bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 4.0 International license.

1

1 Regulation of positive and negative selection and TCR signaling

2 during thymic T cell development by capicua

3

4 Soeun Kim1, Guk-Yeol Park1, Jong Seok Park1, Jiho Park1, Hyebeen Hong1, and

5 Yoontae Lee1,2,*

6 1Department of Life Sciences, Pohang University of Science and Technology

7 (POSTECH), Pohang, Gyeongbuk 37673, Republic of Korea

8 2Institute of Convergence Science, Yonsei University, Seoul 03722, Republic of Korea

9

10 *Correspondence to: Yoontae Lee, Room 388, POSTECH Biotech Center, 77

11 Cheongam-Ro, Nam-Gu, Pohang, Gyeongbuk 37673, Republic of Korea, +82-54-279-

12 2354 (phone), +82-54-279-0659 (fax), [email protected] bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 4.0 International license.

2

13 ABSTRACT

14 Central tolerance is achieved through positive and negative selection of thymocytes

15 mediated by T cell receptor (TCR) signaling strength. Thus, the dysregulation of the

16 thymic selection process often leads to autoimmunity. Here, we show that capicua

17 (CIC), a transcriptional repressor that suppresses autoimmunity, controls the thymic

18 selection process. Loss of CIC prior to T-cell lineage commitment impaired both

19 positive and negative selection of thymocytes. CIC deficiency attenuated TCR

20 signaling in CD4+CD8+ double-positive (DP) cells, as evidenced by a decrease in CD5

21 and phospho-ERK levels and calcium flux. We identified Spry4, Dusp4, Dusp6, and

22 Spred1 as CIC target genes that could inhibit TCR signaling in DP cells. Furthermore,

23 impaired positive selection and TCR signaling were partially rescued in Cic and Spry4

24 double mutant mice. Our findings indicate that CIC is a transcription factor required for

25 thymic T cell development and suggest that CIC acts at multiple stages of T cell

26 development and differentiation to prevent autoimmunity. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

3

27 INTRODUCTION

28 T cells play a crucial role in the adaptive immune system’s defense against external

29 invasion. To distinguish between self and non-self, T cells use their T cell receptors

30 (TCRs) to recognize peptide-loaded major histocompatibility complex (MHC)

31 molecules and respond to many types of antigens with an enormous TCR repertoire.

32 T cells with a specific TCR are generated in the thymus through a series of processes,

33 ranging from random rearrangement of TCR gene segments to selection processes

34 that entail apoptosis of inappropriate cells (Klein, Kyewski, Allen, & Hogquist, 2014).

35 T cell development in the thymus is tightly regulated to prevent the generation of

36 nonfunctional or self-reactive T cells. Progenitor cells from bone marrow (BM) become

37 CD4-CD8- double-negative (DN) cells and undergo a process called -selection, which

38 selects only T cells with a functional TCR chain. DN cells that have passed -

39 selection then become CD4+CD8+ double-positive (DP) cells and undergo positive

40 selection, which selects T cells that bind to self-peptide ligands loaded onto MHC (self-

41 pMHC) molecules. These CD4+CD8+ DP cells then develop into CD4+CD8- single-

42 positive (CD4+ SP) or CD4-CD8+ single-positive (CD8+ SP) T cells. Negative selection,

43 also known as clonal deletion, selectively removes T cells that bind with high affinity

44 to self-pMHC during the DP and SP stages. Because the intensity and duration of TCR

45 signaling based on TCR affinity to self-pMHC are the major determinants of selection

46 (Gascoigne, Rybakin, Acuto, & Brzostek, 2016), defects in the TCR signaling

47 component lead to abnormal T cell development and alteration of the TCR repertoire

48 (Guoping Fu et al., 2010; Sakaguchi et al., 2003). Impairment of thymic selection

49 caused by decreased TCR signaling destroys central tolerance, and consequently

50 induces autoimmunity accompanied by the expansion of the CD44hiCD62Llo activated bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

4

51 effector/memory T cell population (Guoping Fu et al., 2010; Sakaguchi et al., 2003;

52 Sommers et al., 2002).

53 Capicua (CIC) is an evolutionarily conserved transcriptional repressor that regulates

54 the receptor tyrosine kinase (RTK) signaling pathway in Drosophila and mammals

55 (Jiménez, Guichet, Ephrussi, & Casanova, 2000; Jiménez, Shvartsman, & Paroush,

56 2012). CIC is expressed as two different isoforms: long (CIC-L) and short (CIC-S),

57 which differ in their amino termini (Y. Lee, 2020). CIC recognizes specific octameric

58 DNA sequences (5ʹ-T(G/C)AATG(A/G)(A/G)-3ʹ) within its target gene promoter

59 regions and represses their expression (Kawamura-Saito et al., 2006; Shin & Hong,

60 2014; Weissmann et al., 2018). Several genomic and transcriptomic analyses have

61 identified CIC target genes, including ETV1, ETV4, ETV5, SPRY4, SPRED1, DUSP4,

62 and DUSP6, in various types of cells (Fryer et al., 2011; Weissmann et al., 2018; Yang

63 et al., 2017). Activation of the RTK/RAS/MAPK signaling pathway phosphorylates and

64 inactivates CIC via dissociation from target gene promoters, cytoplasmic translocation,

65 and/or proteasomal degradation (Jiménez et al., 2012; Keenan et al., 2020; Y. Lee,

66 2020). CIC also mediates the ERK-DUSP6 negative feedback loop to maintain ERK

67 activity within the physiological range in mammals (Ren et al., 2020).

68 We previously reported that murine CIC deficiency spontaneously induces

69 lymphoproliferative autoimmune-like phenotypes (S. Park et al., 2017). Hematopoietic

70 lineage cell-specific (Vav1-Cre-mediated knockout) and T cell-specific (Cd4-Cre-

71 mediated knockout) Cic-null mice commonly exhibit autoimmune-like symptoms

72 including hyperglobulinemia, increased serum anti-dsDNA antibody levels, and tissue

73 infiltration of immune cells, accompanied by increased frequency of CD44hiCD62Llo T

74 and follicular helper T (Tfh) cells in the spleen (S. Park et al., 2017). However, these

75 phenotypes were more severe in Cicf/f;Vav1-Cre mice than in Cicf/f;Cd4-Cre mice; bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

5

76 moreover, enlargement of secondary lymphoid organs was observed in Cicf/f;Vav1-

77 Cre mice but not in Cicf/f;Cd4-Cre mice (G. Park et al., 2020; S. Park et al., 2017).

78 These results indicated that deletion of Cic alleles in hematopoietic stem and

79 progenitor cells leads to more severe peripheral T cell hyperactivation and

80 autoimmunity than the Cd4-Cre-mediated Cic deletion in CD4+CD8+ DP thymocytes.

81 We also reported that enhanced peripheral T cell hyperactivation in Cicf/f;Vav1-Cre

82 mice relative to Cicf/f;Cd4-Cre mice was caused by CIC deficiency in T cells rather

83 than in other types of immune cells (G. Park et al., 2020), suggesting that this

84 abnormality could be caused by impaired control of early thymic T cell development in

85 Cicf/f;Vav1-Cre mice. It was reported that the frequency of CD4-CD8-CD44+CD25- DN1

86 cells was increased in adult stage-specific Cic-null mice (Tan et al., 2018). Taken

87 together, these studies suggest that CIC plays a crucial role in thymic T cell

88 development.

89 In this study, we found that CIC regulates thymic T cell development from the CD4-

90 CD8- DN stage and positive and negative selection of thymocytes, primarily during the

91 CD4+CD8+ DP stage. TCR signaling was significantly attenuated in DP cells of

92 Cicf/f;Vav1-Cre mice, thereby impairing both positive and negative selection in

93 Cicf/f;Vav1-Cre mice. We also identified Spry4, Dusp4, Dusp6, and Spred1 as CIC

94 target genes that could potentially contribute to the reduced TCR signaling strength

95 and impaired thymic selection processes in Cicf/f;Vav1-Cre mice. Our findings

96 demonstrate that CIC is a critical regulator of TCR signaling in DP cells and thymic T

97 cell development. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

6

98 RESULTS

99 Changes in the frequencies of thymic T cell subsets over development in

100 Cicf/f;Vav1-Cre mice

101 To examine the role of CIC in thymic T cell development, we first evaluated the levels

102 of CIC in multiple subsets of developing thymocytes using homozygous FLAG-tagged

103 Cic knock-in (CicFLAG/FLAG) mice (S. Park, Park, Kim, & Lee, 2019). Flow cytometry for

104 FLAG-CIC revealed that CIC levels were relatively high in CD4-CD8- DN and immature

105 CD8+ single positive (ISP) cells compared to cells at later developmental stages, such

106 as CD4+CD8+ DP and CD4+ or CD8+ SP subsets (Fig. 1A). We then determined the

107 frequency and number of thymic T cell subsets in Cicf/f (WT), Cicf/f;Cd4-Cre, and

108 Cicf/f;Vav1-Cre mice at 7 weeks of age. The total numbers of thymocytes were

109 comparable among WT, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice (Fig. 1B). However,

110 T cell development during the DN stage was abnormal in Cicf/f;Vav1-Cre mice, as

111 evidenced by an increased frequency of CD44hiCD25hi DN2 and CD44loCD25hi DN3

112 subsets at the expense of the CD44loCD25lo DN4 subset (Fig. 1C). As expected, these

113 changes were not detected in Cicf/f;Cd4-Cre mice (Fig. 1C), because Cd4-Cre was not

114 expressed in DN cells (P. P. Lee et al., 2001). We previously reported that the

115 frequency of DN, DP, and SP cells was comparable between WT and Cicf/f;Vav1-Cre

116 mice at 9 weeks of age (S. Park et al., 2017). However, at 7 weeks of age, Cicf/f;Vav1-

117 Cre mice exhibited a mild block in formation of SP cells in the thymus: the frequency

118 of DP cells was significantly increased in Cicf/f;Vav1-Cre mice compared to WT and

119 Cicf/f;Cd4-Cre mice, whereas that of SP cells decreased (Fig. 1D). This difference was

120 not statistically significant when calculated using cell numbers (Fig. 1D). To clarify

121 whether CIC deficiency affected thymic SP cell formation, we performed the same

122 analysis using WT and Cicf/f;Vav1-Cre mice at 1 week of age, when the proportion of bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

7

123 peripheral T cells that recirculated into the thymus was negligible (Hale, Boursalian,

124 Turk, & Fink, 2006). The frequency and number of CD4+ and CD8+ SP cells were

125 significantly decreased in 1-week-old Cicf/f;Vav1-Cre mice (Fig. 1E), suggesting that

126 CIC regulates SP cell development in the thymus. Consistent with this result, both

127 CD4+ and CD8+ T cell populations were substantially reduced in the spleens of 1-

128 week-old Cicf/f;Vav1-Cre mice (Fig. S1). Together, these data demonstrate that CIC is

129 involved in the regulation of thymic T cell development.

130

131 Stable CIC expression in DP cells of Cicf/f;Cd4-Cre mice

132 After obtaining the results shown in Fig. 1D, we wondered why the frequency of DP

133 cells was comparable between WT and Cicf/f;Cd4-Cre mice because Cic alleles were

134 supposed to be deleted in DP cells by Cd4-Cre (P. P. Lee et al., 2001). Western

135 blotting for CIC in multiple developing thymic T cell subsets from WT, Cicf/f;Cd4-Cre,

136 and Cicf/f;Vav1-Cre mice provided the answer. Although CIC expression disappeared

137 in all tested thymic T cell subsets of Cicf/f;Vav1-Cre mice, CIC was still substantially

138 expressed in the DP cells of Cicf/f;Cd4-Cre mice (Fig. 2A and Fig. S4C). This appeared

139 to be caused by CIC protein stability, because the loxP site-flanked genomic regions

140 containing exons 9–11 of Cic were completely removed in DP cells by Cd4-Cre and

141 Vav1-Cre (Fig. 2B). These data suggest that the impaired SP cell formation in the

142 thymus of Cicf/f;Vav1-Cre mice is attributed to the loss of CIC in DN and/or DP cells

143 rather than that in SP cells, because CIC expression was not detected in SP

144 thymocytes from Cicf/f;Cd4-Cre and Cicf/f;Vav1-Cre mice (Fig. 2A and Fig. S4C).

145

146 Impaired thymic positive selection in Cicf/f;Vav1-Cre mice bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

8

147 Defects in thymic positive selection that occurs during the CD4+CD8+ DP

148 developmental stage often result in a decrease in the CD4+ and CD8+ SP cell

149 populations (Fischer, Katayama, Pagès, Pouysségur, & Hedrick, 2005; Lesourne et

150 al., 2009; Neilson, Winslow, Hur, & Crabtree, 2004; D. Wang et al., 2012). Therefore,

151 we examined thymic positive selection in Cicf/f;Vav1-Cre mice. First, we analyzed

152 thymocytes from 7-week-old WT, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice for the

153 expression of CD69 and TCRusing flow cytometry. The frequency of CD69+TCRhi

154 cells, which represent post-positive-selection thymocytes (Guo Fu et al., 2009), was

155 significantly decreased in Cicf/f;Vav1-Cre mice compared with that of WT or Cicf/f;Cd4-

156 Cre mice (Fig. 3A). This reduction was also observed in the thymus of Cicf/f;Vav1-Cre

157 mice at 1 week of age (Fig. 3B). Next, we determined the effect of CIC deficiency on

158 thymic positive selection in mice expressing the MHC class I-restricted H-Y TCR

159 transgene specific to the H-Y male antigen (Kisielow, Blüthmann, Staerz, Steinmetz,

160 & Von Boehmer, 1988) or the MHC class II-restricted OT-II TCR transgene specific to

161 ovalbumin (Barnden, Allison, Heath, & Carbone, 1998) at 7–9 weeks of age. Defects

162 in SP thymocyte development were much more severe in TCR-transgenic

163 hematopoietic lineage cell-specific Cic-null (H-Y;Cicf/f;Vav1-Cre or OT-II;Cicf/f;Vav1-

164 Cre) mice than in non-transgenic polyclonal Cicf/f;Vav1-Cre mice: CIC deficiency

165 dramatically reduced the populations of CD8+ and CD4+ SP cells in female H-Y TCR

166 transgenic and OT-II TCR transgenic mice, respectively (Figs. 3C and D, Fig. S2A).

167 These results indicate that CIC controls the positive selection of thymocytes.

168

169 Impaired thymic negative selection in Cicf/f;Vav1-Cre mice

170 Most thymocytes expressing autoreactive TCRs are eliminated through negative

171 selection during the DP and SP developmental stages (Klein et al., 2014). The strong bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

9

172 TCR signal induced by the interaction of a self-reactive TCR with the self-pMHC

173 molecule triggers apoptosis of autoreactive T cells (Hogquist & Jameson, 2014).

174 Interestingly, we found that the expression of Bcl2, a representative anti-apoptotic

175 gene (Vaux, Cory, & Adams, 1988) and an inhibitor of negative selection (Williams,

176 Norton, Halligey, Kioussis, & Brady, 1998), was markedly increased in DP cells but

177 not in DN and SP cells from Cicf/f;Vav1-Cre mice (Fig. 4A). Furthermore, the TCR

178 stimulation-induced expression of Nur77, an orphan receptor that promotes apoptosis

179 and negative selection of thymocytes (Cainan, Szychowski, Ka-Ming Chan, Cado, &

180 Winoto, 1995), was significantly reduced in DP cells from Cicf/f;Vav1-Cre mice

181 compared to that in WT mice (Fig. 4B). To determine whether CIC regulates thymic

182 negative selection, we analyzed thymocytes from male H-Y;Cicf/f and H-Y;Cicf/f;Vav1-

183 Cre mice by flow cytometry. As previously reported (Kisielow et al., 1988), massive

184 negative selection of DP and CD8+ SP thymocytes was observed in male H-Y TCR

185 transgenic mice (Fig. 4C), in which the male specific H-Y autoantigens are expressed

186 in thymic antigen-presenting cells. The CD8+ SP population and total thymic cell

187 number significantly increased in male H-Y;Cicf/f;Vav1-Cre mice compared to male H-

188 Y;Cicf/f mice (Figs. 4C and D, Fig. S2B), indicating thymic negative selection defects

189 in the absence of CIC. Taken together, our data demonstrate that CIC is required for

190 the negative selection mediated by TCR activation-induced apoptosis.

191

192 Changes in TCR repertoire of CD4+ SP thymocytes due to CIC deficiency

193 Developing T cells mature into SP cells through thymic selection processes;

194 consequently, numerous T cells with various TCRs constitute the TCR repertoire.

195 Because positive and negative selection determine the fate of T cells based on their

196 TCR, defects in these processes can lead to changes in the TCR repertoire (J. Lu et bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

10

197 al., 2019; Martínez-Riaño et al., 2019). To determine whether CIC deficiency alters the

198 TCR repertoire of SP thymocytes, we prepared thymic regulatory T (Treg,

199 CD4+CD25+GFP+) and non-Treg (CD4+CD25-GFP-) cells from Foxp3-GFP;Cicf/f and

200 Foxp3-GFP;Cicf/f;Vav1-cre mice, and performed RNA sequencing for TCR and TCR

201 chain genes in each sample. The analysis of hypervariable complementarity-

202 determining region 3 (CDR3) of TCR and TCR chains revealed that in CIC-deficient

203 cells, the CDR3 amino acid length distributions of TCR were generally longer than in

204 WT cells, with some differences between Treg and non-Treg cells or between TCR

205 and TCR chains (Fig. S3A). This result suggests that CIC regulates the thymic

206 selection process because pre-selection TCR chains in DN and DP thymocytes that

207 have not undergone positive and negative selection show longer CDR3 amino acid

208 length distributions than mature post-selection TCRs in CD44-CD62L+ naive T cells (J.

209 Lu et al., 2019).

210 We also analyzed the usage of the variable (V) and joining (J) segments of the TCR

211 and TCR chains. The diversity of V and J segment combinations in the TCR chain

212 was increased in CIC-deficient Treg cells compared to WT Treg cells, whereas it was

213 comparable between WT and CIC-deficient non-Treg cells (Figs. S3B). Moreover, for

214 both TCR and TCR chains, the usage frequency of multiple V and J segments was

215 different between WT and CIC-deficient Treg cells, whereas there were no significant

216 differences in the non-Treg cell population (Figs. S3C-F). Taken together, these data

217 suggest that impaired thymic selection due to CIC deficiency generated an abnormal

218 Treg cell population with a distinct TCR repertoire in the thymus.

219

220 Attenuated TCR signaling in CIC-deficient DP thymocytes bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

11

221 Because both positive and negative selection are mediated by TCR signaling

222 strength (Klein et al., 2014), we investigated whether CIC regulates the TCR signaling

223 pathway. First, we analyzed the surface expression of CD5, which has levels strongly

224 correlated with TCR signaling intensity (Azzam et al., 1998), on DP and SP thymocytes

225 from WT, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. CD5 levels were substantially

226 decreased in DP cells from Cicf/f;Vav1-Cre mice compared to those from WT or

227 Cicf/f;Cd4-Cre mice, whereas this change was subtle in the CD4+ SP cell population

228 and absent in the CD8+ SP cell population (Fig. 5A). This result was recapitulated in

229 TCR transgenic mice (Figs. S2C and D). Next, we examined the activation of the TCR

230 signaling pathway in thymocytes from WT, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice

231 upon TCR stimulation. As with CD5 levels, calcium influx sharply decreased in DP

232 cells from Cicf/f;Vav1-Cre mice, but not in SP thymocytes (Fig. 5B). We also observed

233 a moderate decrease in calcium influx in DP cells from Cicf/f;Cd4-Cre mice (Fig. 5B),

234 suggesting that CIC sensitively regulates TCR activation-induced calcium influx in DP

235 thymocytes. To further evaluate CIC regulation of TCR signaling in DP thymocytes,

236 we assessed the activation of key TCR signaling cascade components in DP cells of

237 WT, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice after TCR stimulation by treatment with

238 anti-CD3 and anti-CD4. Among the components tested, including ZAP-70, PLC, ERK,

239 JNK, and p38, phospho-ERK levels were significantly decreased in DP cells from

240 Cicf/f;Vav1-Cre mice compared to those from WT and Cicf/f;Cd4-Cre mice (Figs. 5C

241 and D). Together, these results demonstrate that CIC deficiency attenuates TCR

242 signaling by inhibiting calcium influx and ERK activation, especially in DP thymocytes.

243

244 Abnormal thymic T cell development and reduced TCR signaling intensity in

245 Cicf/f;Vav1-Cre mice are T cell-intrinsic bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

12

246 Although most of the cells that make up thymocytes are developing T cells, other

247 types of immune cells, including B cells and dendritic cells, also exist in small

248 proportions and participate in the regulation of T cell development (Klein et al., 2014).

249 Because CIC expression was abolished in whole immune cells of Cicf/f;Vav1-Cre mice,

250 it is unclear whether abnormal thymic T cell development and decreased TCR

251 signaling in Cicf/f;Vav1-Cre mice were T cell-intrinsic or -extrinsic. To clarify this issue,

252 we generated mixed BM chimeric mice by transferring the same number of BM cells

253 from Thy1.1/Thy1.2 heterozygous WT and Thy1.1/Thy1.1 homozygous Cicf/f;Vav1-

254 Cre mice into irradiated Thy1.2/Thy1.2 homozygous WT recipient mice (mixed

255 WT:Cicf/f;Vav1-Cre BM chimera), and analyzed thymic T cells in the chimeras after 8

256 weeks of reconstitution (Fig. 6A). Similar to the observations in Cicf/f;Vav1-Cre mice

257 (Figs. 1C and D), the frequency of DN4 and CD4+ SP cells was significantly decreased

258 in the CIC-deficient cell compartment compared to the WT cell counterpart in the same

259 mixed WT:Cicf/f;Vav1-Cre BM chimeric mice (Figs. 6B and C). The frequency of

260 CD69+TCRhi cells was also significantly lower in the CIC-deficient cell compartment

261 relative to the WT cell compartment (Fig. 6D), indicating that the impaired positive

262 selection in Cicf/f;Vav1-Cre mice was T cell-intrinsic. Furthermore, similar to the results

263 in Fig. 5A, CD5 levels were most strongly reduced in DP thymocytes derived from

264 Cicf/f;Vav1-Cre BM cells (Fig. 6E). Of note, the frequency of CD69+TCRhi cells and

265 CD5 levels in DP thymocytes were also significantly decreased in Cicf/f;pLck-Cre mice,

266 another T cell-specific Cic null mice, compared to that in WT mice; however, the fold

267 decrease was more dramatic in Cicf/f;Vav1-Cre mice than in Cicf/f;pLck-Cre mice

268 (Figs.S4A and B). Although pLck-Cre is expressed during the DN2 cell stage (P. P.

269 Lee et al., 2001), a small amount of CIC still existed in DP thymocytes from Cicf/f;pLck-

270 Cre mice (Fig. S4C), explaining why Cicf/f;pLck-Cre mice showed a less dramatic bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

13

271 decrease in the frequency of CD69+TCRhi cells and CD5 levels in DP thymocytes

272 compared to Cicf/f;Vav1-Cre mice. Overall, these data suggest that defects in thymic

273 T cell development and TCR signaling in Cicf/f;Vav1-Cre mice are T cell-intrinsic.

274

275 Identification of CIC target genes responsible for attenuated TCR signaling in

276 CIC-deficient DP thymocytes

277 To understand how CIC regulates the thymic selection process and TCR signaling in

278 DP cells at the molecular level, we analyzed gene expression profiles of DP

279 thymocytes from WT and Cicf/f;Vav1-cre mice by RNA sequencing. A total of 482

280 differentially expressed genes (DEGs; fold change > 2 and P-value < 0.05), including

281 263 upregulated and 219 downregulated genes, were identified in CIC-deficient DP

282 cells (Table S1). Gene Ontology (GO) analysis revealed that genes involved in the

283 inactivation of MAPKs and anti-apoptotic processes were significantly enriched among

284 the upregulated DEGs (Table S2), consistent with the characteristics found in CIC-

285 deficient DP thymocytes. We also identified several known CIC target genes among

286 the DEGs, including Etv1, Etv4, Etv5, Spry4, Dusp6, Dusp4, and Spred1 (Fryer et al.,

287 2011; Weissmann et al., 2018; Yang et al., 2017) (Fig. 7A and Table S1). Of these,

288 Spry4, Dusp4, Dusp6, and Spred1 were of particular interest because they are

289 negative regulators of ERK activation (Kidger & Keyse, 2016; Sasaki et al., 2003;

290 Wakioka et al., 2001). Murine SPRY4 suppresses Ras-independent ERK activation by

291 binding to Raf1 (Sasaki et al., 2003), whereas it represses insulin receptor and EGFR-

292 induced ERK signaling upstream of Ras in humans (Leeksma et al., 2002). SPRED1

293 also inhibits the activation of ERK by suppressing Raf activation (Wakioka et al., 2001).

294 DUSP6 specifically dephosphorylates activated ERK1/2, whereas DUSP4 acts on

295 ERK as well as other MAPKs, such as JNK and p-38 (Caunt & Keyse, 2013). bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

14

296 Additionally, SPRY4 suppresses calcium mobilization in HEK293T cells by inhibiting

297 phosphatidylinositol 4,5-biphosphate (PIP2) hydrolysis induced by VEGF-A (Ayada et

298 al., 2009). Thus, we investigated the association between derepression of Spry4,

299 Dusp4, Dusp6, and Spred1 and attenuated TCR signaling in CIC-deficient DP

300 thymocytes. First, we measured the levels of Spry4, Dusp4, Dusp6, and Spred1 in DN,

301 DP, CD4+ SP, and CD8+ SP thymocytes from WT and Cicf/f;Vav1-Cre mice by qRT-

302 PCR. The expression of all four genes was significantly upregulated in DP cells from

303 Cicf/f;Vav1-Cre mice relative to those from WT mice (Fig. 7B), verifying the RNA

304 sequencing data (Table S1 and Fig. 7A). The expression of all four genes was also

305 upregulated in CD4+ and CD8+ SP thymocytes from Cicf/f;Vav1-Cre mice; Spry4 and

306 Dusp6 were more dramatically upregulated in DP cells than in SP cells (Fig. 7B).

307 Considering the different gene expression changes in each cell type, as well as the

308 previously known functional significance in the regulation of ERK activation and

309 calcium influx, two of the four CIC target genes, Spry4 and Dusp6, were selected for

310 further investigation of their effect on TCR signaling in DP thymocytes.

311 We infected thymocytes with retroviruses expressing Spry4 or Dusp6 and analyzed

312 phospho-ERK and CD5 levels and calcium influx by flow cytometry. The levels of

313 phospho-ERK were greatly decreased in both SRPY4- and DUSP6-overexpressing

314 DP thymocytes treated with anti-CD3 (Fig. 7C), confirming their inhibitory effect on

315 ERK activation (Groom, Sneddon, Alessi, Dowd, & Keyse, 1996; Muda et al., 1996;

316 Sasaki et al., 2003). In contrast, calcium influx induced by TCR stimulation was almost

317 completely suppressed in DP thymocytes by SPRY4 overexpression but not by

318 DUSP6 overexpression (Fig. 7D). Moreover, CD5 expression was substantially

319 reduced in SRPY4-overexpressing cells, but not in DUSP6-overexpressing cells (Fig. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

15

320 7E). These data imply that Spry4 derepression might critically contribute to attenuated

321 TCR signaling in CIC-deficient DP thymocytes.

322 Finally, to determine whether impaired thymic selection and TCR signaling in

323 Cicf/f;Vav1-Cre mice were caused by derepression of Spry4, we generated Cic and

324 Spry4 double mutant (Spry4-/-;Cicf/f;Vav1-Cre) mice and analyzed thymic T cells in WT,

325 Cicf/f;Vav1-Cre, and Spry4-/-;Cicf/f;Vav1-Cre mice at 7–8 weeks of age. The decreased

326 frequency of CD69+TCRhi cells and CD5 levels of DP thymocytes in Cicf/f;Vav1-Cre

327 mice were significantly, but only partially, rescued in Spry4-/-;Cicf/f;Vav1-Cre mice (Figs.

328 7F and G), suggesting a partial recovery of the CIC deficiency-mediated defective

329 positive selection process by loss of SPRY4. However, the decreased DN4 cell

330 frequency and Nur77 expression in DP thymocytes were not rescued in Spry4-

331 /-;Cicf/f;Vav1-Cre mice (Figs. S5A and B), indicating that Spry4 derepression alone is

332 not sufficient to induce abnormal DN cell development in Cicf/f;Vav1-Cre mice and

333 suppress TCR activation-induced apoptosis in CIC-deficient DP thymocytes. Taken

334 together, our findings demonstrate that CIC tightly controls thymic T cell development

335 and TCR signaling by repressing multiple CIC target genes, including Dusp4, Dusp6,

336 Spred1, and Spry4, in a cell type- and/or developmental stage-specific manner. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

16

337 DISCUSSION

338 Our study uncovered the role of CIC in thymic T cell development via in-depth

339 analyses of thymocytes in various CIC-deficient mouse models. CIC deficiency

340 decreased the frequency of DN4 thymocytes that underwent -selection mediated by

341 pre-TCR and Notch signaling (Dose et al., 2006; Kreslavsky et al., 2012). However,

342 the decreased DN4 cell population did not prevent the formation of thymic DP cells in

343 Cicf/f;Vav1-Cre mice. Thus, CIC deficiency could impair the -selection process;

344 nonetheless, this defect was not sufficient to significantly affect the transition from the

345 DN to DP stage of thymic T cell development in Cicf/f;Vav1-Cre mice. Interestingly,

346 similar to Cicf/f;Vav1-Cre mice, ERK-deficient mice also exhibit a partial block in DN3-

347 to-DN4 maturation without defects in maturation to the DP stage of development

348 (Fischer et al., 2005). Because TCR stimulation-induced ERK activation was markedly

349 suppressed in CIC-deficient DP thymocytes (Figs. 5C and D), we inferred that

350 formation of the abnormal DN subset in the thymus of Cicf/f;Vav1-Cre mice was, at

351 least in part, caused by reduced ERK activity in DN cells deficient in CIC. Consistent

352 with this inference, the Spry4, Dusp4, Dusp6, and Spred1 genes, involved in the

353 inhibition of ERK activation (Kidger & Keyse, 2016; Sasaki et al., 2003; Wakioka et al.,

354 2001), were significantly derepressed in CIC-deficient DN cells (Fig. 7B). Further study

355 on how CIC controls pre-TCR signaling is critical to better understand CIC regulation

356 of thymic T cell development during the DN stage.

357 CIC deficiency, especially in DP thymocytes, disrupts the positive and negative

358 selection of thymocytes, as evidenced by the impaired thymic selection process in

359 Cicf/f;Vav1-Cre mice but not in Cicf/f;Cd4-Cre mice, which express significant amounts

360 of CIC proteins in DP thymocytes. Moreover, CIC loss substantially suppressed TCR

361 signaling in DP thymocytes but not in SP cells. Although Spry4, Dusp4, Dusp6, and bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

17

362 Spred1 were all significantly derepressed in SP and DP cells from Cicf/f;Vav1-Cre mice,

363 among these four CIC target genes, Spry4 and Dusp6 were more dramatically

364 derepressed in DP cells than in SP cells in the absence of CIC (Fig. 7B). These results

365 suggest that Spry4 and Dusp6 may be CIC target genes primarily responsible for the

366 CIC deficiency-mediated dysregulation of TCR signaling in DP cells and thymic

367 selection processes. However, removal of the Spry4 alleles only partially recovered

368 the TCR signal intensity and positive selection, indicating that CIC regulates multiple

369 target genes, including Spry4, to elaborately control TCR signaling and thymic T cell

370 development. Notably, the hyperactivation of peripheral T cells and expansion of the

371 Tfh cell population in Cicf/f;Vav1-Cre mice were also not rescued in Spry4-/-;Cicf/f;Vav1-

372 Cre mice (Figs. S5C and D). These findings suggest that the partial recovery of the

373 TCR signal intensity and positive selection process by loss of SPRY4 was insufficient

374 to ameliorate the abnormal peripheral T cell phenotypes in Cicf/f;Vav1-Cre mice or that

375 derepression of CIC target genes other than Spry4 could lead to T cell hyperactivation

376 and enhanced Tfh cell formation in Cicf/f;Vav1-Cre mice. Concordantly, we have

377 shown that CIC deficiency promotes Tfh cell differentiation via Etv5 derepression (S.

378 Park et al., 2017). In contrast, CIC deficiency-induced thymic T cell phenotypes were

379 not rescued in Cicf/f;Etv5f/f;Vav1-Cre mice (data not shown). Overall, these findings

380 suggest that CIC regulates various target genes with differential effects to broadly

381 control thymic T cell development and peripheral T cell activation and differentiation.

382 Our sequencing analysis for Tcra and Tcrb mRNAs revealed that CIC deficiency

383 substantially altered the TCR repertoire of thymic Treg cells. On the other hand, this

384 alteration was not clearly detected in the non-Treg cell population. Based on the affinity

385 model of thymocyte selection, thymic Treg cell differentiation occurs within a window

386 between positive selection (low affinity of the TCR-self-pMHC interaction) and bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

18

387 negative selection (high affinity of the TCR-self-pMHC interaction) (Klein et al., 2014).

388 Due to impaired negative selection, some Treg cells could be derived from self-

389 reactive T cells that were supposed to be eliminated by negative selection in

390 Cicf/f;Vav1-Cre mice. Therefore, the changed TCR repertoire in CIC-deficient Treg

391 cells might result from defective negative selection. The distribution of CDR3 amino

392 acid length of the TCRchain was changed in CIC-deficient non-Treg cells (Fig. S3A).

393 On the other hand, the diversity of V and J segment combinations in the TCR chain

394 and the usage frequency of V and J segments were comparable between WT and

395 CIC-deficient non-Treg cells (Figs. S3B-F). However, these data are limited in

396 determining whether the TCR repertoire is normal in CIC-deficient non-Treg cells,

397 since there is no information on the combination of TCR and TCR chains at the

398 single cell level. To clarify this issue, single cell RNA sequencing or TCR repertoire

399 analysis using transgenic mice with fixed TCR/TCR chain should be conducted.

400 This study provides insight into how CIC controls autoimmunity. Because defects in

401 thymic selection lead to the breakdown of central tolerance (Xing & Hogquist, 2012),

402 it is conceivable that CIC deficiency during thymic T cell development generates

403 autoreactive T cells, inducing autoimmunity. Therefore, CIC potentially suppresses

404 autoimmunity by controlling the thymic selection process, as well as Tfh cell

405 differentiation (S. Park et al., 2017). The more severe autoimmune-like phenotypes in

406 Cicf/f;Vav1-Cre mice than in Cicf/f;Cd4-Cre mice (S. Park et al., 2017) highlight the

407 significant contribution of the impaired thymic selection process to CIC deficiency-

408 induced autoimmunity. To clarify the importance of each regulatory step in the

409 suppression of autoimmunity, it will be necessary to better understand the molecular

410 mechanisms underlying CIC regulation of thymic T cell development and Tfh cell

411 differentiation. Additionally, studies on the role of CIC in T cell development and Tfh bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

19

412 cell differentiation in humans should be conducted in the future to improve our

413 understanding of the pathogenesis of autoimmune diseases, such as systemic lupus

414 erythematosus. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

20

415 MATERIALS AND METHODS

416

417 Mice

418 All mice were maintained on a C57BL/6 background. Cic-floxed (H. C. Lu et al., 2017;

419 S. Park et al., 2017), Vav1-Cre (de Boer et al., 2003), CD4-Cre (P. P. Lee et al., 2001),

420 pLck-Cre (P. P. Lee et al., 2001), FLAG-tagged Cic knock-in (CicFLAG/FLAG) (S. Park et

421 al., 2019), H-Y TCR transgenic (Kisielow et al., 1988), OT-II TCR transgenic (Barnden

422 et al., 1998), and Foxp3EGFP (Lin et al., 2007) mice have been described previously.

423 Spry4-/- mice were generated using Spry4tm1a(KOMP)Mbp embryonic stem cells obtained

424 from the UC Davis KOMP repository. All experiments were conducted with age-

425 matched mice, and 7–9 week-old mice were used unless otherwise indicated. Mice

426 from both sexes were randomly allocated to experimental groups. All mice were

427 maintained in a specific pathogen-free animal facility under a standard 12 h light/12 h

428 dark cycle. Mice were fed standard rodent chow and provided with water ad libitum.

429 All experiments were approved by the Institutional Animal Care and Use Committee

430 of Pohang University of Science and Technology.

431

432 Cell line

433 The Platinum-E (Plat-E) retroviral packaging cell line (Cell Biolabs) was grown in

434 Dulbecco’s modified Eagle’s medium (DMEM, Welgene) supplemented with 10% fetal

435 bovine serum (FBS, Welgene), and penicillin/streptomycin (Gibco). Cells were

436 cultured in a 37°C incubator with 5% CO2. Mycoplasma contamination was routinely

437 tested using the e-Myco plus Mycoplasma detection kit (INtRON Bio).

438

439 Flow cytometry bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

21

440 Surface staining was performed using the following fluorescence-labeled antibodies

441 against: CD4 (RM4-5; Biolegend, BD Biosciences), CD5 (53-7.3; Biolegend,

442 eBioscience), CD8 (53-6.7; Biolegend), CD11b (M1/70; Biolegend), CD11c (N418,

443 Biolegend), CD19 (1D3; BD Biosciences), CD25 (PC61.5; Tonbo Biosciences), CD44

444 (IM7, BD Biosciences), CD62L (MEL-14; BD Biosciences), CD69 (H1.2F3; Biolegend),

445 CD90.1 (HIS51; eBioscience), CD90.2 (53-2.1; Biolegend), CXCR5 (2G8; BD

446 Biosciences), TCR (H57-597; Tonbo Biosciences), TCR/ (GL-3; eBioscience), NK-

447 1.1 (PK136; Biolegend), TER-119 (TER-119; eBioscience), Gr-1 (RB6-8C5;

448 eBioscience), TCR H-Y (T3.70; eBioscience), and PD-1 (RMP1-30; eBioscience). For

449 CXCR5 staining, cells were incubated with biotinylated anti-CXCR5 for 30 min and

450 then sequentially incubated with APC- or PerCP- Cy5.5-labelled streptavidin

451 (eBioscience) with other surface antibodies. For Live/dead staining was performed

452 using Fixable Viability Dye (FVD) eFluor 780 (eBioscience) or Ghost Dye Violet 510

453 (Tonbo Biosciences). Intracellular staining was performed using the Foxp3 staining

454 buffer set (eBioscience). For FLAG-CIC and Nurr77 staining, fluorochrome-labeled

455 antibodies to FLAG (L5; Biolegend) and Nur77 (12.14; eBioscience) were used. For

456 phospho-ERK staining, cells were stained with FVD, fixed with BD Cytofix, and

457 permeabilized with cold methanol. Permeabilized cells were stained with the p-ERK

458 antibody (Cell Signaling Technology), and then fluorochrome-labeled antibodies to

459 surface markers and secondary antibody against rabbit IgG (Biolegend) were added.

460 The stained cells were analyzed using an LSRII Fortessa flow cytometer (BD

461 Biosciences) or CytoFLEX LX (Beckman Coulter). Data were analyzed using FlowJo

462 software (TreeStar).

463

464 Cell sorting bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

22

465 Single-cell suspensions of thymocytes were stained for surface markers including

466 lineage (TCR/, NK-1.1, TER-119, Gr-1, CD11b, CD11c, CD19) and then lin- DN

467 (CD4-CD8-), ISP (CD4-CD8+TCRloCD24hi), DP (CD4+CD8+TCRlo), CD4+ SP

468 (CD4+CD8-TCRhi), and CD8+ SP (CD4-CD8+TCRhi) cells were sorted. To sort DN3

469 (lin-CD4-CD8-CD44loCD25hi) and DN4 (lin-CD4-CD8-CD44loCD25lo) cells, CD8-

470 thymocytes were obtained by negative selection using the EasySep Mouse

471 Streptavidin Rapid Spheres Isolation Kit (Stem Cell Technologies), then subjected to

472 cell sorting. A MoFlo-XDP (Beckman Coulter) was used for cell sorting.

473

474 Generation of BM chimeric mice

475 To create a mixed BM chimera, 1 × 106 BM cells from each donor were mixed and

476 injected intravenously into C57BL/6 recipient mice that had been irradiated (10 Gy).

477 After 8 weeks of recovery, the mice were sacrificed; thymi were harvested and

478 homogenized to prepare single-cell suspensions. Cells were then stained for flow

479 cytometry.

480

481 Analysis of Nurr77 expression

482 To measure Nur77 expression, freshly isolated thymocytes were incubated with plate-

483 coated anti-CD3 (5 g/ml) and anti-CD28 (10 g/ml) for 2 h, followed by staining of

484 surface markers (CD4, CD8, and TCR) and intracellular staining of Nur77. Samples

485 were analyzed using an LSRII Fortessa flow cytometer or CytoFLEX LX. Data were

486 analyzed using FlowJo software.

487

488 In vitro TCR stimulation bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

23

489 Total thymocytes or sorted DP cells were rested for 30 min in Roswell Park Memorial

490 Institute (RPMI) medium (Welgene) at 37°C. Cells were then washed with T-cell

491 medium (TCM, RPMI supplemented with 10% FBS [Welgene], 1%

492 penicillin/streptomycin [Gibco], and 0.1% -mercaptoethanol [Gibco]) and incubated

493 with biotin-conjugated anti-CD3 (60 g/ml, 145-2C11) and anti-CD4 (60 g/ml, GK1.5)

494 for 20 min on ice. Cells were washed and incubated for 5 min in ice with streptavidin

495 (60 g/ml, SouthernBiotech) for cross-linking and incubated at 37°C. To analyze p-

496 ERK levels, unconjugated anti-CD3 (10 g/ml, 145-2C11) and goat anti-hamster IgG

497 (25 g/ml, Jackson Immunoresearch) were used to stimulate TCR. One milliliter of

498 cold PBS was added at the end of stimulation and cell pellets were lysed for western

499 blotting or further stained with anti-p-ERK antibody for flow cytometry.

500

501 Calcium influx measurement

502 Freshly isolated thymocytes or virus-transduced thymocytes were incubated with 4 M

503 Indo-1-AM (Invitrogen) in TCM at 37°C for 40 min. Cells were washed twice with TCM

504 and incubated with soluble anti-CD3 (10 g/ml) and fluorochrome-conjugated

505 antibodies for surface markers (CD4, CD8, and TCR) in TCM on ice for 20 min. Cells

506 were then washed with TCM and warmed before cross-linking. Goat anti-hamster IgG

507 (25 g/ml) was added to cross-link anti-CD3 and the signals were measured by flow

508 cytometry. Ionomycin was added to ensure that T cells were effectively loaded with

509 Indo-1. The emission wavelength ratios of Ca2+-bound to unbound Indo-1 were

510 analyzed using an LSRII Fortessa flow cytometer.

511

512 Plasmids and retroviral transduction bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

24

513 The coding sequences (CDSs) of mouse Spry4 or Dusp6 with enzyme sites

514 XhoI/EcoRI was amplified by PCR and cloned into the MigR1 retroviral vector (Pear et

515 al., 1998). Virus was generated through transient cotransfection of the Plat-E cells with

516 the cloned retroviral vectors and pCL-Eco helper plasmid (Imgenex) (Naviaux,

517 Costanzi, Haas, & Verma, 1996). Briefly, 2.5 × 106 Plat-E cells were plated in 100-mm

518 plates. The next day, cells were transfected with 3.6 g of retroviral vector and 2.4 g

519 of pCL-Eco using FuGENE HD transfection reagent (Promega). Retrovirus-containing

520 supernatants were harvested twice at 48 h and 72 h after transfection and filtered with

521 a 0.22 m syringe filter (Millipore). For retroviral transduction, 107 thymocytes were

522 mixed with 0.5 ml of the viral supernatant and 1.5 ml of fresh TCM in the presence of

523 4 g/ml polybrene (Sigma) and spin-infected at 1000 g for 90 min at 25°C.

524

525 Western blotting

526 Sorted cells or stimulated cells were lysed in RIPA buffer (50mM Tris-HCl pH 7.4,

527 150mM NaCl, 1mM PMSF, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 1x

528 Roche Complete Protease Inhibitor Cocktail, and 1x Roche Phosphatase Inhibitor

529 Cocktail). Protein concentrations were measured using a BCA kit (Pierce). Equal

530 amounts of protein samples were separated by 9% SDS-PAGE gel and transferred

531 onto nitrocellulose membranes (Bio-Rad). The following primary antibodies were used:

532 anti-CIC (homemade) (Kim et al., 2015), anti-PLC1 (#5690, Cell Signaling), anti-p-

533 PLC1 (#2821, Cell Signaling), anti-ZAP-70 (#2705, Cell Signaling), anti-p-ZAP-70

534 (#2701, Cell Signaling Technology), anti-JNK (#9252, Cell Signaling), anti-p-JNK

535 (#9251, Cell Signaling), anti-ERK (#9102, Cell Signaling), anti-p-ERK (#4370, Cell

536 Signaling), anti-p38 (#9212, Cell Signaling), anti-p-p38 (#9211, Cell Signaling), anti-- bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

25

537 actin (#sc-47778, Santa Cruz), and anti--tubulin (#sc-398103, Santa Cruz).

538 Membranes were incubated with secondary antibodies conjugated to horseradish

539 peroxidase (HRP) and developed using Clarity Western ECL Substrate (Bio-Rad) or

540 SuperSignal West Dura (Thermo Scientific). Images were acquired using an

541 ImageQuant LAS 500 instrument (GE Healthcare).

542

543 RNA isolation, cDNA synthesis, and qRT-PCR

544 Total RNA was extracted from sorted cells with RiboEx (GeneAll) and 0.2–1 g was

545 subjected to cDNA synthesis using the GoScript Reverse Transcription system

546 (Promega) according to the manufacturer’s instructions. SYBR Green real-time PCR

547 master mix (TOYOBO) was used for qRT-PCR analysis. Primers used for qRT-PCR

548 are listed in Table S3.

549

550 RNA sequencing and data analysis

551 Thymi from Cicf/f and Cicf/f;Vav1-Cre mice were dissected and homogenized into

552 single-cell suspensions for cell sorting. DP thymocytes were sorted on the basis of

553 surface markers of lin-CD4+CD8+TCRlo, and total RNA was extracted with RiboEx.

554 The library for mRNA sequencing was generated using the TruSeq Stranded Total

555 RNA LT Sample Prep Kit (Illumina) and sequencing was performed with NovaSeq

556 6000 (Illumina). Trimmed reads were mapped to the mouse reference genome (mm10

557 RefSeq) with HISAT2, and the transcripts were assembled using StringTie. DEGs

558 were generated using edgeR and genes with fold changes > 2 and P-values < 0.05

559 were selected for Gene Ontology (GO) analyses on the basis of biological processes

560 using the DAVID website (https://david.ncifcrf.gov/).

561 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

26

562 TCR repertoire sequencing

563 Treg (CD4+CD8-CD25+GFP+) and non-Treg (CD4+CD8-GFP-) cells were prepared

564 from Foxp3-GFP;Cicf/f and Foxp3-GFP;Cicf/f;Vav1-cre mice by cell sorting. Total RNA

565 was extracted from 3-5 × 104 Treg and 3 × 106 non-Treg cells, then subjected to

566 sequencing of mRNA encoding TCR and TCR chains, performed by iRepertoire,

567 Inc. (Huntsville, AL). Briefly, RNA was amplified using a commercially available

568 multiplex primer mix covering the TCRand TCR chains in two separate PCR

569 reactions. Reverse transcription and subsequent PCR amplification (RT-PCR1) were

570 performed using the Qiagen OneStep RT PCR mix (Qiagen). The cDNA was selected

571 and unused primers were removed by SPRIselect bead selection (Beckman Coulter)

572 followed by a second round of amplification performed with a pair of primers specific

573 for communal sites engineered onto the 5ʹ end of the C- and V- primers used during

574 RT-PCR1. The final constructed library included Illumina sequencing adapters and a

575 6-nucleotide internal barcode associated with the C-gene primer. After library

576 preparation, paired-end sequencing was performed using the Illumina Miseq v3 600-

577 cycle Reagent Kit (Illumina). The output of the immune receptor sequence covered the

578 second framework region through the beginning of the constant region, including both

579 CDR2 and CDR3. Raw sequencing data were analyzed using the previously described

580 iRmap program (C. Wang et al., 2010).

581

582 Data availability

583 The Gene Expression Omnibus (GEO) accession number for the RNA sequencing

584 data of DP thymocytes reported in this paper is GSE173909. The GEO accession

585 number for the TCR repertoire RNA sequencing is GSE173996. All data generated or bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

27

586 analysed during this study are included in the manuscript and supporting files. Source

587 data files have been provided for Figures 1, 2, 3, 4, 5, 6, and 7.

588

589 Statistical analysis

590 Statistical analysis was performed using GraphPad Prism 7 (www.Graphpad.com, La

591 Jolla, CA, USA). All experiments were independently performed more than three times.

592 Student’s t-test (two-tailed, two-sample unequal variance) was used to obtain P-

593 values between groups. Statistical significance was set at P < 0.05. Error bars indicate

594 the standard error of the mean (SEM). bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

28

595 ACKNOWLEDGMENTS

596 We thank Dr. Jaeho Cho and the Lee lab members for their helpful discussions and

597 comments on this study. This work was supported by grants from the Samsung

598 Science and Technology Foundation under project number SSTF-BA1502-14 and the

599 National Research Foundation (NRF) of Korea (NRF-2021R1A2C3004006 and -

600 2017R1A5A1015366). JSP and JP were supported by the BK21 Program. HH was

601 supported by a Global PhD Fellowship (NRF-2017H1A2A1042705).

602

603 AUTHOR CONTRIBUTIONS

604 S.K. and Y.L. designed the study. S.K., G.Y.P., J.S.P., J.P., and H.H. performed the

605 experiments. S.K. analyzed the data. S.K. and Y.L. wrote the manuscript.

606

607 COMPETING INTERESTS

608 The authors declare no competing interests. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

29

609 REFERENCES

610 Ayada, T., Taniguchi, K., Okamoto, F., Kato, R., Komune, S., Takaesu, G., &

611 Yoshimura, A. (2009). Sprouty4 negatively regulates protein kinase C activation

612 by inhibiting phosphatidylinositol 4,5-biphosphate hydrolysis. Oncogene, 28(8),

613 1076–1088. https://doi.org/10.1038/onc.2008.464

614 Azzam, B. H. S., Grinberg, A., Lui, K., Shen, H., Shores, E. W., & Love, P. E. (1998).

615 CD5 Expression Is Developmentally Regulated By T Cel Receptor ( TCR )

616 Signals and TCR Avidity. Journal of Experimental Medicine, 188(12).

617 Barnden, M. J., Allison, J., Heath, W. R., & Carbone, F. R. (1998). Defective TCR

618 expression in transgenic mice constructed using cDNA- based α- and β-chain

619 genes under the control of heterologous regulatory elements. Immunology and

620 Cell Biology, 76(1), 34–40. https://doi.org/10.1046/j.1440-1711.1998.00709.x

621 Cainan, B. J., Szychowski, S., Ka-Ming Chan, F., Cado, D., & Winoto, A. (1995). A

622 role for the orphan steroid receptor Nur77 in apoptosis accompanying antigen-

623 induced negative selection. Immunity, 3(3), 273–282.

624 https://doi.org/10.1016/1074-7613(95)90113-2

625 Caunt, C. J., & Keyse, S. M. (2013). Dual-specificity MAP kinase phosphatases

626 (MKPs): Shaping the outcome of MAP kinase signalling. FEBS Journal, 280(2),

627 489–504. https://doi.org/10.1111/j.1742-4658.2012.08716.x

628 de Boer, J., Williams, A., Skavdis, G., Harker, N., Coles, M., Tolaini, M., … Kioussis,

629 D. (2003). Transgenic mice with hematopoietic and lymphoid specific expression

630 of Cre. European Journal of Immunology, 33(2), 314–325.

631 https://doi.org/10.1002/immu.200310005

632 Dose, M., Khan, I., Guo, Z., Kovalovsky, D., Krueger, A., Von Boehmer, H., …

633 Gounari, F. (2006). c-Myc mediates pre-TCR-induced proliferation but not bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

30

634 developmental progression. Blood, 108(8), 2669–2677.

635 https://doi.org/10.1182/blood-2006-02-005900

636 Fischer, A. M., Katayama, C. D., Pagès, G., Pouysségur, J., & Hedrick, S. M. (2005).

637 The Role of Erk1 and Erk2 in Multiple Stages of T Cell Development. Immunity,

638 23(4), 431–443. https://doi.org/10.1016/J.IMMUNI.2005.08.013

639 Fryer, J. D., Yu, P., Kang, H., Mandel-Brehm, C., Carter, A. N., Crespo-Barreto, J.,

640 … Zoghbi, H. Y. (2011). Exercise and Genetic Rescue of SCA1 via the

641 Transcriptional Repressor Capicua. Science, 334(6056), 690–693.

642 https://doi.org/10.1126/science.1212673

643 Fu, Guo, Vallée, S., Rybakin, V., McGuire, M. V., Ampudia, J., Brockmeyer, C., …

644 Gascoigne, N. R. J. (2009). Themis controls thymocyte selection through

645 regulation of T cell antigen receptor-mediated signaling. Nature Immunology,

646 10(8), 848–856. https://doi.org/10.1038/ni.1766

647 Fu, Guoping, Chen, Y., Yu, M., Podd, A., Schuman, J., He, Y., … Wen, R. (2010).

648 Phospholipase Cγ1 is essential for T cell development, activation, and

649 tolerance. Journal of Experimental Medicine, 207(2), 309–318.

650 https://doi.org/10.1084/jem.20090880

651 Gascoigne, N. R. J., Rybakin, V., Acuto, O., & Brzostek, J. (2016). TCR Signal

652 Strength and T Cell Development. Annual Review of Cell and Developmental

653 Biology, 32, 327–348. https://doi.org/10.1146/annurev-cellbio-111315-125324

654 Groom, L. A., Sneddon, A. A., Alessi, D. R., Dowd, S., & Keyse, S. M. (1996).

655 Differential regulation of the MAP, SAP and RK/p38 kinases by Pyst1, a novel

656 cytosolic dual-specificity phosphatase. EMBO Journal, 15(14), 3621–3632.

657 https://doi.org/10.1002/j.1460-2075.1996.tb00731.x bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

31

658 Hale, J. S., Boursalian, T. E., Turk, G. L., & Fink, P. J. (2006). Thymic output in aged

659 mice. Proceedings of the National Academy of Sciences of the United States of

660 America, 103(22), 8447–8452. https://doi.org/10.1073/pnas.0601040103

661 Hogquist, K. A., & Jameson, S. C. (2014). The self-obsession of T cells: How TCR

662 signaling thresholds affect fate “decisions” and effector function. Nature

663 Immunology, 15(9), 815–823. https://doi.org/10.1038/ni.2938

664 Jiménez, G., Guichet, A., Ephrussi, A., & Casanova, J. (2000). Relief of gene

665 repression by Torso RTK signaling: Role of capicua in Drosophila terminal and

666 dorsoventral patterning. Genes and Development, 14(2), 224–231.

667 https://doi.org/10.1101/gad.14.2.224

668 Jiménez, G., Shvartsman, S. Y., & Paroush, Z. (2012). The Capicua repressor - A

669 general sensor of RTK signaling in development and disease. Journal of Cell

670 Science, 125(6), 1383–1391. https://doi.org/10.1242/jcs.092965

671 Kawamura-Saito, M., Yamazaki, Y., Kaneko, K., Kawaguchi, N., Kanda, H., Mukai,

672 H., … Nakamura, T. (2006). Fusion between CIC and DUX4 up-regulates PEA3

673 family genes in Ewing-like sarcomas with t(4;19)(q35;q13) translocation. Human

674 Molecular Genetics, 15(13), 2125–2137. https://doi.org/10.1093/hmg/ddl136

675 Keenan, S. E., Blythe, S. A., Marmion, R. A., Djabrayan, N. J. V., Wieschaus, E. F.,

676 & Shvartsman, S. Y. (2020). Rapid Dynamics of Signal-Dependent

677 Transcriptional Repression by Capicua. Developmental Cell, 52(6), 794-801.e4.

678 https://doi.org/10.1016/j.devcel.2020.02.004

679 Kidger, A. M., & Keyse, S. M. (2016). The regulation of oncogenic Ras/ERK

680 signalling by dual-specificity mitogen activated protein kinase phosphatases

681 (MKPs). Seminars in Cell and Developmental Biology, 50, 125–132.

682 https://doi.org/10.1016/j.semcdb.2016.01.009 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

32

683 Kim, E., Park, S., Choi, N., Lee, J., Yoe, J., Kim, S., … Lee, Y. (2015). Deficiency of

684 Capicua disrupts bile acid homeostasis. Scientific Reports, 5.

685 https://doi.org/10.1038/srep08272

686 Kisielow, P., Blüthmann, H., Staerz, U. D., Steinmetz, M., & Von Boehmer, H.

687 (1988). Tolerance in T-cell-receptor transgenic mice involves deletion of

688 nonmature CD4+8+ thymocytes. Nature, 333(6175), 742–746.

689 https://doi.org/10.1038/333742a0

690 Klein, L., Kyewski, B., Allen, P. M., & Hogquist, K. A. (2014). Positive and negative

691 selection of the T cell repertoire: What thymocytes see (and don’t see). Nature

692 Reviews Immunology, 14(6), 377–391. https://doi.org/10.1038/nri3667

693 Kreslavsky, T., Gleimer, M., Miyazaki, M., Choi, Y., Gagnon, E., Murre, C., … von

694 Boehmer, H. (2012). β-Selection-Induced Proliferation Is Required for αβ T Cell

695 Differentiation. Immunity, 37(5), 840–853.

696 https://doi.org/10.1016/j.immuni.2012.08.020

697 Lee, P. P., Fitzpatrick, D. R., Beard, C., Jessup, H. K., Lehar, S., Makar, K. W., …

698 Wilson, C. B. (2001). A critical role for Dnmt1 and DNA methylation in T cell

699 development, function, and survival. Immunity, 15(5), 763–774.

700 https://doi.org/10.1016/S1074-7613(01)00227-8

701 Lee, Y. (2020). Regulation and function of capicua in mammals. Experimental and

702 Molecular Medicine, 52(4), 531–537. https://doi.org/10.1038/s12276-020-0411-3

703 Leeksma, O. C., Van Achterberg, T. A. E., Tsumura, Y., Toshima, J., Eldering, E.,

704 Kroes, W. G. M., … De Vries, C. J. M. (2002). Human sprouty 4, a new ras

705 antagonist on 5q31, interacts with the dual specificity kinase TESK1. European

706 Journal of Biochemistry, 269(10), 2546–2556. https://doi.org/10.1046/j.1432-

707 1033.2002.02921.x bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

33

708 Lesourne, R., Uehara, S., Lee, J., Song, K. D., Li, L. Q., Pinkhasov, J., … Love, P. E.

709 (2009). Themis, a T cell-specific protein important for late thymocyte

710 development. Nature Immunology, 10(8), 840–847.

711 https://doi.org/10.1038/ni.1768

712 Lin, W., Haribhai, D., Relland, L. M., Truong, N., Carlson, M. R., Williams, C. B., &

713 Chatila, T. A. (2007). Regulatory T cell development in the absence of functional

714 Foxp3. Nature Immunology, 8(4), 359–368. https://doi.org/10.1038/ni1445

715 Lu, H. C., Tan, Q., Rousseaux, M. W. C., Wang, W., Kim, J. Y., Richman, R., …

716 Zoghbi, H. Y. (2017). Disruption of the ATXN1-CIC complex causes a spectrum

717 of neurobehavioral phenotypes in mice and humans. Nature Genetics, 49(4),

718 527–536. https://doi.org/10.1038/ng.3808

719 Lu, J., Van Laethem, F., Bhattacharya, A., Craveiro, M., Saba, I., Chu, J., … Sun, P.

720 D. (2019). Molecular constraints on CDR3 for thymic selection of MHC-restricted

721 TCRs from a random pre-selection repertoire. Nature Communications, 10(1),

722 1–14. https://doi.org/10.1038/s41467-019-08906-7

723 Martínez-Riaño, A., Bovolenta, E. R., Boccasavia, V. L., Ponomarenko, J., Abia, D.,

724 Oeste, C. L., … Alarcon, B. (2019). RRAS2 shapes the TCR repertoire by

725 setting the threshold for negative selection. Journal of Experimental Medicine,

726 216(10), 2427–2447. https://doi.org/10.1084/jem.20181959

727 Muda, M., Boschert, U., Dickinson, R., Martinou, J. C., Martinou, I., Camps, M., …

728 Arkinstall, S. (1996). MKP-3, a novel cytosolic protein-tyrosine phosphatase that

729 exemplifies a new class of mitogen-activated protein kinase phosphatase.

730 Journal of Biological Chemistry, 271(8), 4319–4326.

731 https://doi.org/10.1074/jbc.271.8.4319 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

34

732 Naviaux, R. K., Costanzi, E., Haas, M., & Verma, I. M. (1996). The pCL vector

733 system: rapid production of helper-free, high-titer, recombinant retroviruses.

734 Journal of Virology, 70(8), 5701–5705. https://doi.org/10.1128/jvi.70.8.5701-

735 5705.1996

736 Neilson, J. R., Winslow, M. M., Hur, E. M., & Crabtree, G. R. (2004). Calcineurin B1

737 is essential for positive but not negative selection during thymocyte

738 development. Immunity, 20(3), 255–266. https://doi.org/10.1016/S1074-

739 7613(04)00052-4

740 Park, G., Lee, G.-W., Kim, S., Hong, H., Park, J. S., Cho, J., & Lee, Y. (2020).

741 Deletion Timing of Cic Alleles during Hematopoiesis Determines the Degree of

742 Peripheral CD4 + T Cell Activation and Proliferation. Immune Network, 20(5), 1–

743 11. https://doi.org/10.4110/in.2020.20.e43

744 Park, S., Lee, S., Lee, C. G., Park, G. Y., Hong, H., Lee, J. S., … Lee, Y. (2017).

745 Capicua deficiency induces autoimmunity and promotes follicular helper T cell

746 differentiation via derepression of ETV5. Nature Communications, 8(May), 1–13.

747 https://doi.org/10.1038/ncomms16037

748 Park, S., Park, J., Kim, E., & Lee, Y. (2019). The Capicua/ETS Translocation Variant

749 5 Axis Regulates Liver-Resident Memory CD8+ T-Cell Development and the

750 Pathogenesis of Liver Injury. Hepatology, 70(1), 358–371.

751 https://doi.org/10.1002/hep.30594

752 Pear, W. S., Miller, J. P., Xu, L., Pui, J. C., Soffer, B., Quackenbush, R. C., …

753 Baltimore, D. (1998). Efficient and rapid induction of a chronic myelogenous

754 leukemia-like myeloproliferative disease in mice receiving p210 bcr/abl-

755 transduced bone marrow. Blood, 92(10), 3780–3792.

756 https://doi.org/10.1182/blood.v92.10.3780 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

35

757 Ren, Y., Ouyang, Z., Hou, Z., Yan, Y., Zhi, Z., Shi, M., … Shao, Y. (2020). CIC Is a

758 Mediator of the ERK1/2-DUSP6 Negative Feedback Loop. IScience, 23(11),

759 101635. https://doi.org/10.1016/j.isci.2020.101635

760 Sakaguchi, N., Takahashi, T., Hata, H., Nomura, T., Tagami, T., Yamazaki, S., …

761 Sakaguchi, S. (2003). Altered thymic T-cell selection due to a mutation of the

762 ZAP-70 gene causes autoimmune arthritis in mice. Nature, 426(6965), 454–460.

763 https://doi.org/10.1038/nature02119

764 Sasaki, A., Taketomi, T., Kato, R., Saeki, K., Nonami, A., Sasaki, M., … Yoshimura,

765 A. (2003). Mammalian Sprouty4 suppresses Ras-independent ERK activation by

766 binding to Raf1. Nature Cell Biology, 5(5), 427–432.

767 https://doi.org/10.1038/ncb978

768 Shin, D. H., & Hong, J. W. (2014). Capicua is involved in Dorsal-mediated repression

769 of Zerknüllt expression in Drosophila embryo. BMB Reports, 47(9), 518–523.

770 https://doi.org/10.5483/BMBRep.2014.47.9.122

771 Sommers, C. L., Park, C. S., Lee, J., Feng, C., Fuller, C. L., Grinberg, A., … Love, P.

772 E. (2002). A LAT mutation that inhibits T cell development yet induces

773 lymphoproliferation. Science, 296(5575), 2040–2043.

774 https://doi.org/10.1126/science.1069066

775 Tan, Q., Brunetti, L., Rousseaux, M. W. C., Lu, H. C., Wan, Y. W., Revelli, J. P., …

776 Zoghbi, H. Y. (2018). Loss of Capicua alters early T cell development and

777 predisposes mice to T cell lymphoblastic leukemia/lymphoma. Proceedings of

778 the National Academy of Sciences of the United States of America, 115(7),

779 E1511–E1519. https://doi.org/10.1073/pnas.1716452115

780 Vaux, D. L., Cory, S., & Adams, T. M. (1988). Bcl-2 promotes the survival of

781 haemotopoeitic cells and cooperates with c-myc to immortalize pre-B cells. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

36

782 Nature, 258, 1955-1957. Retrieved from

783 https://www.nature.com/articles/335440a0.pdf

784 Wakioka, T., Sasaki, A., Kato, R., Shouda, T., Matsumoto, A., Miyoshi, K., …

785 Yoshimura, A. (2001). Spred is a sprouty-related suppressor of Ras signalling.

786 Nature, 412(6847), 647–651. https://doi.org/10.1038/35088082

787 Wang, C., Sanders, C. M., Yang, Q., Schroeder, H. W., Wang, E., Babrzadeh, F., …

788 Han, J. (2010). High throughput sequencing reveals a complex pattern of

789 dynamic interrelationships among human T cell subsets. Proceedings of the

790 National Academy of Sciences of the United States of America, 107(4), 1518–

791 1523. https://doi.org/10.1073/pnas.0913939107

792 Wang, D., Zheng, M., Lei, L., Ji, J., Yao, Y., Qiu, Y., … Lu, L. (2012). Tespa1 is

793 involved in late thymocyte development through the regulation of TCR-mediated

794 signaling. Nature Immunology, 13(6), 560–568. https://doi.org/10.1038/ni.2301

795 Weissmann, S., Cloos, P. A., Sidoli, S., Jensen, O. N., Pollard, S., & Helin, K.

796 (2018). The tumor suppressor CIC directly regulates MAPK pathway genes via

797 histone deacetylation. Cancer Research, 78(15), 4114–4125.

798 https://doi.org/10.1158/0008-5472.CAN-18-0342

799 Williams, O., Norton, T., Halligey, M., Kioussis, D., & Brady, H. J. M. (1998). The

800 action of bax and bcl-2 on T cell selection. Journal of Experimental Medicine,

801 188(6), 1125–1133. https://doi.org/10.1084/jem.188.6.1125

802 Xing, Y., & Hogquist, K. A. (2012). T-Cell Tolerance: Central and Peripheral. Cold

803 Spring Harbor Perspectives in Biology, 4(6), a006957–a006957.

804 https://doi.org/10.1101/cshperspect.a006957

805 Yang, R., Chen, L. H., Hansen, L. J., Carpenter, A. B., Moure, C. J., Liu, H., … Yan,

806 H. (2017). Cic loss promotes gliomagenesis via aberrant neural stem cell bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

37

807 proliferation and differentiation. Cancer Research, 77(22), 6097–6108.

808 https://doi.org/10.1158/0008-5472.CAN-17-1018

809 D A Cell no. (x 106) % of thymocytes CD4 CIC ∆MFI (x103) CD8 2 4 6 0 0 1 2 3 1.95 0 1 2 3 4 5 9.77 DN1 DN2 D D (which wasnotcertifiedbypeerreview)istheauthor/funder,whohasgrantedbioRxivalicensetodisplaypreprintinperpetuity.It bioRxiv preprint Cic * N N DN3 f/f DN4 80.1 3.43 1 1 2 2 7 7 8 8 9 9 0 5 0 5 5 0 5 0 5 0 5 0 0 0 0 0 0 0 ISP SM 4SP DP doi: 1.52 7.02 D Cic D SM 8SP * https://doi.org/10.1101/2021.07.11.451936 P P * * f/f

;Cd4-Cre M 4SP 1 1 1 1 2 2 5 0 5 0 M 8SP 0 5 0 5 0 5 85.1 2.76 C C D D * B 4 4 made availableundera *

*

S S 2.40

7.47 6 Cic P P Cell no. (x 10 ) 1 2 3 C C C 0 0 0 f/f i i i 0 2 4 6 8 0 0 0 0 1 2 3 4 5 c c c 0 ;Vav1-Cre t f f f / h / / f f f C ; ; p C y V C = D m Total D a 0 d4 8 . v o 84.7 8 0 1.96 *

1 5 c S -

* S 7 Cr - y * P Cr 4 CC-BY 4.0Internationallicense ; t P e this versionpostedJuly12,2021. e s e E C

6

Cell no. (x 10 ) % of thymocytes1

0 1 2 CD4 CD44 CD8 CD25 0 0 2 4 6 8 5.84 % of DN 22.6 1.79 5.00 0 1 2 3 4 5 DN1 D D 4 D N Cic Cic N 3 2 N 1 f/f f/f 84.0 1 1 2 2 65.9 9.69 1.28 7 7 8 8 9 9 1 1 2 0 5 0 5 0 5 . 0 5 0 5 0 5 0 0 5 0 0 5 The copyrightholderforthispreprint D 7.34 Cic 17.1 D 3.10 2.42 D Cic P * * N * * P 2 f/f * f/f ;Vav1-Cre ;Cd4-Cre

5 5 6 7 8 0 0 0 0 0 0 2 4 6 8

Cell no. (x1 10 1 ) 0 5 0 5 85.6 66.8 13.7 0.59 C D C D * * N D * 4 3 SP * 4 10.2 2.07 C C Cic

SP 6 3 2 1 i i Cell no. (x 10 ) c c 3 2 1 0 1 2 0 0 0 0 f/f f f 0 0 0 0 / t / f f h ;Vav1-Cre ; 0 1 2 3 4 5 V y Fig 1 C D * a m Total C D * * v N o D * * 8 1 4 17.2 c * * * 70.5 SP * - 8 y Cr *

t SP e e s bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Figure 1. Altered T cell development in Cicf/f;Vav1-cre mice.

(A) CIC protein levels in thymic T cell subsets. Thymocytes of CicFLAG/FLAG mice (N = 7) were subjected to flow cytometry using anti-FLAG antibody. The ΔMFI of CIC-FLAG was calculated by subtraction of the mean fluorescence intensity (MFI) value obtained by isotype control from that obtained by the anti-FLAG antibody. DN1: CD4-CD8-CD44hiCD25lo, DN2: CD4-CD8-CD44hiCD25hi, DN3: CD4-CD8-CD44loCD25hi, DN4: CD4-CD8-CD44loCD25lo; ISP: immature CD8+ single positive cells (CD4-CD8-TCRβloCD24hi), DP: CD4+CD8+, SM: semi-mature (CD69+TCRβhi), and M: mature (CD69-TCRβhi). (B-D) Flow cytometry of thymocytes from 7-week-old Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. (B) Total numbers of thymocytes for each genotype. N = 11, 7, and 12 for Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice, respectively. (C) Proportions of DN1-4 subsets for each genotype. N = 8, 7, and 7 for Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice, respectively. (D) Frequencies and numbers of DN, DP, CD4+ SP, and CD8+ SP cells for each genotype. N = 11, 7, and 12 for Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice, respectively. (E) Flow cytometry of thymocytes of 1-week-old Cicf/f and Cicf/f;Vav1-Cre mice using CD4 and CD8 markers. Total thymocyte numbers, and the frequencies and numbers of DN, DP, CD4 SP, and CD8 SP subsets in mice of each genotype, as well as representative plot images, are presented. N = 5 and 4 for Cicf/f and Cicf/f;Vav1-Cre mice, respectively. Data are representative of two independent experi- ments. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 1-source data 1. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig 2 A

DN3 DN4 ISP DP CD4 SP CD8 SP Cicf/f;-Cre: - Cd4 Vav1 - Cd4 Vav1 - Cd4 Vav1 - Cd4 Vav1 - Cd4 Vav1 - Cd4 Vav1 CIC-L CIC-S β-actin

B FRT loxP loxP Cic floxed E9 E10 E11 E12

Cicf/f Cicf/f;Cd4-Cre Cicf/f;Vav1-Cre DN DP 4SP 8SP DN DP 4SP 8SP DN DP 4SP 8SP

Figure 2. Stable CIC expression in DP cells of Cicf/f;Cd4-Cre mice.

(A) Western blotting for CIC levels in DN3, DN4, ISP, DP, CD4+ SP, and CD8+ SP cells from Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. Lin- (Lineage (CD11b, CD11c, CD19, NK1.1, Gr-1, γδTCR, and TER119)-negative) gated DN3, DN4, ISP, DP, CD4+ SP, and CD8+ SP cells were sorted from mice of each genotype. (B) PCR analysis of Cic knock-out efficiency in DN, DP, CD4+ SP, and CD8+ SP cells from Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. Genomic DNA was extracted from sorted DN, DP, CD4+ SP, and CD8+ SP cells and subjected to PCR for a part of Cic floxed allele. Upper panel, schematic of the Cic floxed allele. Arrowheads indicate the primers used for PCR. Lower panel, representative agarose gel image of PCR bands. An arrow indicates PCR bands of the Cic floxed allele. See also Figure 2-source data 1. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig 3

A 7-week-old B 1-week-old Cicf/f Cicf/f;Cd4-Cre Cicf/f;Vav1-Cre CD69+TCRβhi CD69+TCRβhi 8 **** 4 ** * 2.90 2.67 1.64 3 6.84 6.55 4.57 6 4 2 2 1 % of thymocytes 79.9 5.06 80.1 5.68 83.9 6.31 % of thymocytes CD69 0 0 Cicf/f Cicf/f TCRβ Cicf/f;Cd4-Cre Cicf/f;Vav1-Cre Cicf/f;Vav1-Cre

C H-Y;Cicf/f (F) H-Y;Cicf/f;Vav1-Cre (F) 100 80 * f/f

) H-Y;Cic (F) 6 f/f 13.8 59.3 12.8 69.8 60 80 H-Y;Cic ;Vav1-Cre (F) 60 40 40 20 ** * * Cell no. (x 10 20

11.3 15.6 14.2 3.12 % of thymocytes CD4 0 0 CD8 DP DN DP DN 4SP 8SP 4SP 8SP D OT-Ⅱ;Cicf/f OT-Ⅱ;Cicf/f;Vav1-Cre 100 150 **** * OT-Ⅱ;Cicf/f ) f/f 25.8 60.3 5.71 85.5 80 6 OT-Ⅱ;Cic ;Vav1-Cre 100 60 40 *** 50

20 *** Cell no. (x 10 * 12.9 0.93 7.26 1.57 % of thymocytes CD4 * 0 0 CD8 DP DP DN 4SP 8SP DN 4SP 8SP

Figure 3. Defective positive selection in the absence of CIC.

(A) Thymocytes from 7-week-old Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice were analyzed for surface expression of CD69 and TCRβ. Representative flow cytometry plots (left) and the frequency of post-positive selection subset (CD69+TCRβhi) are shown (right). The CD69+TCRβhi cell population is highlighted with a red box in the FACS plots. N = 4 for each group. (B) Frequencies of the post-positive selection subset (CD69+TCRβhi) of 1-week-old Cicf/f and Cicf/f;Vav1-Cre mice. N = 5 and 4 for Cicf/f and Cicf/f;Vav1-Cre mice, respectively. (C) Thymocytes from female H-Y;Cicf/f (N = 3) and H-Y;Cicf/f;Vav1-Cre (N = 4) mice were analyzed for CD4 and CD8 expression. Representative flow cytometry plots (left) and frequencies (middle) and numbers (right) of DN, DP, CD4+ SP (4SP), and CD8+ SP (8SP) cells are shown. (D) Thymocytes from OT-II;Cicf/f and OT-II;Cicf/f;Vav1-Cre mice were analyzed for CD4 and CD8 expression. Representative flow cytometry plots (left) and the frequencies (middle) and numbers (right) of DN, DP, CD4+ SP (4SP), and CD8+ SP (8SP) cells are shown. N = 5 for each group. The data are representative of at least two independent experiments. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 3-source data 1. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig 4

A Bcl2 B f/f f/f 4 100 Cic Cic ;Vav1-Cre ** f/f

Cic ) 3 Cicf/f;Vav1-cre 80 50 * 3 3 * 40 60 2 2 30

40 (% of DP) Events

+ 20

7 1 * (% of Max)

20 7

1 r 10 u

0 N 0 0 Nur77 MFI (x10 Relative expression 3 4 5 6 0 0 10 10 10 10 DN DP 4SP 8SP Nur77

C H-Y;Cicf/f (M) D H-Y;Cicf/f;Vav1-Cre (M) Total H-Y;Cicf/f (M) H-Y;Cicf/f;Vav1-Cre (M) 100 30 thymocytes * ) 80 6 40 8.26 4.83 8.16 8.59 ) * 20 6 60 30 40 20 10 ** * 20 * 10 Cell no. (x 10 83.7 76.2 % of thymocytes

3.23 7.10 Cell no. (x 10 CD4 0 0 0 CD8 DP DN 4SP 8SP DN DP 4SP 8SP

Figure 4. Defective negative selection in the absence of CIC.

(A) qRT-PCR quantification of Bcl2 expression in DN, DP, CD4+ SP (4SP), and CD8+ SP (8SP) cells of Cicf/f and Cicf/f;Vav1-Cre mice. N = 3 for each group. (B) Flow cytometry of Nur77 expression in DP thymocytes from Cicf/f and Cicf/f;Vav1-Cre mice. Freshly isolated thymocytes were treated with plate-coated anti-CD3 (5 µg/ml) and anti-CD28 (10 µg/ml) for 2 h and then subjected to flow cytometry. Representative histograms for Nur77 expression in DP thymo- cytes of Cicf/f (black line) and Cicf/f;Vav1-Cre (red line) mice overlaid with isotype control (gray shaded) and unstimulated control (dotted line) histograms (left), the frequency of Nur77+ DP cells (middle), and Nur77 MFI of DP thymocytes (right) are presented. N = 3 for each genotype. Data are representative of two independent experiments. (C and D) Thymocytes from male H-Y;Cicf/f (N = 3) and H-Y;Cicf/f;Vav1-Cre (N = 6) mice were ana- lyzed for CD4 and CD8 expression. (C) Representative flow cytometry plots (left) and frequencies (middle) and numbers (right) of DN, DP, CD4+ SP (4SP), and CD8+ SP (8SP) cells are shown. (D) Total thymocyte numbers. Data are representative of two independent experiments. Bar graphs show data as means with SEM. *P < 0.05 and **P < 0.01. The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 4-source data 1. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig 5

A Cicf/f B Cicf/f Cicf/f;Cd4-Cre Cicf/f;Vav1-Cre Cicf/f;Cd4-Cre Cicf/f;Vav1-Cre 4 1.5K **** DP DP **** 200K 3 1.0K 2 150K ionomycin 500 1 100K anti-hamster 0 0 )

3 20 300 CD4 SP ** CD4 SP 15 * 200K 200 10 150K 100 5 100K

0 CD5 MFI (x10 0 8 100 CD8 SP CD8 SP 80 6 200K 60 4 150K 40 2 20 100K

Counts 0

0 Indo-1 Ratio 0 1 2 3 4 5 10 10 10 10 10 10 20 40 60 80 100 120 CD5 Time

C Cicf/f; Cicf/f; D f/f p-ERK p-PLCγ1 p-ZAP-70 Cic Cd4-Cre Vav1-Cre 60 6 8 (min) 0 2 5 15 0 2 5 15 0 2 5 15 *** 6 p-ERK 40 # 4 4

Ratio 20 2 t-ERK 2 (p-form / total) p-ZAP-70 0 0 0 (min) 0 2 5 15 0 2 5 15 0 2 5 15 t-ZAP-70 p-JNK p-PLCγ1 p-p38 2 3 Cicf/f t-PLCγ1 Cicf/f;Cd4-Cre 2 Cicf/f;Vav1-Cre p-JNK 1 1 t-JNK 0 0 p-p38 0 2 5 15 0 2 5 15

t-p38

α-tubulin bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Figure 5. Attenuated TCR signaling in CIC-deficient DP thymocytes.

(A) Thymocytes from 7-week-old Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice were FACS gated into the DP, CD4+ SP, and CD8+ SP cell population and CD5 MFI was measured. Representative histo- gram of CD5 expression in each cell population (left) and calculated CD5 MFI (right) are shown (N = 4 per group). (B) TCR stimulation-induced Ca2+ influx in DP, CD4+ SP, and CD8+ SP thymocytes from 7-week-old Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. Data are representative of at least three independent experiments. (C and D) Western blot analysis for the activation of TCR cascade components in DP thymocytes from 7-week-old Cicf/f, Cicf/f;Cd4-Cre, and Cicf/f;Vav1-Cre mice. The sorted DP cells were stimulated with soluble anti-CD3 and anti-CD4 for indicated time points. (C) Representative western blot images. (D) Band densities were measured using ImageJ software and are presented as ratios of phosphory- lated to total forms of each protein (N = 3). Graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. #P< 0.05 (comparison between Cicf/f;Cd4-Cre and Cicf/f;Vav1-Cre mice). The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 5-source data 1, 2. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 A Cicf/f;Thy1.1/1.2 made available under aCC-BY 4.0 International license. Fig 6 B6 (Thy1.2/1.2)

+ KO WT 8 weeks Thy1.1 Cicf/f;Vav1-Cre;Thy1.1/1.1 Thy1.2 B WT (Thy1.1/1.2) KO (Thy1.1/1.1) DN1 DN2 DN3 DN4 2.05 6.31 1.57 19.4 4 ** 50 ** 80 40 ** 3 40 60 30 30 2 40 20 20

% of DN 1 20 10 36.8 54.9 14.5 64.5 10 CD44 0 0 0 0 CD25 WT KO WT KO WT KO WT KO

C 1.1/1.2 1.1/1.1 WT (Thy ) KO (Thy ) DN DP CD4 SP CD8 SP 18.3 74.9 12.0 83.9 3 100 25 * 6 80 20 4 2 60 15 10 1 40 2 2.85 3.99 2.23 1.83 20 5 CD4

% of thymocytes 0 0 0 0 CD8 WT KO WT KO WT KO WT KO

D CD69+TCRβhi E WT (Thy1.1/1.2) KO (Thy1.1/1.1) DP CD4 SP CD8 SP 15 ** 5 25 15 ) 10.3 5.20 3 ** ** ** 4 20 10 3 15 10 2 10 5 5 1 5 CD69 CD5 MFI (x10 % of thymocytes TCRβ 0 0 0 0 WT KO WT KO WT KO WT KO

Figure 6. Altered T cell development and TCR intensity in Cicf/f;Vav1-Cre mice are caused by CIC loss in T cells.

(A) Schematic for the generation and analysis of mixed BM chimeric mice. Equal numbers of BM cells from Cicf/f;Thy1.1/1.2 (WT) and Cicf/f;Vav1-Cre;Thy1.1/1.1 (KO) were mixed and transferred to irradiated B6 (Thy1.2/1.2) recipient mice (N = 4). A representative FACS plot image showing the thymocytes of each origin is presented. (B-D) Flow cytometry of thymocytes from the mixed BM chimeras for the frequen- cies of DN subsets based on CD44 and CD25 expression (B), DN, DP, CD4+ SP, and CD8+ SP cells (C), and post-positive selection subset (CD69+TCRβhi). The CD69+TCRβhi cell population is highlighted with a red box in the flow cytometry plots (D). (E) Flow cytometry of surface expression levels of CD5 in DP, CD4+ SP, and CD8+ SP thymocytes derived from WT and KO BM cells in the same BM chimeric mice. Data are representative of two independent experiments. Graphs show data as means with SEM. *P < 0.05, **P < 0.01, and ***P < 0.001. The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 6-source data 1. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig 7

A B Cicf/f Cicf/f;Vav1-cre 25 Etv5 Etv4 Spry4 Dusp4 Dusp6 Spred1 Spry4 20 80 * 15 ** 8 ** 5 ** Dusp6 **** Dusp4 4 **** ** 15 60 6 ** 10 *** Etv1 ** **** 3 **** 40 4 ** 10 *** 2 5 **** Spred1 20 2 1 5 Bcl2 Relative expression

-log10 (adjusted p-value) 0 0 0 0

0 DN DP DN DP DN DP DN DP -5 0 5 4SP 8SP 4SP 8SP 4SP 8SP 4SP 8SP log2 fold change C E G isotype unstim CTRL CTRL SPRY4 SPRY4 CTRL DP DUSP6 DUSP6 SPRY4 DUSP6

100 25 ) 3 100 * ) 15 5 * 2 *** *** ** 80 80 20 **** 10 60 4 60 15 ** 3 40 40 10 5 20 2 20 5 CD5 MFI (x10 0 CD5 MFI (x10 Events (% of Max) 1 Events (% of Max) MFI of p-ERK 0 0 0 3 4 5 3 4 5 0 10 10 10 (relative to unstim) 0 0 10 10 10 p-ERK CD5

D F -/- f/f CTRL Spry4 ;Cic ; CD69+TCRβhi SPRY4 DUSP6 Cicf/f Cicf/f;Vav1-Cre Vav1-Cre 8 *** * 160K 6.18 4.97 5.90 5.27 3.70 4.23 6 130K 4 100K anti-hamster 2 70K ionomycin 2.68 3.36 3.87

Indo-1 Ratio 81.4 84.4 82.0 CD69

% of thymocytes 0 40K TCRβ 10 40 70 100 Cicf/f Time Cicf/f;Vav1-Cre Spry4-/-;Cicf/f;Vav1-Cre bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Figure 7. Identification of CIC target genes regulating TCR signaling in DP thymocytes.

(A) Volcano plot showing the DEGs in CIC-deficient DP thymocytes (fold change >2 and adjusted P-value < 0.05). CIC target genes and Bcl2 are marked at each corresponding dot. (B) qRT-PCR quantification of expression of Spry4, Dusp4, Dusp6, and Spred1 in DN, DP, CD4+ SP (4SP), and CD8+ SP (8SP) thymocytes from Cicf/f and Cicf/f;Vav1-Cre mice. N = 3 for each group. (C-E) Effects of overexpression of SPRY4 and DUSP6 on TCR signaling in DP cells. Thymocytes were infected with retroviruses expressing either Spry4 or Dusp6 and subjected to flow cytometry for ERK activation (C), Ca2+ influx (D), and CD5 expression (E) in DP thymocytes. Three independent experiments were performed. (F) Thymocytes from 7-week-old Cicf/f, Cicf/f;Vav1-Cre, and Spry4-/-;Cicf/f;Vav1-Cre mice were analyzed for the surface expression of CD69 and TCRβ. Representative FACS plots (left) and the frequency of CD69+TCRβhi cells (right) are shown. The CD69+TCRβhi cell population is highlighted with a red box in the FACS plots. N = 4, 6, and 4 for Cicf/f, Cicf/f;Vav1-Cre, and Spry4-/-;Cicf/f;Vav1-Cre mice, respective- ly. (G) CD5 levels in DP thymocytes from the mice used in (F). N = 3, 5, and 3 for Cicf/f, Cicf/f;Vav1-Cre, and Spry4-/-;Cicf/f;Vav1-Cre mice, respectively. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. The unpaired two-tailed Student’s t-test was used to calculate P values. See also Figure 7-source data 1. Supplemental Figure 1.DecreasedsplenicTcellpopulations in show data asmeans withSEM. *P <0.05 and**P <0.01. and totalsplenocyte numbersare shown.Datarepresent twoindependentexperiments. Bargraphs images ofCD4 andCD8expression insplenocytes,frequency andnumberof CD4 Flow cytometryofsplenocytes from1-week-old of age. CD4 CD8 (which wasnotcertifiedbypeerreview)istheauthor/funder,whohasgrantedbioRxivalicensetodisplaypreprintinperpetuity.It bioRxiv preprint 1.97 Cic f/f doi: 0.74 https://doi.org/10.1101/2021.07.11.451936 Cic f/f 0.61 ;Vav1-Cre made availableundera 0.20 Cic CC-BY 4.0Internationallicense ; Cell no. (x 104) this versionpostedJuly12,2021.

1 2 3 % of splenocytes f/f 1 2 3 0 0 0 0 0 andCic C * D * * 4 f/f ;Vav1-Cre 0 1 1 1 . . 5 5 0 1 0 5 0 5 Cic . The copyrightholderforthispreprint mice.Representative plot f/f C ;Vav1-Cre * D * * * 8 + andCD8 Cic Cic mice at1week 6

Cell no. (x 10 ) f/f f/f 2 1 1 ;Vav1-Cre 0 5 0 0 5 sp l e Total + Tcells, n Fig S1 o c y t e s bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig S2

A H-Y;Cicf/f (F) H-Y;Cicf/f;Vav1-Cre (F) T3.70+CD8+ 30 ** 10.3 24.2 10.9 7.49 H-Y;Cicf/f (F) H-Y;Cicf/f;Vav1-Cre (F) 20

10 10.9 13.8 54.5 67.7 % of thymocytes T3.70 0 CD8 B H-Y;Cicf/f (M) H-Y;Cicf/f;Vav1-Cre (M) T3.70+CD8+ 87.2 5.61 68.5 23.0 30 *** H-Y;Cicf/f (M) H-Y;Cicf/f;Vav1-Cre (M) 20

10 4.28 2.94 3.53 5.02 T3.70 % of thymocytes 0 CD8 C DP CD4 CD8 f/f

) H-Y;Cic (F) 3 10 * 25 6 * H-Y;Cicf/f;Vav1-Cre (F) 8 20 6 15 4 4 10 2 2 5 CD5 MFI (x10 0 0 0 D DP CD4 CD8 )

2 50 **** 100 25 OT-Ⅱ;Cicf/f 40 20 OT-Ⅱ;Cicf/f;Vav1-Cre 30 15 50 20 10 10 5 CD5 MFI (x10 0 0 0

Supplemental Figure 2. Analysis of H-Y TCR transgenic CD8+ T cells and CD5 levels in TCR transgenic thymocytes.

(A and B) Flow cytometry of thymocytes from female and male H-Y;Cicf/f and H-Y;Cicf/f;Vav1-Cre mice for H-Y TCR (T3.70) and CD8 expression. Frequencies of H-Y TCR+ CD8+ cells among total thymo- cytes are presented. (A) Results for female mice. N = 3 and 4 for female H-Y;Cicf/f and H-Y;Cicf/f;Vav1-Cre mice, respectively. (B) Results for male mice. N = 3 and 6 for male H-Y;Cicf/f and H-Y;Cicf/f;Vav1-Cre mice, respectively. Data are representative of two independent experiments. (C and D) Thymocytes from 7-week-old WT and CIC-deficient TCR transgenic mice were FACS gated into the DP, CD4+ SP, and CD8+ SP cell population and CD5 MFI was measured. (C) Results for female H-Y TCR transgenic mice. N = 3 and 4 for H-Y;Cicf/f and H-Y;Cicf/f;Vav1-Cre mice, respectively. (D) Results for OT-II TCR transgenic mice. N = 5 for each group. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig S3 A non-Treg Treg non-Treg Treg Percentage (%) WT KO WT KO WT KO WT KO 1 38 2 35 32 3 30 4 28 5 26 6 24 7 22 20 8 18 9 16 10 14 11 12 12 10 8 13 6 14 4 15 2 CDR3 length (amino acid residues) 16 0.5 17 0.2 0.01 18 0.001 19 0 20 TCRα TCRβ

B non-Treg Treg mTRBV1 mTRBV2 mTRBV3 mTRBV4 mTRBV5 mTRBV12-1 mTRBV12-2 mTRBV13-1 mTRBV13-2 mTRBV13-3 mTRBV14 mTRBV15 mTRBV16 mTRBV17 mTRBV19 mTRBV20 mTRBV23 mTRBV24 mTRBV26 mTRBV29 mTRBV30 mTRBV31 mTRBV1 mTRBV2 mTRBV3 mTRBV4 mTRBV5 mTRBV12-1 mTRBV12-2 mTRBV13-1 mTRBV13-2 mTRBV13-3 mTRBV14 mTRBV15 mTRBV16 mTRBV17 mTRBV19 mTRBV20 mTRBV23 mTRBV24 mTRBV26 mTRBV29 mTRBV30 mTRBV31 mTRBJ1-1 mTRBJ1-1 mTRBJ1-2 mTRBJ1-2 mTRBJ1-3 mTRBJ1-3 mTRBJ1-4 mTRBJ1-4 mTRBJ1-5 mTRBJ1-5 WT mTRBJ2-1 mTRBJ2-1 mTRBJ2-2 mTRBJ2-2 mTRBJ2-3 mTRBJ2-3 mTRBJ2-4 mTRBJ2-4 mTRBJ2-5 mTRBJ2-5 mTRBJ2-7 mTRBJ2-7

mTRBJ1-1 mTRBJ1-1 mTRBJ1-2 mTRBJ1-2 mTRBJ1-3 mTRBJ1-3 mTRBJ1-4 mTRBJ1-4 mTRBJ1-5 mTRBJ1-5 KO mTRBJ2-1 mTRBJ2-1 mTRBJ2-2 mTRBJ2-2 mTRBJ2-3 mTRBJ2-3 mTRBJ2-4 mTRBJ2-4 mTRBJ2-5 mTRBJ2-5 mTRBJ2-7 mTRBJ2-7 0 3000 6000 9000 12000 15000 18000 21000 24000 27000 30000 0 20000 40000 60000 80000 100000 120000 140000 160000 180000 200000 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. The copyright holder for this preprint (which was not certifiedTCRα by V peer chain review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is C made available under aCC-BY 4.0 International license. (continued) Fig S3 WT KO mTRAV1 mTRAV8D-2 mTRAV12D-3 mTRAV3-1 mTRAV9D-1 mTRAV13D-1 mTRAV3-3 mTRAV9D-3 mTRAV13-2 mTRAV3D-3 mTRAV9D-4 mTRAV13D-2 nonTreg mTRAV3-4 mTRAV10 mTRAV13D-3 mTRAV4-2 mTRAV10D mTRAV13-4/DV7 mTRAV4D-3 mTRAV11 mTRAV13D-4 mTRAV4D-4 mTRAV12-1 mTRAV14-1 mTRAV5-1 mTRAV12D-1 mTRAV14D-1 mTRAV5D-4 mTRAV12D-2 mTRAV14-2 mTRAV6-1 mTRAV7-1 mTRAV14D-3/DV8 mTRAV6-2 mTRAV7D-2 mTRAV14-3 Treg mTRAV6D-3 mTRAV7-3 mTRAV16D/DV11 mTRAV6-4 mTRAV7D-3 mTRAV16 mTRAV6D-4 mTRAV7D-4 mTRAV17 mTRAV6D-5 mTRAV7-5 mTRAV19 mTRAV6D-6 mTRAV7-6 mTRAV21/DV12 mTRAV6-7/DV9 mTRAV7D-6 D TCRβ V chain mTRAV8D-1 WT KO mTRAV8-2 mTRBV1 mTRBV15 mTRBV2 mTRBV16 nonTreg mTRBV3 mTRBV17 mTRBV4 mTRBV19 mTRBV5 mTRBV20 mTRBV12-1 mTRBV23 mTRBV12-2 mTRBV24 mTRBV13-1 mTRBV26 Treg mTRBV13-2 mTRBV29 mTRBV13-3 mTRBV30 mTRBV14 mTRBV31

E TCRα J chain WT KO mTRAJ2 mTRAJ23 mTRAJ39 mTRAJ5 mTRAJ24 mTRAJ40 mTRAJ6 mTRAJ26 mTRAJ42 nonTreg mTRAJ9 mTRAJ27 mTRAJ43 mTRAJ11 mTRAJ28 mTRAJ45 mTRAJ12 mTRAJ30 mTRAJ48 mTRAJ13 mTRAJ31 mTRAJ49 mTRAJ15 mTRAJ32 mTRAJ50 mTRAJ16 mTRAJ33 mTRAJ52 mTRAJ17 mTRAJ34 mTRAJ53 Treg mTRAJ18 mTRAJ35 mTRAJ56 mTRAJ21 mTRAJ37 mTRAJ57 mTRAJ22 mTRAJ38 mTRAJ58

F TCRβ J chain WT KO mTRBJ1-1 mTRBJ1-2 nonTreg mTRBJ1-3 mTRBJ1-4 mTRBJ1-5 mTRBJ2-1 mTRBJ2-2 mTRBJ2-3 mTRBJ2-4 Treg mTRBJ2-5 mTRBJ2-7 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Supplemental Figure 3. Analysis of TCR repertoires of CIC-deficient CD4+ SP thymocytes.

(A-E) Sequencing analysis of Tcra and Tcrb mRNAs in thymic non-Treg (CD4+CD8-GFP-) and Treg (CD4+CD8-CD25+GFP+) cells from Foxp3-GFP;Cicf/f (WT) and Foxp3-GFP;Cicf/f;Vav1-Cre (KO) mice. Heat map showing the distribution of CDR3 amino acid length in the TCRα and TCRβ chains (A). Heat map showing the diversity of V and J segment combinations in the TCRβ chain (B). Pie-chart graphs for the usage frequency of V segments of TCRα (C) and TCRβ (D) chains, and J segments of TCRα (E) and TCRβ (F) chains. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig S4

A + hi B DP Cicf/f Cicf/f;pLck-Cre Cicf/f;Vav1-Cre CD69 TCRβ 10 ** 4 ) **** * ** 3 *** 8 **** 7.43 5.05 3.36 3 1.60 1.15 0.65 6 2 4 2 1 5.34 5.08 5.39 CD5 MFI (x10 % of thymocytes 0 0

CD69 81.8 85.4 88.0 Cicf/f TCRβ Cicf/f;pLck-Cre Cicf/f;Vav1-Cre

C DN3 DN4 DP CD4 SP CD8 SP Cicf/f; -Cre : - Cd4 Vav1 pLck - Cd4 Vav1 pLck - Cd4 Vav1 pLck - Cd4 Vav1 pLck - Cd4 Vav1 pLck CIC-L CIC-S

β-actin

Supplemental Figure 4. Analysis of thymocytes in Cicf/f;pLck-Cre mice.

(A and B) Flow cytometry of thymocytes from 7-week-old Cicf/f (N = 3), Cicf/f;pLck-Cre (N = 6), and Cicf/f;Vav1-Cre (N = 5) mice for the frequency of CD69+TCRβhi thymocytes (A) and CD5 levels in DP thymocytes (B). The CD69+TCRβhi cell population is highlighted with a red box in the FACS plots. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. (C) Western blotting for levels of CIC in DN3, DN4, ISP, DP, CD4+ SP, and CD8+ SP cells from Cicf/f , Cicf/f ;Cd4-Cre, Cicf/f ;Vav1-Cre, and Cicf/f;pLck-Cre mice. Lin- gated DN3, DN4, ISP, DP, CD4+ SP, and CD8+ SP cells were sorted from mice of each genotype. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license. Fig S5

-/- f/f A Spry4 ;Cic ; DN1 DN2 DN3 DN4 Cicf/f Cicf/f;Vav1-Cre Vav1-Cre 3 20 90 ** 40 ** * ** 1.36 14.5 0.60 11.1 0.64 9.42 15 80 30 2 70 10 20 1 60 5 10 % of DN 50 24.8 59.3 18.2 70.1 16.4 73.6 0 0 0 0 CD44 Cicf/f CD25 Cicf/f;Vav1-Cre Spry4-/-;Cicf/f;Vav1-Cre

B DP 100 * ) 80 * 3 8 *** 80 ** 60 60 6 40 4 40 (% of DP)

+ 7

20 7 20 2 r u Events (% of Max) N 0 Nur77 MFI (x10 4 5 6 0 0 0 10 10 10 Nur77

C Spry4-/-;Cicf/f; Cicf/f Cicf/f;Vav1-Cre Vav1-Cre CD4 TEM 60 *

T **

17.7 37.6 42.3 + 40 + CD4 T cell 20

% of CD4 0 CD8 TEM 4.48 25.4 25.0 50 T + 40 * CD8+ T cell **** 30 20

CD44 10 % of CD8 CD62L 0

D Tfh 0.89 7.52 8.58 15

T * + ** 10 CD4+ T cell 5 % of CD4 PD-1 0 CXCR5 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Supplemental Figure 5. Analysis of T cell subsets in the thymus and spleen of Spry4-/-;Cicf/f;Vav1-Cre mice.

(A) Flow cytometry of thymocytes from 7-week-old Cicf/f (N = 4), Cicf/f;Vav1-Cre (N = 6), and Spry4-/-;Cicf/f;Vav1-Cre (N = 4) mice for the frequency of DN1–4 subsets. (B) FACS analysis of Nur77 expression in DP thymocytes from 7-week-old Cicf/f (N = 3), Cicf/f;Vav1-Cre (N = 5), and Spry4-/-;Cicf/f;Vav1-Cre (N = 3) mice. Freshly isolated thymocytes were treated with plate-coated anti-CD3 (5 µg/ml) and anti-CD28 (10 µg/ml) for 2 h and then subjected to flow cytometry. Representative histograms of Nur77 expression in DP thymocytes of Cicf/f (black line), Cicf/f;Vav1-Cre (red line), and Spry4-/-;Cicf/f;Vav1-Cre (burgundy line) mice overlaid with isotype control (gray shaded) and unstimulated control (dotted line) histograms (left), the frequency of Nur77+ DP cells (middle), and Nur77 MFI of DP thymocytes (right) are presented. (C and D) Flow cytometry of the frequency of CD44hiCD62Llo effector memory T (TEM) (C) and Tfh (D) cells in the spleen of 7-week-old Cicf/f (N = 4), Cicf/f;Vav1-Cre (N = 6), and Spry4-/-;Cicf/f;Vav1-Cre (N = 4) mice. Bar graphs show data as means with SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Table S1. Differentially expressed genes (DEGs) in Cic-deficient DP thymocytes Gene_ID Gene_Symbol KO/WT.fc KO/WT.raw.pval KO/WT.bh.pval 18612 Etv4 231.247 1.675E-63 2.682E-59 24066 Spry4 39.348 2.365E-24 1.262E-20 104156 Etv5 21.748 9.148E-46 7.322E-42 28240 Trpm2 18.631 3.214E-05 5.254E-03 17260 Mef2c 18.250 4.837E-16 7.040E-13 100134861 Gm44502 17.490 9.483E-13 1.168E-09 216082 Gm4798 14.136 2.127E-10 1.480E-07 115488204 Smim32 12.326 4.438E-09 2.153E-06 66785 4933433G08Rik 12.168 1.803E-07 5.890E-05 102633937 Gm31639 10.733 1.253E-09 7.430E-07 214239 Ccdc9b 10.485 4.345E-07 1.364E-04 14009 Etv1 10.148 4.035E-16 6.459E-13 214791 Sertad4 10.069 1.486E-17 2.975E-14 108169205 LOC108169205 9.615 2.595E-08 1.123E-05 13390 Dlx1 9.411 1.766E-10 1.285E-07 16071 Igkc 8.900 6.811E-04 6.945E-02 115486208 LOC115486208 7.544 8.095E-04 7.950E-02 115487281 LOC115487281 7.377 8.131E-03 3.708E-01 56644 Clec7a 6.804 4.540E-05 7.197E-03 17381 Mmp12 6.580 4.092E-08 1.590E-05 105244595 Gm40185 6.506 1.597E-08 7.520E-06 102633038 Gm30959 6.181 2.152E-04 2.817E-02 16517 Kcnj16 6.123 1.150E-03 1.017E-01 12978 Csf1r 5.942 2.407E-03 1.751E-01 12262 C1qc 5.831 4.420E-06 9.434E-04 269610 Chd5 5.624 1.941E-03 1.501E-01 66011 Ranbp17 5.470 1.484E-04 2.031E-02 237625 Pla2g3 5.459 2.126E-03 1.605E-01 228731 Nkx2-4 5.312 1.291E-03 1.096E-01 677321 Gm9712 5.285 1.536E-02 5.554E-01 319520 Dusp4 4.869 2.347E-20 7.515E-17 16409 Itgam 4.864 4.975E-18 1.138E-14 100316663 Mir669k 4.820 9.158E-03 4.017E-01 271849 Shc4 4.762 1.354E-06 3.675E-04 16971 Lrp1 4.761 1.483E-10 1.131E-07 66395 Ahnak 4.590 2.725E-17 4.848E-14 432572 Specc1 4.562 4.614E-03 2.610E-01 100416030 Gm17879 4.546 1.012E-03 9.310E-02 115488120 LOC115488120 4.415 2.166E-03 1.622E-01 102637682 Gm26682 4.375 5.217E-05 8.108E-03 258571 Olfr1033 4.282 3.346E-04 3.998E-02 79221 Hdac9 4.260 8.312E-07 2.420E-04 115489725 LOC115489725 4.247 6.149E-03 3.165E-01 67603 Dusp6 4.234 5.354E-22 2.143E-18 57269 Olfr1507 4.233 9.944E-04 9.255E-02 108168851 Gm38237 4.231 6.789E-04 6.945E-02 14369 Fzd7 4.228 3.292E-07 1.054E-04 319616 Fzd10os 4.215 4.333E-04 4.851E-02 117167 Steap4 4.164 4.031E-06 8.881E-04 115487985 LOC115487985 4.100 2.569E-02 7.082E-01 13040 Ctss 4.074 1.125E-05 2.218E-03 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

100038577 Gm10790 4.069 1.743E-02 5.913E-01 106522 Pkdcc 4.062 1.846E-03 1.449E-01 228911 Tshz2 4.044 1.413E-04 1.951E-02 105242938 Gm39010 4.016 3.485E-02 8.090E-01 23831 Car14 4.000 3.948E-03 2.403E-01 18627 Per2 3.998 8.699E-07 2.487E-04 71398 5430427O19Rik 3.976 3.317E-02 7.907E-01 17386 Mmp13 3.952 5.542E-03 2.927E-01 102465194 Mir6374 3.895 1.850E-02 6.140E-01 433365 Teddm1b 3.870 1.204E-03 1.054E-01 75202 Spaca6 3.813 2.860E-02 7.442E-01 114715 Spred1 3.795 6.548E-10 4.032E-07 100503338 Gm14636 3.752 2.023E-04 2.676E-02 105246219 Gm41532 3.717 3.071E-02 7.621E-01 102640671 Gm36685 3.684 1.657E-02 5.833E-01 59036 Dact1 3.666 3.114E-08 1.278E-05 223927 Gtsf2 3.637 1.319E-02 5.064E-01 115490302 LOC115490302 3.636 1.304E-02 5.020E-01 12862 Cox6a2 3.612 1.277E-03 1.094E-01 72748 Hdhd3 3.601 1.044E-02 4.295E-01 56461 Kcnip3 3.581 2.019E-06 5.131E-04 105244320 Gm39954 3.577 2.034E-02 6.390E-01 102644 Oaf 3.571 2.557E-04 3.249E-02 432800 Gm5454 3.569 4.508E-02 9.056E-01 207393 Elfn2 3.550 2.363E-02 6.821E-01 100503495 A930016O22Rik 3.506 3.055E-02 7.617E-01 115489534 LOC115489534 3.483 2.039E-03 1.547E-01 216190 Appl2 3.478 7.844E-03 3.655E-01 216197 Ckap4 3.434 1.489E-02 5.418E-01 105246679 Gm37053 3.416 2.449E-06 6.031E-04 115489127 LOC115489127 3.409 4.741E-02 9.222E-01 230612 Slc5a9 3.395 3.438E-09 1.720E-06 17533 Mrc1 3.391 2.500E-03 1.772E-01 19341 Rab4a 3.381 1.463E-02 5.365E-01 17067 Ly6c1 3.372 7.657E-05 1.114E-02 102643244 Gm16252 3.362 7.738E-03 3.647E-01 320706 Soga1 3.354 2.892E-04 3.553E-02 93708 Pcdhgc5 3.325 9.674E-03 4.108E-01 105246945 Gm42139 3.313 4.165E-03 2.460E-01 102631635 Gm29927 3.307 3.721E-03 2.320E-01 231014 9330182L06Rik 3.303 9.538E-04 8.982E-02 105244068 Gm39746 3.302 3.045E-05 5.102E-03 320051 Exph5 3.294 2.145E-02 6.572E-01 115488309 LOC115488309 3.283 3.171E-02 7.741E-01 102640071 Gm36227 3.262 4.600E-02 9.111E-01 14012 Mpzl2 3.253 2.473E-03 1.772E-01 100040500 Gm2808 3.246 1.270E-07 4.325E-05 105244392 Gm40011 3.245 4.795E-03 2.665E-01 102637109 Gm34004 3.237 8.677E-03 3.891E-01 20720 Serpine2 3.235 4.480E-05 7.173E-03 26912 Gcat 3.231 4.597E-02 9.111E-01 69565 2310015K22Rik 3.231 7.588E-03 3.612E-01 329152 Hecw2 3.211 2.787E-06 6.659E-04 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

105243151 Gm39150 3.197 7.112E-03 3.482E-01 14800 Gria2 3.164 4.337E-06 9.382E-04 102640320 Gm36417 3.105 2.562E-03 1.783E-01 100040471 Gm2792 3.091 1.451E-06 3.808E-04 54486 Hpgds 3.090 3.225E-02 7.788E-01 20271 Scn5a 3.075 3.493E-10 2.237E-07 12443 Ccnd1 3.062 2.703E-02 7.271E-01 320642 A630066F11Rik 3.039 2.712E-02 7.286E-01 215303 Camk1g 3.027 1.750E-05 3.295E-03 69456 Commd10 3.014 5.636E-08 2.050E-05 23963 Tenm1 3.012 4.921E-08 1.832E-05 69660 Tmbim1 3.010 2.573E-02 7.082E-01 329324 Syt14 3.007 2.745E-10 1.831E-07 65962 Slc9a3r2 2.978 7.194E-03 3.490E-01 277154 Nynrin 2.976 7.794E-05 1.124E-02 101631 Pwwp2b 2.969 5.076E-03 2.802E-01 12259 C1qa 2.966 2.465E-03 1.772E-01 73619 1700122E12Rik 2.964 1.641E-02 5.821E-01 54377 Cacng4 2.957 5.039E-03 2.791E-01 11676 Aldoc 2.951 2.191E-02 6.644E-01 17133 Maff 2.941 1.169E-07 4.070E-05 100043475 Gm4462 2.901 2.575E-02 7.082E-01 320609 Strip2 2.884 1.159E-07 4.070E-05 66808 9030624G23Rik 2.877 3.740E-02 8.326E-01 13008 Csrp2 2.868 8.384E-04 8.086E-02 23805 Apc2 2.866 9.368E-03 4.032E-01 110183 Rn18s-rs5 2.864 1.084E-02 4.404E-01 11768 Ap1m2 2.849 3.164E-02 7.741E-01 17240 Mdfi 2.831 3.493E-03 2.228E-01 72575 C430049B03Rik 2.826 9.715E-03 4.109E-01 102631998 Gm30181 2.812 9.366E-03 4.032E-01 100042856 Gm4070 2.808 3.697E-02 8.275E-01 108167997 Gm46347 2.791 4.400E-03 2.507E-01 108167802 Gm46218 2.791 1.493E-02 5.422E-01 234577 Cpne2 2.790 4.654E-03 2.624E-01 115490468 LOC115490468 2.784 4.464E-02 8.989E-01 14299 Ncs1 2.771 2.817E-02 7.412E-01 19699 Reln 2.767 2.329E-02 6.820E-01 74272 1700054O19Rik 2.761 1.975E-02 6.319E-01 19261 Sirpa 2.724 1.838E-02 6.117E-01 102632297 Gm30409 2.713 1.871E-02 6.158E-01 102639290 Gm11963 2.694 6.904E-03 3.420E-01 110454 Ly6a 2.682 7.745E-03 3.647E-01 102465887 Mir8095 2.681 4.051E-02 8.590E-01 68428 Steap3 2.679 4.138E-02 8.648E-01 14345 Fut4 2.672 1.949E-02 6.267E-01 22598 Slc6a18 2.669 2.678E-02 7.257E-01 384783 Irs2 2.667 1.177E-10 9.916E-08 100042846 Gm12057 2.657 3.630E-02 8.245E-01 115487425 LOC115487425 2.656 2.954E-02 7.506E-01 108153 Adamts7 2.654 3.252E-02 7.818E-01 115487746 LOC115487746 2.648 6.462E-03 3.253E-01 100381209 Gm17823 2.639 9.135E-03 4.017E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

100041290 D4Ertd617e 2.636 4.899E-02 9.416E-01 102637465 Gm34268 2.633 2.650E-03 1.829E-01 12505 Cd44 2.630 1.854E-04 2.473E-02 18548 Pcsk1 2.617 3.400E-06 7.666E-04 230126 Shb 2.596 2.241E-02 6.693E-01 12043 Bcl2 2.593 1.951E-08 8.925E-06 105244629 Gm40210 2.590 3.766E-04 4.401E-02 115489728 LOC115489728 2.584 4.023E-03 2.421E-01 330222 Sdk1 2.583 2.425E-04 3.106E-02 231834 Snx8 2.538 3.111E-02 7.650E-01 108168075 Gm46408 2.526 4.086E-04 4.640E-02 14960 H2-Aa 2.516 1.155E-02 4.598E-01 115488600 LOC115488600 2.513 4.000E-02 8.590E-01 99296 Hrh3 2.511 9.997E-03 4.147E-01 108075 Ltbp4 2.510 6.697E-03 3.350E-01 432881 Gm5466 2.499 9.308E-03 4.032E-01 115490428 LOC115490428 2.490 4.049E-02 8.590E-01 78134 Lpar4 2.484 9.317E-04 8.825E-02 102634049 Gm31728 2.481 2.293E-02 6.795E-01 330409 Cecr2 2.473 8.259E-04 8.062E-02 16364 Irf4 2.463 8.652E-05 1.237E-02 15165 Hcn1 2.459 9.925E-03 4.138E-01 115486464 LOC115486464 2.449 9.213E-03 4.019E-01 100125929 0610009E02Rik 2.449 3.537E-02 8.148E-01 432530 Adcy1 2.444 7.415E-07 2.240E-04 69358 Lrrc51 2.434 3.580E-02 8.187E-01 105244435 Gm40043 2.431 9.897E-03 4.137E-01 231296 Lrrc66 2.431 4.230E-02 8.747E-01 83554 Fstl3 2.429 1.559E-02 5.608E-01 69282 1700001J03Rik 2.421 2.776E-02 7.394E-01 104099 Itga9 2.420 9.728E-03 4.109E-01 102640631 Gm36649 2.417 3.595E-02 8.199E-01 100121 Tdrd7 2.395 3.072E-02 7.621E-01 227377 Farp2 2.384 3.685E-02 8.273E-01 239319 Card6 2.383 4.615E-02 9.111E-01 17096 Lyn 2.382 1.274E-02 4.940E-01 83674 Cnnm1 2.380 4.622E-02 9.112E-01 382421 Gm5176 2.379 1.699E-02 5.854E-01 100526560 Mir3109 2.378 4.518E-02 9.065E-01 209012 Ulk4 2.367 2.524E-02 7.027E-01 108169013 Gm46894 2.362 2.179E-03 1.622E-01 105352 Dusp22 2.341 6.542E-03 3.283E-01 433586 Maml3 2.337 2.855E-05 4.863E-03 105244436 Gm37548 2.337 2.772E-08 1.168E-05 115489936 LOC115489936 2.330 5.224E-03 2.842E-01 80890 Trim2 2.323 2.366E-02 6.821E-01 108168987 Gm13205 2.323 1.949E-02 6.267E-01 231510 Gpat3 2.320 3.217E-05 5.254E-03 101113 Snx21 2.320 3.885E-02 8.486E-01 22329 Vcam1 2.315 2.601E-02 7.134E-01 16768 Lag3 2.301 4.028E-04 4.639E-02 268527 Greb1 2.283 2.083E-02 6.495E-01 67260 Cers4 2.283 7.603E-03 3.612E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

100043335 Gm4371 2.280 3.544E-02 8.152E-01 21416 Tcf7l2 2.279 2.380E-02 6.821E-01 12723 Clcn1 2.264 1.242E-02 4.861E-01 73442 Hspa12a 2.260 4.441E-02 8.971E-01 70375 Ica1l 2.259 1.010E-03 9.310E-02 15360 Hmgcs2 2.253 2.224E-05 3.870E-03 20430 Cyfip1 2.253 1.481E-03 1.210E-01 100417687 Gm12846 2.252 2.439E-03 1.767E-01 319939 Tns3 2.246 1.822E-05 3.353E-03 320225 Catsperg1 2.244 3.004E-02 7.564E-01 66725 Lrrk2 2.235 3.526E-02 8.145E-01 546840 Ldlrad1 2.233 1.457E-02 5.363E-01 72007 Fndc3b 2.233 6.421E-03 3.253E-01 16149 Cd74 2.232 4.101E-03 2.450E-01 17476 Mpeg1 2.230 4.397E-02 8.955E-01 102637099 Gm38495 2.224 1.562E-03 1.256E-01 115486901 LOC115486901 2.220 1.479E-03 1.210E-01 108168376 Gm46620 2.214 6.210E-03 3.176E-01 108168205 Gm46489 2.209 3.251E-02 7.818E-01 13992 Khdrbs3 2.198 2.875E-02 7.443E-01 101809 Spred3 2.197 1.814E-02 6.063E-01 319743 9630013D21Rik 2.192 8.822E-03 3.923E-01 21873 Tjp2 2.190 1.419E-02 5.319E-01 12260 C1qb 2.183 3.747E-02 8.326E-01 100503470 Gm16619 2.180 1.688E-02 5.843E-01 229715 Amigo1 2.173 3.729E-03 2.320E-01 17181 Matn2 2.170 1.478E-03 1.210E-01 107815 Scml2 2.159 1.452E-02 5.361E-01 102634592 Gm32133 2.153 4.216E-02 8.744E-01 102640410 Gm36475 2.148 2.802E-02 7.412E-01 208727 Hdac4 2.147 4.049E-06 8.881E-04 77411 Esrp2 2.144 3.784E-05 6.119E-03 73094 Sgip1 2.140 4.393E-03 2.507E-01 108167732 Gm46162 2.139 2.105E-02 6.532E-01 13982 Esr1 2.131 4.263E-03 2.502E-01 100764 Rita1 2.129 1.179E-03 1.037E-01 71302 Arhgap26 2.092 2.090E-05 3.677E-03 386454 Rnf39 2.073 3.425E-02 8.054E-01 433520 Gm14403 2.064 1.884E-02 6.158E-01 110197 Dgkg 2.056 6.982E-07 2.149E-04 67425 Eps8l1 2.055 1.056E-03 9.553E-02 52055 Rab11fip5 2.050 3.300E-02 7.896E-01 208080 Ubap1l 2.046 1.680E-02 5.843E-01 621304 Gm6209 2.044 7.908E-04 7.863E-02 18563 Pcx 2.036 1.363E-02 5.185E-01 226751 Cdc42bpa 2.036 4.267E-03 2.502E-01 18029 Nfic 2.035 1.409E-02 5.310E-01 71354 Wdr31 2.029 2.828E-02 7.421E-01 108167659 LOC108167659 2.025 7.957E-03 3.683E-01 667481 Trav6d-5 2.023 9.011E-03 3.996E-01 19250 Ptpn14 2.012 3.151E-06 7.312E-04 360213 Trim46 2.002 1.690E-02 5.843E-01 229731 Slc25a24 2.001 1.313E-03 1.100E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

399591 Tmsb15l -2.018 4.308E-03 2.507E-01 20620 Plk2 -2.025 4.574E-02 9.111E-01 74050 4921525O09Rik -2.028 4.669E-04 5.155E-02 56043 Akr1e1 -2.030 3.107E-02 7.650E-01 17064 Cd93 -2.031 2.832E-02 7.421E-01 102637967 Gm34647 -2.040 4.912E-02 9.417E-01 23969 Pacsin1 -2.042 9.998E-03 4.147E-01 115490493 LOC115490493 -2.056 3.644E-02 8.245E-01 50785 Hs6st1 -2.061 1.439E-06 3.808E-04 65945 Clstn1 -2.063 2.820E-02 7.412E-01 240120 Zfp119b -2.081 3.489E-02 8.090E-01 17724 mt-Rnr1 -2.088 1.999E-02 6.336E-01 67775 Rtp4 -2.091 2.213E-02 6.685E-01 68992 Zfp580 -2.092 4.083E-02 8.623E-01 71091 Cdkl1 -2.094 5.211E-03 2.842E-01 15024 H2-T10 -2.106 3.059E-05 5.102E-03 381417 Gm14085 -2.112 1.013E-05 2.027E-03 115487084 LOC115487084 -2.119 3.085E-02 7.623E-01 19331 Rab19 -2.122 5.392E-04 5.754E-02 75782 Lca5 -2.130 1.602E-04 2.155E-02 14168 Fgf13 -2.131 1.959E-03 1.508E-01 59010 Sqor -2.134 5.816E-05 8.867E-03 115488064 LOC115488064 -2.135 3.551E-02 8.156E-01 52076 Tmem38b -2.138 1.760E-02 5.958E-01 233552 Gdpd5 -2.144 3.379E-02 7.993E-01 73991 Atl1 -2.155 2.697E-02 7.271E-01 115490050 LOC115490050 -2.156 3.258E-02 7.820E-01 58220 Pard6b -2.163 1.323E-02 5.066E-01 81913 Bambi-ps1 -2.164 2.046E-05 3.639E-03 107022 Gramd3 -2.165 2.744E-04 3.405E-02 102640217 Gm36338 -2.167 4.234E-02 8.747E-01 666348 Apol7e -2.168 4.025E-04 4.639E-02 105242677 Gm38828 -2.170 3.448E-03 2.208E-01 72121 Dennd2d -2.183 9.243E-06 1.873E-03 14114 Fbln1 -2.183 8.037E-03 3.692E-01 100217422 Snord13 -2.195 2.435E-02 6.899E-01 100034675 Gm11335 -2.205 3.198E-02 7.769E-01 102632039 Gm30214 -2.219 4.017E-02 8.590E-01 654824 Ankrd37 -2.221 3.175E-03 2.066E-01 664779 Gm44505 -2.229 9.247E-03 4.023E-01 217882 Cep170b -2.244 7.412E-03 3.553E-01 77032 Tstd3 -2.248 2.370E-02 6.821E-01 386655 Eid2 -2.250 3.800E-02 8.360E-01 414094 Dgkeos -2.255 1.861E-05 3.386E-03 108637 Snord14c -2.255 2.028E-02 6.390E-01 140577 Ankrd6 -2.261 9.148E-06 1.873E-03 242126 Slc22a15 -2.264 4.738E-03 2.643E-01 278679 Apol7b -2.269 3.760E-03 2.324E-01 56325 Abcb9 -2.271 2.977E-06 7.009E-04 21942 Tnfrsf9 -2.294 3.844E-02 8.440E-01 20408 Sh3gl3 -2.302 1.387E-02 5.262E-01 115486004 LOC115486004 -2.302 3.803E-03 2.333E-01 102633785 Gm31527 -2.305 2.330E-02 6.820E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

106952 Arap3 -2.305 3.418E-02 8.054E-01 22436 Xdh -2.308 3.326E-03 2.156E-01 108167368 LOC108167368 -2.312 3.043E-02 7.613E-01 19122 Prnp -2.312 5.151E-04 5.610E-02 102637188 Gm34065 -2.323 2.376E-02 6.821E-01 14073 Faah -2.324 1.904E-05 3.424E-03 115487687 LOC115487687 -2.328 1.136E-05 2.218E-03 241489 Pde11a -2.334 6.455E-03 3.253E-01 115487522 LOC115487522 -2.335 5.152E-03 2.824E-01 214230 Pak6 -2.340 2.533E-03 1.774E-01 67252 Cap2 -2.344 1.734E-02 5.912E-01 105246174 Gm41508 -2.349 2.885E-02 7.450E-01 16574 Kif5c -2.356 2.927E-02 7.504E-01 211027 Trav3-4 -2.360 1.540E-02 5.554E-01 666244 Tmsb15b1 -2.369 1.895E-02 6.178E-01 76220 6530402F18Rik -2.369 1.070E-02 4.368E-01 666240 Gm8000 -2.369 3.559E-02 8.162E-01 665367 Gm12350 -2.372 3.789E-02 8.360E-01 115486514 LOC115486514 -2.410 3.692E-02 8.275E-01 100039355 Rps16-ps2 -2.414 4.396E-03 2.507E-01 57440 Ehd3 -2.418 2.246E-06 5.619E-04 378937 Lrrc24 -2.421 3.877E-02 8.480E-01 53314 Batf -2.426 1.827E-03 1.441E-01 12819 Col15a1 -2.452 7.498E-03 3.583E-01 83671 Sytl2 -2.452 7.192E-04 7.287E-02 74466 Mfsd13b -2.453 3.026E-02 7.592E-01 140488 Igf2bp3 -2.481 9.048E-04 8.622E-02 51789 Tnk2 -2.487 3.358E-09 1.720E-06 102633360 Gm31204 -2.489 2.243E-02 6.693E-01 102633439 Gm14154 -2.501 2.524E-03 1.774E-01 100042235 Gm3739 -2.501 4.285E-03 2.504E-01 15504 Dnajb3 -2.522 4.160E-02 8.675E-01 73379 Dcbld2 -2.531 2.724E-02 7.288E-01 595140 Gm6044 -2.547 4.954E-02 9.434E-01 65116 Prrg2 -2.551 2.723E-02 7.288E-01 102636372 Gm33457 -2.560 2.871E-02 7.443E-01 115488085 LOC115488085 -2.562 3.005E-02 7.564E-01 240066 Zfp870 -2.574 1.487E-02 5.418E-01 229672 Bcl2l15 -2.581 1.912E-02 6.222E-01 20660 Sorl1 -2.584 1.569E-04 2.129E-02 102636200 Gm33335 -2.591 1.969E-02 6.316E-01 243374 Gimap8 -2.598 2.949E-09 1.574E-06 12608 Cebpb -2.620 3.078E-03 2.036E-01 22351 Vill -2.621 2.170E-02 6.629E-01 20667 Sox12 -2.625 2.301E-02 6.795E-01 100101942 Trdv2-1 -2.629 3.454E-02 8.060E-01 13819 Epas1 -2.629 2.807E-02 7.412E-01 102638019 Gm34687 -2.634 2.435E-02 6.899E-01 115487892 LOC115487892 -2.635 1.223E-06 3.374E-04 102640377 Gm36449 -2.642 1.991E-02 6.325E-01 67795 Rnls -2.648 3.266E-02 7.827E-01 18559 Pctp -2.655 8.587E-03 3.870E-01 70080 Igsf23 -2.664 2.255E-04 2.911E-02 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

115487690 LOC115487690 -2.666 6.614E-05 9.896E-03 102631948 Gm30149 -2.689 7.959E-03 3.683E-01 12772 Ccr2 -2.691 4.015E-03 2.421E-01 102635138 Gm38474 -2.701 3.000E-02 7.564E-01 77428 9430083A17Rik -2.714 1.978E-02 6.319E-01 626578 Gbp10 -2.760 7.743E-07 2.295E-04 332579 Card9 -2.796 1.022E-06 2.870E-04 100503915 Smpd5 -2.798 7.376E-03 3.546E-01 105245263 Gm40741 -2.804 2.308E-03 1.687E-01 224129 Adcy5 -2.808 1.297E-02 5.016E-01 115490291 LOC115490291 -2.813 3.184E-02 7.758E-01 105841 Dennd3 -2.832 1.102E-03 9.852E-02 76469 Cmya5 -2.839 8.853E-05 1.254E-02 108168823 Gm46781 -2.868 3.324E-02 7.907E-01 330483 Ceacam16 -2.877 2.047E-02 6.413E-01 105243145 Gm39145 -2.878 1.453E-02 5.361E-01 115489019 LOC115489019 -2.885 1.466E-02 5.365E-01 229665 Ampd1 -2.907 4.860E-05 7.628E-03 73919 Lyrm1 -2.912 9.796E-03 4.116E-01 21753 Tes -2.955 4.173E-08 1.590E-05 26458 Slc27a2 -2.961 1.853E-02 6.140E-01 115488025 LOC115488025 -2.962 4.017E-02 8.590E-01 102640419 Gm16322 -2.967 1.539E-02 5.554E-01 75744 Svip -3.005 1.944E-02 6.267E-01 115490087 LOC115490087 -3.030 3.103E-02 7.650E-01 230379 Acer2 -3.039 1.052E-02 4.308E-01 11981 Atp9a -3.049 6.786E-03 3.384E-01 56643 Slc15a1 -3.057 6.942E-03 3.420E-01 15208 Hes5 -3.061 2.218E-02 6.686E-01 100702 Gbp6 -3.070 2.303E-12 2.634E-09 100041622 Gm3436 -3.070 3.597E-02 8.199E-01 218763 Lrrc3b -3.076 8.003E-03 3.692E-01 64113 Moap1 -3.081 2.332E-02 6.820E-01 666031 Gm7899 -3.115 1.770E-02 5.971E-01 11490 Adam15 -3.121 7.814E-03 3.655E-01 100041925 Gm14586 -3.141 1.743E-02 5.913E-01 105246085 Gm41432 -3.170 5.933E-03 3.084E-01 70727 Rasgef1a -3.174 1.410E-02 5.310E-01 19023 Ppef2 -3.199 1.264E-02 4.924E-01 225825 Cd226 -3.212 1.666E-05 3.213E-03 13052 Cxadr -3.224 1.698E-03 1.352E-01 442803 A830005F24Rik -3.241 9.595E-03 4.085E-01 16400 Itga3 -3.265 3.683E-03 2.320E-01 20440 St6gal1 -3.267 1.965E-09 1.123E-06 20613 Snai1 -3.271 8.051E-03 3.692E-01 105242589 Gm38759 -3.287 9.572E-03 4.085E-01 225392 Rell2 -3.289 7.206E-03 3.490E-01 115489965 LOC115489965 -3.300 3.934E-03 2.403E-01 115489981 LOC115489981 -3.346 7.441E-04 7.445E-02 14200 Fhl2 -3.452 5.904E-04 6.218E-02 71141 4933407L21Rik -3.523 2.947E-02 7.506E-01 12877 Cpeb1 -3.543 2.726E-05 4.692E-03 280047 Gm5075 -3.546 3.977E-02 8.590E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

58233 Dnaja4 -3.550 2.198E-02 6.651E-01 243923 Rgs9bp -3.590 1.875E-02 6.158E-01 70839 P2ry12 -3.591 1.272E-02 4.940E-01 434782 Gm5637 -3.609 4.771E-02 9.247E-01 70821 4921507P07Rik -3.632 2.381E-02 6.821E-01 100043225 Gm16378 -3.669 2.796E-02 7.412E-01 102631731 Gm29998 -3.696 2.670E-03 1.835E-01 71981 Tdrd12 -3.703 3.707E-02 8.277E-01 433739 Gm12799 -3.743 1.335E-02 5.102E-01 77794 Adamtsl2 -3.753 1.420E-07 4.737E-05 22055 n-THgtg3 -3.767 4.868E-02 9.390E-01 102634915 Gm12268 -3.789 1.452E-02 5.361E-01 620772 Gm6180 -3.793 2.179E-02 6.631E-01 100043843 Rps15a-ps7 -3.829 3.055E-02 7.617E-01 381404 Pabpc1l -3.866 2.215E-03 1.642E-01 115488237 LOC115488237 -3.867 1.989E-03 1.523E-01 12671 Chrm3 -3.906 1.808E-02 6.054E-01 105245236 LOC105245236 -3.947 3.321E-02 7.907E-01 18762 Prkcz -4.008 3.257E-08 1.304E-05 433938 Mn1 -4.012 3.134E-03 2.051E-01 100862247 Gm21586 -4.036 1.301E-02 5.019E-01 16776 Lama5 -4.056 1.382E-03 1.153E-01 218695 Gm10044 -4.137 1.178E-02 4.675E-01 18227 Nr4a2 -4.179 2.172E-03 1.622E-01 115487436 LOC115487436 -4.193 2.603E-02 7.134E-01 108167895 Gm46277 -4.219 2.140E-09 1.181E-06 115487786 LOC115487786 -4.287 4.114E-02 8.624E-01 105193 Nhlrc1 -4.296 1.209E-02 4.742E-01 74901 Kbtbd11 -4.308 3.885E-12 3.887E-09 231603 A630023P12Rik -4.324 2.491E-03 1.772E-01 433070 Vmn2r96 -4.328 1.096E-03 9.852E-02 22652 Mkrn3 -4.360 2.517E-02 7.019E-01 231932 Gimap7 -4.468 2.681E-06 6.504E-04 665004 Gm7443 -4.471 1.095E-02 4.426E-01 14702 Gng2 -4.583 3.692E-04 4.346E-02 329628 Fat4 -4.596 6.037E-05 9.118E-03 108168021 Gm46370 -4.616 2.986E-02 7.564E-01 102636184 Gm33323 -4.629 9.653E-04 9.037E-02 100416808 Zfp422-ps -4.810 2.186E-02 6.639E-01 676984 Gm9701 -4.813 5.389E-03 2.886E-01 272009 Srsf12 -4.853 4.080E-04 4.640E-02 667977 Gm8909 -5.021 2.470E-08 1.098E-05 218476 Gcnt4 -5.060 1.798E-05 3.346E-03 115488400 LOC115488400 -5.669 2.503E-02 7.018E-01 319574 Gm20735 -5.729 2.727E-03 1.866E-01 20564 Slit3 -5.827 1.230E-04 1.727E-02 12672 Chrm4 -5.903 3.434E-03 2.208E-01 69717 Gm10499 -6.067 1.221E-11 1.086E-08 387134 Mir16-1 -6.070 2.503E-02 7.018E-01 102467530 n-TGgcc2 -6.671 4.451E-03 2.527E-01 15530 Hspg2 -6.799 2.943E-12 3.141E-09 50930 Tnfsf14 -7.277 1.343E-10 1.075E-07 115487043 LOC115487043 -7.728 6.346E-03 3.235E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

16336 Insl3 -7.905 2.019E-02 6.389E-01 259277 Klk8 -8.308 1.819E-06 4.697E-04 100736249 Mira -8.640 2.164E-04 2.817E-02 102634419 Gm32004 -8.899 1.230E-19 3.281E-16 102634148 Gm31803 -10.816 1.129E-11 1.064E-08 12895 Cpt1b -14.440 1.430E-15 1.908E-12 115485636 LOC115485636 -16.598 1.294E-03 1.096E-01 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Table S2-1. Gene Ontology (GO) analysis of the DEGs in Cic-deficient DP thymocytes (upregulated DEGs) List Pop Pop Fold Category Term Count % PValue Genes Bonferroni Benjamini FDR Total Hits Total Enrichment TRPM2, GO:0071277 MEF2C, GOTERM_BP ~cellular CPNE2, 6 2.586 0.000 145 53 18082 14.117 0.071 0.074 0.074 _DIRECT response to ADCY1, calcium ion SCN5A, SLC25A24 GO:0000188 SPRED3, GOTERM_BP ~inactivation DUSP4, 4 1.724 0.000 145 15 18082 33.254 0.218 0.123 0.122 _DIRECT of MAPK SPRED1, activity DUSP6 APC2, STRIP2, GO:0016477 RELN, GOTERM_BP ~cell 8 3.448 0.001 FNDC3B, 145 191 18082 5.223 0.633 0.241 0.240 _DIRECT migration SIRPA, CDC42BPA, TNS3, CD44 GO:0051966 GRIA2, ~regulation of GOTERM_BP MEF2C, synaptic 4 1.724 0.001 145 24 18082 20.784 0.646 0.241 0.240 _DIRECT SERPINE2, transmission, LRRK2 glutamatergic GO:0030178 ~negative APC2, LRP1, GOTERM_BP regulation of 5 2.155 0.001 CCND1, MDFI, 145 56 18082 11.134 0.700 0.241 0.240 _DIRECT Wnt signaling DACT1 pathway DLX1, CYFIP1, GO:0007275 MEF2C, GOTERM_BP ~multicellular PKDCC, 19 8.190 0.001 145 1029 18082 2.303 0.803 0.270 0.269 _DIRECT organism LRP1, development SERPINE2, FZD7, AMIGO1, bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

SPRY4, TSHZ2, TDRD7, SHB, SPRED3, SPRED1, CSRP2, RELN, NKX2- 4, MDFI, DACT1 APC2, CSF1R, GO:0008283 CD74, GOTERM_BP ~cell 8 3.448 0.002 TCF7L2, 145 220 18082 4.535 0.896 0.323 0.322 _DIRECT proliferation LRP1, BCL2, IRS2, APPL2 GO:0043523 ~regulation of MEF2C, GOTERM_BP neuron 4 1.724 0.002 TRIM2, 145 33 18082 15.116 0.930 0.332 0.330 _DIRECT apoptotic KCNIP3, ESR1 process GO:0050679 TCF7L2, ~positive GOTERM_BP HRH3, regulation of 5 2.155 0.003 145 76 18082 8.204 0.976 0.413 0.412 _DIRECT ESRP2, epithelial cell SCN5A, ESR1 proliferation KCNIP3, GO:0034765 KCNJ16, ~regulation of GOTERM_BP SCN5A, ion 6 2.586 0.005 145 135 18082 5.542 0.995 0.472 0.470 _DIRECT CLCN1, transmembra CACNG4, ne transport HCN1 GO:0070932 PER2, GOTERM_BP ~histone H3 3 1.293 0.005 HDAC4, 145 13 18082 28.778 0.996 0.472 0.470 _DIRECT deacetylation HDAC9 GO:0030890 ~positive GOTERM_BP CD74, MEF2C, regulation of 4 1.724 0.005 145 43 18082 11.600 0.997 0.472 0.470 _DIRECT BCL2, IRS2 B cell proliferation bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

TRPM2, GRIA2, STEAP3, SLC5A9, CNNM1, GO:0006811 GOTERM_BP STEAP4, ~ion 12 5.172 0.007 145 584 18082 2.562 1.000 0.607 0.605 _DIRECT KCNIP3, transport KCNJ16, SCN5A, CLCN1, CACNG4, HCN1 GO:0019882 ~antigen GOTERM_BP CD74, RAB4A, processing 4 1.724 0.009 145 54 18082 9.237 1.000 0.607 0.605 _DIRECT H2-AA, CTSS and presentation DLX1, CYFIP1, MEF2C, PKDCC, GO:0030154 SERPINE2, GOTERM_BP ~cell 14 6.034 0.010 LRRK2, 145 780 18082 2.238 1.000 0.607 0.605 _DIRECT differentiation AMIGO1, CHD5, TDRD7, SHB, DUSP6, ETV5, CSRP2, MDFI

bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Table S2-2. Gene Ontology (GO) analysis of the DEGs in Cic-deficient DP thymocytes (downregulated DEGs) List Pop Pop Fold Category Term Count % PValue Genes Bonferroni Benjamini FDR Total Hits Total Enrichment CEBPB, ITGA3, GOTERM_BP GO:0007613 6 3.141 0.000 PLK2, PAK6, 113 83 18082 11.568 0.114 0.121 0.121 _DIRECT ~memory FGF13, KLK8 LAMA5, COL15A1, CXADR, GO:0007155 GOTERM_BP ADAM15, ~cell 10 5.236 0.003 113 485 18082 3.299 0.905 1.000 1.000 _DIRECT CD93, ITGA3, adhesion CLSTN1, CD226, FAT4, HES5 LAMA5, GO:0016477 GOTERM_BP RASGEF1A, ~cell 6 3.141 0.007 113 191 18082 5.027 0.993 1.000 1.000 _DIRECT TNK2, SNAI1, migration SORL1, PRKCZ GO:0042832 GOTERM_BP ~defense GBP6, GBP10, 3 1.571 0.015 113 30 18082 16.002 1.000 1.000 1.000 _DIRECT response to BATF protozoan GO:0045408 ~regulation of GOTERM_BP CEBPB, interleukin-6 2 1.047 0.018 113 3 18082 106.678 1.000 1.000 1.000 _DIRECT CARD9 biosynthetic process SH3GL3, TNK2, GOTERM_BP GO:0006897 SORL1, 5 2.618 0.026 113 181 18082 4.420 1.000 1.000 1.000 _DIRECT ~endocytosis ATP9A, PACSIN1 bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

GO:0070885 ~negative regulation of GOTERM_BP calcineurin- 2 1.047 0.031 PRNP, CMYA5 113 5 18082 64.007 1.000 1.000 1.000 _DIRECT NFAT signaling cascade GO:0006869 APOL7B, GOTERM_BP ~lipid 4 2.094 0.032 PCTP, SORL1, 113 111 18082 5.766 1.000 1.000 1.000 _DIRECT transport APOL7E GO:0030198 LAMA5, GOTERM_BP ~extracellular 4 2.094 0.034 ADAMTSL2, 113 114 18082 5.615 1.000 1.000 1.000 _DIRECT matrix FBLN1, HSPG2 organization GO:0007157 ~heterophilic cell-cell GOTERM_BP adhesion via CXADR, 3 1.571 0.041 113 52 18082 9.232 1.000 1.000 1.000 _DIRECT plasma CD226, FAT4 membrane cell adhesion molecules GO:0007197 ~adenylate cyclase- inhibiting G- GOTERM_BP protein CHRM3, 2 1.047 0.043 113 7 18082 45.719 1.000 1.000 1.000 _DIRECT coupled CHRM4 acetylcholine receptor signaling pathway GO:0007207 ~phospholipa GOTERM_BP se C- CHRM3, 2 1.047 0.043 113 7 18082 45.719 1.000 1.000 1.000 _DIRECT activating G- CHRM4 protein coupled bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

acetylcholine receptor signaling pathway GO:0001973 ~adenosine GOTERM_BP P2RY12, receptor 2 1.047 0.049 113 8 18082 40.004 1.000 1.000 1.000 _DIRECT ADCY5 signaling pathway GO:0072006 GOTERM_BP ~nephron 2 1.047 0.049 ITGA3, FAT4 113 8 18082 40.004 1.000 1.000 1.000 _DIRECT development GO:0015833 GOTERM_BP SLC15A1, ~peptide 2 1.047 0.049 113 8 18082 40.004 1.000 1.000 1.000 _DIRECT ABCB9 transport

bioRxiv preprint doi: https://doi.org/10.1101/2021.07.11.451936; this version posted July 12, 2021. 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 4.0 International license.

Table S3. Oligonucleotide sequences used for qRT-PCR Primer name Sequences Bcl2 (Forward) 5ʹ-ATGCCTTTGTGGAACTATATGGC-3ʹ Bcl2 (Reverse) 5ʹ-GGTATGCACCCAGAGTGATGC-3 Spry4 (Forward) 5ʹ-GCAGCGTCCCTGTGAATCC-3ʹ Spry4 (Reverse) 5ʹ-TCTGGTCAATGGGTAAGATGGT-3ʹ Dusp4 (Forward) 5ʹ-TCCCCGTCGAAGACAACCA-3ʹ Dusp4 (Reverse) 5ʹ-CTTTACTGCGTCGATGTACTCG-3ʹ Dusp6 (Forward) 5ʹ-ATAGATACGCTCAGACCCGTG-3ʹ Dusp6 (Reverse) 5ʹ-ATCAGCAGAAGCCGTTCGTT-3ʹ Spred1 (Forward) 5ʹ-GAGATGACTCAAGTGGTGGATG-3’ Spred1 (Reverse) 5ʹ- TCTGAAAGGTAAGGCCAAACTTC-3ʹ Hprt (Forward) 5ʹ-TCAGTCAACGGGGGACATAAA-3ʹ Hprt (Reverse) 5ʹ- GGGGCTGTACTGCTTAACCAG-3ʹ