Ann. Hum. Genet. (1990), 54, 1-15
1
Printed in Great Britain
Common glucose-6-phosphate dehydrogenase (G6PD) variants from the Italian population : biochemical and molecular characterization G. VIGLIETTO," V. MONTANARO,t* V. CALABRO,? D. VALLONE,? M. D'URSO,* M. G. PERSICO" AND G. BATTISTUZZI? * Istituto Internazionale d i Genetica e Biojisica, CNR, Naples, Italy Dipartimento d i Genetica, Biologia Generale e Molecolare, UniversitcE d i Napoli, Naples, Italy
SUMMARY
By biochemical characterization of glucose-6-phosphate dehydrogenase (G6PD)from the red cells of seventeen subjects of the population of Matera (Southern Italy) we have identified six genetically determined common variants. Among these, G6PD Metaponto and G6PD A( - ) Matera had been already fully characterized. We have now found that A(-) Matera is genetically heterogeneous since one of two subjects examined had the two mutations at codons 68 and 126 characteristic of a typical A( - ) variant, while the other subject had only the codon 126 mutation. G6PD Pisticci and G6PD Tursi are two new variants whose molecular lesion is not yet known. G6PD Cagliari-like has biochemical characteristics reminiscent of G6PD Cagliari, isolated in Sardinia, and was found to have the same nucleotide substitution as G6PD Mediterranean. G6PD Montalbano is a new variant, with nearly normal properties, due t o a G-tA transition which causes an Arg+His amino acid replacement a t position 285.
INTRODUCTION
Glucose-6-phosphate dehydrogenase (G6PD) is the first and rate-limiting enzyme of the hexose monophosphate pathway. I t s metabolic role, the provision of adequate amounts of NADPH for biosynthetic or detoxifying reactions, is probably essential to all organisms and cell types (Luzzatto & Battistuzzi, 1985). Human G6PD is encoded by an X-linked gene (Childs et al. 1978). It has been mapped near the end of the long arm of the X chromosome a t band Xq28 by somatic cell hybrid analysis (Pai et al. 1980) and cDNA probed by in situ hybridization experiments (Szabo et al. 1984). The G6PD coding sequence and genomic structure have recently been elucidated (Persico et al. 1986; Takizawa et al. 1986; Martini et al. 1986). The G6PD locus is polymorphic in several human populations. Extensive biochemical characterization of variant phenotypes has led to the notion that polymorphism in different populations is often due to the occurrence of different genetically determined G6PD variants. I n fact, since the pioneering work by Kirkman and coworkers (Kirkman et al. 1965), who first reported evidence of heterogeneity within the so-called G6PD-Mediterranean phenotype in the While this manuscript was in preparation, Professor Giorgio Battistuzzi passed away, leaving a profound sense of sorrow in his friends, colleagues and students. We have lost an excellent geneticist whose invaluable contribution in the analytical study of the G6PD gene variability will remain a guideline for us. Address for correspondence and reprints : Dr M. Graziella Persico, Istituto Internazionale di Genetica e Biofisica, CPU'R, Via Marconi. 10. 80125 Naples, Italy. I
H(: I.: 51
2
G. VIGLIETTO AND
OTHERS
population of Greece, evidence has been accumulating that within a population the degree of genetic heterogeneity may be considerable. As a result, among the more than 300 G6PD variants so far described, 87 appear to be common in a t least one population (Luzzatto & Battistuzzi, 1985). At variance with most human red cell enzyme polymorphisms (Battistuzzi et al. 1977), common inherited variants at the G6PD locus are mostly associated with an abnormally low level of enzyme activity. The enzyme deficiency is clinically relevant since it is often associated with the risk of acute haemolytic episodes, triggered by a variety of exogenous agents (Beutler, 1983). On the other hand, G6PD variants associated with altered levels of enzyme activity appear to have adaptive value in malarial environments (Allison, 1960; Motulsky, 1960 ; Siniscalco et al. 1966; Luzzatto, 1979). Identification of the amino-acid substitution, which is responsible for the altered properties of G6PD variants, can provide information on the genetic relationships between different variants and on the relationship between structure and function of the protein molecule. The G6PD protein is, however, produced at a relatively low level in all cell types, making it virtually impossible to obtain sufficient amounts for sequence analysis, especially in cases of enzyme deficiency. Recently, however, thanks to the availability of the G6PD B DNA sequence (Persico et al. 1986; Takizawa et al. 1986) and of improved methods for DNA cloning and sequencing, it has become possible to determine rapidly and reliably the specific alteration in the DNA sequence, and the amino-acid substitution thereof (Hirono & Beutler, 1988 ; Vulliamy et al. 1988). This paper reports the biochemical and molecular characterization of six G6PD genetic variants commonly occurring in the population of Southern Italy.
MATERIAL AND METHODS
Male carriers of a variant G6PD phenotype had been identified during a population survey in the Matera district, Southern Italy (Calabrb et al. in preparation). From each subject a 20-30 ml blood sample was obtained by venepuncture, collected into EDTA containing tubes and processed within 24 h after collection. To each specimen was added a 0.1 volume of a 3% (w/v) solution of dextran (mol. wt 200300 kDa). After mixing by inversion, the sample was allowed to sediment for 15-20 min at room temperature, The plasma fraction was then transferred to a fresh tube and spun a t 2000 r.p.m. for 10 min in a table centrifuge. The supernatant fraction and red blood cells were reunited and thoroughly mixed by inversion. This procedure was repeated twice more. The resulting pellets containing white blood cells were resuspended into calcium-magnesium-free phosphatebuffered saline (PBS), reunited and centrifuged a t 1600 r.p.m. for 6 min. The pellets were washed once more with PBS and kept frozen a t - 80 OC until the DNA was extracted according to the procedure of Battistuzzi et al. (1985). Red blood cells were transferred to an ice bath and washed three times with cold PBS. Lysates for G6PD activity and purification were obtained according to the procedure of Battistuzzi et al. (1977). Red-blood-cell activity levels and electrophoretic phenotypes were determined according to Battistuzzi et al. (1977) and Betke et al. (1967), respectively. Partial purification and biochemical characterization of the G6PD enzyme protein were perfckrned as
Common GGPD variants in Southern Italy
3
previously described (Betke et al. 1967; Beutler et al. 1968; Modiano et al. 1979; Perona et al. 1983). Cloning and sequencing of variant genes DNA samples (250 pg) from male carriers of variant G6PD were digested to completion with EcoRI restriction endonuclease and size-fractionated on 10-40 % sucrose gradients. DNA from fractions positively hybridizing to G6PD specific DNA probes were cloned into a h GTWes phage. h GTWes EcoRI arms were obtained from Bethesda Research Laboratories. Ligated material was packaged using Stratagene extracts as recommended by the suppliers. The libraries were screened by filter hybridization (Benton & Davis, 1977) using G6PD cDNA probes labelled by nick translation. Phages bearing G6PD specific inserts were isolated and amplified (Maniatis et al. 1982). Inserts were subcloned into a pUC18 plasmid vector using E . coli strain HBlOl as a recipient cell for sequencing by the dideoxy method (Sanger et al. 1977). Sequencing reactions were primed by using a set of oligonucleotides homologous to sequences immediately flanking intron-exon junctions (Fig. 1 ) . Evaluation of genomic DNA by using allele speci$c oligonucleotide probes BamHI-digested, ethanol-precipitated genomic DNA ( 1 pg) was dissolved in a final volume of 100 p1 of 10 mM Tris-HC1 p H 8.3, 50 mM-KCI, 2 mM-MgC1 and 0.01 % gelatine. Intron specific 20mers (0.5 pg each) were added and the mixture was incubated a t 100 "C for 15 min for DNA denaturation. Reannealing was then allowed to occur for 15 min at room temperature. dNTPs were added a t a final concentration of 0 2 mM each. The polymerization reaction was started by the addition of 1 U of Taq-polymerase (Anglican Biotec. Ltd). DNA was allowed to elongate a t 65OC for 4min, denatured a t 95OC for 2 min and renatured a t 45°C for 2 min. The elongation-denaturation-renaturation cycle was repeated 30 times. At cycle nos. 10 and 20, 1 U of Taq-polymerase was added; a t cycle no. 20 dNTPs were added a t a final concentration of 0.2 mM each. Amplified DNAs were denatured by adding NaOH a t a final concentration of 60 mM, fractionated on a 1% alkaline agarose gel (Maniatis et al. 1982) and transferred to Z-bind nylon membranes (CUNO, Mass., USA). Allele specific 19mers (Table 3) were chemically synthesized with an Applied Biosystem Synthesizer and 5-labelled with [y-32P]ATPby T4 polynucleotide kinase (Maniatis et al. 1982). Hybridization of allele specific oligonucleotides to filter immobilized amplified genomic DNA was allowed to proceed for 24 h in 5 x SSPE, 5 x Denhardt's, 0 3 % SDS and 200 pg/ml yeast total RNA, a t a temperature calculated as proposed by Thein & Wallace (1986).The filters were washed with 6 x SSC for 60 min at room temperature with 3 changes, for 2 min a t the hybridization temperature and again for 60 min at room temperature. The washed filters were air-dried and auto-radiographed for 1-2 h a t - 80 "C with intensifying screens.
RESULTS
During a screening of the population of Matera district (Southern Italy), we had identified, out of 1525 boys (5-10 years old), 42 subjects with altered red cell G6PD electrophoretic mobility and/or enzyme activity (Calabrb et al. in preparation). I n 17 cases i t was possible to
'7 24 3 9
Tursi
Matera ( A - )
Cagliari-like
Posillipo
I00
IOO
90
I00
111
93
93
K, K, NADP PJADPH ~
Substrate analogues utilization*
DEAE5
x K
n.d. n.d. S
n.d.
x
n.d. n.d.
n.d.
E ?I'
h'
8
c3
Y 0
Z
0 k
c3
e
M
G
c A
? L
K
A
PH' ?I'
h'
T6
In table mean value sf^.^. are reported. The number of determinations on which values are based is reported in parentheses below each value. Kumber of subjects. 'Enzyme activity in red blood cells is expressed as pM/g H b (IU) and as percent of normal (%). Electrophoretic mobility as percent of normal q6PD B. Assignment to each electrophoretic class has been done on a side-by-side comparison basis. Expressed as percent of natural, G6P and NADP, substrates. 5Expressed as the mM concentration of chloride ions at peak elution. Thermostability is rJ for normal ; L for lower than normal stability. 'pH dependence is pu' for normal; A for altered profile, the variant shows a broad p H maximum extending from pH 7.5 to pH 9.5.
GGPD B
5.46 + O I 6
21
Montalbano
0-7
20
Metaponto
Mediterranean
90
69 90
EM3
%
Kame of variant Pisticci
Activity'
Table 1. Comparison of the biochemical characteristics of GGPD genetic variants found in the populution of Southern Italy
Common G6PD variants in Southern Italy
_158
1?0_ C
C
T
C
C
'
P
G
T
C
C
C
C
A
C
C
C
A
5
C
T
T
C
T
M
~
C
A
C
A
C
A
C
C
~
~
C
C
C
A
C
C
T
CAlTCACACAGGCTCC
,269 C A C T C C C T G A C A T C C C C A M C A C A C ' P G A C C C C C T T C T T C M C C ~ ~ T C T C A ~ C T C C C C ~ C C C ~ T T C ~ C C C T C T C T A C C A G - - - - - - - - - -I -V- S- -I -. - - - - - - - - - ?
CGACCGCCTCCCACCM
270, M
G
A
C
A
G
C
C
C
C
T
C
A
C
A
T
C
T
C
T
C
T
C
~
~
T
C
~
T
C
T
C
C
C
T
C
A
C
T
A
C
* AnATGCAGCCTCCTACCACCCCCTCMCACCCACA~M~CCTCCACC~TCACACCCCMCCCCCKTTCTACC~CCTn;CCCCCCACCCTCTACCACCCCCTCACCMCA
,481 ACATTCACCACTCCTCCATCACCCACA~TMCCC~CCTn;CCCTC~C'CCCCCCT
- - - - - - - - - - - - IVS5 - - - - - - - - - -
ACTCCCCGMCACCCCTKMCCCCCTMC
MCCGCAGGCMGCCCCCA
*88, CCACCCTCCCU;CTCCCACCACACCCn;CMCCCCATCn iCCn;CACMCCCCTTCGGCACGCACC'PGCACACCTC~ACCCCC~TCCMCCACATC~CTCCC~~CCC~ACGACC
C
G. VIGLIETTO AND
6
OTHERS
550
Ill
EZ
560
e +
2:1
J
5'0
E5
El2
E l l
Common G6PD variants in Southern Italy
7
Table 2. Nucleotide substitutions found in the coding sequence of GGPD gene for variants from Southern Italy population Exon Codon Amino-acid Reference G6PD variant no. position Codon change substitution Metaponto 4 58 GAT + AAT Asp + Asn Vulliamy et al. ( I 988) Matera (A-) 4 68 GTG-rATG + Met Vulliamy et al. (1988) Matera (A - ) 5 126 AAT-GAT Am-Asp 6 188 TCC TTC Cagliari-like 1 Ser -+ Present paper Cagliari-like 1 11 437 TAC-rTAT Tyr + Tyr Cagliari-like 2 6 188 TCC + TTC Ser --* Present paper 11 437 TAC-TAT Tyr-Tyr Cagliari-like 2 Montalbano 8 285 CGT +CAT Arg + His Present paper Mediterranean 6 188 TCC -+ TTC Ser --t Vulliamy et al. (1988) 11 437 TAC-TAT Tyr -r Tyr Mediterranean Codon position number refers to the cDNA sequence reported by Persico et al. (1986) (see also Fig. 1).
-
partially purify red cell G6PD in order to carry out a full characterization (see Table 1). This has enabled us to distinguish six new G6PD variants which were provisonally named G6PD Pisticci, G6PD Metaponto, G6PD Montalbano, G6PD Tursi, G6PD A( - ) Matera and G6PD Cagliari-like. Their properties are summarized in Table 1 together with those of G6PD Posillipo and G6PD Mediterranean, two variants previously observed in Southern Italy (ColonnaRomano et al. 1985), and those of the ‘normal’ control G6PD B. I n each case the variant phenotype was seen to be transmitted from mother to son (data not shown). The G6PD of subjects classified as carriers of G6PD Metaponto, G6PD A( - ) Matera and GGPD Cagliari-like have properties very similar to those of the homonymous variants previously described by Vulliamy et al. (1988), by Fenu et al. (1982) and by De Vita et al. (1989). To be able to determine the nucleotide mutations occurring in the different variants we extended the nucleotide sequence analysis of the wild type G6PD gene reported by Persico et al. (1986) and Martini et al. (1986) inside the intron sequence. Figure 1A shows the sequence of 3650 nucleotides of the 18 kb region of the G6PD gene comprehensive of the 13exons, the intron flanking sequences, the 5’ and 3’ non-coding ends. From this sequence oligonucleotide sequences (underlined in Fig. 1A) were derived to perform the sequence analysis by the dideoxy method. Sequence analysis of exonic DNA fragments was performed on one subject carrier of G6PD Montalbano and on two subjects with G6PD Cagliari-like. A single nucleotide substitution was found in the sequence of G6PD Montalbano, a G/C+A/T transition in the second position of codon 285 (Table 2). The nucleotide change caused an Arg residue on the wild-type polypeptide Fig. 1. Nucleotide sequence of GBPD gene exonic and flanking intronic DNA and PCR strategy. (A) The updated sequence of exons and of neighbouring intronic regions is reported. Exons are bounded by arrows. Figures above arrows indicate first and last nucleotide of the exon, numbered according to Persico et al. (1986). Underlined and overlined sequences correspond to oligonucleotide sequences used in in vitro amplification experiments. Asterisks indicate nucleotides for which substitutions have so far been observed. (B) Exons are represented as boxes including the identification code. Introns are represented by horizontal lines. Figures indicate length, in base pairs, of the corresponding sequence. Horizontal arrows indicate the position of primers used for in witro amplification experiments. Amplified regions are bracketed. Vertical arrows indicate the approximate position of nucleotide substitution analysed in this study.
G. VIGLIETTOAND
8
OTHERS
Table 3. Allele-speci$c oligonucleotides used for the detection of single ,nucleotide substitutions
........................................................................ oligonucleotide E4 MT2 wt wt sequence
TGGAAGTAGCACCCGATAC
.....AAACACCTTCATCGTGGGCTATGCCCG.....
GACAAGGCCCTACCGGAAG
oligonucleotide E4 MT3 wt wt sequence
.....GTGGCTGTTCCGGGATGGCCTTCTGCC.....
oligonucleotide E5 wt wt sequence
.....CAACAGCCACATGAATGCCCTCCACCT .....
oligonucleotide E6 wt wt sequence
ACCACATCTCCTCCCTGTT TCCAACCACATCTCCTCCCTGTTCCGT... ..
G6PD Mediterranean 1 sequence oligonucleotide E6 mut
AGCCACATGAATGCCCTCC
:
oligonucleotide E8 wt wt sequence
T TGGTGTAGAAGAGGGACAA
.....
TACTGCAGGCACTACTCTT
.....TCAGATGACGTCCGTGATGAGAAGGTA.... .
oligonucleotide Ell wt
TGACGCCTACGAGCGCCTC
wt sequence
G6PD Mediterranean 2 sequence oligonucleotide E l l mut
..... .....
:
.....TCCCTGACGCCTACGAGCGCCTCATCC..... ..... T ..... ACTGCGGATACTCGCGGAG
For mutant sequences only the variant nucleotide is reported. G6PD A(-) 1 and 2 refer to substitutions occurring a t exons 4 and 5 respectively ; G6PD Mediterranean 1 and 2 refer to substitution occurring a t exons 6 and 11 respectively (Table 2).
A+
Mediterranean silent
Mediterranean
-
+ +-
+
-
-
+ +
-
+ +
-
+ + + +-
-
MTlO
MT9
MT8
+
-
MT13
+
+ ++ +-
-
MT15
-
+ +
MTl6 -
+ + -
+
-
-
MT17
+ ++ ++ + -
MT19
+ +
-
+ ++ +
-
-
MT20
+ ++ ++ + -
-
wt
+ +
+ +
+
-_
+ +
-
-
+ +
+
-
+ + -
A(-) Cagliari- Mont- MetaMatera like 2 albano p n t o -
Control plasmid
E8wt E8mut E6wt E6mut Ellwt E l lmut MT8, MT9, MT10, and Cagliari-like 2 are representative of G6PD Cagliari-like. MT13 and Montalbano are representative of GGPD Montalbano. MT15 is representative of G6PD Tursi. MTl6 and Metaponto are representative of G6PD Metaponto. MT17 and A( - ) Matera are representative of GGPD Matera. MT19 and MT20 are representative of GGPD Pisticci. Control plasmids containing the EcoRI genomic DNA fragment spanning the relevant mutation (Martini et al. 1986).
E5wt E5mut
Montalbano
oligo E4wt E4mut E4wt E4mut
Variant Metaponto
Table 4.Allele-specific oligonucleotide analysis of G6PD variants
W
k 8 F
2
(b
E
3
8'
8
Q
3.
s
3 3
10
G. VIGLIETTO AND
OTHERS
Fig. 2. Allele-specific oligonucleotide hybridization on amplified genomic DNA. (A) PCR-amplified exon 5 either from genomic DPU’A or plasmids were electrophoresed, blotted in duplicates and hybridized to wild-type or mutant probes as indicated. Samples 1 and 3 were amplified from cloned plasmids and samples 2, 4, 5, 6, 7, 8 from genomic DKA. GGPD variants are the following: (1) G6PD B, (2) Matera A(-), (3)G6PD B, (4)MT8, (5)MT19, (6) MT15, (7) MT17, (8) G6PD B. (B) Same as in (A)except that exon 4 was amplified. Probes are used as shown and samples are (1) Matera A( -), (2) G6PD B, (3) MT17.
Common GGPD variants in Southern Italy
11
sequence to be substituted by a His in the variant. The G6PD sequence of both Cagliari-like subjects carried a C/G-+T/A transition in exon 6 in the second position of codon 188, causing a Ser-+Phe change in the polypeptide sequence (Table 2). This mutation had also been observed to occur in one subject classified as G6PD Mediterranean (Vulliamy et al. 1988) and in five subjects variously classified as G6PD Mediterranean, Cagliari and Sassari (De Vita et al. 1989). Both Cagliari-like subjects had a second nucleotide substitution in their sequences, a C/G--+T/A transition a t the third position of codon 437 (Table 2). This silent mutation also occurs in the five Sardinian subjects reported by De Vita et al. (1989) but not in the G6PD Mediterranean from Calabria reported by Vulliamy et al. (1988). In order to demonstrate that the nucleotide substitution shown by DNA sequencing of an index case was shared by subjects having a similar phenotype, for each subject the DNA region encompassing each point mutation was in vitro amplified and probed with allele specific oligonucleotides (ASOH) as follows. We applied ASOH on nine samples, not otherwise analysed a t the molecular level, namely two G6PD Pisticci (MT 19, 20), one G6PD Montalbano (MT 13), three G6PD Cagliari-like (MT 8, 9, lo), one G6PD Tursi (MT 15) and one G6PD A( - ) Matera (MT 17), in order to identify the molecular lesion responsible for the variant phenotype and to provide molecular support to the proposed classification of G6PD( - ) subjects based on the biochemical characteristics of the G6PD enzyme (Table 1). Eight 20-bp-long oligonucleotides complementary to intronic sequences flanking on both sides exons 3 4 , 5-6, 8 and 10-1 1 were synthesized and used as primers for DNA amplification reactions (Fig. 1 B). Nineteen-bp-long oligonucleotides complementary to the wild-type and mutated sequence of G6PD variants A( - ), Montalbano, Metaponto and Mediterranean (Table 3) were used as probes for ASOH experiments. As a rule, biochemical classification of G6PD variant types was confirmed by ASOH (Table 4). In fact, the subjects with G6PD Cagliari-like, Metaponto and Montalbano were all found to have the mutation shown by the sequenced index case and, as expected, no known mutation could be identified in carriers of G6PD Tursi or Pisticci. A notable exception was found with regard to G6PD A( - ) Matera. Indeed, while one of the subjects, whose sequence had previously been reported by Vulliamy et al. (1988), showed both mutations described in A( - ) alleles (Hirono & Beutler, 1988), the other, MT 17, showed a positive hybridization signal when probed with the A + specific oligonucleotide but not when probed with the ( - ) specific oligonucleotide, in spite of a typical G6PD A( - ) biochemical behaviour (Table 4 and Fig. 2).
DISCUSSION
The reduced level of G6PD activity in red blood cells is a relatively common trait in several Italian populations. The highest proportion of G6PD( - ) subjects is found in Sardinia where approximately 13Y0 of the male population is affected. G6PD deficiency is also common in Sicily, in Southern Italy and in the Po delta area where the proportion of affected males ranges from 2.2 to 2.5% (Colonna-Romano et al. 1985; Rickards et al. 1988; Calabrb et al. in preparation). On the basis of the enzyme level found in red cell extracts it is convenient to separate samples with a severe reduction of the level of enzyme activity (less than 5% of the normal) and samples with a moderate reduction 15-50% of the normal (Rickards et al. 1988). The relative proportion of these two phenotypic classes varies considerably from one population
12
G. VIGLIETTO AND
OTHERS
to another. Severe deficiency accounts for 30% of the deficiency in the population of Naples (Colonna-Romano et al. 1985), for 70% in the population of Matera (Calabrb et al. in preparation ) and for as much as 96% in the Sardinian population of Sassari (Testa et al. 1980). An even greater degree of heterogeneity becomes manifest when biochemical characteristics of partially purified preparations of G6PD are compared. By this means, four common variants were observed to be segregating in the Xardinian population, namely G6PD Mediterranean, Sassari, Cagliari and Seattle-like (Lenzerini et al. 1969; Testa et al. 1980; Fenu et al. 1982), eight were found in a random sample of the population of Naples (Colonna-Romano et al. 1985), and now this paper reports six in the population of the Matera district. Sequencing of genomic DNA fragments and hybridization analysis of in v i t r o amplified genomic DNA with allele specific probes confirm the biochemical findings for the population of Matera in the majority of the cases. G6PD Pisticci and G6PD Tursi are two apparently new variants, associated with a very mild or a moderate degree of deficiency (Table 1). Their molecular lesion has not yet been determined. An additional new variant appears to be G6PD Montalbano. It is a slow mobility variant associated with moderate deficiency and apparently normal kinetic properties (Table 1). It is generated by a G/C+A/T transition in the second position of codon 285 resulting in an Arg-His amino-acid substitution (Table 2). G6PD Metaponto has already been reported in this population (Vulliamy et al. 1988). It is caused by a G/C+A/T transition in exon 4 a t residue position 172 resulting in an Asp-Asn substitution. The properties of G6PD A( - ) Matera are identical to those reported for a variant detected in the same population in a subject with acute haemolytic anaemia (Calabrb et al. 1989) and found by molecular analysis to bear the same mutations of G6PD A( - ) common in populations of African ancestry (Vulliamy et al. 1988). The two subjects included in this group share the same 376 A/T-+G/C substitution, reported also in G6PD A subjects. However, mutation 202 G/C-A/T, specific to the A( - ) variant, is absent in one of the subjects (MT 17), suggesting that two different mutations have occurred on a pre-existing GGPD A gene to give rise to two deficient variants with very similar properties. A similar finding has recently been reported for Afro-Americans (Hirono & Beutler, 1988). It is, therefore, likely that the two A ( - ) variants found in the population of Matera are a remnant of ancient events of genetic admixture with African populations. The variant we have provisionally named G6PD Cagliari-like is characterized by two mutations on the nucleotide sequence. The one occurring in exon 6 is a C/G-T/A transition, determining a phenylalanine for serine amino acid substitution a t residue 188 of the protein Sequence. The substitution occurs at a site only ten amino acids downstream of an evolutionallyconserved sequence (Jeffery et al. 1985; Persico et al. 1986; Fouts et al. 1988; Ho et al. 1988). The sequence contains a lysyl residue considered to be important for the correct positioning of G6P within the G6PD molecule and hence essential for catalytic activity (Camardella et al. 1981, 1988; Jeffery et al. 1985). A substitution in this region is, therefore, expected to affect the catalytic properties. The observed substitution of an acidic residue with a bulky aromatic amino acid must, however, also introduce a considerable distortion in the molecule structure since in vivo and in witro stability of the molecule is greatly impaired. This very same mutation has also been reported in a subject with a typical G6PD Mediterranean variant (Vulliamy et al. 1988) and, more recently, in the biochemically defined G6PD Cagliari and G6PD Sassari from
Common G6PD variants in Southern Italy
13
Table 5 . Nucleotide substitutions of GGPD variants affect restriction mdonuclease recognition sequences (RERS) G6PD variant Metaponto ( A - ) Matera Montalbano
Wild-type sequence GGATC CGTG GAATG CGTG
Mutated sequence GAATG CATV GGATG CATG
RERS GAATG CATC GGATG CATG
Enzyme Fok I N a I11 Fok I Nla I11
the Sardinian population (De Vita et al. 1989). It is not clear as yet by which mechanism G6PD proteins with different biochemical properties (Table 1 ) can be generated from apparently identical genes. I n all the five G6PD Cagliari-like variants we found a second C/G+T/A transition in cxon 11. The mutation occurred in the third position of codon 437 and, therefore, did not result in an amino-acid substitution. This mutation had not been found in the previously reported sequence of a G6PD Mediterranean subject from the neighbouring area of Calabria (Vulliamy et al. 1988) but is reported to occur in all Sardinian subjects described by De Vita et al. (1989). Further typing of G6PD deficient subjects will clarify whether the subject reported by Vulliamy et al. (1988) is unique or whether the codon 437 mutation is itself a polymorphism within the G6PD Mediterranean variants. Of the six single nucleotide substitutions so far reported in Southern Italy G6PD variants, three can, in principle, be recognized by digesting genomic DNA with restriction endonucleases (Table 5). As an alternative t o ASOH, in vitro DNA amplification followed by DNA fragmentation with a specific enzyme may thus be introduced for the recognition of these variants in hemizygous or heterozygous condition. We thank Mrs Maria Terracciano, Mrs Rosaria Terracciano and Mrs Carmela Salzano for their excellent technical assistance. This work was supported by grants from the Italian Ministry of Education and from the National Research Council to G. B., from the Progetto Finalizzato Ingegneria Genetica CNR and from the EEC (No. ST2,T-0172-2-1)to M.G. P.
REFERENCES
ALLISON,A. C . (1960).Glucose-6-phosphate dehydrogenase deficiency in red blood cells of East Africans. Nature 186,531. BATTISTUZZI, C.,D’URSO,&I., TONIOLO,D., PERSICO, G . M. & LUZZATTO, L. (1985). Tissue specific levels of human glucose-6-phosphate dehydrogenase correlate with methylation of specific sites a t the 3’ end of the gene. Proc. Natl. Acad. Sci. USA 82,1465-1469. BATTISTUZZI, G.,ESAN,G . J. F., FASUAN, F. A., MODIANO,G., & LUZZATTO, L. (1977).Comparison of GdA and GdR activities in Nigerians. A study of the variation of glucose-&phosphate dehydrogenase activity. Am. J . H’um. Genet. 29, 3 1-36. BENTON, W. T). & DAVIS, R . W. (1977). Screening Agt recombinant clones by hybridisation to single playues in situ. AEcience 196, 180-182. BETKE,K., BREWER, G. J., KIRKMAN, H. N., LUZZATTO, L., MOTULSKY,A. G . , RAMOT, B. & SINISCALCO, M. (1967).Standardization of procedures for the study of glucose-&phosphate dehydrogenase : report of a WHO scientific group. WHO Tech. Rep. Series 366. BEUTLER, E. (1983).Glucose-6-phosphate dehydrogenase deficiency. In The Metabolic Basis of Inherited Disease, 5th edn (ed. J. B. Stanbury, J. B. Wyngaarden, D. 8. Fredrickson, J . L. Goldstein & M. S. Braun). pp. 1629-1653. New York: McGraw-Hill.
14
G. VIGLIETTO AND
OTHERS
BEUTLER, E., MATHAI,C. K. & SMITH,J. E. (1968). Biochemical variants of glucose-6-phosphate dehydrogenase giving rise to congenital nonspherocytic hemolytic disease. Blood 31, 131-150. C A L A B R ~V., , CASCONE,A., MALASPINA,P. & BATTISTUZZI, G. (1989). Glucose-6-phosphate dehydrogenase deficiency in Southern Italy: a case of G6PD A( - ) associated with favism. Haemutologica 74 (in the press). CAMARDELLA, L., ROMANO, M., DI PRISCO, G., DESCALZI-CANCEDDA, F. (1981). Human erythrocyte glucose-6phosphate dehydrogenase : labelling of a reactive lysyl residue by pyridoxal-5’-phosphate. Biochem. Biophys. Res. Commun. 103, 1384-1389. CAMARDELLA, L., CARUSO,C., RUTIQLIANO, B., ROMANO, M., DI PRISCO, G. & DESCALZI-CANCEDDA, F. (1988). Human erythrocyte glucose-6-phosphate dehydrogenase. Identification of a reactive lysyl residue labelled with pyridoxal-5‘-phosphate. Eur. J . Biochem. 171, 485-489. CHILDS,B., ZINKHAM, W., BROWN,E. A,, KIMBRO,E. L. & TORBET, J . V. (1978). A genetic study of a defect in glutathion met,abolism of the erythrocyte. Bull. Johns Hopkins Hospital 102, 21. COLONNA-ROMANO, S., IOLASCON, A., LIPPO, S., PINTYO, L., CUTILLO,S. & BATTISTUZZI, G. (1985). Genetic heterogeneity at the glucose-6-phosphate dehydrogenase locus in Southern Italy : a study on the population of Naples. Hum. Genet. 69, 228-232. M., SANPIETRO, M., CAPPELLINI,M. D., FIORELLI, G. & TONIOLO, D. (1989). Two point DE VITA,G., ALCALAY, mutations are responsible for G6PD polymorphism in Sardinia. Am. J. Hum. Genet. 44,233-240. FENU,M. P., FINAZZI, G., MANOUSSAKIS,C., PALOMBA, V. & FIORELLI, G. (1982). Glucose-6-phosphate dehydrogenase deficiency : genetic heterogeneity in Sardinia. Ann. Hum. Genet. 46, 105-1 14. POUTS, D., GANQULY, R., GUTIERREZ, G. A., LUCCHESI, J. C. & MANNINQ,J. E. (1988). Nucleotide sequence of the Drosophila glucose-6-phosphate gene and comparison with the homologous human gene. Gene 63, 261-275. HIRONO,A. & BEUTLER, E. (1988). Molecular cloning and nucleotide sequence of cDNA for human glucose-6phosphate dehydrogenase variant A( -). Proc. Natl. Acad. Sci. USA 85, 3951-3954. Ho, Y. S., HOWARD, A. J. & CRAPO,K. D. (1988). Cloning and sequence of a cDNA encoding rat glucose-6phosphate dehydrogenase. Nucl. Acids Res. 15, 7746. JEFFERY, J., HOBBS,L. & JORNVALL, H. (1985). Glucose-6-phosphate dehydrogenase from Saccharomyces cerevisiae : characterization of a reactive lysine residue labelled with acetylsalicylic acid. Biochemistry 24, 66-7 1. KIRKMAN,H. N., DOXIADIS,S. A., VALEAS,T., TASSOPOULOS, N. & BRINSON,A. G. (1965). Diverse characteristics of glucose-6-phosphate dehydrogenase from Greek children. J . Lab. Clin. Med. 65, 212-221. LENZERINI,L., MEERA KHAN,P., FILIPPI,G., RATTAZZI,M. C., KAY, A. K. & SINISCALCO, M. (1969). Characterization of glucose-6-phosphate dehydrogenase variants. I. Occurrence of G6PD Seattle-like variant in Sardinia and its interaction with the G6PD Mediterranean variant. Am. J. Hum. Genet. 21, 142-153. LUZZATTO, L. (1979). Genetics of red cell and susceptibilitv to malaria. Blood 54, 961-976. LUZZATTO, L. & BATTISTUZZI, G. (1985). Glucose-6-phosphate dehydrogenase. Advances in Human Genetics 14, 217-329. MANIATIS,T., FRITSCH, E. F. & SAMBROOK, J. (1982). Molecular Cloning. Cold Spring Harbor Laboratory. MARTINI, G., TONIOLO, D., VULLIAMY, T., LUZZATTO, L., DONO,R., VIQLIETTO, G., PAONESSA, G., D’URSO,M. & PERSICO, M. G. (1986). Structural analysis of the X-linked gene encoding human glucose-6-phosphate dehydrogenase. EMBO J. 5, 1849-1855. MODIANO,G., BATTISTUZZI, G., ESAN,G. J. F., TESTA,U. & LUZZATTO, L. (1979). Genetic heterogeneity of ‘normal ’ erythrocyte glucose-6-phosphate dehydrogenase : an isoelectrophoretic polymorphism. Proc. Natl. Acad. Sci. USA 76, 852-856. MOTULSKY,A. G. (1960). Metabolic polymorphism and the role of infectious diseases in human populations. Hum. Biol. 32, 28-62. PAI,G. S., SPRENKLE, A, , DO, T. T., MARENI, C. E. & MIQEON,B. R. (1980).Localisation of loci for HPRT and G6PD and biochemical evidence of non-random X-chromosome expression from studies of a human Xautosome translocation. Proc. Natl. Acad. Sci. USA 77, 2810-2813. PERONA, G., GUIDI, G. C., TUMMARELLO, D., MARENI, C., BATTISTUZZI, G. & LUZZATTO, L. (1983). A new glucose-6-phosphate dehydrogenase variant (G6PD Verona) in a patient with myelodysplastic syndrome. S c a d . J. Haematol. 30, 407414. PERSICO, M. G., VIQLIETTO,G., MARTINI, G., TONIOLO,D., PAONESSA, G., MOSCATELLI, C., DONO,R., VULLIAMY, T., LUZZATTO, L. & D’URSO,M. (1986). Isolation of human glucose-6-phosphate dehydrogenase cDNA clones: primary structure of the protein and unusual 5’ coding region. Nucl. Acids Res. 14,2511-2522, 7822. RICKARDS, O., BIONDI,G., DE STEFANO, G. F. & BATTISTUZZI, G. (1988). Distribution of genetically determined deficient variants of glucose-6-phosphate dehydrogenase (G6PD) in Southern Italy. In Advances in Haenwgenetics, Vol. 2 . (ed. W. R. Mayr). Berlin, Heidelberg: Springer Verlag. SANQER, F., NICKLEN,S. L COULSON, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.
Common
GGPD variants in Southern Italy
15
SINISCALCO, M., BERNINI, L., FILIPPI, G., LATTE,B., MEERA KHAN,P., PIOMELLI, S. & RATTAZZI, M. (1966). Population genetics of haemoglobin variants, thalassaemia and glucose-6-phosphate dehydrogenase deficiency with particular reference to the malaria hypothesis. Bull. WHO 34, 379-385. P., PURRELLO, M., ROCCHI, M., ARCHIDIACONO, N., ALHADEFF, B., FILIPPI, G., TONIOLO, D., MARTINI, SZABO, G., LUZZATTO, L. & SINISCALCO, M. (1984). Cytological mapping of the human glucose-6-phosphate dehydrogenase gene distal to the fragile-X site suggests a high rate of meiotic recombination across this site. Proc. Natl. Acad. Sei. USA 81, 7855-7859. TAKIZAWA, T., HUANG, I. Y., IKUTA, T. & YOSHIDA, A. (1986). Human glucose-6-phosphate dehydrogenase: primary structure and cDNA cloning. Proc. Natl. Acad. Sci. USA 83, 41574161. A,, BATTISTUZZI, G., CUTILLO,S. & LUZZATTO, L. (1980).Genetic heterogeneity TESTA, U., MELONI, T., LANIA, of G6PD deficiency in Sardinia. Hum. Genet. 56, 99-105. R. B. (1986). The use of synthetic oligonucleotides as specific hybridisation probes THEIN,S. L. & WALLACE, in the diagnosis of genetic disorders. In Human Genetic Diseases: A Practical Approach (ed. K. E. Davies). Oxford, Washington : IRL Press. VULLIAMY,T. J., D’URSO,M., BATTISTUZZI, G., ESTRADA, M., FOULKES, N. S., MARTINI, G., C A L A B R ~V., , POGGI, V., GIORDANO, R., TOWN, M., LUZZATTO, L. & PERSICO, M. G. (1988).Diverse point mutations in the human glucose-6-phosphate dehydrogenase gene cause enzyme deficiency and mild or severe hemolytic anemia. Proc. Natl. Acad. Sci. USA 85, 5171-5175.
1
Printed in Great Britain
Common glucose-6-phosphate dehydrogenase (G6PD) variants from the Italian population : biochemical and molecular characterization G. VIGLIETTO," V. MONTANARO,t* V. CALABRO,? D. VALLONE,? M. D'URSO,* M. G. PERSICO" AND G. BATTISTUZZI? * Istituto Internazionale d i Genetica e Biojisica, CNR, Naples, Italy Dipartimento d i Genetica, Biologia Generale e Molecolare, UniversitcE d i Napoli, Naples, Italy
SUMMARY
By biochemical characterization of glucose-6-phosphate dehydrogenase (G6PD)from the red cells of seventeen subjects of the population of Matera (Southern Italy) we have identified six genetically determined common variants. Among these, G6PD Metaponto and G6PD A( - ) Matera had been already fully characterized. We have now found that A(-) Matera is genetically heterogeneous since one of two subjects examined had the two mutations at codons 68 and 126 characteristic of a typical A( - ) variant, while the other subject had only the codon 126 mutation. G6PD Pisticci and G6PD Tursi are two new variants whose molecular lesion is not yet known. G6PD Cagliari-like has biochemical characteristics reminiscent of G6PD Cagliari, isolated in Sardinia, and was found to have the same nucleotide substitution as G6PD Mediterranean. G6PD Montalbano is a new variant, with nearly normal properties, due t o a G-tA transition which causes an Arg+His amino acid replacement a t position 285.
INTRODUCTION
Glucose-6-phosphate dehydrogenase (G6PD) is the first and rate-limiting enzyme of the hexose monophosphate pathway. I t s metabolic role, the provision of adequate amounts of NADPH for biosynthetic or detoxifying reactions, is probably essential to all organisms and cell types (Luzzatto & Battistuzzi, 1985). Human G6PD is encoded by an X-linked gene (Childs et al. 1978). It has been mapped near the end of the long arm of the X chromosome a t band Xq28 by somatic cell hybrid analysis (Pai et al. 1980) and cDNA probed by in situ hybridization experiments (Szabo et al. 1984). The G6PD coding sequence and genomic structure have recently been elucidated (Persico et al. 1986; Takizawa et al. 1986; Martini et al. 1986). The G6PD locus is polymorphic in several human populations. Extensive biochemical characterization of variant phenotypes has led to the notion that polymorphism in different populations is often due to the occurrence of different genetically determined G6PD variants. I n fact, since the pioneering work by Kirkman and coworkers (Kirkman et al. 1965), who first reported evidence of heterogeneity within the so-called G6PD-Mediterranean phenotype in the While this manuscript was in preparation, Professor Giorgio Battistuzzi passed away, leaving a profound sense of sorrow in his friends, colleagues and students. We have lost an excellent geneticist whose invaluable contribution in the analytical study of the G6PD gene variability will remain a guideline for us. Address for correspondence and reprints : Dr M. Graziella Persico, Istituto Internazionale di Genetica e Biofisica, CPU'R, Via Marconi. 10. 80125 Naples, Italy. I
H(: I.: 51
2
G. VIGLIETTO AND
OTHERS
population of Greece, evidence has been accumulating that within a population the degree of genetic heterogeneity may be considerable. As a result, among the more than 300 G6PD variants so far described, 87 appear to be common in a t least one population (Luzzatto & Battistuzzi, 1985). At variance with most human red cell enzyme polymorphisms (Battistuzzi et al. 1977), common inherited variants at the G6PD locus are mostly associated with an abnormally low level of enzyme activity. The enzyme deficiency is clinically relevant since it is often associated with the risk of acute haemolytic episodes, triggered by a variety of exogenous agents (Beutler, 1983). On the other hand, G6PD variants associated with altered levels of enzyme activity appear to have adaptive value in malarial environments (Allison, 1960; Motulsky, 1960 ; Siniscalco et al. 1966; Luzzatto, 1979). Identification of the amino-acid substitution, which is responsible for the altered properties of G6PD variants, can provide information on the genetic relationships between different variants and on the relationship between structure and function of the protein molecule. The G6PD protein is, however, produced at a relatively low level in all cell types, making it virtually impossible to obtain sufficient amounts for sequence analysis, especially in cases of enzyme deficiency. Recently, however, thanks to the availability of the G6PD B DNA sequence (Persico et al. 1986; Takizawa et al. 1986) and of improved methods for DNA cloning and sequencing, it has become possible to determine rapidly and reliably the specific alteration in the DNA sequence, and the amino-acid substitution thereof (Hirono & Beutler, 1988 ; Vulliamy et al. 1988). This paper reports the biochemical and molecular characterization of six G6PD genetic variants commonly occurring in the population of Southern Italy.
MATERIAL AND METHODS
Male carriers of a variant G6PD phenotype had been identified during a population survey in the Matera district, Southern Italy (Calabrb et al. in preparation). From each subject a 20-30 ml blood sample was obtained by venepuncture, collected into EDTA containing tubes and processed within 24 h after collection. To each specimen was added a 0.1 volume of a 3% (w/v) solution of dextran (mol. wt 200300 kDa). After mixing by inversion, the sample was allowed to sediment for 15-20 min at room temperature, The plasma fraction was then transferred to a fresh tube and spun a t 2000 r.p.m. for 10 min in a table centrifuge. The supernatant fraction and red blood cells were reunited and thoroughly mixed by inversion. This procedure was repeated twice more. The resulting pellets containing white blood cells were resuspended into calcium-magnesium-free phosphatebuffered saline (PBS), reunited and centrifuged a t 1600 r.p.m. for 6 min. The pellets were washed once more with PBS and kept frozen a t - 80 OC until the DNA was extracted according to the procedure of Battistuzzi et al. (1985). Red blood cells were transferred to an ice bath and washed three times with cold PBS. Lysates for G6PD activity and purification were obtained according to the procedure of Battistuzzi et al. (1977). Red-blood-cell activity levels and electrophoretic phenotypes were determined according to Battistuzzi et al. (1977) and Betke et al. (1967), respectively. Partial purification and biochemical characterization of the G6PD enzyme protein were perfckrned as
Common GGPD variants in Southern Italy
3
previously described (Betke et al. 1967; Beutler et al. 1968; Modiano et al. 1979; Perona et al. 1983). Cloning and sequencing of variant genes DNA samples (250 pg) from male carriers of variant G6PD were digested to completion with EcoRI restriction endonuclease and size-fractionated on 10-40 % sucrose gradients. DNA from fractions positively hybridizing to G6PD specific DNA probes were cloned into a h GTWes phage. h GTWes EcoRI arms were obtained from Bethesda Research Laboratories. Ligated material was packaged using Stratagene extracts as recommended by the suppliers. The libraries were screened by filter hybridization (Benton & Davis, 1977) using G6PD cDNA probes labelled by nick translation. Phages bearing G6PD specific inserts were isolated and amplified (Maniatis et al. 1982). Inserts were subcloned into a pUC18 plasmid vector using E . coli strain HBlOl as a recipient cell for sequencing by the dideoxy method (Sanger et al. 1977). Sequencing reactions were primed by using a set of oligonucleotides homologous to sequences immediately flanking intron-exon junctions (Fig. 1 ) . Evaluation of genomic DNA by using allele speci$c oligonucleotide probes BamHI-digested, ethanol-precipitated genomic DNA ( 1 pg) was dissolved in a final volume of 100 p1 of 10 mM Tris-HC1 p H 8.3, 50 mM-KCI, 2 mM-MgC1 and 0.01 % gelatine. Intron specific 20mers (0.5 pg each) were added and the mixture was incubated a t 100 "C for 15 min for DNA denaturation. Reannealing was then allowed to occur for 15 min at room temperature. dNTPs were added a t a final concentration of 0 2 mM each. The polymerization reaction was started by the addition of 1 U of Taq-polymerase (Anglican Biotec. Ltd). DNA was allowed to elongate a t 65OC for 4min, denatured a t 95OC for 2 min and renatured a t 45°C for 2 min. The elongation-denaturation-renaturation cycle was repeated 30 times. At cycle nos. 10 and 20, 1 U of Taq-polymerase was added; a t cycle no. 20 dNTPs were added a t a final concentration of 0.2 mM each. Amplified DNAs were denatured by adding NaOH a t a final concentration of 60 mM, fractionated on a 1% alkaline agarose gel (Maniatis et al. 1982) and transferred to Z-bind nylon membranes (CUNO, Mass., USA). Allele specific 19mers (Table 3) were chemically synthesized with an Applied Biosystem Synthesizer and 5-labelled with [y-32P]ATPby T4 polynucleotide kinase (Maniatis et al. 1982). Hybridization of allele specific oligonucleotides to filter immobilized amplified genomic DNA was allowed to proceed for 24 h in 5 x SSPE, 5 x Denhardt's, 0 3 % SDS and 200 pg/ml yeast total RNA, a t a temperature calculated as proposed by Thein & Wallace (1986).The filters were washed with 6 x SSC for 60 min at room temperature with 3 changes, for 2 min a t the hybridization temperature and again for 60 min at room temperature. The washed filters were air-dried and auto-radiographed for 1-2 h a t - 80 "C with intensifying screens.
RESULTS
During a screening of the population of Matera district (Southern Italy), we had identified, out of 1525 boys (5-10 years old), 42 subjects with altered red cell G6PD electrophoretic mobility and/or enzyme activity (Calabrb et al. in preparation). I n 17 cases i t was possible to
'7 24 3 9
Tursi
Matera ( A - )
Cagliari-like
Posillipo
I00
IOO
90
I00
111
93
93
K, K, NADP PJADPH ~
Substrate analogues utilization*
DEAE5
x K
n.d. n.d. S
n.d.
x
n.d. n.d.
n.d.
E ?I'
h'
8
c3
Y 0
Z
0 k
c3
e
M
G
c A
? L
K
A
PH' ?I'
h'
T6
In table mean value sf^.^. are reported. The number of determinations on which values are based is reported in parentheses below each value. Kumber of subjects. 'Enzyme activity in red blood cells is expressed as pM/g H b (IU) and as percent of normal (%). Electrophoretic mobility as percent of normal q6PD B. Assignment to each electrophoretic class has been done on a side-by-side comparison basis. Expressed as percent of natural, G6P and NADP, substrates. 5Expressed as the mM concentration of chloride ions at peak elution. Thermostability is rJ for normal ; L for lower than normal stability. 'pH dependence is pu' for normal; A for altered profile, the variant shows a broad p H maximum extending from pH 7.5 to pH 9.5.
GGPD B
5.46 + O I 6
21
Montalbano
0-7
20
Metaponto
Mediterranean
90
69 90
EM3
%
Kame of variant Pisticci
Activity'
Table 1. Comparison of the biochemical characteristics of GGPD genetic variants found in the populution of Southern Italy
Common G6PD variants in Southern Italy
_158
1?0_ C
C
T
C
C
'
P
G
T
C
C
C
C
A
C
C
C
A
5
C
T
T
C
T
M
~
C
A
C
A
C
A
C
C
~
~
C
C
C
A
C
C
T
CAlTCACACAGGCTCC
,269 C A C T C C C T G A C A T C C C C A M C A C A C ' P G A C C C C C T T C T T C M C C ~ ~ T C T C A ~ C T C C C C ~ C C C ~ T T C ~ C C C T C T C T A C C A G - - - - - - - - - -I -V- S- -I -. - - - - - - - - - ?
CGACCGCCTCCCACCM
270, M
G
A
C
A
G
C
C
C
C
T
C
A
C
A
T
C
T
C
T
C
T
C
~
~
T
C
~
T
C
T
C
C
C
T
C
A
C
T
A
C
* AnATGCAGCCTCCTACCACCCCCTCMCACCCACA~M~CCTCCACC~TCACACCCCMCCCCCKTTCTACC~CCTn;CCCCCCACCCTCTACCACCCCCTCACCMCA
,481 ACATTCACCACTCCTCCATCACCCACA~TMCCC~CCTn;CCCTC~C'CCCCCCT
- - - - - - - - - - - - IVS5 - - - - - - - - - -
ACTCCCCGMCACCCCTKMCCCCCTMC
MCCGCAGGCMGCCCCCA
*88, CCACCCTCCCU;CTCCCACCACACCCn;CMCCCCATCn iCCn;CACMCCCCTTCGGCACGCACC'PGCACACCTC~ACCCCC~TCCMCCACATC~CTCCC~~CCC~ACGACC
C
G. VIGLIETTO AND
6
OTHERS
550
Ill
EZ
560
e +
2:1
J
5'0
E5
El2
E l l
Common G6PD variants in Southern Italy
7
Table 2. Nucleotide substitutions found in the coding sequence of GGPD gene for variants from Southern Italy population Exon Codon Amino-acid Reference G6PD variant no. position Codon change substitution Metaponto 4 58 GAT + AAT Asp + Asn Vulliamy et al. ( I 988) Matera (A-) 4 68 GTG-rATG + Met Vulliamy et al. (1988) Matera (A - ) 5 126 AAT-GAT Am-Asp 6 188 TCC TTC Cagliari-like 1 Ser -+ Present paper Cagliari-like 1 11 437 TAC-rTAT Tyr + Tyr Cagliari-like 2 6 188 TCC + TTC Ser --* Present paper 11 437 TAC-TAT Tyr-Tyr Cagliari-like 2 Montalbano 8 285 CGT +CAT Arg + His Present paper Mediterranean 6 188 TCC -+ TTC Ser --t Vulliamy et al. (1988) 11 437 TAC-TAT Tyr -r Tyr Mediterranean Codon position number refers to the cDNA sequence reported by Persico et al. (1986) (see also Fig. 1).
-
partially purify red cell G6PD in order to carry out a full characterization (see Table 1). This has enabled us to distinguish six new G6PD variants which were provisonally named G6PD Pisticci, G6PD Metaponto, G6PD Montalbano, G6PD Tursi, G6PD A( - ) Matera and G6PD Cagliari-like. Their properties are summarized in Table 1 together with those of G6PD Posillipo and G6PD Mediterranean, two variants previously observed in Southern Italy (ColonnaRomano et al. 1985), and those of the ‘normal’ control G6PD B. I n each case the variant phenotype was seen to be transmitted from mother to son (data not shown). The G6PD of subjects classified as carriers of G6PD Metaponto, G6PD A( - ) Matera and GGPD Cagliari-like have properties very similar to those of the homonymous variants previously described by Vulliamy et al. (1988), by Fenu et al. (1982) and by De Vita et al. (1989). To be able to determine the nucleotide mutations occurring in the different variants we extended the nucleotide sequence analysis of the wild type G6PD gene reported by Persico et al. (1986) and Martini et al. (1986) inside the intron sequence. Figure 1A shows the sequence of 3650 nucleotides of the 18 kb region of the G6PD gene comprehensive of the 13exons, the intron flanking sequences, the 5’ and 3’ non-coding ends. From this sequence oligonucleotide sequences (underlined in Fig. 1A) were derived to perform the sequence analysis by the dideoxy method. Sequence analysis of exonic DNA fragments was performed on one subject carrier of G6PD Montalbano and on two subjects with G6PD Cagliari-like. A single nucleotide substitution was found in the sequence of G6PD Montalbano, a G/C+A/T transition in the second position of codon 285 (Table 2). The nucleotide change caused an Arg residue on the wild-type polypeptide Fig. 1. Nucleotide sequence of GBPD gene exonic and flanking intronic DNA and PCR strategy. (A) The updated sequence of exons and of neighbouring intronic regions is reported. Exons are bounded by arrows. Figures above arrows indicate first and last nucleotide of the exon, numbered according to Persico et al. (1986). Underlined and overlined sequences correspond to oligonucleotide sequences used in in vitro amplification experiments. Asterisks indicate nucleotides for which substitutions have so far been observed. (B) Exons are represented as boxes including the identification code. Introns are represented by horizontal lines. Figures indicate length, in base pairs, of the corresponding sequence. Horizontal arrows indicate the position of primers used for in witro amplification experiments. Amplified regions are bracketed. Vertical arrows indicate the approximate position of nucleotide substitution analysed in this study.
G. VIGLIETTOAND
8
OTHERS
Table 3. Allele-speci$c oligonucleotides used for the detection of single ,nucleotide substitutions
........................................................................ oligonucleotide E4 MT2 wt wt sequence
TGGAAGTAGCACCCGATAC
.....AAACACCTTCATCGTGGGCTATGCCCG.....
GACAAGGCCCTACCGGAAG
oligonucleotide E4 MT3 wt wt sequence
.....GTGGCTGTTCCGGGATGGCCTTCTGCC.....
oligonucleotide E5 wt wt sequence
.....CAACAGCCACATGAATGCCCTCCACCT .....
oligonucleotide E6 wt wt sequence
ACCACATCTCCTCCCTGTT TCCAACCACATCTCCTCCCTGTTCCGT... ..
G6PD Mediterranean 1 sequence oligonucleotide E6 mut
AGCCACATGAATGCCCTCC
:
oligonucleotide E8 wt wt sequence
T TGGTGTAGAAGAGGGACAA
.....
TACTGCAGGCACTACTCTT
.....TCAGATGACGTCCGTGATGAGAAGGTA.... .
oligonucleotide Ell wt
TGACGCCTACGAGCGCCTC
wt sequence
G6PD Mediterranean 2 sequence oligonucleotide E l l mut
..... .....
:
.....TCCCTGACGCCTACGAGCGCCTCATCC..... ..... T ..... ACTGCGGATACTCGCGGAG
For mutant sequences only the variant nucleotide is reported. G6PD A(-) 1 and 2 refer to substitutions occurring a t exons 4 and 5 respectively ; G6PD Mediterranean 1 and 2 refer to substitution occurring a t exons 6 and 11 respectively (Table 2).
A+
Mediterranean silent
Mediterranean
-
+ +-
+
-
-
+ +
-
+ +
-
+ + + +-
-
MTlO
MT9
MT8
+
-
MT13
+
+ ++ +-
-
MT15
-
+ +
MTl6 -
+ + -
+
-
-
MT17
+ ++ ++ + -
MT19
+ +
-
+ ++ +
-
-
MT20
+ ++ ++ + -
-
wt
+ +
+ +
+
-_
+ +
-
-
+ +
+
-
+ + -
A(-) Cagliari- Mont- MetaMatera like 2 albano p n t o -
Control plasmid
E8wt E8mut E6wt E6mut Ellwt E l lmut MT8, MT9, MT10, and Cagliari-like 2 are representative of G6PD Cagliari-like. MT13 and Montalbano are representative of GGPD Montalbano. MT15 is representative of G6PD Tursi. MTl6 and Metaponto are representative of G6PD Metaponto. MT17 and A( - ) Matera are representative of GGPD Matera. MT19 and MT20 are representative of GGPD Pisticci. Control plasmids containing the EcoRI genomic DNA fragment spanning the relevant mutation (Martini et al. 1986).
E5wt E5mut
Montalbano
oligo E4wt E4mut E4wt E4mut
Variant Metaponto
Table 4.Allele-specific oligonucleotide analysis of G6PD variants
W
k 8 F
2
(b
E
3
8'
8
Q
3.
s
3 3
10
G. VIGLIETTO AND
OTHERS
Fig. 2. Allele-specific oligonucleotide hybridization on amplified genomic DNA. (A) PCR-amplified exon 5 either from genomic DPU’A or plasmids were electrophoresed, blotted in duplicates and hybridized to wild-type or mutant probes as indicated. Samples 1 and 3 were amplified from cloned plasmids and samples 2, 4, 5, 6, 7, 8 from genomic DKA. GGPD variants are the following: (1) G6PD B, (2) Matera A(-), (3)G6PD B, (4)MT8, (5)MT19, (6) MT15, (7) MT17, (8) G6PD B. (B) Same as in (A)except that exon 4 was amplified. Probes are used as shown and samples are (1) Matera A( -), (2) G6PD B, (3) MT17.
Common GGPD variants in Southern Italy
11
sequence to be substituted by a His in the variant. The G6PD sequence of both Cagliari-like subjects carried a C/G-+T/A transition in exon 6 in the second position of codon 188, causing a Ser-+Phe change in the polypeptide sequence (Table 2). This mutation had also been observed to occur in one subject classified as G6PD Mediterranean (Vulliamy et al. 1988) and in five subjects variously classified as G6PD Mediterranean, Cagliari and Sassari (De Vita et al. 1989). Both Cagliari-like subjects had a second nucleotide substitution in their sequences, a C/G--+T/A transition a t the third position of codon 437 (Table 2). This silent mutation also occurs in the five Sardinian subjects reported by De Vita et al. (1989) but not in the G6PD Mediterranean from Calabria reported by Vulliamy et al. (1988). In order to demonstrate that the nucleotide substitution shown by DNA sequencing of an index case was shared by subjects having a similar phenotype, for each subject the DNA region encompassing each point mutation was in vitro amplified and probed with allele specific oligonucleotides (ASOH) as follows. We applied ASOH on nine samples, not otherwise analysed a t the molecular level, namely two G6PD Pisticci (MT 19, 20), one G6PD Montalbano (MT 13), three G6PD Cagliari-like (MT 8, 9, lo), one G6PD Tursi (MT 15) and one G6PD A( - ) Matera (MT 17), in order to identify the molecular lesion responsible for the variant phenotype and to provide molecular support to the proposed classification of G6PD( - ) subjects based on the biochemical characteristics of the G6PD enzyme (Table 1). Eight 20-bp-long oligonucleotides complementary to intronic sequences flanking on both sides exons 3 4 , 5-6, 8 and 10-1 1 were synthesized and used as primers for DNA amplification reactions (Fig. 1 B). Nineteen-bp-long oligonucleotides complementary to the wild-type and mutated sequence of G6PD variants A( - ), Montalbano, Metaponto and Mediterranean (Table 3) were used as probes for ASOH experiments. As a rule, biochemical classification of G6PD variant types was confirmed by ASOH (Table 4). In fact, the subjects with G6PD Cagliari-like, Metaponto and Montalbano were all found to have the mutation shown by the sequenced index case and, as expected, no known mutation could be identified in carriers of G6PD Tursi or Pisticci. A notable exception was found with regard to G6PD A( - ) Matera. Indeed, while one of the subjects, whose sequence had previously been reported by Vulliamy et al. (1988), showed both mutations described in A( - ) alleles (Hirono & Beutler, 1988), the other, MT 17, showed a positive hybridization signal when probed with the A + specific oligonucleotide but not when probed with the ( - ) specific oligonucleotide, in spite of a typical G6PD A( - ) biochemical behaviour (Table 4 and Fig. 2).
DISCUSSION
The reduced level of G6PD activity in red blood cells is a relatively common trait in several Italian populations. The highest proportion of G6PD( - ) subjects is found in Sardinia where approximately 13Y0 of the male population is affected. G6PD deficiency is also common in Sicily, in Southern Italy and in the Po delta area where the proportion of affected males ranges from 2.2 to 2.5% (Colonna-Romano et al. 1985; Rickards et al. 1988; Calabrb et al. in preparation). On the basis of the enzyme level found in red cell extracts it is convenient to separate samples with a severe reduction of the level of enzyme activity (less than 5% of the normal) and samples with a moderate reduction 15-50% of the normal (Rickards et al. 1988). The relative proportion of these two phenotypic classes varies considerably from one population
12
G. VIGLIETTO AND
OTHERS
to another. Severe deficiency accounts for 30% of the deficiency in the population of Naples (Colonna-Romano et al. 1985), for 70% in the population of Matera (Calabrb et al. in preparation ) and for as much as 96% in the Sardinian population of Sassari (Testa et al. 1980). An even greater degree of heterogeneity becomes manifest when biochemical characteristics of partially purified preparations of G6PD are compared. By this means, four common variants were observed to be segregating in the Xardinian population, namely G6PD Mediterranean, Sassari, Cagliari and Seattle-like (Lenzerini et al. 1969; Testa et al. 1980; Fenu et al. 1982), eight were found in a random sample of the population of Naples (Colonna-Romano et al. 1985), and now this paper reports six in the population of the Matera district. Sequencing of genomic DNA fragments and hybridization analysis of in v i t r o amplified genomic DNA with allele specific probes confirm the biochemical findings for the population of Matera in the majority of the cases. G6PD Pisticci and G6PD Tursi are two apparently new variants, associated with a very mild or a moderate degree of deficiency (Table 1). Their molecular lesion has not yet been determined. An additional new variant appears to be G6PD Montalbano. It is a slow mobility variant associated with moderate deficiency and apparently normal kinetic properties (Table 1). It is generated by a G/C+A/T transition in the second position of codon 285 resulting in an Arg-His amino-acid substitution (Table 2). G6PD Metaponto has already been reported in this population (Vulliamy et al. 1988). It is caused by a G/C+A/T transition in exon 4 a t residue position 172 resulting in an Asp-Asn substitution. The properties of G6PD A( - ) Matera are identical to those reported for a variant detected in the same population in a subject with acute haemolytic anaemia (Calabrb et al. 1989) and found by molecular analysis to bear the same mutations of G6PD A( - ) common in populations of African ancestry (Vulliamy et al. 1988). The two subjects included in this group share the same 376 A/T-+G/C substitution, reported also in G6PD A subjects. However, mutation 202 G/C-A/T, specific to the A( - ) variant, is absent in one of the subjects (MT 17), suggesting that two different mutations have occurred on a pre-existing GGPD A gene to give rise to two deficient variants with very similar properties. A similar finding has recently been reported for Afro-Americans (Hirono & Beutler, 1988). It is, therefore, likely that the two A ( - ) variants found in the population of Matera are a remnant of ancient events of genetic admixture with African populations. The variant we have provisionally named G6PD Cagliari-like is characterized by two mutations on the nucleotide sequence. The one occurring in exon 6 is a C/G-T/A transition, determining a phenylalanine for serine amino acid substitution a t residue 188 of the protein Sequence. The substitution occurs at a site only ten amino acids downstream of an evolutionallyconserved sequence (Jeffery et al. 1985; Persico et al. 1986; Fouts et al. 1988; Ho et al. 1988). The sequence contains a lysyl residue considered to be important for the correct positioning of G6P within the G6PD molecule and hence essential for catalytic activity (Camardella et al. 1981, 1988; Jeffery et al. 1985). A substitution in this region is, therefore, expected to affect the catalytic properties. The observed substitution of an acidic residue with a bulky aromatic amino acid must, however, also introduce a considerable distortion in the molecule structure since in vivo and in witro stability of the molecule is greatly impaired. This very same mutation has also been reported in a subject with a typical G6PD Mediterranean variant (Vulliamy et al. 1988) and, more recently, in the biochemically defined G6PD Cagliari and G6PD Sassari from
Common G6PD variants in Southern Italy
13
Table 5 . Nucleotide substitutions of GGPD variants affect restriction mdonuclease recognition sequences (RERS) G6PD variant Metaponto ( A - ) Matera Montalbano
Wild-type sequence GGATC CGTG GAATG CGTG
Mutated sequence GAATG CATV GGATG CATG
RERS GAATG CATC GGATG CATG
Enzyme Fok I N a I11 Fok I Nla I11
the Sardinian population (De Vita et al. 1989). It is not clear as yet by which mechanism G6PD proteins with different biochemical properties (Table 1 ) can be generated from apparently identical genes. I n all the five G6PD Cagliari-like variants we found a second C/G+T/A transition in cxon 11. The mutation occurred in the third position of codon 437 and, therefore, did not result in an amino-acid substitution. This mutation had not been found in the previously reported sequence of a G6PD Mediterranean subject from the neighbouring area of Calabria (Vulliamy et al. 1988) but is reported to occur in all Sardinian subjects described by De Vita et al. (1989). Further typing of G6PD deficient subjects will clarify whether the subject reported by Vulliamy et al. (1988) is unique or whether the codon 437 mutation is itself a polymorphism within the G6PD Mediterranean variants. Of the six single nucleotide substitutions so far reported in Southern Italy G6PD variants, three can, in principle, be recognized by digesting genomic DNA with restriction endonucleases (Table 5). As an alternative t o ASOH, in vitro DNA amplification followed by DNA fragmentation with a specific enzyme may thus be introduced for the recognition of these variants in hemizygous or heterozygous condition. We thank Mrs Maria Terracciano, Mrs Rosaria Terracciano and Mrs Carmela Salzano for their excellent technical assistance. This work was supported by grants from the Italian Ministry of Education and from the National Research Council to G. B., from the Progetto Finalizzato Ingegneria Genetica CNR and from the EEC (No. ST2,T-0172-2-1)to M.G. P.
REFERENCES
ALLISON,A. C . (1960).Glucose-6-phosphate dehydrogenase deficiency in red blood cells of East Africans. Nature 186,531. BATTISTUZZI, C.,D’URSO,&I., TONIOLO,D., PERSICO, G . M. & LUZZATTO, L. (1985). Tissue specific levels of human glucose-6-phosphate dehydrogenase correlate with methylation of specific sites a t the 3’ end of the gene. Proc. Natl. Acad. Sci. USA 82,1465-1469. BATTISTUZZI, G.,ESAN,G . J. F., FASUAN, F. A., MODIANO,G., & LUZZATTO, L. (1977).Comparison of GdA and GdR activities in Nigerians. A study of the variation of glucose-&phosphate dehydrogenase activity. Am. J . H’um. Genet. 29, 3 1-36. BENTON, W. T). & DAVIS, R . W. (1977). Screening Agt recombinant clones by hybridisation to single playues in situ. AEcience 196, 180-182. BETKE,K., BREWER, G. J., KIRKMAN, H. N., LUZZATTO, L., MOTULSKY,A. G . , RAMOT, B. & SINISCALCO, M. (1967).Standardization of procedures for the study of glucose-&phosphate dehydrogenase : report of a WHO scientific group. WHO Tech. Rep. Series 366. BEUTLER, E. (1983).Glucose-6-phosphate dehydrogenase deficiency. In The Metabolic Basis of Inherited Disease, 5th edn (ed. J. B. Stanbury, J. B. Wyngaarden, D. 8. Fredrickson, J . L. Goldstein & M. S. Braun). pp. 1629-1653. New York: McGraw-Hill.
14
G. VIGLIETTO AND
OTHERS
BEUTLER, E., MATHAI,C. K. & SMITH,J. E. (1968). Biochemical variants of glucose-6-phosphate dehydrogenase giving rise to congenital nonspherocytic hemolytic disease. Blood 31, 131-150. C A L A B R ~V., , CASCONE,A., MALASPINA,P. & BATTISTUZZI, G. (1989). Glucose-6-phosphate dehydrogenase deficiency in Southern Italy: a case of G6PD A( - ) associated with favism. Haemutologica 74 (in the press). CAMARDELLA, L., ROMANO, M., DI PRISCO, G., DESCALZI-CANCEDDA, F. (1981). Human erythrocyte glucose-6phosphate dehydrogenase : labelling of a reactive lysyl residue by pyridoxal-5’-phosphate. Biochem. Biophys. Res. Commun. 103, 1384-1389. CAMARDELLA, L., CARUSO,C., RUTIQLIANO, B., ROMANO, M., DI PRISCO, G. & DESCALZI-CANCEDDA, F. (1988). Human erythrocyte glucose-6-phosphate dehydrogenase. Identification of a reactive lysyl residue labelled with pyridoxal-5‘-phosphate. Eur. J . Biochem. 171, 485-489. CHILDS,B., ZINKHAM, W., BROWN,E. A,, KIMBRO,E. L. & TORBET, J . V. (1978). A genetic study of a defect in glutathion met,abolism of the erythrocyte. Bull. Johns Hopkins Hospital 102, 21. COLONNA-ROMANO, S., IOLASCON, A., LIPPO, S., PINTYO, L., CUTILLO,S. & BATTISTUZZI, G. (1985). Genetic heterogeneity at the glucose-6-phosphate dehydrogenase locus in Southern Italy : a study on the population of Naples. Hum. Genet. 69, 228-232. M., SANPIETRO, M., CAPPELLINI,M. D., FIORELLI, G. & TONIOLO, D. (1989). Two point DE VITA,G., ALCALAY, mutations are responsible for G6PD polymorphism in Sardinia. Am. J. Hum. Genet. 44,233-240. FENU,M. P., FINAZZI, G., MANOUSSAKIS,C., PALOMBA, V. & FIORELLI, G. (1982). Glucose-6-phosphate dehydrogenase deficiency : genetic heterogeneity in Sardinia. Ann. Hum. Genet. 46, 105-1 14. POUTS, D., GANQULY, R., GUTIERREZ, G. A., LUCCHESI, J. C. & MANNINQ,J. E. (1988). Nucleotide sequence of the Drosophila glucose-6-phosphate gene and comparison with the homologous human gene. Gene 63, 261-275. HIRONO,A. & BEUTLER, E. (1988). Molecular cloning and nucleotide sequence of cDNA for human glucose-6phosphate dehydrogenase variant A( -). Proc. Natl. Acad. Sci. USA 85, 3951-3954. Ho, Y. S., HOWARD, A. J. & CRAPO,K. D. (1988). Cloning and sequence of a cDNA encoding rat glucose-6phosphate dehydrogenase. Nucl. Acids Res. 15, 7746. JEFFERY, J., HOBBS,L. & JORNVALL, H. (1985). Glucose-6-phosphate dehydrogenase from Saccharomyces cerevisiae : characterization of a reactive lysine residue labelled with acetylsalicylic acid. Biochemistry 24, 66-7 1. KIRKMAN,H. N., DOXIADIS,S. A., VALEAS,T., TASSOPOULOS, N. & BRINSON,A. G. (1965). Diverse characteristics of glucose-6-phosphate dehydrogenase from Greek children. J . Lab. Clin. Med. 65, 212-221. LENZERINI,L., MEERA KHAN,P., FILIPPI,G., RATTAZZI,M. C., KAY, A. K. & SINISCALCO, M. (1969). Characterization of glucose-6-phosphate dehydrogenase variants. I. Occurrence of G6PD Seattle-like variant in Sardinia and its interaction with the G6PD Mediterranean variant. Am. J. Hum. Genet. 21, 142-153. LUZZATTO, L. (1979). Genetics of red cell and susceptibilitv to malaria. Blood 54, 961-976. LUZZATTO, L. & BATTISTUZZI, G. (1985). Glucose-6-phosphate dehydrogenase. Advances in Human Genetics 14, 217-329. MANIATIS,T., FRITSCH, E. F. & SAMBROOK, J. (1982). Molecular Cloning. Cold Spring Harbor Laboratory. MARTINI, G., TONIOLO, D., VULLIAMY, T., LUZZATTO, L., DONO,R., VIQLIETTO, G., PAONESSA, G., D’URSO,M. & PERSICO, M. G. (1986). Structural analysis of the X-linked gene encoding human glucose-6-phosphate dehydrogenase. EMBO J. 5, 1849-1855. MODIANO,G., BATTISTUZZI, G., ESAN,G. J. F., TESTA,U. & LUZZATTO, L. (1979). Genetic heterogeneity of ‘normal ’ erythrocyte glucose-6-phosphate dehydrogenase : an isoelectrophoretic polymorphism. Proc. Natl. Acad. Sci. USA 76, 852-856. MOTULSKY,A. G. (1960). Metabolic polymorphism and the role of infectious diseases in human populations. Hum. Biol. 32, 28-62. PAI,G. S., SPRENKLE, A, , DO, T. T., MARENI, C. E. & MIQEON,B. R. (1980).Localisation of loci for HPRT and G6PD and biochemical evidence of non-random X-chromosome expression from studies of a human Xautosome translocation. Proc. Natl. Acad. Sci. USA 77, 2810-2813. PERONA, G., GUIDI, G. C., TUMMARELLO, D., MARENI, C., BATTISTUZZI, G. & LUZZATTO, L. (1983). A new glucose-6-phosphate dehydrogenase variant (G6PD Verona) in a patient with myelodysplastic syndrome. S c a d . J. Haematol. 30, 407414. PERSICO, M. G., VIQLIETTO,G., MARTINI, G., TONIOLO,D., PAONESSA, G., MOSCATELLI, C., DONO,R., VULLIAMY, T., LUZZATTO, L. & D’URSO,M. (1986). Isolation of human glucose-6-phosphate dehydrogenase cDNA clones: primary structure of the protein and unusual 5’ coding region. Nucl. Acids Res. 14,2511-2522, 7822. RICKARDS, O., BIONDI,G., DE STEFANO, G. F. & BATTISTUZZI, G. (1988). Distribution of genetically determined deficient variants of glucose-6-phosphate dehydrogenase (G6PD) in Southern Italy. In Advances in Haenwgenetics, Vol. 2 . (ed. W. R. Mayr). Berlin, Heidelberg: Springer Verlag. SANQER, F., NICKLEN,S. L COULSON, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.
Common
GGPD variants in Southern Italy
15
SINISCALCO, M., BERNINI, L., FILIPPI, G., LATTE,B., MEERA KHAN,P., PIOMELLI, S. & RATTAZZI, M. (1966). Population genetics of haemoglobin variants, thalassaemia and glucose-6-phosphate dehydrogenase deficiency with particular reference to the malaria hypothesis. Bull. WHO 34, 379-385. P., PURRELLO, M., ROCCHI, M., ARCHIDIACONO, N., ALHADEFF, B., FILIPPI, G., TONIOLO, D., MARTINI, SZABO, G., LUZZATTO, L. & SINISCALCO, M. (1984). Cytological mapping of the human glucose-6-phosphate dehydrogenase gene distal to the fragile-X site suggests a high rate of meiotic recombination across this site. Proc. Natl. Acad. Sei. USA 81, 7855-7859. TAKIZAWA, T., HUANG, I. Y., IKUTA, T. & YOSHIDA, A. (1986). Human glucose-6-phosphate dehydrogenase: primary structure and cDNA cloning. Proc. Natl. Acad. Sci. USA 83, 41574161. A,, BATTISTUZZI, G., CUTILLO,S. & LUZZATTO, L. (1980).Genetic heterogeneity TESTA, U., MELONI, T., LANIA, of G6PD deficiency in Sardinia. Hum. Genet. 56, 99-105. R. B. (1986). The use of synthetic oligonucleotides as specific hybridisation probes THEIN,S. L. & WALLACE, in the diagnosis of genetic disorders. In Human Genetic Diseases: A Practical Approach (ed. K. E. Davies). Oxford, Washington : IRL Press. VULLIAMY,T. J., D’URSO,M., BATTISTUZZI, G., ESTRADA, M., FOULKES, N. S., MARTINI, G., C A L A B R ~V., , POGGI, V., GIORDANO, R., TOWN, M., LUZZATTO, L. & PERSICO, M. G. (1988).Diverse point mutations in the human glucose-6-phosphate dehydrogenase gene cause enzyme deficiency and mild or severe hemolytic anemia. Proc. Natl. Acad. Sci. USA 85, 5171-5175.
