Combined immunodeficiency in a patient with deficiency of c-Rel

Capsule summary:

This study reports a homozygous mutation in REL abrogating c-Rel protein expression in a patient with combined immunodeficiency characterized by susceptibility to Mycobacterium tuberculosis, Salmonella, Cryptosporidium, and cytomegalovirus.

To the Editor:

c-Rel is a transcription factor in the NF-κB family, critical for T and B cell function. c-Rel binds to the promoters of genes that encode cytokines important for immunity against infectious pathogens, including IL-2, IFN-γ, IL-12, IL-21, and IL-23.¹ Rel-deficient mice have impaired T and B cell proliferation, hypogammaglobulinemia, and cytokine secretion.^2–5 Human patients with mutations in c-Rel have not been previously published. Here, we describe a homozygous mutation abrogating c-Rel expression in a patient with impaired T and B cell activation, susceptibility to opportunistic infections, and Mycobacterium tuberculosis.

The patient is the son of first-degree cousins of Arabic descent. He received live vaccinations against poliovirus, measles, and Bacille Calmette-Guerin (BCG) without clinical sequelae. At 30 months of age, he developed chronic diarrhea associated with multiple infections, including Salmonella enteritica, cytomegalovirus (CMV), and Cryptosporidium. In the setting of chronic cryptosporidiosis, he developed hepatomegaly and sclerosing cholangitis. Immunologic evaluation at six years of age was notable for leukocytosis, lymphocytosis, and thrombocytosis (Table 1). He had increased numbers of CD4⁺ and CD8⁺ T cells with decreased memory CD4⁺CD45RO⁺ T cells, and reduced proliferation to phytohemagglutinin (PHA) (Table 1). He had B cell lymphopenia, impaired B cell proliferation upon stimulation with CD40 ligand + IL-21, reduced IgG and undetectable IgA levels, and non-protective titers to the tetanus and diphtheria vaccines despite booster vaccinations (Table 1). Despite vaccination with BCG, he developed femoral Mycobacterium tuberculosis osteomyelitis at 7 years of age that was successfully treated with isoniazid, rifampin, pyrazinamide, and ethambutol. He also developed one lobar pneumonia confirmed by chest X-ray. Additional immunophenotyping done at 9 years of age revealed a predominance of naïve CD19⁺IgD⁺CD27⁻ B cells (85%; normal 47.3 – 77%) and reduced percentages of switched memory CD19⁺CD27⁺IgD⁻ B cells (0.3%; reference range 10.0 – 30.4). He is currently treated with intravenous immunoglobulin and prophylactic antibiotics while undergoing evaluation for hematopoietic stem cell transplantation.

Table 1.

Immunological profile of the proband at 6 years of age. Bold values are outside the normal range. Age-matched normal ranges are shown in parentheses.

Hemogram 103 cells/μL	6 years
White blood cells	12,600 (5,700 – 9,900)
Neutrophils	5,800 (2,800 – 6,300)
Lymphocytes	5,900 (1,200 – 4,700)
Platelets	447,000 (227,000 – 350,000)
Lymphocyte subpopuations
CD3+, 103 cells/μL	4,969 (1,200 – 2,600)
CD3+CD45RO+, % CD3+	10.3% (15 – 41)
CD3+CD4+, 103 cells/μL	1,843 (650 – 1,500)
CD3+CD8+, 103 cells/μL	2,483 (370 – 1,100)
CD19+, 103 cells/μL	103 (270 – 860)
CD3− CD56+, 103 cells/μL	70 (100 – 480)
T cell proliferation, % control
PHA	47.3%
PMA + ionomycin	108%
B cell proliferation, % control
CD40L + IL21	6.4%
Serum Immunoglobulins mg/dL (normal)
IgG	150 (650-1,150)
IgM	150 (50-150)
IgA	Undetectable (40-160)
Anti-diphtheria, IU/mL	0.015 (>0.1 IU/mL)
Tetanus, IU/mL	0.041 (>0.1 IU/mL)

Whole-exome sequencing of genomic DNA from the proband identified 80 rare variants in the proband which were absent from the gnomAD and 1000 genomes databases (Table E1 in this article’s Online Repository). Of these, a homozygous mutation in the canonical donor splice-site after exon 5 of REL (NM_002908.3: c.535+1G>A) was considered the most likely pathogenic variant due to the known contribution of c-Rel in T and B cell activation and cytokine secretion. Sanger sequencing confirmed that the mutation is homozygous in the proband and heterozygous in his parents and healthy brother (Fig E1 in this article’s Online Repository). Amplification of the patient’s REL cDNA using primers specific to sequences in exons 2 and 7 identified an alternatively spliced transcript with lower expression than that of wild-type REL from control cDNA (Fig 1). Sanger sequencing of the patient’s mutant REL transcript indicated that the mutant utilizes a cryptic donor splice site in exon 5 (NM_002908.3: c.483) and a cryptic acceptor splice site near the start of exon 6 (NM_002908.3: c.538) (Fig 1B, top). The mutant transcript lacks 54 nucleotides encoding 18 residues within the Rel homology domain (Fig 1B, bottom). If expressed, the molecular mass of the mutant c-Rel protein is predicted to be 66 kilodaltons. However, immunoblotting using an antibody specific to the protein’s N-terminus demonstrated the absence of c-Rel protein in the patient’s peripheral blood mononuclear cells (PBMCs) with no detectable truncation products, indicating that the mutation abrogates protein expression (Fig 1C).

Table E1.

Variants identified by WES in the proband which are absent from the gnomAD and 1000 genomes databases. The mutation in REL is bolded.

Gene	Chr:Pos	Ref/Alt	Zygosity	Alteration(s)
TTLL10	1:1120348	G/A	Heterozygous	NM_001130045.1:c.1261-1G>A
PRDM16	1:3347585	C/T	Heterozygous	NM_022114.3:c.3434C>T
MST1L	1:17085015	C/T	Heterozygous	NM_001271733.1:c.1460G>A
IPO13	1:44433161	A/T	Heterozygous	NM_014652.3:c.2788A>T
BRDT	1:92457845	C/T	Heterozygous	NM_001242806.2:c.2101C>T
OR10J1	1:159409676	T/C	Heterozygous	NM_012351.2:c.128T>C
IGSF8	1:160061714	A/G	Heterozygous	NM_052868.5:c.1741T>C
PLEKHA6	1:204192629	C/A	Heterozygous	NM_014935.4:c.3116G>T
MARK1	1:220809217	C/G	Heterozygous	NM_001286124.1:c.1319C>G
OBSCN	1:228525005	C/T	Heterozygous	NM_001271223.2:c.19592C>T
HEATR1	1:236721531	G/C	Heterozygous	NM_018072.5:c.5204+6C>G
REL	2:61144153	G/A	Homozygous	NM_002908.3:c.535+1G>A
ANKRD36C	2:96616506	G/C	Heterozygous	NM_001310154.1:c.1286C>G
ANKRD36C	2:96616507	C/A	Heterozygous	NM_001310154.1:c.1285G>T
SSFA2	2:182756857	C/T	Heterozygous	NM_001130445.2:c.64C>T
GIGYF2	2:233712226	CACAGCAGCAGCAGCAGCAGC/-	Heterozygous	NM_001103147.1:c.3693_3713del
COL6A5	3:130110144	G/C	Heterozygous	NM_001278298.1:c.2539G>C
NMD3	3:160939747	TTTTTTTTTTTTTTTTT/-	Heterozygous	NM_001320227.1:c.-20-22_-20-6del
CYTL1	4:5018591	T/C	Heterozygous	NM_018659.2:c.299A>G
SORCS2	4:7436509	C/T	Heterozygous	NM_020777.2:c.548+38427C>T
CEP135	4:56835228	G/C	Heterozygous	NM_025009.4:c.1045-1G>C
THEGL	4:57467169	C/A	Heterozygous	NM_001256475.1:c.1177C>A
CEP44	4:175223340	A/G	Heterozygous	NM_001145314.1:c.237+3A>G
CNPY3	6:42902265	G/T	Heterozygous	NM_001318842.1:c.204G>T
IMPG1	6:76744476	A/T	Heterozygous	NM_001563.3:c.330T>A
MDN1	6:90494802	G/C	Heterozygous	NM_014611.2:c.1378C>G
TIAM2	6:155450909	C/A	Homozygous	NM_012454.3:c.552C>A
GET4	7:916358	G/A	Heterozygous	NM_015949.2:c.76G>A
DNAH11	7:21599405	G/T	Heterozygous	NM_001277115.1:c.877G>T
SEMA3A	7:83590893	T/C	Heterozygous	NM_006080.2:c.2110A>G
RP1L1	8:10480232	C/A	Heterozygous	NM_178857.5:c.480G>T
SH2D4A	8:19251032	G/A	Heterozygous	NM_022071.3:c.1252G>A
LYN	8:56866432	A/G	Heterozygous	NM_002350.3:c.679A>G
ZNF517	8:146033370	G/A	Heterozygous	NM_213605.2:c.1069G>A
RIC1	9:5763776	T/C	Heterozygous	NM_020829.3:c.2749T>C
OR1L1	9:125424679	C/A	Heterozygous	NM_001005236.3:c.685C>A
SLC16A12	10:91198403	T/A	Heterozygous	NM_213606.3:c.986A>T
CPEB3	10:94000082	T/C	Heterozygous	NM_014912.4:c.26A>G
TACC2	10:123848028	C/T	Homozygous	NM_206862.3:c.5495C>T
MUC6	11:1016613	G/C	Heterozygous	NM_005961.2:c.6188C>G
OR52J3	11:5068002	A/G	Heterozygous	NM_001001916.2:c.247A>G
OR56A3	11:5969049	T/A	Heterozygous	NM_001003443.2:c.473T>A
ARNTL	11:13393880	GGG/-	Heterozygous	NM_001297719.1:c.992_994delGGG
LMO2	11:33886347	C/T	Heterozygous	NM_005574.3:c.265G>A
LRRC4C	11:40136226	C/G	Heterozygous	NM_020929.2:c.1617G>C
CCDC85B	11:65658456	C/T	Heterozygous	NM_006848.2:c.202C>T
SHANK2	11:70332235	G/A	Heterozygous	NM_133266.4:c.2399C>T
APLP2	11:129940533	C/G	Heterozygous	NM_001642.2:c.105+556C>G
VWF	12:6167043	G/C	Heterozygous	NM_000552.4:c.1701C>G
ATN1	12:7050180	C/T	Heterozygous	NM_001007026.1:c.3352C>T
PIK3C2G	12:18762523	A/T	Heterozygous	NM_001288772.1:c.4142A>T
CELA1	12:51740388	CTGGACCATATCC/-	Heterozygous	NM_001971.5:c.16+7_16+19delGGATATGGTCCAG
CELA1	12:51740403	TT/-	Heterozygous	NM_001971.5:c.16+3_16+4delAA
G2E3	14:31061589	C/T	Heterozygous	NM_017769.4:c.298C>T
GPATCH2L	14:76633107	G/A	Heterozygous	NM_017926.3:c.727+37G>A
SLC24A4	14:92790151	G/C	Heterozygous	NM_153647.3:c.-23-1G>C
WARS	14:100820895	T/G	Heterozygous	NM_004184.3:c.431A>C
PKM	15:72492836	C/T	Heterozygous	NM_001206796.2:c.1690G>A
CD276	15:74005281	G/A	Heterozygous	NM_001024736.1:c.1589G>A
BCL2A1	15:80253513	T/C	Heterozygous	NM_004049.3:c.424A>G
CEMIP	15:81224250	A/C	Heterozygous	NM_018689.2:c.2663A>C
NME4	16:448386	G/T	Heterozygous	NM_005009.2:c.92-604G>T
PARN	16:14704541	-/T	Heterozygous	NM_002582.3:c.514dupA
GTF3C1	16:27476729	G/C	Heterozygous	NM_001520.3:c.5207C>G
CNGB1	16:57945772	T/C	Heterozygous	NM_001297.4:c.2377A>G
MRM3	17:685929	T/C	Heterozygous	NM_018146.3:c.311T>C
ZBTB4	17:7385592	C/T	Heterozygous	NM_020899.3:c.-81+1731G>A
ATAD5	17:29196590	A/G	Heterozygous	NM_024857.4:c.3533A>G
MYO1D	17:30986172	A/C	Heterozygous	NM_015194.2:c.2306T>G
SOCS7	17:36523760	G/A	Heterozygous	NM_014598.3:c.1192-5G>A
KRTAP4-8	17:39254133	G/T	Heterozygous	NM_031960.2:c.204C>A
MRPS23	17:55926758	C/T	Heterozygous	NM_016070.3:c.58G>A
TNFAIP8L1	19:4651887	C/A	Heterozygous	NM_152362.2:c.6C>A
SH2D3A	19:6753521	G/A	Heterozygous	NM_005490.2:c.1516C>T
ZNF492	19:22817275	CG/-	Heterozygous	NM_020855.2:c.-95_-94delCG
FCGBP	19:40364256	G/C	Heterozygous	NM_003890.2:c.14386C>G
LILRB2	19:54783690	G/T	Heterozygous	NM_005874.4:c.311C>A
CHODL	21:19618290	G/A	Heterozygous	NM_024944.2:c.79+671G>A
IL17RA	22:17566097	G/A	Heterozygous	NM_014339.6:c.116G>A
CELSR1	22:46768834	C/G	Heterozygous	NM_014246.1:c.7705G>C

Figure E1.

Chromatograms of Sanger sequencing of the REL mutation in the patient, his parents, and healthy brother.

Figure 1.

A. Representative electrophoresis (left) and densitometry quantification of REL cDNA from EBV-transformed lymphoblastoid cell lines (EBV-LCLs) from controls and the patient. B. Schematic of REL cDNA and protein with the patient’s deleted nucleotides and residues in red. C. Representative immunoblot of c-Rel in lysates of peripheral blood mononuclear cells (PBMCs) from the patient and controls. D. CD3⁺ T cell proliferation to phytohemagglutinin (PHA), measured by CellTrace Violet (CTV) dilution. E. IL-2 secretion from patient and control PBMCs, with or without PHA stimulation. F. CD3⁺ T cell proliferation to PHA+IL-2. G, H. Secretion of IFN-γ (G) and IL-23 (H) by PBMCs from the patient and controls cultured with the indicated stimuli. I, J. Secretion of IL-12 (I) and IL-8 (J) by EBV-LCLs from the patient and controls stimulated with PDBu. K. CD19⁺ B cell proliferation to anti-CD40+IL-21. L. IL-21 secretion by PHA blasts from the patient or controls with or without PHA+IL-23 stimulation. M. Schematic of the contribution of c-Rel to adaptive and innate immunity. *p<0.05, **p<0.01. Data in A, E, G – J, and L are pooled from two independent experiments; data in C, D, F, K are representative of two to three independent experiments.

The patient’s history of infections with CMV and cryptosporidiosis with secondary sclerosing cholangitis indicated defective T cell function. Mice deficient in c-Rel have impaired T cell proliferation and defective IL-2 secretion; the addition of exogenous IL-2 normalizes T cell proliferation.^2,3 After PHA stimulation, the patient’s T cells had reduced proliferation, as demonstrated by minimal proliferation beyond the third round of division, and negligible IL-2 secretion (Fig 1D and E). As has been seen in c-Rel-deficient mice, addition of IL-2 normalized the proliferation of the patient’s T cells stimulated with PHA (Fig 1F).

c-Rel binds to the promoters of the genes encoding IFN-γ, IL-23, and IL-12, all of which contribute to host defense against mycobacteria and Salmonella.^1,6 IFN-γ was undectable after PHA stimulation of the patient’s T cells and failed to increase with the addition of IL-2 (Fig 1G). Antigen presenting cells are the predominant sources of IL-12 and IL-23.⁶ Stimulation of the patient’s PBMCs with LPS did not result in IL-23 secretion, as compared to the control (Fig 1H). Similarly, the patient’s EBV-LCLs secreted minimal amounts of IL-12 after phorbol 12, 13-dibutyrate stimulation (PDBu) (Fig 1I). In contrast, secretion of IL-8, a cytokine independent of c-Rel, was comparable between control and patient EBV-LCLs (Fig 1J). Although the patient demonstrated susceptibility to extrapulmonary Mycobacterium tuberculosis, he had no evidence of complications from the less virulent, attenuated Mycobacterium bovis BCG strain. The residual T cell proliferation in the patient may have helped to prevent complications from the BCG vaccine.

In B cells, c-Rel is among the first NF-kB transcription factors activated after CD40 stimulation.⁴ c-Rel-deficient mice have severely impaired B cell proliferation following CD40 ligation.^2,4 Similarly, after anti-CD40 stimulation, nearly all of the patient’s CD19⁺ B cells were limited to only 1 round of division, compared to up to 4 rounds of division for the control (Fig 1K). In addition to its role in B cell proliferation, c-Rel also binds to the promoter of the gene encoding IL-21, a cytokine important for the development of germinal center B cells, isotype switching, and the antibody response to T dependent antigens.⁷ In vitro stimulation of PHA blasts with PHA+IL-23 revealed defective secretion of IL-21 in the patient compared to control (Fig 1L). Our patient shares features found in individuals with deficiency of IL-21 or the IL-21 receptor that include hypogammaglobulinemia, recurrent gastrointestinal infections, pneumonia, and a predominance of naïve B cells.⁷ In addition to its role in B cell function, IL-21 enhances the effector function of CD8⁺ T cells by increasing the expression of IFN-γ, perforin, and granzyme, all of which facilitate the elimination of Cryptosporidium.7

The patient’s combined immunodeficiency mirrors the defects in T cell activation, B cell proliferation, innate immunity and humoral immunity seen in c-Rel-deficient mice.^2–5 His susceptibility to mycobacteria, Salmonella, and opportunistic pathogens illustrate the pleiotropic contributions of c-Rel to host immunity in humans, as illustrated in Fig 1M. Small molecule inhibitors of c-Rel are under investigation as therapies for lymphoid tumors and graft versus host disease, but the potential risks of infections with these drugs have not yet been studied.⁸ This novel immunodeficiency highlights the potential susceptibility to a broad range of pathogens with therapeutic strategies targeting c-Rel.

Materials and Methods

Study oversight.

Written informed consent was obtained from the patient’s parents. This study was approved by the Institutional Review Board of Boston Children’s Hospital.

Immunophenotyping.

Antibodies and viability dyes used for flow cytometry studies of T and B cells are as follows: Zombie Aqua™ Fixable Viability Kit, anti-human CD4-APC/Cy7 (clone OKT4), anti-human CD8-PE (clone SK1), anti-human CD3-FITC (clone OKT3), Annexin V, Alexa Fluor® 647, anti-human CD27-APC/Cy7 (clone O323), anti-human CD19-FITC (clone HIB19), all from Biolegend, Dedham, MA, USA). The analysis of flow cytometric data was performed with FlowJo software version 10.5.0 (TreeStar).

Cell cultures.

T cell proliferation was quantified using dilution of CellTrace™ Violet Cell Proliferation Kit (Invitrogen, Waltham, MA, USA) after stimulation with PHA (5 μg/mL; Sigma Aldrich, St. Louis, MO, USA) ± IL-2 (40 U/mL; R&D Systems, Minneapolis, MN, USA) for 2.5 days, at which point cell culture supernatants were also harvested for measure of IFN-γ and IL-2. B cell proliferation was quantified using dilution of CellTrace Violet after stimulation with anti-CD40 monoclonal antibody (clone G28-5, at a concentration of 10 μg/mL, Enzo Biochem, Farmingdale, NY, USA) + IL-21 (50 ng/mL, Cell Sciences, Newburyport, MA, USA) for 3 days. For IL-21 secretion, PBMCs were stimulated with PHA for 7 days, followed by PHA + IL-23 (100 ng/mL, R&D Systems, Minneapolis, MN, USA) for 72 hours. For IL-23 secretion, patient and control PBMCs were stimulated with LPS (100 ng/mL, InvivoGen, San Diego, CA, USA) for 18 hours. For IL-12 secretion, live B-LCLs were isolated after eliminating dead cells using the Dead Cell Removal Kit (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) per the manufacturer’s guidelines. Remaining live B-LCLs were stimulated with pDBU (10⁻⁷ M, Sigma Aldrich, St. Louis, MO, USA) for 18 hours as previously described.¹

Cytokine quantification.

Concentrations of IL-21, IFN-γ, IL-2, and IL-12 in cell culture supernatants were measured by Cytometric Bead Array (BD Biosciences, Singapore) per the manufactuer’s guidelines. IL-23 was measured in cell culture supernatants by ELISA.

Whole exome sequencing (WES).

Library preparation for WES was performed using Agilent SureSelect. WES was performed on genomic DNA from the proband using the Illumina HiSeq-2000 (Illumina Inc., San Diego, CA). The average coverage of the exome was 53x. The Burrows-Wheeler Aligner (BWA) was used to map reads to the human reference genome assembly GRCh37.² Variants were called using the Genome Analysis Toolkit (GATK).³

Primers.

Sanger sequencing of the proband’s genomic DNA for the c-REL mutation was performed with the following primers: (F: 5′-GATATTGAAGATTGTGACCTCA-3′, R: 5′- TATTTCTCTAATGTCTGGAAGT-3′). RT-PCR was performed using the following primers: c-REL (F: 5′-GCACAGACAACAACCGAA-3′, R: 5′- GTCAGAAGGTCTCCGCA-3′); 18S (F: 5′-CGGCTACCACATCCAAGGAA-3′, R: 5′- GCTGGAATTACCGCGGCT-3′). ImageJ analyzer software (1.48v) for quantitation.

Capillary immunoassay.

Control or patient PBMCs were lysed in sample buffer (Pierce, Appleton, WI, USA) and 2-mercaptoethanol (Sigma Aldrich, St. Louis, MO, USA Electrophoresis) and transferred to 12 – 230 kDa Wes Separation Modules (Protein Simple) per the manufacter’s guidelines with the following antibodies, as indicated in figure legends: anti-c-REL (clone D4Y6M, Cell Signaling Technology, Danvers, Massachusetts, USA), and anti-β-actin (Sigma Aldrich, St. Louis, MO, USA). Densitometry was performed using the Compass Software for Simple Western.

Statistics.

Student’s t-test was used for comparisons of two groups; one- or two-way ANOVA with the with Holm-Šídák post-comparison test was used for the comparison of more than 2 groups, as indicated. Data are graphed as mean with SEM. All experiments were performed with two to three controls. Statistical analysis was performed using GraphPad Prism.

Abbreviations:

BCG

Bacille Calmette-Guerin

CMV

cytomegalovirus

EBV-LCLs

Epstein Barr virus-transformed lymphoblastoid cell lines

LPS

lipopolysaccharide

NF-κB

nuclear factor-kappa B

PDBu

phorbol 12, 13-dibutyrate

PBMCs

peripheral blood mononuclear cells

PHA

phytohemagglutinin

PMA

phorbol 12-myristate 13-acetate