Inherited CARD9 Deficiency in a Child with Invasive Disease Due to Exophiala dermatitidis and Two Older but Asymptomatic Siblings

概要

1

Title page

2
3

x

4

The name(s) of the author(s)
Yusuke Imanaka

5
6

x

A concise and informative title

7

Inherited CARD9 deficiency in a child with invasive disease due to Exophiala dermatitidis and two

8

older but asymptomatic siblings

9
10

x

The affiliation(s) and address(es) of the author(s)

11
12

ʀYusuke Imanaka

13

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

14

Hiroshima, Japan

15

lowiqyou@yahoo.co.jp

16
17

࣭Maki Taniguchi

18

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

19

Hiroshima, Japan

20

taniguchi-mk@hiroshima-u.ac.jp

21
22

࣭Takehiko Doi

23

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

24

Hiroshima, Japan

25

take-doi02@hiroshima-u.ac.jp

26
27

࣭Miyuki Tsumura

28

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

29

Hiroshima, Japan

30

m055@hiroshima-u.ac.jp

31
32

࣭Rie Nagaoka

33

Division of Infectious Diseases Laboratory Medicine, Hiroshima University Hospital, Hiroshima, Japan

34

pmarie@hiroshima-u.ac.jp

35
36

࣭Maiko Shimomura





37

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

38

Hiroshima, Japan

39

shimomai0105@hiroshima-u.ac.jp

40
41

࣭Takaki Asano

42

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

43

Hiroshima, Japan

44

St. Giles Laboratory of Human Genetics of Infectious Diseases, Rockefeller Branch, Rockefeller

45

tasano@rockefeller.edu

46
47

࣭Reiko Kagawa

48

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

49

Hiroshima, Japan

50

ykagawa@ja2.so-net.ne.jp

51
52

࣭Yoko Mizoguchi

53

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

54

Hiroshima, Japan

55

ymizoguchi@gmail.com

56
57

࣭Shuhei Karakawa

58

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

59

Hiroshima, Japan

60

kara1224@hiroshima-u.ac.jp

61
62

࣭Koji Arihiro

63

Department of Anatomical Pathology, Hiroshima University Hospital, Hiroshima, Japan

64

arihiro@hiroshima-u.ac.jp

65
66

࣭Kohsuke Imai

67

Department of Pediatrics and Developmental Biology, Graduate School of Medical and Dental Sciences,

68

Tokyo Medical and Dental University, Tokyo, Japan

69

kimai.ped@tmd.ac.jp

70
71

࣭Tomohiro Morio





72

Department of Pediatrics and Developmental Biology, Graduate School of Medical and Dental Sciences,

73

Tokyo Medical and Dental University, Tokyo, Japan

74

tmorio.ped@tmd.ac.jp

75
76

࣭Jean-Laurent Casanova

77

St. Giles Laboratory of Human Genetics of Infectious Diseases, Rockefeller Branch, Rockefeller University,

78

New York, NY, United States

79

Laboratory of Human Genetics of Infectious Diseases, Necker Branch, INSERM UMR 1163, Imagine

80

Institute, Paris, France

81

University of Paris, Paris, France, EU

82

Howard Hughes Medical Institute, New York, USA

83

casanova@rockefeller.edu

84
85

࣭Anne Puel

86

St. Giles Laboratory of Human Genetics of Infectious Diseases, Rockefeller Branch, Rockefeller University,

87

New York, NY, United States

88

Laboratory of Human Genetics of Infectious Diseases, Necker Branch, INSERM UMR 1163, Imagine

89

Institute, Paris, France

90

University of Paris, Paris, France, EU

91

anne.puel@inserm.fr

92
93

࣭Osamu Ohara

94

Department of Applied Genomics, Kazusa DNA Research Institute, Kisarazu, Japan

95

ohara@kazusa.or.jp

96
97

࣭Katsuhiko Kamei

98

Department of Medical Mycology Research Center, Chiba University, Japan

99

kkamei-chiba@umin.ac.jp

100
101

࣭Masao Kobayashi

102

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

103

Hiroshima, Japan

104

Japan Red Cross, Chugoku-Shikoku Block Blood Center, Hiroshima, Japan

105

masak@hiroshima-u.ac.jp

106
107

࣭Satoshi Okada





108

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,

109

Hiroshima, Japan

110

sokada@hiroshima-u.ac.jp

111
112

x

113

Correspondence to Satoshi Okada

114

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Sciences

115

1-2-3 Kasumi, Minami-Ku, Hiroshima-Shi, Hiroshima, 734-8551, Japan

116

E-mail: sokada@hiroshima-u.ac.jp

117

Tell: +81-82-257-5212

118

Fax: +81-82-257-5214

The e-mail address, telephone and fax numbers of the corresponding author

119





120

Inherited CARD9 deficiency in a child with invasive disease due to Exophiala dermatitidis and two

121

older but asymptomatic siblings

122

123

Authors

124

Yusuke Imanaka1, Maki Taniguchi1, Takehiko Doi1, Miyuki Tsumura1, Rie Nagaoka2, Maiko

125

Shimomura1, Takaki Asano1, 3), Reiko Kagawa1, Yoko Mizoguchi1, Shuhei Karakawa1, Koji Arihiro4,

126

Kohsuke Imai5, Tomohiro Morio5, Jean-Laurent Casanova3, 6, 7, 8, Anne Puel3, 6, 7, Osamu Ohara9,

127

Katsuhiko Kamei10, Masao Kobayashi1, 11), Satoshi Okada1

128

129

Institutions

130

1

131

Hiroshima, Japan

132

2

Division of Infectious Diseases Laboratory Medicine, Hiroshima University Hospital, Hiroshima, Japan

133

3

St. Giles Laboratory of Human Genetics of Infectious Diseases, Rockefeller Branch, Rockefeller

134

University, New York, NY, United States

135

4

Department of Anatomical Pathology, Hiroshima University Hospital, Hiroshima, Japan

136

5

Department of Pediatrics and Developmental Biology, Graduate School of Medical and Dental Sciences,

137

Tokyo Medical and Dental University, Tokyo, Japan



Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Science,



138

6

139

Institute, Paris, France

140

7

University of Paris, Paris, France, EU

141

8

Howard Hughes Medical Institute, New York, USA

142

9

Department of Applied Genomics, Kazusa DNA Research Institute, Kisarazu, Japan

143

10

144

11

Laboratory of Human Genetics of Infectious Diseases, Necker Branch, INSERM UMR 1163, Imagine

Department of Medical Mycology Research Center, Chiba University, Japan

Japan Red Cross, Chugoku-Shikoku Block Blood Center, Hiroshima, Japan

145

146

)

current affiliation

147

148

Corresponding Author

149

Satoshi Okada, MD, PhD

150

Department of Pediatrics, Hiroshima University Graduate School of Biomedical and Health Sciences

151

1-2-3 Kasumi, Minami-Ku, Hiroshima-Shi, Hiroshima, 734-8551, Japan

152

Tell: +81-82-257-5212

153

Fax: +81-82-257-5214

154

E-mail: sokada@hiroshima-u.ac.jp

155
156

Abstract





157

Purpose

158

Autosomal recessive CARD9 deficiency predisposes patients to invasive fungal disease. Candida and

159

Trichophyton species are major causes of fungal disease in these patients. Other CARD9-deficient patients

160

display invasive disease caused by other fungi, such as Exophiala spp. The clinical penetrance of CARD9

161

deficiency regarding fungal disease is surprisingly not complete until adulthood, though the age remains

162

unclear. Moreover, the immunological features of genetically confirmed yet asymptomatic individuals with

163

CARD9 deficiency have not been reported.

164

Methods

165

Identification of CARD9 mutations by gene panel sequencing and characterization of the cellular phenotype

166

by quantitative PCR, immunoblot, luciferase reporter, and cytometric bead array assays were performed.

167

Results

168

Gene panel sequencing identified compound heterozygous CARD9 variants, c.1118G>C (p.R373P) and

169

c.586A>G (p.K196E), in a 4-year-old patient with multiple cerebral lesions and systemic lymphadenopathy

170

due to Exophiala dermatitidis. The p.R373P is a known disease-causing variant, whereas the p.K196E is a

171

private variant. Although the patient’s siblings, a 10-year-old brother and an 8-year-old sister, were also

172

compound heterozygous, they have been asymptomatic to date. Normal CARD9 mRNA and protein

173

expression were found in the patient’s CD14+ monocytes. However, these cells exhibited markedly





174

impaired pro-inflammatory cytokine production in response to fungal stimulation. Monocytes from both

175

asymptomatic siblings displayed the same cellular phenotype.

176

Conclusions

177

CARD9 deficiency should be considered in previously healthy patients with invasive Exophiala

178

dermatitidis disease. Asymptomatic relatives of all ages should be tested for CARD9 deficiency. Detecting

179

cellular defects in asymptomatic individuals is useful for diagnosing CARD9 deficiency.

180
181

Keywords: CARD9 deficiency, invasive fungal disease (IFD), Exophiala dermatitidis, asymptomatic

182

siblings, cytokine production

183

184

Declarations

185

Funding

186

This study was supported by Grants-in-Aid for Scientific Research from the Japan Society for the

187

Promotion of Science (16H05355 and 19H03620 to SO), Promotion of Joint International Research from

188

the Japan Society for the Promotion of Science (18KK0228 to SO), and the Practical Research Project for

189

Rare/Intractable Diseases from Japan Agency for Medical Research and Development, AMED (Grant

190

Number: JP16ek0109179, JP19ek0109209, and JP20ek0109480) to S.O.

191

192

Conflicts of Interest





193

The authors declare that they have no conflicts of interest.

194

195

Availability of data and material

196

The datasets during and/or analyzed during the current study are available from the corresponding author

197

on reasonable request.

198

199

Code availability

200

Not applicable

201

202

Authors’ contributions

203

All authors contributed to the accrual of subjects and/or data. SO contributed to the conception and design

204

of the study. YI, TA, AP, and JLC drafted the manuscript. YI, MT, RK, and YM performed cellular assay

205

and gene expression experiment. MT, TD, RN, MS, SK, KA, KI, TM, KK, and MK performed the clinical

206

work and collected data. OO and SO analyzed data obtained by gene panel sequencing. All authors have

207

revised the manuscript for important intellectual content and approved the final version.

208

209

Ethics approval





210

The study was approved by the Ethics Committees and Institutional Review Board of Hiroshima University.

211

All experiments were carried out with adherence to the Declaration of Helsinki.

212

213

Consent to participate

214

Informed consent was obtained from the guardians of the pediatric patients or directly from participants.

215

216

Consent for publication

217

Informed consent was obtained from the guardians of the pediatric patients for publication of this case

218

report and accompanying images.

219

220

Introduction

221

Caspase-associated recruitment domain-9 (CARD9) deficiency is an autosomal recessive (AR) primary

222

immunodeficiency caused by loss-of-function mutations in the CARD9 gene(1), which encodes a signaling

223

protein located downstream of C-type lectin receptors that recognizes fungal pathogen-associated

224

molecular patterns. Accordingly, AR CARD9 deficiency results in specific susceptibility to invasive and/or

225

superficial fungal disease (2, 3). Since its first report in 2009, AR CARD9 deficiency has been identified

226

in 78 patients from 55 kindreds from 17 countries, with 28 mutations identified as disease causing (1, 3-

227

20). With descriptions of an increasing number of patients, the clinical characteristics, pathophysiology,





228

and genetic background of AR CARD9 deficiency are gradually being deciphered. Nonetheless, many

229

questions remain unanswered (3).

230

AR CARD9 deficiency is characterized by invasive fungal diseases (IFD) that often affect the central

231

nervous system (CNS) (21). Candida and Trychophyton represent the two major disease-causing fungal

232

species in patients with AR CARD9 deficiency (Fig. S1) (3); Aspergillus (8, 14, 19, 22), Auerobasidum

233

(23), Corynespora (7, 24), Exophiala (13, 14, 17, 25), Microsprorum (9), Mucor (6), Ochroconis (17),

234

Pallidocercospora (11), Phialophora (10, 26), Saprochaete (15), and Trichosporon (4) species have less

235

frequently been reported. In particular, IFD caused by Exophiala dermatitidis has only been reported in 2

236

previously healthy patients with AR CARD9 deficiency at the ages of 8 and 23 years (13, 25).

237

Because all patients with disease-causing CARD9 mutations develop fungal disease, the clinical

238

penetrance of AR CARD9 deficiency is thought to be complete (3). However, the age at onset ranges from

239

childhood to adulthood (3.5–58 years) (3, 27), suggesting that there are asymptomatic children or adults

240

who carry disease-causing mutations in CARD9, and such individuals are expected to develop fungal

241

disease later in life. Overall, the mortality rate of CARD9-deficient patients who develop IFD is >20% (3-

242

5, 7, 8, 12, 13, 15, 16, 18). Therefore, to reduce the mortality rate, it is important to diagnose patients with

243

AR CARD9 deficiency prior to the onset of IFD. Presymptomatic diagnosis of this disorder enables us to

244

monitor the patient closely and consider institutional therapy with antifungal prophylaxis. Although

245

diagnosing AR CARD9 deficiency is relatively easy when patients display characteristic clinical features





246

and carry previously reported disease-causing mutations, it becomes more challenging when patients

247

display an atypical clinical course, carry novel CARD9 variants, or carry reported disease-causing mutations

248

but are asymptomatic.

249

250

Materials and methods

251

Fungal identification

252

PrepManTM Ultra Sample Preparation Reagent (Applied Biosystems, Waltham, Massachusetts, USA) was

253

used to extract genomic DNA from a lymph node biopsy that was cultured in Sabouraud dextrose agar

254

according to the manufacturer’s protocol. The DNA was amplified and sequenced from the D2 region of

255

the nuclear large subunit ribosomal RNA gene using MicroSEQ TM D2 rDNA Fungal Identification Kit

256

(Applied Biosystems) according to the manufacturer’s protocol. For species assignment, sequences were

257

aligned using BLAST (NCBI, Washington, DC).

258

259

DNA sequencing

260

Genomic DNA was extracted from peripheral blood leukocytes and subjected to gene panel sequencing

261

and/or Sanger sequencing. The former revealed enriched PID-related genes reported in IUIS2017 (28). The

262

detailed method was described previously (29).

263





264

Isolation of CD14+ monocytes from peripheral whole blood

265

Peripheral blood mononuclear cells (PBMCs) were isolated from peripheral whole blood by density

266

gradient centrifugation using Lymphoprep TM (Alere Technologies AS, Oslo, Norway). CD14+ monocytes

267

were separated from PBMCs using IMag™ Cell Separation System (BD Biosciences, San Jose, CA, USA)

268

according to the manufacturer’s protocol and resuspended in RPMI 1640 medium (Gibco, Thermo Fischer

269

Scientific, Waltham, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS)

270

(HyClone, Logan, UT, USA) and 100 μg/ml penicillin/streptomycin.

271

272

Quantitative PCR

273

Total RNA was extracted from isolated CD14+ monocytes with Qiagen RNeasy Mini kit (Qiagen, Hilden,

274

Germany) according to the manufacturer’s protocol and transcribed by using Superscript III Reverse

275

Transcriptase (Invitrogen, Carlsbad, CA, USA). Quantitative PCR was performed in triplicate using

276

TaqMan primer/probe sets for CARD9 (Hs00364485_m1), GAPDH (Hs99999905_m1) (Applied

277

Biosystems), TaqMan Fast Advanced Master Mix Reagents Kit (Applied Biosystems) according to the

278

manufacturer’s protocol and the StepOne Real-Time PCR system (Applied Biosystems). GAPDH was used

279

as normalization control. The data were analyzed with the 2-ΔΔCT method.

280

281

Immunoblot analysis





282

Equal amounts of protein from isolated CD14+ monocytes were separated by 10% SDS-PAGE and

283

transferred to PVDF membranes (Merck KgaA, Darmstadt, Germany). The membranes were blocked with

284

low-fat bovine milk. Proteins were probed with a rabbit anti-human CARD9 polyclonal antibody (Protein

285

Tech, Thermo Fisher Scientific, Waltham, MA, USA, catalog 10669-1-AP) or a mouse anti-ß-actin

286

monoclonal antibody (Sigma-Aldrich, St. Louis, MO, USA, catalog A5316). HRP-conjugated goat anti-

287

mouse and anti-rabbit antibodies (GE Healthcare, Buckinghamshire, England, UK) were used as secondary

288

antibodies. Antibody binding was detected using enhanced chemiluminescence reagent (Thermo Fisher

289

Scientific, Waltham, MA, USA), and the band intensity was quantified using ImageJ software (National

290

Institutes of Health, Bethesda, MD).

291

292

Mutagenesis and transient transfections

293

We used pcDNA3.1 V5-His-wild-type (WT)-CARD9 and -mutant-CARD9 (p.R35Q and p.R70W), as

294

described previously (21), for this study. We generated expression vectors encoding p.K196E and p.R373P

295

CARD9 variants using PCR-based mutagenesis of the pcDNA3.1 V5-His-WT-CARD9 vector with

296

mismatched PCR primers. The primer sequences and PCR conditions are available upon request.

297

HEK293T cells were plated for 18 h in 6-well plates at 7.5×105 cells/well in DMEM (Gibco) supplemented

298

with 100 μg/ml penicillin/streptomycin. Then, plasmids carrying the WT CARD9 allele or each mutant

299

CARD9 allele were used to transfect HEK293T cells with Lipofectamine LTX Reagent (Thermo Fisher





300

Scientific) according to the manufacturer’s protocol. After 24 h, the transfected HEK293T cells were

301

subjected to immunoblot analysis.

302

303

Luciferase reporter assay

304

HEK293T cells were plated for 18 h in 96-well plates at 2.5×104 cells/well in DMEM (Gibco) supplemented

305

with 100 μg/ml penicillin/streptomycin. The cells were transfected with DECTIN-, SYK-, and BCL10-

306

expressing pcDNA3.1 vectors with the WT CARD9- or mutant CARD9 (p.R70W, p.K196E or p.R373P)-

307

expressing pcDNA3.1 vector, Igkcona-Luc (provided by S. Yamaoka) and pRL-TK (Promega, Madison,

308

Wisconsin, USA) using Lipofectamine LTX Reagent according to the manufacturer’s protocol. The cells

309

were stimulated with heat-killed Exophiala dermatitidis (1×106 particles/well) for 24 h. Luciferase reporter

310

gene activities were determined with Dual-Luciferase Reporter Assay System (Promega). The experiments

311

were performed in triplicate, and data are expressed in relative luciferase units (RLU).

312

313

Cytokine analysis

314

Isolated CD14+ monocytes were cultured in 96-well plates at 4×104 cells/well in RPMI 1640 medium and

315

stimulated with lipopolysaccharide (LPS) (from Escherichia coli, serotype O111: B4; Sigma-Aldrich) (10

316

ng/ml) for 2 h or with heat-killed Candida albicans (1×106 particles/well), heat-killed Candida glabrata

317

(1×106 particles/well), or heat-killed Exophiala dermatitidis (1×106 particles/well) for 24 h. The details of





318

heat-killed fungus preparation are described previously (30). Cytokine levels (TNF-α, IL-6) were measured

319

in the culture supernatants using a cytometric bead array (CBA) (BD Biosciences) and analyzed according

320

to the manufacturer’s instructions using CBA Flex Set (BD Biosciences). The experiments were performed

321

in triplicate.

322

323

Results

324

Case report

325

The patient was a previously healthy 4-year-old Japanese girl born to non-consanguineous parents. There

326

was no history of any severe disease in her parents or her two siblings, a 10-year-old brother and an 8-year-

327

old sister. She received all the vaccines for her age, according to the recommendation by the Japan Pediatric

328

Society, without any adverse effects.

329

At the age of 4 years, she was hospitalized with speech disorder and right hemiparesis that continued for

330

one month. Physical examination showed muscle weakness in the right upper and lower limbs.

331

Lymphadenopathies in the supraclavicular and axillary regions (10 mm) and a mass in the abdomen (30

332

mm) were also noted. Brain magnetic resonance imaging (MRI) revealed multiple masses up to 20 mm in

333

diameter on the left side of the cerebellum, mesencephalon, temporal lobe and basal ganglia (Fig. 1Aa, b).

334

Chest and abdominal computed tomography (CT) scans showed supraclavicular, axillary, and intra-

335

abdominal lymphadenopathies and multiple low-density lesions in the spleen (Fig. 1Ac, d). Cerebrospinal





336

fluid (CSF) leukocyte counts were normal, as were CSF levels of protein and glucose. Blood and CSF

337

cultures were negative for bacterial, fungal, and acid-fast bacilli; gastric juice culture was also negative for

338

acid-fast bacilli. Interferon-gamma release assays (IGRAs) showed negative results, ruling out

339

Mycobacterium tuberculosis infection. Based on histopathology of the axillary lymph nodes, necrotizing

340

granuloma with low neutrophil infiltration was present (Fig. 1Ba, b). Periodic acid Schiff (PAS) and

341

Grocott staining revealed yeast-like fungi (Fig. 1Bc, d). Exophiala dermatitidis was suspected by direct

342

microscopic examination of the fungal culture (Fig. 1C) and was confirmed by sequencing the D2 region

343

of the large subunit ribosomal RNA gene. The patient was thus diagnosed with invasive

344

phaeohyphomycosis (brain, lymph nodes, spleen) due to E. dermatitidis.

345

The patient was initially treated with a 16-mg voriconazole/kg/day infusion as empiric therapy. Her

346

symptoms gradually improved with a month of treatment, though with little impact on the multiple cerebral

347

lesions and systemic lymphadenopathies. She then received 2.5 mg liposomal amphotericin B/kg/day in

348

addition to voriconazole based on the identification and drug sensitivity of E. dermatitidis, and the multiple

349

cerebral lesions and systemic lymphadenopathy gradually improved. After 5 months of administration of

350

liposomal amphotericin B, the multiple cerebral lesions shrank and stabilized, but not fully disappeared.

351

Then the patient was subsequently treated with oral 800 mg voriconazole, and 125 mg terbinafine has been

352

continued to date. Follow-up at 2 years indicated no evidence of recurrence.

353





354

Identification of CARD9 variants

355

Due to the IFD caused by E. dermatitidis in this otherwise healthy 4-year-old girl, we suspected the

356

possibility of an inborn error of immunity and performed gene panel sequencing. After the filtering process

357

(minor allele frequency (MAF) <0.01), 15 rare variants were identified (Table S1). Among them, rare

358

variants in AK2, BCL11B, IL10RA, IL17RC, IRAK1, KMT2D, LRBA, ORAI1, PRF1, SH3BP2, and

359

SLC29A3 were unlikely to be disease causing based on their inheritance patterns or the patient’s clinical

360

phenotype. As no other candidate rare variants that could explain the patient’s manifestations were

361

identified by gene panel sequencing, two variants, c.586A>G (p.K196E) and c.1118G>C (p.R373P), of

362

CARD9 (Fig. 2) were considered to be the best candidates. Both variants were confirmed by Sanger

363

sequencing (Fig. 3A). The p.K196E variant, which was inherited from her asymptomatic mother, has never

364

been reported. In contrast, the p.R373P variant, inherited from her asymptomatic father, has previously

365

been reported as disease causing, either in the homozygous or compound heterozygous state (9, 11, 31).

366

The patient’s 10-year-old brother and 8-year-old sister were totally asymptomatic, even though they were

367

both compound heterozygous for CARD9 p.K196E and p.R373P, similarly to their affected sister (Fig. 3B).

368

Computational assessment of the predicted pathological significance of these two variants using combined

369

annotation-dependent depletion (CADD) showed that their CADD scores (p.K196E: 22.9; p.R373P: 16.0)

370

were higher than the 99% confidence mutation significant cutoff (MSC: 10.26) (32-34); in addition, a low

371

MAF (p.K196E: 4.4 × 10-5; p.R373P: 2.3 × 10-5) in the general population was determined for both. These





372

compound heterozygous variants were thus expected to be very rare, even though each MAF was not much

373

different from that of heterozygous variants reported in the general population (Fig. S2). Moreover, disease-

374

causing nonsense, frameshift, and essential splicing mutations showed lower MAFs and/or higher CADD

375

scores than the homozygous variants reported in the general population. In contrast, some disease-causing

376

missense variants, including the two identified variants p.K196E and p.R373P, had MAFs and/or CADD

377

scores equivalent to those of some homozygous variants reported in the general population (Fig. 3C).

378

Collectively, these data suggest that the identified biallelic CARD9 variants are disease causing and

379

strengthen the importance of functional testing to validate the pathogenicity of identified variants.

380

381

CARD9 mRNA and protein expression

382

We first investigated CARD9 mRNA expression levels in peripheral blood by quantitative PCR. CARD9

383

mRNA was strongly expressed in the neutrophils, monocytes, and natural killer (NK) cells of healthy

384

donors (Fig. S3). Therefore, we assessed CARD9 mRNA levels in the CD14+ monocytes of the patient and

385

found levels comparable to those of two controls tested in parallel (Fig. 4A). We next assessed CARD9

386

protein expression in her CD14+ monocytes by immunoblotting and found levels similar to those of control

387

cells (Fig 4B, C). Taken together, the biallelic variants of CARD9 did not affect mRNA or protein

388

expression in the patient’s cells. To confirm these findings, we transiently expressed WT or mutant

389

p.K196E, p.R373P, p.R35Q, or p.R70W CARD9 alleles in HEK293T cells; p.R35Q and p.R70W have





390

previously been reported as disease causing (8, 18, 21, 35). In cells transfected with the p.K196E or

391

p.R373P allele, CARD9 protein levels were similar to those in cells transfected with the WT, p.R35Q, or

392

p.R70W allele (Fig. 4D, E).

393

394

Functional impact of p.K196E and p.R373P CARD9 alleles

395

We next evaluated the functional impact of each CARD9 allele using an NF-κB reporter assay, as previously

396

reported (21). In cells transfected with the CARD9 p.K196E or p.R373P allele, NF-κB transcriptional

397

activity was comparable to that in cells transfected with the WT allele, both at the basal level and after

398

stimulation with E. dermatitidis. In contrast, cells transfected with the CARD9 p.R70W allele displayed

399

impaired NF-κB transcriptional activity, consistent with a previous report (Fig. 4F) (21). Therefore, the

400

NF-κB reporter assay using HEK293T cells did not allow us to draw a conclusion about the impact of the

401

identified CARD9 variants, and further analyses were carried out.

402

403

Cytokine production in response to fungal stimulation

404

We next evaluated the biological impact of the p.K196E and p.R373P variants by measuring the production

405

of pro-inflammatory cytokines from CD14+ monocytes from the patient, patient’s mother’s or siblings

406

stimulated with heat-killed C. albicans, C. glabrata, E. dermatitidis and LPS. The patient’s CD14+

407

monocytes (p.K196E/p.R373P) displayed markedly impaired TNF-α and IL-6 production after stimulation





408

with C. albicans, C. glabrata, and E. dermatitidis compared with cells from healthy controls (Fig. 5A, B).

409

In contrast, cytokine production following LPS stimulation was normal in the patient’s CD14+ monocytes.

410

Similarly, the CD14+ monocytes from the patient’s asymptomatic siblings (p.K196E/p.R373P) were also

411

markedly impaired in TNF-α and IL-6 production in response to fungal stimulation, which were normal in

412

response to LPS. The CD14+ monocytes from the patient’s mother (p.K196E/WT) displayed an

413

intermediate cellular phenotype; cells from her father (p.R373P/WT) were not available. Altogether, these

414

results showed monocytes carrying biallelic variants, p.K196E/p.R373P, to be impaired with regard to

415

TNF-α and IL-6 production in response to various fungal ligands but normal in response to LPS. These ex

416

vivo observations, together with the clinical manifestations of the patient, suggested that both CARD9

417

mutations are pathogenic.

418

419

Immunological findings

420

The immunological findings for the patient at the age of 4 (before starting antifungal treatment) and 5 (after

421

treatment) years are shown in Tables S1 and S2. Briefly, blood analysis indicated normal percentages of

422

neutrophils, monocytes, and lymphocytes; however, leukocyte counts were high at 15,540/mm3, and the

423

percentages of eosinophils were also high, at 26.1%, before treatment. The serum level of IgE was normal,

424

whereas that of IgG was high at 4,254 mg/dL. The leukocytosis, including eosinophilia, and elevated IgG

425

resolved after antifungal treatments. T lymphocyte proliferation was normal in response to PHA and Con-





426

A. In addition, the leukocyte oxidative burst, as assessed by the dihydrorhodamine (DHR) test, was normal.

427

HIV infection was ruled out by laboratory testing. Furthermore a detailed lymphocyte subpopulation

428

analysis was performed by multicolor flow cytometry, as previously described (36), and the percentages of

429

T, B, and NK cells were within the normal ranges; however, slightly decreased Th17 cell (CCR6+CXCR3-

430

/CD3+CD4+CD45RO+) percentages were noticed.

431

Immunological findings for the patient’s brother (at 12 years) and sister (at 11 years) as well as her mother

432

are shown in Tables S1 and S2. Briefly, blood analysis in the patient’s siblings revealed normal percentages

433

of leukocytes, neutrophils, lymphocytes, and monocytes, though the percentages of eosinophils in her

434

brother were slightly high at 8.9%; serum levels of IgE in the brother and sister were also high, at 339

435

IU/mL and 342 IU/mL, respectively. The percentages of T cells, B cells, and NK cells in the patient’s

436

siblings and mother were within normal ranges, with no decrease in Th17 cell counts.

437

438

Discussion

439

We report a patient with compound heterozygous CARD9 mutations who developed IFD caused by E.

440

dermatitidis, a dematiaceous fungus distributed in the environment (37). Although E. dermatitidis is found

441

worldwide, it is particularly common in East Asia (38). E. dermatitidis is a pathogen that causes a number

442

of clinical manifestations of phaeohyphomycosis, including skin, subcutaneous, and sinus infections. In

443

rare instances, it can cause invasive phaeohyphomycosis in the CNS and liver (13). In a summary report of





444

43 patients with invasive phaeohyphomycosis caused by E. dermatitidis, the state of secondary

445

immunosuppression, including presenting with malignant tumors, cystic fibrosis, and steroid treatment,

446

was reported to involve host factors in 18 patients. Moreover, primary immunodeficiency (AR CARD9

447

deficiency in 1 patient and chronic granulomatous disease in 1 patient (39)) was reported as a host factor;

448

no known host factors were reported for the other 23 cases (25). In patients with primary immunodeficiency,

449

the onset of invasive phaeohyphomycosis caused by E. dermatitidis has only been reported in 1 additional

450

patient aside from those previously mentioned, and this patient was diagnosed with AR CARD9 deficiency

451

(13). Among two patients with AR CARD9 deficiency, one died by severe pneumonia and central nervous

452

infection which resulted in brain herniation (13). The other patient developed IFD, but successfully treated

453

with antifungal therapy. She is alive, although she experienced the recurrence of invasive

454

phaeohyphomycosis caused by E. dermatitidis in spite of antifungal prophylaxis (25). Therefore, our case

455

is the third report of invasive phaeohyphomycosis caused by E. dermatitidis in association with AR CARD9

456

deficiency. The target organs in our patient were the brain, systemic lymph nodes, and spleen. The

457

histopathology of the lymph nodes in our patient showed not only the presence of fungi, but also necrotizing

458

granuloma with low neutrophil infiltration. These findings are consistent to the previous studies which

459

described impaired neutrophil infiltration to the infection sites, such as CSF (21, 31, 40), skin (17, 19, 37),

460

lymph node (22), and adrenal masses (22), in patients with AR CARD9 deficiency. Lack of CXC-

461

chemokine induction at the infection sites have been reported as a cause of impaired neutrophil infiltration





462

(40, 41). CNS disease was reported in both patients with AR CARD9 deficiency who developed invasive

463

E. dermatitidis disease (13, 25). Nevertheless, fungal disease of the CNS has been frequently reported in

464

patients with AR CARD9 deficiency; among 26 patients who developed invasive Candida species disease,

465

20 (76.9%) developed CNS disease (3, 4, 8, 16, 18, 23). Overall, it is suspected that many patients who

466

develop invasive phaeohyphomycosis caused by E. dermatitidis without known host factors have not

467

undergone genetic evaluations. Among these, AR CARD9 deficiency may require differentiation,

468

particularly in patients with CNS disease.

469

In our patient, AR CARD9 deficiency was diagnosed based on the presence of various symptoms,

470

identification of CARD9 mutations and impaired production of pro-inflammatory cytokines specific to

471

fungal stimulation in CD14+ monocytes. Although p.K196E and p.R373P, identified in our patient, are

472

considered loss-of-function mutations, impaired function caused by each mutation could not be adequately

473

evaluated in vitro or computational analysis, MAFs and CADD scores. The CARD9 gene contains 13 exons;

474

the encoded protein has CARD and coiled-coil (CC) domains (42). The mutation p.K196E located in exon

475

4 within the CC domain and p.R373P in exon 8 within the CC domain. p.K196E is a novel mutation,

476

whereas p.R373P is a known disease-causing mutation identified in 3 patients from 3 kindreds (9, 11, 31).

477

CARD9 protein expression in patients with p.R373P homozygous mutations is reportedly normal (11),

478

though it is impaired in patients with p.R373P/p.G72S compound heterozygous mutations (31).

479

Accordingly, there is no consensus on the effect of p.R373P mutation on CARD9 protein expression. In





480

our patient, levels of both CARD9 mRNA and protein expression were normal; hence, p.R373P was

481

determined to be normally expressed at the protein level. The transient gene expression experiment

482

confirmed this finding. Indeed, both p.K196E and p.R373P alleles were normally expression in protein

483

level. Subsequently, we sought to assess the pathological significance of p.K196E and p.R373P mutations

484

using transient gene expression experiments; however, the results of NF-κB transcriptional activity

485

assessment failed to demonstrate dysfunction. Previous study investigated CARD9 mutants in CARD

486

domain (p.R18W, p.R35Q, and p.R70W) and CC domain (p.Q289* and p.Q295*) by NF-κB transcriptional

487

activity. This assay revealed impaired NF-κB activity in three mutations in CARD domain, whereas two

488

mutations in CC domain predicted to have normal activity (21, 25). Since two mutations in CC domain are

489

nonsense and recurrently found in patients with IFD, they should be pathogenic. Therefore, NF-kB reporter

490

assay might not be suitable for evaluating pathogenicity of mutations in CC domain. We thus suspect that

491

NF-kB reporter assay failed to confirm the pathogenicity of p.R373P and p.K196E allele because they

492

locate in CC domain. Including our study, there have been no in vitro evaluations that can accurately

493

measure the effects of CARD9 mutations, and this is a topic for future study.

494

Although our patient’s siblings, a 10-year-old brother and an 8-year-old sister, did not develop fungal

495

disease, similar to the patient, both harbored p.K196E/p.R373P CARD9 mutations. Thus, asymptomatic

496

siblings of all ages should be tested for AR CARD9 deficiency. Because cases of adulthood onset have

497

been reported, it is possible that there are individuals with AR CARD9 deficiency who do not develop





498

fungal disease in childhood. Nonetheless, there have been no reports to date on detailed investigations in

499

presymptomatic individuals carrying disease-causing CARD9 mutations. Indeed, this is the first report of

500

impaired production of pro-inflammatory cytokines against fungi in a patient prior to the onset of fungal

501

disease. This may fit with a previous observation which described complete penetrance of AR CARD9

502

deficiency (3). We started antifungal prophylaxis with oral fluconazole (100 mg/day) and close monitoring

503

of patient’s siblings because they are considered at high risk for future fungal disease. After starting

504

prophylaxis, they have no episodes of fungal infections. On the other side, we need to say that there still

505

remains a possibility that the penetrance of AR CARD9 deficiency is not complete because some of the

506

patients with AR CARD9 deficiency are asymptomatic until middle age (3). Further accumulation of the

507

cases is required to fully understand a global epidemiology of this disorder. Regardless of the presence or

508

absence of fungal disease, a reduction in the production of pro-inflammatory cytokines was demonstrated

509

in this study by using a cellular assay for CD14+ monocytes from both patients and presymptomatic

510

individuals, and this evaluation system might be used to assess the biological effects of CARD9 variants of

511

unknown pathological significance identified using comprehensive genetic analyses.

512

513

Appendix

514

515

Acknowledgment





516

The sequence analysis was supported by the Analysis Center of Life Science, Natural Science Center for

517

Basic Research and Development, Hiroshima University.

518

519

References

520

1.

521

CARD9 mutation in a family with susceptibility to fungal infections. N Engl J Med. 2009;361(18):1727-

522

35.

523

2.

524

Microbiol. 2017;40:123-30.

525

3.

526

Ascomycete Fungi in Previously Healthy Children and Adults. J Clin Immunol. 2018;38(6):656-93.

527

4.

528

patient with CARD9 deficiency and a review of 48 cases. Br J Dermatol. 2019;180(5):1221-5.

529

5.

530

zopfii Colitis in Inherited CARD9 Deficiency. J Infect Dis. 2018;218(3):485-9.

531

6.

532

in a patient with CARD9 deficiency. Br J Dermatol. 2019;180(1):213-4.

533

7.

534

Early-Onset Invasive Infection Due to Corynespora cassiicola Associated with Compound Heterozygous

535

CARD9 Mutations in a Colombian Patient. J Clin Immunol. 2018;38(7):794-803.

536

8.

537

CARD9 Founder Mutation Disrupts NF-kappaB Signaling by Inhibiting BCL10 and MALT1 Recruitment

538

and Signalosome Formation. Front Immunol. 2018;9:2366.

539

9.

540

Microsporum ferrugineum in a patient with CARD9 mutations. Br J Dermatol. 2019;181(5):1093-5.

541

10.

542

americana with CARD9 mutation and 20-year literature review in China. Mycoses. 2019;62(10):908-19.

543

11.

544

Pallidocercospora crystallina-Caused Localized Subcutaneous Phaeohyphomycosis in a Patient with a

545

Homozygous Missense CARD9 Mutation. J Clin Immunol. 2019;39(7):713-25.



Glocker EO, Hennigs A, Nabavi M, Schaffer AA, Woellner C, Salzer U, et al. A homozygous

Shiokawa M, Yamasaki S, Saijo S. C-type lectin receptors in anti-fungal immunity. Curr Opin

Corvilain E, Casanova JL, Puel A. Inherited CARD9 Deficiency: Invasive Disease Caused by

Quan C, Li X, Shi RF, Zhao XQ, Xu H, Wang B, et al. Recurrent fungal infections in a Chinese

Sari S, Dalgic B, Muehlenbachs A, DeLeon-Carnes M, Goldsmith CS, Ekinci O, et al. Prototheca

Wang X, Wang A, Wang X, Li R, Yu J. Cutaneous mucormycosis caused by Mucor irregularis

Arango-Franco CA, Moncada-Velez M, Beltran CP, Berrio I, Mogollon C, Restrepo A, et al.

De Bruyne M, Hoste L, Bogaert DJ, Van den Bossche L, Tavernier SJ, Parthoens E, et al. A

Zhang Y, Mijiti J, Huang C, Song Y, Wan Z, Li R, et al. Deep dermatophytosis caused by

Huang C, Zhang Y, Song Y, Wan Z, Wang X, Li R. Phaeohyphomycosis caused by Phialophora

Guo Y, Zhu Z, Gao J, Zhang C, Zhang X, Dang E, et al. The Phytopathogenic Fungus



546

12.

547

mutation in a patient with chronic invasive dermatophyte infection (tinea profunda). J Cutan Pathol.

548

2020;47(2):166-70.

549

13.

550

Exophiala dermatitidis in a Chinese CARD9-Deficient Patient: A Case Report and Literature Review. Front

551

Neurol. 2019;10:938.

552

14.

553

Deficiency in a Patient with Both Exophiala spinifera and Aspergillus nomius Severe Infections. J Clin

554

Immunol. 2020;40(2):359-66.

555

15.

556

capitata Infection in a Patient with Autosomal Recessive CARD9 Deficiency and a Review of the Literature.

557

J Clin Immunol. 2020;40(3):466-74.

558

16.

559

infection with CARD9 deficiency after G-CSF treatment. Comp Immunol Microbiol Infect Dis.

560

2020;70:101417.

561

17.

562

in CARD9-Deficient Patients with Phaeohyphomycosis. J Invest Dermatol. 2018;138(3):607-17.

563

18.

564

Child With Headaches and Prolonged Fever Associated With Oral and Nail Lesions. Open Forum Infect

565

Dis. 2019;6(11):ofz229.

566

19.

567

Patient with CARD9 Deficiency and Aspergillus Susceptibility of Card9 Knockout Mice. J Clin Immunol.

568

2020.

569

20.

570

Allogenic Stem Cell Transplantation in Patients with Inherited CARD9 Deficiency. J Clin Immunol.

571

2019;39(5):462-9.

572

21.

573

CARD9 deficiency in otherwise healthy children and adults with Candida species-induced

574

meningoencephalitis, colitis, or both. J Allergy Clin Immunol. 2015;135(6):1558-68 e2.

575

22.

576

Aspergillus infection in patients with CARD9 deficiency. JCI Insight. 2016;1(17):e89890.

577

23.

578

splice mutations in CARD9 causing adult-onset Candida endophthalmitis. Mycoses. 2017;12:3218-21.

579

24.

580

cassiicola infection in a Chinese patient. Br J Dermatol. 2016;174(1):176-9.



Nazarian RM, Lilly E, Gavino C, Hamilos DL, Felsenstein D, Vinh DC, et al. Novel CARD9

Wang C, Xing H, Jiang X, Zeng J, Liu Z, Chen J, et al. Cerebral Phaeohyphomycosis Caused by

Perez L, Messina F, Negroni R, Arechavala A, Bustamante J, Oleastro M, et al. Inherited CARD9

Erman B, Firtina S, Aksoy BA, Aydogdu S, Genc GE, Dogan O, et al. Invasive Saprochaete

Du B, Shen N, Hu J, Tao Y, Mo X, Cao Q. Complete clinical remission of invasive Candida

Wang X, Zhang R, Wu W, Song Y, Wan Z, Han W, et al. Impaired Specific Antifungal Immunity

Martin S, Balligand E, Peeters J, Nassogne MC, Mondovits B, Loop M, et al. A 7-Year-Old

Zhang Y, Huang C, Song Y, Ma Y, Wan Z, Zhu X, et al. Primary Cutaneous Aspergillosis in a

Queiroz-Telles F, Mercier T, Maertens J, Sola CBS, Bonfim C, Lortholary O, et al. Successful

Lanternier F, Mahdaviani SA, Barbati E, Chaussade H, Koumar Y, Levy R, et al. Inherited

Rieber N, Gazendam RP, Freeman AF, Hsu AP, Collar AL, Sugui JA, et al. Extrapulmonary

Gavino C, Mellinghoff S, Cornely OA, Landekic M, Le C, Langelier M, et al. Novel bi-allelic

Yan XX, Yu CP, Fu XA, Bao FF, Du DH, Wang C, et al. CARD9 mutation linked to Corynespora



581

25.

582

deficiency in 2 unrelated patients with invasive Exophiala infection. J Infect Dis. 2015;211(8):1241-50.

583

26.

584

phaeohyphomycosis and TH17 cell deficiencies. J Allergy Clin Immunol. 2014;133(3):905-8 e3.

585

27.

586

Genet. 2020;139(6):1011-22.

587

28.

588

Phenotypic Classification for Primary Immunodeficiencies. J Clin Immunol. 2018;38(1):129-43.

589

29.

590

Variant Detection in Cost-Effective Gene Panel Testing by Next-Generation Sequencing. J Mol Diagn.

591

2018;20(5):572-82.

592

30.

593

neutrophil functions against Phialophora verrucosa. Mycopathologia. 2015;179(5-6):347-57.

594

31.

595

fungal infection and impaired neutrophil killing in human CARD9 deficiency. Blood. 2013;121(13):2385-

596

92.

597

32.

598

mutation significance cutoff: gene-level thresholds for variant predictions. Nat Methods. 2016;13(2):109-

599

10.

600

33.

601

visualizing minor allele frequencies and damage prediction scores of human genetic variations.

602

Bioinformatics. 2018;34(24):4307-9.

603

34.

604

deleteriousness of variants throughout the human genome. Nucleic Acids Res. 2019;47(D1):D886-D94.

605

35.

606

Chronic and Invasive Fungal Infections in a Family with CARD9 Deficiency. J Clin Immunol.

607

2016;36(3):204-9.

608

36.

609

Cytometry for the Diagnosis of Primary Immunodeficiency Diseases. J Clin Immunol. 2017;37(5):486-95.

610

37.

611

opportunistic fungal pathogen. Virulence. 2019;10(1):984-98.

612

38.

613

neurotropic black yeast Exophiala dermatitidis has a possible origin in the tropical rain forest. Stud Mycol.

614

2008;61:145-55.



Lanternier F, Barbati E, Meinzer U, Liu L, Pedergnana V, Migaud M, et al. Inherited CARD9

Wang X, Wang W, Lin Z, Wang X, Li T, Yu J, et al. CARD9 mutations linked to subcutaneous

Puel A. Human inborn errors of immunity underlying superficial or invasive candidiasis. Hum

Bousfiha A, Jeddane L, Picard C, Ailal F, Bobby Gaspar H, Al-Herz W, et al. The 2017 IUIS

Fujiki R, Ikeda M, Yoshida A, Akiko M, Yao Y, Nishimura M, et al. Assessing the Accuracy of

Liang P, Wang X, Wang R, Wan Z, Han W, Li R. CARD9 deficiencies linked to impaired

Drewniak A, Gazendam RP, Tool AT, van Houdt M, Jansen MH, van Hamme JL, et al. Invasive

Itan Y, Shang L, Boisson B, Ciancanelli MJ, Markle JG, Martinez-Barricarte R, et al. The

Zhang P, Bigio B, Rapaport F, Zhang SY, Casanova JL, Abel L, et al. PopViz: a webserver for

Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the

Alves de Medeiros AK, Lodewick E, Bogaert DJ, Haerynck F, Van Daele S, Lambrecht B, et al.

Takashima T, Okamura M, Yeh TW, Okano T, Yamashita M, Tanaka K, et al. Multicolor Flow

Kirchhoff L, Olsowski M, Rath PM, Steinmann J. Exophiala dermatitidis: Key issues of an

Sudhadham M, Prakitsin S, Sivichai S, Chaiyarat R, Dorrestein GM, Menken SB, et al. The



615

39.

Kenny RT, Kwon-Chung KJ, Waytes AT, Melnick DA, Pass HI, Merino MJ, et al. Successful

616

treatment of systemic Exophiala dermatitidis infection in a patient with chronic granulomatous disease.

617

Clin Infect Dis. 1992;14:235-42

618

.

619

40.

620

Dependent Neutrophil Recruitment Protects against Fungal Invasion of the Central Nervous System. PLoS

621

Pathog. 2015;11(12):e1005293.

622

41.

623

CARD9(+) microglia promote antifungal immunity via IL-1beta- and CXCL1-mediated neutrophil

624

recruitment. Nat Immunol. 2019;20(5):559-70.

625

42.

626

caspase recruitment domain-containing protein that interacts with BCL10/CLAP and activates NF-kappa

627

B. J Biol Chem. 2000;275(52):41082-6.

Drummond RA, Collar AL, Swamydas M, Rodriguez CA, Lim JK, Mendez LM, et al. CARD9-

Drummond RA, Swamydas M, Oikonomou V, Zhai B, Dambuza IM, Schaefer BC, et al.

Bertin J, Guo Y, Wang L, Srinivasula SM, Jacobson MD, Poyet JL, et al. CARD9 is a novel

628





629
630

Figure 1

631

Image findings and features of the fungus infecting the patient. A Radiological examination of the patient.

632

a and b Brain MRI showed high-intensity lesions on the left side of the mesencephalon, temporal lobe and

633

basal ganglia. c and d Abdominal CT scan showed multiple intra-abdominal lymphadenopathies and

634

multiple low-density lesions in the spleen. B Histopathological and microbiological features of the fungus

635

in the patient. a and b Hematoxylin-eosin staining of the lymph node biopsy specimen showed necrotizing

636

granulomas with low neutrophil infiltration (a 200×, b 400×). c and d Fungi were noted in the lymph node





637

biopsy by Periodic acid Schiff and Grocott staining, respectively (c 200×, d 400×). C Macroscopic

638

appearance of the fungus. Rough colonies of black color on Sabouraud dextrose agar.

639

640





641
642

Figure 2

643

Schematic representation of the human CARD9 protein with the CARD domain (residues 7-98) and coiled-

644

coiled domain (CCD) (residues 140-420). The proband’s variants (p.K196E and p.R373P) are shown in red,

645

among other previously reported pathogenic mutations. The 13 exons are indicated by Roman numerals,

646

and the first exon is nonprotein coding.

647





648
649

Figure 3

650

Identification of CARD9 variants and computational analysis. A Sanger sequencing results. The

651

heterozygous p.K196E variant in exon 4 was present in the patient and her mother. The heterozygous

652

p.R373P variant in exon 8 was present in the patient and her father. B Pedigree of the family. The arrow

653

indicates the proband. C In silico analysis of CARD9 variants. The graph shows the MAF and CADD v1.6

654

scores for disease-causing mutations previously reported in AR CARD9 deficiency and homozygous





655

variants in the general population, gnomAD v2.1.1 (https://gnomad.broadinstitute.org). The red dotted line

656

shows the CADD-MSC score (99% confidence interval) for CARD9. The variants identified in our patient

657

are indicated in red circles. Missense, nonsense, frameshift/essential splicing, and UTR (others) mutations

658

reported in AR CARD9 deficiency are indicated by light blue diamonds, yellow squares, blue squares, and

659

black triangles, respectively. Homozygous missense and essential splicing variants reported in the general

660

population are indicated by white circles and white squares, respectively. CADD scores were calculated at

661

http://cadd.gs.washington.edu. WT, wild-type; MAF, minor allele frequency; CADD, combined

662

annotation-dependent depletion; MSC, mutation significance cutoff.

663
664





665
666

Figure 4

667

CARD9 mRNA and protein expression and NF-κB transcriptional activity. A Relative CARD9 mRNA

668

expression normalized to GAPDH in CD14+ monocytes of the patient and healthy controls (n=2) by

669

quantitative PCR. B, C Immunoblot (B) and quantitative analysis (C) of CARD9 expression in CD14+





670

monocytes of the patient and healthy controls (n=4). The results in C show the ratio of CARD9 to β-actin

671

of each individual analyzed. D, E Immunoblot (D) and quantitative (E) analyses of CARD9 expression in

672

transfected HEK293T cells. The results in E show the ratio of CARD9 to β-actin of each individual analyzed.

673

F NF-κB transcriptional activity in transfected HEK293T cells by the NF-κB luciferase assay. HC, healthy

674

control; WT, wild-type; RLU, relative luciferase units.

675

676





677
678

Figure 5

679

Cytokine production in CD14+ monocytes of the patient (p.K196E/p.R373P), the patient’s brother

680

(p.K196E/p.R373P), the patient’s sister (p.K196E/p.R373P), the patient’s mother (p.K196E/WT) and

681

healthy controls (n=2), stimulated with LPS for 2 h or heat-killed Exophiala dermatitidis, Candida albicans,

682

or Candida glabrata for 24 h, as measured by cytometric bead array analysis. A TNF-α production. B IL-

683

6 production. NS, not stimulated; HC, healthy control





684



rs188378450

rs145374241

.

.

.

.

.

rs141919534

rs12161733

heterozygous

heterozygous

heterozygous

heterozygous

heterozygous

heterozygous

heterozygous

heterozygous

heterozygous

homozygous

heterozygous

heterozygous

.

CARD9

CARD9

IL10RA

IL17RC

IRAK1

KMT2D

KMT2D

KMT2D

LRBA

ORAI1

PRF1

SH3BP2

SLC29A3

rs2252997

rs764213233

rs768281299

rs149712114

.

heterozygous

BCL11B

rs202182972

heterozygous

dbSNP

AK2

Gene

.

0.000009982

0.0014

.

.

.

.

.

.

0.0001

0.00009914

0.00005277

0.00004542

.

0.005

ExAC_ALL

.

0.000013

0.000902

.

.

.

.

.

.

0.000118

0.000081

0.000042

0.000022

.

0.000016

gnomAD_ALL

Table S1 Summary of candidate genes by gene panel sequencing

ProArgProGlyAspThrGlyArg

GCGGCCCGGGGACACGGGCCG

SerSerGlyGlyGln

insAGCTCAGGTGGGCA

c.714_715delTGinsCA

c.1234C>T

c.10C>T

c.138_143delACCGCC

c.80C>G

TCCAGGGGCT

c.5918_5919insAGCCCG

c.5920A>T

p.ThrVal238ThrIle

p.His412Tyr

p.Arg4Cys

p.Pro47_Pro48del

p.Pro27Arg

p.Trp1973fs

p.Thr1974Ser

p.Thr4629Pro

p.CysProProGluLeu485

c.1453_1466delTGCCCACCTGAGCT

c.13885A>C

p.Gly219Ser

p.Gly105Ser

p.Lys196Glu

c.655G>A

c.313G>A

c.586A>G

p.Arg373Pro

p.Arg384_Gly385insCysIleGlyLeu

c.1151_1152insGTGCATAGGGTTGCC

c.1118G>C

p.Gly205Glu

HGVS.p

c.614G>A

HGVS.c



686

685



26.1
3.4
29.4
4,254
83
163
3.2
150
33
375
192
86.9
negative

Eosinophils (%)

Monocytes (%)

Lymphocytes (%)

IgG (mg/dl)

IgA (mg/dl)

IgM (mg/dl)

IgE (IU/ml)

C3 (mg/dl)

C4 (mg/dl)

Proliferative response of lymphocytes to PHA (SI)

Proliferative response of lymphocytes to ConA (SI)

Leukocyte oxidative burst (DHR)(%)

HIV serology

SI: stimulation index, DHR: dihydrorhodamine, NA: not available

40.4

Neutrophils (%)

NA

NA

NA

NA

NA

NA

NA

142

42

920

56.6

5.4

2.5

35.3

8,160

(after treatment)

(before treatment)
15,540

5 years old

4 years old

Leukocytes (/ml)

Patient at

Patient at

Table S2 㻌 Characteristics of the patient, and her siblings

NA

86.5

343

905

15

91

339

114

171

928

44.6

5.8

8.9

40.2

NA

99.3

274

500

24

105

342

118

128

972

39.1

4.5

1.2

54.9

7,270

11 years old

12 years old
5,830

Sister at

Brother at

negative

>80

74.1-1,793

102-2,644

17-45

86-160

<232

46-260

110-410

870-1,700

21.3-50.2

2.7-7.6

0.2-7.3

38.3-71.1

3,040-8,540

Normal values



688

687



85.6

CD45RA+/CD3+CD4+ (naïve) (%)

13.3

CCR6+CXCR3-/CD3+CD4+CD45RO+ (Th17) (%)

18.2
1.89

CD19+/lymphocyte (%)

CD16+CD56+/Lym (%)

B cells

NK cells

Th: helper T, NK: natural killer

24.7

(regulatory T) (%)

2.70

60.1

CCR6-CXCR3-/CD3+CD4+CD45RO+ (Th2) (%)

IL-7R-CD25+/CD3+CD4+ CCR4+

22.2

14.7

CCR6-CXCR3+/CD3+CD4+CD45RO+ (Th1) (%)

(effector memory) (%)

CCR7-CD62 L-/CD3+CD4+CD45RO+

(central memory) (%)

40.2

68.8

CD4+/CD3+ (%)

CCR7+CD62 L+/CD3+CD4+CD45RO+

67.8

8.56

18.4

35.9

2.60

24.6

20.9

25.9

17.2

41.2

64.1

53.1

68.9

old

old

CD3+/lymphocyte (%)

12 years

5 years

CD8+/CD3+ (%)

T cells

Brother at

Patient at

Table S3 Lymphocyte subpopulations of the patient, her siblings, and her mother

10.7

17.0

32.9

1.42

26.7

15.3

33.7

17.6

49.7

71.9

56.4

70.1

old

11 years

Sister at

13.4

10.2

39.2

2.41

41.3

13.3

22.7

18.1

51.3

35.5

56.1

67.5

old

37 years

Mother at

8.8 ± 6.5

16.1 ± 7.4

29.7 ± 6.7

1.65 ± 0.83

22.2 ± 6.2

41.4 ± 10.6

25.0 ± 9.5

24.0 ± 8.8

41.9 ± 11.7

75.9 ± 8.5

60.7 ± 7.3

69.0 ± 9.0

old

2-6 years

7.1 ± 5.8

12.4 ± 6.3

33.4 ± 9.0

2.13 ± 0.60

25.7 ± 4.7

40.2 ± 16.5

23.7 ± 11.1

27.9 ± 10.3

33.0 ± 20.5

65.4 ± 6.0

59.4 ± 4.5

74.9 ± 12.3

old

7-19 years

Normal values

13.4 ± 4.1

12.2 ± 4.4

34.1 ± 8.7

3.11 ± 1.02

23.7 ± 4.3

35.3 ± 13.8

22.6 ± 8.7

30.9 ± 7.9

30.9 ± 7.9

47.2 ± 9.3

59.9 ± 9.9

67.8 ± 5.4

old

>20 years



689

690

Figure S1

691

Causative fungi in patients with AR CARD9 deficiency. The percentage of each fungus causing invasive

692

disease in patients with AR CARD9 deficiency is shown.

693





694

695

696

Figure S2

697

In silico analysis of CARD9 variants. The graph shows the MAF and CADD v1.6 scores for disease-causing

698

variants identified in our patient and heterozygous variants in the general population, gnomAD v2.1.1

699

(https://gnomad.broadinstitute.org). The red dotted line shows the CADD-MSC score (99% confidence

700

interval) for CARD9. The variants identified in our patient are indicated in red circles. Missense, nonsense,

701

frameshift and essential splicing variants in the general population are indicated by light blue diamonds,

702

blue squares, yellow squares, and gray circles, respectively. CADD scores were calculated at

703

http://cadd.gs.washington.edu. MAF, minor allele frequency; CADD, combined annotation-dependent

704

depletion; MSC, mutation significance cutoff.





705
706

707

Figure S3

708

CARD9 mRNA expression in peripheral blood subpopulations. Relative CARD9 mRNA expression

709

normalized to GAPDH in CD66b+ neutrophils, CD14+ monocytes, CD16+56+ NK cells, CD19+ B cells,

710

CD3+4+ T cells and CD3+8+ T cells of healthy controls by quantitative PCR.

711





712

Supplemental materials and methods

713

Cell sorting

714

Peripheral blood cells from healthy donors after the removal of erythrocytes were stained with fluorescently

715

conjugated anti-human CD3, CD4, CD8, CD14, CD16, CD19, CD56, and CD66b (BD Biosciences)

716

antibodies. After surface staining, CD66b+ neutrophils, CD14+ monocytes, CD16+56+ NK cells, CD19+ B

717

cells, CD3+CD4+ T cells and CD3+CD8+ T cells were sorted using a BDFACS AriaTM Cell Sorter (BD

718

Biosciences).

719
720

Quantitative PCR

721

Total RNA was extracted from the sorted cells with the Qiagen RNeasy Mini kit (Qiagen) according to the

722

manufacturer’s protocol. The detailed method of quantitative PCR is described in the materials and methods.

723