Protein kinase cδ deficiency causes mendelian systemic lupus erythematosus with B cell-defective apoptosis and hyperproliferation

Abstract

Objective: Systemic lupus erythematosus (SLE) is a prototype autoimmune disease that is assumed to occur via a complex interplay of environmental and genetic factors. Rare causes of monogenic SLE have been described, providing unique insights into fundamental mechanisms of immune tolerance. The aim of this study was to identify the cause of an autosomal-recessive form of SLE.

Methods: We studied 3 siblings with juvenile-onset SLE from 1 consanguineous kindred and used next-generation sequencing to identify mutations in the disease-associated gene. We performed extensive biochemical, immunologic, and functional assays to assess the impact of the identified mutations on B cell biology.

Results: We identified a homozygous missense mutation in PRKCD, encoding protein kinase δ (PKCδ), in all 3 affected siblings. Mutation of PRKCD resulted in reduced expression and activity of the encoded protein PKCδ (involved in the deletion of autoreactive B cells), leading to resistance to B cell receptor- and calcium-dependent apoptosis and increased B cell proliferation. Thus, as for mice deficient in PKCδ, which exhibit an SLE phenotype and B cell expansion, we observed an increased number of immature B cells in the affected family members and a developmental shift toward naive B cells with an immature phenotype.

Conclusion: Our findings indicate that PKCδ is crucial in regulating B cell tolerance and preventing self-reactivity in humans, and that PKCδ deficiency represents a novel genetic defect of apoptosis leading to SLE.

Figure 1

Clinical features and genetics analyses. A, Pedigree of the consanguineous family in which 3 siblings have manifestations of systemic lupus erythematosus. Solid symbols represent affected individuals; open symbols represent unaffected individuals. The diagonal line indicates that the patient is deceased. B, Clinical and histologic characteristics of subject V-1. Left, Photograph of the patient shows a photosensitive cutaneous rash. Right, Biopsy sections of the kidney show World Health Organization classes IV–V lupus nephritis (top) and C1q deposits (bottom). C, Sequence trace of the G510 residue in subject V-2, the unaffected sibling with the homozygous wild-type (WT) allele, and in subject V-1, showing the homozygous nonsynonymous variant (c.1528G>A; p.G510S [★]). D, Sequence alignment of the activation loop (boxed area) of protein kinase Cδ (PKCδ) from vertebrates to yeast. E, Schematic representation of the structure of PKCδ, illustrating important functional domains. The relative positions of relevant phosphorylation sites are indicated. F, Conservation of the G510S residue (boxed area) in PKCδ and other members of the AGC family of protein kinases. G, Structural modeling of PKCδ G510S. Because the x-ray structure of PKCδ is unavailable, we modeled the G510S substitution in PKCδ using the homologous PKCθ. X-ray structures of PKCθ (green) are compared with the structures of wild-type PKCδ (orange) (left) and G510S PKCδ (purple) (right). Mutation to a serine residue causes disruption of the activation loop that houses the G510 residue (thin black arrow).

Figure 2

Effect of the G510S mutation on the stability and function of PKCδ. A, Western blot showing weak phosphorylation of the T507, S645, and S662 sites in G510S PKCδ–transfected and wild-type PKCδ–transfected H157 and HEK 293T cells. Phosphorylation of the PKCδ substrates MARCKS and Mark2 was not increased by G510S. B, Western blot showing no induction of Mark2 phosphorylation in cells expressing G510S PKCδ, 5 minutes after phorbol myristate acetate (PMA) stimulation. C, Expression of wild-type PKCδ and G510S PKCδ constructs after transfection. Wild-type PKCδ expression persisted for up to 96 hours posttransfection, while the G510S mutant was undetectable at 96 hours and displayed significantly decreased levels at 72 hours. D, Effect of the G510S mutation on PRKCD mRNA integrity. Reverse transcription–polymerase chain reaction was performed to detect PRKCD and GAPDH in RNA extracted from lymphoblastoid cell lines (LCLs) established from affected siblings V-4 and V-1, their father (IV-2), and unaffected sibling (V-2) and from HEK 293T cells. E and F, Western blot analyses of LCL lysates harvested from family members, showing that patients with the G510S mutation express lower levels of PKCδ and reduced phosphorylation at the T507, S645, and S662 sites. G, Effect of PMA stimulation of LCLs from subjects V-2 and V-3. Stimulation did not induce PKCδ activation in homozygous patient V-3. See Figure 1 for other definitions.

Figure 3

Fluorescence-activated cell sorting (FACS) analysis of B cell subsets. A, Representative FACS plots showing increased percentages of transitional B cells in peripheral blood mononuclear cells isolated from subject V-4 (an affected sibling) but not in those obtained from either healthy donors (Ctl) or subject IV-1 (his heterozygous carrier mother). Transitional B cells were defined, according to the results of flow cytometry, as the CD24^highCD38^high population within CD19+ cells. B, Assessment of CD5 expression by naive (CD19+CD27−CD38^lowIgD+), pre-naive (CD19+ CD27−CD38^intermediateIgD+), and transitional (CD19+CD27−CD38^highIgD+) B cells from healthy donors, subject IV-1, and subject V-4. C, Decreased numbers of CD19+CD27+IgD+ and CD19+CD27+IgD− memory B cells in subject V-4 but not in healthy donors or subject IV-1. The percentage of IgD−CD27− memory B cells was unchanged, while the percentage of IgD+CD27− naive B cells was strongly increased, in subject V-4.

Figure 4

Functional characterization of mutation and phenotype reversal. A, Percentage of viable CD19+ B cells isolated from wild-type (WT) healthy donors (Ctl), subject V-4, his mother (IV-1), and wild-type patients with juvenile-onset systemic lupus erythematosus (JSLE). B cells were cultured for 24 hours in the presence of medium alone (open bars) or the Toll-like receptor 9 ligand CpG (solid bars). Values are the mean ± SEM (n = 16 healthy donors, 3 patients with juvenile-onset SLE, and 3 independent samples each from subjects IV-1 and V-4). B, Proliferation of CD19+ B cells in response to anti-IgM, CD40L, and CpG stimulation of B cells from healthy donors, subjects V-4 and IV-1, and patients with juvenile-onset SLE. Values are the mean ± SEM (n = 10 healthy donors, 4 patients with juvenile-onset SLE, and 3 independent samples each from subjects IV-1 and V-4). C, Percentages of viable CD19+ B cells isolated from wild-type healthy donors and subjects V-4 and IV-1 after culturing for 24 hours in the presence of medium alone (open bars) or thapsigargin (solid bars). Values are the mean ± SEM (n = 10 healthy donors and 2 independent samples each from subjects IV-1 and V-4. D and E, Percentages of specific apoptosis of lymphoblastoid cell lines established from subjects V-1, V-2, V-4, or IV-1 following culturing for 24 hours in the presence of anti-IgM (D) or thapsigargin (E). Statistical analyses were performed using analysis of variance with Dunnett’s correction. Symbols represent individual data points; horizontal lines and error bars show the mean ± SEM (n = 6–9 independent experiments each). F, Left, Percentages of specific apoptosis of LCLs established from subjects V-2 and V-1 that were either nontransduced (NT) or were transduced with protein kinease Cδ (PKCδ) G510S mutant construct (left) or PKCδ wild-type construct (right) and cultured for 24 hours with thapsigargin. Right, Western blot showing PKCδ expression in the corresponding LCLs. Results for transduced and nontransduced LCLs were compared by paired 2-tailed t-test. NS = not significant.