145 reasoned that if STING were to have a significant role in systemic disease, this would provide an

146 opportunity to test whether targeting STING activation dampens systemic inflammation in COPA

147 syndrome patients. Splenocytes from CopaE241K/+ mice had a significant increase in type I ISGs

148 that completely reverted to wild-type levels in CopaE241K/+xSting1gt/gt mice (Extended Data Fig.

149 4a). An analysis of peripheral T cell populations revealed that STING deficiency reversed the

150 significant increase in activated effector memory cells and cytokine-secreting T cells caused by

151 mutant COPA (Fig. 5a and Extended Data Fig. 4b-d). Finally, in strong support of STING being a

152 critical mediator of disease in COPA syndrome pathogenesis, we found that loss of STING rescued

153 embryonic lethality of homozygous CopaE241K/E241K mice (Extended Table 2). In aggregate, these

154 data indicate that STING contributes to immune dysregulation in CopaE241K/+ mice and that

155 targeting STING in COPA syndrome may be an important therapy for patients.

156 Activation of STING requires palmitoylation at the Golgi22 and recent studies report that

157 inhibition of activation-induced palmitoylation of STING with small molecules prevents

158 multimerization of the protein and recruitment of downstream signaling partners23. We treated

159 splenocytes from CopaE241K/+ mice with C-176 mouse STING inhibitor and found that this was

160 highly effective at dampening activation of type I ISGs (Fig. 5b). We then took peripheral blood

161 mononuclear cells (PBMCs) from a COPA syndrome patient or healthy control subject and treated

162 them with the small molecule human STING inhibitor H-151. In parallel, we treated PBMCs with

163 a JAK-STAT inhibitor since drugs in this class have recently been reported as a clinical therapy

164 for COPA syndrome24. Although both the JAK-STAT inhibitor and STING inhibitor reduced type

165 I ISGs, only the STING inhibitor was able to cause a decrease in the levels of the upstream cytokine

166 IFN-b (Fig. 5c, d). Our results suggest that targeting STING activation with small molecule

167 inhibitors might be an effective treatment for COPA syndrome patients with increased efficacy

168 over current therapies.

169 This work establishes COPA as a critical regulator of STING transport that maintains

170 immune homeostasis by mediating retrieval of STING from the Golgi. Defects in COPA function

171 cause STING to multimerize and become spontaneously activated at the Golgi even in the absence

172 of STING ligand, suggesting that at steady state low levels of STING continuously cycle between

173 the ER and Golgi. In support of our findings, Mukai et. al used cultured cells to show that

174 retrograde transport by COPI is essential to maintaining STING in its dormant state independent

175 of cGAMP synthase (cGAS)28. Although we used a candidate approach to identify SURF4 as a

176 putative cargo receptor that engages STING for incorporation into COPI vesicles by COPA, Mukai

177 et. al directly tested 18 STING-binding proteins containing C-terminal dilysine motifs and found

178 that knockdown of SURF4 was the only one that caused STING to localize on the Golgi28. Taken

179 together, our data demonstrates an important role for SURF4 in mediating the retrieval of STING

180 from the Golgi by COPI and provides new insight into the steady state dynamics of STING

181 transport in resting cells.

182 Missense mutations in COPA that lie in the WD40 domain impair retrograde transport of

183 a broad range of COPI cargo proteins containing C-terminal dilysine motifs1. Despite this, our data

184 suggests that impaired trafficking of STING in particular is key to the pathogenesis of COPA

185 syndrome, since not only does loss of STING reverse many of the immunological derangements

186 in our mouse model, it also strikingly rescues embryonic lethality of homozygous mutant mice.

187 Further study should be undertaken to determine whether other COPA cargo contribute to COPA

188 syndrome pathogenesis independent of STING.

189 The ubiquitous expression of COPA has made it difficult to identify the cell types that are

190 responsible for causing disease particularly since COPA is not enriched in any specific immune or

191 lung cell. The tissue specificity of STING may provide additional insight to understanding COPA

192 syndrome and the organs most affected. One unexpected outcome of our work was the finding that

193 STING has a functional role in thymic stromal tissue. Although thymic epithelial cells are known

194 to be a significant source of type I interferons21,25, the mechanisms regulating interferon secretion

195 in the thymus remain largely unexplored. Interestingly, STING also modulates autophagy8 which

196 in thymic epithelial cells is essential for processing self-antigen peptides for presentation to

197 thymocytes26. Additional research might address whether activated STING alters autophagic

198 function in thymic epithelial cells and if so, whether this impacts T cell selection.

199 Our work indicates that COPA syndrome belongs to a category of diseases defined by

200 STING activation5. Going forward, clinicians may want to compare and contrast clinical features

201 and treatment approaches in COPA syndrome and SAVI to establish optimal care of patients.

202 Understanding the mechanisms by which STING causes interstitial lung disease has the potential

203 to substantially improve outcomes in both disorders6,7. For those with COPA syndrome, the

204 dramatic reversal of immune dysregulation we observed in CopaE241K/+xSting1gt/gt mice provides

205 some hope that small molecule STING inhibition can be an effective molecularly targeted

206 approach for treating this highly morbid disease.

207 Main References:

208 1. Watkin, L. B. et al. COPA mutations impair ER-Golgi transport and cause hereditary

209 autoimmune-mediated lung disease and arthritis. Nature genetics 47, 654–660 (2015).

210 2. Volpi, S. et al. Type I interferon pathway activation in COPA syndrome. Clin Immunol 187,

211 33–36 (2018).

212 3. Adolf, F. et al. Proteomic Profiling of Mammalian COPII and COPI Vesicles. Cell Reports 26,

213 250–265.e5 (2019).

214 4. Eugster, A., Frigerio, G., Dale, M. & Duden, R. COP I domains required for coatomer integrity,

215 and novel interactions with ARF and ARF-GAP. EMBO J. 19, 3905–3917 (2000).

216 5. Uggenti, C., Lepelley, A. & Crow, Y. J. Self-Awareness: Nucleic Acid-Driven Inflammation

217 and the Type I Interferonopathies. Annu Rev Immunol 37, 247–267 (2019).

218 6. Tsui, J. L. et al. Analysis of pulmonary features and treatment approaches in the COPA

219 syndrome. ERJ Open Research 4, 00017–2018 (2018).

220 7. Liu, Y. et al. Activated STING in a vascular and pulmonary syndrome. N Engl J Med 371,

221 507–518 (2014).

222 8. Gui, X. et al. Autophagy induction via STING trafficking is a primordial function of the cGAS

223 pathway. Nature 567, 262–266 (2019).

224 9. Dobbs, N. et al. STING Activation by Translocation from the ER Is Associated with Infection

225 and Autoinflammatory Disease. Cell Host & Microbe 18, 1–13 (2015).

226 10. Ablasser, A. & Hur, S. Regulation of cGAS- and RLR-mediated immunity to nucleic acids.

227 Nature Immunology 21, 17–29 (2020).

228 11. Srikanth, S. et al. The Ca2+ sensor STIM1 regulates the type I interferon response by retaining

229 the signaling adaptor STING at the endoplasmic reticulum. Nature Immunology 20, 152–162

230 (2019).

231 12. Liu, S. et al. Phosphorylation of innate immune adaptor proteins MAVS, STING, and TRIF

232 induces IRF3 activation. Science 347, aaa2630–aaa2630 (2015).

233 13. Lee, M. N. et al. Identification of regulators of the innate immune response to cytosolic DNA

234 and retroviral infection by an integrative approach. Nature Immunology 14, 179–185 (2012).

235 14. Huttlin, E. L. et al. Architecture of the human interactome defines protein communities and

236 disease networks. Nature 545, 505–509 (2017).

237 15. Shang, J. et al. Quantitative Proteomics Identified TTC4 as a TBK1 Interactor and a Positive

238 Regulator of SeV-Induced Innate Immunity. Proteomics 18, 1700403–13 (2018).

239 16. Li, S., Wang, L., Berman, M., Kong, Y.-Y. & Dorf, M. E. Mapping a Dynamic Innate

240 Immunity Protein Interaction Network Regulating Type I Interferon Production. Immunity 35,

241 426–440 (2011).

242 17. Mitrovic, S., Ben-Tekaya, H., Koegler, E., Gruenberg, J. & Hauri, H.-P. The cargo receptors

243 Surf4, endoplasmic reticulum-Golgi intermediate compartment (ERGIC)-53, and p25 are required

244 to maintain the architecture of ERGIC and Golgi. Mol. Biol. Cell 19, 1976–1990 (2008).

245 18. Emmer, B. T. et al. The cargo receptor SURF4 promotes the efficient cellular secretion of

246 PCSK9. Elife 7, 154 (2018).

247 19. Deng, Z. et al. A Defect in Thymic Tolerance Causes T Cell–Mediated Autoimmunity in a

248 Murine Model of COPA Syndrome. The Journal of Immunology 204, ji2000028–2373 (2020).

249 20. Manils, J. et al. Double deficiency of Trex2 and DNase1L2 nucleases leads to accumulation

250 of DNA in lingual cornifying keratinocytes without activating inflammatory responses. Sci Rep 7,

251 11902 (2017).

252 21. Xing, Y., Wang, X., Jameson, S. C. & Hogquist, K. A. Late stages of T cell maturation in the

253 thymus involve NF-[kappa]B and tonic type I interferon signaling. Nature Immunology 17, 565–

254 573 (2016).

255 22. Mukai, K. et al. Activation of STING requires palmitoylation at the Golgi. Nat Commun 7,

256 11932 (2016).

257 23. Haag, S. M. et al. Targeting STING with covalent small-molecule inhibitors. Nature 559, 269–

258 273 (2018).

259 24. Krutzke, S., Rietschel, C. & Horneff, G. Baricitinib in therapy of COPA syndrome in a 15-

260 year-old girl. Eur J Rheumatol 1–4 (2019). doi:10.5152/eurjrheum.2019.18177

261 25. Otero, D. C., Baker, D. P. & David, M. IRF7-dependent IFN-β production in response to

262 RANKL promotes medullary thymic epithelial cell development. J Immunol 190, 3289–3298

263 (2013).

264 26. Nedjic, J., Aichinger, M., Emmerich, J., Mizushima, N. & Klein, L. Autophagy in thymic

265 epithelium shapes the T-cell repertoire and is essential for tolerance. Nature 455, 396–400 (2008).

266 27. UniProt Consortium. UniProt: a worldwide hub of protein knowledge. Nucleic Acids Research

267 47, D506–D515 (2019).

268 28. Mukai, K. et al. Homeostatic regulation of STING by Golgi-to-ER membrane traffic. In

269 preparation.

270

271 Figure 1

272 273 Fig. 1. COPA syndrome patient fibroblasts demonstrate spontaneous STING activation. 274 (A) Real-time PCR was performed for ISG expression in primary lung fibroblasts from a healthy 275 control and COPA syndrome subject. (B) Primary lung fibroblasts from a healthy control and 276 COPA syndrome subject were reconstituted with wild-type EFGP-STING retrovirus. The 277 reconstituted cells were stained with indicated antibody and localization of STING observed using 278 a confocal microscope. ERGIC 53 is an ERGIC marker. (C) Immunoblots of endogenous STING 279 in primary lung fibroblasts from 2 healthy controls and a COPA syndrome subject under non- 280 reducing SDS-PAGE conditions with and without cGAMP (0.5µg/ml) stimulation. (D) 281 Immunoblots of indicated antibodies in primary lung fibroblasts with cGAMP (2µg/ml) 282 stimulation for indicated time (n=2 per group). Data in (A) from two independent experiments 283 represent means ± SD. *p<0.05, **p<0.01, (two-tailed Mann-Whitney U test). Scale bar, 20µm in 284 (B).

285 286

287 Figure 2

288 Fig. 2. Activated STING is retained on the Golgi after failed retrieval of SURF4 by mutant 289 COPA. (A) HeLa cells reconstituted with wild-type EGFP-STING retrovirus and transfected with 290 wild-type COPA or mutant COPA (E241K) for 24 hours. Cells were stained with indicated

291 antibodies and localization of STING observed using a confocal microscope. GM130 is a Golgi 292 marker. (B) HEK293T cells with and without EGFP-STING were transfected with wild-type or 293 mutant COPA (E241K) for 24 hours and then immunoblot performed with indicated antibodies. 294 (C) HEK293T cells with and without EGFP-STING were transfected with wild-type or mutant 295 COPA (E241K) for 24 hours and then real-time PCR performed for ISG expression. (D) FLAG 296 immunoprecipitates (IP) from lysates of HEK293 cells overexpressing FLAG-tagged COPA 297 (WT/E241K) were immunoblotted for indicated antibodies. (E) FLAG immunoprecipitates from 298 lysates of HEK293T cells overexpressing FLAG-tagged COPA (WT/E241K) and HA-tagged 299 SURF4 (WT/mutant) immunoblotted for indicated antibodies. Data in (C) from 2 independent 300 experiments represent means ± SD. **p<0.01 (two-tailed Mann-Whitney U test). Scale bar, 20µm 301 in (A). 302

303 Figure 3

A C

304

305 Fig. 3 CopaE241K/+ mice exhibit STING activation with elevated type I interferon signaling. 306 (A) Real-time PCR was performed for ISG expression in immortalized MEF cells derived from 3 307 WT and 3 CopaE241K/+ mice. (B) Immortalized MEF cells from WT, CopaE241K/+ and 308 CopaE241K/E241K mice were reconstituted with wild-type EGFP-STING retrovirus. The 309 reconstituted cells were stained with the indicated antibodies and localization of STING 310 observed using a confocal microscope. GM130 is a Golgi marker. (C) Immunoblot was 311 performed with the indicated antibodies in splenocytes from WT or CopaE241K/+ mice (n=3 for 312 each genotype). Data in (A, C) were means ± SD. *p<0.05, **p<0.01(unpaired two-tailed 313 Student’s t tests). Scale bar, 20µm in (B). 314

315 Figure 4 316 A B

D C

317 Fig. 4 Activated STING perturbs thymocyte development by increasing type I interferons in 318 the thymus. (A) Medullary thymic epithelial cells sorted from wild-type and CopaE241K mice were 319 used for RNA-seq analysis (n=3 each genotype). Volcano plot of the differentially affected genes 320 (FDR<0.1, red; FDR<0.1, ISG from viral response GO categories, purple). (B) Real-time PCR 321 performed for Ifit1, Rsad2 and Isg15 expression in SP thymocytes (top:CD4SP; bottom:CD8SP). 322 (n=4 mice each genotype, 2 independent experiments). (C) left: Percentages of Qa2high population 323 among SP thymocytes (top: CD4SP; bottom: CD8SP). right: Representative flow analysis of Qa2 324 expression on SP thymocytes (top: CD4SP; bottom: CD8SP). (n=5 each genotype, 3 independent 325 experiments). (D) top: CD69 versus MHCI expression on SP thymocytes (top:CD4SP; 326 bottom:CD8SP). bottom: Percentages of SM and M2 population among SP thymocytes (top: 327 CD4SP; bottom: CD8SP). (WT: n=5 each genotype, 3 independent experiments). (E) Real-time 328 PCR performed for Ifnb1, Ifit1 and Isg15 expression in thymic epithelial cells (CD45- 329 ,Epcam+,Ly51-,MHC-IIhighmTEC) (n≥3 each genotype, 2 independent experiments). Data in (B- 330 F) represent means ± SD. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, ns, not significant 331 (unpaired, parametric, two-tailed student’s t-test). 332

333 Figure 5 A

B C

334

335 Fig. 5 Loss of STING function dampens inflammation caused by mutant COPA (A) Top left: 336 intracellular levels of IFNγ and IL17A in splenic CD4+ T cells after PMA/Ionomycin stimulation. 337 Top right: percentages of IFNγ producing CD4+ T cells. (n≥3 each genotype, >3 independent 338 experiments). Bottom left: intracellular IFNγ production in splenic CD8+ T cells after 339 PMA/Ionomycin stimulation. Bottom right: percentages of IFNγ producing CD8+ T cells. (n≥3 340 each genotype, >3 independent experiments). (B) Real-time PCR for ISG expression in 341 splenocytes harvested from 4 WT mice and 4 CopaE241K/+mice treated with or without STING 342 inhibitor C-176 (10µM) for 24 hours. (C) Real-time PCR performed in duplicate for ISG 343 expression in PBMCs from a healthy control and COPA syndrome subject treated with or without

344 STING inhibitor H-151 (10µM) or (D) JAK inhibitor Tofacitinib (2µM) for 24 hours. Data 345 represent means ± SD. **p<0.01, ***p<0.001, ns, not significant. (unpaired, parametric, two- 346 tailed Student’s t tests). 347

348 Methods:

349 Reagents

350 2′3′-cGAMP and CP-690550 (Tofacitinib) were purchased from InvivoGen. STING

351 inhibitors N-(4-iodophenyl)-5-nitrofuran-2-carboxamide (C176) and 1-(4-ethylphenyl)-3-(1H-

352 indol-3-yl)urea (H151) were synthesized by SYNthesis med chem at or greater than 99% purity

353 by HPLC.

354 Plasmids

355 Human STING and SURF4 were subcloned from HEK293T cDNA into pCMV6-AC

356 (Origene) with a FLAG tag at the C-terminus or a HA tag at the N-terminus, respectively. Plasmids

357 expressing FLAG tagged wild-type and mutant E241K human COPA were previously generated1.

358 Retroviral plasmids expressing EGFP tagged human and mouse STING were kindly provided by

359 Dr. Tomohiko Taguchi (Tohoku University).

360 Study subjects

361 Subjects were selected on the basis of COPA syndrome diagnosis, with their written

362 informed consent, were studied via protocols approved by the Research Ethics Board of Toronto

363 General Hospital and the Institutional Review Boards for the protection of human subjects of

364 Cleveland Clinic or the University of California San Francisco.

365 Isolation of human lung fibroblasts

366 Control lung tissue was harvested from anonymous brain-dead donors from the Northern

367 California Transplant Donor Network. Screening criteria for selection of healthy lung for specimen

368 collection were previously described2. Mutant lung explants were from two COPA syndrome

369 patients receiving lung transplants. Fibroblasts were isolated as described3. In brief, tissue was

370 minced with scissors into 5 mm pieces and digested three times with 0.25% trypsin (GE Healthcare

371 Life Sciences) for 10 minutes at 37°C. The resulting cell and tissue suspension was collected,

372 neutralized with complete media (45% Ham’s F12, 45% DMEM, 10% FBS), pelleted, plated onto

373 FBS coated 60 mm dishes and cultured in 5% CO2 at 37°C. Fibroblasts were ready to passage and

374 use one week following isolation.

375 Isolation of mouse embryonic fibroblasts

376 MEFs were isolated as described4. In brief, day 13.5 embryos were washed with PBS,

377 minced with scissors into 1-2 mm pieces and digested three times with 0.25% trypsin for 10

378 minutes at 37°C. The cell suspension was neutralized with complete media (DMEM, 10% FBS),

379 pelleted, resuspended in complete media, and cultured in 5% CO2 at 37°C.

380 To create immortal MEFs, Phoenix packaging cells (kindly gifted by Dr. Mark Anderson,

381 UCSF) were transfected with pBABE-neo largeTcDNA plasmid (a gift from Bob Weinberg;

382 Addgene plasmid # 1780; http://n2t.net/addgene:1780; RRID:Addgene_1780) 5, and viral

383 supernatant was collected two days later. Early passage MEFs were incubated in viral supernatant

384 for two days and then transformed cells were selected with G418 (Teknova).

385 siRNA knockdown

386 Predesigned siRNA oligomers (ON-TARGETplus SMARTPool) for SURF4 and

387 transfection control (siGENOME RISC-Free) were obtained from Dharmacon and resuspended in

388 RNase-free water at 10 µM. siRNAs were transfected in Opti-Mem media (Life Technologies) for

389 48 h with Lipofectamine RNAiMAX (Life Technologies) according to manufactuer’s protocol.

390 Immunoblotting and antibodies

391 Cells were lysed in Cold Spring Harbor NP40 lysis buffer (150 mM NaCl, 50 mM Tris pH

392 8.0, 1.0% Nonidet-P40) supplemented with protease and phosphatase inhibitors (PMSF, NaF,

393 Na3VO4, Roche PhosSTOP) and then centrifuged at 12,000g for 15 mins at 4°C to get cellular

394 lysate. Equal amounts of protein were loaded and size separated on an SDS–PAGE gel and wet

395 transfered onto PVDF membrane. The membrane was blocked in TBS-T buffer with 5% milk for

396 1 h at room temperature, followed by overnight incubation with primary antibodies diluted in TBS-

397 T with 5% BSA. Membrane was washed 3 times with TBS-T buffer for 10 mins, incubated at room

398 temperature with HRP-conjugated IgG secondary antibody (Jackson Immunoresearch), washed 3

399 times with TBS-T for 10 mins followed once with TBS buffer for 10 mins, and then developed

400 with SuperSignal West Femto Chemiluminescent Substrate (Life Technologies).

401 Rabbit antibodies against TBK1, p-TBK1 (Ser172), STING, p-STING (Ser365), p-STING

402 (Ser366), FLAG, HA and GFP were from Cell Signaling Technology. Rabbit antibody against

403 SURF4 was from Novus. Mouse antibodies against GAPDH and β-Tubulin were from Santa Cruz

404 Biotechnology. Mouse antibody against GM130 was from BD Biosciences.

405 Co-immunoprecipitation

406 HEK293 or HEK293T cells transfected with indicated plasmids were collected, washed

407 once with PBS, lysed in NP40 lysis buffer supplemented with protease and phosphatase inhibitors

408 and centrifuged at 12,000g for 15 min at 4°C. The supernatant was mixed with FLAG-M2 beads

409 (Sigma) and incubated overnight at 4°C. Ten percent of lysate was saved as input. The following

410 day, the beads were washed 3 times with IP washing buffer (10 mM Tris, pH 7.4, 1 mM EDTA,

411 150 mM NaCl and 1% Triton X-100) for 10 mins. 2× SDS loading buffer (100 mM Tris pH 6.8,

412 4% SDS, 20% glycerol, 10% β-mercaptoethanol, 0.1% bromophenol blue) was added into the IP

413 complex and boiled at 95 °C for 5 min. Samples were analyzed by immunoblot as described above.

414 Confocal microscopy

415 Cells were seeded onto glass coverslips and treated as indicated. Cells then were fixed with

416 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 and blocked with 3% BSA. Slides

417 were incubated with primary antibody overnight at 4°C, incubated with fluorescent-conjugated

418 secondary antibody for 1 h at room temperature and followed with DAPI incubation for 10 mins

419 at room temperature. Slides were then mounted with FluorSave Reagent (Millipore) and kept at

420 4°C in the dark. Images were captured with a Leica TCS SPE microscope.

421 Mouse antibody against ERGIC-53 was from Santa Cruz Biotechnology.

422 Mice strains

423 CopaE241K knock in mice were generated in our lab6. C57BL/6J-Sting1gt/J (Stinggt/gt) mice

424 were purchased from the Jackson Laboratory. All mice were maintained in the specific pathogen

425 free (SPF) facility at UCSF, and all protocols were approved by UCSF’s Institutional Animal Care

426 and Use Committee (IACUC).

427 Flow cytometry and antibodies

428 Single cell suspension of thymocytes and splenocytes were prepared by mechanically

429 disrupting the thymus and spleen. Cells were filtered through 40 µm filters (Genesee Scientific)

430 into 15 ml conical tubes and maintained in RPMI 1640 containing 5% FBS on ice. Splenocytes

431 were further subjected to red blood cells lysis (Biolegend). For evaluation of surface receptors,

432 cells were blocked with 10 µg/ml anti-CD16/32 for 15 minutes at room temperature and then

433 stained with indicated antibodies in FACS buffer (PBS, 2% FBS) on ice for 40 minutes.

434 For intracellular cytokine detection, freshly isolated splenocytes were stimulated by PMA

435 (Sigma) and ionomycin (Sigma) in the presence of brefeldin A (Biolegend) for four hours. Cells

436 were collected and stained with Ghost Dye Violet 450 (Tonbo Biosciences) followed by surface

437 staining. Cells were then washed, fixed at room temperature for 15 minutes (PBS, 1% neutral

438 buffered formalin), permeabilized and stained with antibodies against cytokines on ice (PBS, 2%

439 FBS, 0.2% saponin). All samples were acquired on LSRFortessa or FACSVerse (BD Bioscience)

440 and analyzed using FlowJo software V10.

441 Antibodies to the following were purchased from Biolegend: CD8 (53-6.7), IFNγ

442 (XMG1.2), Epcam (G8.8); from eBioscience: IL17A (eBio17B7), IL13 (eBio13A), MHCII (I-A)

443 (NIMR-4), Qa2 (69H1-9-9); from Tonbo Biosciences: CD4 (GK1.5), MHCII (I-A/I-E)

444 (M5/114.15.2); The CD16/CD32 (2.4G2) antibody was obtained from UCSF’s Monoclonal

445 Antibody Core.

446 Thymic epithelial cell isolation

447 Thymic epithelial cells were isolated as described7. Briefly, thymi from 4-week-old mice

448 were minced with razor blades into 1-2 mm3 pieces. The tissue fragments were collected and

449 enzymatically digested three times (DMEM, 2% FBS, 100 µg/ml DNase I and 100 µg/ml

450 Liberase™). The single-cell suspensions from each digestion were pooled into 20 ml of MACS

451 buffer (PBS, 0.5% BSA, 2 mM EDTA) on ice. Total cells were washed and centrifugated using a

452 three-layer Percoll gradient with specific gravities of 1.115, 1.065 and 1.0. Thymic stromal cells

453 were enriched and collected from the Percoll-light fraction, between the 1.065 and 1.0 layers.

454 Thymic stromal cells were washed, stained and sorted to isolate MHC-IIhighCD80high mTEC

455 (mTEChigh) by FACS.

456 RNA isolation and quantitative real-time PCR analysis

457 RNA was isolated with the EZNA Total RNA kit (Omega Bio-tek) was reverse transcribed

458 to cDNA with SuperScript III Reverse Transcriptase and oligo d(T)16 primers (Invitrogen).

459 Quantitative real-time PCR was performed on Bio-Rad CFX thermal cyclers with TaqMan Gene

460 Expression assays from Life Technologies (GAPDH, Hs02786624_g1; IFNB1, Hs01077958_s1;

461 IFI6, Hs00242571_m1; IFI27, Hs01086373_g1; IFI44L, Hs00915292_m1; ISG15,

462 Hs01921425_s1; IFIT1, Hs03027069_s1; IFITM1, Hs00705137_s1; RSAD2, Hs00369813_m1;

463 Gapdh, Mm99999915_g1; Ifnb1, Mm00439552_s1; Ifit1, Mm00515153_m1; Ifi44l,

464 Mm00518988_m1; Isg15, Mm01705338_s1; Rsad2, Mm00491265_m1; S1pr1: Mm00514644).

465 RNA-Seq and data analysis:

466 RNA-sequencing libraries by first using the Nugen Ovation method (Tecan 7102-A01) to

467 create cDNA from the isolated RNA, the sequencing library was then created from cDNA using

468 the Illumina Nextera XT method (Illumina FC-131-1096). All libraries were combined and

469 sequenced on Illumina HiSeq4000 lanes, yielding approximately 300 million single end, 50bp

470 reads. Sequencing reads were then aligned to the mouse reference genome and the ensemble

471 annotation build (GRCm38.78) using STAR8 (v2.4.2a). Read counts per gene were used as input

472 to DESeq29 (v1.26.0) to test for differential gene expression between conditions using a Wald test.

473 Genes passing a multiple testing correct p-value of 0.1 (FDR method) were considered significant.

474 Pathway analysis was carried out using DAVID10 and the R/Bioconductor package

475 RDAVIDWebService11 (v1.24.0).

476 Statistical analysis

477 All statistical analysis was performed using Prism 7 (GraphPad Software) or R 4.0.0 (R

478 Foundation for Statistical Computing). Where indicated, two-tailed Mann-Whitney U-test and

479 unpaired, parametric, two-tailed student’s t-test were used to evaluate the statistical significance

480 between two groups, and p <0.05 was considered statistically significant.

481 Methods References:

482 1. Watkin, L. B. et al. COPA mutations impair ER-Golgi transport and cause hereditary

483 autoimmune-mediated lung disease and arthritis. Nature genetics 47, 654–660 (2015).

484 2. Lee, J. W., Fang, X., Gupta, N., Serikov, V. & Matthay, M. A. Allogeneic human

485 mesenchymal stem cells for treatment of E. coli endotoxin-induced acute lung injury in

486 the ex vivo perfused human lung. Proceedings of the national academy of Sciences 106,

487 16357–16362 (2009).

488 3. Ramos, C. et al. Fibroblasts from idiopathic pulmonary fibrosis and normal lungs differ in

489 growth rate, apoptosis, and tissue inhibitor of metalloproteinases expression. Am J Respir

490 Cell Mol Biol 24, 591–598 (2001).

491 4. Durkin, M. E., Qian, X., Popescu, N. C. & Lowy, D. R. Isolation of Mouse Embryo

492 Fibroblasts. Bio Protoc 3, (2013).

493 5. Hahn, W. C. et al. Enumeration of the simian virus 40 early region elements necessary for

494 human cell transformation. Mol. Cell. Biol. 22, 2111–2123 (2002).

495 6. Deng, Z. et al. A Defect in Thymic Tolerance Causes T Cell–Mediated Autoimmunity in a

496 Murine Model of COPA Syndrome. The Journal of Immunology 204, ji2000028–2373

497 (2020).

498 7. Miller, C. N. et al. Thymic tuft cells promote an IL-4-enriched medulla and shape

499 thymocyte development. Nature 559, 627–631 (2018).

500 8. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21

501 (2013).

502 9. Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion

503 for RNA-seq data with DESeq2. Genome Biol. 15, 550–21 (2014).

504 10. Huang, D. W. et al. The DAVID Gene Functional Classification Tool: a novel biological

505 module-centric algorithm to functionally analyze large gene lists. Genome Biol. 8, R183–

506 16 (2007).

507 11. Fresno, C. & Fernández, E. A. RDAVIDWebService: a versatile R interface to DAVID.

508 Bioinformatics 29, 2810–2811 (2013).

509 Acknowledgements: Suzy A. A. Comhair and Serpil C. Erzurum from Cleveland Clinic

510 Foundation for help providing lung explants. Paul Wolters, Tien Peng, Chaoqun Wang and N.

511 Soledad Reyes de Mochel at UCSF for help with providing lung explants. David J. Erle for

512 discussions and assistance with RNA-sequencing. Mark S. Anderson for discussions and review.

513 Funding: UCSF Program for Breakthrough Biomedical Research, funded in part by the Sandler

514 Foundation (AKS), NIH/NIAID R01AI137249 (AKS), NIH/NHLBI R01HL122533 (AKS),

515 NIH/NHLBI R00HL135403 (WLE), NIH/NHLBI PO1HL103453 (SAAC, SCE). Author

516 contributions: ZD and ZC designed and performed experiments, analyzed data. KM and TT

517 provided technical advice. CSL performed experiments and analyzed data. FOH performed

518 experiments. BJB assisted with small molecule experimental design. WLE analyzed RNA-

519 sequencing data. TM provided lung explant tissue. AKS directed the study and wrote the

520 manuscript. Competing interest declaration: Authors declare no competing interests. Data and

521 materials availability: All data that support the findings of this study are available from the

522 corresponding author upon reasonable request. RNA-sequencing data will be uploaded into a

523 public respository and accession codes and unique identifiers will be provided.

524

525 Extended Data: 526 527 Extended Data Figure 1 528

529 530 531 Extended Data Fig. 1 532 COPA syndrome patient fibroblasts demonstrate STING activation. (A) Real-time PCR was 533 performed for IFNB1 expression in primary lung fibroblasts from a healthy control and COPA 534 syndrome subject treated with cGAMP (2µg/ml) for indicated time. Data pooled from 2 535 independent experiments are means ± SD, two-tailed Mann-Whitney U test. (B) Real-time PCR 536 for ISG expression in primary lung fibroblasts from a healthy control and COPA syndrome subject 537 treated with cGAMP (2µg/ml) for indicated time. Data are means ± SD, unpaired two-tailed 538 Student’s t test. (C) Immunoblots of primary lung fibroblasts for indicated antibodies. **p<0.01, 539 ***p<0.001, ****p<0.0001. 540

541 Extended Data Figure 2 542

543 544 545 Extended Data Fig. 2 546 SURF4 functions as a cargo receptor that mediates retrieval of STING by COPA. (A) FLAG 547 immunoprecipitates (IP) from lysates of HEK293T cells overexpressing FLAG-tagged STING and 548 HA-tagged SURF4 were immunoblotted for indicated antibodies. (B) HEK293 cells were 549 transfected with FLAG-tagged COPA and si-SURF4 for 48 hours. FLAG immunoprecipitates (IP) 550 from lysates of HEK293 cells were immunoblotted for indicated antibodies. (C) Real-time PCR 551 for ISG expression in immortalized MEF cells with EGFP-STING after si-SURF4 for 48 hours. 552 Data represent means ± SD. *p<0.05, **p<0.01 (unpaired two-tailed Student’s t tests). siNeg = 553 non-targeting siRNA control. 554

555 Extended Data Figure 3 556 557

558 559 560 Extended Data Fig. 3 561 DAVID Gene Ontology analysis of bulk RNA sequencing in thymic epithelial cells. DAVID 562 Gene Ontology analysis comparing differentially expressed genes in MHC-IIhighCD80high mTECs 563 from WT and COPAE241K/+ mice. Top GO clusters with the key word are shown. (n=3 of each 564 genotype). 565

566 Extended Data Figure 4 A

567 568

569 Extended Data Fig. 4 570 Activated STING in CopaE241K/+ mice results in type I interferon driven inflammation. 571 (A) Real-time PCR performed for ISG expression in splenocytes from WT (n=5), CopaE241K/+ 572 (n=5), Stinggt/gt (n=5) and CopaE241K/+xStinggt/gt (n=6) mice. (B) left: Representative flow plots 573 showing expression of CD62L versus CD44 on T cells (top: CD4+ T cells; bottom: CD8+ T cells). 574 right: Percentages of naive and effector memory T cells (left: CD4+ T cells; right: CD8+ T cells). 575 (WT: n=6; CopaE241K/+: n=4; Stinggt/gt: n=5; CopaE241K/+xStinggt/gt: n = 5). (C) Left: intracellular 576 levels of IL13 in splenic CD4+ T cells after PMA/Ionomycin stimulation. right: percentages of 577 IL13 producing CD4+ T cells. (WT: n=6; CopaE241K/+: n=4; Stinggt/gt: n=5; CopaE241K/+xStinggt/gt: 578 n=5, > 3 independent experiments). (D) Left: intracellular TNFα production in splenic CD8+ T 579 cells after PMA/Ionomycin stimulation. right: percentages of TNFα producing CD8+ T cells. (WT: 580 n=4; CopaE241K/+: n=3; Stinggt/gt: n=5; CopaE241K/+xStinggt/gt: n=5, >3 independent experiments). 581 Data in (A-D) represent means ± SD. *p<0.05, **p<0.01, ****p<0.0001 (unpaired two-tailed 582 Student’s t tests). 583

584 Extended Table 1 585

Subcellular Protein C terminus Function location

CCDC47 KQIKVKAM ER Calcium regulation protein DDOST MKEKEKSD ER Glycosyltransferase enzyme ER, Plasma HM13 KGLEKKEK ER-resident intramembrane protease membrane MATR3 ERRQKKET Nucleus RNA binding protein SAC1 LVQKEKID ER, Golgi Phosphoinositide phosphatase Catalytic subunit of oligosaccharyl STT3B KKISKKTV ER transferase SURF4 MDEKKKEW ER, ERGIC, Golgi Cargo receptor 586 587 Extended Table 1: 588 589 STING interacting partners containing a C-terminal dilysine motif. Seven STING interacting 590 partners with C-terminal dilysine motifs were identified in published studies describing STING 591 affinity purification-mass spectrometry (AP-MS) data13-16,27. Shown for each protein are its C- 592 terminus sequence with dilysine motif highlighted, subcellular location and function. 593

594 Extended Table 2 Loss of STING rescues embryonic lethality of Copa E241K/E241K mice 595 596 597

Stingwt/wt Stinggt/gt

Neonates from 5 litters Neonates from 5 litters

Copa +/+ 7 (24%) 11(22%)

Copa E241K/+ 22 (76%) 22(44%)

Copa E241K/E241K 0 (0%) 17 (34%)

598 599 600