Dominant-activating germline mutations in the gene encoding the PI(3)K catalytic subunit p110δ result in T cell senescence and human immunodeficiency

Abstract

The p110δ subunit of phosphatidylinositol-3-OH kinase (PI(3)K) is selectively expressed in leukocytes and is critical for lymphocyte biology. Here we report fourteen patients from seven families who were heterozygous for three different germline, gain-of-function mutations in PIK3CD (which encodes p110δ). These patients presented with sinopulmonary infections, lymphadenopathy, nodular lymphoid hyperplasia and viremia due to cytomegalovirus (CMV) and/or Epstein-Barr virus (EBV). Strikingly, they had a substantial deficiency in naive T cells but an over-representation of senescent effector T cells. In vitro, T cells from patients exhibited increased phosphorylation of the kinase Akt and hyperactivation of the metabolic checkpoint kinase mTOR, enhanced glucose uptake and terminal effector differentiation. Notably, treatment with rapamycin to inhibit mTOR activity in vivo partially restored the abundance of naive T cells, largely 'rescued' the in vitro T cell defects and improved the clinical course.

Figures

Figure 1. A novel primary immunodeficiency characterized by T and B cell accumulation with lymphoid nodules and defective memory cell populations

(a) Endoscopic imaging reveals nodules in airways (top, patient E.1) and gastrointestinal tract (bottom, patient A.1) with lymphocytic infiltrates. Data are representative of 6 patients examined. (b) H & E staining of histological sections of a lymphoma biopsy from patient F.II.1 reveals a nodular sclerosis form of classical Hodgkin lymphoma (top); LMP-1 stain revealed EBV latent membrane antigen (brown coloration, 3,3’-Diaminobenzidine) (bottom). (c) Hematoxylin and eosin (H & E) staining (i) and immunoperoxidase staining for IgG (ii) and IgM (iii) on histological sections of lymph node from patient G.1. (d) The proportion (numbers in quadrants are percentages) of transitional (CD10+CD27−), naïve (CD10−CD27−) and memory (CD10−CD27+) cells within the B-cell population of a healthy control (Ctrl, top left) or patient B.III.1 (top right). Representative of 4 independent experiments. Each symbol in the bottom graphs shows cumulative data for memory (bottom left, P = 0.004 by two-tailed, unpaired t-test) and transitional (bottom right, P = 0.0005 by two-tailed, unpaired t-test) B cells represents an individual healthy control (Ctrl, n = 6) or patient (Pt, n = 7), and the horizontal bars represent means. Peripheral blood CD4+ T cell counts (e) and CD8+ T cell counts (f) in indicated patients as a function of age in years over time in all nine patients. (g) Thymidine incorporation assay for proliferation in response to the tetanus recall antigen, comparing normal controls (n = 9) and patients (n =9) and showing mean and standard deviation as horizontal lines. Data are cumulative from 9 independent experiments. SI = stimulation index of the value with antigen divided by that with no antigen. P = 0.01 by Mann-Whitney test. (h) Composite graph showing Tnaive (CD45RA+CD62L+), TCM (CD45RA−CD62L+), TEM (CD45RA−CD62L−), and TEMRA (CD45RA+CD62L−) populations among CD8+ T cells in all living patients (Pt, n = 11) compared to the clinical normal range derived from 50 healthy controls (Ctrl). (i) Representative flow cytometric dot plots of PBMCs from patient G.1 or a health control (Ctrl) stained for CD8+ versus CD4+ T cells (gated on lymphocytes, top) or CCR7 versus CD45RA gated on CD4+ (middle) or CD8+ (bottom) lymphocytes, showing Tnaive (CD45RA+CCR7+), TCM (CD45RA−CCR7+), TEM (CD45RA−CCR7−), and TEMRA (CD45RA+CCR7−) populations. Events falling into each quadrant are given as a percentage of the total. Data are representative of 5 independent experiments on multiple patients (see Supplementary Fig. 1D).

Figure 2. Heterozygous, gain-of-function mutation in PIK3CD in all affected individuals causes AKT hyperphosphorylation

(a) Schematic of p110δ protein domains (as described in the text) with patient mutations labeled with red lines. (b) Location of patient mutations (green spheres) shown on the structure (PDB 3HHM) of human p110α H1047R mutant (blue) in complex with p85α (pink). Positively charged amino acids along the membrane-binding face of p110 are shown as grey sticks. Location of the p110αH1047R mutation indicated by grey sticks. (c) Pedigrees of affected families with mutation-positive (black), unaffected (white), unscreened (grey), and deceased (diagonal) individuals. (d) Immunoblot on serum-starved, activated T cells using antibodies for p110δ, Akt, p-Akt (S473), and β-tubulin as indicated for patients D.II.1, D.II.2, and E.1 compared to three normal controls (Ctrl) with 10 min anti-CD3 stimulation (+) or not (−). Data are quantified in Supplementary Fig. 4A and are representative of p-Akt (S473) results from 4 independent experiments. (e) Immunoblot for Flag, p-Akt (S473), and total Akt on lysates from unactivated healthy control PBMCs overexpressing empty vector (EV), WT p110δ, or mutant p110δ 16 hr post-transfection. Data are representative of 3 independent experiments that were quantified in Supplementary Fig. 4F.

Figure 3. Patient PBMCs contain virus-specific CD8 T cells but fail to proliferate upon TCR stimulation in vitro due to terminal differentiation and senescence

(a) Flow cytometry analysis of EBV-specific tetramer staining of patient G.1 and control PBMCs for detection of EBV lytic and latent antigen-specific CD8+ T cells. Numbers in quadrants represent the percentage of total. Data are representative of 4 independent experiments. (b) CFSE dilution on gated CD8+ T cells for patients A.1 or B.III.1 compared to control 72 hrs after stimulation of PBMCs with the indicated mitogen. PHA = phytohemagglutinin. The percentage of cells that have divided is indicated. Data are representative of 8 patients examined in 6 independent experiments. (c) IL-2 secretion 48–72 hrs after activation of PBMCs from normals (n = 13 from combined experiments) and patients (n = 13 from combined experiments) with anti-CD3/CD28 Dynabeads. Horizontal line indicates mean value, and P = 0.0006 by Mann Whitney test. Data are representative of 4 independent experiments. (d) Intracellular flow cytometry for IFN-γ production and T-bet expression in activated CD8+ T cells stimulated with low-dose, immobilized anti-CD3. Numbers in quadrants represent the percentage of total. Data are quantified in Supplementary Fig. 5D and are representative of 3 independent experiments. (e) Granzyme B expression in activated CD8+ T cells from patients A.1 and G.1 compared to control. Combined data from independent experiments are quantified in Supplementary Fig. 5E. (f) Flow cytometric measurement of LAMP-1 cycling in unstimulated (US) or stimulated (as indicated) CD8+ T cells from patients (Pt, n = 4) compared to healthy controls (Ctrl, n = 3). Patients include A.1, E.1, F.II.1, and G.1. Horizontal line indicates mean value, and P = 0.01 obtained using the unpaired, two-tailed t-test. Data are representative of 3 independent experiments. (g) CD57 expression on CD4+ (left) or CD8+ (right) T cells from patient B.III.1 compared to healthy control (Ctrl). Data are representative of 4 patients examined in 3 independent experiments, which are quantified in Supplementary Fig. 5H.

Figure 4. Patient T cells exhibit normal TCR signaling responses but elevated mTOR activity and glucose uptake in vitro

(a) Calcium flux (as measured by the ratio of Fluo-4 to FuraRed) induced by anti-CD3 stimulation in T cell blasts from patients A.1, E.1, and F.II.1 compared to controls. The calcimycin ionophore shows is used as a positive control. Data are representative of 2 independent experiments examining cells from 5 different patients. (b) Confocal microscopy showing NF-κB p65 (green, left panels) translocation into the nucleus stained by 4’,6-Diamidino-2-Phenylindole, Dihydrochloride (DAPI, blue, colors merged right panels) in serum-starved T cell blasts following TCR stimulation (+) or not (−). Data are representative of 3 patients examined. (c) Phosflow analysis for phosphorylation of S6 (at residues S235 and S236) in unstimulated T cell blasts (fixed in complete medium) from patients G.1, D.II.1, and D.II.2 compared to controls. Data are representative of 3 independent experiments and are quantified in Supplementary Fig. 8A. (d) Phosflow analysis of S6 phosphorylation at S235, S236 at baseline (0 min) and 10 min or 20 min after stimulation with anti-CD3 plus protein A for patient G.1 (blue) and a healthy control (red). Data are representative of 2 independent experiments. (e) Phosflow analysis of S6 phosphorylation at residues S240 and S244 in unstimulated T cell blasts from patient F.II.1 treated for 20 min with DMSO, 10 µM IC87114 (p110δ inhibitor), or 50 nM rapamycin (mTOR inhibitor) compared to healthy control (Ctrl). Data are representative of 3 independent experiments. (f) Glucose uptake measured by flow cytometry analysis of the fluorescent deoxyglucose analog 2-NBDG in glucose-starved T cell blasts from patient G.1 compared to control (Ctrl). Data are representative of 4 experiments with a quantified analysis shown in Supplementary Fig. 8C. (g) Positron Emission Tomography (PET) scan of patient G.1 following administration of 18F-fludeoxyglucose analog shows glucose uptake most pronounced in the iliac chains and inguinal regions bilaterally. Data are representative of 3 patients examined.

Figure 5. In vivo rapamycin treatment improves in vitro and in vivo disease phenotypes

(a) CD4+ and CD8+ T cell counts from patient A.1 before and after rapamycin treatment. Dotted lines represent the range of T cell counts for normal controls. (b) PBMCs from patient A.1 before (UT, untreated) and after in vivo rapamycin treatment (Rapa) displaying CD8+ versus CD4+ (CD3+ gate, top) and CD62L versus CD45RA (CD4+ gate, middle, or CD8+ gate, bottom). Numbers in quadrants represent the percentage of total. (c)In vitro IL-2 production measured by Luminex after 24-hr activation (anti-CD3/CD28 Dynabeads) of PMBCs from an untreated normal control (n = 5) or patient A.1 before (UT, n = 3 independent measurements) rapamycin treatment (Rapa, left) and after (n = 2 independent measurements, right). Horizontal line indicates mean value, and P = 0.04 by the Mann-Whitney test. (d) Cell proliferation measured by CFSE dilution 72 hrs after activation (red) or not (black) of patient or control PBMCs with the indicated stimulus, shown before (UT, top) and after (bottom) in vivo treatment of patient A.1 with rapamycin (Rapa). The percentage of cells that have divided is shown.

Figure 6. Proposed model of effects of activating mutations (red asterisk) in p110δ

Augmented PI(3)K signaling results in increased glycolysis, causing CD8+ T cells to differentiate into effector cells that proliferate vigorously and then senesce. This results in impaired generation of long-lived memory CD8+ T cells and, consequently, poor control of EBV and CMV infection.