Abstract
Nijmegen breakage syndrome (NBS) is a rare autosomal recessive disorder characterized by spontaneous chromosomal instability with predisposition to immunodeficiency and cancer. In order to assess the cellular basis of the compromised immune response of NBS patients, the distribution of functionally distinct lymphocyte subsets in peripheral blood was evaluated by means of double-colour flow cytometry. The study involved the 36 lymphopenic patients with a total lymphocyte count ≤1500 µl (group A) and seven patients (group B) having the absolute lymphocyte count comparable with the age-matched controls (≥3000 µl). Regardless of the total lymphocyte count the NBS patients showed: (1) profound deficiency of CD4+ and CD3/CD8+ T cell subsets and up to fourfold increase in natural killer (NK) cells, almost lack of naive CD4+ T cells expressing CD45RA isoform, unchanged percentage of naive CD8+ cell subset (CD8/CD45RA+) but bearing the CD8 receptor of low density (CD8low); (2) normal expression of CD45RA isoform in the CD56+ lymphocyte subset, profound decrease in αβ but up to threefold increase in γδ-T cell-receptor (TCR)-positive T cells; (3) shift towards the memory phenotype in both CD4+ and CD8+ lymphocyte subpopulations expressing CD45RO isoform (over-expression of CD45RO in terms of both the fluorescence intensity for CD45RO isoform and the number of positive cells); and (4) an increase in fluorescence intensity for the CD45RA isoform in NK cells population. These results indicate either a failure in T cell regeneration in the thymic pathway (deficiency of naive CD4+ cells) and/or more dominant contribution of non-thymic pathways in lymphocyte renewal reflected by an increase in the population of CD4+ and CD8+ memory cells, γδ-TCR positive T as well as NK cell subsets.
Keywords: immunodeficiency, lymphocytes, NBS, receptors
INTRODUCTION
Nijmegen breakage syndrome (NBS) is an autosomal recessive disorder characterized by chromosomal rearrangements in cultured T lymphocytes with a hypersensitivity to ionizing radiation as well as a predisposition to malignancies, particularly those of lymphoid origin and severe, progressive microcephaly associated with different degrees of growth and intelligence retardation [1–4]. NBS is caused by mutations in the NBS1 gene located on chromosome 8q21 that encodes a novel protein called nibrin. This protein is a component of the hMre11/hRad50 protein complex that is involved in repairing DNA double-strand breaks [5]. Cytogenetic abnormalities reveal a high incidence of rearrangements in chromosome 7 and/or 14 with exchange points in chromosomes 7p13 and 7q34 as well as 14q11 and 14q32 [2,6]. Because these loci correlate with those encoding the T cell receptor (TCR) and the immunoglobulin heavy chain, they may give rise to developmental and functional lymphocyte abnormalities in NBS patients [6].
In many papers concentrating specifically on NBS [2–8], including our recently published data [9], defective Ig production was observed but the clinical presentation of immunodeficiency is different from hypo- or agammaglobulinaemia: NBS patients have opportunistic infections and are highly predisposed to develop malignancies [2–4]. T cell defects described originally in the 1980s by Seemanova [4], Weemeas [1,7] and Conley [8], such as deficiency in T cell numbers and their low proliferative response to mitogens, have also been confirmed by us in a single determination [3]. In this work, which evaluates a large group of patients, we explore further the nature of T cell defects by showing an imbalance in peripheral T cell subsets which indicate impaired T cell renewal and repertoire formation in NBS patients.
MATERIALS AND METHODS
Patients
We studied 43 NBS patients. The patients’ detailed characteristics are provided in Tables 1 and 2. The diagnosis was based on characteristic clinical symptoms and cytogenetic studies, confirmed subsequently by mutation analysis. All children tested here were homozygous for the common 657del5 mutation in the NBS1 gene, as described previously [10]. None of the patients suffered from acute infections or other diseases at the time of the study. Forty healthy, age-matched children served as normal controls.
Table 1.
General characteristics of NBS patients
| Number of subjects | 43 |
| Female (F)/male (M) ratio | 26/17 |
| Median age at examination: years (range) | 6·33 (2 months−18 years) |
| Mean head circumference at birth cm (range) | |
| F | 29·5 (28–31) |
| M | 31·5 (29–36) |
| Mean weight at birth (range) | |
| F | 2760 (1900–3600) |
| M | 3020 (2170–3950) |
| Microcephaly at examination | 43/43 |
| Infections | |
| Recurrent bronchopneumonia | 26 |
| Bronchiectasis | 16 |
| Sinusitis | 29 |
| Otitis media | 17 |
| Mastoiditis | 6 |
| Urinary tract infection | 13 |
| Diarrhoea (recurrent) | 14 |
| Persistent HBV,or HCV infection | 4/43, 3/43 |
| IVIG supplementation | 28 |
| Death caused by malignancy | 15 |
| Respiratory insufficiency | 2 |
| Other | 2 |
Table 2.
The presence of malignancies among NBS patients
| NBS | ||
|---|---|---|
| Malignancies | Group A n = 34 | Group B n = 7 |
| Malignancy: total † | 15/34 | 5/7 |
| Lymphoid origin | 14/15 | 5/5 |
| B cell (B-NHL; B-LBL/ALL) | 8*/14 | 3/5 |
| T cell (T-NHL; T-LBL/ALL | 5/14 | 1/5 |
| Hodgin's disease | 1/14 | 1/5 |
| Solid tumour (medulloblastoma) | 1/15 | – |
| Median age at diagnosis of malignancy: years (range) | 8·5 (4·0–24) | 8·0 (4·5–12·5) |
| Death due to cancer or complication of therapy | 10/15** | 5/5 |
| Median age at death: years (range) | 11·0 (4·0–20·5) | 8·0 (4·5–12·5) |
Two further cases suspected clinically (no autopsy performed).
Relapse of the same type of lymphoma in two patients,
five patients alive (under remission or therapy).
The study was approved by the the Ethics Committee of the Children's Memorial Health Institute and informed consent was obtained from all parents.
Lymphocyte isolation
Peripheral blood mononuclear cells (PBMC) of controls and NBS patients were isolated from heparinized blood by Ficoll/Hypaque density centrifugation. The cells were washed twice in cold phosphate buffered saline (PBS) and submitted immediately to cytofluorometric analysis.
Flow cytometry analysis
Freshly isolated PBMC were double-stained simultaneously with phycoerythrin (PE) and fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies (MoAbs), and analysed for direct double-coloured fluorescence by the EPICS XL-MCL fluorocytometer. The panel of FITC-labelled MoAbs included CD3 (Leu-4a), CD16 (Leu-11a) (Becton-Dickinson Immunocytometry System, Mountain View, CA, USA), CD45RA, CD45RO and pan-γ/δ MoAbs (Immunotech, Marseille, France). The panel of PE-conjugated MoAbs consisted of CD3, CD4 (Leu-3a), CD8 (Leu-2a) and CD56 (Leu-19) (Becton-Dickinson). Pairs of FITC/PE-conjugated MoAbs stained for CD45RA/CD4, CD45RO/CD4, CD45RA/CD8, CD45RO/CD8, CD16/CD56 and pan-γδ/CD3. CD45-FITC/CD14-PE (Simultest Leuco-Gate) as well as γ1-FITC γ2-PE (Simultest control; Becton-Dickinson) were included in each staining panel. The fluorescence intensity (FI) was calculated as a relative medial channel fluorescence for each surface molecule, as described previously [11,12]. The peak of the green fluorescence histogram of the beads was placed approximately in channel 78 by adjusting the green high voltage gain to 850 ± 30 and the peak of the red fluorescence histogram in channel 156 by adjusting the red high voltage gain to 700 ± 20.
Statistical analysis
According to their total lymphocyte count the 43 NBS patients were divided into group A (36 lymphopenic patients) and group B (seven patients with normal total lymphocyte count). The control group (C) consisted of 40 age-matched healthy children. The NBS patients (groups A and B) were compared to the control (group C). Accordingly, the P-levels resulting from comparisons of C versus A and C versus B were designed as P1 and P2, respectively. In a subanalysis of the NBS patients the results of group A were compared to B and P-level was described as P3. The absolute number of lymphocytes, the percentage of lymphocyte subpopulations and the FI for a given surface molecule were calculated. In each patient at least three consecutive measurements were performed over a period of 2 years. The data were pooled and evaluated statistically. Normality of distribution was checked by Kolmogorov–Smirnov and Shapiro–Wilkinson's tests, the log-normality by Kolmogorov–Smirnov and χ2 tests. The normally distributed data are expressed as their arithmetic means with standard deviations, the log-normally distributed variables in geometric means accompanied by 95% confidence intervals (CI). Comparisons between the different groups were performed with anova Kruskall–Wallis rank test using statistica (StatSoft) software.
RESULTS
General phenotypic characteristics of lymphocytes: expression of CD3, CD4, CD8 and CD56 receptors (Table 3)
Table 3.
Phenotypic characteristics of lymphocytes in controls and NBS patients
| Lymphocyte types | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| Lymphocytes | ||||
| % (average ± s.d.) | 45 ± 9 | 24 ± 8*** | 44 ± 10 | P < 0·001 |
| Abs.cell count/mm3 | ||||
| (geom. mean) | 3355 | 1365*** | 3391 | P < 0·001 |
| < 3094÷3638 >† | < 1206÷1545 > | < 2858÷4025 > | ||
| CD3+ | ||||
| % (average ± s.d.) | 77 ± 6 | 43 ± 13*** | 51 ± 19*** | P < 0·05 |
| FI | 22·1 ± 4·9 | 18·7 ± 5·7 | 20·2 ± 5·9 | |
| Abs.cell count/mm3 | 2683 ± 713 | 650 ± 237*** | 1670 ± 698*** | P < 0·001 |
| CD4+ | ||||
| % (average ± s.d.) | 48 ± 7 | 23 ± 7*** | 28 ± 19*** | P = 0·95 |
| FI (geom. mean) | 32 | 34 | 31 | |
| < 31·2÷34·3 > | < 31·7÷37·9 > | < 29·6÷32·5 > | ||
| Abs. cell count/mm3 | 1610 | 329*** | 835*** | P < 0·001 |
| < 1610÷1539 > | < 283÷383 > | < 5021÷389 > | ||
| CD8+ | ||||
| % (average ± s.d.) | 27 ± 5 | 23 ± 6 | 25 ± 12 | |
| FI | 42·2 ± 5·3 | 27·9 ± 7·6** | 28·1 ± 7·2** | P = 0·95 |
| Abs. cell count/mm3 | 911 ± 198 | 344 ± 140** | 921 ± 674 | P < 0·001 |
| CD4/CD8 ratio (geom. mean) | 1·8 | 0·8*** | 1·4* | P < 0·01 |
| < 1·7÷1·9 > | < 0·7÷0·9 > | < 0·2÷1·5 > | ||
| CD56+ | ||||
| % (average ± s.d.) | 8 ± 3 | 31 ± 10*** | 28 ± 12*** | P = 0·68 |
| FI | 7·05 ± 2·2 | 7·1 ± 1·8 | 7·6 ± 1·9 | |
| Abs. cell count/mm3 (geom. mean) | 245 | 407*** | 992*** | P < 0·001 |
| < 215÷278 > | < 332÷500 > | < 652÷1510 > | ||
P < 0·05,
P < 0·01,
P < 0·001, FI: fluorescence intensity, %: percentage of cells, bold type: statistically significant values,
95% CI (confidential interval). Comparisons: P1: C versus A, P2: C versus B, and P3: A versus B.
A profound deficiency of CD3- and CD4-bearing lymphocytes (percentage, absolute number) was seen in both the lymphopenic (group A) and non-lymphopenic (group B) of the NBS patients (P1 = P2 < 0·001) compared with controls, with no changes in FI for the CD3 and CD4 receptors. In contrast, the percentage of the total CD8+ cell population remained normal but the FI for CD8 receptor per cell was reduced in both NBS groups (P1 = P2 < 0·001) with no differences between them. The CD4+/CD8+ ratio was below 1·0 in group A (P1 < 0·001) and slightly above 1 in group B (P2 < 0·05) compared with controls, remaining significantly lower in group A than group B (P3 < 0·01). In contrast, the percentage of CD56 receptor-bearing natural killer (NK) cell population was highly elevated in both groups (P1 = P2 < 0·001) compared with controls, but without differences between the groups and with unchanged fluorescence intensity for the CD56 receptor.
Expression of αβ and γδ-TCR (Table 4)
Table 4.
Expression of αβ and γδ T-cell receptors in controls and NBS patients
| Lymphocyte receptors | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| αβ-TcR+ | ||||
| % (average ± s.d.) | 62 ± 9 | 33 ± 11*** | 32 ± 14** | P = 0·38 |
| FI | 3·6 ± 0·6 | 5·10 ± 1·44 | 4·08 ± 1·73 | |
| Abs. cell count/mm3 (geom. mean) | 2489 | 454*** | 919*** | |
| < 2210÷2804 >† | < 378÷546 > | < 617÷1371 > | P < 0·0001 | |
| γδ-TcR+ | ||||
| % | 4 | 11*** | 16*** | P = 0·19 |
| < 3·6÷4·3 > | < 9·1–14·3 > | < 9·9÷25·2 > | ||
| FI (average ± s.d.) | 4,9 ± 1,3 | 6·3 ± 1·9* | 5·5 ± 1·3* | P = 0·26 |
| Abs. cell count/mm3 | 163 ± 55 | 187 ± 108 | 631 ± 450*** | P < 0·0001 |
| αβ/γδ TcR+ ratio | ||||
| (average ± s.d.) | 20·0 ± 7·2 | 3·4 ± 2·1*** | 2·4 ± 1·5*** | P = 0·20 |
See Table 3 for explanation.
The percentage of αβ-TCR+ cells was equally reduced in both the lymphopenic and non-lymphopenic NBS patients (P1 = P2 < 0·001) compared with controls, but FI for the αβ receptor remained normal. In contrast, both the percentage of γδ-TCR-bearing cells as well as the FI for γδ-TCR were increased significantly in the NBS patients of groups A and B compared with controls (P1 < 0·001, P2 < 0·05, respectively). As a result, the αβ-TCR/γδ-TCR ratio was several times lower in the NBS patients than in controls (P1 = P2 < 0·001).
Expression of CD45RA and CD45RO isoforms in the CD4+ lymphocyte subset (Table 5)
Table 5.
Expression of CD45RA and CD45RO isoforms in the CD4+ lymphocyte subset in controls and NBS patients
| Lymphocyte receptors | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| CD45RA/CD4+ | ||||
| % (average ± s.d.) | 42 ± 6 | 6 ± 4*** | 6 ± 3*** | P = 0·83 |
| CD45RA+ | ||||
| FI (geom. mean) | 6,9 | 7,6 | 6,8 | |
| CD4+ | < 6·3÷7·6 >† | < 6·9÷8·3 > | < 8·4÷10·5 > | |
| FI | 31·2 | 14·4*** | 15·5*** | P = 0·89 |
| < 29·7÷32·8 > | < 12·7÷16·3 > | < 11,9÷17,6 > | ||
| Abs. cell count/mm3 | 1387 | 73*** | 208*** | P = 0·001 |
| < 1254÷1534 > | < 54÷99 > | < 133÷329 > | ||
| CD45RO/CD4+ | ||||
| % (average ± s.d.) | 7 ± 3 | 18 ± 6*** | 22·0 ± 9*** | P = 0·29 |
| CD45RO+ | ||||
| FI (geom.mean) | 4·3 | 10·2*** | 8·4*** | P = 0·23 |
| CD4+ | < 3·7÷4·4 > | < 8·9÷11·8 > | < 6·5÷10·9 > | |
| FI | 33·3 | 36·7 | 31·9 | |
| < 31·7÷35·6 > | < 34·3÷39·4 > | < 26·9÷38·1 >1 | ||
| Abs. cell count/mm3 | 245 | 254 | 619* | P = 0·1 |
| < 214÷280 > | < 220÷291 > | < 432÷888 > | ||
| CD45RA/CD4+ | 6·0 | 0·30*** | 0·45*** | P = 0·81 |
| CD45RO/CD4+ ratio (geom.mean) | < 5·1÷6·3 > | < 0·23÷0·42 > | < 0·17÷1·53 > | |
The percentage of naive CD4+ T cells expressing CD45RA was reduced greatly in both groups of the NBS patients compared with controls (P1 = P2 < 0·0001), with no differences between the groups and unchanged FI for the CD45RA isoform. However, FI for the CD4 receptor of this T cell subset was significantly lower than in controls (P1 = P2 < 0·001). Virtual lack of the circulating naive CD4+ cells in NBS was shown further by their minimal absolute number in the both groups of NBS patients compared with controls (P1 = P2 < 0·0001).
Conversely, the proportion of memory CD4+ cells expressing CD45RO was significantly higher (more than twofold) in the NBS patients than in controls (P1 = P2 < 0·001), with no significant differences between the NBS groups. Additionally, memory CD4+ cells of NBS patients showed approximately a twofold increase in FI for the CD45RO isoform (P1 = P2 < 0·001), showing no differences between the groups and normal FI for the CD4 receptor. The naive/memory CD4+ ratio was about 15 times lower in the NBS patients than in controls (P1 = P2 < 0·001), without significant differences between the groups.
Expression of CD45RA and CD45RO isoforms in the CD8+ lymphocyte subset (Table 6)
Table 6.
Expression of CD45RA and CD45RO isoforms in the CD8+ lymphocyte subset in controls and NBS patients
| Lymphocyte receptors | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| CD45RA/CD8+ | ||||
| % (average ± s.d.) | 25 ± 4 | 20 ± 6 | 21 ± 8 | |
| CD45RA+ | ||||
| FI (geom. mean) | 12·4 | 15·7 | 13·4 | |
| CD8+ | < 10·9÷13·2 > | < 14·1÷17·3 > | < 11·1÷16·3 > | |
| FI | 42·9 | 18·6*** | 16·2*** | |
| < 40·7÷45·2 > | < 17·3÷20·2 > | < 14·6÷18·8 > | P = 0·12 | |
| Abs. cell count/mm3 (geom. mean) | 888 | 299** | 652*** | P = 0·05 |
| < 823÷959 > | < 251÷366 > | < 387÷1098 > | ||
| CD45RO/CD8+ | ||||
| % (average ± s.d.) | 3 ± 1 | 13 ± 7*** | 15 ± 3*** | P = 0·07 |
| CD45RO+ | ||||
| FI (geom. mean) | 3·4 | 6·9*** | 5·2*** | P = 0·09 |
| CD8+ | < 3·0÷3·7 > | < 6·3÷7·7 > | < 4·0÷6·8 > | |
| FI | 41·1 | 38·7 | 37·1 | |
| < 39·9÷43·1 > | < 36·4÷41·7 > | < 32·7÷41·9 > | ||
| Abs. cell count/mm3 (geom. mean) | 114 | 158*** | 466*** | P = 0·001 |
| < 106÷124 > | < 134÷186 > | < 282÷771 > | ||
| CD45RA/CD8+ | 8·8 | 1·8*** | 1·7*** | P = 0·65 |
| CD45RO/CD8+ ratio | < 7·9÷9·9 > | < 1·5÷2·1 > | < 1·2÷1·8 > | |
In striking contrast to the CD4+ subset, only a slight and statistically non-significant decrease in the percentage of naive CD8+ cells expressing the CD45RA isoform with normal FI was found in both groups of NBS patients. In contrast, the FI for CD8 receptor (groups A and B) was diminished equally compared with controls (P1 = P2 < 0·001). Furthermore, the absolute number of naive CD8+ cells was highly reduced in group A (P1 < 0·001) and slightly but significantly decreased in group B (P2 < 0·05) compared with controls. In contrast to the naive CD8+ cell subset, both the percentage of memory CD8+ cell subpopulation and the FI of its CD45RO isoform were significantly higher in both groups of NBS patients compared with controls (P1 < 0·0001, P2 < 0·001, respectively), showing no differences between the groups and having unchanged FI for the CD8 receptor. The CD8+CD45RA+/CD8+CD45RO+ ratio was greatly diminished in NBS patients compared with controls (P1 = P2 < 0·001), with no differences between the groups.
Expression of CD45RA and CD45RO isoforms in the NK cell subset of NBS patients (Table 7)
Table 7.
Expression of CD45RA and CD45RO isoforms in the CD56+ lymphocyte subset in controls and NBS patients
| Lymphocyte surface receptors | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| CD45RA/CD56+ | ||||
| % (average ± s.d.) | 9 ± 3 | 35 ± 12*** | 27 ± 14*** | P = 0·18 |
| CD45RA+ | ||||
| FI | 11·9 ± 2·9 | 17·6 ± 4·5*** | 17·7 ± 3·5*** | P = 0·81 |
| FI | 5·6 ± 1·5 | 5·9 ± 1·8 | 4·5 ± 1·2 | |
| CD56+ | ||||
| Abs. cell count/mm3 (average ± SD) | 310 ± 98 | 532 ± 270*** | 957 ± 542*** | P < 0·001 |
| CD45RO/CD56+ | ||||
| % (average ± s.d.) | 1 ± 0·6 | 6 ± 3** | 7 ± 3** | P = 0·55 |
| CD45RO+ | ||||
| FI | 3·6 ± 1·2 | 4·9 ± 2·1 | 5·3 ± 2·7 | |
| CD56+ | ||||
| FI | 3·8 ± 1·5 | 3·4 ± 1·7 | 3·7 ± 1·2 | |
| Abs. cell count/mm3 | 29 | 75*** | 191*** | P < 0·0001 |
| (geom. mean) | < 24÷35 > | < 61÷93 > | < 93÷389 > | |
| CD45RA/CD56+ | ||||
| CD45RO/CD56+ ratio | ||||
| (average ± s.d.) | 11 ± 3 | 7 ± 3*** | 5 ± 2*** | P < 0·05 |
One of the most characteristic and striking phenotypical features of lymphocytes in NBS patients (groups A and B) was an increase in the NK cell subset and changes in the expression of its RA and RO isoforms. The percentage and absolute count of CD56+ cells expressing CD45RA and CD45RO isoforms and the FI of the CD45RA isoform were elevated in the NBS patients (P1 = P2 < 0·001) compared with controls, showing no changes in FI for both the CD56 receptor and CD45RO isoform. An increase in the CD45RO+/CD56+ population led to a significantly lower CD45RA+CD56+/CD45RO+CD56+ ratio in both groups of NBS patients (P1 = P2 < 0·0001), more pronounced in group B than group A (P3 < 0·05).
Expression pattern of CD8 in a population of CD3+ T and CD3– non-T cell subsets of NBS patients(Table 8).
Table 8.
Expression pattern of the CD8 receptor in CD3/CD8+, CD3–/CD8+ and CD8/CD56+ lymphocyte subsets of control and NBS patients
| Lymphocyte surface receptors | Control (C) n = 40 | NBS (A) n = 36 P1 | NBS (B) n = 7 P2 | P3 |
|---|---|---|---|---|
| CD3/CD8+ | ||||
| % (average ± SD) | 25 ± 4 | 19 ± 8*** | 20 ± 8*** | P = 0·67 |
| CD3+ | ||||
| FI | 22·1 ± 4·6 | 20·7 ± 4·5 | 20·1 ± 5·6 | P = 0·28 |
| CD8+ | ||||
| FI | 43·3 ± 6·8 | 40·9 ± 6·5 | 38·5 ± 8·5 | |
| Abs. cell count/mm3 | 852 ± 192 | 280 ± 154** | 667 ± 347*** | P < 0·05 |
| CD3–/CD8+ | ||||
| % (geom. mean) | 3 | 10*** | 7*** | P = 0·09 |
| CD8+ | < 3÷4 >† | < 8÷12 > | < 5÷9 > | |
| FI (average ± s.d.) | 9·3 ± 2·2 | 6·2 ± 2·1 | 7·7 ± 1·8 | P = 0·09 |
| Abs. cell count/mm3 (geom. mean) | 110 | 142*** | 230*** | P < 0·0001 |
| < 94÷128 > | < 117÷172 > | < 176÷300 > | ||
| CD8/CD56+ | ||||
| % (average ± SD) | 5 ± 2 | 13 ± 5*** | 11 ± 6*** | P = 0·36 |
| CD8+ | ||||
| FI | 13·6 ± 4·5 | 9·8 ± 2·3* | 9·3 ± 3·1* | P = 0·86 |
| CD56+ | ||||
| FI | 5·2 ± 2·1 | 6·7 ± 2·6 | 5·8 ± 2·4 | |
| Abs. cell count/mm3 | 160 | 160 | 293 | P < 0·05 |
| < 136÷187 > | < 128÷202 > | < 137÷627 > | ||
| CD56+ | ||||
| CD8/CD56+ ratio | ||||
| (average ± SD) | 2·9 ± 0·7 | 3·2 ± 0·9 | 3·1 ± 0·9 | |
The whole CD8+ cell population consists of at least two subsets: CD3+/CD8high T cell subset and CD3–/CD8low non-T cell subset, containing the CD8+/CD56+ NK lymphocytes [13]. Therefore, we evaluated the distribution of the CD8 receptor in: (1) the CD3+/CD8high T cell; (2) the CD3–/CD8low non-T cell; and (3) the CD8lowCD56+ NK cell subsets. The percentage and absolute numbers of the CD3+/CD8high T cell subset were diminished significantly in both groups A and B compared with controls (P1 = P2 < 0·001), with no changes in FI for both CD3 and CD8 receptors. In contrast, the CD3–/CD8low non-T cell subset was significantly elevated (percentage, absolute numbers) in NBS (P1 = P2 < 0·001) compared with controls, but FI for the CD8 receptor was much lower in NBS than in controls (P1 = P2 < 0·05). As mentioned above, the CD3–/CD8+ non-T cell subset corresponds to the subpopulation of CD8+/CD56+-bearing cells. Again, the proportion of this cell subset was elevated significantly in both NBS groups (P1 = P2 < 0·0001) compared with controls, with decreased FI for the CD8 receptor (P1 = P2 < 0·05) and normal FI for the CD56 receptor. Due to the high number of total CD56+ NK cell populations, the proportion of CD8+ cells within the CD8/CD56+ cell subset remained unchanged because the CD56+/CD56CD8+ cell ratio in NBS was comparable to controls.
DISCUSSION
Regardless of the total lymphocyte count the distribution of lymphocyte subsets in peripheral blood of NBS patients had several features in common: (1) a profound deficiency of the naive CD4+ subset with only a few cells of this type remaining but with CD4low; (2) a normal proportion of naive CD8/CD45RA+ cells but with CD8low; (3) a shift towards the memory phenotype in CD4+ and CD8+ lymphocyte subsets (over-expression of CD45RO isoform in terms of density, percentage and absolute number); and (4) an increase in the γδ TcR+ T cell subset and NK cell population.
These findings may reflect an imbalance between thymic-dependent and independent pathways of T cell renewal. During thymic selection the proper rearrangement of the T cell receptor (TCR) genes could be affected by mutations in the NBS1 gene encoding a protein involved in repairing double-strand breaks of DNA [5]. The NBS1 protein and histone γ-H2AX, one of the primary DNA damage sensors, have been identified recently in normal, double-positive thymocytes at the sites of TCR gene-specific breaks introduced during V(D)J recombination. Thus, surveillance of TCR recombination intermediates by NBS1 protein might prevent oncogenic translocations and control the selection process of thymocytes [14].
The expression of CD45RA and CD45RO isoforms of the CD45 common leucocyte antigen is regulated strictly during intrathymic T cell selection. Normally, CD3-negative, double-positive (DP) cortical thymocytes express CD45RO. Positively selected DP thymocytes become single CD4+ or CD8+ switch their CD45 isotype from CD45RO to CD45RA and leave the thymus as newly generated post-thymic lymphocytes. In contrast, the CD45RO+ thymocytes are eliminated by apoptosis [15]. Thus, CD45RO-bearing memory CD4+ and CD8+ cells could represent either immature lymphocytes that escaped apoptosis, or activated peripheral T cells (transition of RA to RO isoform). On the other hand, expression of the CD45RA isoform is not a unique marker of thymic origin because the T cell subset of putative thymic progeny may still contain the CD45RA ‘revertants’ (transition of CD4RO to CD45RA after antigen elimination), which are identical phenotypically to the naive T cell population with thymic progeny [16,17].
Selective deficiency of the naive CD4+ T cell subsets in NBS patients (CD56+ and CD8+ cells expressed CD45RA) might not imply altered expression or splicing of the CD45 gene, but suggests a failure in intrathymic CD4+ T cell selection and/or prevalence of extrathymic pathways in T cell generation. Children normally generate new T cells via thymic output of CD45RA+ naive T cell subsets, with only a minimal contribution of antigen-driven peripheral expansion of the CD45RO+ memory T cell pool [17]. Thus, the diminished naive CD45RA/CD4+ T cell subset with CD4low in both lymphopenic and non-lymphopenic NBS children possibly represents a restricted thymic output of immature naive CD4+ T cells rather than ‘revertants’ from the periphery, which should possess CD4 receptors of higher density [17]. The naive CD4+ cells proliferate well in response to mitogens [phytohaemagglutinin (PHA), concanavalin-A (ConA)][18], autologous and allogeneic cells [19] and act as a suppressor–inducer T cell subset that activates suppressor CD8+ T cells which in turn decrease IgG synthesis by B cells [20,21]. Thus, deficiency of the naive CD4+ T cell subset might contribute to the defects of cellular and humoral immune response described previously in NBS patients [2–4,6–9,22,23].
Thymus-independent T cell generation is due to peripheral expansion of mature T cells and/or extrathymic differentiation of bone marrow progenitors. These pathways predominate in advancing age as well as in impaired thymic function, as in NBS. Extrathymic expansion or differentiation takes place in lymph nodes, gut, bone marrow and liver, often leading to an excess of non-classical lymphocyte subsets such as TCR-γδ+ T cells or marrow-derived NK cells. The newly generated T cells express CD45ROhigh isoforms and have TCRs of restricted diversity. Their absolute number remains either reduced throughout life (possibly as in group A) or is elevated [15,16], which results probably from either a response to infectious agents or expansion of malignant lymphocyte clones (possibly as in group B). Other characteristics of T lymphocytes generated by peripheral expansion fit well with our results: first, the density of CD45RO isoforms in CD4+ and CD8+ memory cell subsets was high. Had they have survived intrathymic apoptosis and left the thymus as CD45RO+, they should have resembled the cells of immature phenotype-expressing CD4, CD8 receptors of low density. However, the density of these receptors and others (CD3, αβ-TCR, γδ-TCR) did not decrease. Secondly, double-positive T cells should be present in the periphery but they were absent (data not shown). Thirdly, the T cell population bearing the memory phenotype is possibly not fully active in the NBS patients. Normal memory CD4+ T cells exert a helper effect on B cells and participate in the induction of class I-restricted cytotoxic T cells protecting against infection. However, despite a relative excess of the memory CD4+ T cell population, which should enhance IgG synthesis, NBS patients had hypo- or dysgammaglobulinaemia [1–4,6–9] and were negative for the presence of specific serum IgG antibodies against Streptococcus pneumoniae and HBs-Ag vaccination [9], and did not respond to diphtheria and tetanus vaccination [23]. A deficiency in naive CD4+ T cells, a shift towards a memory phenotype in this T cell subset and functional T cell defects resembling those found in NBS have also been described in ataxia–telangiectasia (A–T) [24]. Therefore, it would be difficult at present to suggest some specific role for NBS1 protein in the development of specific T cell defects in NBS which would be different to those found in A–T.
The NK lymphocytes contain CD3–/CD8low cells [13]. Due to an excess of NK cells the CD8low subset of NBS patients represented about 50% of the whole CD8+ population versus 5% in controls. This might account for the decline in the value of the mean fluorescence intensity for CD8 receptors expressed in the total CD8+ population and in the naive CD8/CD45RA+ cell subset (both contain NK cells).
The proportion of the whole CD8+ population and the naive CD8+ cell subset were unchanged, regardless of the total lymphocyte count. In contrast, the absolute number of total CD8+ cells as well as of naive CD8/CD45RA+ and CD3/CD8+ subsets, was decreased more than twofold in group A compared to the control or group B. Thus, lymphopenia was caused by the loss of naive CD4+ and CD8+ T cells, but not of CD45RA/CD56+ or CD45RA/CD8low non-T cell subsets (part of CD56+ subpopulation).
The deletional mechanism which preferentially induces the loss of naive CD4+ and CD8+ T cell subsets (group A) or only the naive CD4+ T cell subset (group B) is not known, but may involve apoptosis. Fas-independent apoptosis has been described in naive CD4+ and CD8+ lymphocytes of lymphopenic CVID patients [25] and in cultured human T cell blasts following activation in interleukin (IL)-2-deprived medium [26]. The T cells of NBS patients are poor producers of IL-2, thus an increase in their susceptibility to apoptosis may be due to similar mechanisms. Preferential loss of the naive CD4+ cell subset (group B) may indicate a prevelance of Fas-dependent apoptosis of CD4+ T cells, because CD8+ T cells (relatively less affected) are eliminated mainly by TNF receptor triggering [27].
Impaired renewal of CD4+ and CD8+ T cell subsets could be compensated partially by an excess of NK and TcR-γδ T cell subsets. NK cells do not express a rearranged TcR and the NBS1 gene seems not to interfere with the maturation of NK cells. The ratio of CD45RA/CD56+ to CD45RO/CD56+ was decreased significantly due to over-expression of CD45RO isoform, thus at least some of the NK cells were activated in vivo. The CD45 isoforms in the NK population are critical in providing signals necessary for optimal cytotoxic activity [28]. Density and distribution of the CD16 receptor within the CD56+ NK population were similar to controls (data not shown). The CD16 receptor is engaged in cytokine synthesis and ADCC reaction [29]. As the distribution of the CD8 receptor in the NK cell subpopulation was also unaffected, NK lymphocytes in NBS patients are possibly functionally active. Elevation of NK cells in single cases of NBS patients have been reported by us [5] and others [30].
An increase in γδ-TCR-bearing T cells results possibly from extensive extrathymic T cell expansion or differentiation and reflects further the failure of T cell regeneration in thymic-dependent pathways [31]. Increased numbers of γδ-TCR T cells have been described in patients with A–T as an inherited defect of genetic recombination, in SCID patients, in the elderly, in the response to mycobacterial and other infectious antigens [32–34].
Thus, in NBS patients γδ-TCR T cells together with NK lymphocytes probably form a significant line of defence against infectious agents. These cells are highly active, poorly specific cytotoxic effectors capable of recognizing antigens in the context of relatively non-polymorphic antigen-presenting molecules not encoded in the MHC [35].
In summary, in connection with chromosomal breaks the changes in the distribution of functionally different lymphocyte subsets may give rise to impaired immunosurveillance that may predispose to malignancy found often in NBS. Very characteristic phenotypes of lymphocytes of NBS patients, together with clinical features, could be helpful in the diagnosis of NBS, which should be definitively verified subsequently by mutation analysis within the NBS1 gene.
REFERENCES
- 1.Weemaes CMR, Hustings TWJ, Scheres JMJC, van Munster PJJ, Bakkeren JAJM, Taalman RDFM. A new chromosomal instability disorder: the Nijmegen breakage syndrome. Acta Pediatr Scand. 1980;70:557–64. doi: 10.1111/j.1651-2227.1981.tb05740.x. [DOI] [PubMed] [Google Scholar]
- 2.Van der Burgt I, Chrzanowska KH, Smeets D, Weemaes CMR. Nijmegen breakage syndrome. J Med Genet. 1996;33:153–6. doi: 10.1136/jmg.33.2.153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chrzanowska KH, Kleijer WJ, Krajewska-Walasek M, et al. Eleven Polish patients with microcephaly, immunodeficiency, and chromosomal instability: the Nijmegen breakage syndrome. Am J Med Genet. 1995;57:462–71. doi: 10.1002/ajmg.1320570321. [DOI] [PubMed] [Google Scholar]
- 4.Seemanova E, Passarge E, Beneskova D, Houstek J, Kasal P, Sevcikova M. Familial microcephaly microcephaly with normal intelligence, immunodeficiency and risk for lymphoreticular malignancies: a new autosomal recessive disorder. Am J Med Genet. 1985;20:639–48. doi: 10.1002/ajmg.1320200410. [DOI] [PubMed] [Google Scholar]
- 5.Varon R, Vissinga C, Platzer M, et al. Nibrin, a novel DNA double-strand break protein, is mutated in Nijmegen breakage syndrome. Cell. 1998;93:467–76. doi: 10.1016/s0092-8674(00)81174-5. [DOI] [PubMed] [Google Scholar]
- 6.Wegner RD, Chrzanowska KH, Sperling K, Stumm M. Ataxia telangiectasia variants (Nijmegen breakage syndrome) In: Ochs HD, Smith CIE, Puck JM, editors. Immunodeficiency disorders. A molecular and genetic approach. Oxford: Oxford University Press; 1999. pp. 324–34. [Google Scholar]
- 7.Weemeas CMR, The TH, Van Munster JJ, Bakkeren JAJM. Antibody responses in vivo in chromosome instability syndromes with immunodeficiency. Clin Exp Immunol. 1984;57:529–34. [PMC free article] [PubMed] [Google Scholar]
- 8.Conley ME, Spinner NB, Emanuel BS, Novell PC, Nichols WW. A chromosome breakage syndrome with profound immunodeficiency. Blood. 1986;67:1251–6. [PubMed] [Google Scholar]
- 9.Gregorek H, Chrzanowska KH, Michańkiewicz J, Syczewska M, Madaliłski K. Heterogeneity of humoral immune abnormalities in children with Nijmegen breakage syndrome: an 8-year follow-up study in a single centre. Clin Exp Immunol. 2002;130:319–24. doi: 10.1046/j.1365-2249.2002.01971.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Varon R, Seemanova E, Chrzanowska K, et al. Clinical ascertainment of Nijmegen breakage syndrome (NBS) and prevalence of the major mutation, 657del5, in three Slav populations. Eur J Hum Genet. 2000;8:900–2. doi: 10.1038/sj.ejhg.5200554. [DOI] [PubMed] [Google Scholar]
- 11.Michałkiewicz J, Stachowski J, Barth C, Patzer J, Dzierżanowska D, Madaliński K. Effect of Pseudomonas aeruginosa exotoxin A on IFN-γ synthesis: expression of costimulatory molecules on monocytes and activity of NK cells. Immunol Lett. 1999;69:359–67. doi: 10.1016/s0165-2478(99)00121-2. [DOI] [PubMed] [Google Scholar]
- 12.Socha P, Michałkiewicz J, Stachowski J, Barth C, Socha J, Madalinski K. The expression of CD45RA isoform of CD45 common leukocyte antigen in CD4+ T lymphocytes in children with infantile cholestasis. Immunol Lett. 2001;75:179–84. doi: 10.1016/s0165-2478(00)00305-9. [DOI] [PubMed] [Google Scholar]
- 13.Lanier LL. NK cell receptors. Ann Rev Immunol. 1998;16:359–93. doi: 10.1146/annurev.immunol.16.1.359. [DOI] [PubMed] [Google Scholar]
- 14.Chen HT, Bhandola A, Difilippantonio MJ, et al. Response to RAG-mediated V (D) J cleavage by NBS1 and γ-H2X. Science. 2000;290:1962–4. doi: 10.1126/science.290.5498.1962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pilarski LM, Gillitzer R, Zola H, Shortman K, Scollay R. Definition of the thymic generative lineage by selective expression of high molecular weight isoforms of CD45 (T-200) Eur J Immunol. 1989;19:589–97. doi: 10.1002/eji.1830190403. [DOI] [PubMed] [Google Scholar]
- 16.Bell EB, Sparshott SM, Bunce C. CD4+ T-cell memory, CD45R subsets and the persistence of antigen-a unifying concept. Immunol Today. 1998;19:60–4. doi: 10.1016/s0167-5699(97)01211-5. [DOI] [PubMed] [Google Scholar]
- 17.Mackall C, Hakim TF, Grees RE. T-cell regeneration − all repertoires are not created equal. Immunol Today. 1997;18:245–51. doi: 10.1016/s0167-5699(97)81664-7. [DOI] [PubMed] [Google Scholar]
- 18.Akbar AN, Terry L, Timms A, Beverly PCL, Jannossy G. Loss of CD45 and gain of UCHL-1 reactivity is a feature of primed T cells. J Immunol. 1988;140:2171–8. [PubMed] [Google Scholar]
- 19.Dohlsten M, Sjogren HG, Carlsson R. Two subsets of human CD4+ T helper cell differing in kinetics and capacities to produce interleukin-2 and interferon-gamma can be defined by the Leu-18 and UCHL-1 monoclonal antibodies. Eur J Immunol. 1988;18:1173–8. doi: 10.1002/eji.1830180805. [DOI] [PubMed] [Google Scholar]
- 20.Morimoto C, Letvin NL, Distao JA, Aldrich WA, Schlossman SF. The isolation and characterization of the human suppressor–inducer T cell subset. J Immunol. 1988;138:1508–14. [PubMed] [Google Scholar]
- 21.Smith SH, Brown MH, Rowe D, Callard RE, Beverley CL. Functional subsets of human helper-inducer cells defined by a new monoclonal antibody, UCHL1. Immunology. 1986;58:63–70. [PMC free article] [PubMed] [Google Scholar]
- 22.Wegner RD, Metzger M, Hanefeld F. A new chromosome instability disorder confirmed by complementation studies. Clin Genet. 1988;33:20–32. [PubMed] [Google Scholar]
- 23.Der Kaloustian VM, Kleijer W, Booth A, et al. Possible new variant of Nijmegen breakage syndrome. Am J Med Genet. 1996;65:21–6. doi: 10.1002/(SICI)1096-8628(19961002)65:1<21::AID-AJMG3>3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
- 24.Paganelli R, Scala E, Scarselli E, et al. Selective deficiency of CD4+/CD45RA+ lymphocytes in patients with ataxia–telangiectasia. J Clin Immunol. 1992;12:84–91. doi: 10.1007/BF00918137. [DOI] [PubMed] [Google Scholar]
- 25.Iglesias I, Matamoros N, Raga S, Ferrer JM, Mila J. CD95 expression and function on lymphocyte subpopulations in common variable immunodeficiency (CVID) related to increased apoptosis. Clin Exp Immunol. 1999;117:138–46. doi: 10.1046/j.1365-2249.1999.00946.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hieronymus T, Blank N, Gruenke M, et al. CD95-independent mechanism of IL-2 deprivation-induced apoptosis in activated human lymphocytes. Cell Death Diff. 2000;7:538–47. doi: 10.1038/sj.cdd.4400684. [DOI] [PubMed] [Google Scholar]
- 27.Suzuki I, Fink PI. The dual functions of Fas ligand in the regulation of peripheral CD8+ and CD4+ T cells. PNAS. 2000;97:1707–12. doi: 10.1073/pnas.97.4.1707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Shen F, Xu X-L, Graf LH, Chong AS-F. CD45-cross linking stimulates IFN-γ production in NK cells. J Immunol. 1995;154:644–52. [PubMed] [Google Scholar]
- 29.Trinchieri G. Biology of natural killer cells. Adv Immunol. 1989;47:187–376. doi: 10.1016/S0065-2776(08)60664-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Green AJ, Yates JRW, Taylor AMR, et al. Severe microcephaly with normal intellectual development: the Nijmegen breakage syndrome. Arch Dis Child. 1995;73:431–4. doi: 10.1136/adc.73.5.431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Parker CM, Groh V, Band H, et al. Evidence for extrathymic changes in T cell receptor γ/δ repertoire. J Exp Med. 1990;171:1597–612. doi: 10.1084/jem.171.5.1597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Carbonari M, Cherchi M, Paganelli R, et al. Relative increase of T cels expressing the γδ rather than the αβ receptor in ataxia telangietasia. N Engl J Med. 1990;322:73–6. doi: 10.1056/NEJM199001113220201. [DOI] [PubMed] [Google Scholar]
- 33.Pfeffer K, Schoel B, Plesnila LA. Lectin-binding, protease-resistant mycobacterial ligand specifically activates V gamma 9+ human gamma delta T cells. J Immunol. 1992;148:575–83. [PubMed] [Google Scholar]
- 34.Modlin RL, Pirmez C, Hofman FM. Lymphocytes bearing antigen-specific γδ T-cell receptors accumulate in human infectious disease lesions. Nature. 1989;339:544–8. doi: 10.1038/339544a0. [DOI] [PubMed] [Google Scholar]
- 35.Schild H, Mavaddat N, Litzenberger C. The nature of major histocompatibility complex recognition by gamma-delta T cells. Cell. 1994;76:29–37. doi: 10.1016/0092-8674(94)90170-8. [DOI] [PubMed] [Google Scholar]