Chronic Granulomatous Disease-Like Presentation of a Child with Autosomal Recessive PKCδ Deficiency

Background Autosomal recessive (AR) PKCδ deficiency is a rare inborn error of immunity (IEI) characterized by autoimmunity and susceptibility to bacterial, fungal, and viral infections. PKCδ is involved in the intracellular production of reactive oxidative species (ROS). Material and Methods We studied a 5-year old girl presenting with a history of Burkholderia cepacia infection. She had no history of autoimmunity, lymphocyte counts were normal, and no auto-antibodies were detected in her plasma. We performed a targeted panel analysis of 407 immunity-related genes and immunological investigations of the underlying genetic condition in this patient. Results Consistent with a history suggestive of chronic granulomatous disease (CGD), oxidative burst impairment was observed in the patient’s circulating phagocytes in a dihydrorhodamine 123 (DHR) assay. However, targeted genetic panel analysis identified no candidate variants of known CGD-causing genes. Two heterozygous candidate variants were detected in PRKCD: c.285C > A (p.C95*) and c.376G > T (p.D126Y). The missense variant was also predicted to cause abnormal splicing, as it is located at the splice donor site of exon 5. TOPO-TA cloning confirmed that exon 5 was completely skipped, resulting in a truncated protein. No PKCδ protein was detected in the patient’s neutrophils and monocyte-derived macrophages. The monocyte-derived macrophages of the patient produced abnormally low levels of ROS, as shown in an Amplex Red assay. Conclusion PKCδ deficiency should be considered in young patients with CGD-like clinical manifestations and abnormal DHR assay results, even in the absence of clinical and biological manifestations of autoimmunity. Supplementary Information The online version contains supplementary material available at 10.1007/s10875-022-01268-8.


Introduction
Protein kinase C delta (PKCδ) is a protein kinase that is ubiquitously expressed and has been implicated in various pathways involved in both innate and adaptive immunity [1][2][3][4]. Autosomal recessive (AR) PKCδ deficiency is a very rare inborn error of immunity (IEI), which has been reported in 17 patients from 10 unrelated kindreds worldwide [5][6][7][8][9][10][11][12][13]. Fifteen of these 17 patients presented clinical features of autoimmunity and had antinuclear antibodies (ANA) or anti-double-stranded DNA (anti-dsDNA) antibodies. Two patients presented no signs of autoimmunity and were negative for both ANA and anti-dsDNA antibodies [5]. Thirteen patients were diagnosed with early-onset systemic lupus erythematosus (SLE) and displayed photosensitivity, malar rash, alopecia, and hepatosplenomegaly. One patient developed B-cell lymphoproliferative disease and was diagnosed with autoimmune lymphoproliferative syndrome (ALPS) type III (OMIM #615,559) (8). A progressive decline in CD19 + B cells was frequently observed, with high proportions of CD21 low B cells and low proportions of memory B cells. Besides autoimmunity, most PKCδ-deficient patients presented with recurrent infections with various fungi, bacteria, or viruses [5][6][7][8][9][10][11][12][13]. Only one asymptomatic individual has been reported to date, and this lack of symptoms may be explained by his young age at diagnosis [5]. The susceptibility to fungal and bacterial infections of these patients has been shown to be partially due to the impaired production of reactive oxygen species (ROS) by the patients' phagocytes and myeloid cells [5]. Defective ROS production by phagocytes is the main cellular phenotype observed in patients with chronic granulomatous disease (CGD), who present with susceptibility to bacterial and fungal infections [14][15][16]. PKCδ deficiency and CGD are both characterized by impaired ROS production by phagocytes, but they also overlap in clinical presentation in terms of susceptibility to infectious disease. We report here a new patient with AR PKCδ deficiency and no history of autoimmunity who was initially suspected to have CGD.

DNA Extraction and Sanger Sequencing
The presence of the PRKCD variants was confirmed on genomic DNA extracted from whole blood or saliva samples with the DNeasy blood and tissue kit (#69,504, Qiagen, Hilden, Germany). Exons 4 and 5 of PRKCD were amplified with the specific primer pairs 5′-TGG GAA AAC  ACT GGT GCA GA-3′ and 5′-TCA CAC CCA GGA CAT GTT  GG-3′ and 5′-AGG CTG CCC TCA CTG ACC TTG TTC -3′  and 5′-TCA CGT GCA TAC ACT CTC TGA GCG C-3′, respectively, with the GoTaq DNA polymerase (#M3005, Promega

Targeted Mutagenesis and Transient Transfection
Site-directed mutagenesis was performed on the PRKCD-WT pCMV6 plasmid (#RC221652, OriGene) with the Pfu Ultra II Fusion HS DNA (#600,674, Agilent) polymerase as previously described [5]. The DNA was then digested with DpnI (#R0176L, New England Biolabs). PCR products were purified with the NucleoSpin Gel and PCR Clean-up kit (#740,609.50, Machery-Nagel). For the deletion of exon 5, blunting and ligation were performed with the Quick Blunting (#E1201S, New England Biolabs) and Quick Ligation (#M2200S, New England Biolabs) kits. Plasmids were amplified in NEB-10 β competent Escherichia coli cells (#C3019H, New England Biolabs) and purified with the QIAprep Spin Miniprep kit (#27106X4, Qiagen). HEK293T cells were transiently transfected with 1 µg of plasmid DNA in the presence of X-tremeGene9 DNA transfection reagent (#6,365,809,001, Merck), in accordance with the manufacturer's instructions.

TOPO-TA Cloning
Total RNA was extracted with the Quick-RNA Microprep Kit (#R1051, Zymo Research) from monocyte-derived macrophages (MDMs) from the patient and a healthy, unrelated control. RNA was transcribed to generate cDNA with the SuperScript II Reverse Transcriptase (#18,064,014, Thermo Fisher Scientific) and full-length PRKCD was amplified with the GoTaq DNA Polymerase (Promega). PCR products were inserted into the pCR-4 TOPO vector (#K457502, Thermo Fisher Scientific) and amplified in NEB-10 β competent E. coli cells (New England Biolabs). We analyzed 60 colonies for both the patient and control, using M13 forward and reverse primers to amplify the inserted cDNA.

RT-qPCR
Gene expression was analyzed on 50 ng of cDNA with the TaqMan Universal PCR Master Mix (Thermo Fisher Scientific) and Taq-Man probes for GUSB (#4326320E, Thermo Fisher Scientific) and PRKCD (#Hs01090047_m1 and #Hs01090051_m1, Thermo Fisher Scientific). Normalization was performed for each sample with GUSB (ΔcT) and values are expressed as 2 −ΔcT .

ROS Production Assays
ROS production by neutrophils and monocytes was analyzed by incubating whole blood with dihydrorhodamine 123 (DHR) and catalase in the presence or absence of PMA. Red blood cells were lysed with BD FACS lysing solution (#349,202, BD) and stained with the following antibodies: CD16 APC-H7 (#561,306, BD), CD66b V450 (#561,649, BD), and CD45 V500-C (#655,873, BD). Analysis was performed on a BD FACSCanto II (BD

Case Report
The patient (II.2) is a 5-year-old girl born to non-consanguineous parents originating from and living in the USA (Fig. 1a). Both parents and her older brother are healthy. She was vaccinated against diphtheria, tetanus and pertussis (DTaP), hepatitis A and B, Haemophilus influenzae Type b (Hib), and influenza and received a conjugated polysaccharide vaccine against Staphylococcus pneumoniae (PCV13) without adverse reaction. Vaccination with live vaccines against varicella and measles, mumps, and rubella (MMR) was also uneventful. She was not vaccinated with the live Bacillus Calmette-Guérin (BCG) vaccine. From the age of 2 years, the patient suffered from recurrent upper respiratory infections and multiple cervical lymphadenitis, presenting with a large submandibular lymph node. No fever, joint symptoms, malar rash, photosensitivity, or hepatosplenomegaly was reported. Fine-needle aspiration biopsy studies of the lymph node revealed an absence of granulomas, and cotrimoxazole-susceptible Burkholderia cepacia grew in culture. The lymphadenopathy resolved on oral cotrimoxazole therapy. One month later, the patient presented with another inflamed and enlarged lymph node, in the left lower anterior cervical chain (Fig. 1b). An excisional biopsy was performed and pathology studies reported an abscess without granuloma formation. Bacterial and fungal cultures were negative. Universal microbial PCR was positive for Mycobacterium lentiflavum. The patient was treated with cotrimoxazole, isoniazid, and rifampicin. Four months after the excision biopsy, her lymph nodes had returned to a minimal size and the patient was symptom-free. Whole-blood cell counts revealed normal counts of leukocytes, including polymorphonuclear neutrophils (PMNs) and monocytes, at several time points. Lymphocyte immunophenotyping by flow cytometry confirmed that the numbers of total T, B, and NK cells but decreased relative proportions  (Table 1; Fig. S1). Her proportion of CD21 low B cells was within normal range. At presentation, the patient tested negative for ANA, anti-dsDNA, anti-ribonucleoprotein and anti-Smith antibodies, and for autoantibodies against IFN-γ. An evaluation of the patient's circulating phagocytes with the DHR assay revealed abnormally low levels of ROS production by PMNs and monocytes in response to PMA stimulation (Fig. 1c). Based on the patient's clinical presentation and immunological findings, a diagnosis of CGD was suspected. However, a genetic analysis of CGD-associated genes -CYBA, CYBB, NCF1, NCF2, NCF4 and CYBC1 -was normal, and no copy-number variants were identified.

Identification of Compound Heterozygous PRKCD Variants
The patient was investigated by target panel sequencing covering 407 immune system-related genes. Two heterozygous candidate variants of PRKCD were identified and predicted to be deleterious. One of the variants was a single-nucleotide substitution (c.285C > A) predicted to lead to a premature stop codon (p.C95*); the other was a single-nucleotide substitution (c.376C > T) leading to a missense variant (p.D126Y). The presence of the two variants and an AR pattern of inheritance were confirmed by Sanger sequencing (Fig. 1d). The p.D126Y variant was inherited from the mother, whereas the p.C95* variant was probably de novo. A combined annotation-dependent depletion (CADD) score of 34 was predicted for p.D126Y and of 36 for p.C95*. These values are well above the mutation significance cutoff (MSC), with a 99% confidence interval of 25.1 for PRKCD. The variants are private and have never been reported in public databases such as gnomADv2.1.1 and v3.1.2, ExAC, BRAVO/Topmed, or ATAV or in patients with AR PKCδ deficiency (Fig. 1e). Target panel sequencing also identified a private intronic variant, c.376 + 6A > G, in the same allele as p.D126Y. This variant had a low CADD score (15.28), was not predicted to abolish splicing, and is, therefore, probably not disease-causing. However, biallelic variants of PRKCD have been reported to underlie susceptibility to infections similar to those observed in patients with CGD [5]. Taken together, these results suggest that the patient is suffering from AR PRKδ deficiency.

The Patient's Variants Impair Splicing and Are Loss-of-Expression
Both potentially deleterious variants are located in the N-terminal part of the PKCδ protein, between the C2-like and C1 domains (Fig. 1f). The p.D126Y variant is located at the end of exon 5 and is predicted, by GeneSplicer, NNSPLICE, and MaxEntScan, to disrupt the 5′ splice donor site. We studied the potential impact of the variants on mRNA splicing by performing TOPO-TA cloning and sequencing on cDNA from MDMs from the patient and a healthy control. Three different transcripts were detected for the patient, all of which were absent from the healthy control: 30% of the transcripts corresponded to the nonsense variant p.C95*, whereas 60% of transcripts displayed a skipping of exon 5 (c.316_376del) due to aberrant splicing and were predicted to encode a truncated protein (p.L106Ifs*17) ( Fig. 2a and  b). About 10% of transcripts corresponded to the p.D126Y variant, probably due to leaky splicing. Importantly, the variants were detected on different transcripts, confirming that the patient was compound heterozygous. We studied the impact of the different PRKCD transcripts by transiently transfecting HEK293T cells with plasmids encoding the wild-type (WT) or mutant PRKCD cDNAs, including a previously reported loss-of-function (LOF) mutant serving as negative control (p.K378M [9]). All constructs were C-terminally tagged with a Myc-DDK tag. Protein levels and basal activity were analyzed by western blotting whole-cell lysates. The WT, and the p.D126Y and p.K378M variants were produced at the expected molecular weight of 77 kDa, whereas the p.C95* and c.316_376del variants were undetectable with an antibody directed against the N-terminus of PKCδ (Fig. 2c). When an anti-DDK antibody was used, no protein was detected for the p.C95* and c.316_376del

Fig. 2 Functional analysis of the patient's variants by overexpression and investigation of primary cells. a Proportions of the various
PRKCD transcripts identified by TOPO-TA cloning on monocytederived macrophages (MDMs) from the patient (P) and a healthy control (Ctrl). b Schematic representation of the different splice variants found in P and their predicted consequences for the PKCδ protein. The positions of affected amino acids are indicated by red triangles. c Western blot of total cell extract from HEK293T cells not transfected (NT), or transfected with the pCMV6 empty vector (EV), wild-type (WT), or mutated PRKCD pCMV6. PKCδ phosphorylation was detected with antibodies against the phosphorylated p.T505 and p.S643 residues. d RT-qPCR for PRKCD on MDMs from a healthy control (Ctrl) and the patient (P), with probes spanning the junction between exons 3 and 4 (left) and exons 17 and 18 (right). GUSB was used for normalization (technical duplicates ± SD). e Western blot of total protein lysates isolated from polymorphonuclear neutrophils (PMNs) from a healthy control (Ctrl), the patient (P), her mother (M), and father (F). The asterisk indicates nonspecific bands. f Extracellular H 2 O 2 release, measured in the Amplex Red assay after PMA stimulation with or without IFN-γ priming, for MDMs from a local control (Ctrl), the patient (P), and her WT/WT father (F) (technical triplicates ± SD). g Western blot on whole-cell lysates from MDMs of the patient (P) and a healthy control (Ctrl) not stimulated (NS) or stimulated with PMA for 15 min. The phosphorylation of p40 phox was assessed with an antibody against the phosphorylated p.T154 site variants, excluding the possibility of a re-initiation of translation. Both variants are therefore loss-of-expression (LOE). Basal PKCδ activity was evaluated with antibodies against the autophosphorylated p.T505 and p.S643 residues. No autophosphorylation was observed at either of these residues, for the p.K378M negative control, whereas the p.D126Y variant presented normal autophosphorylation comparable to the WT protein.
These results indicate that most of the transcripts identified in this patient are LOE when overexpressed.

The Patient's Phagocytes Display a Loss of PKCδ Expression and an Impaired Oxidative Burst
We then assessed the impact of the variants on mRNA and protein production in primary and MDMs from the patient. Quantitative PCR on MDMs with two different probes flanking the patient's variants revealed that PRKCD mRNA levels in the patient were 75% decreased compared to those in a healthy control (Fig. 2d). PKCδ protein levels were analyzed by western blotting with whole-cell lysates from PMNs. No protein was detected in the cells of the patient, even at a lower molecular weight (Fig. 2e). Taken together, these results suggest that the patient's mRNA is largely subjected to nonsense-mediated mRNA decay, resulting in an absence of detectable protein. We completed our study of the impact of the patient's PKCδ variants on ROS production, by studying the NADPH oxidase activity of MDMs. Extracellular H 2 O 2 release was evaluated in MDMs from the patient, a healthy control and the wild-type father, who served as a travel control. The patient displayed much lower levels of H 2 O 2 production after PMA stimulation, in the presence or absence of IFN-γ priming, than the two controls (Fig. 2g). Furthermore, western blotting was performed on MDMs with and without PMA stimulation, to evaluate the PKCδ-dependent phosphorylation of p40 phox on the p.T154 residue. The absence of PKCδ protein was confirmed in the patient's MDMs, together with abnormally low levels of p40 phox phosphorylation on p.T154 after PMA stimulation (Fig. 2h). Taken together, these results indicate that the patient's circulating phagocytes and MDMs have impaired ROS production associated with a decrease in the phosphorylation of p40 phox .

Normal IFN-γ-Mediated Immunity in the Patient's Cells
The patient presented with M. lentiflavum infection. Susceptibility to mycobacterial infections has been associated with an impaired or disrupted IFN-γ mediated immunity [17][18][19]. We studied the impact of AR PKCδ deficiency on the production of, or response to IFN-γ in vitro. The amounts of IL-12p40 secreted by whole blood of the patient in response to BCG or BCG plus IFN-γ were within the range of healthy controls (Fig. 3a). The secretion of IFN-γ was evaluated in whole blood of the patient following BCG or BCG plus IL-12 stimulation and was comparable to those observed in healthy controls (Fig. 3b). Additionally, we used conventional flow cytometry to analyze the production of IFN-γ and TNF in the patient's primary T cells following stimulation with CD2/CD3/CD28 beads. The production of both cytokines in cells of the patient was comparable to those of the travel control (Fig. 3c). Collectively, these results indicate that AR PKCδ deficiency does not alter the global production or response to IFN-γ in vitro.

Discussion
This newly identified patient with AR PKCδ deficiency presented with clinical and immunological findings common to the other reported patients -high susceptibility to infectious disease and impaired ROS production by various circulating phagocytes and MDMs -but no autoimmunity. She had B. cepacia infection, which is commonly seen in CGD patients [20,21]. CGD can also be associated with the formation of granulomatous lesions [22]. While the patient we report here showed no granuloma formation, granulomatous inflammation in the lymph nodes has been reported in several PKCδ-deficient patients [5]. These observations highlight the overlap between the infectious and inflammatory phenotypes in patients with CGD and PKCδ deficiency. Interestingly, the patient also suffered from an infection caused by M. lentiflavum, a slow-growing environmental mycobacterium (EM) that is frequently isolated from water and soil samples [23]. Susceptibility to mycobacterial disease manifesting as BCG-itis after BCG vaccination has already been reported in four PKCδ-deficient patients [5], but this is the first such patient to present M. lentiflavum infection. Unlike B. cepacia infections, EM infections are rarely seen in patients with CGD and are more common in patients with Mendelian susceptibility to mycobacterial disease (MSMD) [19,24,25]. Patients with MSMD display a high susceptibility to EM due to genetic defects disrupting IFN-γ-mediated immunity [19]. Autoantibodies against IFN-γ are also known to underlie mycobacterial disease [26,27], but no such antibodies were found in this patient. While some studies in mice and human cell lines suggest a role of PKCδ in host defense against M. tuberculosis and the upregulation of IFN-γ-dependent genes after IFN-γ stimulation, the production and response to IFN-γ was normal in cells of the patient [28,29]. Susceptibility to mycobacterial disease has been previously associated with a defective ROS production in MDMs as shown in patients with X-linked recessive MSMD due to hypomorphic mutations in CYBB [30]. The defective ROS production by PKCδ-deficient MDMs could thus contribute to the susceptibility to mycobacterial infection observed in AR PKCδ deficiency. Strikingly, this patient presented no signs of autoimmunity at the time of diagnosis and had normal proportion of CD21 low B cells as well as no detectable auto-antibodies. Only two other PKCδ-deficient patients with no signs of autoimmunity have been described, one of whom had no history of infectious disease [5]. All three patients were below the age of 7 years, and the possibility of SLE development later in life cannot, therefore, be ruled out, but the mean age at onset of autoimmune manifestations in PKCδ-deficient patients is 3 years (± 2 years) [5,8,10,11]. In conclusion, PKCδ deficiency can manifest clinically as a syndromic form resembling CGD. We suggest that PRKCD should be included in targeted panel analyses for the investigation of young patients presenting with a CGD-like infectious phenotype and abnormal DHR results.

Declarations
Ethics Approval Written informed consent was obtained in accordance with local regulations, with approval from the institutional review board (IRB). The experiments were performed in the USA and France, in accordance with local regulations, and with the approval of the IRB Necker Hospital for Sick Children, France.

Consent to Participate
Written informed consent for participation in the study was obtained from the patient's parents.
Consent for Publication Consent to publish was obtained from the patient's parents. All the authors approved the final version of the manuscript.

Conflict of Interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.