Mc Bgll. *.. Psd. I. I. I I. I. I. I. I. 1. 72. 142 159. 318. I a Bglll. FIG. 1. Restriction map of p47-phox cDNA. (a) Probes used for preliminary analysis of AR-CGDRNA.
Proc. Nati. Acad. Sci. USA Vol. 88, pp. 2753-2757, April 1991 Medical Sciences
Autosomal recessive chronic granulomatous disease caused by deletion at a dinucleotide repeat (PCR/mutation/DNA sequencing/NADPH oxidase)
COLIN M. CASIMIR*t, HANAN N. BU-GHANIM*, ADAM R. F. RODAWAY*, DAVID L. BENTLEYt, PETER ROWE*, AND ANTHONY W. SEGAL* *Department of Medicine, Rayne Institute, University College London, University Street, London WC1E 6JJ, United Kingdom; and tImperial Cancer Research Fund Laboratories, Lincoln's Inn Fields, London WC2A 3PX, United Kingdom
Communicated by Walter Bodmer, December 26, 1990 (received for review November 22, 1990)
ABSTRACT Chronic granulomatous disease (CGD) is a rare inherited condition rendering neutrophils incapable of killing invading pathogens. This condition is due to the failure of a multicomponent microbicidal oxidase that normally yields a low-midpoint-potential b cytochrome (cytochrome bm). Although defects in the X chromosome-linked cytochrome account for the mijority of CGD patients, as many as 30% of CGD cases are due to an autosomal recessive disease. Of these, >90% have been shown to be defective in the synthesis of a 47-kDa cytosolic component of the oxidase. We demonstrate here in three unrelated cases of autosomal recessive CGD that the identical underlying molecular lesion is a dinucleotide deletion at a GTGT tandem repeat, corresponding to the acceptor site of the first intron-exon junction. Slippage of the DNA duplex at this site may contribute to the high frequency of defects in this gene.
a
EooRi
I
PO I
BamHl
318
634
I
Neil Smal I I
EcoRi
916
1400
AI BernI
Psd
: A
~
*
%.
A
1060
a Narl
~A Smal
~~~~~~~
II
*.~~~~~~~~~~~~~~~~~~~~~~~~~.
.
*I.
Failure of a membrane-bound, multicomponent electron transport chain of neutrophils to generate superoxide gives rise to chronic granulomatous disease (CGD), a condition characterized by extreme susceptibility to infection (1, 2). The end point of the electron transport chain is a low midpoint-potential b cytochrome (cytochrome b-245) (2-4), and defects in this X chromosome-linked cytochrome account for the majority of CGD patients. A large proportion of CGD cases (30%), however, are due to an autosomal recessive disease (AR-CGD), making defective autosomes actually -'200 times more common than defective X chromosomes. In >90%o ofAR-CGD a 47-kDa cytosolic protein (5-9) has been shown to be the defective component of the microbicidal oxidase (p47-phox) (8-12). We describe here the precise nature of the gene defect in the p47-phox protein.§
MATERIALS AND METHODS Cell Culture. B cell lines and HL-60 human promyelocytic leukemia cells were maintained at 37TC in Hepes-buffered RPMI 1640 medium/10%o fetal calf serum/2 mM L-glutamine/penicillin at 100 ,ug/ml/streptomycin at 100 pg per ml. Induction of myeloid differentiation in HL-60 cells was by addition of 1% dimethyl sulfoxide to the culture medium and growth for a further 7 days. Probe Construction. Probe constructs were made by digesting a complete cDNA cloned into the modified BlueScribe vector pVZ1 (12) with enzymes that cut once in the polylinker sequence 3' to the cDNA and once in the cDNA insert itself. When recircularized, this procedure generated
b EWR1
I I
Xmnl
I I 72
Mc
Bgll
*..
I 1I I
I
142 159
I
Psd 318
a
Bglll
FIG. 1. Restriction map of p47-phox cDNA. (a) Probes used for preliminary analysis ofAR-CGD RNA. The precise areas covered by the probes were as follows: A Pst I probe, nucleotides (nt) 1-318; A BamHI probe, nt 343-634; A Sma I probe, nt 727-1020; A Nar I probe, nt 916-1400. (b) Expanded view of the Pst I probe region showing the probes from the 5' end of the p47-phox cDNA used to localize position of the defect in AR-CGD.
The T3 RNA promoter was thus brought into proximity with different regions of the cDNA for the synthesis of antisense cRNA probes. This nested set of subclones was converted to a series of overlapping probe templates covering distinct regions of the cDNA by linearizing the subclones at appropriate positions increasingly 5' within the cDNA inserts, as illustrated in Fig. 1. The Bgl II probe did not require separate construction but was transcribed from the original Pst I probe by linearizing at the Bgl II site.
five subclones all having the 5' end of the cDNA sequence but terminating at different 3' end points in the cDNA sequence.
Abbreviations: CGD, chronic granulomatous disease; p47-phox, 47-kDa protein component of the microbicidal oxidase; AR-CGD, autosomal recessive CGD; X-CGD, X chromosome-linked CGD. *To whom reprint requests should be addressed. §The sequences reported in this paper have been deposited in the GenBank data base (accession nos. M60941 and M60942).
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
2753
2754
Proc. Natl. Acad. Sci. USA 88 (1991)
Medical Sciences: Casimir et al.
collected into a 1.5-ml microcentrifuge tube, phenol/ chloroform/isoamyl alcohol (25:24:1) extracted once, and ethanol precipitated. The DNA was resuspended in 20-30 ,ul ofTE buffer (10 mM Tris, pH 8.0/1 mM EDTA), and 1 A.1 was used as template for sequencing reactions. The sequencing reactions were done as described by Keohavong and Thilly (18) but [32P]dCTP was used for labeling. The sequence was established on both strands of the DNA by using the PCR primers to prime the sequencing reactions. The genomic DNA subclone sequencing was performed as described in the Sequenase manual (United States Biochemical). Sequencing gel electrophoresis was done exactly as described for the RNase protection analysis.
RNase Mapping. Total RNA was isolated from EpsteinBarr virus-transformed B lymphocytes (13, 14) on guanidinium thiocyanate/cesium chloride gradients (15). For hybridization the RNA (1-4 pug) was mixed with 20 Ag of yeast tRNA and 4 ng of cRNA probe prepared by transcription of 1 ,ug of linearized vector with bacteriophage T3 polymerase and hybridized for at least 16 hr at 420C in siliconized glass capillary tubes. After hybridization, the contents of the capillary were flushed out and digested as described by Melton et al. (16) by using RNase A at 16.7 gg/ml and RNase T1 at 167 units/ml. The reaction products were analyzed on standard denaturing 5% polyacrylamide gels. The gel was then dried under vacuum and autoradiographed with x-ray film (Hyperfilm MP; Amersham) for up to 24 hr at room temperature or at - 70'C with or without intensifying screens, depending on the strength of the signal. DNA Amplification and Sequencing. Sequencing template was prepared from the total cell RNA by synthesis of cDNA using random hexanucleotide primers and PCR (30 cycles) with p47-phox-specific primers as described by Kawasaki and Wang (17). PCR conditions were as follows: denature at 920C for 30 sec (1 min, 1st cycle); hybridize at 520C for 1 min; extend at 720C for 1 min (5 min final cycle). The PCR products were chloroform extracted (equal vol) and separated on a 0.8% high-gelling temperature/2.5% wide-range agarose composite gel. The DNA-containing gel slice was excised from the gel, frozen in liquid nitrogen (5-10 min), and thawed, and the DNA was then purified from the gel by centrifugation at 13,000 x g for 5 min. The eluate from the gel slice was
1 2 3 4
5 6 7
Pst 8 9 10 11 12 13
r0t ~ 11
47 kDaF lop
4 5
T-7 8
9 i0 o
14 15
9
*
_R,
4194 n.
A _
310 nt m
A47kDa 1
Ac
b
a
Cofilin l
RESULTS RNase Protection Mapping of AR-CGD Lesion. To identify the molecular lesion rendering the p47-phox protein nonfunctional in AR-CGD patients, we first established that patients lacking the p47-phox protein did, indeed, express an mRNA. Epstein-Barr virus-transformed B cell lines have been shown to contain a functional oxidase identical to that of phagocytes (13, 14) and provided an abundant and convenient source of RNA from CGD patients and normal individuals. Northern (RNA) blot hybridizations of the patients' RNA, probed with the insert from a complete p47-phox cDNA clone (12) revealed the presence of an mRNA corresponding identically in size to the normal p47-phox messenger (C.M.C., unpublished work) in agreement with observations made in monocytes (8). Clearly, the defect in AR-CGD patients did not involve
-mm.
O
0
_
@281 271
142> 0
Sol
aS4. 234
a 194
pii.*t***
|
FIG. 2. RNase protection analysis of RNA from normals and AR-CGD patients. (a) Total cellular RNA was hybridized in solution to a 32P-labeled cRNA probe, and the resulting hybrids were digested with a mixture of RNases A and T1. To control for equal loading of RNA the Pst I probe was used in combination with an unrelated constitutive gene probe, that for cofilin, which was included in the same hybridization reactions as the p47-phox protein probe. The cofilin probe was prepared from a subclone encoding the 5' 250 nt of the cDNA. Arrows denote the protected probe species; positions of the full-length protection products are given by the protection products in lanes 10 and 11. The slight size discrepancy between size of the protected species in lanes 10 and 11 and those in the remaining lanes is from the vector sequences common to the probe and the synthetic mRNA, which are absent in the RNA extracted from cells. The fully protected Pst I probe probably appears as a doublet due to initiation of mRNA at two slightly differing sites. Other probes [used here (Acc I, see below) and previously (12)] also complementary to the 5' terminus of the mRNA have behaved similarly. The arrow (47 kDa) at 220 nt indicates the position of a p47-phox-specific partial-protection product, which is the only species seen with the AR-CGD RNA. Note that in uninduced HL-60 cells, which express only very small amounts of p47-phox, this species, as well as the full-length product, are absent, demonstrating their common origin. Lane designation is as follows: RNA from X-CGD heterozygote carrier (lane 1), X-CGD (lane 2), AR-CGD (lanes 3-5), normal individual (lane 6), normal polymorphonuclear leukocytes (lane 7), undifferentiated HL-60 cells (lane 8), differentiated HL-60 cells (lane 9), positive control "synthetic p47-phox mRNA" prepared by transcription of cDNA clone to produce a "sense" RNA transcript (lane 10), "synthetic" cofilin mRNA (lane 11), tRNA controls and p47-phox probe (lane 12), cofilin probe (lane 13), undigested probes, p47-phox Pst I probe (lane 14), cofilin probe (lane 15). (b) RNase mapping of p47-phox mRNA with Acc I probe. RNA from X-CGD "carrier" (lane 1), AR-CGD (lanes 2-4), X-CGD (lane 5), normal individual (lane 6), normal polymorphonuclear leukocytes (lane 7), undifferentiated HL-60 cells (lane 8), "synthetic" mRNA (lane 9), tRNA (lane 10), undigested Acc I probe (lane 11).
Proc. Nati. Acad. Sci. USA 88 (1991)
Medical Sciences: Casimir et al.
deletion of the entire gene or gross structural rearrangements but was due to a more subtle alteration' in coding potential. With the object of locating such a change in the mRNA sequence, a series of RNase mapping experiments (16) were done. The cDNA sequence was divided into four subclones (see Materials and Methods), together covering the complete mRNA sequence, except for m100 nt (positions 635-726 and 319-342) (see Fig. la). These four 'subclones were transcribed into cRNA probes (16) by using the vector-encoded bacteriophage T3 promoter and hybridized in solution to total cellular RNA from normal individuals, X chromosome-linked CGD (X-CGD) patients, or AR-CGD patients. After RNase digestion of the RNA/RNA hybrids'and denaturing gel electrophoresis, only the Pst I probe, which contained sequences complementary to the 5' terminal 318 nt of the mRNA, showed evidence for discriminating between ARCGD and normal p47-phox mRNA. Results obtained with this probe are shown in Fig. 2a, where the hybridizations also contain a constitutive control gene probe, that for the actinbinding protein cofilin (19). In all three cases of AR-CGD virtually no fully protected (318 nt) probe was visible; instead a partial protection product of -220 nt was seen indicating a divergence in sequence between the probe and AR-CGD mRNA s100 nt from one end of the Pst I probe. We note that the' size of this partial product also corresponds with a similar partial protection product visible in all the "wild-type" cellular RNAs, including that from the HL-60 promyelocytic cell line. This signal probably derives from a combination of the presence of unprocessed nuclear RNA, some incorrectly spliced mRNA, and possibly also slippage of the two strands in the hybrid (see below). Owing to the uniform labeling of the probe it was not possible to locate unequivocally the position of the sequence difference between the normal and AR-CGD cases. To define this difference in greater detail, a series of additional probes was prepared, complementary to different areas of the region of the mRNA encompassed by the Pst I probe (Fig. lb). The Bgl II probe proved unable to distinguish AR-CGD patients' RNA from normal RNA (data not shown), which indicated that the defect in the AR-CGD patient's p47-phox mRNA could not be located in the region of overlap between the Bgl II and Pst I probes but was, therefore, located within 100 nt of the 5' end of the mRNA. This pattern was confirmed by use of the Acc I probe, which spans the 5' terminal 142 nt of the mRNA, in contrast with the Bgl II probe, and behaved similarly to the Pst I probe in revealing a clear difference between AR-CGD and normal (or X-CGD) p47-phox mRNA (Fig. 2b). With this probe a strong partial protection product (95-100 nt in length) was observed with AR-CGD, but no full-length species of 142 nt. This technique located the AR-CGD defect to a position :'45-50 nt from one end of the Acc I probe. As the AR-CGD defect had been mapped to 100 nt from the 5' end of the mRNA using the Pst I probe, it was highly probable that the partial-protection product from the Acc I probe was missing sequences from the 3' end of the probe. This result would then also place the AR-CGD lesion at -100 nt into the mRNA. This interpretation was confirmed by using the Xmn I probe, which spans the first 70 nt of the p47-phox mRNA. As expected, the Xmn I probe was unable to detect any sequence discrepancy between normal and AR-CGD p47-phox mRNA, although the probe was clearly able to accurately identify accumulation of this mRNA during (dimethyl sulfoxide-induced) myeloid differentiation in HL-60 cells (data not shown) (7, 12). DNA Sequence Analysis of AR-CGD. The localization of the defect in AR-CGD to approximately position 100 in the mRNA enabled us to elucidate the sequence of the mRNA by direct sequencing of a PCR-amplified partial cDNA (Fig. 3a). A PCR product that spanned the defect was obtained by synthesizing oligonucleotide primers identical to the extreme
2755
5' end of the mRNA sequence and at a location overlapping the Bgl II site (position 170). After reverse transcription, these primers were able to direct the synthesis of a PCR product of 178 nt (Fig. 3b). The PCR product was purified, and its sequence was determined (Fig. 3a). In all three cases of AR-CGD a GT dinucleotide at a GT-GT repeat was found deleted from the sequence at positions 95 and 96 (see also Fig. 3b). The consequences of this 2-nt deletion are (i) to frameshift the mRNA such that amino acids subsequent to amino acid 24 are incorrect and (ii) to introduce a chain-terminating codon (TAA) at amino acid 52 (nt 175-177). This alteration predicts a protein product of m6000 Mr from the AR-CGD p47-phox mRNA (Fig. 3b). At present we have no evidence regarding the existence of such a molecule in the cells from AR-CGD patients. Localization of the AR-CGD defect to this position in the mRNA was of interest as our analysis of genomic DNA by Southern blotting (H.B.-G and C.M.C., unpublished data) had indicated the presence of an intron-exon junction close to the Xmn I site and upstream from the Acc I site. DNA sequencing of a genomic subclone demonstrated that the AR-CGD deletion lay exactly at this junction (Fig. 4), which comes at position 94/95 in the wild-type mRNA sequence (Fig. 3b). The possibility remained, therefore, that the pri-
a
+ A C G T x
A
AR-
AR -
AC G T
AC G T
-G -is T --@i-Gw
-
*
i_-G -
_W
A T
_
"I'A
."w
we
~
i.;4f ~~~~~~
A
~~CG 4 ~~~A
I_
b
1
+ICGD
+/CGD +/CGD CGD
25
GGG GAGCACTGGAGGCCACCCAGTC ATG met gly
GAC ACC TTC asp thr phe
50 XmnI CGT CAC ATC GCC GTG CTG GGC TTT GAG AAG CGC ile arg his ile ala leu leu gly phe glu lys arg
ATC
** 100 75 TTC GTA CCC AGC CAG CAC TAT GTG TAC ATG TTC CTG his val ser val tyr met phe leu tyr pro phe gln GTA CAT GTT CCT GGT val his val pro gly
+
125 ACC I GTG AAA TGG CAG GAC CTG TCG GAG AAG GTG GTC TAC val lys trp g1n asp ieu ser glu lyS val val tyr
CGD
GAA ATG GCA GGA CCT GTC GGA GAA GGT GGT CTA CCG g1u met ala gly pro val gly glu gly giy leu pro
+
CGD
150 Bgl II CGG CGC TTC ACC GAG ATC TAC
175
GAG TTC CAT AAA ACC arg arg phe thr glu ile tyr glu phe his iys thr
GCG CTT CAC CGA GAT CTA CGA GTT CCA TAA AAC CTT ala leu his arg asp leu arg val pro *
FIG. 3. Sequence analysis of AR-CGD p47-phox mRNA. (a) Sequencing ladders from PCR-amplified p47-phox cDNA from ARCGD (AR) and normal (+) RNA. The sequence of the RNA was determined with T7 DNA polymerase on template obtained by reverse transcription and PCR amplification. O, Bases deleted in AR-CGD. (b) Nucleotide and encoded amino acid sequences surrounding the defect in the p47-phox mRNA. Nucleotide positions are shown above the sequence, and primers for PCR amplification and
sequencing are underlined. Relevant restriction enzyme sites are indicated on the sequence. The boldface GTGT dinucleotide repeat corresponds to that from which a GT dinucleotide (**) was found deleted in AR-CGD. Sequences downstream from the deletion are shown for both normal and mutant alleles. ***, Termination codon.
2756
Medical Sciences: Casimir et al.
TTT GAG AAG CGC TTC GTA CCC AGC CAG CAC TAT gtgagttagctggtggagg phe glu lys arg phe val pro ser gin his tyr
Proc. Natl. Acad. Sci. USA 88 (1991) nomenon may also occur quite frequently in processing of the
transcript for the p47-phox-encoding gene. gcatcccgtggggggaatacgggagggacagcacggccacccttgcagtc cccagggggaacc E The mutagenicity of a dinucleotide repeat could result from a tendency for the DNA strands to slip at this site. This aqctccaqtqaqqactaacg ----------slippage could generate deletions during copying of this sequence by DNA polymerase. Alternatively, a dinucleotide ---------g tttgtgccctttctqcaatccaoqacaaccqcaaagatggtcctcaccc caatcctctgggc repeat could act as a site of unequal crossing over. In this work some support for the former hypothesis is provided in the RNase mapping experiments. The partial protection ttcctccagtgggtagtgggatcctggatgcaccagcaaagcctctttgc gaggctgaatggg products seen with normal RNA originated from cleavage at GT( A-AA TGG CAG gtcccccgactctggctttcccccag a site identical to that of the patients' deletions, suggesting val tyr met phe leu val lys trp gin that in a proportion of the hybrids, slippage was occurring at FIG. 4. Nucleotide sequence of p47-phox-encoxding genomic the dinucleotide repeat; the resulting unpaired bases at this DNA at intron-exon junctions. The intron sequencee is shown in location would then be nuclease sensitive. GTG TAC ATG TTC CTG
lowercase letters, and the exon sequence is shown in uppercase letters. Boldface sequences are as for Fig. 3b. Underlirned sequences
represent primers for PCR amplification and sequenc ing.
mary defect in the AR-CGD patients was not the 2-nt deletion but a different underlying change, resulting in delfective RNA processing. To investigate this possibility, genoi iic DNA was amplified across the )' splice junction by usin, g the downstream cDNA primer previously described and a sequence chosen from the intron sequence 224 nt upstreain, as determined from the sequence of the subcloned geniomic DNA. The sequence of this 224-nt PCR product was doetermined in identical fashion to the mRNA amplification Iproduct. Sequencing revealed that the genomic DNA from t.he AR-CGD patients also lacked the GT dinucleotide found rnissing from the mRNA (Fig. 4). The presence of the GT deletion has enabled uIS to develop a simple diagnostic test for p47-phox-deficient AR-CGD by using the restriction enzyme Dra III, which ciuts only the wild-type cDNA sequence and not the mutated form. We have used this test to confirm that another fouir individuals most probably carry the dinucleotide deletion and also to identify one family that appear to have a le-sion in the p47-phox-encoding gene distinct from the one de scribed here (data not shown).
We thank Dr. Phil Marsh (King's College, London) for synthesis of oligonucleotide primers and for being a fount of useful knowledge; Dr. Carolyn Dent (University College London) for synthesis of oligonucleotide primers; Drs. A.-M. Frischauf and M. Bucan (Imperial Cancer Research Fund) for the gift of the human cosmid library; Dr. Gordon (Jimmy) Stewart for helpful discussions and moral support; and Margaret Chetty for excellent technical assistance. This work was supported by grants from the Medical Research
Council and the Wellcome Trust. 1. Teahan, C., Rowe, P., Parker, P., Totty, N. & Segal, A. W. (1987) Nature (London) 327, Invest. 720-721. 1785-1793. A. W. (1989) J. Clin. 2. Segal, 83, Nature (London) 3. Segal, A. W. & Jones, 0. T. G. (1978) 276, 5557 4. Segal, A. W. (1988) Hematol. Oncol. Clin. North Am. 2,
213-223.
5. Volpp, B. D., Nauseef, W. M., Donelson, J. E., Moser, D. R. & Clark, R. A. (1989) Proc. Natl. Acad. Sci. USA 86, 7195-
7199. 6. Nunoi, H., Rotrosen, D., Gallin, J. I. & Malech, H. L. (1988) Science 242, 1298-1301. 7. Volpp, B. D., Nauseef, W. M. & Clark, R. A. (1988) Science 242, 1295-1297. 8. Lomax, K. J., Leto, T. L., Nunoi, H., Gallin, J. I. & Malech, H. L. (1989) Science 245, 409-412. 9. Teahan, C. G., Totty, N., Casimir, C. M. & Segal, A. W. DISCUSSION (1990) Biochem. J. 267, 485-489. AR-CGD accounts for about one-third of all CGD cases, 10. Clark, R. A., Malech, H. L., Gallin, J. I., Nunoi, H., Volpp, B. D., Pearson, D. W., Nauseef, W. M. & Curnutte, J. T. signifying that -1 in 1700 individuals carry a del fective allele for p47-phox. It is of significance, therefore, tdhat all three (1989) N. EngI. J. Med. 321, 647-652. 11. Segal, A. W., Heyworth, P. G., Cockcroft, S. & Barrowman, patients studied here, despite coming from totallly unrelated M. M. (1985) Nature (London) 316, 547-549. populations, are all homozygous for the same deftective allele. This mutation in the p47-phox-encoding gene i!s, therefore, Rodaway, 12. A. A. R. F., C. G.,Cell. Casimir, C. M., Segal, W. & Bentley, D. Teahan, L. (1990) Mol. Biol. 10, 5388-53%. extremely prevalent, a situation that contrasts m; 13. Maly, F. E., Cross, A. R., Jones, 0. T., Wolf Vorbeck, G., wcth a recent study on the very rare AR-CGD cause rkedlby d by defects Walker, C. & De Weck, A. L. (1988) J. Immunol. 140, 2334in the a subunit of the cytochrome, where fc)ur different 2339. defective alleles were identified in three differ'ent patients 14. Maly, F. E., Nakamura, M., Gauchat, A., Urwyler, C., (20). At present we have no definitive answer to why the Walker, C., Dahinden, C. A., Cross, A. R., Jones, 0. T. & De GTGT repeat in the gene for p47-phox should b e so suscepWeck, A. L. (1990) J. Immunol. 142, 1260-1267. tible to mutation. IC seems unlikely that the mutation is 15. Maniatis, T., Fritsch, E. & Sambrook, J. (1989) Molecular maintained in any way by selection or confers aliy heterozyCloning:A Laboratory Manual-(Cold Spring Harbor Lab., Cold Spring Harbor, gote advantage, as appears to be the case in the tthalassemias 16. Melton, D. A., NY). Krieg, P. A., Rebagliati, M. R., Maniatis, T., (21) and in cystic fibrosis (22). It is possible t hatfeatules tZinn, K. & Green, M. R. (1984) Nucleic Acids Res. 12, 7035intrinsic to the DNA sequence might render it suisceptible to 7056. deletion. Dinucleotide repeats have been cited (23) as spuyta17. Kawasaki, E. S. & Wang, A. W. (1989) in PCR Technology: genic, and a brief literature search has revealed 1three similar Principles and Applications for DNA Amplification, ed. Erlich, cases (24, 25), the most obvious parallel to the p4[7-phox gene H. A. (Stockton, New York), pp. 89-97. deletion being in a particular case of a-thalasse-mia (26). In 18. Keohavong, P. & Thilly, W. G. (1989) Proc. Natl. Acad. Sci. this instance the lesion was found at an AGAG repeat, the USA 86, 9253-9257. donor site of the first intron-exon junction. Also in common 19. Matsuzaki, F. M., Matsumoto, S., Yahara, I., Yonezawa, N., with the gene defect for p47-phox, repetition of th e conserved Nishida, E. & Sakai, H. (1988) J. Biol. Chem. 263, 11564nucleotides at the splice junction prevents the g veneration of 20. Dinauer, M. C., Pierce, E. A., Bruns, G. A. P., Curnutte, J. T. an abnormal splice in the deleted mRNA. A simi lar mutation & Orkin, S. H. (1990) J. Clin. Invest. 86, 1729-1737. has also been found in a c-myc cDNA clone (27), probably on 21. Kazian, H. H., Orkin, S. H., Markham, A. F., Chapman, this occasion due to faulty RNA splicing. DatLa described C. R., Youssoufian, H. & Waber, P.G. (1984) Nature (London) earlier on the RNase mapping suggests that a similar phe310,152-154.
Medical Sciences: Casimir et al. 22. Jorde, L. B. & Lathrop, G. M. (1988) Am. J. Hum. Genet. 42, 808. 23. White, M. B., Amos, J., Hsu, J. M. C., Gerrard, B., Finn, P. & Dean, M. (1990) Nature (London) 344, 665-667. 24. Golding, G. B. & Glickman, B. W. (1986) Can. J. Genet. Cytol. 28, 483-496.
Proc. Nadl. Acad. Sci. USA 88 (1991)
2757
25. Sifers, R. N., Brashears-Macatee, S., Kidd, V. J., Muensch, H. & Woo, S. L. (1988) J. Biol. Chem. 263, 7330-7335. 26. Safaya, S. & Rieder, R. F. (1988) J. Biol. Chem. 263, 43284332. 27. Bentley, D. L. & Groudine, M. T. (1986) Mol. Cell. Biol. 6, 3481-3489.
