1 SUPPLEMENTAL FIGURES 2 RAB11FIP5 fl/fl mice CMV-Cre mice (C57BL/6-Rab11fip5 tm1Sud /J; Stock# 021776) (B6.C-Tg(CMV-cre)1Cgn/J; Stock# 006054)

F1 Genotype RAB11FIP5 fl / fl Genotype RAB11FIP5 wt / wt CMV-Cre X / X X CMV-Cre X Cre / Y

RAB11FIP5 fl/fl mice (C57BL/6-Rab11fip5tm1Sud /J; Stock# 021776)

F2 Genotype RAB11FIP5 fl / wt Genotype RAB11FIP5 fl / wt Genotype RAB11FIP5 fl / fl CMV-Cre X / Y CMV-Cre X Cre / X X CMV-Cre X / Y

Genotype RAB11FIP5 fl / wt Genotype RAB11FIP5 fl / fl Genotype RAB11FIP5 fl / fl Genotype RAB11FIP5 fl / fl CMV-Cre X Cre / X CMV-Cre X Cre / X CMV-Cre XCre / Y CMV-Cre X / Y F3 In 50% animals, CMV-Cre gene In 50% animals, CMV-Cre gene is in the inactive X-chromosome. is in the active X-chromosome. CMV-Cre express in 100% animals. RAB11FIP5 was NOT knocked out. RAB11FIP5 was knocked out. RAB11FIP5 was knocked out.

Genotype RAB11FIP5 fl / fl Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO CMV-Cre X Cre / X CMV-Cre X Cre / X X CMV-Cre X Cre / Y

F4 Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO CMV-Cre X Cre / X Cre CMV-Cre X Cre / X X CMV-Cre X / Y CMV-Cre X Cre / Y

F5 Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO Genotype RAB11FIP5 KO / KO CMV-Cre X Cre / X CMV-Cre X / X X CMV-Cre X / Y CMV-Cre X Cre / Y

-/- 3 RAB11FIP5 mice 4 Figure S1. Generation of constitutive RAB11FIP5 knockout (RAB11FIP5-/-) mice. 5 RAB11FIP5-/- mice were generated by breeding RAB11FIP5 floxed mice (RAB11FIP5fl/fl) with CMV-Cre mice. In 6 the RAB11FIP5fl/fl conditional mutant mouse, exon 2 of the RAB11FIP5 gene is flanked by loxP sites. Mice 7 homozygous for this allele are viable, fertile and produce normal levels of Rab11Fip5 protein. When crossed with a 8 Cre-expressing strain, the floxed exon 2 is excised and therefore the expression of functional RAB11FIP5 mRNA is 9 abolished. In our study, as the CMV-Cre transgene is X-chromosome linked, F1 female RAB11FIP5fl/fl mice were 10 crossed with male CMV-Cre mice. F2 female littermates were kept and bred with RAB11FIP5fl/fl mice. In F3 11 approximately half of the female littermates still had the RAB11FIP5 gene due to X-chromosome inactivation in the 12 CMV-Cre positive allele, while males and the remaining females had RAB11FIP5 deletion. By inbreeding F3 13 RAB11FIP5-/- male and females, we eventually generated constitutive RAB11FIP5-/- mice independent of Cre. All 14 experiments in this study were conducted with mice after F5. 15

1 30 ) 0.25 ) g ( g ( A B t t 0.20 gh i gh i 25 e

e 0.15

W RAB11FIP5 fl/fl W n cm y 0.10

20 ee l od p

B 0.05 RAB11FIP5 -/- S 15 0.00 -/- fl/fl -/- fl/fl C Gating strategy for B cells E Gating strategy for T and NK cells FSC-H SSC-A SSC-H FSC-W SSC-W Live/dead FSC-A SSC-A FSC-A FSC-A SSC-A FSC-A CD8 B220 CD19 CD138 CD62L Live/dead FSC-A IgM B220 TER119 CD4 CD44 NK1-1 CD62L CD43 CD93 CD11b CD3 CD44 B220 B220 CD5 CD25 PD1 CD62L IgD

GL7 Thy1.2 CD127 CXCR5 CD21

IgM CD95 CD23 SSC-A PD1

CXCR5 Thy1.2 D F Spleen B cells Spleen T & NK cells

) 6 n o i 10 l 6 g il 5 RAB11FIP5 fl/fl ) o n l 0

m RAB11FIP5 fl/fl o ( i

1 5 r RAB11FIP5 -/- l

4 l s i e l

l RAB11FIP5 -/- og p l m

e 4 ( r s

c 3 s ll e ll n

p 3 e e ce 2 s c e l . l l o n 2 p e ce s 1 N e . l 1 o 0 p s N 0 Bone Marrow B cells

) 6 Bone Marrow T & NK cells n o i 10 6 l g il 5 RAB11FIP5 fl/fl ) n o 0 l m o RAB11FIP5 fl/fl ( i

1 5 l r 4 RAB11FIP5 -/- l i s e l og l l m p 4 RAB11FIP5 -/- ( e r

s 3 c s e ll ll n p 3 e e ce s

2 c l e l . l

n 2 o p e ce

s 1 N e . l 1 o p s 0 N 0 Lymph Nodes B cells Lymph Nodes T & NK cells ) 6 n

o 6 i 10

l RAB11FIP5 fl/fl ) g

il 5 n

0 RAB11FIP5 fl/fl o o l i m RAB11FIP5 -/- 1 5 l ( l r 4 i s og

e RAB11FIP5 -/- l l m l 4 ( p e r s

s 3 c e ll ll

p 3 n e s c e ce 2 l l e . n l 2 o e p ce s 1 e N . l 1 o p s 0 N 0 ls ls ls ls s ls s ls ls ls s ls l l l l ll l ll l l l ll l ls ls ls ls lls ls ls lls ls ls lls lls ls ls ls ls ls lls ls ls e e e e e e e e e e e e l l l l l l e lls lls lls l l e e l l lls l lls l l e l l c c c c c c c c c c c c e e e e e e e c e e e e e c c e e e e e e e c e e d c c c c c c c c c c c c c c c c c c c c c a a B B B B B B B B i T d B m m ll o l r e r lo i Tfr s s A r re re a te n la e ro Tfh la la P P tu n n u y th All T Treg io e Zo ic P P a it l ll M + - m s C a Ery Total NK All CD4Non-Treg T M n l n o M g Im a i F Thy1.2- Tfh Ig I ra in rg Naive CD8 T Thy1.2+ Tfh Naive CD4 T T m a Effector CD8 T Effector CD4 T r M Ge Thy1.2-CD62L+Thy1.2+CD62L+Thy1.2+CD62L-Thy1.2-CD62L+ NKThy1.2-CD62L+ NKNK1.1+CD3+ NK NK NK NKT Central memoryEffector CD8 memory T CD8 T Central memory CD4 T 16 Effector memory CD4 T

2 17 Figure S2. No major perturbations in lymphocyte subsets in RAB11FIP5-/- mice. 18 (A) Pictures of spleens of representative RAB11FIP5-/- and control RAB11FIP5fl/fl mice. 19 (B) Body and spleen weights of RAB11FIP5-/- (n=11) and control RAB11FIP5fl/fl mice (n=10). 20 (C-D) Phenotypic analysis of B cells in RAB11FIP5-/- (n=3) and control RAB11FIP5fl/fl (n=3) mice. (C) Gating 21 strategy for B cell subsets. (D) Cells from lymph nodes, spleen, and bone marrow were analyzed and the bar graphs 22 show the absolute number of total cells, IgM+ plasma cells (PC; CD138+), IgM- PC, total B cells (CD19+B220+), pro 23 B cells (CD+B220+), pre-B cells (IgM-IgD-), immature (Imm) B cells (B220+CD93+IgM+IgD-), transitional (Trans) B 24 cells (B220+CD93+IgM+IgD+/low), germinal center (GC) B cells (B220+CD93-GL7+CD95+CD38low), marginal zone 25 (MZ) B cells (B220+CD93-GL7-CD95-IgG-CD21highCD23low/-), follicular (Fo) B cells (B220+CD93-GL7-CD95-IgG- 26 CD21intCD23high), myeloid cells (IgM-CD19-B220-CD11b+CD5-), and T cells (IgM-CD19-B220-CD11b-CD5+). 27 (E-F) Phenotypic analysis of T cells and NK cells in RAB11FIP5 -/- and fl/fl mice. (E) Gating strategy for T cell and 28 NK cell subsets. (F) Cells from lymph node, spleen and bone marrow were analyzed and the bar graphs show the 29 absolute number of total cells, erythroid cells (TER119+B220-), B cells (TER119-B220+), total NK cells (CD3- 30 NK1.1+), Thy1.2-CD62L+ NK cells, Thy1.2+CD62L+ NK cells, Thy1.2+CD62L- NK cells, Thy1.2-CD62L- NK cells, 31 total T cells (CD3+), CD8+ T cells, naive CD8+ T cells (CD8+CD44-CD62L+), central memory (CM) CD8+ T cells 32 (CD8+CD44+CD62L+), effector CD8+ T cells (CD8+CD44+CD62L-), effector memory (EM) CD8+ T cells 33 (CD8+CD44-CD62L-), CD4+ T cells, non-Treg (CD4+CD127highCD25low), Tfh (CD4+CD127highCD25lowCXCR5+PD- 34 1+), Thy1.2- Tfh, Thy1.2+ Tfh, Treg (CD4+CD127lowCD25high), Tfr (CD4+CD127lowCD25highCXCR5+PD-1+), naive 35 CD4+ T cells (CD4+CD44-CD62L+), central memory (CM) CD4+ T cells (CD4+CD44+CD62L+), effector CD4+ T 36 cells (CD4+CD44+CD62L-), and effector memory (EM) CD4+ T cells (CD4+CD44-CD62L-). 37 38

3 RAB11FIP1 RAB11FIP2 RAB11FIP3

100000 10000 100000

10000 10000 1000 M M M K K K 1000 1000 P P P F F F 100 100 100

10 10 10 s s s s s s s s s s s s s s s ll ll ll ll ll ll ll ll ll ll ll ll ll ll ll e e e e e e e e e e e e c c c ce c c c c ce c c c c ce c r r r B fh f T K B fh f T K B fh f T T T 8 N T T 8 N T T 8 NK CD CD CD

RAB11FIP4 RAB11FIP5 RAB11A

10000 10000 1000000

1000 1000 100000 M M M K K K P P F FP F 100 100 10000

10 10 1000 s s s s s s s s s s s ls ls s ls ll ll ll ll ll ll ll ll ll ll ll l l ll l e e e e e e e e e e e e e c c c ce c c c c c c c c c ce c r r h r T B fh f T K B fh f T K B f f K T T 8 N T T 8 N T T 8 N D CD C CD

RAB11B RAB27A

100000 100000 RAB11FIP5fl/fl mice RAB11FIP5-/- mice 10000 M M

K 10000 K P P

F F 1000

1000 100 s s s s s s s s s s ll ll ll ll ll ll ll ll ll ll e e e e e e e e e c c c ce c c c c c c r r B fh f T K B fh f T K T T 8 N T T 8 N D 39 CD C 40 Figure S3. RAB11FIP5 deficiency has no effect on expression of other RAB11 family genes and RAB11 family 41 interacting protein (Fip) genes. FPKMs of RAB11A, RAB11B, and RAB11FIPs genes in B cells, Tfh cells, Tfr 42 cells, CD8+ T cells and NK cells are shown as dots; each dot indicates data from one animal. While the deletion of 43 exon 2 in RAB11FIP5 alleles results in the absence of a complete and functional RAB11FIP5 mRNA, 44 incomplete/partial RAB11FIP5 transcripts may still exist at various levels in different cell types as indicated by 45 decreased FPKM in the RAB11FIP5-/- mice. 46

4 GC B cells A 90.5 5.47 86.4 95.7 SSC-A FSC-W SSC-W Live/dead

91.9 AF647-gp120

FSC-A SSC-A FSC-A FSC-A BV421-gp120 CD38low IgG+ GC B cells 17.0 0.079 0.19 2.24 2.67 5.03 0.5 B220 CD93 CD138 CD95

16.9 CD38 4.55 96.2 91.9 19.3 AF647-gp120 52.3 IgM CD19 B220 GL7 IgG1/2/3 BV421-gp120

IgG+ memory B cells 2.45 2.41

IgG1/2/3 96.1 AF647-gp120

IgM BV421-gp120

B

Bone Marrow

C

Lymph Node

D

Spleen

E

Payer’s Patch

47 48 Figure S4. Analysis of antigen-specific B cell responses in immunized mice. 49 (A) Gating strategy to identify HIV-Env binding cells (double-positive for binding to CH505 gp120 tetramers 50 labeled with two different fluorochromes) in germinal center (GC) B cells, CD38low IgG+ GC B cells, and IgG+ 51 memory B cells. 52 (B-E) HIV-1 CH505 TF SOSIP-specific GC B cells, CD38low IgG+ GC B cells, and memory B cells were compared 53 between RAB11FIP5-/- mice (n=21) and control RAB11FIP5fl/fl mice (n=20) in different tissues. Cells from bone 54 marrow (B), spleen (C), lymph node (D), and Peyer’s patch (E) were analyzed and absolute numbers of SOSIP- 55 binding cells are shown in each tissue. Bars represent mean ± SD and each dot indicates one animal. Combined data 56 from three independent immunization studies are shown. The statistical significance of differences between groups 57 was determined by Mann-Whitney U test, and no significant differences were found.

5