Monogenic mutations differentially affect the quantity and quality of T follicular helper cells in patients with human primary immunodeficiencies

quality of T follicular helper cells in patients with human primary immunodeficiencies

Cindy S. Ma, PhD,

^a,b

Natalie Wong, BSc Hons,

^a

Geetha Rao, MSc,

^a

Danielle T. Avery, BApplSci,

^a

James Torpy, MPhil,

^a

Thomas Hambridge, BSc Hons,

^a

Jacinta Bustamante, MD, PhD,

^c,d

Satoshi Okada, MD, PhD,

^e

Jennifer L. Stoddard, BS,

^f

Elissa K. Deenick, PhD,

^a,b

Simon J. Pelham, MSc,

^a,b

Kathryn Payne, BSc Hons,

^a

St ephanie Boisson-Dupuis, PhD,

^c,g

Anne Puel, PhD,

^c

Masao Kobayashi, MD, PhD,

^e

Peter D. Arkwright, FRCPCH, DPhil,

^h

Sara Sebnem Kilic, MD,

ⁱ

Jamila El Baghdadi, PhD,

^j

Shigeaki Nonoyama, MD, PhD,

^k

Yoshiyuki Minegishi, MD, PhD,

^l

Seyed Alireza Mahdaviani, MD,

^m

Davood Mansouri, MD,

^m

Aziz Bousfiha, MD,

ⁿ

Annaliesse K. Blincoe, BHB, MBChB,

^o

Martyn A. French, MB ChB, MD, FRACP, FRCPath, FRCP,

^p,q

Peter Hsu, FRACP, PhD,

^r

Dianne E. Campbell, FRACP, PhD,

^r

Michael O. Stormon, MBBS, FRACP,

^r

Melanie Wong, MBBS, PhD, FRACP, FRCPA,

^r

Stephen Adelstein, MBBCh, PhD, FRACP, FRCPA,

^s

Joanne M. Smart, MBBS, FRACP,

^t

David A. Fulcher, MBBS, PhD, FRACP, FRCPA,

^u

Matthew C. Cook, MBBS, PhD, FRACP, FRCPA,

^v,w,x

Tri Giang Phan, MBBS, PhD, FRACP, FRCPA,

^a,b

Polina Stepensky, MD,

^y

Kaan Boztug, MD,

^z,aa

Aydan Kansu, MD,

^bb

Aydan _ Ikincio gullari, MD,

^cc

Ulrich Baumann, MD,

^dd

Rita Beier, MD,

^ee

Tony Roscioli, FRACP, PhD,

^b,ff,gg

John B. Ziegler, MD, FRACP,

^hh

Paul Gray, FRACP,

^hh

Capucine Picard, MD, PhD,

^c,d

Bodo Grimbacher, MD,

ⁱⁱ

Klaus Warnatz, MD, PhD,

ⁱⁱ

Steven M. Holland, MD,

^jj

Jean-Laurent Casanova, MD, PhD,

^c,g,kk,ll

Gulbu Uzel, MD,

^jj

and Stuart G. Tangye, PhD

^a,b

Darlinghurst, Melbourne, Perth, Westmead, Acton, Canberra, and Randwick, Australia, Paris, France, Hiroshima, Saitama, and Tokushima, Japan, Bethesda, Md, New York, NY, Manchester, United Kingdom, Bursa and Ankara, Turkey, Rabat and Casablanca, Morocco, Tehran, Iran, Auckland, New Zealand, Jerusalem, Israel, Vienna, Austria, and Hannover, Essen, and Freiburg, Germany

From^athe Immunology Research Program, Garvan Institute of Medical Research, Darling- hurst;^bSt Vincent’s Clinical School, UNSW Australia, Melbourne;^cthe Laboratory of Human Genetics of Infectious Diseases, Necker Branch, INSERM U1163, Institut IMA- GINE, Necker Medical School, University Paris Descartes, Paris;^dthe Study Center for Primary Immunodeficiencies, Assistance Publique-H^opitaux de Paris (AP-HP), Necker Hospital for Sick Children, Paris;^ethe Department of Pediatrics, Hiroshima University Graduate School of Biomedical & Health Sciences, Hiroshima;^fthe Clinical Center, Na- tional Institutes of Health, Bethesda;^gSt Giles Laboratory of Human Genetics of Infec- tious Diseases, Rockefeller Branch, Rockefeller University, New York;^hthe University of Manchester, Royal Manchester Children’s Hospital, Manchester;ⁱthe Department of Pe- diatric Immunology, Uludag University Medical Faculty, G€or€ukle, Bursa;^jthe Genetics Unit, Military Hospital Mohamed V, Hay Riad, Rabat;^kthe Department of Pediatrics, Na- tional Defense Medical College, Tokorozawa, Saitama;^lthe Division of Molecular Med- icine, Institute for Genome Research, University of Tokushima;^mPediatric Respiratory Diseases Research Center, National Research Institute of Tuberculosis and Lung Dis- eases (NRITLD), Shahid Beheshti University of Medical Sciences, Tehran;ⁿthe Clinical Immunology Unit, Pediatric Infectious Diseases Department, Averroes University Hos- pital, King Hasan II University, Casablanca;^oStarship Children’s Hospital, Auckland;

pthe Department of Clinical Immunology, Royal Perth Hospital;^qthe School of Pathology and Laboratory Medicine, University of Western Australia, Perth;^rChildren’s Hospital at Westmead;^sClinical Immunology, Royal Prince Alfred Hospital, Sydney;^tthe Depart- ment of Allergy and Immunology, Royal Children’s Hospital Melbourne;^uthe Depart- ment of Immunology, Westmead Hospital, University of Sydney;^vAustralian National University Medical School and^wthe John Curtin School of Medical Research, Australian National University, Acton;^xthe Department of Immunology, Canberra Hospital, Acton;

yPediatric Hematology-Oncology and Bone Marrow Transplantation Hadassah, Hebrew University Medical Center, Jerusalem;^zCeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna;^aathe Department of Paediatrics and Adolescent Medicine, Medical University of Vienna;^bbthe Department of Pediatric Gastroenterology and^ccthe Department of Pediatric Immunology and Allergy, Ankara University Medical School;^ddPaediatric Pulmonology, Allergy and Neonatology, Hano- ver Medical School;^eePediatric Haematology and Oncology, University Hospital Essen;

ffthe Kinghorn Centre for Clinical Genomics, Victoria St Darlinghurst;^ggthe Department of Medical Genetics, Sydney Children’s Hospital, Randwick;^hhSydney Children’s Hos- pital, Randwick, and School of Women’s and Children’s Health, University of New South Wales, Sydney;ⁱⁱthe Center for Chronic Immunodeficiency, University Medical Center Freiburg, University of Freiburg;^jjthe Laboratory of Clinical Infectious Diseases, Na- tional Institute of Allergy and Infectious Diseases, National Institutes of Health, Be- thesda;^kkthe Pediatric Hematology and Immunology Unit, Necker Hospital for Sick Children, AP-HP, Paris; and^llthe Howard Hughes Medical Institute, New York.

Supported by project and program grants from the National Health and Medical Research Council (NHMRC) of Australia (to E.K.D., S.G.T., C.S.M., D.A.F., and M.C.C.;

596813, 1016953, 1066694, and 1027400), the German Federal Ministry of Education and Research (BMBF 01EO1303 to B.G. and K.W.), and Rockefeller University Center for 541 Clinical and Translational science (5UL1RR024143, to J.L.C.). C.S.M.

is a recipient of a Career Development Fellowship (1008820), and S.G.T. a Principal Research Fellowship (1042925) from the NHMRC of Australia.

Disclosure of potential conflict of interest: E. K. Deenick has received research support from the National Health and Medical Research Council, Australia. S. J. Pelham has received research support from University of New South Wales (Australian Postgraduate Award [PhD Scholarship]). D. E. Campbell is employed by NSW Health, has received research support from NHMRC, the Ilhan Food Allergy Foundation, and the University of Sydney; has received payment for development of educational presentations from the Australian Medical Council; and has received travel support from ALLSA, NSW Health, and ASCIA. M. C. Cook has received research support from the National Health and Medical Research Council, Australia. K. Boztug has received research support from the Austrian Science Fund (FWF; grant no. P24999). B. Grimbacher has received research support from BMBF, the European Union, Helmholtz, DFG, and DLR; is employed by UCL and UKL-FR; and has received lecture fees from CSL, Baxter, and Biotest. K.

Warnatz has received consultancy fees from Biotest, CSL Behring, and Baxter; has received research support from Baxter; and has received lecture fees from Baxter, CSL Behring, Pfizer, Biotest, Novartis Pharma, Roche, Octapharma, Glaxo Smith Kline, Stallergenes AG, UCB Pharma, and the American Academy of Allergy, Asthma &

Immunology. J.-L. Casanova has received research support from the National Institutes of Health (5UL1RR024143); has received consultancy fees from Biogen Idec, Merck, Sanofi-Aventis, Regeneron, and Bioaster; and has received research support from Merck, Biogen Idec, and Pfizer. S. G. Tangye has received research support from the National Health and Medical Research Council of Australia and has received travel support from ESID. The rest of the authors declare that they have no relevant conflicts of interest.

Received for publication February 20, 2015; revised May 4, 2015; accepted for publica- tion May 13, 2015.

Corresponding author: Stuart G. Tangye, PhD, or Cindy S. Ma, PhD, Immunology Divi- sion, Garvan Institute of Medical Research, 384 Victoria St, Darlinghurst, NSW 2010, Australia. E-mail:s.tangye@garvan.org.au. Or:c.ma@garvan.org.au.

0091-6749/$36.00

Ó2015 American Academy of Allergy, Asthma & Immunology. Published by Elsevier Ltd. All rights reserved

http://dx.doi.org/10.1016/j.jaci.2015.05.036

1

Background: Follicular helper T (T

FH

) cells underpin T cell–

dependent humoral immunity and the success of most vaccines.

T

FH

cells also contribute to human immune disorders, such as autoimmunity, immunodeficiency, and malignancy.

Understanding the molecular requirements for the generation and function of T

FH

cells will provide strategies for targeting these cells to modulate their behavior in the setting of these immunologic abnormalities.

Objective: We sought to determine the signaling pathways and cellular interactions required for the development and function of T

FH

cells in human subjects.

Methods: Human primary immunodeficiencies (PIDs) resulting from monogenic mutations provide a unique opportunity to assess the requirement for particular molecules in regulating human lymphocyte function. Circulating follicular helper T (cT

_FH

) cell subsets, memory B cells, and serum immunoglobulin levels were quantified and functionally assessed in healthy control subjects, as well as in patients with PIDs resulting from mutations in STAT3, STAT1, TYK2, IL21, IL21R, IL10R, IFNGR1/2 , IL12RB1 , CD40LG , NEMO , ICOS , or BTK . Results: Loss-of-function (LOF) mutations in STAT3 , IL10R , CD40LG , NEMO , ICOS , or BTK reduced cT

FH

cell

frequencies. STAT3 and IL21/R LOF and STAT1 gain-of- function mutations skewed cT

FH

cell differentiation toward a phenotype characterized by overexpression of IFN-g and programmed death 1. IFN-g inhibited cT

FH

cell function in vitro and in vivo , as corroborated by

hypergammaglobulinemia in patients with IFNGR1/2, STAT1, and IL12RB1 LOF mutations.

Conclusion: Specific mutations affect the quantity and quality of cT

FH

cells, highlighting the need to assess T

FH

cells in patients by using multiple criteria, including phenotype and function.

Furthermore, IFN-g functions in vivo to restrain T

FH

cell–

induced B-cell differentiation. These findings shed new light on T

FH

cell biology and the integrated signaling pathways required for their generation, maintenance, and effector function and explain the compromised humoral immunity seen in patients with some PIDs. (J Allergy Clin Immunol 2015;nnn:nnn-nnn.) Key words: Follicular helper T cells, humoral immunity, primary immunodeficiencies, cytokine signaling

Naive CD4

¹

T cells differentiate into distinct populations of effector cells with specialized functions. Such fine tuning en- sures the generation of appropriate immune responses that effi- ciently clear pathogens and generate long-term protective immunity after infection or vaccination.

¹

The CD4

¹

T cells responsible for mediating the differentiation of naive B cells into memory cells and plasma cells, thereby providing effective humoral immunity against T-dependent (TD) antigen, are follic- ular helper T (T

_FH

) cells.

^2-5

T

_FH

cells express increased levels of CXCR5, programmed death 1 (PD-1), B-cell lymphoma 6 (Bcl- 6), and several molecules involved in T-cell/B-cell interactions and localize to follicles of secondary lymphoid tissues.

^2-4

Dif- ferentiation of naive CD4

¹

T cells into T

FH

cells is a complex process requiring integration of signals delivered by dendritic cells, B cells, cytokines, specific signaling pathways, and tran- scription factors.

^2-5

The critical role of T

_FH

cells in eliciting long-lived humoral immunity is evidenced by impaired genera- tion of germinal centers (GCs), memory B cells, and antibodies to TD antigen in mice and human subjects who lack genes that

promote T

_FH

cell formation.

^2-6

Antibody-mediated autoimmune conditions can also be caused by dysregulated T

_FH

cell func- tion.

^7-9

Thus delineating molecular requirements underlying T

FH

cell generation and function are important in understanding how these cells operate and in identifying pathways that could be targeted in the settings of vaccination, immunodeficiency, or autoimmunity.

Although studies of mice and some human immune disorders have taught us much about T

_FH

cells, our understanding of hu- man T

_FH

cell biology remains incomplete, largely because of limited access to lymphoid tissues, where T

FH

cells are located.

However, progress has been made by studying circulating CD4

¹

CXCR5

¹

T cells as correlates of tissue T

FH

cells. Subsets of circulating follicular helper T (cT

_FH

) cells have been reported, with CCR6

¹

, CCR6

¹

PD-1

¹

, or CCR7

^lo

PD-1

^hi

subsets being superior to other subsets in providing B-cell help.

^10,11

These subsets correlated with antibody responses to influenza virus after vaccination in young adults, but not older adults,

¹²

and are increased in patients with autoimmune dis- eases

^8,10,13-16

and decreased in patients with HIV infection.

¹¹

Similarly, CD4

¹

CXCR5

¹

PD-1

¹

CXCR3

²

T cells were identi- fied as the circulating counterpart of lymphoid T

FH

cells, and their frequencies positively correlated with neutralizing anti- bodies in patients with HIV infection.

¹⁷

Although these studies generally confirmed that PD-1

¹

CXCR3

²

/CCR6

¹

cT

_FH

cells are a reliable correlate of T

_FH

cells in human lymphoid tissue, the identity of the circulating B-helper human CD4

¹

T cells remains contentious because other studies demonstrated that CXCR5

¹

CXCR3

¹

or even CXCR5

²

CD4

¹

T cells exhibit detectable B-helper function

11,15,18,19

and correlate with influenza vaccine responsiveness.

^18,20

To assess the molecular requirements for the generation and function of human cT

FH

cells, we investigated more than 110 subjects with 14 different monogenic mutations that underlie primary immunodeficiencies (PIDs). Our findings identify mutations that have distinct quantitative effects, qualitative effects, or both on human cT

_FH

cells, providing an explanation for humoral immune defects in some PIDs, as well as insights into mechanisms regulating human T

_FH

cell differentiation and function.

Abbreviations used Bcl-6: B-cell lymphoma 6

cTFH: Circulating follicular helper T DN: Double-negative

DP: Double-positive GC: Germinal center GOF: Gain of function IL-10R: IL-10 receptor IL-12R: IL-12 receptor IL-21R: IL-21 receptor LOF: Loss of function PD-1: Programmed death 1

PID: Primary immunodeficiency

STAT: Signal transducer and activator of transcription TD: T-dependent

TFH: Follicular helper T TFR: Follicular regulatory T Treg: Regulatory T

J ALLERGY CLIN IMMUNOL nnn2015

2 MA ET AL

METHODS Human samples

PBMCs were isolated from healthy control subjects (Australian Red Cross) and patients with PIDs. Human spleens were obtained from cadaveric organ donors (NSW Organ Transplant Registry). All studies were approved by institutional human research ethics committees.

Antibodies and reagents

eFluor660–anti–IL-21, PerCP-Cy5.5–anti–IFN-g, fluorescein isothiocya- nate–anti-CD45RA, and biotin–PD-1 were from eBioscience (San Diego, Calif). Alexa Fluor 647–anti-CXCR5 and anti–phosphorylated signal trans- ducer and activator of transcription 1 (pSTAT1), allophycocyanin–anti-CD10, allophycocyanin-Cy7–anti-CD4, BV605–anti-IgG, phycoerythrin–anti- pSTAT3 and anti-CCR6, PE-Cy7–anti-CD25 and anti-CD27, PerCpCy5.5–

anti-CD127, biotin–anti-IgA, SA-PerCpCy5.5, and recombinant IFN-gwere from Becton Dickinson (Franklin Lakes, NJ). BV421–anti-CXCR3, Pacific Blue–anti-CD20, and SA-BV605 were from BioLegend (San Diego, Calif).

Lymphocyte phenotyping and isolation

T cells.

PBMCs were incubated with mAbs to CD4, CD45RA, CD127, CD25, CXCR5, CXCR3, CCR6, and PD-1, and proportions of regulatory T (Treg) cells (CD4¹CD127^loCD25^hi), total memory cells (CD4¹CD45RA²), and cTFHcells (CD4¹CD45RA²CXCR5¹), as well as subsets of non-cTFH

memory and cTFHcells defined according to CXCR3 and CCR6 expression, were determined.^10,19 To isolate these subsets, Treg cells were excluded, and the remaining population was sorted into naive (CD45RA¹ CXCR5²CXCR3²CCR6²), non-TFH memory (CD45RA²CXCR5²), and cTFHcells. Subsets of non-cTFHand cTFHcells were identified according to dif- ferential CXCR3 and CCR6 expression.¹⁰All populations were sorted on a FACSAria (Becton Dickinson) to greater than 98% purity.

B cells.

PBMCs were incubated with mAbs to CD20, CD27, CD10, IgG, and IgA, and the frequency of total memory (CD20¹CD27¹CD10²) and switched memory B cells was determined.^21,22

Expression of phospho-STATs

EBV-transformed lymphoblastoid cell lines established from healthy donors, patients withIFNGR2LOFmutations, or patients withSTAT1LOFmu- tations were stimulated with IFN-gor IL-21 for 30 minutes. Cells were fixed, permeabilized, and stained for anti-pSTAT1 and anti-pSTAT3.²¹

Analysis of CD4

¹

T-cell function in vitro

Isolated CD4¹T-cell populations were cultured with T-cell activation and expansion beads (anti-CD2/CD3/CD28; Miltenyi Biotech, Bergisch Glad- bach, Germany) in 96-well round-bottom plates. After 5 days, supernatants were harvested, and production of IL-4, IL-5, IL-10, IL-13, IL-17A, IL-17F, IFN-g, and TNF-awas determined by using cytometric bead arrays (Becton Dickinson); secretion of IL-22 (eBioscience) and CXCL13 (R&D Systems, Minneapolis, Minn) was determined by using ELISA. For cytokine expres- sion, activated CD4¹T cells were restimulated with phorbol 12-myristate 13-acetate (100 ng/mL)/ionomycin (750 ng/mL) for 6 hours, with Brefeldin A (10mg/mL) added after 2 hours. Cells were then fixed, and expression of intracellular cytokines was detected.^19,22,23For gene expression, RNA was ex- tracted and transcribed into cDNA. Expression ofTBX21,GATA3,RORC, and BCL6 was determined by using quantitative PCR and standardized to GAPDH.^19,24

T-cell/B-cell coculture assays and immunoglobulin determination

CD4¹T-cell subsets were treated with mitomycin C (100mg/mL; Sigma, St Louis, Mo) and then cocultured at a 1:1 ratio (50310³/200mL per well) with allogeneic total splenic B cells.^19,22,24In some experiments exogenous IFN-gwas added to the cultures. After 7 days, immunoglobulin secretion

was determined by means of ELISA.^19,21Serum IgG and IgM levels were determined by using nephelometry.

Statistical analysis

Significant differences were determined by using 1-way ANOVA (Prism;

GraphPad Software, La Jolla, Calif).

RESULTS

Cytokine and transcription factor expression by human naive and memory CD4

¹

T cells

To determine the requirements for generating human T

_FH

cells in vivo, we first defined parameters that identify different CD4

¹

T-cell subsets in peripheral blood. Memory cells were the pre- dominant producers of all cytokines examined, producing approximately 5- to 100-fold higher levels of T

H

1 (IFN-g), T

_H

2 (IL-4, IL-5, and IL-13), and T

_H

17 (IL-17A, IL-17F, and IL-22) cytokines, as well as B-cell helper/T

FH

cytokines (IL-10 and IL- 21), than naive cells (Fig 1, A-D). Memory CD4

¹

T cells also ex- pressed significantly higher levels of TBX21 (T-bet), GATA3, and RORC (retinoic acid–related orphan receptor gt) than naive cells (Fig 1, E-G). There was no significant difference in BCL6 expres- sion between naive and memory cells (Fig 1, H), which is consis- tent with other studies reporting Bcl-6 levels are similar in circulating human CD4

¹

T-cell subsets.

8,10,12,15,17,25

Delineation of memory CD4

¹

T cells into defined populations of T

H

1, T

H

2, T

H

17, and cT

FH

cells and subsets

Human memory T

_H

1, T

_H

2, T

_H

17, and cT

_FH

cells can be defined according to differential expression of CXCR3, CCR6, and CXCR5,

^26,27

with T

_H

1 cells being CD45RA

²

CXCR5

²

CXCR3

¹

CCR6

²

, T

H

17 cells being CD45RA

²

CXCR5

²

CXCR3

²

CCR6

¹

, T

_H

2 cells being CD45RA

²

CXCR5

²

CXCR3

²

CCR6

²

, and cT

_FH

cells being CD45RA

²

CXCR5

¹

(Fig 1, I-K). In contrast, CD45RA

¹

naive cells lack these chemokine receptors (Fig 1, I and J). We extended these findings by demonstrating T

H

1 cells were enriched for IFN-g secretion (Fig 1, L); T

_H

2 cells produced the greatest amounts of IL-4, IL-5, and IL-13 (Fig 1, M); and T

_H

17 cells secrete the most IL-17A, IL-17F, and IL-22 (Fig 1, N). Impor- tantly, the highest proportion of IL-21–expressing cells was de- tected in the cT

FH

subset, which also contained a greater proportion of IL-10–expressing cells than the T

_H

1 and T

_H

2 subsets (Fig 1, O). Indeed, of the cytokines examined, IL-10 and IL-21 were the only ones produced by cT

_FH

cells at levels significantly greater than those seen in naive cells (P < .005; Fig 1, L-O), which is consistent with the B-cell helper function of these cytokines.

²⁸

The proportions of cT

_FH

cells expressing IL-10 and IL-21 were also significantly greater than naive cells (P < .005), T

H

1 cells (P < .05), T

_H

2 cells (P < .05), and T

_H

1/T

_H

17 cells (P < .05) but not T

H

17 cells. The memory CD4

¹

T-cell population coexpressing CCR6 and CXCR3 (termed T

_H

1/T

_H

17 cells) produced both T

_H

1 and T

H

17 but only low T

H

2 cytokine levels (Fig 1, L-O). Thus these cells represent proinflammatory cells, which is consistent with their detection in inflamed tissues.

²⁹

TBX21, GATA3, and RORC expression also significantly correlated with their respec- tive production of T

_H

1, T

_H

2, and T

_H

17 cytokines (Fig 1, P-S).

Having successfully characterized T

H

1, T

H

2, T

H

17, and cT

FH

cells in peripheral blood, we next investigated cT

_FH

cells using

a similar approach: dividing them into subpopulations according

J ALLERGY CLIN IMMUNOL nnn2015

4 MA ET AL

to CCR6 and CXCR3 (Fig 2, A and B)

¹⁰

and assessing cytokine production and B-cell helper function. The different cT

FH

cell subsets produced lower levels of cytokines than memory

(CD45RA

²

CXCR5

²

) cells; however, IL-4 and IL-13 were en- riched in the CXCR3

²

CCR6

²

(double-negative [DN]) subset, IL-17A/F and IL-22 were enriched in the CCR6

¹

CXCR3

²

FIG 1. Identification of effector subsets within populations of human memory CD4¹T cells.A-H,Naive and memory CD4¹T cells were sorted from healthy control subjects and stimulated with T-cell activation and expansion (anti-CD2/CD3/CD28) beads. Secretion/expression of the indicated cytokines (Fig 1, A-D;

mean6SEM; n525-27) or transcription factors (Fig 1,E-H; mean6SEM; n512-19) were determined after 5 days. I, Resolving blood naive (CD45RA¹CXCR5²), non-TFH memory (CD45RA²CXCR5²), and cTFH

(CD45RA²CXCR5¹) cells from healthy control subjects.JandK, CXCR3 and CCR6 expression on naive and non-TFHmemory cells. Fig 1,K, depicts the percentage of TH1 (CXCR3¹CCR6²), TH2 (CXCR3²CCR6²), TH17 (CXCR3²CCR6¹), and TH1/TH17 (CXCR3¹CCR6¹) subsets among the non-TFHmemory population (n555-58).L-S,Secretion/expression of the indicated cytokines (Fig 1,L-O; mean6SEM; n510-15) or tran- scription factors (Fig 1,P-S; mean6SEM; n57-10) by naive, TH1, TH2, TH17, TH1/TH17, and cTFHcell subsets after 5 days of culture with T-cell activation and expansion beads. Significant differences (1-way ANOVA) between naive and memory CD4¹T cells or subsets are indicated. *P< .05, **P< .01, ***P< .005, and

****P< .001.ns, Not significant.

=

FIG 2.Delineating subsets of human circulating TFH cells. A and B, cTFH cells were defined as CD45RA²CXCR5¹CD4¹T cells. CXCR3¹CCR6², CXCR3²CCR6², CXCR3²CCR6¹, and CXCR3¹CCR6¹sub- sets were then detected (n 5 55-58). C-G,Naive, non-TFH memory, total cTFH and CXCR3¹CCR6², CXCR3²CCR6¹, CXCR3²CCR6²(DN), and CXCR3¹CCR6¹(DP) cTFHcell subsets were sorted from peripheral blood and stimulated with T-cell activation and expansion beads. After 5 days, secretion or expression of the indicated cytokines were determined (n55-6).H-J,purified subsets of blood CD4¹T cells were cocul- tured with allogeneic B cells and T-cell activation and expansion beads. IgM, IgG, and IgA secretion was determined after 7 days.

(hereafter referred to as CCR6

¹

) subset, and IFN-g was enriched in the CXCR3

¹

CCR6

²

(‘‘CXCR3

¹

’’) and CXCR3

¹

CCR6

¹

(double-positive [DP]) subsets (Fig 2, C-F). In contrast, levels of IL-10, IL-21 (Fig 2, F), and CXCL13 (Fig 2, G), which are pro- duced by T

_FH

cells,

^25,30

were comparable in all cT

_FH

cell subsets.

Total cT

FH

cells induced greater B-cell differentiation than naive and memory cells (Fig 2, H), as well as T

_H

1, T

_H

2, T

_H

17, and T

_H

1/

T

H

17 memory cell subsets (Fig 2, I). Among cT

FH

cell subsets, the CCR6

¹

population consistently induced the most immunoglob- ulin secretion by cocultured B cells over the CXCR3

¹

, DN, and DP subsets (Fig 2, J). Thus, consistent with recent studies, blood CD4

¹

CXCR5

¹

T cells have T

_FH

function, with the CCR6

¹

sub- set being most effective.

^10,11,17

Mutations causing PIDs affect CD4

¹

T-cell function and B-cell memory formation

These data established the ability to identify effector functions of human CD4

¹

T-cell subsets, thereby providing a framework to

determine the consequences of gene mutations on CD4

¹

T-cell differentiation in vivo. We examined CD4

¹

T-cell subsets and function in patients with PIDs with monoallelic or biallelic muta- tions in surface receptors or cytokine signaling pathways that might affect humoral immunity in the context of T

_FH

cell forma- tion and long-lived protective antibody responses.

⁴

Loss-of-function (LOF) mutations in CD40LG, NEMO, IL12RB1, or TYK2 compromised IFN-g production by memory CD4

¹

T cells (Fig 3, A, and see Table E1 in this articles Online Repository at www.jacionline.org), which is consistent with known roles in eliciting T

_H

1 responses.

³¹

Production of T

_H

2 cy- tokines was enhanced by IL12RB1, IFNGR1/2, or TYK2 LOF mu- tations, confirming that T

H

1 cells suppress T

H

2 cells,

³²

as well as by LOF mutations in IL10R, IL21/R, or STAT3 and gain-of- function (GOF) mutations in STAT1 (Fig 3, B, and see Table E1). This revealed novel roles for IL-10, IL-21, and STAT3 as reg- ulators of human T

H

2 immunity, which differs from studies in mice.

^33,34

B cells also clearly regulate CD4

¹

T-cell differentia- tion, as revealed by reduced IFN-g and increased T

H

2 cytokine

FIG 3.Defects in cytokine production because of monogenic mutations causing different PIDs.A-D,mem- ory CD4¹T cells were sorted from healthy control subjects or patients with the indicated gene mutations and then stimulated with T-cell activation and expansion beads. Production of IFN-g(Fig 3,A); IL-4, IL-5, and IL-13 (TH2 cytokines; Fig 3,B); IL-17A, IL-17F, and IL-22 (TH17 cytokines; Fig 3,C); and IL-10 and IL-21 (TFHcytokines; Fig 3,D) was determined after 5 days (means6SEMs). Control subjects, n526-35. Patients with mutations in the following genes:STAT3LOF, n56;STAT1GOF, n57-9;STAT1LOF, n55-7;IL21/R, n55;

IL10R, n52;CD40LG, n54;NEMO, n55-6;BTK, n56;ICOS, n54;IL12R, n55;IFNGR1/2, n56; and TYK2, n53. SeeTable E1for more detailed results.EandF, proportions of total (Fig 3,E) and class- switched (Fig 3,F) memory B cells in healthy control subjects (n537) and patients with gene mutations (STAT3LOF, n511-12;STAT1GOF, n516;STAT1LOF, n57;IL21/R, n55;IL10R, n53-4;CD40LG, n54;

NEMO, n57-10;ICOS, n53;IL12R, n58;IFNGR1/2, n57-8; andTYK2, n54) were determined. Significant differences (1-way ANOVA) between control subjects and patients are indicated. **P< .01, ***P< .005, and

****P< .001.

J ALLERGY CLIN IMMUNOL nnn2015

6 MA ET AL

levels in B cell–deficient subjects with BTK mutations (Fig 3, A and B, and see Table E1). STAT3, NEMO, ICOS, IL12RB1, or TYK2 LOF mutations strongly diminished T

H

17 cytokine levels (Fig 3, C) and RORC expression (not shown). STAT1 GOF and IL21/R LOF mutations also reduced T

H

17 cytokine levels, although to a lesser extent than these other mutations (Fig 3, C, and see Table E1). These findings are consistent with susceptibil- ity of some subjects with these mutations to Candida species infec- tion (STAT3

_LOF

, STAT1

_GOF

, IL21R, IL12RB1, and NEMO)

^31,35

and the requirement for ICOS in human T

H

17 cell generation.

³⁶

In contrast, STAT1

_LOF

mutations had no or only a partial effect on IL-17A/F or IL-22 levels (Fig 3, C, and see Table E1), mirroring intact immunity against Candida species in these patients,

^31,35

whereas IL10R mutations resulted in excessive production of IL- 17A/F and IL-22 (see Table E1). IL-10 production was compro- mised by STAT3, TYK2, IL10R, and IL21R LOF and STAT1 GOF mutations but unaffected by other mutations, whereas IL-21 pro- duction was reduced by STAT1 GOF, and IL21R, IFNGR1/2, and NEMO LOF mutations (Fig 3, D, and see Table E1).

To establish regulators of human lymphocyte differentiation in the context of TD B-cell function, we assessed patients with PIDs for memory B cells. There were marked reductions in memory B- cell numbers in patients with STAT3, IL21/R, IL10R, ICOS, CD40LG, and NEMO LOF and STAT1 GOF mutations (Fig 3, E). Thus it is likely that signaling through the IL-10 receptor (IL-10R) and IL-21 receptor (IL-21R) underlie the paucity of memory B cells in STAT3-deficient patients. Although memory B-cell numbers were reduced in patients with STAT3 LOF, STAT1 GOF, IL10R, and NEMO mutations, the residual memory cells still underwent class switching. In contrast, a lack of CD40 ligand, ICOS, or IL-21/IL-21R signaling impaired isotype switching. The frequencies of memory B cells were unaffected by LOF mutations in STAT1, TYK2, IL12RB1, or IFNGR1/2; how- ever, there was a trend for more switched memory B cells in most of these patients (Fig 3, F). Together, these data demonstrated the validity of using patients with PIDs as models for defects in B-cell and CD4

¹

T-cell differentiation and cytokine production.

Specific gene mutations compromise the generation of human circulating T

FH

cells

We next used these patients to examine the relationship between PID-associated gene mutations and phenotypically defined subsets of CD4

¹

T cells. T

H

1 cell numbers were signifi- cantly increased in patients with STAT3

_LOF

and STAT1

_GOF

muta- tions, whereas T

_H

17 cells were significantly reduced in both these patients and NEMO-, ICOS-, or BTK-deficient patients (Table I), which is consistent with poor production of IL-17A/IL-17F and IL-22 by memory cells from most of these patient groups (Fig 3, C, and see Table E1). The reduction in CCR6

¹

CD4

¹

T-cell numbers because of STAT3 mutations likely reflects the inability of mutant STAT3 to induce CCR6, which has been shown to be a direct target of STAT3 in mice.

³⁷

T

H

2 and T

H

1/T

H

17 phenotype cells were unaffected by most of the mutations examined (Table I). When cT

_FH

cells were measured, significant reductions were observed in patients with mutations in STAT3, CD40LG, BTK, and ICOS, which is consistent with previous studies,

^19,38,39

as well as in patients with IL10R or NEMO mutations (Fig 4, A and B). In our cohort of CD40LG-deficient patients, 3 had clini- cally milder disease than those with classic hyper-IgM syndrome (ie, detectable levels of IgG/IgA, less severe infections, and later age of diagnosis). These patients had normal cT

_FH

cell fre- quencies, thus correlating with disease severity (Fig 4, A and B). Although there were trends for fewer cT

_FH

cells when signaling through IL-21R, IL-12 receptor (IL-12R), or TYK2 was compromised, these differences were not significant (Fig 4, A and B). None of the other mutations examined affected T

FH

for- mation, as determined by quantifying circulating CD4

¹

CXCR5

¹

T cells. In addition to T

FH

cells that promote B-cell differentia- tion, there is also a population of cells termed follicular regulatory T (T

_FR

) cells that restrain T

_FH

cell function to regulate production of specific antibodies.

⁴⁰

These cells can be defined as CXCR5

¹

cells within the population of Treg cells. Although specific gene mutations affected the formation of cT

FH

cells (Fig 4, A and B), the proportions of CXCR5

¹

Treg cells (putative T

_FR

cells) in all groups of patients with PID were similar to those in healthy

TABLE I.Treg cells and CD4¹CXCR5²memory subsets in patients with PIDs

Genetic diseases Treg cells

CXCR5¹Treg cells (% Treg cells)

Total memory

CD4¹CXCR5²memory cells TH1

(CXCR3¹CCR6²)

TH2 (CXCR3²CCR6²)

TH17 (CXCR3²CCR6¹)

TH1/TH17 (CXCR3¹CCR6¹)

Control subjects (n547-71) 6.160.3 8.260.9 52.261.8 15.561.0 47.361.8 23.061.1 14.161.1 STAT3LOFmutations (n512-22) 7.460.7 8.761.7 29.663.7à 34.464.3à 46.365.1 5.360.7à 10.061.5 STAT1GOFmutations (n516-23) 5.660.7 12.061.6 41.063.7 27.463.4* 56.963.7 6.961.05à 9.561.6 STAT1LOFmutations (n55-6) 7.561.1 15.965.5 38.066.1 22.466.2 37.967.6 20.564.4 19.265.0 IL-21/Rmutations (n55) 5.861.1 6.662.4 32.267.1 25.1611.2 57.6610.5 12.063.3 5.361.1 IL10Rmutations (n54) 6.461.2 9.865.3 31.569.4 20.1617.0 52.8612.3 23.0610.9 4.161.8 CD40LGmutations (n58) 3.660.8 6.061.4 25.963.8  6.962.0 70.766.0* 17.962.8 4.662.0 NEMOmutations (n58-12) 9.061.7 8.361.7 40.267.6 2.661.6 85.264.0à 10.762.3  1.460.7  BTKmutations (n54-11) 4.960.7 5.561.1 30.065.2  12.363.6 71.466.5 11.261.8* 5.261.6 ICOSmutations (n56) 8.860.9 10.162.8 53.167.1 30.566.3 50.164.5 10.461.4* 8.961.7 IL12RB1mutations (n53) 4.161.8 10.862.2 16.463.0* 12.862.1 67.867.6 16.368.4 3.061.2 IFNGR1/2mutations (n54-7) 15.865.0à 10.561.5 49.167.0 15.763.6 55.965.8 16.862.6 11.562.9 TYK2mutations (n54) 5.460.6 3.361.2 56.2610.5 18.365.6 43.465.3 18.864.6 19.565.1

PBMCs from healthy control subjects or patients with specific gene mutations were labeled with mAbs against CD4, CD127, CD25, CD45RA, CXCR5, CXCR3, and CCR6.

Proportions of Treg cells among all CD4¹T cells, proportions of Treg cells expressing CXCR5 (putative TFRcells), proportions of memory cells within the non–Treg cell population, and proportions of CD4¹CD45RA²CXCR5²memory cells with a TH1 (CXCR3¹CCR6²), TH2 (CXCR3²CCR6²), TH17 (CXCR3²CCR6¹), or TH1/TH17 (CXCR3¹CCR6¹) phenotype were then determined by means of flow cytometric analysis. Values represent means6SEMs for the indicated number of healthy control subjects or patients.

Significant differences (1-way ANOVA) between healthy control subjects and patients are indicated as follows: *P< .05, P< .01, andàP< .001.

FIG 4.Effect of monogenic mutations on generation of human cTFHcells.AandB, cTFHcells were identified among the population of non-Treg cells as CD45RA²CXCR5¹CD4¹T cells. cTFHcell frequencies in healthy donors (n572) and patients with mutations inSTAT3(n524),STAT1(GOF [n522] and LOF [n57]),IL21R/

IL21(n55),IL10R(n54),CD40LG(n510),NEMO(n511),BTK(n511),ICOS(n56),IL12RB1(n58), IFNGR1/2(n57), orTYK2(n54) were determined.CandD, proportions of CXCR3¹CCR6², CXCR3²CCR6¹, CXCR3²CCR6², or CXCR3¹CCR6¹subsets among total cTFHcells (healthy donors, n571;STAT3, n521, STAT1GOF, n520;STAT1LOF, n57;IL21R/IL21, n55;IL10R, n53;CD40LG, n58;NEMO, n57;

BTK, n55;ICOS, n56;IL12RB1, n53;IFNGR1/2, n56; andTYK2, n54).E,Production of IL-17A, IFN- g, IL-21, and IL-10 by sorted cTFHcells from healthy donors, patients withSTAT3LOFmutations, or patients withSTAT1GOFmutations (mean6SEM; n53-4). Significant differences (1-way ANOVA) between control subjects and patients are indicated. *P< .05 and ****P< .001.

J ALLERGY CLIN IMMUNOL nnn2015

8 MA ET AL

control subjects. Thus the effects of the gene mutations studied here are specific to cT

FH

cells, suggesting distinct origins of cT

_FH

and cT

_FR

cells. This is consistent with T

_FR

cells arising from Treg cells, which are largely unaffected by the mutations studied here (Table I), rather than T

_FH

cells.

⁴⁰

STAT3

LOF

or STAT1

GOF

mutations skew cT

FH

cell differentiation to a nonhelper phenotype

When we analyzed cT

_FH

cell subsets defined by CXCR3 and CCR6 expression (Fig 2, D),

¹⁰

we found significant reductions in numbers of CCR6

¹

cT

FH

cells, the most proficient B-helper cT

_FH

cell population (Fig 2, H-J),

¹⁰

in patients with STAT3

_LOF

, STAT1

GOF

, and IL21/IL21R mutations and corresponding in- creases in CXCR3

¹

cT

_FH

cell numbers in patients with STAT3

_LOF

and STAT1

GOF

mutations (Fig 4, C and D). The loss of CCR6

¹

cT

_FH

cells in these patients was even more striking when ratios of CCR6

¹

to CXCR3

¹

cT

FH

cells were calculated (control sub- jects, 1.6; patients with STAT3

LOF

mutations, 0.18 [P < .001]; pa- tients with STAT1

GOF

mutations, 0.6 [P < .05]; and patients with IL21/IL21R mutations, 0.6). NEMO mutations resulted in more DN cT

_FH

cells, whereas the DP subset was not affected by any mutations examined (Fig 4, D). Thus not only did STAT3

LOF

mu- tations compromise cT

_FH

cell generation, they also prevented for- mation of the cT

FH

cell subset most capable of inducing B-cell help. This perturbation to cT

_FH

cell differentiation was mirrored by STAT1

_GOF

and, to a lesser extent, IL21/IL21R

_LOF

mutations (Fig 4, A-D).

These data suggest that cT

_FH

cells from patients with STAT3

_LOF

and STAT1

GOF

mutations would exhibit skewed cytokine produc- tion. Indeed, cT

_FH

cells from patients with STAT3

_LOF

mutations exhibited 4-fold fewer IL-17A–expressing and 4-fold more IFN-g–expressing cells than those from healthy control subjects (Fig 4, E). Although IL-21 was expressed by comparable fre- quencies of control and STAT3

_LOF

cT

_FH

cells, IL-10 secretion by STAT3-deficient cT

_FH

cells was markedly reduced (6-fold;

Fig 4, E). STAT1

GOF

mutant cT

FH

cells exhibited similar but less extreme perturbations to cytokine production (Fig 4, E).

Thus altered phenotypes of cT

FH

cells in patients with some PIDs correlated with altered function with respect to cytokine production, demonstrating these mutations not only affect cT

FH

cell generation but also their quality, predominantly yielding a population with limited B-cell helper capacity, which is consis- tent with impaired humoral immunity in patients with STAT3

_LOF

and STAT1

_GOF

mutations.

^21,41

IFN-g restrains T

FH

-induced B-cell differentiation IFN-g can impede immunoglobulin secretion by human PBMCs.

^42-46

Taken together with our findings of skewed differen- tiation of STAT3

_LOF

and STAT1

_GOF

cT

_FH

cells to an IFN- g–secreting subset, we hypothesized that IFN-g production by CXCR3

¹

cT

_FH

cells would compromise their B-helper function (Fig 2, J).

¹⁰

To investigate this, it was important to demonstrate that IFN-g directly signals in human B cells. IFN-g induced STAT1 phosphorylation in EBV-transformed lymphoblastoid cell lines from healthy control subjects but not from STAT1- or IFNGR-deficient subjects, whereas IL-21–induced STAT3 activa- tion was unaffected by such mutations (Fig 5, A). Next, we cocul- tured cT

_FH

and allogeneic B cells with or without exogenous IFN-g. Exogenous IFN-g suppressed TD differentiation of

normal B cells by 40% to 50% but had no effect on immunoglob- ulin secretion by IFNGR1- or STAT1-deficient B cells (Fig 5, B), demonstrating B cell–intrinsic signaling through IFN-g receptor 1/STAT1 mediates the repressive effect of IFN-g on T

FH

-induced B-cell differentiation.

If IFN-g plays a substantial role in regulating immunoglobulin production in vivo, mutations in molecules regulating its produc- tion (ie, IL12RB1) or signaling (ie, IFNGR1 or STAT1) should cause hypergammaglobulinemia. Indeed, 60% (9/15), 38% (6/

16), and 53% (17/32) of patients with IFNGR1/2, STAT1, and IL12RB1 LOF mutations, respectively, had serum IgG levels greater than those of age-matched control subjects (Fig 5, C), whereas 60% and 25% of IL12RB1- or STAT1-deficient patients also had increased serum IgM levels (Fig 5, D). Notably, several patients whose immunoglobulin levels fell within the normal range were actually at the upper end of normal (Fig 5, C and D). Thus a key function of IFN-g in vivo is to suppress immuno- globulin production.

Mutations in STAT3 , STAT1 , and IL21R cause aberrant expression of PD-1 on CD4

¹

T cells

Although PD-1 is highly expressed on T

FH

cells,

^24,47

only a sub- set of cT

_FH

cells exhibit increased PD-1 expression.

9,11,12,15,17

PD- 1

^hi

cT

FH

cells have been used as a biomarker of humoral immunity in health and disease.

8,11,12,14,15,17

Because cT

_FH

development was perturbed by different PID-causing mutations, we examined PD-1 expression in these patients. PD-1 was expressed at low levels on naive CD4

¹

T cells, irrespective of genotype, and increased on non-cT

FH

memory cells from healthy control subjects and patients (Fig 6, A and B). Non-T

_FH

memory cells from patients with STAT1

GOF

and IL21/R

LOF

mutations and cT

FH

cells from patients with STAT3

_LOF

and STAT1

_GOF

mutations expressed significantly more PD-1 than those from control subjects (Fig 6, A and B).

None of the other mutations affected PD-1 expression, suggesting that signaling through STAT3/STAT1, possibly downstream of IL- 21, regulates PD-1 expression on memory CD4

¹

T cells.

Because STAT3

_LOF

, STAT1

_GOF

, and, to a lesser extent, IL21R mutations skewed cT

FH

differentiation toward a CXCR3

¹

pheno- type and upregulated PD-1 expression, we tested whether these 2 observations were related. PD-1 was expressed by all cT

FH

cell subsets from healthy control subjects, with significantly greater expression on CXCR3

¹

and CXCR3

¹

CCR6

¹

subsets compared with CCR6

¹

and CCR6

²

CXCR3

²

subsets (Fig 6, C). Despite this, PD-1 expression was significantly higher on all cT

_FH

cell sub- sets from patients with STAT3

LOF

mutations and for most cT

FH

cell subsets from STAT1

_GOF

and IL21R-deficient patients compared with control subjects (Fig 6, C and D). Thus STAT3

LOF

, STAT1

GOF

, or IL21/IL21R mutations dysregulate PD-1 expression.

DISCUSSION

It was first reported in 1965 that thymus-derived cells are

required for antigen-specific antibody responses.

⁴⁸

Although it

was initially proposed that T

_H

2 cells mediated B-cell differentia-

tion,

³²

it is now clear that this is mediated by T

_FH

cells. For this

reason, T

FH

cells are being used as a biomarker for humoral im-

munity.

8-15,17,20,49

Furthermore, targeting T

_FH

cells might consti-

tute a viable approach of enhancing the efficacy of some vaccines

and treating diseases caused by autoantibodies.

^2,4,5,7

Tracking

T

FH

cells in human subjects requires a clear understanding of their

circulating counterparts. Recent studies revealed heterogeneity within circulating CD4

¹

CXCR5

¹

T cells, with CCR7

^lo

PD1

¹

or CXCR3

²

/CCR6

¹

PD1

¹

subsets emerging as the most efficient helpers for B-cell differentiation.

^8,11,17

Notably, frequencies of these subsets correlated with in vivo T

_FH

cell behavior, such as generating neutralizing antibodies after infection/vaccina- tion,

^12,17

excessive activity in autoimmunity,

^8,9,13-15

or impaired function in HIV infection.

^11,49

However, other studies found that CXCR3

¹

cT

_FH

cells

²⁰

or CD4

¹

CXCR5

²

T cells

¹⁸

predicted antibody responses after vaccination. Thus despite substantial ad- vances in our understanding of human T

_FH

cell biology, the exact correlates between successful antibody responses and CD4

¹

T- cell subsets remain controversial and present a roadblock to trans- lating these findings to the clinic.

Here we examined a wide spectrum of PIDs to delineate quantitative and qualitative consequences of gene mutations on cT

FH

cells. Consistent with previous studies, cT

FH

cell numbers were reduced by mutations in CD40LG, ICOS, BTK, or STAT3.

^19,38,39

Our finding of reduced cT

FH

cell numbers in pa- tients with NEMO deficiency revealed another similarity between disease caused by CD40LG and NEMO mutations, which is consistent with NEMO being required for CD40 signaling.

³⁵

The significant reduction in cT

_FH

cell numbers in patients with IL-10R deficiency was unexpected because IL-10/IL-10R signaling in mice reduced T

_FH

cell formation.

⁴

This points to possible species-specific differences in the role of IL-10 in

regulating human and murine T

FH

cells, which parallels the distinct effects of IL-10 on B-cell differentiation in these spe- cies.

²⁸

The deficits in cT

FH

cell numbers in these groups of pa- tients with PID are likely a direct effect of the mutation, rather than being secondary to infection, because cT

_FH

cells were de- tected at normal frequencies in STAT1- and IFNGR1/2-deficient subjects. Our findings highlight that interactions between CD4

¹

T cells, dendritic cells, and B cells are required for T

FH

forma- tion.

^2-5

It is likely that CD40 ligand/CD40/NEMO signaling in B cells contributes to the maintenance of cT

FH

cells in human sub- jects. However, a T cell–intrinsic role for NEMO cannot be dis- counted. The importance of B cells is also evident from studies reporting diminished cT

_FH

cell frequencies in patients with severe B-cell deficiency caused by E47

⁵⁰

or NFKB2

⁵¹

mutations. Our data also shed light on the division of labor between STAT3- activating cytokines in generating cT

_FH

cells, as well as on the functionality and requirements for cT

FH

cell subset formation.

Mutations in STAT3 or IL10R, but not IL21/IL21R, had compara- ble deleterious effects on cT

FH

cell frequencies, implying IL-10/

STAT3 signaling is important in establishing the pool of cT

_FH

cells. However, STAT3

_LOF

, STAT1

_GOF

, and IL21/R

_LOF

mutations, but not IL-10R deficiency, reduced CCR6

¹

cT

FH

cell numbers, most likely because of STAT3 directly regulating CCR6 expres- sion.

³⁷

The resultant population of cT

FH

cells in these subjects was skewed to a CXCR3

¹

IFN-g

¹¹

IL-10

^low

phenotype resem- bling STAT3-deficient murine T

FH

cells, which also adopt a

FIG 5.IFN-gsuppresses B-cell differentiationin vitroandin vivo.A,Phosphorylation of STAT1 or STAT3 in EBV-transformed lymphoblastoid cell lines from healthy control subjects(red histograms)or patients with IFNGR2 (blue)orSTAT1 (green)LOF mutations in response to IFN-gor IL-21.Gray histogram, Response of unstimulated cells from healthy control subjects.B,cTFHcells from healthy control subjects were cocultured with allogeneic normal, IFN-greceptor 1– or STAT1-deficient B cells in the absence or presence of exoge- nous IFN-g. Immunoglobulin secretion was determined after 7 days. Values represent the mean percentage of immunoglobulin secretion (6SEM) in the presence of IFN-grelative to its absence (defined as 100%).C andD, serum levels of IgG and IgM in patients withIFNGR1/IFNGR2,IL12RB1, orSTAT1LOFmutations.

Thesolid upperandlower linescorrespond to normal serum immunoglobulin levels in age-matched con- trol subjects.

J ALLERGY CLIN IMMUNOL nnn2015

10 MA ET AL

FIG 6.Aberrant expression of PD-1 on cTFHcells from patients withSTAT3LOF,STAT1GOF, andIL21/RLOF

mutations.AandB, PD-1 expression on naive, non-TFHmemory, and cTFHcells. Contour and histogram plots in Fig 6,A, represent individual healthy control subjects or patients. The graph in Fig 6,B, depicts average PD-1 expression (mean fluorescent intensity[MFI]6SEM) for all subjects tested (healthy donors, n545; patients with mutations in the following genes:STAT3, n515;STAT1GOF, n518;STAT1LOF, n57;IL21R/IL21, n55;IL10R, n54;CD40LG, n53;NEMO, n55-6;BTK, n55-6;ICOS, n55;IL12RB1, n54;IFNGR1/2, n54; andTYK2, n54).CandD, PD-1 expression on CXCR3¹CCR6², CXCR3²CCR6¹, CXCR3²CCR6², and CXCR3¹CCR6¹cTFHcell subsets cells from healthy control subjects (n525) or pa- tients withSTAT3LOF(n510),STAT1GOF(n510), orIL21/RLOFmutations (n55). Significant differences (1-way ANOVA) between control subjects and patients are indicated. *P< .05, **P< .01, ***P< .005, and

****P< .001.

T

_H

1 fate.

⁵²

This revealed a dominant role for IL-21/STAT3 signaling in specifying cT

FH

cell subsets. These data demonstrate that specific mutations cause qualitative changes in cT

_FH

cells that would not be apparent by quantifying CD4

¹

CXCR5

¹

T cells.

The mechanism underlying greater helper function of CCR6

¹

and corresponding poor function of CXCR3

¹

in cT

FH

cells is un- known. Although it was reported that CCR6

¹

cT

_FH

cells produce more IL-21 than other subsets,

¹⁰

this was not confirmed by us or others

^11,13,17

and is unlikely to explain their ability to promote B- cell differentiation. Rather, on the basis of several lines of evi- dence, including our data, we propose this results from reduced IFN-g production.

First, exogenous IFN-g reduced immunoglobulin production by human PBMCs

^43,44

or in B-cell/T

_FH

cell cocultures.

⁴⁵

This was not observed for IFNGR1- or STAT1-deficient B cells, demonstrating a B cell–intrinsic inhibitory effect of IFN-g.

Second, Mycobacterium leprae–specific IgG levels negatively correlated with IFN-g production by PBMCs in response to M leprae antigens.

⁴²

Third, serum immunoglobulin levels were increased in many patients with mutations in the IFN-g signaling pathway.

Fourth, there were trends for increased class-switched memory B cells in patients with IFNGR1/2, TYK2, or STAT1

_LOF

mutations.

Overall, IFN-g appears to attenuate antigen-specific antibody responses in human subjects in vivo, thereby explaining the poor helper function of CXCR3

¹

cT

_FH

cells. Because IFN-g is induced by IL-12/IL-12R/Tyk2 signaling and itself signals through STAT1, it is likely that the similar phenotypes detected in patients with IL12RB1, TYK2, IFNGR1/2, and STAT1 LOF mutations reflect the contribution of a common pathway to the regulation of B-cell differentiation and immunoglobulin secretion.

The abundance of IFN-g–producing cT

_FH

cells in patients with STAT3

_LOF

and STAT1

_GOF

mutations, together with reduced CCR6

¹

cT

FH

cell numbers, would thus contribute to impaired hu- moral immune responses.

^21,41

This would be further compounded by reduced production of IL-10, an important cytokine for B-cell differentiation in the context of T

_FH

/B-cell interactions.

^22,28,45

CCR6

¹

cT

FH

cell numbers are also reduced in patients with HIV infection,

¹¹

in whom T

_FH

cells exhibit impaired B-helper function.

^11,49,53

The similarities in cT

_FH

cell dysregulation in pa- tients with HIV infection and PIDs identifies intrinsic (specific gene mutations) and extrinsic/environmental (chronic viral infec- tion/pathogen exposure) factors as modifiers of T

FH

cell differen- tiation and function. Thus it would be interesting to determine whether T

FH

cells in HIV

¹

patients also have a skewed cytokine profile with increased IFN-g and reduced IL-10 levels. Our find- ings regarding IFN-g and impaired humoral immunity also explain the paradoxical data of Schmitt et al

⁵⁴

showing that cT

_FH

cell numbers are reduced in patients with IL-12Rb1 defi- ciency, yet these patients have intact, if not heightened, antibody responses to vaccination or infection. Here, although impaired IL- 12R signaling might reduce cT

FH

cell numbers,

⁵⁴

there will also be less IL-12–induced IFN-g production, resulting in a qualitative change in the cT

FH

cytokine repertoire with reduced IFN- g–mediated suppression of B-cell responses.

A defining feature of T

_FH

cells is increased PD-1 expression.

⁴⁷

Although most cT

FH

cells express low PD-1 levels, a subset with increased PD-1 expression corresponds to the most efficient pop- ulation of B-helper cells.

^8,11,12,17

Given our finding of fewer cT

FH

cells in patients with STAT3

_LOF

mutations and skewing of cT

_FH

cell subsets in patients with STAT1

GOF

or STAT3

LOF

mutations

to a phenotype with reduced B-helper function, it might be pre- dicted that their cT

FH

cells would have reduced PD-1 expression.

However, this was not the case because PD-1 expression was significantly increased on cT

FH

cells in these patients, raising several important possibilities.

First, because PD-1 engagement impedes T

FH

formation in vivo,

^55-57

exaggerated signaling through overexpressed PD-1 on STAT1

_GOF

and STAT3

_LOF

cT

_FH

cells might further impair func- tion. Interestingly, the PD-1 ligand PD-L1 is aberrantly expressed on CD4

¹

T cells from patients with STAT1

_GOF

mutations,

⁴¹

revealing a possible autocrine mechanism of restrained cT

FH

func- tion. Similarly, although PD-1 is expressed comparably on T

_FH

cells from healthy donors and HIV-infected patients, more GC B cells from HIV-infected patients expressed PD-L1 than those from healthy donors.

⁵³

Importantly, this heightened expression impaired T

FH

cell IL-21 production, suppressing effector func- tion.

^53,58

PD-1

^hi

–expressing T

_FH

cells also undergo accelerated death compared with cells expressing lower PD-1 levels.

⁵⁹

Thus dysregulated expression of PD-1 or its ligand or ligands could compromise T

FH

cell function or survival in patients with these PIDs.

41,53,58,59

Interestingly, cT

_FH

cells in STAT3-deficient patients also expressed CD57 (M. C. Cook, personal communica- tion). Although CD57 is expressed by a large proportion of T

_FH

cells in lymphoid tissues, it is usually absent from cT

_FH

cells in healthy control subjects.

^9,17,24

Expression of CD57 by T

FH

cells has been associated with an exhausted/senescent phenotype and a susceptibility to rapidly undergo death relative to CD57

²

CD4

¹

T cells (M. C. Cook, personal communication).

⁶⁰

Thus it is possible that cT

FH

cells in patients with some PIDs correspond to ‘‘exhausted’’ cells, and this would further affect their function in the setting of humoral responses in these patients. These possi- bilities are currently being investigated.

Second, our data caution against solely quantifying cT

_FH

cells with respect to PD-1 as a biomarker for humoral immunity.

Although this correlation has been established in vaccination,

¹²

HIV infection,

^11,17

and autoimmunity,

^8,9,15,16

it might be misleading in patients with some monogenic PIDs. Thus cT

_FH

cells in patients with these and other immunopathologies should be assessed for phenotype and function. The importance of doing so is highlighted by a recent study reporting impaired cT

_FH

cell function in elderly subjects, even though their cT

FH

cells ex- pressed sufficient IL-21 and ICOS levels.

¹²

Third, STAT3 signaling during T

FH

cell differentiation is required to not only generate the CCR6

¹

subset but also regulate PD-1 and minimize suppression through PD-1/PD-L1 interac- tions. The mechanism underlying STAT3-mediated PD-1 repres- sion is unknown but might involve IL-21 because IL21R deficiency partially recapitulated this defect. Interestingly, because STAT3 can protect T

_FH

cells from the inhibitory effects of IFN-g

⁵²

and IFN-g induces PD-1 on murine CD4

¹

T cells,

⁶¹

increased PD-1 expression on STAT3

_LOF

cT

_FH

cells might result from heightened IFN-g signaling in the absence of STAT3.

Intriguingly, our findings also revealed overlapping cellular phenotypes caused by STAT3

LOF

and STAT1

GOF

mutations, including reduced production of T

_H

17 cytokines by memory CD4

¹

T cells, skewed differentiation of and PD-1 expression by cT

FH

cells, and fewer memory B cells. These shared functional defects are consistent with similar clinical features of these distinct molecular entities, such as mucocutaneous candidiasis and impaired antibody responses.

^21,35,41

Although the mechanism underlying these common features is unknown, the data suggest

J ALLERGY CLIN IMMUNOL nnn2015

12 MA ET AL