Professional Documents
Culture Documents
Copa (Inglés)
Copa (Inglés)
Unbiased genetic studies have uncovered surprising molecular might also be anticipated to cause autoimmunity, as disruptions in
mechanisms in human cellular immunity and autoimmunity1. protein trafficking lead to endoplasmic reticulum (ER) stress and
We performed whole-exome sequencing and targeted activation of the unfolded protein response (UPR), both of which
sequencing in five families with an apparent mendelian have been implicated in autoimmune disease10.
syndrome of autoimmunity characterized by high-titer We identified five families with a previously undescribed mendelian
autoantibodies, inflammatory arthritis and interstitial lung syndrome of autoimmunity manifested by high-titer autoantibodies,
disease. We identified four unique deleterious variants in interstitial lung disease and inflammatory arthritis (Fig. 1a–d, Table 1
the COPA gene (encoding coatomer subunit a) affecting the and Supplementary Table 1). The average age of presentation was
same functional domain. Hypothesizing that mutant COPA 3.5 years, with a range of 6 months to 22 years. Several patients pre-
leads to defective intracellular transport via coat protein sented with pulmonary hemorrhage requiring immunosuppression,
complex I (COPI)2–4, we show that COPA variants impair and all patients have lung disease (Table 1 and Supplementary Table 1).
binding to proteins targeted for retrograde Golgi-to-ER A comparison of lung biopsies from unrelated patients identified
transport. Additionally, expression of mutant COPA results lymphocytic interstitial infiltration with germinal center formation
npg
in ER stress and the upregulation of cytokines priming for a (Fig. 1b,c), findings consistent with the interstitial lung disease occur-
T helper type 17 (TH17) response. Patient-derived CD4+ T cells ring in systemic autoimmune syndromes11. Immunohistochemical
also demonstrate significant skewing toward a TH17 phenotype staining of lungs identified CD20 + B cells within the germinal
that is implicated in autoimmunity5,6. Our findings uncover centers and substantial numbers of lung-infiltrating CD4+ T cells
an unexpected molecular link between a vesicular transport (Fig. 1b,c). Autoantibodies were detected in 86% of the affected
protein and a syndrome of autoimmunity manifested by patients, including anti-nuclear antibodies (ANAs), anti-neutrophil
lung and joint disease. cytoplasmic antibodies (ANCAs) and rheumatoid factor (RF) (Table 1
and Supplementary Table 2). Immunoglobulin levels, absolute
Monogenic disorders have proven powerful in elucidating biological lymphocyte counts and CD4/CD8 cell ratios were largely normal
mechanisms underlying autoimmunity7,8 by showing that autoimmu- (Supplementary Table 3). Ninety-five percent of the patients have
nity can arise from perturbations in several non-classical pathways7. arthritis, and some initially presented with joint pain. Four unrelated
Defects in the intracellular trafficking mechanisms of immune cells9 patients underwent renal biopsies that demonstrated immune-mediated
1Department of Pediatrics, Baylor College of Medicine, Houston, Texas, USA. 2Texas Children’s Hospital Center for Human Immuno-Biology, Houston, Texas, USA.
3Department of Medicine, University of California, San Francisco, San Francisco, California, USA. 4Department of Molecular and Human Genetics, Baylor College
of Medicine, Houston, Texas, USA. 5Department of Medicine, Baylor College of Medicine, Houston, Texas, USA. 6Center for Statistical Genetics, Baylor College
of Medicine, Houston, Texas, USA. 7Cardiovascular Research Institute, University of California, San Francisco, San Francisco, California, USA. 8Department of
Dermatology, University of California, San Francisco, San Francisco, California, USA. 9Department of Pathology, Texas Children’s Hospital, Houston, Texas, USA.
10Department of Pathology, University of California, San Francisco, San Francisco, California, USA. 11Human Genome Sequencing Center, Baylor College of Medicine,
Houston, Texas, USA. 12Division of Respiratory Medicine, Hospital for Sick Children, Toronto, Ontario, Canada. 13Department of Pediatrics, University of California,
San Francisco, San Francisco, California, USA. 14Houston, Texas, USA. 15Human Genetics Center and Institute of Molecular Medicine, University of Texas–Houston
Health Science Center, Houston, Texas, USA. 16These authors contributed equally to this work. 17These authors jointly supervised this work. Correspondence should
be addressed to A.K.S. (anthony.shum@ucsf.edu) or J.S.O. (orange@bcm.edu).
Received 16 October 2014; accepted 19 March 2015; published online 20 April 2015; doi:10.1038/ng.3279
a d I
1 2
1 2 3 4 5 6 7 8 9 10 11
R II
Family A
+/m +/+ +/m +/+
1 2 3 4 5 6 * 7 8 9 10
III
+/+ +/m +/m +/m +/m
1 * 2
b IV
+/m
1 2
I
+/m +/+
Family B
1 2 3 4 5 6 7 8
II
+/+ +/m +/+ +/m +/+ +/+
1 * 2 3 4 5 6 7 8
H&E CD20
III
+/m +/m +/m +/m +/+ +/+ +/+ +/+
1 2
I
1 2 3 4 5 6
II
© 2015 Nature America, Inc. All rights reserved.
1 2 * 3 4 5 6 7 8 9 10 * 11 12 13 14 15 16 17 18
CD8 CD4
III
c +/+ +/+ +/+
Family C
1 2 *
I 1 2
+/m I
Family E
1* 2 3 4 5*
Family D
* * +/m
II 1 2
+/+ +/m +/m +/+ II
1* 2 * 3* 4 * 5* 6*
+/m +/+
II
CD8 CD4 +/m +/+ +/m +/+ +/m +/+
Figure 1 Genetic segregation of interstitial lung disease and arthritis with COPA mutations. (a) Left, a chest computed tomography (CT) scan of patient
npg
B.III.1 shows basilar ground glass opacities (circle), reticulations (red arrow) and cystic formations (black arrow). Right, a bilateral knee radiograph
from A.III.6 before restorative surgery. The radiograph shows generalized osteopenia with pathologic fracture of the distal femur. There is bilateral
avascular necrosis of the metaphysis and diaphysis of the distal femur and proximal tibia with slackening of the condyles and articular cartilage
destruction. (b,c) Lung biopsy sections from patients A.III.6 (b) and B.III.1 (c) stained with hematoxylin and eosin (H&E) show cellular interstitial
infiltrates and germinal center formation; immunohistochemical analysis demonstrates the presence of CD20 + B cells and CD4+ and CD8+ T cells.
Scale bars, 100 µm. (d) The pedigrees for families A–E are shown with the presence of wild-type (+) or mutant (m) COPA alleles. Family A (c.698G>A)
shows incomplete penetrance, with one unaffected carrier over four generations. Family B (c.728A>G) shows incomplete penetrance of severe disease
in females affected over three generations. Family C (c.721G>A) shows incomplete penetrance of severe disease and segregation of the mutation
within affected patients over six generations. In family D (c.690G>T), there is an unaffected mutation carrier who is 1 year old, below the age of onset.
Family E (c. 698G>A) shows complete penetrance over two generations. Square, male; circle, female; black filled, affected with severe disease;
unfilled, unaffected; slash, deceased; asterisk, submitted for whole-exome sequencing; vertical line, unaffected carrier. Roman numerals represent
the generation within the family, and Arabic numerals designate individuals within the generation.
renal disease (Supplementary Fig. 1 and Supplementary Table 1), a c.698G>A (p.Arg233His) variant and individual B.III.1 from family
providing further evidence of autoimmunity. All patients have B having a c.728A>G (p.Asp243Gly) variant (NM_004371.3). These
required long-term immunosuppression. variants had not been identified in several large databases12. The
We enrolled patients for whole-exome sequencing to identify variants cosegregated with the disease phenotype, and we identified
genetic variants potentially associated with disease. Family pedi- them in all other affected family members (Supplementary
grees suggested autosomal dominant inheritance with incomplete Fig. 2). For family C, we performed whole-exome sequencing analysis
penetrance (Fig. 1d). We obtained whole-exome sequencing data for and direct sequencing for five affected individuals (C.IV.1, C.IV.2,
patients A.III.6, A.IV.1 and B.III.1 and carried out analysis to identify C.IV.15, C.V.1 and C.V.9) and four unaffected pedigree members
shared variants. There was only one gene, COPA, observed to have (C.III.2, C.III.10, C.IV.16 and C.V.8) and identified a c.721G>A
variants in both families, with the two individuals in family A having (p.Glu241Lys) COPA variant on chromosome 1 to be the only rare,
Table 1 Demographic and clinical characteristics of affected four variants of the COPA gene in five affected families, that the rare
individuals with COPA mutations variants cosegregated with the disease and that the variation at this
Demographic and clinical characteristics Patients n (%) locus is linked to the autoimmunity phenotype, we conclude that it
Total 21 is highly likely that pathological variants in COPA are the underlying
Age at presentation <5 years 16 (76) cause of this syndrome.
Sex COPA encodes the COPA subunit of COPI. COPI and COPII are
Male 8 (38) carrier complexes required for membrane trafficking between the ER
Female 13 (62) and the Golgi. COPII facilitates the anterograde movement of newly
Symptom at initial presentation created proteins or lipids from the ER toward the cell surface, whereas
Tachypnea, cough, hemoptysis 14 (67) COPI is engaged in the retrograde movement of proteins from the
Joint pain 5 (24) Golgi to the ER14. To investigate potential mechanisms by which the
Arthritis 20 (95) COPA variants might lead to disease, we measured the levels of COPA
Pulmonary manifestations transcript and COPA protein and found no difference in expression
Hemorrhage or interstitial lung disease 21 (100)
levels between patient and control B-lymphoblastoid cell lines (BLCLs)
Autoantibodies 18 (86)
(Supplementary Fig. 3a–c). Imaging analysis demonstrated that the
ANAs 14 (67)
COPA variants did not cause aggregation or mislocalization of COPA
ANCAs 15 (71)
protein thereby impairing its function4 (Supplementary Fig. 3d).
RF 9 (43)
Given that COPA was expressed at normal levels in patients, we
Response to immunosuppression 21 (100)
sought to determine a functional consequence of the COPA variants.
Notably, point mutations mapping to the COPA WD40 domain in
nonsynonymous variant that segregated with the disease phenotype yeast, analogous to the disease-associated damaging alleles identified
© 2015 Nature America, Inc. All rights reserved.
(Table 2 and Supplementary Fig. 2). In family D, whole-exome in the patients, impair binding of dilysine-tagged proteins and cause a
sequencing of subjects D.I.2, D.II.1–D.II.5 and D.III.1–D.III.6 iden- defect in retrograde transport2,3,15. To test whether binding of dilysine
tified another COPA variant, c.690G>T (p.Lys230Asn), that cosegre- proteins to the COPA mutants identified was similarly impaired, we
gated with the disease phenotype. In family E, given that the clinical generated a synthetic peptide corresponding to the cytoplasmic tail of
phenotype was nearly identical to that for the other patients with yeast Wbp1 (Wbp1p peptide), which bears a dilysine retrieval signal
COPA alterations, we performed targeted Sanger sequencing of that binds to mammalian COPA2,16. We incubated recombinant COPA
COPA exons 8 and 9 and identified the c.698G>A (p.Arg233His) protein with Wbp1p peptide and found that Glu241Lys mutant COPA
missense variant that we also discovered in family A (Table 2 and had impaired binding to Wbp1p peptide in comparison to wild-type
Supplementary Fig. 2). Interestingly, genotype analysis of families COPA (Fig. 2b). We next transfected cell lines stably expressing
A and E identified the same ancestral haplotype in the COPA locus Glu241Lys mutant COPA with a construct for a reporter protein bear-
spanning approximately 140 kb and 29 SNPs. The presence of a com- ing a dilysine retrieval signal. In the subsequent immunoprecipitation
mon haplotype provides a potential explanation for the co-occurrence assays, we also found impaired binding of the dilysine reporter to
of the COPA c.698G>A (p.Arg233His) variant in these two apparently Glu241Lys mutant COPA (Fig. 2c). We obtained comparable results in
unrelated families (Supplementary Table 4). Overall, we employed experiments using Lys230Asn mutant COPA (Supplementary Fig. 4).
whole-exome sequencing and targeted sequencing at two institutions Taken together, these results demonstrate that mutant COPA bearing
and independently identified four unique variants in five affected a WD40 variant identified in patients behaves in an anomalous way,
families. Within these 5 families, 21 of 30 COPA mutation carriers analogous to the altered activity observed for the point mutations in
were affected by disease whereas 9 were unaffected clinically, consist- yeast shown to cause a defect both in the binding of dilysine-tagged
npg
ent with incomplete penetrance. Notably, all variants identified were proteins and retrograde transport2,3,15.
predicted to be deleterious and mapped to a 14-residue region in the We next sought to identify physiological effects of mutant COPA
highly conserved, functionally important WD40 domain of the COPA that might be linked mechanistically to autoimmune disease. Defects
protein3 (Fig. 2a and Table 2).
To provide statistical evidence that variants Table 2 Bioinformatic evaluation of segregating missense variants in COPA
in COPA are involved in disease etiology, we
Family ID A and E B C D
performed an affected-only parametric link-
hg19 physical position 160,293,229 160,283,894 160,283,901 160,293,237
age analysis on the four families that were
phyloP (100wayall) 7.12 7.53 7.26 2.81
informative, analyzing the COPA missense
13 PhastCons 1 1 1 1
variants in these families . No accurate
GERP 4.73 5.62 5.62 2.87
estimate of the penetrance for this rare pheno- cDNA change c.698G>A c.728A>G c.721G>A c.690G>T
type exists. We assumed that the disease allele Amino acid change p.Arg233His p.Asp243Gly p.Glu241Lys p.Lys230Asn
frequency and variant allele frequency were PolyPhen-2 HVAR Probably damaging Probably damaging Probably damaging Probably damaging
both 0.001. The four informative pedigrees SIFT Damaging Damaging Damaging Damaging
provided the following logarithm of odds LRT Damaging Damaging Damaging Damaging
(LOD) scores at θ (recombination fraction) = 0 MutationAssessor Low Neutral Low Medium
for the identified COPA variants: family A, MutationTaster Disease causing Disease causing Disease causing Disease causing
1.5; family B, 0.90; family C, 2.7; family D, CADD (scaled) 32 26.7 35 19.81
0.60. The combined statistically significant phyloP, phastCons and GERP scores for nucleotide conservation were derived from the UCSC Genome Browser.
LOD score of 5.7 for these families provides PolyPhen-2, SIFT, LRT, MutationAssessor and MutationTaster provided a functional prediction for each missense
variant. For Combined Annotation-Dependent Depletion (CADD), a scaled C score ≥10 places a variant within the
statistical evidence that COPA is involved top 10% of likely deleterious variants, whereas a C score ≥20 indicates that the variant is within the top 1% of likely
in disease etiology. Given that we observed deleterious variants across the genome.
COPA-Flag (% of WT)
100
GFP-CD8-KKTN. Cell lysates were precipitated with antibody to CD8, Wbp1p beads Input
and immunoprecipitates were assayed for COPA-Flag expression and COPA-Flag COPA-Flag
75
GFP-CD8 enrichment. The number below each lane is the relative ***
IB WT E241K WT E241K 50
protein amount expressed as the percentage of wild-type COPA or
Flag 25
GFP-CD8 levels in cells expressing wild-type COPA. Statistical analysis
of five independent experiments is shown. ***P < 0.001. 100 58 100 115 0
WT E241K
COPA-Flag (% of WT)
COPA-Flag COPA-Flag 100
COPA2,3,15, have been shown to affect retrograde trafficking and to IB WT E241K WT E241K
© 2015 Nature America, Inc. All rights reserved.
75
also indirectly affect COPII-mediated anterograde trafficking from Flag
***
the ER to the Golgi4. Thus, we hypothesized that the COPA variants 50
100 55 100 90
in patients likewise impair intracellular protein trafficking and, as a GFP 25
result, lead to ER stress and activation of the UPR4,17. Notably, ER (CD8)
0
stress and activation of the UPR have been linked to autoimmunity and 100 100 100 107
WT E241K
lung disease10,18,19. To determine whether patients exhibited increased
ER stress, we examined lung biopsies for expression of the molecular in this pathway. Fluorescent imaging of cells showed an increase in
chaperone binding immunoglobulin protein (BiP) 20. We found autophagosome and endolysosome size (Supplementary Figs. 6a–f
that BiP levels were markedly increased in the lung epithelium and and 7a–d), and patient-derived BLCLs demonstrated a decrease in the
alveolar macrophages of patients in comparison to controls (Fig. 3a levels of the autophagy substrate p62 and torin-induced p62 catabo-
and Supplementary Fig. 5a,b), consistent with our hypothesis. lism (Supplementary Fig. 8a,b). This constellation of findings sug-
We next performed dynamic functional assessments of patient- gests a baseline defect in autophagic function that might be due to
derived cells for evidence of enhanced ER stress caused by mutant impaired early endosomal function in patients24. Thus, it is possible
COPA. We treated BLCLs from patients with low levels of thapsigargin, that impaired autophagy exacerbates the increase in ER stress caused
an inhibitor of the ER Ca2+ ATPase used experimentally to induce ER by mutant COPA25. In aggregate, our findings demonstrate that the
stress. Cells from patients with COPA variants had significantly elevated expression of mutant forms of COPA in patients is sufficient for the
levels of transcript for HSPA5 (which encodes BiP) (Fig. 3b and induction of ER stress.
Supplementary Fig. 5c) in comparison to control cells from related Finally, to investigate potential immunological links between
npg
and unrelated individuals (P < 0.05). Interestingly, the clinically asymp- mutant COPA and the clinical phenotype, we examined CD4+ T cells,
tomatic COPA mutation carriers (with apparent non-penetrance) given their high numbers in diseased lungs (Fig. 1b,c) and their critical
demonstrated a weaker increase in ER stress in comparison to those function in autoimmune disease26. Interestingly, recent studies have
severely affected (Fig. 3b), suggesting that the mutation’s effect on shown that ER stress has a role in the generation of TH17 cells27–30,
healthy carriers is below a threshold required for clinical disease. a CD4+ effector T cell population implicated in autoimmunity5,6,
To establish the specific role of mutant COPA in the induction of by causing antigen-presenting cells (APCs) to secrete cytokines
ER stress, we performed small interfering RNA (siRNA)-mediated necessary for the induction and expansion of TH17 populations.
knockdown experiments in HEK (human embryonic kidney) cells. Thus, we hypothesized that mutant COPA might lead to the production
Indeed, we found that reducing COPA expression resulted in higher of TH17 cells through a similar mechanism.
levels of ER stress, as indicated by a marked increase in HSPA5 (BiP) Given that B cells are professional APCs, we tested our BLCLs for
expression (Fig. 3c). To demonstrate that the increased ER stress the expression of cytokines involved in the generation of the CD4 +
in patient-derived cells was specifically due to mutant COPA, we over- TH1, TH2 and TH17 helper T cell subsets. Patient-derived BLCLs with
expressed wild-type or mutant COPA in HEK cells (Supplementary mutant COPA showed a significant elevation in the levels of tran-
Fig. 5e,f). Consistent with a dominant-negative effect of the muta- scripts encoding interleukin (IL)-1β, IL-6 and IL-23 (p19 and p40
tions, presence of any of the four COPA mutations identified in subunits) (Fig. 4a), cytokines that are critical for the priming and
patients resulted in elevated HSPA5 (BiP) levels in comparison to expansion of TH17 populations31, and this effect was enhanced by
cells transfected with wild-type COPA (Fig. 3d and Supplementary ER stress. In contrast, transcript levels for the TH2-priming cytokine
Fig. 5d). The levels of additional ER stress markers such as ATF4 and IL-4 in patients were similar to those in controls, whereas transcript
DDIT3 (CHOP) were also significantly increased when COPA expres- levels for the TH1-priming cytokine IL-12 (p35 subunit) were very
sion was reduced or mutant COPA was present (Supplementary low in both groups (Fig. 4a). Transcript levels for IL-10 and trans-
Fig. 5g,h). Given the role of autophagy in alleviating ER stress21 and forming growth factor (TGF)-β were not different between patients
regulating immune responses22,23, we sought to evaluate alterations and controls, and interferon (IFN)-α (IFNA2) levels in patients were
a
Control E241K D243G R233H
***
0 0 0 0
TG – + – + – + – + – + – + – + – + – + – + – + CTRL COPA CTRL COPA WT E241K D243G R233H
CO1 CO2 CO3 CO4 HC1 PA1 PA2 HC2 PA4 PA5 PA6 siRNA siRNA siRNA siRNA
Figure 3 Mutant COPA leads to an increase in ER stress. (a) Immunohistochemistry for the ER stress component BiP in lung biopsies from patients with
© 2015 Nature America, Inc. All rights reserved.
the Glu241Lys, Asp243Gly and Arg233His COPA variants is shown. A non-carrier is shown as a control. Scale bars, 40 µm. Enlarged insets emphasize
BiP staining in lung epithelial cells. (b) Quantitative PCR (qPCR) of HSPA5 (BiP) expression in untreated (white; −) and thapsigargin (TG)-treated
(striped, gray and black; +) BLCLs. The leftmost graph shows the results from two unrelated controls (CO1, CO2) and two related non-carrier controls
(CO3, CO4). Results were normalized to values for untreated CO1. Data shown are from healthy carriers HC1 and HC2 (C.III.2, B.III.4) and patients
PA1, PA2 and PA4–PA6 (C.IV.15, C.V.1, B.III.3, A.II.5, A.III.6). One-way ANOVA followed by Dunnett’s multiple-comparison test was used to compare
thapsigargin-treated experimental groups with thapsigargin-treated CO1 as a reference. Means and s.e.m. of data pooled from 2–4 independent
experiments are shown. *P < 0.05, **P < 0.01, ***P < 0.001. (c) COPA knockdown in HEK293 cells. The left graph demonstrates >90% reduction
in COPA mRNA levels after treatment with COPA siRNA. The right graph shows induction of ER stress after COPA knockdown as measured by HSPA5
(BiP) levels. Means and s.e.m. of data pooled from 3 independent experiments each with n = 3–4 (pooled n = 11) are shown. Statistical comparisons
were made using Student’s unpaired t tests with Welch’s correction. **P < 0.01, ***P < 0.001. CTRL, control. (d) HEK293 cells were transfected
with vectors for wild-type COPA or mutant Glu241Lys, Asp243Gly or Arg233His COPA, and HSPA5 (BiP) mRNA levels were measured 48 h after
transfection. Means and s.e.m. are shown. One-way ANOVA followed by Dunnett’s multiple-comparison test was used to compare experimental groups
with wild-type control. Results are shown for data pooled from three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.
increased and decreased in comparison to controls (Supplementary disorders in which alterations to widely expressed proteins lead
Fig. 9). To determine whether CD4+ T cells in patients demonstrated to distinct clinical phenotypes32,33. For instance, four mendelian
an immunological phenotype consistent with this pattern of cytokine syndromes with disparate clinical manifestations result from
expression, we profiled peripheral blood mononuclear cells (PBMCs). mutations affecting the COPII trafficking machinery34. We can only
Intracellular cytokine staining demonstrated equal frequencies of speculate that, in patients with COPA mutations, there are tissue-
npg
IL-13–secreting TH2 cells among patient and control cells (Fig. 4b,c). specific cells central to the disease that are sensitive to alterations in
By contrast, patient CD4+ T cells demonstrated a marked skewing COPA function. In addition, given the reduced penetrance of disease,
away from IFN-γ–secreting TH1 cells and a significant increase there may be differences attributable to sex or other genetic and envi-
in the frequency of IL-17A–secreting TH17 cells (Fig. 4b,c and ronmental factors that dictate whether a carrier of a mutant COPA
Supplementary Fig. 10). Thus, mutant COPA leads both to higher allele develops clinical disease.
levels of ER stress in patient-derived BLCLs and the production of Future studies may elucidate how defects in COPA lead to ER
cytokines that promote the expansion of TH17 populations. stress and autoimmunity. ER stress and the UPR have been strongly
Through genetic studies in five unrelated families, we provide associated with the pathogenesis of inflammatory and autoimmune
evidence that point mutations affecting a 14-residue region of the disorders, in part through the intersection of key pathways that lead
WD40 domain of COPA cause a systemic autoimmune syndrome of to activation of nuclear factor (NF)-κB or alterations in antigen pres-
lung and joint disease. Functional analyses demonstrate that mutant entation10,18. Furthermore, because the COPA mutations identified
COPA is unable to bind proteins with dilysine-based motifs and pro- are predicted to impair retrograde transport, proteins destined for
vide evidence of defective intracellular trafficking. Specifically, we return to the ER may instead escape the cell or be presented in tissue
show that expression of mutant COPA in cells induces higher levels of sites to prime autoreactive immune responses35. Interestingly, both
ER stress, which may be further exacerbated in patients by impaired ER stress and TH17 cells have previously been implicated in patients
autophagy25. Finally, we establish a putative immunological mecha- with interstitial lung disease19,36,37 and inflammatory arthritis10,38,39,
nism for COPA mutations in disease by showing that mutant COPA including rheumatoid arthritis5. It remains unknown whether TH17
via ER stress leads to a cytokine milieu that promotes the generation cells in COPA-mutant patients are pathogenic or arise as a result of
of TH17 cells, a cell population shown to mediate autoimmunity5,6. inflammation. The role of regulatory T cells also needs to be deter-
It may seem unusual that mutation of a ubiquitously expressed mined. In conclusion, the discovery of a molecular link between a
protein in an essential trafficking pathway leads to an autoimmune defect in the COPI vesicular trafficking machinery and a new syn-
syndrome restricted to the lung and joints. However, there are several drome of autoimmunity defines a starting point for understanding
a 100 250 40 10 10
IL23A *** IL12B IL1B 105 IL6 IL12A IL4
50
200 *** *** 10 4
*** 8 8
30
cytokine/GAPDH
8
Fold increase
150 *** 10 3 6 6
6 20
100 102 4 4
4
*** 10
2 50 101 2 2
*
0 0 0 100 0 0
TG – + – + – + – + –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+ –+
CO1 CO4 PA3 PA2 CO1 CO4 PA3 PA2 CO1 CO4 PA3 PA2 CO1 CO4 PA3 PA2 CO1 CO4 PA3 PA2 CO1 CO4 PA3 PA2
IFN-γ
IL-13
+
numbers. (a) Cytokine 0 84 0 83 0 99.6 30 3
expression was measured 10
0 37 0 3 0 0.7 20 2
in BLCLs from CO1, CO4
(C.III.10), PA2 (C.V.1) and 5
Patient
10 1
PA3 (C.IV.2) either untreated
+
(white bars; −) or treated 0 0 0
(gray and black bars; +) with 0 63 0 97 0 99.3
ls
ls
ls
s
© 2015 Nature America, Inc. All rights reserved.
nt
nt
nt
tro
tro
tro
tie
tie
tie
thapsigargin. Samples were CD4
on
on
on
Pa
Pa
Pa
C
C
analyzed via qPCR for mRNA
levels of IL-23 p19 (IL23A), IL-12 p40 (IL12B), IL-12 p35 (IL12A), IL-4 (IL4), IL-1β (IL1B) and IL-6 (IL6). Shown are mRNA levels normalized against
the average value for healthy control 1 (CO1) in the absence of stimulation. Error bars indicate s.e.m. for three technical replicates in the assay. Results
are representative of three independent experiments. *P < 0.05, ***P < 0.001. (b,c) CD4+ T cells from three COPA-mutant patients with stable disease
(C.IV.1 and C.IV.2 were on mycophenolate; C.IV.15 was on no treatment) and three age-matched controls were analyzed for intracellular levels of the
IL-17A, IFN-γ and IL-13 cytokines after stimulation with PMA (phorbol 12-myristate 13-acetate) and ionomycin. (b) Representative FACS plots for CD4+
T cells producing IL-17A, IFN-γ and IL-13 are shown. Plots are gated on live CD3+CD4+ T cells. (c) Graphs demonstrating the frequency of TH1, TH2 and
TH17 cells in patients and controls. More than 350,000 cells in each of 3 patients and 3 controls were analyzed in a single experiment. The frequency of
IL-17A–secreting TH17 cells is significantly increased in patients in comparison to controls (patient mean 39% versus control mean 19%; **P < 0.01;
n = 3), whereas patients show a significant decrease in IFN-γ–producing TH1 cells (patient mean 4% versus control mean 13%; *P < 0.05; n = 3).
No significant difference was found in the frequency of IL-13–secreting T H2 cells. Student’s unpaired t tests were used for statistical evaluation.
the role of intracellular transport in the induction of human autoim- (National Human Genome Research Institute) grant U54HG003273 (R.A.G.),
munity and as a potential therapeutic target in disease. US NIH grant K23NS078056 (W.W.) and US NIH grant AI053831 (L.B.W.).
AUTHOR CONTRIBUTIONS
URLs. BCM-HGSC protocol, https://hgsc.bcm.edu/sites/default/
L.B.W., B.J. and W.W. designed and performed experiments, analyzed data and
files/documents/Illumina_Barcoded_Paired-End_Capture_Library_ wrote the manuscript. T.J.V., L.R.F., D.L.C., S.K.N., S.D.D. and M.R.W. provided
Preparation.pdf; Burrows-Wheeler aligner (BWA), http://bio-bwa.
npg
clinical care and analyzed clinical data. J.H. and K.D.J. performed pathology
sourceforge.net/; Genome Analysis Toolkit (GATK), http://www. services. A.S.-P., K.L., T.G., R.L.P.S.-C., C.-G.J., N.Z. and L.F.T. analyzed data.
broadinstitute.org/gatk/; GRCh37 human reference genome, http:// Y.S., E.M.M., D.L., K.N., M.T., M.J. and C.S.L. performed experiments and analyzed
data. G.M., N.T.E., F.R.P. and M.D.R. designed experiments and analyzed data.
www.ncbi.nlm.nih.gov/projects/genome/assembly/grc/human/; S.N.J., S.P., D.M.M., E.B., R.A.G., M.S.A. and P.-Y.K. organized studies. S.M.L.,
HGSC Mercury analysis pipeline, https://www.hgsc.bcm.edu/ R.L.P.S.-C. and M.H.C. analyzed data and assisted with writing the manuscript.
software/mercury. J.S.O., J.R.L. and A.K.S. supervised the study and wrote the manuscript.
Acknowledgments 1. Zenewicz, L.A., Abraham, C., Flavell, R.A. & Cho, J.H. Unraveling the genetics of
We thank the patients and their families for their involvement in these studies. autoimmunity. Cell 140, 791–797 (2010).
The group at the University of California San Francisco thanks R. Schekman 2. Letourneur, F. et al. Coatomer is essential for retrieval of dilysine-tagged proteins
for useful comments, S. McClymont, O. Estrada, R. Wadia and M. Newasker for to the endoplasmic reticulum. Cell 79, 1199–1207 (1994).
assistance with subject recruitment and the University of California San Francisco 3. Eugster, A., Frigerio, G., Dale, M. & Duden, R. COP I domains required for coatomer
Genomics Core. Funding was provided by US National Institutes of Health (NIH) integrity, and novel interactions with ARF and ARF-GAP. EMBO J. 19, 3905–3917
grant R01AI067946 (J.S.O.), the Jeffrey Modell Foundation (J.S.O.), US NIH grant (2000).
4. Popoff, V., Adolf, F., Brügger, B. & Wieland, F. COPI budding within the Golgi stack.
K08HL095659 (A.K.S.), the Foundation of the American Thoracic Society (A.K.S.),
Cold Spring Harb. Perspect. Biol. 3, a005231 (2011).
the Pulmonary Fibrosis Foundation (A.K.S.), the Nina Ireland Lung Disease 5. Leipe, J. et al. Role of Th17 cells in human autoimmune arthritis. Arthritis Rheum.
Program (A.K.S.), US NIH (National Human Genome Research Institute/National 62, 2876–2885 (2010).
Heart, Lung, and Blood Institute) grant U54HG006542 and (National Institute 6. Miossec, P., Korn, T. & Kuchroo, V.K. Interleukin-17 and type 17 helper T cells.
of Neurological Disorders and Stroke) grant 2RO1NS058529 (J.R.L.), US NIH N. Engl. J. Med. 361, 888–898 (2009).
7. Cheng, M.H. & Anderson, M.S. Monogenic autoimmunity. Annu. Rev. Immunol. 30, 24. Razi, M., Chan, E.Y.W. & Tooze, S.A. Early endosomes and endosomal coatomer
393–427 (2012). are required for autophagy. J. Cell Biol. 185, 305–321 (2009).
8. Rieux-Laucat, F. & Casanova, J.L. Autoimmunity by haploinsufficiency. Science 345, 25. Adolph, T.E. et al. Paneth cells as a site of origin for intestinal inflammation.
1560–1561 (2014). Nature 503, 272–276 (2013).
9. Garg, S. et al. Lysosomal trafficking, antigen presentation, and microbial 26. Palmer, M.T. & Weaver, C.T. Autoimmunity: increasing suspects in the CD4+
killing are controlled by the Arf-like GTPase Arl8b. Immunity 35, 182–193 T cell lineup. Nat. Immunol. 11, 36–40 (2010).
(2011). 27. Wheeler, M.C. et al. KDEL-retained antigen in B lymphocytes induces a
10. Todd, D.J., Lee, A.-H. & Glimcher, L.H. The endoplasmic reticulum stress response proinflammatory response: a possible role for endoplasmic reticulum stress in
in immunity and autoimmunity. Nat. Rev. Immunol. 8, 663–674 (2008). adaptive T cell immunity. J. Immunol. 181, 256–264 (2008).
11. Strange, C. & Highland, K.B. Interstitial lung disease in the patient who has 28. Glatigny, S. et al. Proinflammatory Th17 cells are expanded and induced by
connective tissue disease. Clin. Chest Med. 25, 549–559 (2004). dendritic cells in spondylarthritis-prone HLA-B27–transgenic rats. Arthritis Rheum.
12. Bamshad, M.J. et al. The Centers for Mendelian Genomics: a new large-scale 64, 110–120 (2012).
initiative to identify the genes underlying rare Mendelian conditions. Am. J. Med. 29. DeLay, M.L. et al. HLA-B27 misfolding and the unfolded protein response augment
Genet. A. 158A, 1523–1525 (2012). interleukin-23 production and are associated with Th17 activation in transgenic
13. Fishelson, M. & Geiger, D. Exact genetic linkage computations for general pedigrees. rats. Arthritis Rheum. 60, 2633–2643 (2009).
Bioinformatics 18 (suppl. 1), S189–S198 (2002). 30. Goodall, J.C. et al. Endoplasmic reticulum stress-induced transcription factor,
14. Brandizzi, F. & Barlowe, C. Organization of the ER-Golgi interface for membrane CHOP, is crucial for dendritic cell IL-23 expression. Proc. Natl. Acad. Sci. USA
traffic control. Nat. Rev. Mol. Cell Biol. 14, 382–392 (2013). 107, 17698–17703 (2010).
15. Schröder-Köhne, S., Letourneur, F. & Riezman, H. α-COP can discriminate between 31. Gaffen, S.L., Jain, R., Garg, A.V. & Cua, D.J. The IL-23–IL-17 immune axis: from
distinct, functional di-lysine signals in vitro and regulates access into retrograde mechanisms to therapeutic testing. Nat. Rev. Immunol. 14, 585–600 (2014).
transport. J. Cell Sci. 111, 3459–3470 (1998). 32. Bolze, A. et al. Ribosomal protein SA haploinsufficiency in humans with isolated
16. Goldberg, J. Decoding of sorting signals by coatomer through a GTPase switch in congenital asplenia. Science 340, 976–978 (2013).
the COPI coat complex. Cell 100, 671–679 (2000). 33. Gissen, P. & Maher, E.R. Cargos and genes: insights into vesicular transport from
17. Claerhout, S. et al. Abortive autophagy induces endoplasmic reticulum stress and inherited human disease. J. Med. Genet. 44, 545–555 (2007).
cell death in cancer cells. PLoS ONE 7, e39400 (2012). 34. Russo, R., Esposito, M.R. & Iolascon, A. Inherited hematological disorders due to
18. Hasnain, S.Z., Lourie, R., Das, I., Chen, A.C.-H. & McGuckin, M.A. The interplay defects in coat protein (COP)II complex. Am. J. Hematol. 88, 135–140 (2013).
between endoplasmic reticulum stress and inflammation. Immunol. Cell Biol. 90, 35. Weber, C.K. et al. Antibodies to the endoplasmic reticulum–resident chaperones
260–270 (2012). calnexin, BiP and Grp94 in patients with rheumatoid arthritis and systemic lupus
19. Tanjore, H., Blackwell, T.S. & Lawson, W.E. Emerging evidence for endoplasmic erythematosus. Rheumatology 49, 2255–2263 (2010).
© 2015 Nature America, Inc. All rights reserved.
reticulum stress in the pathogenesis of idiopathic pulmonary fibrosis. Am. J. Physiol. 36. Olewicz-Gawlik, A. et al. Interleukin-17 and interleukin-23: importance in the
Lung Cell. Mol. Physiol. 302, L721–L729 (2012). pathogenesis of lung impairment in patients with systemic sclerosis. Int. J. Rheum.
20. Samali, A., Fitzgerald, U., Deegan, S. & Gupta, S. Methods for monitoring Dis. 17, 664–670 (2014).
endoplasmic reticulum stress and the unfolded protein response. Int. J. Cell Biol. 37. Weaver, C.T., Elson, C.O., Fouser, L.A. & Kolls, J.K. The Th17 pathway and
2010, 830307 (2010). inflammatory diseases of the intestines, lungs, and skin. Annu. Rev. Pathol. 8,
21. Ogata, M. et al. Autophagy is activated for cell survival after endoplasmic reticulum 477–512 (2013).
stress. Mol. Cell. Biol. 26, 9220–9231 (2006). 38. Nistala, K. et al. Interleukin-17–producing T cells are enriched in the joints of
22. Deretic, V., Saitoh, T. & Akira, S. Autophagy in infection, inflammation and children with arthritis, but have a reciprocal relationship to regulatory T cell
immunity. Nat. Rev. Immunol. 13, 722–737 (2013). numbers. Arthritis Rheum. 58, 875–887 (2008).
23. Bhattacharya, A. & Eissa, N.T. Autophagy and autoimmunity crosstalks. 39. Benham, H. et al. Th17 and Th22 cells in psoriatic arthritis and psoriasis. Arthritis
Front. Immunol. 4, 88 (2013). Res. Ther. 15, R136 (2013).
npg
purification, the amount and size distribution of the pre-capture LM-PCR dbSNP132. A filtering analysis of the exome variants in the nine subjects in
product was determined using the LabChip GX electrophoresis system family C was performed in VarSifter44. The filtering strategy was designed
(PerkinElmer). to find variants shared among 6 carriers that were not seen in 3 related
Capture enrichment. The pre-capture libraries were pooled as a 4-plex controls or in 20 other whole-exome sequences generated for unrelated
(approximately 500 ng/sample, 1 µg per pool) and hybridized in solution control subjects. We also excluded variants with a minor allele frequency
to the HGSC CORE design40 (52 Mb; NimbleGen) according to the manu- (MAF) >0.001. This filtering analysis yielded a single rare variant in COPA
facturer’s protocol (NimbleGen SeqCap EZ Exome Library SR User’s Guide) (c.721G>A) that was shared by affected subjects and not present in any control.
with minor revisions. Human Cot-1 DNA and full-length Illumina adaptor- The filtering strategy in family D was performed similarly to that employed
specific blocking oligonucleotides were added into the hybridization to block for family C to identify a variant in COPA (c.690G>T) that cosegregated with
repetitive genomic sequences and the adaptor sequences, and post-capture the disease phenotype.
LM-PCR amplification was performed using 2× SOLiD Library High-Fidelity
Amplification Mix with 14 cycles of amplification. After the final SPRI bead Plasmids and stably transfected cell lines. pCMV6-Entry vectors encoding
purification step, the quantity and size of the capture library was analyzed Myc-DDK–tagged or untagged human COPA (Origene) were used as the
using an Agilent BioAnalyzer 2100 DNA Chip 7500. The efficiency of the parental vectors for mutant plasmid generation. Mutant plasmids were created
capture was evaluated by performing a qPCR-based quality assay on the four via site-directed mutagenesis using the QuikChange Lightning kit (Agilent
standard NimbleGen internal controls. Successful enrichment of the capture Technologies) according to the manufacturer’s instructions or a third-party
libraries was estimated to range from a ∆Ct value of 6 to 9 over the non- service (Epoch Life Sciences). CD8A sequence spanning residues 1 through 207
enriched samples. was subcloned from human cDNA into the pCMV6-Entry vector. A sequence
Sequencing. Library templates were prepared for sequencing using Illumina’s encoding mTagGFP at the N terminus was added to the CD8α sequence,
npg
cBot cluster-generation system with TruSeq PE Cluster Generation kits. Briefly, and a sequence encoding residues KYKSRRSFIDEKKTN including the
these libraries were denatured and diluted in hybridization buffer to achieve dilysine motif KKTN at the C terminus was added. Cell lines stably expressing
a load density of ~800,000 clusters/mm40. The library pool was loaded on a wild-type or Glu241Lys mutant COPA-Flag fusion protein in the presence
single lane of a HiSeq flow cell, with 2% phiX control library spiked into the of doxycycline were generated with the Flp-In T-REx system (Invitrogen).
lane for run quality control. The sample libraries then underwent bridge ampli- Full-length sequences were amplified from patient cDNA and subcloned into
fication to form clonal clusters, followed by hybridization with the sequencing pcDNA5/FRT/TO with a sequence encoding a C-terminal Flag tag. Complete
primer. The sequencing run was performed in paired-end mode using the vectors were then transfected into Flp-In T-REx 293 cells (Invitrogen) and
Illumina HiSeq 2000 platform. Using TruSeq SBS kits, sequencing-by-synthesis selected according to the manufacturer’s protocol. T-REx 293 cells are certified
reactions were extended for 101 cycles from each end, with an additional by the manufacturer to be free of mycoplasma.
7 cycles for the index read. With the sequencing run yielding an average of
~8.4 Gb per sample, samples achieved an average of 89% of the targeted exome Cell transfection. For gene expression studies, HEK293 cells were transiently
bases covered to a depth of 20× or greater. transfected with 5 µg of plasmid DNA using TransIT-293 Transfection Reagent
Primary data analysis for families A and B. Initial sequence analysis was (Mirus) according to the manufacturer’s instructions. For COPA knockdown
performed using the HGSC Mercury analysis pipeline41 (see URLs). In sum- experiments, HEK293 cells were transfected with either COPA-targeting
mary, the primary analysis software on the instrument produces .bcl files that siRNA or control scrambled siRNA (Thermo Scientific Dharmacon) using
are transferred off the instrument into the HGSC analysis infrastructure by Lipofectamine RNAiMAX transfection reagent (Life Technologies) following
the HiSeq Real-Time Analysis module. Once the run is complete and all .bcl the manufacturer’s instructions. After 48 h, cells were collected and processed
files are transferred, Mercury runs the vendor’s primary analysis software for RNA isolation. HEK293FT cells (Invitrogen) are certified by the manufac-
(CASAVA), which demultiplexes pooled samples and generates sequence turer to be free of mycoplasma.
reads and base call confidence values (qualities). The next step is the map- For microscopic analysis, HEK293T cells stably expressing GFP-LC3 were
ping of reads to the GRCh37 human reference genome (see URLs) using electroporated using the Amaxa 4D nucleofector system according to the
the Burrows-Wheeler aligner (BWA41; see URLs) and producing a BAM42 manufacturer’s protocol. Briefly, HEK293T cells were grown to 80% conflu-
(binary alignment/map) file. The third step involves quality recalibration using ence. Then, 1 × 106 cells were nucleofected with 2 µg of plasmid DNA encoding
the Genome Analysis Toolkit (GATK; see URLs) and, where necessary, the the COPA mutants. After 24 h, cells were collected and imaged.