Chronic Lymphocytic Leukemia

Loss-of-function mutations within the IL-2 inducible kinase ITK in patients with EBV-associated lymphoproliferative diseases

Abstract

The purpose of this study was the appraisal of the clinical and functional consequences of germline mutations within the gene for the IL-2 inducible T-cell kinase, ITK. Among patients with Epstein-Barr virus-driven lymphoproliferative disorders (EBV-LPD), negative for mutations in SH2D1A and XIAP (n=46), we identified two patients with R29H or D500T,F501L,M503X mutations, respectively. Human wild-type (wt) ITK, but none of the mutants, was able to rescue defective calcium flux in murine Itk−/− T cells. Pulse-chase experiments showed that ITK mutations lead to varying reductions of protein half-life from 25 to 69% as compared with wt ITK (107 min). The pleckstrin homology domain of wt ITK binds most prominently to phosphatidylinositol monophosphates (PI(3)P, PI(4)P, PI(5)P) and to lesser extend to its double or triple phosphorylated derivates (PIP2, PIP3), interactions which were dramatically reduced in the patient with the ITKR29H mutant. ITK mutations are distributed over the entire protein and include missense, nonsense and indel mutations, reminiscent of the situation in its sister kinase in B cells, Bruton's tyrosine kinase.

Introduction

The IL-2 inducible T-cell kinase (ITK) was first described in the early 90s when three groups independently showed that its mRNA becomes strongly upregulated in T cells stimulated by IL-2.1, 2, 3 Even though wild-type (wt) ITK is first and foremost expressed in T cells and mast cells only, its coding region is affected by acquired activating point mutations in diverse human cancers,4 not only in T-cell malignancies.

The T-cell kinase ITK can be regarded as ‘brother or sister kinase’ of Bruton's tyrosine kinase (BTK), which is specifically expressed in B cells and instrumental for their undisturbed development, differentiation and signaling. Inherited hemizygous germline BTK mutations were among the first molecularly defined primary immunodeficiency disorders and more than thousand different BTK mutations have been identified in patients with X-linked agammaglobulinemia including missense, nonsense or splice site mutations (see http://bioinf.uta.fi/BTKbase/).5, 6 In contrast, biallelic germline ITK mutations were more recently identified as the cause for severe Epstein-Barr virus -associated lymphoproliferative diseases (EBV-LPD). We identified two sisters whose T-cell immune response against EBV was severely impaired, leading to their death at the age of 7 and 10 years, respectively.7 Our clinical observation was in good accordance with a severely compromised T-cell function against EBV, but the degree of the ITK malfunction has not yet been proven at the molecular level. To some extend the function of ITK is characterized in mice, and knockout studies have revealed its paramount importance for proper T-cell development, T-cell receptor (TCR) signaling, cytokine release and regulation of differentiation.8, 9, 10, 11 Accordingly, Itk−/− mice show impaired immune response when infected with Toxoplasma gondii or Schistosoma mansoni eggs.12, 13

Itk−/− mice and the few patients with ITK mutations analyzed so far both show almost complete absence of invariant natural killer T cells, indicating its crucial role in the development and survival of this tiny, highly specialized cell population. However, the ubiquitous B-cell pathogen in the human population throughout the world, the Epstein Barr virus, does not infect mouse B cells. It is therefore an open question to what extent the Itk−/− mice resemble the clinical situation in humans with germline mutations in ITK. Thus, we wanted to know whether ITK mutations are specifically accompanied by severe disturbance of the immune control against EBV or if patients have clinical signs of broader, more general T-cell defects making them prone to other virus infections, autoimmunity or non-EBV-associated cancer. Moreover, it is unknown which types of mutations occur, where such mutations reside in the protein and what their functional and biochemical consequences are. In this study, we addressed some of these questions and functionally characterized two new (R29H and D500T,F501L,M503X) and the two known (R335W and Y588X) ITK mutations. Our data strongly argue for a loss-of-function mechanism although to a varying degree.

Materials and methods

The studies were reviewed and approved by the institutional review board (ethics committee) of Heinrich Heine University, Dusseldorf, Germany. The parents gave informed consent to carry out the investigations described below.

The microarray data have been deposited in NCBI's Gene Expression Omnibus14 and are accessible through GEO Series accession number GSE28200 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE28200).

Plasmid construction

The cDNA of human ITK (NM 005546) and ITKR335W was purchased from GenScript Inc. (Piscataway, NY, USA). Point mutation of ITK86G>A (ITKR29H) was accomplished by overlap-extension PCR,15 truncated fragments due to premature stop codons (ITK1497delT leading to ITKD500T,F501,LM503X and ITK1764C>G leading to ITKY588X) were generated by PCR. All of these constructs were inserted into derivatives of a bicistronic expression vector16 (pMC) used previously for stable expression of genes in human cells.17 Here the vector was modified to provide N-terminal fusion with a hemagglutinin (HA) epitope tag (YPYDVPDYA) linked by a five amino-acid spacer (PGPTR) or linked by an additional fusion to either the enhanced cyan or yellow fluorescent protein (CFP/YFP) with a seven amino-acid spacer (KQKVMNH).18 All new constructs were checked by DNA sequencing. To facilitate simultaneous visualization of ITK and ITKR29H in human cells, a tricistronic expression vector was generated, in which HA-CFP-ITKR29H was placed in the first and HA-YFP-ITK in the second cistron, followed by the puromycin-resistance gene Puromycin N-acetyl-transferase in the third cistron.

Calcium measurements and fluorescence-activated cell sorting analyses

Basically, calcium flux analyses in human T cells19 as well as immunophenotyping by flow cytometry7 were performed as described previously. Further details of these methods including complementation analyses in Itk knockout mice20 and antibody descriptions are provided as supplementary data.

Detailed protocols for all other methods are provided as supplementary information.

Results

Occurrence of ITK mutations in EBV-LPD patients

Since the initial discovery of ITK as a cause for an insufficient immune response to EBV in two girls,7 we sequenced all exons of the ITK gene in 46 patients with EBV-LPD (lymphoproliferative disorders), 33 with clinical ALPS, 40 with suspicion of a congenital Hemophagocytic syndrome (hemophagocytic lymphohistiocytosis, HLH) and 12 children with common variable immunodeficiency.

Although the latter cohorts only harbored some heterozygous base exchanges (ALPS: ITKV466V SNP rs17595896 or ITKV587I SNP rs56005928—one patient each, HLH no exchanges, common variable immunodeficiency: one heterozygous ITKY578H), we identified two patients with homozygous ITK mutations among the EBV-LPD cohort. Beneath mutations in the SH2 domain of ITK that we found in our two index patients7 (ITKR335W) or a 32-residue truncation of the catalytic domain (ITKY588X) in three patients from Palestine (patients no. 3–5),21 we now identified mutations in the pleckstrin homology (PH) domain (ITKR29H patient no. 6 originating from Morocco) and another premature stop codon in the kinase domain of ITK in a girl originating from India (ITKD500T,F501,L,M503X patient no. 7). Figures 1a and b show the sequencing results for these newly identified patients no. 6 and 7. Patient no. 6 originates from a consanguineous family, but as patient no. 7 was a street child, who had been adopted by a Finnish family, we were not able to reconstruct her family tree. An overview of all ITK mutations assigned to its protein domains is illustrated in Figure 1c.

Figure 1
figure1

Homozygous missense mutations in the ITK locus. (a) Relevant section of the ITK sequencing data for patient no. 6 (bottom left) compared with his parents (top left, data from mother shown). The corresponding amino acids are indicated above the codons. The pedigree of the affected boy (right) shows a consanguineous loop. Circles indicate females, squares indicate males, the ages of all siblings are given below. (b) Relevant sequencing data for patient no. 7 (bottom) compared with a healthy control (top). (c) Domain structure of ITK and its mutants in patients no. 1–7: ITKwt dark blue, ITKR335W green (according to patients no. 1 and 2), ITKY588X brown (patients no. 3–5), ITKR29H red (patient no. 6) and ITKD500T,F501,L,M503X cyan (patient no. 7). The domain boundaries40 according to the http://www.uniprot.org database v123 are indicated below the wt construct (top). The mutated domains in the four constructs below are emphasized in white.

Clinical features of patients with ITK mutation

Seven ITK-deficient patients from four unrelated families have been identified so far. For detailed patient characteristics please see Table 1, and the clinical reports of the Turkish7 and the Palestinian21 family. The patients between the age of 3 and 11 years presented with persistent symptoms of infectious mononucleosis including recurrent febrile episodes and lymphadenopathies accompanied by extremely high EBV viral load in the blood. Further, EBV-related symptoms like hepatosplenomegaly, bilineage cytopenia or autoimmune phenomena were observed in some of the patients at onset or during the course of the disease (Table 1). Remission after different treatment concepts including virostatic agents, corticosteroids, rituximab or chemotherapy was only brief, and without hematopoietic stem cell transplantation, which has been performed in two patients (patients no. 5 and 7), the disorder has shown to be fatal. Relapse of LPD or lymphoma was always accompanied or preceded by reactivation of the highly replicative EBV infection. Of note, a pulmonary involvement with large interstitial nodules was observed in the majority of patients, as it is known from other EBV-related pathologies like lymphomatoid granulomatosis. Because of intermittent respiratory symptoms, lungs were frequently examined by X-ray, CT or PET-CT scans (see Supplementary Figure S1 for patient no. 5) or even open lung biopsy (patient no. 7). We think that the inability to control EBV infection resulting in an initially polyclonal proliferation of EBV-infected B cells with consecutive malignant transformation seems to be the main pathology in ITK deficiency.

Table 1 Clinical and laboratory findings in the ITK-deficient patients

Common immunological features in ITK-deficient patients are a progressive hypogammaglobulinemia and a progressive loss of CD4+ T cells with a declining proportion of naive phenotype cells. Apart from EBV, patient no. 1 had BK virus and Pneumocystis jirovecii infection shortly before her death. Patients no. 4 and 5 revealed recurrent common respiratory viral infections in their history. Patient no. 6 repeatedly showed copies of CMV in serum and bronchoalveolar fluid. Invariant natural killer T cells were measured once in four of the patients and found to be low in all of them (Table 1 and Figure S1c).

No asymptomatic and EBV-seronegative ITK-deficient person has been identified so far. Therefore, we do not know whether ITK-deficient patients are prone to develop LPD and lymphoma or any immunological abnormalities in the absence of active EBV-infection.

Gene expression of human CD4+ cells with ITK mutation resembles murine Itk−/− cells

In order to perform additional experiments despite limited amounts of primary patient T lymphocytes, we transformed them to stable growth by Herpesvirus saimiri (HVS) infection.22 First, we approached the question of whether our HVS-transformed cell lines harboring the ITK deficiency mimic the situation in murine Itk−/− T cells by gene expression analyses using Affymetrix gene arrays (Affymetrix, Santa Clara, CA, USA). At first, we analyzed unstimulated HVS-transformed T cells of our index patient no. 2 (ITKR335W CD4+) compared with healthy control HVS cells. The average from cells at three different culturing time points led to a set of 1499 annotated genes (Welch's T-test, P<0.05) showing significant misregulation in the ITKR335W cells when compared with HVS-transformed cells from healthy volunteers (data not shown).

Next, we compared our data with corresponding gene expression analyses previously done on murine Itk knock-out CD4+ cells.23 We correlated both the data sets (our 1499 annotated genes with 1320 in mice) and generated a list of common differentially expressed genes consisting of only 26 downregulated and 20 upregulated genes in unstimulated CD4-positive cells (Figure 2a and Supplementary Table 1). A functional clustering based on Gene Ontology terms confirmed a high functional similarity for mice and human cells, particularly for genes involved in key intracellular signaling pathways, for example, the release of sequestered calcium ion into cytosol (Supplementary Figure S2). This is also supported by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses24 revealing, for example, four common genes having a part in the calcium signaling pathway (KEGG id hsa04020: P2RX7, F2R, GNA15 and CAMK4) along with NFAT being a part of the TCR signaling pathway (KEGG id hsa04660). The high scoring similarities between human HVS cell lines and the murine Itk−/− cells in terms of expression profiling, suggest that these transformed patient cell lines may serve as a suitable model for subsequent calcium flux analyses.

Figure 2
figure2

Characterization of ITK deficiency. (a) Correlation of the gene expression data for unstimulated, human HVS T cells harboring the ITKR335W mutation (CD4+, right) with corresponding data from Itk−/− mice (left). Fold changes for all common differentially expressed genes are illustrated as a heat map (downregulated red, upregulated green), matching genes are listed beside (right). Levels of intensities are illustrated by the lightness of red or green and a pale blue line. (b) HVS cells from ITK-deficient patients were subjected to calcium flux analyses (left) or analyzed for CD3 expression (right). Cells are from a healthy control person (ITKwt, dark blue), from patient no. 2 (ITKR335W, green), from patient no. 4 (ITKY588X, brown), from patient no. 6 (ITKR29H, red) and from patient no. 7 (ITKD500T,F501,L,M503X, cyan). Isotype staining for CD3 analyses (right) is shown in black. Stimulation by CD3 antibodies is indicated (open triangle); ionomycin was added as a loading control (filled triangle). (c) ITK mRNA abundance in the same HVS T cells. Quantitative PCR of at least two independent experiments normalized to wt are shown (s.e.m.). (d) Immunoblot analyses of endogenous ITK protein in these cells using N-terminus specific antibodies. Calculated MW of full-length ITK: 72 kDa; ITKY588X 68 kDa; ITKD500T,F501,L,M503X 58 kDa. (e) Calcium signaling of murine Itk−/− T cells rescued by human HA-ITK constructs of the different mutations depicted in Figure 1c (left) and corresponding green fluorescent protein expression (right). The empty vector served as a control in Itk−/− cells (black) as well as in cells from wt mice (magenta).

Mutations in ITK impede calcium mobilization

ITK is a crucial part of the TCR-mediated signaling cascade, which has been shown to be critically involved in calcium signaling in mouse T cells.25 Thus, we assayed calcium response after TCR stimulation by CD3 antibodies in HVS patient T lymphocytes. Figure 2b shows representative measurements from at least three independent experiments. Cells from patients harboring either the R29H mutation in the PH domain, or the R335W mutation in the SH2 domain as well as the cells from patient no. 4 (ITKY588X), showed a clearly reduced or nearly absent calcium response in all experiments. The cells from patient no. 7 (ITKD500T,F501,L,M503X) exhibited remarkable calcium flux ability—although with significant delay when compared with cells with wt ITK. In the light of the fact that the ITK mRNA of patient no. 7 was likely to be degraded due to nonsense-mediated mRNA decay and low protein stability (see below), it came as a surprise that her cells exhibited such a visible residual calcium response.

The premature stop codon in ITK from patient no. 7 results in low abundance of mRNA transcripts

Premature stop codons in an adequate distance of 50–55 nucleotides to an exon–exon boundary may lead to nonsense-mediated decay of transcripts.26 We used our HVS cells to analyze their ITK mRNA abundance by quantitative PCR. In the case of ITKY588X the premature stop codon in exon 16 is located only 27 nucleotides upstream of the last exon–exon junction, and the mRNA measurement did not show significant differences to wt ITK mRNA. The situation is different for the ITKD500T,F501,L,M503X mutant, where the premature stop codon in exon 14 is followed by three exons. We found a strong reduction of the ITK mRNA from patient no. 7 down to about 25%, most likely due to nonsense-mediated decay, whereas the mRNA levels in all other variants were not dramatically changed (Figure 2c). To exclude exon skipping events in ITK of patient no. 7 as a result of the mutation, we sequenced the relevant fragment of the corresponding cDNA (data not shown).

We next asked whether the protein variants are indeed expressed in patient cells. We again used our HVS model to show that the various ITK variants can be detected in most cases, except the largely truncated and presumably very weakly expressed ITK form of patient no. 7, which we could not detect behind the unspecific background of the antibody (Figure 2d). These data are in good accordance with our mRNA measurements from the same cells and with our previous observation of mutant R335W of patient no. 2.7

Human wt but not mutated ITK rescues the diminished TCR-mediated calcium flux in murine Itk−/− cells

To prove that the mutations identified in our patients really affect TCR-mediated calcium mobilization, we aimed to restore calcium flux by reintroducing the wt enzyme in ITK-deficient cells. Unfortunately, several attempts to do this assay in stably HVS-transformed cells failed. We did not succeed in infecting these HVS cells by using retroviral or lentiviral vector constructs. We therefore analyzed the ability to restore calcium signaling of our different ITK mutations (Figure 1c) in murine Itk−/− T lymphocytes,20 which have been shown to exhibit a clearly reduced but still residual (60%) calcium response compared with wt mice.27 Cells from isolated thymus were infected with retroviral vectors containing green fluorescent protein as a marker for ITK expression and analyzed for calcium response after TCR stimulation. Figure 3e shows representative measurements from at least three independent experiments. The ITK wt construct nicely rescued the diminished calcium response of the Itk−/− phenotype, up to the level of our wt control cells. By contrast, the ITK proteins corresponding to all patients did not show an enhanced calcium flux in Itk−/− mouse T cells.

Figure 3
figure3

Different mutations impede the protein half-life of ITK. Pulse-chase analyses of ITK mutants. (a) HEK 293 cells stably expressing only the HA peptide (lanes 1–5) or all five ITK constructs depicted in Figure 1c (lanes 6–30) were subjected to pulse-chase analyses. Cells were cultured in medium containing 35S-labeled amino acids for 20 min (‘pulse’) and ‘chased’ up to 30, 60, 120 and 240 min. HA-ITK proteins were immunoprecipitated, separated by SDS–polyacrylamide gel electrophoresis and exposed to X-ray films. (b) Normalized mean values of at least two experiments were plotted against time. Owing to ongoing translation and protein folding after time point zero (sometimes observed as an uptake of band intensities), we normalized our measurements to the values at 30 min. Calculated half-life times (t1/2) are shown (for details see Materials and Methods section and Supplementary Table 2); reductions compared with wt are given in brackets.

ITK mutations lead to significant reduced protein half-life

Our experiments in the HVS cells as well as our previous data in HEK293(ref. 7) or primary T cells,21 strongly suggest that the ITK mutations may influence its stability. Therefore, we aimed to determine the protein half-life of the ITK mutants more precisely and performed [35S]-Met/Cys pulse-chase analyses in HEK 293 cells, stably expressing HA-ITK or the respective mutants depicted in Figure 1c. Although the protein half-life measured with HA-tagged protein variants may not exactly reflect those of endogenous proteins in primary T cells, we confirmed at least a significant impact of the mutations on the stability (Figure 3a and diagram in Figure 3b; the corresponding data are provided in the Supplementary Table 2). We determined the half-life of wt ITK to be 107 min, whereas each mutant exhibited a destabilizing effect leading to a 25% reduction for the half-life of ITKR335W (80.5 min), 41% reduction for ITKY588X (63 min), 39% reduction for ITKR29H (65 min) and 69% reduction for ITKD500T,F501,L,M503X (33 min), respectively. These data also suggest that, in addition to its mRNA degradation, the ITK protein of patient no. 7 is subjected to rapid degradation, probably as a result of misfolding due to its incomplete C terminus.

The R29H mutation impedes the membrane targeting of ITK

Patient no. 6 harbors a point mutation in the pleckstrin homology domain of ITK, a domain that is involved in ITK membrane targeting.28 Therefore, we first asked for the impact of this mutation on the structure of the PH domain in silico. We modeled a structure of the ITK PH domain in inositol–phosphate complex based on the crystal structure of the PH domain of BTK,29 which is also a member of the TEC kinase family. This model enabled us to compare both ITK PHwt and PHR29H according to their interaction with the inositol–phosphate moiety. As shown in Figure 4a, multiple electrostatic contacts (250–330 pm) with the phosphate group of the inositol–phosphate formed by Arg-29 of ITK PHwt are reduced to only one single contact made by His-29 in PHR29H To verify this observation experimentally, we used a cell-free phospholipid affinity assay analyzing the binding of purified ITK PHwt and PHR29H proteins to phospholipids. Remarkably, ITK PHwt bound with the highest specificity to phosphatidylinositol monophosphates, which are a substantial part of cellular membranes,30 an interaction that was dramatically reduced in the case of the ITK PHR29H mutation (Figure 4b). An interaction with phosphatidylinositol di- and tri-phosphates, although rather weak, was detectable for ITK PHwt only, but not at all in the case of the R29H mutant (longer exposure not shown). The putative loss of the membrane targeting function in ITKR29H was subsequently also confirmed using the full-length protein in living cells, visualized as fusions with fluorescent proteins. N-terminally green fluorescent protein-tagged full-length ITK has previously been shown to maintain function and behavior.31 Figure 4d shows representative images of living HEK 293 cells stably coexpressing EYFP-ITK and ECFP-ITKR29H from the same polycistronic mRNA. The wt ITK shows a homogenous staining of the cellular membrane, whereas the mutated ITKR29H protein does not. We made the same observation when the human sarcoma cell line HT-1080 was used instead of HEK 293 (data not shown). Interestingly, an analog ITK mutant has been tested previously, carrying a cysteine at position 29 instead of histidine (ITKR29C). R29C also impedes membrane recruitment of ITK as demonstrated by exclusive staining of the cytoplasm excluding the cellular membranes.32 These results collectively suggest that ITKR29H lost its ability to recruit to the plasma membrane both in vitro and in vivo.

Figure 4
figure4

The R29H exchange impedes membrane recruitment of ITK. (a) Modeled PH domain of ITK wt (red) and R29H (green) in complex with inositol 3 monophosphate (orange). Protein structures are shown as cartoon, individual amino acids of ITK as well as ligand are shown as sticks. Dashed lines indicate the interaction between ligand and wt (red) or mutated (green) ITK, respectively. (b) Phosphoinositide affinity binding strips were probed with GST-tagged proteins comprising the PH and the SH3 domain of ITK (GST-ITKW239X), either harboring the wt R29 (wt, top) or the mutated residue (R29H, middle). The GST-tagged PH domain of PLCγ1 served as a positive control for PI(4,5)P2 (bottom). The strips were spotted with phosphoinositol-monophosphates (lanes 1–3), -diphosphates (lanes 4–6), -triphosphate (lane 7) or phosphatidic acid (PA, lane 8). (c) HEK 293 cells stably coexpressing ITK fused to YFP and ITKR29H fused to CFP were subjected to green fluorescent protein-directed immunoprecipitation. Precipitates were analyzed by SDS–polyacrylamide gel electrophoresis (8% gels), subjected to western blotting and probed with ITK antibodies. Migration distances of molecular weight marker proteins are indicated on the right margin. Both the proteins have a calculated molecular weight of about 100 kDa. (d) Representative examples of the same cells, visualized by confocal live-cell imaging: transmitted light (left), YFP-ITK (pseudo-colored in red, middle left). CFP-ITKR29H (green, middle right) and the merged image (right). Please see Materials and Methods section for further details.

Discussion

We show that germline mutations in ITK occur in both the sexes and in the children of diverse racial backgrounds. Our mutational screening data establish ITK deficiency as a separate molecular entity, which clinicians should consider in children with B cell LPD. The seven patients identified so far all presented primarily with massive EBV+ B-cell lymphoproliferation although its specific histological classification as B cell LPD, EBV+ Hodgkin's lymphoma or LCBL/lymphomatoid granulomatosis clearly varied. As seen in patients no. 4 and 6, autoimmunity may occur concomitantly and probably indicates an altered T-cell development, which has nicely been documented in Itk−/− mice, both in the CD4+ and CD8+ compartment.9, 10, 33, 34, 35, 36 We think that ITK deficiency is clinically clearly distinct from SAP and XIAP deficiency, two well-known X-linked primary immunodeficiencies that often present as fulminant infectious mononucleosis or HLH after primary infection with EBV.37, 38 ITK-deficient patients showed high EBV viral load, as it is sometimes seen in other primary T-cell deficiencies, for example, Wiskott–Aldrich syndrome or hypomorphic Artemis mutations. SAP- and XIAP-deficient patients are prone to the development of HLH, mostly with rather low EBV copy numbers or sometimes even in complete absence of EBV. SAP-deficient patients primarily develop Burkitt's lymphomas of the ileocecal region, whereas Hodgkin's lymphomas seem to be more prominent in the ITK-deficient cohort. XIAP deficiency may lead to chronic hemorrhagic colitis but not to lymphoma development.37 As no asymptomatic and EBV-seronegative ITK-deficient person has been identified yet, we do not know whether ITK-deficient patients are intrinsically susceptible to develop lymphoma or dysgammaglobulinemia also in the absence of EBV infection. All three disorders have in common that they show a strong reduction of invariant Natural killer T cells, corroborating their immune-modulatory role in EBV defense. However, the occurrence of additional viral infections point to a more general, not purely EBV-specific immune deficiency in our patients. This situation is reminiscent to what can be seen in ITK-deficient mice that are unable to control a number of different infectious agents.12, 13 When we consider BTK and ITK as close relatives expressed in B or T lymphocytes, respectively, the mutational spectrum appears to be similar.39 We found missense, nonsense and indel mutations spanning all protein domains of ITK. The different mutations clearly have different impact on the ITK function, but all data favor a loss-of-function mechanism.

The M503X mutant may likely be subjected to nonsense-mediated mRNA decay because it is located upstream of an exon–exon junction in an appropriate distance.26 The strongly reduced protein half-life when overexpressed in HEK293 cells and its virtual absence in HVS-transformed cells additionally argue for a very low intracellular abundance. Why M503X then exhibited a delayed but still significant calcium response after TCR stimulation, despite its low protein abundance, whereas the smaller truncated variant Y588X showed much less calcium signal? Both the proteins could still harbor a basic kinase activity, as the active site of ITK has been mapped to be upstream of the truncations (amino acids 376–435).40 The complex regulation of the second messenger Ca2+ depends on the relation of the individual TCR signaling components to each other, mode and strength of TCR activation, timing, and the calcium storage pools itself. One possible explanation could be the presence of Ca-flux inhibitory regions, for example, the Tyr-512 residue, located in the truncated region of patient no. 7. Tyr-512 has been shown to be phosphorylated by LCK,40, 41 thought as an important mechanism in ITK regulation. LCK is a specific target of the HVS tyrosine-kinase-interacting protein (Tip). Strong overexpression of Tip inhibits the rise of intracellular calcium concentration and this effect is largely dependent on its interaction with LCK.42 In HVS cells, the complex interplay between Tip, LCK and ITK may well be influenced when intracellular ITK level falls down under a critical threshold. Additionally, this low abundancy might also favor a redundant pathway in the HVS cells as assumed from several knock out studies.43 Of note, like all other ITK mutants, M503X is unable to rescue the defective calcium response of ITK-deficient mice, pointing to subtle species–specific differences. From a diagnostic point of view, genetic ITK mutation analyses may be more valid than functional assessment of Ca2+ response.

We show that the PH domain of wt ITK directly associates with the highest specificity to phosphoinositol monophosphates in vitro, but only to a significant lesser extend with their higher order di- or tri-phosphate derivates. In contrast, the PH domain of BTK preferentially binds to phosphatidylinositol(3,4,5)triphosphate over other phosphorylated forms, including monophosphates.44 The ITK R29H mutant exhibited dramatically reduced membrane binding in living cells, which is also consistent with a pronounced impact on its ability to induce sufficient calcium flux in human cells. Huang et al.45 recently showed by a set of very elegantly performed experiments how the complex interplay between freely soluble inositol polyphosphates (for example, IP4) and phosphatidylinositol(3,4,5)triphosphate influences recruitment of ITK to the membrane. We were able to confirm this observation for the ITK PHwt by utilizing a phospholipids array strip at concentrations 106-fold lower than that used by Huang et al. In addition, we identified phosphatidylinositol monophosphates as the most specific partners for the PH domain of ITK. Under inactive conditions, ITK remains as a monomer within the cytoplasm of a T cell.32, 46 Following TCR activation, ITK becomes activated and incorporated into a multi-protein signalosome by membrane binding via its PH domain.11 It seems likely but remains to be seen whether R29H or other mutations also differently affect ITK kinase activity.

We are aware of the fact that mutations of the homolog Arg-28 of BTK have frequently been shown to be involved in X-linked agammaglobulinemia. The BTKbase lists a total of 28 patients harboring mutations of Arg-28(ref. 5) of which 17 were R28H (Table 2). In addition, a patient with a frameshift indel mutation leading to a premature stop codon in BTK, Y627X,47 completely resembles the Y588X mutation in the ITK protein of patients no. 3–5. Another patient with X-linked agammaglobulinemia is reported to have a frameshift deletion that also leads to a premature stop codon, F540X, a mutation only two amino acids away from the corresponding premature stop-codon mutation of our patient no. 7.

Table 2 BTK homologies to the reported ITK mutations

Thus, we think that the large collection of mutations now assembled into the BTKbase since 1993 may serve as a paradigm for the ITK deficiency as well. The clinical course of patients with ITK deficiency may be more variable and they may be more difficult to treat than patients with Bruton's disease.

Accession codes

Accessions

Gene Expression Omnibus

References

  1. 1

    Tanaka N, Asao H, Ohtani K, Nakamura M, Sugamura K . A novel human tyrosine kinase gene inducible in T cells by interleukin 2. FEBS Lett 1993; 324: 1–5.

  2. 2

    Siliciano JD, Morrow TA, Desiderio SV . Itk, a T-cell-specific tyrosine kinase gene inducible by interleukin 2. Proc Natl Acad Sci USA 1992; 89: 11194–11198.

  3. 3

    Gibson S, Leung B, Squire JA, Hill M, Arima N, Goss P et al. Identification, cloning, and characterization of a novel human T-cell-specific tyrosine kinase located at the hematopoietin complex on chromosome 5q. Blood 1993; 82: 1561–1572.

  4. 4

    Greenman C, Stephens P, Smith R, Dalgliesh GL, Hunter C, Bignell G et al. Patterns of somatic mutation in human cancer genomes. Nature 2007; 446: 153–158.

  5. 5

    Väliaho J, Smith CIE, Vihinen M . Btkbase: the mutation database for X-linked agammaglobulinemia. Hum Mutat 2006; 27: 1209–1217.

  6. 6

    Lindvall JM, Blomberg KEM, Väliaho J, Vargas L, Heinonen JE, Berglöf A et al. Bruton's tyrosine kinase: cell biology, sequence conservation, mutation spectrum, siRNA modifications, and expression profiling. Immunol Rev 2005; 203: 200–215.

  7. 7

    Huck K, Feyen O, Niehues T, Rüschendorf F, Hübner N, Laws H et al. Girls homozygous for an Il-2-inducible T cell kinase mutation that leads to protein deficiency develop fatal EBV-associated lymphoproliferation. J Clin Invest 2009; 119: 1350–1358.

  8. 8

    Sahu N, Venegas AM, Jankovic D, Mitzner W, Gomez-Rodriguez J, Cannons JL et al. Selective expression rather than specific function of Txk and Itk regulate Th1 and Th2 responses. J Immunol 2008; 181: 6125–6131.

  9. 9

    Hu J, Sahu N, Walsh E, August A . Memory phenotype CD8+ T cells with innate function selectively develop in the absence of active Itk. Eur J Immunol 2007; 37: 2892–2899.

  10. 10

    Au-Yeung BB, Katzman SD, Fowell DJ . Cutting edge: Itk-dependent signals required for CD4+ T cells to exert, but not gain, Th2 effector function. J Immunol 2006; 176: 3895–3899.

  11. 11

    Andreotti AH, Schwartzberg PL, Joseph RE, Berg LJ . T-cell signaling regulated by the Tec family kinase, Itk. Cold Spring Harb Perspect Biol 2010; 2: a002287.

  12. 12

    Schaeffer EM, Yap GS, Lewis CM, Czar MJ, McVicar DW, Cheever AW et al. Mutation of Tec family kinases alters T helper cell differentiation. Nat Immunol 2001; 2: 1183–1188.

  13. 13

    Schaeffer EM, Debnath J, Yap G, McVicar D, Liao XC, Littman DR et al. Requirement for Tec kinases Rlk and Itk in T cell receptor signaling and immunity. Science 1999; 284: 638–641.

  14. 14

    Edgar R, Domrachev M, Lash AE . Gene expression omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 2002; 30: 207–210.

  15. 15

    Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR . Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 1989; 77: 51–59.

  16. 16

    Mielke C, Tümmler M, Schübeler D, von Hoegen I, Hauser H . Stabilized, long-term expression of heterodimeric proteins from tricistronic mRNA. Gene 2000; 254: 1–8.

  17. 17

    Linka RM, Porter ACG, Volkov A, Mielke C, Boege F, Christensen MO . C-terminal regions of topoisomerase IIalpha and IIbeta determine isoform-specific functioning of the enzymes in vivo. Nucleic Acids Res 2007; 35: 3810–3822.

  18. 18

    Hu C, Chinenov Y, Kerppola TK . Visualization of interactions among bZip and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol Cell 2002; 9: 789–798.

  19. 19

    Waibler Z, Sender LY, Merten C, Hartig R, Kliche S, Gunzer M et al. Signaling signatures and functional properties of anti-human CD28 superagonistic antibodies. PLoS ONE 2008; 3: e1708.

  20. 20

    Liao XC, Littman DR . Altered T cell receptor signaling and disrupted T cell development in mice lacking Itk. Immunity 1995; 3: 757–769.

  21. 21

    Stepensky P, Weintraub M, Yanir A, Revel-Vilk S, Krux F, Huck K et al. Il-2-inducible T-cell kinase deficiency: clinical presentation and therapeutic approach. Haematologica 2011; 96: 472–476.

  22. 22

    Biesinger B, Müller-Fleckenstein I, Simmer B, Lang G, Wittmann S, Platzer E et al. Stable growth transformation of human T lymphocytes by Herpesvirus saimiri. Proc Natl Acad Sci USA 1992; 89: 3116–3119.

  23. 23

    Blomberg KEM, Boucheron N, Lindvall JM, Yu L, Raberger J, Berglöf A et al. Transcriptional signatures of Itk-deficient CD3+, CD4+ and CD8+ T-cells. BMC Genomics 2009; 10: 233.

  24. 24

    Kanehisa M, Goto S, Furumichi M, Tanabe M, Hirakawa M . KEGG for representation and analysis of molecular networks involving diseases and drugs. Nucleic Acids Res 2010; 38 (Database issue): D355–D360.

  25. 25

    Liu KQ, Bunnell SC, Gurniak CB, Berg LJ . T cell receptor-initiated calcium release is uncoupled from capacitative calcium entry in Itk-deficient T cells. J Exp Med 1998; 187: 1721–1727.

  26. 26

    Nagy E, Maquat LE . A rule for termination-codon position within intron-containing genes: when nonsense affects RNA abundance. Trends Biochem Sci 1998; 23: 198–199.

  27. 27

    Readinger JA, Mueller KL, Venegas AM, Horai R, Schwartzberg PL . Tec kinases regulate T-lymphocyte development and function: new insights into the roles of Itk and Rlk/Txk. Immunol Rev 2009; 228: 93–114.

  28. 28

    Lemmon MA, Ferguson KM . Signal-dependent membrane targeting by pleckstrin homology (PH) domains. Biochem J 2000; 350 (Part 1): 1–18.

  29. 29

    Baraldi E, Djinovic Carugo K, Hyvönen M, Surdo PL, Riley AM, Potter BV et al. Structure of the PH domain from Bruton's tyrosine kinase in complex with inositol 1,3,4,5-tetrakisphosphate. Structure 1999; 7: 449–460.

  30. 30

    Lemmon MA . Membrane recognition by phospholipid-binding domains. Nat Rev Mol Cell Biol 2008; 9: 99–111.

  31. 31

    Woods ML, Kivens WJ, Adelsman MA, Qiu Y, August A, Shimizu Y . A novel function for the Tec family tyrosine kinase Itk in activation of beta 1 integrins by the T-cell receptor. EMBO J 2001; 20: 1232–1244.

  32. 32

    Qi Q, Sahu N, August A . Tec kinase Itk forms membrane clusters specifically in the vicinity of recruiting receptors. J Biol Chem 2006; 281: 38529–38534.

  33. 33

    Raberger J, Schebesta A, Sakaguchi S, Boucheron N, Blomberg KEM, Berglöf A et al. The transcriptional regulator Plzf induces the development of CD44 high memory phenotype T cells. Proc Natl Acad Sci USA 2008; 105: 17919–17924.

  34. 34

    Hu J, August A . Naive and innate memory phenotype CD4+ T cells have different requirements for active Itk for their development. J Immunol 2008; 180: 6544–6552.

  35. 35

    Horai R, Mueller KL, Handon RA, Cannons JL, Anderson SM, Kirby MR et al. Requirements for selection of conventional and innate T lymphocyte lineages. Immunity 2007; 27: 775–785.

  36. 36

    Berg LJ, Finkelstein LD, Lucas JA, Schwartzberg PL . Tec family kinases in T lymphocyte development and function. Annu Rev Immunol 2005; 23: 549–600.

  37. 37

    Pachlopnik Schmid J, Canioni D, Moshous D, Touzot F, Mahlaoui N, Hauck F et al. Clinical similarities and differences of patients with X-linked lymphoproliferative syndrome type 1 (XLP-1/SAP deficiency) versus type 2 (XLP-2/XIAP deficiency). Blood 2011; 117: 1522–1529.

  38. 38

    Latour S, Gish G, Helgason CD, Humphries RK, Pawson T, Veillette A . Regulation of SLAM-mediated signal transduction by SAP, the X-linked lymphoproliferative gene product. Nat Immunol 2001; 2: 681–690.

  39. 39

    Mohaamed AJ, Yu L, Bäckesjö C, Vargas L, Faryal R, Aints A et al. Bruton's tyrosine kinase (Btk): function, regulation, and transformation with special emphasis on the PH domain. Immunol Rev 2009; 228: 58–73.

  40. 40

    Brown K, Long JM, Vial SCM, Dedi N, Dunster NJ, Renwick SB et al. Crystal structures of interleukin-2 tyrosine kinase and their implications for the design of selective inhibitors. J Biol Chem 2004; 279: 18727–18732.

  41. 41

    Heyeck SD, Wilcox HM, Bunnell SC, Berg LJ . Lck phosphorylates the activation loop tyrosine of the Itk kinase domain and activates Itk kinase activity. J Biol Chem 1997; 272: 25401–25408.

  42. 42

    Cho N, Feng P, Lee S, Lee B, Liang X, Chang H et al. Inhibition of T cell receptor signal transduction by tyrosine kinase-interacting protein of Herpesvirus saimiri. J Exp Med 2004; 200: 681–687.

  43. 43

    Pearson H . Surviving a knockout blow. Nature 2002; 415: 8–9.

  44. 44

    Rameh LE, Arvidsson AK, Carraway KL, Couvillon AD, Rathbun G, Crompton A et al. A comparative analysis of the phosphoinositide binding specificity of pleckstrin homology domains. J Biol Chem 1997; 272: 22059–22066.

  45. 45

    Huang YH, Grasis JA, Miller AT, Xu R, Soonthornvacharin S, Andreotti AH et al. Positive regulation of Itk PH domain function by soluble Ip4. Science 2007; 316: 886–889.

  46. 46

    Qi Q, August A . The Tec family kinase Itk exists as a folded monomer in vivo. J Biol Chem 2009; 284: 29882–29892.

  47. 47

    Nomura K, Kanegane H, Karasuyama H, Tsukada S, Agematsu K, Murakami G et al. Genetic defect in human x-linked agammaglobulinemia impedes a maturational evolution of pro-B cells into a later stage of pre-B cells in the B-cell differentiation pathway. Blood 2000; 96: 610–617.

  48. 48

    Bienemann K, Iouannidou K, Schoenberg K, Krux F, Reuther S, Feyen O et al. INKT cell frequency in peripheral blood of caucasian children and adolescent: the absolute iNKT cell count is stable from birth to adulthood. Scand J Immunol 2011; 74: 406–411.

  49. 49

    Shearer WT, Rosenblatt HM, Gelman RS, Oyomopito R, Plaeger S, Stiehm ER et al. Lymphocyte subsets in healthy children from birth through 18 years of age: the pediatric aids clinical trials group p1009 study. J Allergy Clin Immunol 2003; 112: 973–980.

Download references

Acknowledgements

We thank all the patients, their parents and their treating physicians for sharing the clinical data with us. The cell lines HEK 293 and HT-1080, as well as the basic bicistronic expression vectors (pMC) and the coding sequences for ECFP and EYFP were kindly provided by F Boege, Dusseldorf, Germany. We thank S Furlan, S Bellert, M Oellers and U Wiczorek in Dusseldorf, as well as I Mueller-Fleckenstein and M Schmidt in Erlangen, for excellent technical assistance. The ALPS samples were kindly provided by G Lahr. The HLH samples were kindly provided by A Meindl and U zur Stadt. The EBV LPD samples were kindly provided by: J Richter, A Gennery, V Hazar, D K Uygun, P Vorwerk, E Mejstrikova, D Schwabe, M Seidel, C Prada, T Bernig, H von Bernuth, R Schneppenheim, S Choo, M van der Burg, P S Palacin and H Ören. This work was generously supported by the Elterninitative ‘Kinderkrebsklinik e.V.’. SLR, RDv and MRA were supported by the grants of the research committee of the medical faculty of the Heinrich-Heine-University Dusseldorf and the E-Rare project NSEuroNet. AH was supported by the Deutsche Forschungsgemeinschaft (He 2526/7-2). AB was also supported by Grants from the BMBF and the German-Israeli-Foundation.

Author information

Correspondence to A Borkhardt.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Linka, R., Risse, S., Bienemann, K. et al. Loss-of-function mutations within the IL-2 inducible kinase ITK in patients with EBV-associated lymphoproliferative diseases. Leukemia 26, 963–971 (2012). https://doi.org/10.1038/leu.2011.371

Download citation

Keywords

  • kinase
  • ITK
  • loss-of-function
  • primary immunodeficiency
  • EBV
  • lymphoma

Further reading