NLRP3 Inflammasome Contributes to Host Defense Against Talaromyces

bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1 NLRP3 inflammasome contributes to host defense against

2 Talaromyces marneffei infection

3

4 Haiyan Ma1, Jasper FW Chan2, Yen Pei Tan2, Lin Kui1, Chi-Ching Tsang2, Steven LC Pei1, Yu-

5 Lung Lau1, Patrick CY Woo2, Pamela P Lee1*

6

7 1.Department of Pediatrics and Adolescent Medicine, Li Ka Shing Faculty of Medicine, The

8 University of Hong Kong, Hong Kong SAR, China.

9 2.Department of Microbiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong,

10 Hong Kong SAR, China

11

12 *Correspondence author. E-mail: [email protected]

13

14 Abstract

15 Talaromyces marneffei is an important thermally dimorphic pathogen causing disseminated

16 mycoses in immunocompromised individuals in southeast Asia. Previous study has

17 suggested that NLRP3 inflammasome plays a critical role in antifungal immunity. However,

18 the mechanism underlying the role of NLRP3 inflammasome activation in host defense

19 against T. marneffei remains unclear. We show that T. marneffei yeasts but not conidia

20 induce potent IL-1 response, which is differentially regulated in discrete immune cell types.

21 Dectin-1/Syk signaling pathway mediates pro-IL-1 production, and NLRP3 inflammasome is

22 activated to trigger the processing of pro-IL-1 into IL-1. The activated NLRP3 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

23 inflammasome partially promotes Th1 and Th17 immune responses against T. marneffei

24 yeasts. In vivo, mice with NLRP3 or caspase-1 deficiency exhibit higher mortality rate and

25 fungal load compared to wild-type mice. Herein, our study provides the first evidence that

26 NLRP3 inflammasome contributes to host defense against T. marneffei infection, which may

27 have implications for future antifungal therapeutic designs.

28

29 Introduction

30 Talaromyces marneffei is a dimorphic fungus geographically restricted to Southeast Asia. It

31 causes a form of endemic mycosis that predominantly affects individuals with HIV infection

32 and ranks the third most common opportunistic infection in acquired immunodeficiency

33 syndrome (AIDS), following tuberculosis and cryptococcosis (Sirisanthana & Supparatpinyo,

34 1998; Supparatpinyo et al, 1994; Vanittanakom et al, 2006; Cao et al, 2019). In highly

35 endemic regions such as Chiang Mai in Thailand and Haiphong in Vietnam, T. marneffei is in

36 fact the most common systemic opportunistic fungal infection in AIDS (Vanittanakom et al,

37 2006; Son et al, 2014). Compared with HIV-negative individuals, fungemia, hepatomegaly

38 and splenomegaly occur much more commonly in HIV-positive patients, supporting

39 profound T-lymphopenia as the key predisposing factor for disseminated talaromycosis

40 (Kawila et al, 2013).

41

42 Less commonly, talaromycosis occurs in other immunocompromised conditions such as

43 solid organ or hematopoietic stem cell transplantation, autoimmune diseases or primary

44 immunodeficiency diseases (PID) (Chan et al, 2016; Lee et al, 2012; Lee et al, 2014; Lee &

45 Lau, 2017; Luo et al, 2010; Stathakis et al, 2015). T. marneffei shares some common

46 characteristics with other phylogenetically related dimorphic fungi such as Histoplasma and bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

47 Coccidioides. They exist naturally in the soil environment as conidia which enter the human

48 body via inhalation route, and are converted to the yeast form after being phagocytosed by

49 tissue-resident macrophages. These dimorphic fungal pathogens commonly result in

50 systemic infection when immunity is suppressed, and the uncontrolled proliferation of

51 yeasts in the macrophages eventually lead to dissemination via the reticuloendothelial

52 system. Paracoccidioides brasiliensis and P. lutzii, another dimorphic fungi which is endemic

53 in Latin America, is prevalent in immunocompetent individuals and co-infection with HIV is

54 uncommon, whereas talaromycosis, histoplasmosis and coccidioidomycosis are regarded as

55 AIDS-defining conditions in their respective endemic regions (Limper et al, 2017). They have

56 also been reported in patients with autoantibodies against interferon gamma (anti-IFN),

57 and PID characterized by impaired Th17 immune response such as autosomal dominant (AD)

58 hyper-IgE syndrome caused by loss-of-function STAT3 defect and AD chronic

59 mucocutaneous candidiasis (CMC) caused by gain-of-function STAT1 defect, as well as

60 CD40L deficiency and IFN- receptor deficiency. These acquired defects and inborn errors of

61 immunity provide evidence on the critical role of Th1 and Th17 immune response in host

62 defense against these dimorphic fungi (Lee & Lau, 2017; Lee et al, 2019). However,

63 knowledge about host-pathogen interaction in endemic mycoses is much less

64 comprehensive than other common opportunistic fungal pathogens of worldwide

65 distribution. In particular, little is known about how T. marneffei is recognized by innate

66 immune cells and the ensuing cytokine signaling that triggers adaptive immune response.

67

68 Immune responses against fungal infection are mounted by fungal pathogen-associated

69 molecular patterns (PAMPs) that can be recognized by pathogen recognition receptors (PRRs),

70 including Toll-like receptors (TLRs), C-type lectin receptors (CLRs), NOD-like receptors (NLRs) bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

71 (Erwig & Gow, 2016). CLRs are central for fungal recognition, uptake and induction of innate and

72 adaptive immune responses (Vautier et al, 2012). Dectin-1 specifically recognizes -1,3-glucans

73 in fungal cell wall (Taylor et al, 2007), while Dectin-2 and macrophage inducible C-type lectin

74 (Mincle) recognize -mannan and -mannose, respectively (Vautier et al, 2012). Dectin-1,

75 Dectin-2 and Mincle signal through the Syk-CARD9 pathway which induces NF-kB activation as

76 well as cytokine and chemokine gene expression (Mócsai et al, 2010). NLRP3 is a member of

77 NLRs family, which is able to indirectly sense danger signals, e.g. PAMPs (Schroder & Tschopp,

78 2010). Upon recognition, NLRP3 interacts with adaptor protein ASC that can further recruit pro-

79 caspase-1, leading to the assembly of multiprotein complex known as NLRP3 inflammasome

80 (Latz et al, 2013). Thereafter, pro-caspase-1 can be activated through auto-proteolysis, resulting

81 in the release of caspase-1 p10 and p20 (Schroder & Tschopp, 2010). Active caspase-1 not only

82 processes pro-IL-1 and pro-IL-18 into bioactive IL-1 and IL-18, but also mediates pyroptosis

83 (Man & Kanneganti, 2016). The activation of IL-1 transcription and NLRP3 inflammasome is

84 dependent on the Dectin-1/Syk signaling pathway in human macrophages (Kankkunen et al,

85 2010). The activation of Dectin-1 also triggers the induction of IL-23 through the Syk-CARD9

86 signaling pathway and drives robust Th17 response (LeibundGut-Landmann et al, 2007). The IL-

87 17/IL-23 axis has been shown to promote immunity to C. albicans, Aspergillus and dimorphic

88 fungi such as H. capsulatum (Deepe Jr. & Gibbons, 2009; Chamilos et al, 2010; Wu et al, 2013)

89 and Paracoccidioides brasiliensis (Tristão et al, 2017). In a murine model of disseminated

90 candidiasis, NLRP3 inflammasome exerts antifungal activity through driving protective Th1

91 and Th17 immune responses (van de Veerdonk et al, 2011). NLRP3 inflammasome has also

92 been shown to mediate IL-1 and IL-18 signaling in murine DC and macrophages infected by

93 P. brasiliensis (Tavares et al, 2013; Feriotti et al, 2015; Ketelut-Carneiro et al, 2015), and is bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

94 essential for the induction of protective Th1/Th17 immunity against pulmonary

95 paracoccidioidomycosis (Feriotti et al, 2017; de Castro et al, 2018).

96

97 Previous study showed that IL-1 gene transcription is upregulated in human monocyte-

98 derived DC and macrophages stimulated by T. marneffei (Ngaosuwankul et al, 2008).

99 However, the mechanism of IL-1 induction has not been explored, and whether NLRP3

100 inflammasome plays a role in protective immunity against T. marneffei is unknown. From

101 our earlier work, impaired Th1 and Th17 effector functions emerges as the common

102 pathway in various types of PIDs rendering susceptibility to talaromycosis (Lee et al, 2012;

103 Lee et al, 2014; Lee et al, 2019; Lee & Lau, 2017). In this study, we sought to investigate the

104 role of NLRP3 inflammasome in the induction of Th1 and Th17 immune response to T.

105 marneffei. Our findings show that Dectin-1/Syk signaling pathway mediates pro-IL-1

106 production, and NLRP3 inflammasome is assembled to enhance the processing of pro-IL-1

107 and pro-IL-18 into mature IL-1 and IL-18 in response to T. marneffei yeasts. T. marneffei

108 induces Th1 and Th17 cell differentiation, which is attenuated by caspase-1 inhibition.

109 Furthermore, our in vivo studies show that NLRP3 and caspase-1 confer protection against

110 disseminated T. marneffei infection.

111

112 Results

113 T. marneffei yeasts, but not conidia, induce potent IL-1 response in human PBMCs

114 To dissect the production of cytokines elicited by T. marneffei, freshly isolated human

115 peripheral blood mononuclear cells (PBMCs) were co-cultured with T. marneffei yeasts or

116 conidia at 0.5 multiplicity of infection (MOI) for indicated incubation time points. As shown

117 in Fig 1A, IL-1became detectable at 8 hr, peaked at 18 hr, and remained plateaued bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

118 thereafter till day 5 post infection in the culture supernatant of human PBMCs infected with

119 live T. marneffei yeasts. On the other hand, TNF- production was detectable as early as 4

120 hr, and rapidly peaked at 18 hr, and subsequently showed a slight falling trend up to day 5

121 post infection (Fig 1B). These data demonstrated that T. marneffei yeasts were a potent

122 inducer of IL-1 and TNF-a production. In contrast to T. marneffei yeasts, only minimal levels

123 of IL-1 was detected in PBMCs co-cultured with conidia at 5 days post infection (Fig 1C),

124 whereas TNF- production began to increase steadily after 2 days of incubation (Fig 1D).

125 This suggests that T. marneffei conidia were able to trigger TNF- production in PBMCs but

126 failed to induce IL-1response.

127

128 To examine whether the viability of T. marneffei is a prerequisite for the induction of IL-1

129 in PBMCs, yeasts and conidia were inactivated by heat (100C for 40 min) or 4%

130 paraformaldehyde (PFA) treatment for 10 min. Our results showed that there was no

131 significant difference in IL-1 production in PBMCs infected with live, heat-killed or PFA-

132 treated T. marneffei yeasts (Fig EV1A), indicating that the viability of T. marneffei yeasts was

133 not necessary for the induction of IL-1 response. In contrast, IL-1 production in the

134 PBMCs stimulated with heat-killed conidia was significantly higher compared to live and

135 PFA-treated conidia (Fig EV1B), suggesting that heat treatment might promote the exposure

136 of some pro-inflammatory PAMPs on the conidia cell wall to induce IL-1 response. In

137 addition, both yeast and pseudo-hyphal forms of C. albicans were capable of inducing IL-1

138 production in human PBMCs, and no differential IL-1 response was observed for heat- and

139 PFA-treated C. albicans (Fig EV1C and D).

140 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

141 We further evaluated the production of other pro-inflammatory cytokines in T. marneffei

142 yeasts-infected human PBMCs. As shown in Fig 1E-J, IFN- and IL-17A could be detected only

143 after 3 days of co-culture, whereas IL-6, MCP-1, IL-8 and IL-18 were detectable as early as 18

144 hr. Specifically, IFN-production reached its maximum at 3 days and slightly reduced at 5

145 days, while IL-17A production greatly increased from 3 days to 5 days of co-culture (Fig 1E

146 and F). IL-6 and MCP-1 production peaked at 18 hr and remained at similar levels at 3 days

147 and 5 days of co-culture (Fig 1G and H). IL-8 and IL-18 production was maximal at 18 hr and

148 showed a downward trend thereafter (Fig 1I and J). These results suggested that T.

149 marneffei yeasts sequentially activated innate and adaptive immune responses.

150

151 Differential requirement for IL-1 response to T. marneffei yeasts in various human

152 immune cell types

153 It has been suggested that IL-1 production can be mediated by inflammasome activation

154 which requires two signals: the triggering of pro-IL-1 production (signal 1) and the

155 processing of pro-IL-1 into mature IL-1 (Signal 2) (Latz et al, 2013). To investigate whether

156 there was any differential requirement for IL-1 response to yeasts in different cell types,

157 human PBMCs, monocytes, monocytes-derived macrophages and monocytes-derived DCs

158 were primed with or without lipopolysaccharide (LPS) prior to T. marneffei yeasts

159 stimulation. As shown in Fig 2A and B, T. marneffei yeasts alone were able to induce

160 abundant IL-1 production in PBMCs and monocytes, indicating that yeasts provided ‘signal

161 1’ and ‘signal 2’ for inflammasome activation in these cell types. In contrast, LPS priming

162 was required for IL-1 response to T. marneffei yeasts in macrophages and DCs (Fig 2C and

163 D), suggesting that T. marneffei yeasts only provided ‘signal 2’, and the pro-IL-1induced by

164 LPS (‘signal 1’) was essential in triggering inflammasome activation in these two cell types. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

165 On the other hand, high level of TNF- could be detected in human PBMCs (Fig 2E),

166 monocytes (Fig 2F) and DCs (Fig 2H), but not in macrophages (Fig 2G), after T. marneffei

167 yeast stimulation.

168

169 Dectin-1/syk signaling mediates IL-1 response to T. marneffei yeasts in human

170 monocytes

171 Dectin-1 recognizes -1,3-glucans in fungal cell wall, which leads to pro-inflammatory

172 immune responses (Hardison & Brown, 2012). To elucidate the involvement of Dectin-1 in

173 the initiation of IL-1 response to T. marneffei yeasts, human CD14+ monocytes were co-

174 cultured with heat-killed T. marneffei yeasts in the presence of anti-human Dectin-1 or TLR2

175 neutralizing antibodies or isotype controls (mouse IgG or human IgA). As shown in Fig 3A,

176 Dectin-1 blockade resulted in a marked reduction of IL-1 in T. marneffei yeasts-stimulated

177 monocytes compared with mouse IgG treatment group. TNF- production was slightly

178 reduced by Dectin-1 blockade, though not significantly different (Fig 3B). These data

179 suggested that Dectin-1 mediated the production of IL-1 and TNF- in response to T.

180 marneffei yeasts in human monocytes. Conversely, blockade of TLR2 resulted in significantly

181 elevated IL-1 secretion and had no effect on TNF- production in response to yeasts (Fig

182 3C and D), suggesting that TLR2 negatively regulated IL-1 production in the context of T.

183 marneffei stimulation.

184

185 To study the role of Dectin-1/Syk signaling pathway in the induction of IL-1 response to T.

186 marneffei yeasts, we measured Syk phosphorylation in T. marneffei stimulated human

187 CD14+ monocytes by flow cytometry. As expected, T. marneffei yeasts induced Syk

188 phosphorylation compared to the unstimulated cells (Fig 3E). Pre-treatment of CD14+ bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

189 monocytes with R406 (Syk inhibitor) prior to T. marneffei yeasts stimulation resulted in

190 significantly reduced expression of pro-IL-1 in cell lysates (Fig. 3F) as well as IL-1 in culture

191 supernatant (Fig 3G), in a dose-dependent manner. A similar trend of dose-dependent

192 inhibition of TNF- production in monocytes co-cultured with T. marneffei yeasts was also

193 observed (Fig 3H). Collectively, our data suggested that Syk, the molecule that acts

194 downstream to Dectin-1 in the initiation of antifungal immunity (Drummond et al, 2011),

195 played an important role in eliciting IL-1 response to T. marneffei yeasts in human

196 monocytes.

197

198 To verify our hypothesis that caspase-1 activation mediates IL- and IL-18 release in human

199 monocytes infected with T. marneffei yeasts, we pre-treated human CD14+ monocytes with

200 caspase-1 inhibitor (Z-YVAD, 10µM and 20µM) prior to co-culture with heat-killed T.

201 marneffei yeasts. IL-1 and IL-18 secretion was significantly attenuated in Z-YVAD-treated

202 monocytes (Fig 3I and J), whereas inhibition of caspase-1 activity had no obvious effect on

203 TNF-production (Fig 3K), demonstrating that caspase-1 activation contributed to IL-1 and

204 IL-18 release in response to T. marneffei.

205

206 T. marneffei yeasts trigger IL-1 production via the NLRP3 inflammasome

207 To delineate whether IL-1release induced by T. marneffei yeasts was mediated by NLRP3

208 inflammasome, murine bone marrow-derived DCs (BMDCs) from WT, Nlrp3-/-, Casp1-/- mice

209 were co-cultured with heat-killed yeasts, and then NLRP3, caspase-1 p20 and p45

210 expression were analyzed by immunoblots. As shown in Fig 4A, we found that NLRP3

211 expression was obviously upregulated by yeasts in BMDCs from WT and Casp-1-/- mice.

212 Moreover, pro-caspase-1 (p45) expression in cell lysates was comparable between bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

213 unstimulated and yeasts-stimulated cells in either WT or Nlrp3-/- mice. Active caspase-1 p20

214 was detectable in the concentrated supernatant of yeasts-stimulated BMDCs from WT mice,

215 but it was absent in the supernatant from cells deficient in NLRP3 or caspase-1, suggesting

216 that caspase-1 activation is dependent on NLRP3. LPS plus nigericin, which are considered as

217 the activators of classical NLRP3 inflammasome, could activate caspase-1 since active

218 caspase-1 p20 was observed in both cell lysates and supernatant of WT BMDCs.

219

220 To determine whether T. marneffei yeasts could induce ASC specks to mediate IL-1

221 processing, BMDCs from WT and Nlrp3-/- mice were co-cultured with yeasts, and the

222 formation of ASC pyroptosomes (“specks”) was examined. Confocal microscopy revealed

223 that ASC pyroptosomes, appearing as a single cytoplasmic speck, was present in WT cells

224 but not in Nlrp3-/- cells stimulated with T. marneffei yeasts or LPS plus nigericin (Fig 4B). We

225 also quantified the percentage of cells containing ASC specks and observed that around 8%

226 of WT cells formed ASC specks while it was undetectable in Nlrp3-/- cells after T. marneffei

227 yeasts stimulation (Fig 4C). Similarly, LPS plus nigericin could induce about 29% of WT cells

228 containing ASC specks (Fig 4C). Our results indicated that T. marneffei yeasts induced ASC

229 speck formation, which was dependent on NLRP3, to induce IL-1 processing.

230

231 Next, to further test the hypothesis that IL-1 response to T. marneffei yeasts was

232 dependent on NLRP3 inflammasome, BMDCs from WT, Casp-1-/-, and Nlrp3-/- mice were co-

233 cultured with heat-killed T. marneffei yeasts, and IL-1 and TNF- in the supernatant were

234 examined. As shown in Fig 4D, LPS induced a small amount of IL-1 in these BMDCs,

235 whereas there was remarkably lower IL-1in Casp-1-/- and Nlrp3-/- BMDCs than WT cells

236 after stimulation with LPS plus nigericin. More importantly, we observed that BMDCs from bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

237 Casp-1-/- and Nlrp3-/- mice had significantly impaired IL-1 response to T. marneffei yeasts

238 when compared with WT mice (Fig 4D). Furthermore, an intriguing observation was that IL-

239 1 production in BMDCs from Nlrp3-/- mice was slightly lower than those from Casp-1-/-mice

240 after T. marneffei yeasts stimulation with or without LPS priming (Fig 4D), indicating that IL-

241 1 response to T. marneffei yeasts was more dependent on NLRP3 than caspase-1. In

242 contrast to IL-1 production, BMDCs from WT, Casp-1-/-, and Nlrp3-/- mice produced

243 comparable levels of TNF- when co-cultured with T. marneffei yeasts (Fig 4E). Collectively,

244 these results corroborated that NLRP3 and caspase-1 were required for IL-1 response to T.

245 marneffei yeasts in murine BMDCs.

246

247 T. marneffei yeasts elicit differential Th1 and Th17 immune responses in human CD14+

248 monocytes and monocytes-derived DCs.

249 We have observed that T. marneffei yeasts triggered IFN- and IL-17A production in human

250 PBMCs (Fig 1E and F). To further validate our hypothesis that T. marneffei yeasts would

251 induce Th1 and Th17 immune responses, isolated human CD4+ T cells were co-cultured with

252 autologous monocytes or monocytes-derived DCs with T. marneffei yeasts or C. albicans for

253 7 and 12 days, and intracellular IFN-and IL-17A production in CD4+ T cells was quantified by

254 flow cytometry. Upon co-culture of CD4+ T cells with T. marneffei yeasts, no IFN-- or IL-17A-

255 producing cells could be detected (Figs 5A-C, and EV2A and B). However, when CD4+ T cells

256 were co-cultured with T. marneffei yeasts for 7 and 12 days in the presence of monocytes,

257 robust IFN- and/or IL-17A production was evident (Figs 5A,B, G, and EV2C and D), while no

258 statistically significant differences were observed for IL-17A-producing cells on day 12 (Fig

259 5B) and IFN-+IL-17A+ -producing cells on day 7 and 12 (Fig 5C) after stimulation with T.

260 marneffei yeasts. These findings suggested that the induction of Th1 and Th17 response by T. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

261 marneffei yeasts required the presence of antigen presenting cells (APCs) such as

262 monocytes. We also found that C. albicans yeasts induced much stronger Th1 and Th17

263 immune responses than T. marneffei in the presence of monocytes (Figs 5A-C, and G, and

264 EV2C and D). Consistent with IFN- and IL-17A production, increased T-bet and RORt, the

265 master transcription factors of respective Th1 and Th17 cells, were observed in CD4+ T cells

266 after co-culture with TM yeasts-primed autologous monocytes for 7 days (Fig EV2G and H).

267 Distinct from CD14+ monocytes, monocytes-derived DCs stimulated with T. marneffei or C.

268 albicans yeasts induced delayed Th1 and Th17 cell differentiation; the remarkable increase

269 in the percentage of IFN- -and/or IL-17A-producing CD4+ T cells could mostly be detected

270 on day 12 when T cells were co-cultured with T. marneffei or C. albicans yeasts in the

271 presence of DCs (Figs 5D-F, and EV2E and F).

272

273 In order to further dissect the difference between monocytes and DCs in promoting the

274 differentiation of CD4+ T cells after fungal stimulation, we next assessed the release of IL-

275 12p70 and IL-23, which are considered to be the key cytokines driving Th1 and Th17 cell

276 differentiation respectively, from human CD14+ monocytes and monocyte-derived DCs. We

277 found that the induction of IL-12p70 and IL-23 by T. marneffei was negligible in monocytes

278 (Fig 5I and J). In contrast, IL-23 was induced by T. marneffei in DCs at a level comparable

279 with C. albicans (Fig 5J).

280

281 Taken together, our data showed that T. marneffei yeasts elicited Th1 and Th17 immune

282 responses in the presence of APCs. CD14+ monocytes exhibited earlier induction of IFN--

283 producing and IL-17A-producing CD4+ T cells than monocyte-derived DCs when co-cultured

284 with T. marneffei or C. albicans. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

285

286 Caspase-1 inhibition attenuates Th1 and Th17 immune response to T. marneffei yeasts

287 To characterize the role of activated NLRP3 inflammasome in the Th1 and Th17 immune

288 responses towards T. marneffei yeasts, human PBMCs were co-cultured with T. marneffei

289 yeasts for 5 days in the presence of caspase-1 inhibitor, Z-YVAD. Blocking of caspase-1

290 activation resulted in significantly reduced IFN- and IL-17A release in response to T.

291 marneffei yeasts (Fig 6A). Next, we determined the percentage of IFN-- and/or IL-17A-

292 producing CD4+ T cells upon co-culture of T. marneffei yeasts-stimulated monocytes and

293 CD4+ T cells in the presence of Z-YVAD. Our results showed that caspase-1 inhibition led to

294 approximately 2-fold reduction in overall IL-17A-producing CD4+ T cells (Fig 6B and D) and

295 IFN-/IL-17A-producing CD4+ T cells (Fig 6B and E), although there was no significant effect

296 on the overall intracellular IFN- production (Fig 6B and C). Our data indicated that NLRP3

297 inflammasome activation partially enhanced T. marneffei yeasts-induced Th1 and Th17

298 immune responses in human immune cells in vitro.

299

300 To clarify the role of caspase-1 and NLRP3 in the induction of Th1 and Th17 response to T.

301 marneffei, we isolated murine splenocytes from wild-type, Casp-1-/-, Nlrp3-/- mice and

302 incubated them with T. marneffei yeasts for 2 days or 6 days, and the supernatant was

303 collected for the quantification of IFN- and IL-17A. As shown in Fig 6F, it was observed that

304 splenocytes from Casp-1-/- and Nlrp3-/- mice had impaired IFN- and IL-17A production

305 compared to WT mice at 2 days post stimulation, while there was no statistically significant

306 effect on IL-17A. On day 6 of incubation, IFN- and IL-17A production by Casp-1-/- and Nlrp3-

307 /- splenocytes was sharply increased and reached a similar level as splenocytes from WT bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

308 mice. Our findings suggested that NLRP3 inflammasome activation promoted early stage of

309 IFN- production in murine splenocytes.

310

311 The NLRP3 inflammasome controls antifungal immunity in vivo

312 Our in vitro study has demonstrated that T. marneffei yeasts activated NLRP3

313 inflammasome that promoted Th1 and Th17 immune responses. To examine the protective

314 role of NLRP3 inflammasome in defense against T. marneffei infection in vivo, WT, Casp-1-/-,

315 Nlrp3-/- mice were infected with 5x105 CFU of T. marneffei yeasts per mouse by intravenous

316 injection and their survival was monitored. As shown in Fig 7A, Casp-1-/- and Nlrp3-/- mice

317 were more susceptible to T. marneffei yeasts infection than WT mice. Of note, Casp-1-/- mice

318 showed more resistance to fungal infection than Nlrp3-/- mice. These observations

319 suggested that NLRP3 and caspase-1 conferred host protection against T. marneffei

320 infection, in particular, NLRP3 played a more protective role than caspase-1. To further

321 investigate the factors resulting in the differential survival rates among WT, Casp-1-/-, Nlrp3-

322 /- mice, murine spleens and livers were collected at 7 and 14 days post infection (dpi) for

323 fungal load detection. There was comparable fungal load in spleens from WT, Casp-1-/-,

324 Nlrp3-/- mice at 7 dpi (Fig 7B). Nevertheless, significantly elevated fungal load was observed

325 in spleens from Nlrp3-/- mice when compared to WT mice at 14 dpi (Fig 7B). Similarly, liver

326 specimen from Nlrp3-/- mice exhibited the highest fungal load, followed by Casp-1-/- mice,

327 and WT mice showed the least fungal load in livers at 7 and 14 dpi (Fig 7C). It was worth

328 noting that the fungal load in spleens and livers from WT mice at 14 dpi was similar to that

329 at 7 dpi, whereas the fungal load in these organs from Nlrp3-/- mice at 14 dpi was

330 significantly higher than that 7 dpi (Fig 7B and C), indicating that NLRP3 inflammasome

331 played a pivotal role in the control of T. marneffei proliferation. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

332

333 Histological studies of liver sections from WT, Casp-1-/- and Nlrp3-/- mice at 7 dpi showed a

334 great number of granulomas (Fig 7D, upper panel). By day 14, granulomas were

335 considerably enlarged and almost replaced the liver parenchyma in Casp-1-/- and Nlrp3-/-

336 mice when compared to those at 7 dpi, but liver sections from WT mice exhibited a smaller

337 number of granulomas (Fig 7D, lower panel). HE staining of livers revealed that WT mice had

338 relatively fewer cytoplasmic fungal yeasts within macrophages in the central granulomas

339 than Casp-1-/- and Nlrp3-/- mice (Fig 7D, insets of lower panel). Tissue necrosis was not

340 observed in all three strains of mice. GMS staining further verified that fungal yeasts were

341 present in the center of granulomas (Fig 7E). Slightly fewer yeasts were observed in WT

342 mice than Nlrp3-/- and Casp-1-/- mice at 7 dpi (Fig 7E, upper panel). Importantly, massive

343 number of T. marneffei yeasts were seen in the granulomas of Casp-1-/- and Nlrp3 -/- livers,

344 while T. marneffei yeasts were sparse in the granulomas of WT mice at 14 dpi (Fig 7E, lower

345 panel). Spleen sections of Nlrp3-/- mice showed a more severe loss of follicular structure in

346 the white pulp than WT and Casp-1-/- mice at 7 dpi (Fig EV3A, upper panel). No granulomas

347 were observed in the spleens (Fig EV 3A). We failed to detect T. marneffei in the spleens at 7

348 dpi, but scattered fungal yeasts could be seen in the spleens of WT, Nlrp3-/- and Casp-1-/-

349 mice at 14 dpi by GMS staining (Fig EV3B). In conclusion, these histopathological

350 examinations further demonstrated that NLRP3 inflammasome was essential for host

351 defense against T. marneffei infection in vivo.

352

353 Discussion

354 T. marneffei is an important form of endemic mycoses causing major morbidity and

355 mortality in patients with HIV infection, as well as those with primary and secondary bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

356 immunodeficiencies. However, innate immune recognition of T. marneffei and the induction

357 of adaptive immunity, is largely unknown. For the first time, our study characterizes Syk-

358 and caspase-1 mediated IL-1 response towards T. marneffei in human myeloid cells, and

359 the role of caspase-1 in the induction of Th1 and Th17 response in human lymphocytes.

360 Importantly, our murine model of disseminated talaromycosis demonstrates increased

361 fungal load in the liver and spleen of Casp-1-/- and Nlrp3-/- mice, and correlates with reduced

362 survival.

363

364 Human T. marneffei infection is initiated by inhalation of conidia existing in the environment.

365 After phagocytosis by pulmonary alveolar macrophages, T. marneffei conidia undergo

366 morphological switch and germinate into the pathogenic yeast cells, which readily disseminate

367 throughout the body causing systemic infection. Our findings reveal differential cytokine

368 response towards T. marneffei conidia and yeasts in human PBMCs (Fig 1A-D). These

369 interesting findings indicate that conidia are less immunological than yeasts, as shown by

370 negligible IL-1 and TNF- production in PBMCs during the first 24-48 hours of co-culture.

371 The delayed rise in TNF-, and IL-1 to a lesser extent, after 3 days of co-culture probably

372 represents the switch of conidia to yeast form in vitro, suggesting that the transformation of

373 conidia into yeasts would be required for IL-1 response. The change in cell wall composition

374 and architecture during different life stages of fungi can lead to differential exposure of PAMPs

375 in the inner cell wall, causing varying degree of PRR activation which determines the level of

376 pathogenicity (Cheng et al, 2011).

377

378 Our work demonstrates that IL-1 response to T. marneffei yeasts is partially mediated by

379 Dectin-1/Syk signaling pathway. Blockade of Dectin-1 with neutralizing antibodies impairs bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

380 the production of IL-1 and TNF- to a lesser extent, in response to T. marneffei yeasts in

381 human CD14+ monocytes. This suggests that Dectin-1 plays a role in initiating downstream

382 signaling that leads to pro-IL-1 and TNF- production, which has been demonstrated in

383 other fungal pathogens such as C. albicans and H. capsulatum (Chang et al, 2017; Hise et al,

384 2009). We further show that T. marneffei yeasts are capable of inducing Syk kinase

385 phosphorylation, and blockade of Syk kinase significantly inhibits both pro-IL-1 and IL-1 as

386 well as TNF- in a dose-dependent fashion (Fig 3F-H). Acquired immunity to other

387 dimorphic fungi such as Blastomyces dermatitidis, Histoplasma capsulatum and Coccidioides

388 posadasii infection variably depends on innate sensing by Dectin-1, Dectin-2 and Mincle

389 (Wang et al, 2014), and the role of these CTL receptors in cytokine response against T.

390 marneffei will require further studies.

391

392 TLR2 heterodimerizes with either TLR1 or TLR6 and recognizes triacylated and diacylated

393 lipoprotein. TLR2 and dectin-1 physically associate with one another and synergize to

394 augment anti-fungal response by modulating cytokine production (Dennehy et al, 2009;

395 Gantner et al, 2003; Hise et al, 2009). Our results show that blockade of TLR2 leads to

396 increased IL-1 production but has no impact on TNF- production in T. marneffei yeasts-

397 stimulated human monocytes (Fig 3C and D), suggesting that TLR2 ligation might exert a

398 negative effect on IL-1-mediated immune response against T. marneffei. TLR2-deficient

399 mice infected with P. brasiliensis have preferential activation of Th17 response and lower

400 fungal load, while Treg expansion is diminished and aggravates lung inflammation (Loures et

401 al, 2010). A similar role of TLR2 was also observed in a murine model of disseminated

402 candidiasis, where TLR2 exerts anti-inflammatory effect by promoting IL-10 production and

403 Treg cell proliferation (Netea et al, 2004). bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

404

405 It was previously shown that IL-12p40 production is induced by T. marneffei in murine

406 BMDCs, and is mediated by Dectin-1, TLR2 and MyD88 (Nakamura et al, 2008). Our data

407 showed that IL-12p70 could not be induced by T. marneffei yeasts in human CD14+

408 monocytes and monocyte-derived DCs. This confirms the previous published finding that T.

409 marneffei yeasts do not upregulate IL-12 expression in human monocyte-derived DCs and

410 monocyte-derived macrophages (Ngaosuwankul et al, 2008). Furthermore, our findings

411 provide strong evidence that human Th17 cells are expanded when co-cultured with T.

412 marneffei-stimulated monocytes or monocyte-derived DCs, in contrast to the findings from

413 a recent study showing that murine BMDCs are unable to induce, or actually lower, the

414 percentage of Th17 cells upon co-culture with T. marneffei (Tang et al, 2020). These

415 observations point to the possible difference in Th1 and Th17 immune response towards T.

416 marneffei in human vs murine hosts.

417

418 We demonstrate that IFN- and IL-17A-producing human CD4+ T cells are induced by T.

419 marneffei-primed human CD14+ monocytes (Fig 5A and B and G), albeit less strong as C.

420 albicans-primed monocytes. IL-12 is an important cytokine which induces the differentiation

421 of naïve CD4+ T cell into IFN- secreting Th1 cells, whereas differentiation of naïve CD4+ T-

422 cells to Th17 cells is a result of the combined activity of IL-23 and IL-1 (Goncalves et al,

423 2017; Zhou et al, 2009). As neither IL-12 nor IL-23 could be detectable upon co-culture of T.

424 marneffei-stimulated human CD14+ monocytes (Fig 5I and J), this suggests that cytokines

425 other than IL-12 and IL-23 might be secreted by human monocytes for the differentiation of

426 Th1 and Th17 cells in response to T. marneffei infection. For example, it was found that IL-6

427 contributes to IFN- expression in lung tissues in a murine model of systemic P. brasiliensis bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

428 infection (Tristão et al, 2017). Alternatively, Th17 cells against T. marneffei could originate

429 from memory T cells that are cross-reactive to C. albicans instead of being differentiated de

430 novo, since Th17 cells cross-reactive to C. albicans have been demonstrated to contribute to

431 Th17 response towards A. fumigatus (Bacher et al, 2019).

432

433 On the other hand, IL-23 could be induced by T. marneffei yeasts in monocyte-derived DCs,

434 but not IL-12 (Fig 5J). The preferential induction of IL-23 over IL-12 response in DCs is

435 possibly related to dectin-1-mediated signaling, as previously shown in H. capsulatum

436 (Chamilos et al, 2010). Furthermore, IL-23 is induced by T. marneffei in human DCs at a level

437 comparable to C. albicans, and correlates with a similar percentage of both IFN- and IL-

438 17A-producing CD4+ T-cells upon co-culture with the two fungal pathogens. IL-12 and IL-23

439 are thought to promote mutually exclusive CD4+ Th1 and Th17 cell fates (Teng et al, 2015),

440 However, recent investigations on human monogenic IL12R2 and IL-23R deficiency, which

441 result in defective responses to IL-12 or IL-23 respectively, show that the isolated absence of

442 IL-12 or IL-23 could in part be compensated by its counterpart for the production of IFN-,

443 conferring protection against intracellular pathogen such as mycobacteria (Martínez-

444 Barricate et al, 2018). In the absence of IL-12, IL-17 is the requisite for the protective effect

445 induced by IL-23 in pulmonary histoplasmosis (Deepe Jr. & Gibbons, 2009). This might

446 explain how loss-of-function mutation in STAT3, which encodes the transcription factor

447 activated by IL-23, leads to increased susceptibility to dimorphic fungi such as T. marneffei,

448 C. immitis and H. capsulatum in AD hyper-IgE syndrome (Lee & Lau, 2017).

449

450 IL-1 is essential for Th17 differentiation in human and mice (Acosta-Rodriguez et al, 2007;

451 Chung et al, 2009). IL-1 synergizes with IL-23 in controlling the expression of RORt and IL- bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

452 23 receptor to sustain IL-17 production by effector Th17 cells (Chung et al, 2009). Another

453 report also indicates that NLRP3 inflammasome-derived IL-18 and IL-1 drive protective Th1

454 and Th17 immune responses in disseminated candidiasis (van de Veerdonk et al, 2011).

455 Consistently, treatment of human PBMCs with caspase-1 inhibitor, which suppressed the

456 production of IL-1 and IL-18 production in human monocytes (Fig 3I and J), significantly

457 impaired the secretion of IFN- and IL-17A, and the differentiation of IL-17A-producing-T

458 cells (Fig 6C-E). These findings highlight the important role of caspase-1 in bridging the

459 innate and adaptive immune response, particularly Th17 response towards T. marneffei

460 infection in human cells.

461

462 Furthermore, in line with the role of inflammasome activation in host defense against C.

463 albicans (Hise et al, 2009) and A. fumigatus (Karki et al, 2015), our study defined the critical

464 role of NLRP3 and caspase-1 in the induction of IL-1 response to T. marneffei. In WT

465 BMDCs, T. marneffei yeasts strongly upregulated the expression of NLRP3, cleavage of pro-

466 caspase-1 to caspase-1 p20, and the formation of ASC pyroptosomes (Fig 4A and C). Using

467 the murine model of systemic T. marneffei infection, we found that Nlrp3-/- and Casp-1-/-

468 mice had higher mortality rate. Of note, the fungal load in the spleen and liver of WT mice

469 were comparable on 7 dpi and 14 dpi, while an increase in fungal load became obvious on

470 14 dpi in Casp-1-/- and Nlrp3-/- mice, particularly the latter (Fig 7A-C). This suggests that

471 NLRP3/Caspase-1 signaling is crucial in controlling fungal proliferation in vivo.

472

473 An intriguing observation is that the significantly lower survival rate of Nlrp3-/- mice

474 correlated with higher fungal load, particularly in the liver at 14 dpi, compared with Casp-1-/-

475 mice (Fig 7A-C). These in vivo data correlate well with IL-1 response to T. marneffei in bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

476 murine BMDCs in vitro, reinforcing the critical role of IL-1 in protective immunity against T.

477 marneffei. The residual IL-1 response to T. marneffei in BMDCs from Casp-1-/- and Nlrp3-/-

478 mice (Fig 4D), and particularly the lower fungal load and better survival in Casp-1-/- mice

479 compared to Nlrp3-/- mice (Fig 7A-C), suggests possible contributions from alternate

480 signaling pathways. For example, the engagement of Dectin-1 by P. brasiliensis activates

481 caspase-8 which drives the transcriptional priming and post-translational processing of pro-

482 IL-1. Interestingly, the canonical caspase-1 inflammasome pathway normally restricts the

483 noncanonical caspase-8 inflammasome activation, therefore in the context of caspase-1/11

484 deficiency, the lack of inhibition from caspase-1 enables caspase-8-dependent IL-1

485 processing (Ketelut-Carneiro et al, 2018). A recent study show that IL-1 release via

486 caspase-11-mediated pyroptosis induced by P. brasiliensis reprograms Th17 lymphocytes to

487 produce high levels of IL-17 and enhances effector mechanisms by innate cells (Ketelut-

488 Carneiro et al, 2019). Further studies are required to investigate alternative molecular

489 pathways in the induction of inflammasome activation against T. marneffei infection.

490

491 DCs from patients with HIV was previously shown to exhibit poor activation of NLRP3

492 inflammasome in response to PAMPs and DAMPs, which could be the consequence of

493 increased basal expression and activation of NLRP3 induced by HIV in chronic infection

494 leading to immune exhaustion (Pontillo et al, 2012), as well as the inhibitory effect on

495 NLRP3 inflammasome activity by type I interferon which is commonly elevated in plasma of

496 HIV-infected patients (Reis et al, 2019; Hardy et al, 2013). In response to LPS stimulation,

497 PBMCs from HIV patients express relatively lower levels of activated caspase-1 (Ahmad et al,

498 2002), and IL-1 production in DCs fails to be upregulated by either LPS (Pontillo et al, 2012)

499 or ATP which classically induces canonical activation of NLRP3 (Reis et al, 2019). It is possible bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

500 that functional defect of NLRP3 could represent an additional contributory factor to

501 increased susceptibility to T. marneffei infection in HIV-positive individuals. Further studies

502 on functional evaluation of NLRP3 and caspase-1 activation in HIV-positive patients infected

503 with T. marneffei will be required.

504

505 This study enhances our understanding of host defense mechanisms against T. marneffei.

506 Knowledge about human immune response towards T. marneffei will have therapeutic

507 implications in managing patients suffering from this fatal infection. Delineation of the roles

508 of various cytokines towards protection against T. marneffei will provide important

509 information to the treatment of patients who have secondary immunodeficiency resulting

510 from the use of biologics for immunological disorders. For example, with the increasingly

511 wide clinical use of IL-1receptor antagonist (e.g. Anakinra, Rilonacept) or neutralization

512 antibodies (Canakinumab) for patients with auto-inflammatory or auto-immune diseases,

513 and therefore clinicians should be alerted to the susceptibility and signs of talaromycosis

514 when treating patients who reside in endemic regions where there is increased risk of

515 environmental exposure to T. marneffei. Finally, the study of functional cellular response

516 towards T. marneffei may provide breakthroughs for the discovery of monogenic immune

517 defects in patients with talaromycosis which is not otherwise explained. It is possible that

518 genetic defects in molecules involved in Dectin-1/Syk signaling pathway, NLRP3

519 inflammasome activation and induction of Th1 and Th17 immune responses might

520 predispose to T. marneffei infection.

521

522 To conclude, in this study, we first demonstrate that T. marneffei yeasts rather than conidia

523 induce IL-1response, which is differentially regulated in distinct cell types. Our findings bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

524 show that Dectin-1/Syk signaling pathway mediates pro-IL-1 production upon fungal

525 stimulation. T. marneffei yeasts also trigger the assembly of NLRP3-ASC-caspase-1

526 inflammasome to facilitate IL-1 maturation. The activated inflammasome partially

527 enhances Th1 and Th17 immune responses against fungal infection. Nlrp3-/- and Casp1-/-

528 mice are more susceptible to T. marneffei yeasts with higher fungal load as compared to WT

529 mice. Collectively, our study highlights the importance of NLRP3 inflammasome in host

530 defense against T. marneffei infection and sheds new light on the interaction between host

531 and fungal cells.

532

533 Materials and Methods

534 Fungi

535 T. marneffei and C. albicans were isolated from patients by the Department of Microbiology,

536 Queen Mary Hospital. T. marneffei conidia were cultured on Sabouraud Dextrose Agar (SDA)

537 plates at 25C for 7 days, and then collected in sterile water by wet cotton swabs, washed 3

538 times and re-suspended in sterile water, and finally filtered through 40µm nylon filters. To

539 prepare T. marneffei yeasts, T. marneffei conidia were cultured on SDA plates at 37C for 10

540 days, and then collected in sterile 0.1% PBST by wet cotton swabs, washed 3 times and re-

541 suspended in RPMI 1640 with shaking for 24 hr at 37C, and finally filtered through 40µm

542 nylon filters. C. albicans were cultured on SDA plates at 30℃ for 48 hr. Then, they were

543 transferred to Sabouraud Dextrose (SD) broth and incubated at 37C with shaking for 16–24

544 hr to obtain the yeast form of C. albicans. Simultaneously, C. albicans were cultured with

545 shaking in enriched media consisting of RPMI+10% fetal bovine serum (FBS) at 37C for 24

546 hr to obtain the pseudo-hyphae form of C. albicans. For fixed fungal preparation, fungi were bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

547 washed in sterile PBS, fixed in 4% paraformaldehyde (PFA) solution for 10 min, and then

548 washed three times in sterile PBS. For heat-killed fungal preparation, fungi were boiled at

549 100C for 40 min.

550

551 Cell isolation and culture

552 Peripheral blood mononuclear cells (PBMCs) were isolated from the buffy coats of healthy

553 donors according to the principles of the Declaration of Helsinki and were approved by the

554 responsible ethics committee (Institutional Review Boards of the University of Hong

555 Kong/Hospital Authority Hong Kong West Cluster). Human primary CD14+ monocytes and

556 CD4+ T cells were isolated from PBMCs by magnetic labeling with CD14 or CD4 microbeads

557 respectively, followed by positive selection with MACS columns and separators (Miltenyi

558 Biotec). For preparation of monocyte-derived macrophages, PBMCs were cultured in 10cm

559 dish, and medium was changed gently at 2 hr after cell seeding. Cells were left overnight,

560 and adherent cells were treated with cold 5mM EDTA for 10 min, scrapped and cultured in

561 24-well plates for 12 days. During this period, medium was changed on day 7 and day 9 to

562 remove non-adherent cells. Human PBMCs, CD14+monocytes, monocyte-derived

563 macrophages were cultured in RPMI 1640 supplemented with 5% heat-inactivated

564 autologous sera and 1% penicillin/streptomycin. For monocyte-derived dendritic cells (DCs),

565 CD14+ monocytes were cultured for 5 days in RPMI1640 containing 10% FBS, 1%

566 penicillin/streptomycin, human IL-4 (40ng/ml, PeproTech) and GM-CSF (50ng/ml,

567 PeproTech). Non-adherent or loosely adherent cells were collected.

568

569 Murine bone marrow-derived dendritic cells (BMDCs) were prepared as previously

570 described (Helft et al, 2015). Briefly, bone marrow cells were isolated from the femurs and bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

571 tibias of mice. Red blood cells were lysed on ice for 5 min by using 1x RBC Lysis Buffer

572 (BioLegend, 420301). Bone marrow cells (3x106 per well) were cultured in 6-well plates in 3

573 ml of complete medium (RPMI 1640, 1% penicillin/ streptomycin, 10% FBS, 1% Glutamax, 1%

574 sodium pyruvate, and 55M -mercaptoethanol) supplemented with murine GM-CSF (20

575 ng/ml, Peprotech). Half of the medium was removed on day 2 and fresh medium

576 supplemented with murine GM-CSF (40 ng/ml) was added. The culture medium was entirely

577 aspirated on day 3 and replaced by fresh medium containing GM-CSF (20 ng/ml). Non-

578 adherent cells in the culture supernatant and loosely adherent cells were harvested on day

579 6. The experiments that were performed for Western blot analysis of cell culture

580 supernatant were carried out in Opti-MEMTM serum free medium (Gibco). Murine

581 splenocytes were prepared by homogenization of murine spleen, and subsequent lysis of

582 red blood cells, and cultured in the complete medium.

583

584 Cell stimulation

585 PBMCs (4x106/ml) were co-cultured with live, PFA-treated or heat-killed T. marneffei, as

586 well as live, heat-killed or PFA-treated C. albicans yeasts or pseudohyphae at 0.5 multiplicity

587 of infection (MOI) for indicated time points. PBMCs (4x106/ml), CD14+monocytes (2x106/ml),

588 monocyte-derived macrophages (1x106/ml), or monocyte-derived DCs (1x106/ml) were

589 primed with LPS (10ng/ml for PBMCs and monocytes, 100ng/ml for monocyte-derived

590 macrophages and monocyte-derived DCs, Invivogen) for 3 hr and then stimulated with heat-

591 killed T. marneffei yeasts at 0.5 MOI for 18 hr. Human CD14+ monocytes (2x106/ml) were

592 pre-incubated with human anti-Dectin-1 (10g/ml, Invivogen), anti-TLR2 blocking antibodies

593 (10g/ml, Invivogen), or corresponding isotype controls (10g/ml, Invivogen), or different

594 concentrations of R406 (Invivogen), or Z-YVAD(OMe)-FMK (ALX-260-074, Enzo) for 1 hr bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

595 before co-culture with heat-killed T. marneffei yeasts for 18 hr. Murine BMDCs (2x106/ml)

596 were primed with or without LPS (100 ng/ml) for 3 hr, and then co-cultured with heat-killed

597 T. marneffei yeasts at 1 MOI for additional 18 hr. Alternatively, Nigericin (10 M) was added

598 into LPS-primed cells in the last 40 min for positive control. Total CD4+ T cells (1x105/well)

599 were co-cultured for 7 days or 12 days with monocytes (1x105/well)) or monocyte-derived

600 DCs (0.2x105/well) that were pre-pulsed for 3 hr with heat-killed T. marneffei yeasts

601 (1x105/well) or C. albicans yeasts (1x105/well) in 96-well plates. Moreover, murine

602 splenocytes (2x106/ml) were co-cultured with heat-killed T. marneffei yeasts at 1 MOI for 0,

603 2, 6 day(s).

604

605 Cytokine assays

606 Cytokine concentration in culture supernatant was measured by commercial ELISA kits:

607 human IL-1 (ebioscience), human TNF-, IL-18, and IFN-, IL-17A, IL-12p70, and IL-23 (R&D

608 systems), and murine IL-1, TNF-, IFN- and IL-17A (R&D systems) according to

609 manufacturer’s instructions. IL-6, MCP-1, IL-8, IL-18 and IFN- from human PBMCs were

610 measured by LEGENDplex™ Human Inflammation Panel (Biolegend) according to

611 manufacturer’s instructions.

612

613 Western blot

614 Total cell lysates were prepared using sample buffer containing 5% -mercaptoethanol.

615 Protein concentration was determined by bicinchoninic acid (BCA) assay. Concentrated

616 serum free culture supernatant was prepared as previously described (Hornung et al., 2008).

617 Briefly, cell culture supernatant was precipitated by adding an equal volume of methanol

618 and 0.25 volumes of chloroform. It was vigorously vortexed and centrifuged at 20,000 g for bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

619 10 min, and the upper phase was aspirated and discarded, and 500 l methanol was added

620 and centrifuged at 20,000 g for 10 min. Then protein pellet was dried in 55℃ water baths

621 for 5 min. Proteins from cell lysates (20-50 g) or concentrated supernatant were boiled,

622 separated on SDS-PAGE, and transferred onto PVDF membranes (Bio-Rad Laboratories).

623 Membranes were blocked with 5% bovine serum albumin (BSA) in Tris-buffered saline (TBS)

624 with 0.1% Tween 20 for 1 hr at room temperature, incubated overnight at 4℃ with the

625 following primary antibodies diluted 1:1000 in 5% BSA/TBST: anti-pro-IL-1 (#12703, Cell

626 Signaling Technology) and anti-NLRP3 (AG-20B-0014-C100, Adipogen Life Sciences), and

627 anti-mouse caspase-1 (AG-20B-0042-C100, Adipogen Life Sciences), and then incubated for

628 1 hr with goat anti-rabbit and goat anti-mouse secondary antibodies, respectively, at room

629 temperature. Membranes were washed and exposed by adding enhanced

630 chemiluminescence (ECL) in dark room. When necessary, stained membranes were

631 incubated in stripping buffer at 65℃ for 10 min, and washed 3 times with 0.1% TBS-Tween

632 20, and blocked with 5% BSA for 1 hr, and stained with anti--actin rabbit mAb (4970S, Cell

633 Signaling Technology) followed by goat anti-rabbit secondary antibody staining and film

634 exposure.

635

636 Confocal imaging of ASC specks

637 Murine BMDCs (2x106/ml) were seeded on the chamber slides (154941, Nalge Nunc

638 International) and were co-cultured with heat-killed T. marneffei yeasts at 1 MOI for 18 hr,

639 or were stimulated with LPS (100 ng/ml) for 18 hr with addition of Nigericin (10 M) in the

640 last 40 min. BMDCs were fixed with 4% PFA for 20 min and permeabilized with 0.1% Triton-

641 X-100 for 10 min. Cells were washed 3 times and blocked with 5% BSA for 1 hr, followed by

642 incubation with anti-ASC rabbit polyclonal antibody (SC-22514-R, Santa Cruz Biotechnology) bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

643 overnight at 4C. After washing, cells were incubated with AlexaFlour-488 conjugated goat

644 anti-rabbit secondary antibody (Life Technologies) at room temperature for 1 hr.

645 Subsequently, cells were stained with Alexa Fluor 647 phalloidin (ThermoFisher Scientific,

646 A22287) for 30 min at room temperature avoiding light. Finally, slides with cells were

647 mounted with ProLong Gold Antifade Mountant with DAPI (Invitrogen) and analyzed by

648 confocal microscopy (Zeiss LSM 710). For quantification of ASC speck, 40-150 monocytes per

649 field under microscopy with the magnification of 40x objective were manually analyzed, and

650 the ratio of ASC speck positive cells to total cells was calculated. At least 3 fields per sample

651 were counted.

652

653 Flow cytometry

654 To detect Syk phosphorylation, CD14+ monocytes (2x106/ml) were co-cultured with heat-

655 killed T. marneffei yeasts at 0.5 MOI for 18 hr. Cells were fixed with BD Phosflow Fix Buffer I

656 for 10 min at 37℃, followed by permeabilization with BD Phosflow Perm Buffer III for 30

657 min on ice. After washing, cells were stained with PE-conjugated phospho-Syk (Tyk525/526)

658 antibody (#6485, Cell Signaling Technology) for 30 min on ice, and analyzed by flow

659 cytometer (LSR II, BD). For intracellular staining of IFN- and IL-17A, CD4+ T cells were co-

660 cultured with monocytes or monocyte-derived DCs in the presence of T. marneffei yeasts or

661 C. albicans yeasts for 7 days or 12 days. Cells were then stimulated with PMA and ionomycin

662 for 5 hr and Golgi plug (BD Biosicence) was added in the last 4 hr. Subsequently, cells were

663 stained for 30 min at 4C with FITC-conjugated anti-human CD3 antibody (Biolegend),

664 followed by fixation and permeabilization of cells with BD Cytofix/Cytoperm™ (BD

665 Bioscience). After washing, cells were stained with Alexa Fluor® 647-conjugated anti-human bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

666 IFN- antibody (502516, Biolegend) and PE-conjugated anti-human IL-17A antibody (512305,

667 Biolegend) for 30 min at 4℃. Results were analyzed by flow cytometer (LSR II, BD).

668

669 Murine systemic talaromycosis model

670 For in vivo T. marneffei yeasts infection, WT, Nlrp3-/-, Casp1-/- mice were intravenously

671 injected with 100 l of suspension containing 5x105 colony forming units (CFUs) live T.

672 marneffei yeasts in sterile PBS. Mice survival was daily monitored following infection and

673 mice were sacrificed once they showed signs of the humane endpoints. In the second batch

674 of infection experiments, mice were sacrificed, and spleens and livers were harvested at 7

675 or 14 days post infection (dpi) of T. marneffei yeasts. Part of spleen and liver were weighed

676 and homogenized in sterile PBS, and a series of diluted solutions of cell suspensions were

677 plated onto SDA plates. Fungal load was assessed after culturing for 2 days at 25℃. Part of

678 spleen and liver were fixed in 4% paraformaldehyde (PFA) and paraffin slides were prepared

679 for hematoxylin and esosin (HE) staining and Grocott’s methenamine silver (GMS) staining.

680 The stained sections were mounted and observed under microscope for histopathological

681 assessment.

682

683 Data analysis

684 Statistical analysis was carried out using GraphPad Prism v6.0 software. Figure 5G and H

685 were ploted by R software, version 3.6.1. Data were represented as mean ± SEM. Statistical

686 significance was determined by two-tailed Student’s t test, one-way ANOVA or two-way

687 ANOVA with multiple comparison tests, log-rank test or Gehan-Breslow-Wilcoxon test for

688 comparison of survival rate. p<0.05 was considered statistically significant. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

689

690 Acknowledgements

691 We gratefully acknowledge the Laboratory Animal Unit, the Faculty of Core Facility, and the

692 Histopathological Service Center at the Department of Pathology, the University of Hong

693 Kong, for their professional technical supports. This work was supported by the Edward and

694 Yolanda Wong Fund, the RGC General Research Fund (GRF) (HKU project code: 17111814),

695 and the Health and Medical Research Fund (No. HKM-15-M07 [commissioned project]).

696

697 Author contributions

698 HM, JC, YL, PW, PL designed the experiments and analyzed the results. HM, YT, LK, CT, SP

699 performed the experiments. HM and PL wrote and revised the manuscript.

700

701 Conflict of interest

702 The authors declare that they have no conflict of interest.

703

704 References

705 Acosta-Rodriguez EV, Napolitani G, Lanzavecchia A, Sallusto F (2007) Interleukins 1beta and

706 6 but not transforming growth factor-beta are essential for the differentiation of

707 interleukin 17-producing human T helper cells. Nat immunol 8: 942-9

708 Ahmad R, Sindhu ST, Toma E, Morisset R, Ahmad A (2002) Elevated levels of circulating

709 interleukin-18 in human immunodeficiency virus-infected individuals: role of peripheral

710 blood mononuclear cells and implications for AIDS pathogenesis. J Virol 76: 12448-56

711 Bacher P, Hohnstein T, Beerbaum E, Rocker M, Blango MG, Kaufmann S, Rohmel J,

712 Eschenhagen P, Grehn C, Seidel K, Rickerts V, Lozza L, Stervbo U, Nienen M, Babel N, bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

713 Milleck J, Assenmacher M, Cornely OA, Ziegler M (2019) Human anti-fungal Th17

714 immunity and pathology rely on cross-reactivity against Candida albicans. Cell 176: 1-16

715 Cao C, Xi L, Chaturvedi V (2019) Talaromycosis (Penicilliosis) Due to Talaromyces (Penicillium)

716 marneffei: Insights into the Clinical Trends of a Major Fungal Disease 60 Years After the

717 Discovery of the Pathogen. Mycopathologia 184: 709-720

718 Chamilos G, Ganguly D, Lande R, Gregorio J, Meller S, Goldman WE, Gilliet M, Kontoyiannis

719 DP (2010) Generation of IL-23 producing dendritic cells (DCs) by airborne fungi regulates

720 fungal pathogenicity via the induction of T(H)-17 responses. PLoS One 5: e12955

721 Chan JF, Lau SK, Yuen KY, Woo PC (2016) Talaromyces (Penicillium) marneffei infection in

722 non-HIV-infected patients. Emerg Microbes Infect 5: e19

723 Chang TH, Huang JH, Lin HC, Chen WY, Lee YH, Hsu LC, Netea MG, Ting JP, Wu-Hsieh BA

724 (2017) Dectin-2 is a primary receptor for NLRP3 inflammasome activation in dendritic

725 cell response to Histoplasma capsulatum. PLoS Pathog 13: e1006485

726 Cheng SC, van de Veerdonk FL, Lenardon M, Stoffels M, Plantinga T, Smeekens S, Rizzetto L,

727 Mukaremera L, Preechasuth K, Cavalieri D, Kanneganti TD, van der Meer JW, Kullberg BJ,

728 Joosten LA, Gow NA, Netea MG (2011) The dectin-1/inflammasome pathway is

729 responsible for the induction of protective T-helper 17 responses that discriminate

730 between yeasts and hyphae of Candida albicans. J Leukoc Biol 90: 357-66

731 Chung Y, Chang SH, Martinez GJ, Yang XO, Nurieva R, Kang HS, Ma L, Watowich SS, Jetten

732 AM, Tian Q, Dong C (2009) Critical regulation of early Th17 cell differentiation by

733 interleukin-1 signaling. Immunity 30: 576-87

734 de Castro LF, Longhi LNA, Paião MR, Justo-Júnior ADS, de Jesus MB, Blotta MHSL, Mamoni

735 RL (2018) NLRP3 inflammasome is involved in the recognition of Paracoccidioides bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

736 brasiliensis by human dendritic cells and in the induction of Th17 cells. J Infect 77: 137-

737 144.

738 Deepe GS Jr, Gibbons RS (2009) Interleukins 17 and 23 influence the host response to

739 Histoplasma capsulatum. J Infect Dis 200: 142-51

740 Dennehy KM, Willment JA, Williams DL, Brown GD (2009) Reciprocal regulation of IL-23 and

741 IL-12 following co-activation of Dectin-1 and TLR signaling pathways. Eur J Immunol 39:

742 1379-1386

743 Drummond RA, Saijo S, Iwakura Y, Brown GD (2011) The role of Syk/CARD9 couple C-type

744 lectins in antifungal immunity. Eur J Immunol 41: 276-281

745 Erwig LP, Gow NA (2016) Interactions of fungal pathogens with phagocytes. Nat Rev

746 Microbiol 14: 163-76

747 Feriotti C, Bazan SB, Loures FV, Araújo EF, Costa TA, Calich VL (2015) Expression of dectin-1

748 and enhanced activation of NALP3 inflammasome are associated with resistance to

749 paracoccidioidomycosis. Front Microbiol 6: 913

750 Feriotti C, de Araújo EF, Loures FV, da Costa TA, Galdino NAL, Zamboni DS, Calich VLG (2017)

751 NOD-Like Receptor P3 Inflammasome Controls Protective Th1/Th17 Immunity against

752 Pulmonary Paracoccidioidomycosis. Front Immunol 8: 786

753 Gantner BN, Simmons RM, Canavera SJ, Akira S, Underhill DM (2003) Collaborative

754 Induction of Inflammatory Responses by Dectin-1 and Toll-like Receptor 2. J Exp Med

755 197: 1107-1117

756 Goncalves AC, Ferreira LS, Manente FA, de Faria C, Polesi MC, de Andrade CR, Zamboni DS,

757 Carlos IZ (2017) The NLRP3 inflammasome contributes to host protection during

758 Sporothrix schenckii infection. Immunology 151: 154-166 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

759 Hardison SE, Brown GD (2012) C-type lectin receptors orchestrate antifungal immunity. Nat

760 Immunol 13 (9): 817-822

761 Hardy GA, Sieg S, Rodriguez B, Anthony D, Asaad R, Jiang W, Mudd J, Schacker T, Funderburg

762 NT, Pilch-Cooper HA, Debernardo R, Rabin RL, Lederman MM, Harding CV (2013)

763 Interferon-α is the primary plasma type-I IFN in HIV-1 infection and correlates with

764 immune activation and disease markers. PLoS One 8: e56527

765 Helft J, Bottcher J, Chakravarty P, Zelenay S, Huotari J, Schraml BU, Goubau D, Reis e Sousa C

766 (2015) GM-CSF Mouse Bone Marrow Cultures Comprise a Heterogeneous Population of

767 CD11c(+)MHCII(+) Macrophages and Dendritic Cells. Immunity 42: 1197-211

768 Hise AG, Tomalka J, Ganesan S, Patel K, Hall BA, Brown GD, Fitzgerald KA (2009) An essential

769 role for the NLRP3 inflammasome in host defense against the human fungal pathogen

770 Candida albicans. Cell host microbe 5: 487-97

771 Hornung V, Bauernfeind F, Halle A, Samstad EO, Kono H, Rock KL, Fitzgerald KA, Latz E (2008)

772 Silica crystals and aluminum salts mediate NALP-3 inflammasome activation via

773 phagosomal destabilization. Nat Immunol 9 (8): 847-856

774 Kankkunen P, Teirila L, Rintahaka J, Alenius H, Wolff H, Matikainen S (2010) (1,3)-beta-

775 glucans activate both dectin-1 and NLRP3 inflammasome in human macrophages. J.

776 Immunol 184: 6335-42

777 Karki R, Man SM, Malireddi RK, Gurung P, Vogel P, Lamkanfi M, Kanneganti TD (2015)

778 Concerted activation of the AIM2 and NLRP3 inflammasomes orchestrates host

779 protection against Aspergillus infection. Cell host microbe 17: 357-68

780 Kawila R, Chaiwarith R, Supparatpinyo K (2013) Clinical and laboratory characteristics of

781 penicilliosis marneffei among patients with and without HIV infection in Northern

782 Thailand: a retrospective study. BMC Infect Dis 13: 464 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

783 Ketelut-Carneiro N, Ghosh S, Levitz SM, Fitzgerald KA, da Silva JS (2018) A Dectin-1-Caspase-

784 8 Pathway Licenses Canonical Caspase-1 Inflammasome Activation and Interleukin-1β

785 Release in Response to a Pathogenic Fungus. J Infect Dis 217: 329-339

786 Ketelut-Carneiro N, Souza COS, Benevides L, Gardinassi LG, Silva MC, Tavares LA, Zamboni

787 DS, Silva JS (2019) Caspase-11-dependent IL-1α release boosts Th17 immunity against

788 Paracoccidioides brasiliensis. PLoS Pathog 15: e1007990

789 Latz E, Xiao TS, Stutz A (2013) Activation and regulation of the inflammasomes. Nat Rev

790 Immunol 13: 397-411

791 Lee PP, Chan KW, Lee TL, Ho MH, Chen XY, Li CH, Chu KM, Zeng HS, Lau YL (2012)

792 Penicilliosis in children without HIV infection--are they immunodeficient? Clin Infect Dis

793 54: e8-e19

794 Lee PP, Mao H, Yang W, Chan KW, Ho MH, Lee TL, Chan JF, Woo PC, Tu W, Lau YL (2014)

795 Penicillium marneffei infection and impaired IFN-gamma immunity in humans with

796 autosomal-dominant gain-of-phosphorylation STAT1 mutations. J Allergy Clin Immunol

797 133: 894-6 e5

798 Lee PP, Lau YL (2017) Cellular and Molecular Defects Underlying Invasive Fungal Infections-

799 Revelations from Endemic Mycoses. Front Immunol 8: 735

800 Lee PP, Lao-Araya M, Yang J, Chan KW, Ma H, Pei LC, Kui L, Mao H, Yang W, Zhao X,

801 Trakultivakorn M, Lau YL (2019) Application of Flow Cytometry in the Diagnostics

802 Pipeline of Primary Immunodeficiencies Underlying Disseminated Talaromyces

803 marneffei Infection in HIV-Negative Children. Front Immunol 10: 2189

804 LeibundGut-Landmann S, Gross O, Robinson MJ, Osorio F, Slack EC, Tsoni SV, Schweighoffer

805 E, Tybulewicz V, Brown GD, Ruland J, Reis e Sousa C (2007) Syk- and CARD9-dependent bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

806 coupling of innate immunity to the induction of T helper cells that produce interleukin

807 17. Nat Immunol 8: 630-8

808 Limper AH, Adenis A, Le T, Harrison TS (2017) Fungal infections in HIV/AIDS. Lancet Infect Dis

809 17: e334-e343

810 Loures FV, Pina A, Felonato M, Araújo EF, Leite KR, Calich VL (2010). Toll-like receptor 4

811 signaling leads to severe fungal infection associated with enhanced proinflammatory

812 immunity and impaired expansion of regulatory T cells. Infect Immun 78: 1078-88

813 Luo DQ, Chen MC, Liu JH, Li Z, Li HT (2010) Disseminated Penicillium Marneffei Infection in

814 an SLE Patient: A Case Report and Literature Review. Mycopathologia 171: 191-19

815 Man SM, Kanneganti TD (2016) Converging roles of caspases in inflammasome activation,

816 cell death and innate immunity. Nat Rev Immunol 16: 7-21

817 Martínez-Barricarte R, Markle JG, Ma CS, Deenick EK, Ramírez-Alejo N, Mele F, Latorre D,

818 Mahdaviani SA, Aytekin C, Mansouri D, Bryant VL, Jabot-Hanin F, Deswarte C, Nieto-

819 Patlán A, Surace L, Kerner G, Itan Y, Jovic S, Avery DT, Wong N et al. (2018) Human IFN-γ

820 immunity to mycobacteria is governed by both IL-12 and IL-23. Sci Immunol 3: eaau6759

821 Mócsai A, Ruland J, Tybulewicz VL (2010) The SYK tyrosine kinase: a crucial player in diverse

822 biological functions. Nat Rev Immunol 10: 387-402

823 Nakamura K, Miyazato A, Koguchi Y, Adachi Y, Ohno N, Saijo S, Iwakura Y, Takeda K, Akira S,

824 Fujita J, Ishii K, Kaku M, Kawakami K (2008) Toll-like receptor 2 (TLR2) and dectin-1

825 contribute to the production of IL-12p40 by bone marrow-derived dendritic cells

826 infected with Penicillium marneffei. Microbes Infect 10: 1223-7.

827 Netea MG, Sutmuller R, Hermann C, Van der Graaf CAA, Van der Meer JWM, van Krieken JH,

828 Hartung T, Adema G, Kullberg BJ (2004) Toll-Like Receptor 2 Suppresses Immunity bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

829 against Candida albicans through Induction of IL-10 and Regulatory T Cells. J. Immunol

830 172: 3712-3718

831 Ngaosuwankul P, Pongtanalert P, Engering A, Chaiyaroj SC (2008) Differential gene

832 expression profiles of human monocyte-derived antigen presenting cells in response to

833 Penicillium marneffei: roles of DC-SIGN (CD209) in fungal cell uptake. Asian Pac J

834 Allergy Immunol 26: 151-63.

835 Pontillo A, Silva LT, Oshiro TM, Finazzo C, Crovella S, Duarte AJ (2012) HIV-1 induces NALP3-

836 inflammasome expression and interleukin-1β secretion in dendritic cells from healthy

837 individuals but not from HIV-positive patients. AIDS 26: 11-8

838 Reis EC, Leal VNC, da Silva LT, Dos Reis MML, Argañaraz ER, Oshiro TM, Pontillo A (2019)

839 Antagonistic role of IL-1ß and NLRP3/IL-18 genetics in chronic HIV-1 infection. Clin

840 Immunol 209: 108266

841 Schroder K, Tschopp J (2010) The inflammasomes. Cell 140: 821-32

842 Sirisanthana T, Supparatpinyo K (1998) Epidemiology and management of penicilliosis in

843 human immunodeficiency virus-infected patients. In 8th International Congress on

844 Infectious Diseases, pp 48-53. Boston, Massachusetts:

845 Son VT, Khue PM, Strobel M (2014) Penicilliosis and AIDS in Haiphong, Vietnam: evolution

846 and predictive factors of death. Med Mal Infect 44: 495-501

847 Stathakis A, Lim KP, Boan P, Lavender M, Wrobel J, Musk M, Heath CH (2015) Penicillium

848 marneffei infection in a lung transplant recipient. Transplant infectious disease : an

849 official journal of the Transplantation Society 17: 429-34

850 Supparatpinyo K, Khamwan C, Baosoung V, Nelson KE, Sirisanthana T (1994) Disseminated

851 Penicillium marneffei infection in Southeast Asia. Lancet 344: 110-13 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

852 Tang Y, Zhang H, Xu H, Zeng W, Qiu Y, Tan C, Tang S, Zhang J (2020) Dendritic Cells Promote

853 Treg Expansion but Not Th17 Generation in Response to Talaromyces marneffei Yeast

854 Cells. Infect Drug Resist 13: 805-813

855 Tavares AH, Magalhaes KG, Almeida RD, Correa R, Burgel PH, Bocca AL (2013) NLRP3

856 inflammasome activation by Paracoccidioides brasiliensis. PLoS Negl Trop Dis 7: e25

857 Taylor PR, Tsoni SV, Willment JA, Dennehy KM, Rosas M, Findon H, Haynes K, Steele C, Botto

858 M, Gordon S, Brown GD (2007) Dectin-1 is required for beta-glucan recognition and

859 control of fungal infection. Nat Immunol 8: 31-8

860 Teng MW, Bowman EP, McElwee JJ, Smyth MJ, Casanova JL, Cooper AM, Cua DJ (2015) IL-12

861 and IL-23 cytokines: from discovery to targeted therapies for immune-mediated

862 inflammatory diseases. Nat Med 21: 719-29

863 Tristão FSM, Rocha FA, Carlos D, Ketelut-Carneiro N, Souza COS, Milanezi CM, Silva JS (2017)

864 Th17-Inducing Cytokines IL-6 and IL-23 Are Crucial for Granuloma Formation during

865 Experimental Paracoccidioidomycosis. Front Immunol 8: 949

866 van de Veerdonk FL, Joosten LA, Shaw PJ, Smeekens SP, Malireddi RK, van der Meer JW,

867 Kullberg BJ, Netea MG, Kanneganti TD (2011) The inflammasome drives protective Th1

868 and Th17 cellular responses in disseminated candidiasis. Eur J Immunol 41: 2260-8

869 Vanittanakom N, Cooper CR, Jr., Fisher MC, Sirisanthana T (2006) Penicillium marneffei

870 infection and recent advances in the epidemiology and molecular biology aspects. Clin

871 Microbiol Rev 19: 95-110

872 Vautier S, MacCallum DM, Brown GD (2012) C-type lectin receptors and cytokines in fungal

873 immunity. Cytokine 58: 89-99 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

874 Wang H, LeBert V, Hung CY, Galles K, Saijo S, Lin X, Cole GT, Klein BS, Wüthrich M (2014) C-

875 type lectin receptors differentially induce th17 cells and vaccine immunity to the

876 endemic mycosis of North America. J Immunol 192: 1107-1119

877 Wu SY, Yu JS, Liu FT, Miaw SC, Wu-Hsieh BA (2013) Galectin-3 negatively regulates dendritic

878 cell production of IL-23/IL-17-axis cytokines in infection by Histoplasma capsulatum. J

879 Immunol 190: 3427-37

880 Zhou L, Chong MM, Littman DR (2009) Plasticity of CD4+ T cell lineage differentiation.

881 Immunity 30: 646-55

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

898 Figure Legends

899 Figure 1. T. marneffei yeasts, but not conidia, induce potent IL-1 response in human

900 PBMCs.

901 A, B. Quantification of IL-1 (A) and TNF- (B) by ELISA in human PBMCs (4x106/ml) infected

902 with live T. marneffei yeasts (0.5 MOI) for indicated time points (n=5, mean ± SEM).

903 C, D. ELISA detection of IL-1 (C) and TNF- (D) in human PBMCs (4x106/ml) infected with

904 live T. marneffei conidia (0.5 MOI) for indicated time points (n=5, mean ± SEM).

905 E. Quantification of IFN- by LEGENDplexTM in human PBMCs (4x106/ml) infected with live

906 T. marneffei yeasts (0.5 MOI) for indicated time points (n=5).

907 F. Quantification of IL-17A by ELISA in human PBMCs (4x106/ml) stimulated with heat-

908 killed T. marneffei yeasts (0.5 MOI) for indicated time points (n=4).

909 G-J. Quantitative analyses of cytokines by LEGENDplexTM in human PBMCs (4x106/ml)

910 infected with live T. marneffei yeasts (0.5 MOI) for indicated time points (n=5).

911 Data information: In (A-D), “h” and “d” denote “hours” and “days”, respectively. In (E-J),

912 each symbol represents one healthy donor.

913

914 Figure 2. Differential requirement for IL-1 response to T. marneffei yeasts in various

915 human immune cell types.

916 A-D. Quantitative ELISA detection of IL-1 in human PBMCs (4x106/ml), CD14+ monocytes

917 (2x106/ml), human monocytes-derived macrophages (1x106/ml), CD14+ monocytes-

918 derived dendritic cells (DCs) (1x106/ml), respectively, after stimulation with heat-killed

919 T. marneffei (TM) yeasts (0.5 MOI) for 18 hr with or without LPS priming.

920 E-H. Quantification of TNF- by ELISA in human PBMCs (4x106/ml), CD14+ monocytes

921 (2x106/ml), human monocytes-derived macrophages (1x106/ml), CD14+ monocytes- bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

922 derived dendritic cells (DCs) (1x106/ml), respectively, after stimulation with heat-killed

923 T. marneffei (TM) yeasts (0.5 MOI) for 18 hr with or without LPS priming.

924 Data information: In (A-H), each symbol represents one healthy donor, and data are

925 analyzed by one-way ANOVA. ns =not significant, ****p<0.0001.

926

927 Figure 3. Dectin-1/syk signaling mediates IL-1 response to T. marneffei yeasts in human

928 monocytes.

929 A, B. Quantification of IL-1 and TNF- by ELISA in human CD14+ monocytes (2x106/ml)

930 stimulated with heat-killed T. marneffei (TM) yeasts (0.5 MOI) for 18 hr in the presence

931 of anti-hDectin-1 blocking antibodies or mouse IgG (mIgG) (10 g/ml).

932 C, D. Quantification of IL-1 and TNF- by ELISA in human CD14+ monocytes (2x106/ml)

933 stimulated with heat-killed T. marneffei (TM) yeasts (0.5 MOI) for 18 hr in the presence

934 of anti-hTLR2 blocking antibodies or human IgA (hIgA) (10 g/ml).

935 E. Detection of median fluorescence intensity (MFI) of phospho-Syk by flow cytometry in

936 human CD14+ monocytes (2x106/ml) stimulated with heat-killed T. marneffei (TM)

937 yeasts (0.5 MOI) for 18 hr (n=3, mean ± SEM).

938 F. A representative immunoblot of pro-IL- of cell lysates in human CD14+ monocytes

939 (2x106/ml) stimulated with heat-killed T. marneffei (TM) yeasts (0.5 MOI) for 18 hr in

940 the absence or presence of Syk inhibitor, R406 (n=3).

941 G, H. ELISA quantification of IL-1 and TNF- in the culture supernatant of human CD14+

942 monocytes (2x106/ml) stimulated with heat-killed T. marneffei (TM) yeasts (0.5 MOI)

943 for 18 hr in the absence or presence of Syk inhibitor, R406. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

944 I-K. Quantification of IL-1, IL-18 and TNF-by ELISA in the culture supernatant of human

945 CD14+ monocytes (2x106/ml) stimulated with heat-killed T. marneffei yeasts (0.5 MOI)

946 for 18 hr in the absence or presence of Z-YVAD (caspase-1 inhibitor).

947 Data information: Data are analyzed by paired two-tailed t test (A-E) or by one-way ANOVA

948 (G-K). Each symbol represents one healthy donor. ns =not significant, *p<0.05, **p<0.01,

949 ***p<0.001, ****p<0.0001.

950

951 Figure 4. T. marneffei yeasts trigger IL-1 production via the NLRP3 inflammasome.

952 A. Representative immunoblots of NLRP3, caspase-1 p45 and p20 from cell lysates and

953 caspase-1 p20 from concentrated cell supernatant in WT, Nlrp3-/- and Casp-1-/- murine

954 BMDCs stimulated with heat-killed T. marneffei (TM) yeasts (1 MOI) for 18 hr, or LPS

955 plus nigericin as a positive control (n=3).

956 B, C. Representative confocal micrographs (B) and the percentages (C) of ASC specks

957 formation in WT, Nlrp3-/- murine BMDCs (2x106/ml) stimulated with heat-killed T.

958 marneffei (TM) yeasts (1 MOI) for 18 hr, or LPS plus nigericin as a positive control.

959 D, E. Quantitative detection of IL-1 and TNF- by ELISA in the cell supernatant of WT, Casp-

960 1-/- and Nlrp3-/- murine BMDCs (2x106/ml) stimulated with heat-killed T. marneffei (TM)

961 yeasts (1 MOI) for 18 hr with or without LPS priming, or LPS plus nigericin as a positive

962 control.

963 Data information: In (C), the percentages of ASC speck positive cells are quantified and

964 depicted as mean ± SEM of n=3 mice for each group, and “ND” denotes “not detectable”. In

965 (D, E), data represent mean±SEM of n=4 mice, and data are analyzed by two-way ANOVA.

966 *p<0.05, **p<0.01, ****p<0.0001.

967 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

968 Figure 5. T. marneffei yeasts elicit differential Th1 and Th17 immune responses in human

969 CD14+ monocytes and monocytes-derived DCs.

970 A-C. Analyses of IFN- and/or IL-17A producing CD4 T cells by flow cytometry in human CD4+

971 T cells (1x105/well) co-cultured with heat-killed T. marneffei- or C. albicans-stimulated

972 autologous CD14+ monocytes (1x105/well) for 7 days and 12 days.

973 D-F. Analyses of IFN- and/or IL-17A producing CD4 T cells by flow cytometry in human CD4+

974 T cells (1x105/well) co-cultured with heat-killed T. marneffei- or C. albicans-stimulated

975 autologous monocytes-derived DCs (0.2x105/well) for 7 days and 12 days.

976 G. Scatter plot of overall IFN--producing CD4 T cells (y axis) and overall IL-17A-producing

977 CD4 T cells (x axis) in (A and B).

978 H. Scatter plot of overall IFN--producing CD4 T cells (y axis) and overall IL-17A-producing

979 CD4 T cells (x axis) in (D and E).

980 I. Quantitative ELISA analysis of IL-12p70 in the supernatant of human CD14+ monocytes

981 (Monocytes) (2x106/ml) and monocytes-derived DCs (MDDCs) (1x106/ml) stimulated

982 with T. marneffei yeasts or C. albican yeasts for 18 hr.

983 J. Quantitative ELISA analysis of IL-23 in the supernatant of human CD14+ monocytes

984 (Monocytes) (2x106/ml) and monocytes-derived DCs (MDDCs) (1x106/ml) stimulated

985 with T. marneffei yeasts or C. albican yeasts for 18 hr.

986 Data information: In (A-H), data are given as the percentage of total gated CD3+ cells. In (A-F,

987 and I-J), data are analyzed by two-way ANOVA and depicted as means ± SEM. ns =not

988 significant, *p<0.05, **p<0.01, p<0.0001. Each symbol represents one healthy donor in all

989 the data.

990 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

991 Figure 6. Caspase-1 inhibition attenuates Th1 and Th17 immune responses to T. marneffei

992 yeasts.

993 A. Quantification of IFN- and IL-17A by ELISA in the supernatant of human PBMCs

994 (4x106/ml) stimulated with heat-killed T. marneffei (TM) yeasts (0.5 MOI) in the

995 absence or presence of caspase-1 inhibitor, Z-YVAD (20M), for 5 days.

996 B. Representative plots of intracellular IFN- and/or IL-17A in the CD4+ T cells (1x105/well)

997 co-cultured with heat-killed T. marneffei (TM) yeasts-stimulated autologous CD14+

998 monocytes (1x105/well) in the absence or presence of Z-YVAD (20M) for 7 days.

999 C-E. Statistical analyses of overall IFN--producing cells (C) and overall IL-17A-producing cells

1000 (D) and IFN-+IL-17A+ -producing cells (E) in co-cultured CD4+ T cells and T. marneffei-

1001 primed autologous CD14+ monocytes in the absence or presence of Z-YVAD for 7 days

1002 as described in (B).

1003 F. Detection of murine IFN- and IL-17A by ELISA in the supernatant of WT, Casp-1-/- and

1004 Nlrp3-/- murine splenocytes (2x106/ml) stimulated with heat-killed T. marneffei yeasts

1005 (1 MOI) for indicated time points.

1006 Data information: In (A) and (C-E), each symbol represents one healthy donor, and data are

1007 analyzed by paired two-tailed Student’s t tests. In (F), data are depicted as mean ± SEM of

1008 n=4 mice and are analyzed by two-way ANOVA. ns=not significant, *p<0.05, ***p<0.001,

1009 ****p<0.0001.

1010

1011 Figure 7. The NLRP3 inflammasome controls antifungal immunity in vivo

1012 A. Survival plot of WT, Casp-1-/- and Nlrp3-/- mice intravenously injected with live T.

1013 marneffei yeasts (5x105 CFU per mouse). bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1014 B, C. Fungal load analyses of spleens (B) and livers (C) of WT, Casp-1-/- and Nlrp3-/- mice

1015 infected with live T. marneffei yeasts at 7 and 14 days post infection (dpi).

1016 D. Representative HE staining graphs of murine livers of WT, Casp-1-/- and Nlrp3-/- mice

1017 infected with live T. marneffei yeasts (n=4). Scale bar denotes 500 m. The insets

1018 indicate the magnified area of the smaller box containing granulomas, and the asterisk

1019 indicates massive fungal yeasts in the granulomas.

1020 E. Representative GMS staining graphs of murine livers of WT, Casp-1-/- and Nlrp3-/- mice

1021 infected with live T. marneffei yeasts (n=4). Scale bar denotes 50 m. The insets

1022 indicate the magnified area of the smaller box containing T. marneffei yeasts.

1023 Data information: In (A), the log-rank tests are performed when compared between WT and

1024 Nlrp3-/- mice groups, and between Nlrp3-/- and Casp-1-/- groups; the Gehan-Breslow-

1025 Wilcoxon test is performed between WT and Casp-1-/- groups. In (B and C), data are

1026 depicted as mean ± SEM of n=5 mice and are analyzed by two-way ANOVA. *p<0.05,

1027 **p<0.01, ***p<0.001.

1028

1029 Figure EV1. The viability of T. marneffei yeasts is dispensable for IL-1 response in human

1030 PBMCs

1031 A, B. Quantification of IL-1 by ELISA in the cell supernatant of human PBMCs (4x106/ml)

1032 stimulated with live, heat-killed or PFA-treated T. marneffei yeasts (0.5 MOI) and

1033 conidia (0.5MOI) (n=5) for 18 hr.

1034 C, D. Quantification of IL-1 by ELISA in the cell supernatant of human PBMCs (4x106/ml)

1035 stimulated with live, heat-killed or PFA-treated C. albicans yeasts (0.5 MOI) and

1036 pseudohyphae (0.5MOI) (n=5) for 18 hr. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1037 Data information: In (A-D), data are depicted as mean ± SEM, and are analyzed by one-way

1038 ANOVA. ns =not significant, *p<0.05.

1039

1040 Figure EV2. T. marneffei yeasts elicit differential Th1 and Th17 immune responses in

1041 human CD14+ monocytes and monocytes-derived DCs.

1042 A, B. Representative flow cytometric graphs of IFN- and/or IL-17A producing CD4+ T cells

1043 (1x105/well) in human CD4+ T cells stimulated with heat-killed T. marneffei (TM) yeasts

1044 in the absence of autologous CD14+ monocytes or monocytes-derived DCs for 7 days (A)

1045 and 12 days (B).

1046 C, D. Representative flow cytometric plots of IFN- and/or IL-17A producing CD4+ T cells

1047 (1x105/well) in human CD4+ T cells stimulated with heat-killed T. marneffei (TM) yeasts

1048 in the presence of autologous CD14+ monocytes (1x105/well) for 7 days (C) and 12 days

1049 (D).

1050 E, F. Representative flow cytometric plots of IFN- and/or IL-17A producing CD4+ T cells in

1051 human CD4+ T cells (1x105/well) stimulated with heat-killed T. marneffei (TM) yeasts in

1052 the presence of autologous monocytes-derived DCs (0.2x105/well) for 7 days (E) and 12

1053 days (F).

1054 G, H. Flow cytometric analyses of T-bet (G, n=5) and RORt (H, n=4) in CD4+ T cells co-

1055 cultured with TM yeasts-primed autologous monocytes for 7 days.

1056 Data information: In (G and H), data are depicted as mean ± SEM, and are analyzed by two-

1057 tail Student’s t tests. “MFI” denotes “median fluorescence intensity”. **p<0.01.

1058

1059 Figure EV3. Histopathological analysis of murine livers and spleens. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1060 A. Representative graphs of HE staining of spleens from WT, Nlrp3-/- and Casp-1-/- mice

1061 intravenously infected T. marneffei yeasts (5x105 CFU per mouse) at 7 days and 14

1062 days post infection (dpi) (n=4). Scale bar denotes 500 m.

1063 B. Representative graphs of GMS staining of spleens from WT, Nlrp3-/- and Casp-1-/- mice

1064 intravenously infected T. marneffei yeasts (5x105 CFU per mouse) at 7 days and 14

1065 days post infection (dpi) (n=4). Scale bar denotes 50 m. Arrows indicate T. marneffei

1066 yeasts, and the insets indicate the magnified areas containing T. marneffei yeasts. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

1

2

3 Figure 1. T. marneffei yeasts, but not conidia, induce potent IL-1b response in human

4 PBMCs

5

6 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

7

8

9 Figure 2. Differential requirement for IL-1b response to T. marneffei yeasts in various

10 human immune cell types.

11

12

13

14

15

16 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

17

18

19 Figure 3. Dectin-1/syk signaling mediates IL-1b response to T. marneffei yeasts in human

20 monocytes.

21

22

23

24

25

26

27 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

28

29

30 Figure 4. T. marneffei yeasts trigger IL-1b production via the NLRP3 inflammasome.

31 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

32

33 Figure 5. T. marneffei yeasts elicit differential Th1 and Th17 immune responses in human

34 CD14+ monocytes and monocyte-derived DCs.

35 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

36

37 Figure 6. Caspase-1 inhibition attenuates Th1 and Th17 immune responses to T. marneffei

38 yeasts. bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

39

40

41 Figure 7. The NLRP3 inflammasome controls antifungal immunity in vivo

42

43

44 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

45

46

47

48

49 Figure EV1. The viability of T. marneffei yeasts is for IL-1b response in human PBMCs

50

51

52

53

54

55

56 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

57

58

59 Figure EV2. T. marneffei yeasts elicit differential Th1 and Th17 immune responses in

60 human CD14+ monocytes and monocytes-derived DCs.

61

62

63

64 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

65

66

67 Figure EV3. Histopathological analysis of murine spleens after fungal infection

68

69

70

71

72

73

74

75 bioRxiv preprint doi: https://doi.org/10.1101/2020.08.17.253518; this version posted August 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license.

76 Graphical abstract

77