Chromosomal radiosensitivity in common variable immune deficiency

Mutation Research, 290 (1993)255-264 © 1993 Elsevier Science PublishersB.V. All rights reserved 0027-5107/93/$06.00

255

MUT 05330

Chromosomalradiosensitivityin common variable immune deficiency Igor Vo echovsk

a, David Scott b,,, Mansel R. Haeney c and David A.B. Webster d

a Karolinska Institute, Center for BioTechnology, 14157 Huddinge, Sweden, b Cancer Research Campaign Department of Cancer Genetics, Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Manchester M20 9BX, UK, c Department of Immunology, Hope Hospital, Salford M6 8HD, UK and a Clinical Research Centre, Northwick Park Hospital, Harrow, Middlesex HA1 3UJ, UK (Received31 December 1991) (Revisionreceived30 July 1993) (Accepted4 August 1993)

Keywords: Immunologicdeficiencysyndromes;Deficiencysyndromes,immunologic; Lymphoma;Chromosomeaberrations;X-Rays

Summary From more than 500 tumours reported in human primary immune deficiencies a majority has been observed in two disorders: ataxia telangiectasia (A-T) and common variable immune deficiency (CVID). Since both diseases have an increased risk of lymphomas/leukaemias and gastrointestinal tumours, suggesting a common risk factor, and the ceils derived from A-T patients exhibit an increased chromosomal radiosensitivity we analysed chromosome damage in the G2 lymphocytes of 24 CVID patients and 21 controls after X-irradiation in vitro. There was a significant difference in mean aberration yields between patients and controls. Three CVID patients had yields higher than the mean + 3SD of the controls. Six patients but only one control had yields higher than the mean + 2SD of controls. The patient with the highest chromosomal radiosensitivity subsequently developed a lymphoma. Repeat assays on the same blood sample, with a 24-h delay in setting up the second culture, showed as much variability for control donors as the variation between control donors although for CVID patients inter-individual variation was greater than the difference between results of repeat samples. There was a weak positive correlation between radiosensitivity and age of donor. Chromosomal radiosensitivity of five patients with X-linked hypogammaglobulinaemia was not different from healthy donors. The mean mitotic index (MI) for unirradiated samples from CVID patients was significantly lower than for controls and there was an inverse relationship between MI and aberration yields in the patients, but not in controls. We suggest that the defect in CVID patients that reduces response to mitogenic stimuli may have mechanism(s) in common with those involved in cellular repair processes.

Common variable immune deficiency (CVID) is a heterogeneous group of disorders characterised by depressed serum immunoglobulin lev-

* Correspondingauthor. Tel. 061 446 3126; Fax (061) 446

3109.

els, decreased ability to produce antibodies in response to antigenic challenge, frequent abnormalities in cellular immune functions assayed in vivo and in vitro, and clinically, by severe infections arising after a period of apparently normal immune function (Waldmann, 1978). CVID is usually a sporadic condition, but in the relatives

256 of a number of patients with CVID an immune deficiency has been observed (Fuchs et al., 1984; Vo~echovs~ et al., 1991). Since parental consanguinity was found in some families, an autosomal recessive inheritance has been proposed for some cases (Wollheim et al., 1964; Morrel et al., 1986; Vo~echovsk9 et al., 1991). CVID is associated with an increased risk of tumours (Spector et al., 1978; Kinlen et al., 1985; Cunningham-Rundles et al., 1987; Hermaszewski and Webster, 1993). Of more than 500 tumours so far reported in human primary immune deficiencies, the majority occur in two disorders: in ataxia telangiectasia (A-T) 160 tumours, and in CVID 130 tumours have been reported (Muller, 1990). The most frequent tumour types found in CVID are similar to those found in A-T (i.e. lymphomas/leukaemias and stomach tumours; Kinlen et al., 1985; Hecht and Hecht, 1990). This similar spectrum of tumours suggests a common risk factor. Although both CVID and A-T share some immune defects there are differences in the immunological abnormalities, in particular the rarity of panhypogammaglobulinaemia in A-T (Waldmann, 1982). Furthermore, no relationship between the degree of immune defect and the development of malignancy has been found in these disorders (Spector et al., 1982; Kersey et al., 1988). In addition to immune defects, A-T patients have other clinical symptoms including progressive cerebellar ataxia and oculocutaneous telangiectasia (Bridges and Harnden, 1982). Since A-T cells exhibit an increased number of chromosome aberrations after X-irradiation (Higurashi and Conen, 1973; Rary et al., 1974) and bleomycin (BLM; Taylor et al., 1979) compared to that found in similar cells from normal individuals, we previously used BLM treatment in vitro to assess the chromosomal damage in the lymphocytes of 15 CVID patients (Vofechovsk~ et al., 1989). We found that some of these patients had elevated levels of aberrations compared to healthy controls although not as pronounced as for A-T patients. The sensitive patients did not have any clinical symptoms suggesting A-T. One of the CVID patients with an increased level of BLM-induced chromosome damage developed a myelodysplastic syndrome followed by myelogenous leukaemia. We pro-

posed that the reason for the higher incidence of cancer in these primary immune deficiencies may not be the immune deficiency per se but a genomic instability, manifested by an increased level of chromosomal damage after suitable mutagenic stress in vitro (Vo~echovsk9 et al., 1989, 1990). In view of these findings and the existence of clinical and genetic "variant forms" of A-T or "A-T-like syndromes" (Ying and Decoteau, 1981; Hecht and Kaiser-McCaw, 1982; Fiorilli et al., 1985; Taylor et al., 1987; Jaspers et al., 1988; Yates et al., 1989; Curry et al., 1989; Ziv et al., 1989; Woods et al., 1990), we have now screened the lymphocytes of CVID patients for chromosomal sensitivity after X-irradiation in vitro with a modification of the method used at the National Cancer Institute (NCI), Bethesda, which appears to be the most sensitive in detecting many cancer-prone disorders (Sanford et al., 1989a,b; Parshad et al., 1990), including A-T homozygotes and heterozygotes (Sanford et ai., 1990). Materials and methods

Patients and controls

Heparinised blood samples (20 i u / m l , CP Pharmaceuticals Ltd. UK) from the patients and healthy volunteers (controls) were recieved coded from two UK centres. 24 patients with CVID (12 males aged 26-71 and 12 females aged 20-78) with lymphocytes responsive to phytohaemagglutinin (PHA), and 21 controls (10 males aged 21-62 and 11 females aged 27-62) were included in the study. In addition, 5 patients with X-linked hypogammaglobulinaemia (XLA) were tested since they suffer from chronic infectious complications similar to CVID patients. The diagnosis of CVID and XLA was based on W H O criteria (Scientific Group on Immunodeficiency, 1989). Cultures

Whole blood cultures were set up in T50 plastic culture flasks (Costar) containing RPMI 1640 medium (Flow Labs., UK) supplemented with 15% fetal calf serum (Advanced Protein Products, UK, batch AF1188), glutamine (Flow Laboratories, UK, at a final concentration of 0.58 mg/ml), 1% P H A (Wellcome) and antibiotics (penicillin 50 i u / m l , streptomycin 50 /zg/ml;

257 Gibco). T h e same batch of serum was used throughout the study. Each culture comprised 1.75 ml whole blood in 20 ml of medium prewarmed to 37°C and equilibrated in a 5 % CO 2, 95% air atmosphere. T h e flasks were returned to the gassed incubator with caps loose and after several hours were transferred to a hot room (37°C) with the caps tightened. All samples were incubated for 72 h in the dark. Before irradiation at 72 h the cultured cells were given a medium change (prewarmed to 37°C and pregassed) without the centrifugation step used at the NCI (Parshad et al., 1990) because we have found that centrifugation can lead to cell-cycle arrest (Scott et al., in preparation). Medium was carefully removed with a pipette without disturbing the cells at the bottom of the flask.

Irradiation and harvesting of the cells All procedures were performed at 37°C until the time of addition of fixative. Irradiation (0.25 Gy) was in culture flasks using a 300 kV Pantak unit at 10 mA ( H V T 2.3 m m Cu, dose rate 1.44 Gy/min). Our dose (0.25 Gy) was lower than that used at the NCI (58 R) because we found that the NCI dose produced an unacceptably high suppression of the mitotic index. However, our dose of 0.25 Gy produced a mean aberration yield in our control donors similar to that observed at the NCI after 58 R. 30 min after irradiation colcemid was added at a final concentration of 0.1 /~g/ml for 60 min and the cultures were transferred to centrifuge tubes and spun for 9 min at 150 g. A further 6 min was allowed for handling the tubes before hypotonic treatment (75 mM KCI for 20 min). For the second centrifugation and removal of supernatant an additional 15 min was allowed so that the total time between irradiation and fixation was 140 min. This time has been found at the NCI to give maximum discrimination between controls and cancer-prone individuals (Parshad et al., 1990; Sanford et al., 1990). Samples were fixed in glacial acetic acid and methanol (1:3, v:v) with 3 changes of fixative. T h e slides were stained in 5 % Giemsa solution (Gurr) at room temperature for 3 min. For randomly selected individuals a repeat blood culture was set up within 24 h of setting up the first culture, to determine the reproducibility

of the assay on a given blood sample. Blood was kept at 4°C between the repeats.

Evaluation of chromosomal damage and statistical analysis Scoring of aberrations was undertaken on coded slides. 100 metaphase cells were examined per person in the irradiated samples, 50 in the unirradiated cultures. The majority of aberrations in this assay were chromatid breaks (ctb) and chromatid gaps (ctg). Lesions were scored as ctb if the achromatic region was greater than the width of the chromatid or if the chromatid distal to the lesion was clearly misaligned with respect to the intact sister chromatid. Lesions of less than chromatid width, without misalignment were scored as ctg. To assess the possible influence of the quality of cytogenetic preparations on the number of aberrations scored, a marking system was created, taking into account the number of mitotic cells available for scoring, the degree of chromosome condensation, spreading of the chromosomes and chromatids, and stainability. Since, after decoding, the frequency of ctg scored per individual and ctg/ctb ratio increased with the quality of mitoses (r -- 0.5, p < 0.01) and no correlation was found between the ctb and the quality of mitotic cells we disregarded chromatid gaps in order to obtain a more reliable marker for expression of overall breakage per sample. All chromosomal aberrations except ctg were therefore combined and divided by the number of cells analysed to obtain the frequency of aberrations per cell (APC value). In order to determine whether the variability in the frequency of X-ray-induced chromosome damage was higher between the samples than within the samples (in repeated experiments using the same blood sample), a one-way analysis of variance was used (Table 1). T h e same test was applied to compare intra- and inter-experimental variability (Table 2). Wilcoxon's test was used to test for any systematic error in repeated measurements using the same blood sample. An unpaired t-test was used to compare aberration yields in the controls and the patients. The mitotic index (MI) of each unirradiated sample was determined by counting a total of

258

1000 mononuclear cells. A Mann-Whitney U-test was used to compare controls and patients.

TABLE

Results

T h e A P C values in 14 i n d e p e n d e n t e x p e r i m e n t s u s i n g blood s a m p l e s f r o m u p to 7 individuals p e r e x p e r i m e n t , b o t h f r o m p a t i e n t s a n d controls. O n l y t h o s e e x p e r i m e n t s involving two or m o r e s a m p l e s are shown.

The spontaneous level of chromosome aberrations Aberration levels in unirradiated lymphoeytes from the CVID patients did not differ significantly from those of healthy controls, the frequencies of chromosome-type aberrations per cell being 5.7 x 10 -3 and 3.3 x 10 -3 and of chromatid breaks per cell 6.9 x 10 -3 and 6.6 x 10 -3, respectively. The only dicentric chromosome in unirradiated samples was found to be in a female

TABLE

1

CHROMOSOMAL IN REPEATED SAMPLE FROM ALS Sample

DAMAGE AFTER X-IRRADIATION ASSAYS USING THE SAME BLOOD RANDOMLY SELECTED INDIVIDU-

Irradiation

1

Irradiation

2

2

COMPARISON OF TAL VARIABILITY

Expt. No.

INTER-

AND

INTRA-EXPERIMEN-

S a m p l e No. ( f r o m d i f f e r e n t individuals) 1

2

3

1 2 3

0.64 0.66 0.35

0.35 0.48 0.56

. . 0.80 0.59

4 5 6

0.50 0.53 0.60

0.95 0.60 0.64

7 8 9

0.57 0.53 0.58

10 11 12 13 14

4

5

6

7

0.64

0.92

0.74

0.57

0.74 0.86 0.57

0.78 0.54 -

0.60 0.50 -

0.85 0.40

1.19 -

0.60 0.63 0.41

0.47 . . 0.50

0.50 . . 0.59

0.52 . 0.87

0.55

-

0.40

0.71

0.61 0.50 0.59

0.68 0.66 0.73

0.50 0.52 0.44

0.67 0.78

0.78

0.63

0.40

0.46 0.54

0.48 0.75

0.48 . .

0.78 .

0.74 .

0.51

-

.

.

APC

Q

APC

Q

Control Control Control

0.60 0.52 0.41

3 4 1

0.60 0.54 0.61

3 3 2

Control a Control b Control

0.66 0.50 0.53

4 3 2

0.66 0.57 0.56

3 3 2

control. The patient with the highest level of chromosomal damage after lymphocyte irradiation in vitro did not show an increased level of spontaneous aberrations.

Control Control Control

0.44 0.55 0.73

1 1 3

0.48 0.63 0.48

2 1 2

Comparison of radiosensitivity within and between blood samples

CVID CVID CVID

0.50 0.54 0.71

2 3 2

0.68 0.64 0.67

4 3 3

CVID CVID CVID

0.57 0.78 0.78

3 2 4

0.46 0.77 0.73

4 3 4

CVID XLA

0.63 0.59

2 3

0.40 0.50

1 3

C V I D , p a t i e n t w i t h c o m m o n v a r i a b l e i m m u n e deficiency. XLA, patient with X-linked hypogammaglobulinaemia. A P C , f r e q u e n c y o f t o t a l a b e r r a t i o n s p e r cell ( e x c l u d i n g gaps). Q, quality of p r e p a r a t i o n s classified 1 - 5 , 1 is best. a m different b l o o d s a m p l e f r o m t h e s a m e i n d i v i d u a l was u s e d for t h e s e c o n d i r r a d i a t i o n , t h e interval b e t w e e n s a m p l i n g was 2 m o n t h s . b In t h e s e c o n d e x p e r i m e n t t h e n u m b e r of a n a l y s e d cells was 60.

The results from 17 duplicate experiments using the same blood samples selected randomly from controls and patients are shown in Table 1. Since no systematic change in APC values was found in duplicate measurements (Wilcoxon's test, p > 0.05), we conclude that storage of blood at 4°C for up to 24 h does not alter chromosomal radiosensitivity. After decoding the slides it was found that there were 9 controls, 7 CVID patients and 1 XLA patient. An analysis of variance showed that for controls the variability between blood samples was similar to that from duplicate experiments with the same sample (p > 0.5) so no conclusions about possible inter-individual differences are warranted. On the other hand, for patients there

259 was greater variability between than within samples ( p = 0.03) suggesting real differences between patients. The same statistical test for metaphase quality on all 17 samples showed that the variability between samples was greater than that for duplicates ( p < 0.01) indicating that the quality of cytogenetic preparations after X-irradiation is inherent to a particular sample.

Comparison experiments

of radiosensitivity

within and between

Because up to 7 blood samples were cultured and irradiated at the same time (within one experiment) we have addressed the question whether differences in chromosomal radiosensitivity between blood samples observed throughout the study may be attributable to the different conditions created in independent experiments (Table 2). One way analysis of variance for these data showed that the variability between independent experiments was not significantly higher than that within experiments. The differences observed between the blood samples analysed are therefore not likely to be explained by variable conditions from one experiment to another.

Chromosomal tients

radiosensitivity

in controls and pa-

There was a significant difference in chromosomal radiosensitivity between controls and CVID patients (Table 3, Figs. 1 and 2). There were 6 CVID patients (4 males and 2 females) with APC values (1.19, 0.95, 0.92, 0.86, 0.87 and 0.85) higher than the mean + 2SD of the APC value in controls (0.78). None of these patients and their close relatives had a neurological disorder or telangiectasias. In controls, only one male had an APC value (0.80) above the mean + 2SD. 3 CVID patients and no controls had APC values higher than the mean + 3SD of the controls (0.89). Comparison of all males vs. all females included in the study did not show any differences for any of the endpoints. 5 male patients with XLA did not differ from male controls and showed an almost identical range of APC values (0.48, 0.53, 0.59, 0.40, 0.75). The highest APC value of 1.19 was found in a male CVID patient, aged 40. This was about twice as high as the mean APC value of 3 con-

TABLE 3 COMPARISON OF CHROMOSOMAL DAMAGE AND MITOTIC INDEX IN LYMPHOCYTES OF THE CVID PATIENTS AND CONTROLS AFTER X-IRRADIATION IN VITRO CVID patients

Controls

Significance

Number of individuals 24 21 Age (mean+SD, years) 41 _+15 40 +12 NS APC value (mean+SD) 0.70+ 0.17 0.56+ 0.11 p<0.01 MI (mean+SD) 1.95+ 1.91 3.08+ 1.61 p<0.01 NS, not significant(p > 0.05). SD, standard deviation. trois in the same experiment. However, it was lower than the 3-fold increase in an A-T homozygote compared with a parallel control in an independent experiment using the same technique. The CVID patient had no family history of an immune deficiency or symptoms suggesting A-T, but he recently developed a T cell lymphoma and is receiving cytotoxic therapy.

Age effect There was a weak but significant positive correlation (r = 0.29, p < 0.05) between APC values

8

"~ 7 i-; 6 .-R

5

'~

4

~

a

E -' z

2

----

1

o

/ "~ "x

~ . . . . N 0.3 0.4 0.5 0.6 0.7 0.8 0.9

o

o"

1.0

1.1

1.2

1.3

APC value

Fig. 1. Frequency distribution of APC values in the CVID patients (open circles) and controls (solid circles). Individual APe values were assigned to categories with intervals of 0.1 (0.26-0.35, 0.36-0.45, 0.46-0.55, etc.). Data points are plotted at the midpointof these intervals (0.30, 0.40, 0.50, etc.)

260 1.20 1.10 1.00 0 0

0.90 .2= m > 0.80 0 a.

0 . . . . .

0

0 -0-

0 - (]D .

~II

.

.

.

0

.

.

00

.

.

.

.

.

.

.

.

.

.

.

0

0.70

0

0.60

o

0.50 0.40

V

0.30 10

0

•

• ~ °'Sg o %w % ~ •

•

o •

•

i

i

i

i

i

i

i

20

30

40

50

60

70

80

90

Age (years) Fig. 2. Relationship between age and APC value in the CVID patients (open circles), XLA patients (open triangles) and controls (solid circles). The horizontal line represents the mean + 2SD of the control group (see text).

and age if all 50 donors were included (Fig. 2). Considering the patients and controls separately, the former had a higher correlation coefficient but for neither group was the correlation statistically significant, possibly because of the relatively low number of cases. It should be pointed out that the hypersensitive CVID patients covered a wide age range (Fig. 2) and were not confined to the older age group.

Mitotic index (MI) The mean MI was significantly lower ( p < 0.01, Table 3) for CVID patients (1.95 _+ 1.91) than for controls (3.08_+ 1.61). The mean MI of the 6 CVID patients with APC values higher than the mean + 2SD of the controls was lower than the mean value for all CVID patients (i.e. 1.08_+ 0.47). There was an inverse relationship between MI and APC values for CVID patients which was on the borderline of statistical significance (correlation coefficient, r = 0.38; p = 0.067), but no correlation for controls (r = 0.05; p = 0.84). Discussion

Variability in the technique In our experience there is considerable inherent variability in the results obtained with this assay. There was more than a 2-fold range of

APC values (0.36-0.80) in our 21 control samples (Fig. 1) and a similar degree of variability even in duplicate assays with the same blood sample (Table 1). This variability is greater than that obtained at the NCI (Parshad et al., 1990 and personal communication) but is not a consequence of our omission of pre-irradiation centrifugation of the cells (Scott et al., in preparation). However, in spite of this variability we observed a significant difference in APC values between controls and CVID patients as we had previously found using bleomycin (BLM) although we had hoped to see a better discrimination with X-rays. The use of a single sampling time for such studies has previously been criticised on the grounds that differences in aberration yields between different samples might simply reflect differences in cell-cycle progression (Savage and Papworth, 1991). This criticism is based upon the assumption that changes in aberration frequencies in G 2 ceils represent differences in the intrinsic sensitivity of ceils to the induction of chromosomal damage at different stages of G e. However, it has now been convincingly demonstrated that the rapid decline in radiation-induced aberration yields that occurs between late and early G 2 represents the repair of the lesions that can give rise to the aberrations (Mozdarani and Bryant, 1989; MacLeod et al., 1990; Jayanth and Hittelman, 1991; MacLeod and Bryant, 1992; Scott and Bryant, 1992), cells in early G 2 having more time for repair before reaching metaphase. For example, MacLeod et al. (1990) have shown that a doubling of the duration of G e by reducing the temperature to 33°C had no effect on the aberration yield versus time curve.

Chromosomal radiosensitiuity in other disorders with genetically determined immune deficiency The identification of one CVID patient with a particularly high chromosomal radiosensitivity indicates that immune deficiency and radiosensitivity can occur without the typical clinical signs of A-T. A patient with selective IgA-deficiency, absence of neurological disease and cellular hypersensitivity of skin fibroblasts to the lethal effects of y-irradiation and other chemicals (Teo et al., 1983) has already been described (Webster et al.,

261 1992), but no chromosomal analyses were possible because of nonresponsiveness to mitogens. In this patient two missense mutations in the DNA ligase I gene have been found (Barnes et al., 1992). Although Taalman et al. (1987) did not find any systematic difference in chromosomal radiosensitivity when analysing the lymphocytes of 10 asymptomatic patients with selective IgA-deficiency and 6 controls (dose 1 Gy, cells irradiated in the S/G 2 phase of the cell cycle and sampled at 6 h), two of the patients showed a frequency of aberrations about twice as high as the mean of controls and more than 3SD above the mean control value. These findings are similar to our results in that only some patients with immune deficiency showed a hypersensitive response whereas the majority were indistinguishable from healthy controls. This may be due to heterogeneity in CVID (Wollheim et al., 1964; Waldmann, 1978; Fuchs et al., 1984; Vo[echovsk~ et al., 1991) and is compatible with the fact that only a proportion of CVID patients develop tumours (Spector et al., 1978; Kinlen et al., 1985; CunninghamRundles et al., 1987; Hermaszewski and Webster, 1993). It is noteworthy that CVID and selective IgA deficiency are probably genetically related disorders (Schaffer et al., 1989; Olerup et al., 1992). Clinical course and treatment of CVID patients with tumours Tolerance to antitumour therapy and survival is found to be poor in some CVID patients with lymphomas (Robison et al., 1987; Durham et al., 1987; Fesus et al., 1989; Vo[echovsk~ et al., 1990) but in others there is rapid response without relapse suggesting sensitivity of these tumours (Fesus et al., 1989; Vo~echovsloj et al., 1990). These observations are compatible with our findings that some, but not all, CVID patients show an elevated level of chromosome damage after X-irradiation. A-T patients, who exhibit extreme cellular radiosensitivity in vitro in cell killing or chromosome aberration assays (Bridges and Harnden, 1982), have shown devastating, lifethreatening tissue necrosis after conventional radiotherapy (Gotoff et al., 1967; Cunliffe et al., 1975) and have been treated with substantially

reduced doses with effective tumour control (Abadir and Hakami, 1983; Hart et al., 1987). We have recently found that breast and cervical cancer patients who exhibit excessive normal tissue damage after radiotherapy show, on average, significantly higher G 2 chromosomal radiosensitivity than controls (Jones et al., 1992; Scott et al., 1993). Peters (1990) has proposed that genetic variability in radiosensitivity be taken into account in clinical radiotherapy; reduction of doses may be appropriate for those CVID cases with cellular radiosensitivity. Reasons for increased chromosome damage in some CVID patients We suggest that the increased chromosomal damage after X-irradiation or BLM treatment (Vo~echovsk~ et al., 1989) in vitro is due to the gene(s) responsible for the CVID phenotype. This would be analogous to the A-T (Higurashi and Conen, 1973; Rary et al., 1974) or the scid (severe combined immune deficiency) mutation in mice (Fulop and Phillips, 1990). This offers an alternative explanation for the high incidence of tumours in CVID to the "immune surveillance against cancer" hypothesis, which has been challenged many times (e.g. Kripke et al., 1988). Cancer risk may be related to genomic instability, unmasked by a suitable mutagen treatment in vitro. This instability manifested at the cytogenetic level appears to result from a defect in DNA repair (Parshad et al., 1983; Gantt et al., 1987), which may be responsible both for the cancer susceptibility and hypogammaglobulinaemia. It is possible that the elevated chromosomal radiosensitivity in some CVID patients is simply a reflection of the decreased ratio of CD4/CD8 positive cells described in some CVID patients (Moretta et al., 1977). However, it has recently been shown that these lymphocyte subtypes are equally sensitive to the lethal effects of X-rays (Nakamura et al., 1990) and that there is a close relationship between survival and chromosomal damage in lymphocytes (Prosser et al., 1990). Relationship between chromosome damage and mitotic index The inverse relationship between aberration yields and mitotic index in CVID patients, which

262

was on the borderline of statistical significance, suggests that the defect in CVID patients that reduces response to mitogenic stimuli may have mechanism(s) in common with those involved in cellular repair processes. In support of such an idea are our recent observations on A-T lymphocytes cultured in parallel with controls in which 13/14 A-T patients showed a lower mitotic response and all A-T patients had higher aberration yields (Scott et al., unpublished). We were unable to undertake chromosome studies on 5 CVID patients because their lymphocytes failed to respond to PHA; these patients may also have had a repair defect. There is experimental evidence that DNA repair is important in the early stages of the triggering process for lymphocyte proliferation (Johnstone and Williams, 1982); this is supported by the complete failure of in vitro lymphocyte proliferation using a variety of stimuli in the patient with a DNA ligase I defect mentioned earlier (Webster et al., 1992).

Acknowledgements The work was supported by the Stephen Rutty Research Fund and the Cancer Research Campaign. Thanks are due to Mrs. Ann Spreadborough for mitotic index scoring.

References Abadir, R., and N. Hakami (1983) Ataxia telangiectasiawith cancer, An indication for reduced radiotherapy and chemotherapy doses, Br. J. Radiol.,56, 343-345. Barnes, D.E., A.E. Tomkinson,A.R. Lehmann, A.D.B. Webster and T. Lindahl(1992) Mutations in the DNA ligase I gene of an individualwith immunodeficiencies and cellular hypersensitivityto DNA-damagingcells, Cell, 69, 495-503. Bridges, B.A., and D.G. Harnden (1982)Ataxia-telangiectasia: A cellular and molecular link between cancer, neuropathologyand immune deficiency,Wiley, New York. Cunliffe,P.N., J.R. Mann, A.H. Cameron, K.D. Roberts and H.W.C. Ward (1975)Radiosensitivityin ataxia telangiectasia, Br. J. Radiol.,48, 374-376. Cunningham-Rundles,C., F.P. Siegal, S. Cunningham-Rundies and P. Lieberman (1987) Incidence of cancer in 98 patients with common variable immuodeficiency,J. Clin. Immunol.,7, 294-299. Curry, C.J., J. Tsai, H.T. Hutchinson,N.G.J. Jaspers, D. Wara and R.A. Gatti (1989)ATFrcsno:a phenotype linkingataxia telangiectasiawith the Nijmegenbreakage syndrome, Am. J. Hum. Genet., 45, 270-275.

Durham, J.C., D.S. Stephens, D. Rimland, V.H. Nassar and T.J. Spira (1987) Common variable hypogammaglobulinemia complicated by an unusual T-suppressor/cytotoxic cell lymphoma,Cancer, 59, 271-276. Fesus, S.M., F.B. Hagemeisterand J. Manning(1989)Hodgkin disease in a patient with common variable immunodeficiency, Am. J. Hematol., 32, 138-142. Fiorilli,M., A. Antonelli, G. Russo, M. Crescendi, M. Carbonari and P. Petrinelli(1985)Variant of ataxia telangiectasia with a low level radiosensitivity,Hum. Genet., 70, 274-277. Fuchs, H.B., L. Slater, H. Novey, K. Ong and S. Gupta (1984) Immunologicalanalysis in familial common variable immunodeficiency,Clin. Exp. Immunol.56, 29-33. Fulop, G.M., and R.A. Phillips(1990) The scid mutation in mice causes a general defect in DNA repair, Nature (London),347, 479-482. Gantt, R., R. Parshad, F.M. Price and K.K. Sanford (1987) Biochemicalevidence for deficient DNA repair leading to enhanced G 2 chromatid radiosensitivityand susceptibility to cancer, Radiat. Res., 108, 117-126. Gotoff, S.P., E. Amirnokri and E.J. Liebnor (1967) Ataxiatelangiectasia,neoplasia untoward response to X-irradiation and tuberous sclerosis, Am. J. Dis. Child., 114, 617627. Hart, R.M., B.F. Kimler, R.G. Evans and C.H. Park (1987) Radiotherapeutic management of medulloblastomain a pediatric patient with ataxia telangiectasia,Int. J. Radiat. Oncol. Biol. Phys. 13, 1237-1240. Hecht, F., and B.K. Hecht (1990)Cancer in ataxia telangiectasia patients, Cancer Genet. Cytogenet.,46, 9-19. Hecht, F., and B. Kaiser-McCaw (1982)Ataxia-telangiectasia: genetics and heterogeneity, in: B.A. Bridges and D.G. Harnden (Eds.),Ataxia-telangiectasia--ACellular and Molecular Link between Cancer, Neuropathology, and Immune Deficiency.Wiley, New York, pp. 197-201. Hermaszewski, R.A., and A.D.B. Webster (1993) Primary hypogammaglobulinaemia: a survey of clincial manifestations and complications,Quart. J. Med., 86, 31-42. Higurashi, M., and P.E. Conch (1973) In vitro chromosomal radiosensitivity in 'chromosomal breakage' syndromes, Cancer, 32, 380-383. Jaspers, N.G.J., R.D.F.M. Taalman and C. Baan (1988) Patients with an inherited syndrome characterized by immunodeficiency,microcephaly,and chromosomalinstability: genetic relationship to ataxia telangiectasia, Am. J. Hum. Genet., 42, 66-73. Jayanth, R.V. and W.M. Hittelman (1991) 9-/3-D-Arabinofuranosyl-2-fluoroadenine(F-ARA-A) inhibits both the fast and slow components of chromosome repair, in: J.D. Chapman, W.C. Dewey and G.F. Whitmore (Eds.), Radiation Research: a Twentieth-Century Perspective, Academic Press, New York, p. 411 (Abstr.). Johnstone, A.P., and G.T. Williams (1982) Role of DNA breaks and ADP-ribosyltransferase activity in eukaryotic differentiationdemonstrated in human lymphocytes,Nature (London),300, 368-370. Jones, L.A., D. Scott, S.A. Elyan and R. Cowan (1992)Chro-

263 mosomal radiosensitivityand cancer predisposition:low dose rate irradiationof human lymphocytes,Int. J. Radiat. Biol.,62, 372 (Abstr.). Kersey, J.H., R.S. Shapiro and A.H. Filipovich(1988) Relationship of immunodeficiencyto lymphoid malignancy, Pediatr. Infect. Dis. J., 7, $10-12. Kinlen, L.J., A.D.B. Webster, A.G. Bird, R. Haile, J. Peto, J.F. Soothilland R.A. Thompson(1985)Prospectivestudy of cancer in patients with hypogammaglobulinaemia, Lancet, i, 263-266. Kripke,M.L. (1988)Immunoregulation of carcinogenesis:past, present and future, J. Natl. Cancer Inst., 80, 722-727. Macleod, R.A.F., and P.E. Bryant (1992) Effects of adenine arabinosideand coformycinon the kinetics of G 2 chromatid aberrations in X-irradiated human lymphocytes, Mutagenesis,7, 285-290. Macleod, R.A.F., A.F. Christie,N.D. Costa and P.E. Bryant (1990) Repair kinetics in Chinese hamster ovary cells of X-ray induced DNA damage and chromatid aberrations during a cell-cycle extended by transient hypothermia, Mutagenesis,5, 279-283. Moretta, L., M.C. Mingari, S.R. Webb, E.R. Pearl, P.M. Lydyard, C.E. Grossy, A.R. Lawton and M.D. Cooper (1977)Imbalancesin T cell subpopulationassociatedwith immunodeficiency and autoimmunesyndrome,Eur. J. Immunol.,7, 696-700. Morrel, D., C.L. Chase and M. Swift (1986) Cancer and autoimmune disease in families with common variable immunedeficiency,Genet. Epidemiol.,3, 17-26. Mozdarani, H., and P.E. Bryant (1989)Cytogeneticresponse of normal human and ataxia telangiectasiaG 2 cells exposed to X-rays and araA, Mutation Res., 226, 223-228. Muller, H.-J. (1990)Recessivelyinheriteddeficienciespredisposingto cancer, Anticancer Res., 10, 513-518. Nakamura, N., Y. Kusunoki and M. Akiyama (1990) Radiosensitivityof CD4 or CD8 positivehuman T-lymphocytes by an in vitro colony formationassay, Radiat. Res., 123, 224-227. Olerup, O., C.I.E. Smith,J. Bj6rkanderand L. Hammarstr6m (1992)Shared HLA class II-associatedgenetic susceptibility and resistance,related to the HLA-DQB1gene, in IgA deficiencyand commonvariable immunodeficiency, Proc. Natl. Acad. Sci. (U.S.A.),89, 10653-10657. Parshad, R., K.K. Sanford and G.M. Jones (1983)Chromatid damage after G 2 phase X-irradiation of cells from cancer-prone individualsimplicates deficiency in DNA repair, Proc. Natl. Acad. Sci. (U.S.A.),80, 5612-5616. Parshad, R., K.K. Sanford, K.H. Kraemer, G.M. Jones and R.E. Tarone (1990) Carrier detection in xeroderma pigmentosum,J. Clin. Invest., 85, 135-138. Peters, L.J. (1990) Significanceof genetic variabilityin radiosensitivityin clinicalradiotherapy,J. Jpn. Soc. Ther. Radiol.Oncol.,2, 247-253. Prosser, J.S., A.A. Edwards and D.C. Lloyd(1990)The relationshipbetween colony-formingabilityand chromosomal aberrations induced in human T-lymphocytesafter gamma-irradiation,Int. J. Radiat. Biol.,58, 293-302.

Rary, J.M., M.A Bender and T.E. Kelly (1974) Cytogenetic studies in ataxia telangiectasia,Am. J. Hum. Genet., 26, A70. Robison,L.L.,V. Stoker, G. Frizzera, K. Heinitz,A.T. Meadows and A.H. Filipovich(1987)Hodgkin'sdisease in pediatric patients with naturalllyoccurring immunodeficiency, Am. J. Pediatr. Hematol. Oncol.,9, 189-192. Sanford, K.K., R. Parshad, R. Gantt, R.E. Tarone, G.M. Jones and F.M. Price (1989a)Factors affectingand significance of G 2 phase chromatid radiosensitivityassociated with genetic predispositionto cancer, Int. J. Radiat. Biol., 55, 963-982. Sanford, K.K., R. Parshad, R.R. Gantt and R.E. Tarone (1989b)A deficiencyin chromatinrepair, genetic instability and predispositionto cancer, CRC Crit. Rev. Oncogen., 1, 323-341. Sanford, K.K., R. Parshad, F.M. Price, G.M. Jones, R.E. Tarone, L. Eierman, P. Hale and T.A. Waldmann(1990) Enhanced chromatid damage in blood lymphocytesafter G 2 phase X-irradiation,a marker of the ataxia telangiectasia gene, J. Natl. Cancer Inst., 82, 1050-1054. Savage,J.R.K.,and D.G. Papworth(1991)Excogitationsabout the quantificationof structural chromosomalaberrations, in: G. Obe (Ed.),Advancesin MutagenesisResearch, Vol. 3, Springer,Berlin,pp. 162-189. Schaffer, F.M., J. Palermos, Z.B. Zhu, B.O. Barger, M.D. Cooper and J.E. Volankis (1989) Individualswith IgA deficiencyand common variable immunodeficiency share polymorphismsof major histocompatibility complex class III genes, Proc. Natl. Acad. Sci. (U.S.A.),86, 8015-8019. Scientific Group on Immunodeficiency(1989) Primary immunodeficiencydiseases, Report of a WHO sponsored meeting,Immunodef.Rev., 1, 173-205. Scott, D., and P.E. Bryant (1992)Workshop Report, Second L.H. Gray Workshop,Int. J. Radiat. Biol.,6, 293-297. Scott, D., L.A. Jones, S.A.G. Elyan, A. Spreadborough,R. Cowan and G. Ribiero (1993) Identificationof A-T heterozygotes, in: R.A. Gatti and R.B. Painter (Eds.), Ataxia-Telangiectasia,NATO ASI Series, Subseries H, Cell Biology,Vol. 77, pp. 101-116. Spector, B.D., G.S. Perry III and J.H. Kersey (1978)Genetically determined immunodeficiency diseases (GDID) and malignancy,Report from the ImmunodeficiencyCancer Registry,Clin. Immunol.Immunopathol.,11, 12-29. Spector, B.D., A.H. Filipovich,G.S. III. Perry and J.H. Kersey (1982)Epidemiologyof cancer in ataxia telangiectasia,in: B.A. Bridges and D.G. Harnden (Eds.), Ataxia-telangiect a s i a - - A Cellular and Molecular Link between Cancer Neuropathology,and Immune Deficiency, Wiley, New York, pp. 379-386. Taalman, R.D.F.M., C.M.R. Weemaes, T.W.J. Hustinx, J.M.J.C. Scheres, J.M.E. Clement and G.B.A. Stoelinga (1987)Chromosomestudiesin IgA-deficientpatients,Clin. Genet., 32, 81-87. Taylor, A.M.R., C.M. Rosney and J.B. Campbell(1979) Unusual sensitivityof ataxia-telangiectasia to bleomycin,Cancer Res., 39, 1046-1050.

264 Taylor, A.M.R., E. Flude, B. Laher, M. Stacey, E. McKay, J. Watt, S.H. Green and A.E. Harding (1987)Variant forms of ataxia telangiectasia,J. Med. Genet., 24, 669-677. Teo, I.A., C.F. Arlett, S.A. Harcourt, A. Priestley and B.C. Broughton(1983) Multiple hypersensitivityto mutagens in a cell strain (46BR) derived from a patient with immunodeficiencies,Mutation Res., 107, 371-386. Vo[echovsloj,I., M. Munzarova and J. Lokaj (1989)Increased bleomycin-inducedchromosome damage in lymphocytesof patients with common variable immunodeficiencyindicates an involvementof chromosomal instabilityin their cancer predisposition,Cancer Immunol. Immunother.,29, 303-306. Vo~echovsk~, I., J. Litzman, J. Lokaj, P. Hausner and T. Poch (1990) Common variable immunodeficiencyand malignancy: a report of two cases and possible explanationfor the association,Cancer Immunol. Immunother., 31, 250254. Vo[echovsk~,I., J. Litzman, J. Lokaj and R. Sobotkova(1991) Family studies in common variable immunodeficiency,J. Hyg. Epidemiol.Microbiol.Immunol.,35, 17-26. Waldmann, T.A. (1978) Common variable hypogammaglobulinemia,in: A. Baumbarten and F.F. Richards (Eds.), CRC Handbook Series in Clinical Laboratory Science, Section F. Immunology,Vol. 1, Part 1, CRC Press, Washington, pp. 73-84.

Waldmann,T.A. (1982)Immunologicalabnormalitiesin ataxia telangiectasia,in: D.A. Bridges and D.G. Harnden (Eds.), Ataxia-telangiectasia--ACellular and Molecular Link between Cancer Neuropathology,and Immune Deficiency, Wiley, New York, pp. 37-51. Webster, A.D.B., D.E. Barnes, C.F. Arlett, A.R. Lehmann and T. Lindahl(1992)Growth retardation and immunodeficiency in a patient with mutations in the DNA ligase I gene, Lancet, 339, 1508-1509. Wollheim, F.A., S. Belfrages and S. Coster (1964) Primary 'acquired' hypogamma-globulinemia: clinical and genetic aspects of nine cases, Acta Med. Scand., 176, 1-26. Woods, C.G., S.E. Bundey and A.M.R. Taylor (1990)Unusual features in the inheritance of ataxia telangiectasia,Hum. Genet., 84, 555-562. Yates, J.R.W., A.M.R. Taylor, R. Carachi, J.B.P. Stephenson, E. Boyd and D.G. Young (1989) A new chromosomal instabilitysyndrome with tissue specific radiosensitivity,J. Med. Genet., 26, 208-209. Ying, K.L. and W.E. Decoteau (1981) Cytogenetic anomalies in a patients with ataxia, immune deficiency, and high alpha-fetoproteinin the absence of telangiectasia,Cancer Genet. Cytogenet, 4, 311-317. Ziv, Y., A. Amiel, N.G.J. Jaspers, A.I. Berkel and Y. Shiloh (1989)Ataxia-telangiectasia:a variant with alteled in vitro phenotype of fibroblastcells, Mutation Res., 210, 211-219.