Mutation Research, 234 (1990) 61-70
61
Elsevier MUTENV 08749
An improved method for the detection of genetic variations in DNA with denaturing gradient gel electrophoresis Norio Takahashi, Keiko Hiyama, Mieko Kodaira and Chiyoko Satoh Department of Genetics, Radiation Effects Research Foundation, 5-2 Hijiyama Park, Minami-ku, Hiroshima 732 (Japan)
(Received 25 July 1989) (Revision received 5 October 1989) (Accepted 9 October 1989)
Keywords: Denaturing gradient gel electrophoresis; fl-Globin; DNA polymorphism; RNA : DNA duplexes
Summary We have examined the feasibility of denaturing gradient gel electrophoresis ( D G G E ) of R N A : D N A duplexes to detect variations in genomic and cloned DNAs. The result has demonstrated that employment of R N A : D N A duplexes makes D G G E much more practical for screening a large number of samples than that of D N A : D N A heteroduplexes originally developed by Lerman et al. (1986), because preparation of R N A probes is easier than that of D N A probes. Three different 32p-labeled R N A probes were produced. Genomic or cloned D N A s were digested with restriction enzymes and hybridized to labeled R N A probes, and resulting R N A : D N A duplexes were examined by D G G E . The presence of mismatch(es) was detected as a difference in mobility of bands on the gel. The experimental conditions were determined using D N A segments from cloned normal and 3 thalassemic human fl-globin genes. The results of the experiments on the cloned DNAs suggest that D G G E of R N A : D N A duplexes will detect nucleotide substitutions and deletions in DNA. In the course of these studies, a polymorphism due to a single-base substitution at position 666 of IVS2 (IVS2-666) of the human fl-globin gene was directly identified using genomic D N A samples. A study of 59 unrelated Japanese from Hiroshima was made in which the frequency of the allele with C at IVS2-666 was 0.48 and that of the allele with T was 0.52. This approach was found to be very effective for the detection of heritable variation and should be a powerful tool for the detection of fresh mutations in DNA, which occur outside the known restriction sites.
A study to detect mutations at the protein level to elucidate the genetic effects of atomic bomb radiation was started at the Radiation Effects Research Foundation (RERF) in 1976. The final report (Neel et al., 1988; Satoh and Neel, 1988) of
Correspondence: Dr. N. Takahashi, Department of Genetics, Radiation Effects Research Foundation, 5-2 Hijiyama Park, Minami-ku, Hiroshima 732 (Japan).
the study showed no demonstrable genetic effects of parental exposure. Four fresh mutations were detected in the exposed group and 3 in the control group. The mutation rates per locus per generation of 0 . 6 0 x 10 -s and 0 . 6 4 x 10 -s in the 2 groups, respectively, did not differ significantly. It was emphasized at the same time, however, that a still larger number of gene loci must be examined before this negative conclusion is ascertained with confidence. Recently, this laboratory has under-
0165-1161/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
62 taken pilot studies to detect mutations at the D N A level, in which not only exon regions coding the proteins but also non-coding regions are examined. Many methods have been developed for detecting sequence heterogeneities in D N A molecules. Each has provided important information for gene linkage studies and diagnosis of genetic diseases. Two recently developed methods have significantly increased the number of nucleotides which can be examined at one time for single-base substitutions or more complex sequence differences. One is based on the melting behavior of D N A and uses denaturing gradient gel electrophoresis ( D G G E ) (Lerman et al., 1986; Myers et al., 1985a,b,c, 1987). In this procedure, a singlestranded, radioactive D N A probe was hybridized to cloned or genomic D N A and the resulting D N A - D N A duplex was analyzed using D G G E . D G G E was also employed in order to detect variations in R N A using R N A : R N A duplexes or D N A : R N A duplexes (Smith et al., 1986). The other method, 'ribonuclease (RNase) A cleavage', is based on the susceptibility of single-base mismatches in R N A : D N A duplexes to RNase A cleavage (Myers et al., 1985d; Myers and Maniatis, 1986). Myers et al. (1985d,e) reported that these 2 methods could independently detect known single-base mutations in the /~-globin genes. We have examined whether these methods are suitable for the large-scale screening of D N A samples necessary to the projected new effort to study the genetic effects of the atomic bombs. Preliminary reports have been presented (Hiyama et al., 1988a,b; Takahashi et al., 1988a,b; Satoh et al., 1988). The results obtained from the ' R N a s e cleavage' method will be described elsewhere (Hiyama et al., 1989). The disadvantage of D G G E of D N A : D N A duplexes is that synthesis of radiolabeled singlestranded D N A probes requires many steps. Since synthesis of R N A probes, on the other hand, is much simpler, we developed D G G E of R N A : D N A duplexes. In this report we demonstrate the feasibility of D G G E with R N A probes and the ability of this method to detect a polymorphic substitution at IVS2-666 of the ~-globin gene in both cloned and genomic D N A samples which cannot be directly observed by restriction frag-
ment length polymorphism (RFLP) (review in Orkin, 1984). This method can also be applied for the detection of nucleotide variations such as insertions and deletions.
Materials and methods
High-molecular-weight D N A samples were extracted from cell lines established from EpsteinBarr virus-transformed B lymphocytes obtained from 59 unrelated Japanese from Hiroshima using a modification of the techniques described by Maniatis et al. (1982). The 59 individuals were 35 healthy volunteers, and 11 husband-and-wife pairs and 2 husbands from 13 families belonging to a group of approximately 1000 families of atomic b o m b (A-bomb) survivors, their spouses, children and their controls. Permanent lymphoblastoid cell lines from these families are being established in our laboratory for a future study to examine the potential genetic effects of the A-bombs ( H a m ilton et al., 1985). Genomic D N A s of 20 children of these 11 couples were also extracted from cell lines and examined in the family study. Although some of these husbands and wives had sustained significant radiation exposure at the time of the b o m b detonation, the variation encountered proved to be inherited, so that the question of a radiation effect does not enter into the results. A plasmid containing the normal fl-globin gene (YF204) and plasmids (YF226, 15H-3 and YF227) containing 3 thalassemia genes (Takihara et al., 1984; Fukumaki, personal communication) were obtained from Dr. Y. Fukumaki of Kyushu University. Cloned D N A fragments were obtained by digestion of YF204, YF226, 15H-3 and YF227 with E c o R I and RsaI. The known variations in these fragments are as follows (Fig. 1): thymine (T) at position 1161 relative to the m R N A cap site, which corresponds to IVS2-666 in YF204, is replaced by cytosine (C) in YF226. There is a C to T substitution in 15H-3 at position 1149 (IVS2654). YF227 has deletion of 1 of the 3 Gs between positions 1074 and 1976 (IVS2-579-IVS2-581) (Hiyama et al., 1989) in addition to the substitution noted in 15H-3. Therefore, 3 sequence differences are involved between YF226 and YF227, i.e., 2 base substitutions and 1 base deletion.
63
~MIIIIIH#478 496
1346 724
(229)
956 (461)
~((
1149 16041 1161
~
1394
666)
Exon2~lVS
2"
Number from
cap site (
):Number of lVS 2
Exon 3
B
I
V
i
h(
c
T
V
c
v
E;EcoRI.
YF204-111-917 Y F204-111-439
c
V v i\# B; Barn HI.
YF204
Y F226 K16 15H-3 YF227
1075 (580) a G delete R; Rsa I.
Fig. 1. Schematic diagram showing the location of sequence differences in DNA fragments cloned from the human fl-globin gene used in this study. Thick fines show the fragments which were figated into a transcription vector and the resulting plasmids were used for preparing the RNA probes.
Restriction enzymes were obtained from Takara Shuzo (Kyoto, Japan) or Toyobo (Osaka, Japan); SP6-polymerase, RNasin, p G E M - 3 and p G E M - 3 blue plasmids from Promega (Madison, WI, U.S.A.); [ct-32p]CTP (800 C i / m m o l e , 20 m C i / m l ; 1 Ci = 37 GBq) from Amersham International (Amersham, U.K.); RNase-free DNase I from Worthington (U.S.A.); proteinase K from Boehringer Mannheim (F.R.G.); RNase A from Sigma (St. Louis, MO, U.S.A.), and RNase T1 from Sankyo (Tokyo, Japan). Water used in this study was treated with diethylpyrocarbonate and autoclaved.
Preparation of RNA probes and RNA : DNA duplexes Preparation of uniformly labeled singlestranded R N A probes and R N A : D N A duplexes was carried out as described by Hiyama et al. (1989). Briefly, in order to obtain D N A templates for synthesizing R N A probes, 3 types of plasmid were prepared by recloning each of 3 types of fragments (Fig. 1) into p G E M - 3 or pGEM-3 blue. These plasmids were linearized by EcoRI and used as templates for the preparation of probes using SP6-polymerase. The resulting probes were designated sense YF204-III-917, sense YF204-III-
439 and sense K16, respectively, since all probes used in the present study are designated 'sense probes' which have the same nucleotide sequence (except uracil (U) to T) of the sense strands of the D N A fragments. Both sense YF204-III-917 and sense YF204-III-439 have U at position 1161, while sense K16 has C at the same position. To synthesize the probe for examination of cloned D N A samples, 1 ~g template was treated in the transcription mixture containing 500 /~M each of ATP, G T P and UTP, 20 ~Ci [a-32p]CTP (40 C i / m m o l e , 25/~M). For the R N A probe with high specific activity to examine genomic D N A samples, 100 ~tCi [et-32p]CTP (400 C i / m m o l e , 1 2 / I M ) was used. After transcription, the template D N A was digested with D N a s e I, and the single-stranded R N A probe was resuspended in an appropriate amount of buffer containing 10 m M tris[hydroxymethyl]aminomethane (Tris)-HC1 (pH 7.5), 1 m M ethylenediaminetetraacetic acid (EDTA) and 0.1% sodium dodecyl sulfate. Restriction enzyme-treated cloned D N A s (200 ng) or genomic D N A s (4.5 ~g) were hybridized with the probe (1-5 x 105 cpm of the probe made for cloned D N A samples, 1-2.5 x 10 6 cpm for genomic D N A samples) according to the method of Myers et al. (1985d). The length of sense YF204-III-917, sense YF204-III-439 and sense K16 were 960 nucleotides (nt), 489 nt and 694 nt, respectively, since each probe included the transcript of a polylinker part. All of these probes are longer than the RsaI-EcoRI fragments (439 base pairs (bp)) of examined D N A (Fig. 1). After the hybridization reaction, the single-stranded 'overhang' part of the probe and the unhybridized R N A probe were digested with RNase A and RNase T1. The quantities of R N a s e A and T1 are critical. The quantities must be minimized to the level which is enough for the completion of the above-mentioned reaction because too large an amount of RNases causes cleavage of the probe at mismatches and the resulting products may disturb the pattern of D G G E . After preliminary experiments, we determined the conditions as follows: RNase solution of 300/~1 should contain 0.2 /~g/ml RNase A and 5 n g / m l RNase T1 for examination of cloned D N A s and 5 / ~ g / m l A and 250 n g / m l T1 for duplexes derived from genomic DNAs.
64
Electrophoresis In this study, we used 2 types of gel. One is termed 'parallel gradient gel', since the concentration gradient of the denaturant in a polyacrylamide gel is parallel to the direction of electrophoresis. The other is the 'perpendicular gradient gel' in which the gradient is perpendicular to the direction of electrophoresis (Lerman et al., 1984). 'Parallel gradient gel' electrophoresis was performed using the apparatus (Model SE-520, Vertical Flat Gel) purchased from Hoefer Scientific Instruments (San Francisco, CA, U.S.A.). The gel, measuring 14 cm wide x 31 cm high x 0.75 m m thick, was prepared by the method of Fischer and Lerman (1979) with the following modifications: a 6.5% polyacrylamide gel in TAE buffer (40 mM Tris, 20 m M sodium acetate, 1 mM EDTA, and adjusted with acetic acid to p H 7.4) was prepared 2 cm up from the bottom of the gel to prevent leakage of the denaturant from the gel into the electrode solution during electrophoresis because our equipment had no bottom trough. The gels were composed of a fixed acrylamide concentration (acrylamide in ge1=6.5% (wt/vol); bisacrylamide in total acrylamide = 2.6% (wt/wt)) in T A E buffer with a linearly increasing concentration gradient of the denaturant. In most of the experiments, gels with a denaturant gradient of 0-100% were used (100% denaturant = 7 M urea and 40% formamide). Depending on the situation, gels with gradients of 0-50% or 0-25% were also used. Gradient gels were formed by mixing 2 solutions containing no denaturant and denaturant of the appropriate concentration in a 2-chamber gradient maker (LKB, 92-90-1886, Bromma, Sweden) and poured into the gel cassette from the top. These gels had a short stacking gel containing no denaturant. 'Perpendicular gradient gel electrophoresis' was performed with equipment made according to Fischer and Lerman (1979). Gel solution with 0-25% gradient was poured into the gel cassette and rotated 90 °. The sample was applied as a band across the top of the gel. In both types of electrophoresis, the gel plates were fixed onto an apparatus which had a cathodal chamber at the top, and the whole assembly was completely submerged in a buffer-filled tank that functioned as the anode. A peristaltic p u m p circulated buffer between the 2 chambers. The elec-
trophoresis chambers were maintained at 6 0 ° C with a circulating heater. The 'parallel gel' was run at 200 V for 17 h and the 'perpendicular gel' at 150 V for 10 h. After electrophoresis, the gels were dried on a Gel Drying Processor AE-370 (Atto, Tokyo, Japan). Autoradiography was performed at - 8 0 ° C with Fuji RX X-ray film (Kanagawa, Japan) and a Du Pont Gronex HI-Plus intensifying screen (Wilmington, DE, U.S.A.). Results
Detection of sequence differences in cloned DNA fragments In order to establish the optimum conditions for the detection of base mismatches, we used 4 types of duplexes, i.e., sense K 1 6 / Y F 2 0 4 , sense K 1 6 / Y F 2 2 6 , sense K 1 6 / 1 5 H - 3 and sense K I 6 / Y F 2 2 7 . Sense K 1 6 / Y F 2 0 4 denotes a duplex produced from the 32p-labeled sense K16 probe and a complementary strand of D N A fragment between EcoRI and Rsal of YF204. The general melting behavior of these duplexes was examined in the 'perpendicular gradient gel'. The pattern on the gel containing the gradient of 0-25% denaturant is shown in Fig. 2. Sense K I 6 / Y F 2 2 7 separated from 3 other duplexes in
2
e~
o IM
P
Increasing Denaturant Concentration Fig. 2. Melting profiles of 4 types of R N A : D N A duplexes. D N A s used for making duplexes with labeled sense K16 probe are (A) YF226, (B) YF204, (C) 15H-3, (D) YF227. Electrophoresis was carried out on acrylamide gel containing a denaturant (0-25%) gradient perpendicular to the electric field.
65 the gel where the denaturant concentration was 0-16%. The remaining 3 duplexes separated from each other between denaturant concentrations 9% and 16%. We therefore examined their separation by electrophoresis on 3 types of 'parallel gels' containing 0-25%, 0-50%, and 0-100% denaturant, respectively, for 17 h at 200 V. Under our conditions, the gel containing the gradient 0-100% denaturant provided the best separation. The duplexes were retarded between the denaturant concentrations 10% and 20%. We conducted an experiment to establish the optimum electrophoresis time to separate 3 types of duplexes (sense K16/YF204, sense K 1 6 / Y F 226, sense K 1 6 / Y F 2 2 7 ) on a parallel gel. At 11 h, the band of sense K 1 6 / Y F 2 2 7 was distinguishable from those of sense K 1 6 / Y F 2 0 4 and sense K 1 6 / YF226, but the latter 2 were still indistinguishable. After 15-19 h, these latter 2 bands were clearly separated. When electrophoresis was continued for 27 h, the 3 bands migrated to higher denaturant concentrations, but no marked change was observed in the differences between them. The mobility of the bands decreased rapidly as they migrated to higher concentrations of denaturant. Based on these results, we employed the following electrophoretic conditions: acrylamide concentration 6.5%, denaturant gradient 0-100%, electrophoresis time 17 h, temperature 6 0 ° C . Typical patterns obtained under these conditions are shown in Fig. 3. Sense K16/YF226, which contained no mismatch, migrated farthest. Sense K16/YF227, with 3 types of mismatch, showed the slowest mobility. Sense K16/YF204, with 1 mismatch, and sense K16/15H-3, with 2 mismatches, were intermediate between them. These results are consistent with the theory that duplexes with more mismatches are less stable and melt at lower denaturant concentrations. Sense YF204-III-439 and sense YF204-III-917 were hybridized with the RsaI-EcoRI fragments of YF204, YF226, 15H-3, and YF227, respectively, and electrophoresed on the denaturing gradient gels. The patterns are shown in Fig. 4. In these experiments, the number of mismatches in the duplexes made from fragments of YF204, YF226, 15H-3 and YF227 are 0, 1, 1, 2, respectively, since both probes have U at IVS2-666. In YF226, a T---, C substitution occurs at IVS2-666,
e, om ~a eO e-
eO ¢.9
2 u U.I
--I e-
¢-.
Q
O
e.m
t5 U e-
1 2 3 4 Fig. 3. The typical pattern of DGGE of duplexes between
sense K16 and cloned DNAs. DNAs used for making duplexes are (lane 1) YF226, (lane 2) YF204, (lane 3) 15H-3, (lane 4) YF227. Electrophoresis was carried out in a denaturant (0-100%) gradient gel.
m
1 X
2
3
m
4
5
/
~
Sense YF204-111-439 Sense
6
7
8 /
YF204-HI-917
Fig. 4 . T h e t y p i c a l p a t t e r n o f D G G E of duplexes between either sense YF204-III-439(lanes 1-4) or sense YF204-III-917 (lanes 5-8) and cloned DNAs. DNAs used for making duplexes are shown above the photograph. Electrophoresis was carried out in a denaturant (0-100%) gradient gel.
66
while in 15H-3 a C ~ T substitution occurs at IVS2-654. The different types of substitution at different positions reflected the mobilities of these duplexes in Fig. 4. Theoretically, the duplexes made from a D N A fragment with sense YF204III-439 and sense YF204-III-917, respectively, were the same when they were applied to a gradient gel since the 'overhang' part of the probes had been cleaved. However, in Fig. 4, bands obtained from the shorter probe (sense YF204-II-439) were broader than those from the longer probe (sense YF204-III-917). Since the band of sense YF204III-439/YF204 (lane 1 of Fig. 4) was not clearly separated from that of sense YF204-III-439/ YF226 (lane 2 of Fig. 4), the duplexes made of genomic D N A from an individual exhibiting 2 bands will be likely to be observed as a single broad band. For this reason, sense YF204-III-917 was used in addition to sense K16 in the screening of total genomic D N A s to be described below.
Screening of genomic DNA Total genomic D N A s obtained from 79 individuals (including 20 children) were examined by D G G E . D G G E patterns from 1 family using 2 probes are shown in Fig. 5. When the sense K16 probe was used, 2 clearly identified bands were observed (Fig. 5B). Mobilities of the faster band and the slower band were identical with those of sense K 1 6 / Y F 2 2 6 and sense K 1 6 / Y F 2 0 4 (Fig. 5A), respectively. Since the sense K16 probe, YF226 and YF204 have C, C and T, respectively, at IVS2-666 of the /3-globin gene, the former duplex has no haismatch but the latter has a mismatch that makes its migration slower. The genotype of an individual showing a single faster band was termed C / C while that of an individual showing a single slower band was termed T/T. Accordingly, the genotype of an individual exhibiting 2 bands was C / T . Screening of the same population was also conducted using sense YF204-III-917 which has U at IVS2-666. Two clearly identified bands were observed (Fig. 5D). In this case, duplexes made of genomic DNAs from individuals with genotype C / C migrated slower than those made of genomic D N A s obtained from individuals with genotype T / T . Thus, the genotype of each individual was independently determined using these 2 probes
TIT C/C
TIT
F10
[] ©
'--I--'
i
A
1
,
i
[]
©
C/T
C/T
C/C
~
[]
C/T
B
d
C/T
C
2
1
2
Fig. 5. D G G E patterns of duplexes from human genomic DNAs, together with those from cloned DNAs as comparison. (A) The duplexes with sense K16 and the cloned DNAs. Lane 1: sense K16/YF226 (YF226 has C at IVS2-666). Lane 2: sense K16/YF204 (YF204 has T at IVS2-666). (B) The duplexes with sense K16 and genomic DNAs from 1 nuclear family. The genotypes are shown with the pedigree. The mobilities of faster bands and slower bands are identical with those of sense K16/YF226 (lane 1 of (A)) and sense K16/YF204 (lane 2 of (A)), respectively. (C) The duplexes with sense YF204-III-917 and the cloned DNAs. Lane 1: sense YF204-III-917/YF204. Lane 2: sense YF204-III-917/YF226. (D) The duplexes with sense YF204-II1-917 and the genomic DNAs from the same family shown in (B). The mobilities of faster bands and slower bands are identical with those of sense YF204-III-917/YF204 (lane 1 of (C)) and sense YF204-III-917/YF226 (lane 2 of (C)), respectively.
and the results were in complete agreement. Table 1 presents the distribution of the genotypes and allele frequencies obtained from these experiments TABLE 1 DISTRIBUTION OF GENOTYPES A N D THEIR QUENCIES 1N 59 UNRELATED JAPANESE
FRE-
Genotype
Number observed
Number expected
Frequency
C/C C/T T/T
14 29 16
13.77 29.47 15.76
C = 0.4831 T = 0.5169 X2 = 0.0150, 0.950 > P > 0.900 (dr = 1)
Total
59
59.00
67 TABLE 2 DISTRIBUTION OF GENOTYPES AT IVS2-666 OF THE fl-GLOBIN GENE IN 11 FAMILIES Genotype of parents
Number Number Expected Observed of of genotype genotype couples children of of children children
TT x C/T
1
2
C/T x C/T
4
8
C/C x T/T C/C x C/T
1 4
2 6
C/C × C/C
1
2
C/T T/T T/T C/T C/C C/T C/T c/c C/C
2 x C/T 1 x T/T 4 x C/T 3 x C/C 2 x C/T 5 × C/T 1× c / c 2 x C/C
among 59 unrelated individuals. The frequencies of the alleles C and T were 0.48 and 0.52, respectively. The distribution of the observed genotypes is in good agreement with the expectation predicted from the allele frequencies on the assumption of the Hardy-Weinberg equilibrium. The distribution of genotypes in the children of 11 couples is shown in Table 2. There was no exception to the genotypes among 20 children of the 11 couples expected on the basis of a 2-allele system. Discussion
The results obtained with D G G E show that we detected the polymorphism based on a nucleotide substitution at IVS2-666 (see Results). The samples of the same population were also screened in our laboratory by the RNase cleavage method (Hiyama et al., 1989). The results obtained with this method, in which the location of the mismatch can be estimated from the size of the newly produced fragments from RNase A cleavage, also demonstrated that the polymorphic substitution occurred at IVS2-666. In addition, results of family studies employing these 2 methods demonstrated that the sequence difference that produced the polymorphism is inherited in a Mendelian fashion (Table 2) (Hiyama et al., 1989). This indicates that the alteration of melting behavior observed in this study was not caused by artifacts due to base methylation or partial digestion of examined D N A s with restriction enzymes. To confirm the
results obtained with D G G E and the RNase cleavage method, sequence analysis of this region was carried out (Hiyama et al., 1989). Direct sequencing of the products of polymerase chain reaction (PCR) (Saiki et al., 1985) from the family members shown in Fig. 5, except the daughter, revealed that the nucleotides at IVS2-666 in husband, wife and their son were T, C, and both C and T, respectively. Studies on the fl-globin genes clearly show that the allele in which T is replaced by C at IVS2-666 is not necessarily associated with the fl-thalassemia allele in which 4 bases within codons 41 and 42 of the fl-globin gene are deleted (Kimura et al., 1983; Takihara et al., 1984; Kazazian et al., 1984; Fukumaki et al., 1987). Thus, an individual having an allele with a T to C substitution at IVS2-666 is not always a patient with thalassemia (Orkin et al., 1982). Several methods have been developed for the detection of mutations in genomic DNA. R F L P is the method that was first introduced and has been used most widely (Flavell et al., 1978; Geever et al., 1981). Unfortunately, this approach has one significant limitation in that variations occurring outside the restriction enzyme recognition sites cannot be detected, and even large panels of restriction enzymes recognize only a fraction of the existing sequence variations. In this method, per diploid genome, each combination of a unique probe and a 6-cutter restriction enzyme examines 36 bp consisting of 24 bp for loss of a cutting site and on average 12 bp for gain of a cutting site (Delehanty et al., 1986). Base substitutions which do not alter the restriction enzyme cleavage sites can be detected by the differential hybridization of synthetic oligonucleotide probes, if the location and nature of the mutation and the surrounding nucleotides are known (Kidd et al., 1983; Pirastu et al., 1983). This method, therefore, is applicable only to genes that are well characterized prior to examination. Only 40 bp per diploid genome can be examined at one examination, since, for practical purpose, the maximum length of each probe should be less than 20 nucleotides. For these reasons, the abovementioned 2 methods are not efficient in screening for new mutations. The RNase A cleavage method has been developed to detect single-base substitutions in
68 cloned and genomic D N A (Myers et al., 1985d). Since several hundreds to 1000 bp can be examined at a time with 1 probe, the efficiency of this method is higher than that of the 2 methods described above. Moreover, the advantage of this method is that it not only detects mutations but also localizes them. The limitation of this method, however, is that all types of mismatch are not always cleaved by RNase A. Myers et al. (1985d; Myers and Maniatis, 1986) reported that approximately one-third of all possible single-base substitutions were not detectable, even if 2 types of probe were used, one of which was homologous to one strand of the test D N A and the other homologous to the opposite strand of the test DNA. Based on the results of our study (Hiyama et al., 1989) employing the RNase cleavage method, we are confident that this is a very effective method to detect variations in cloned DNAs or DNAs amplified by PCR (Saiki et al., 1985) although it has this limitation. D G G E of DNA : D N A heteroduplexes has been reported to be a very useful tool for identifying base substitutions in human genomic D N A (Myers et al., 1985e; Myers and Maniatis, 1986). By this method, approximately 600 bp can be analyzed with 1 probe (Noll and Collins, 1987), the limitations of this method, however, were described earlier and until they are resolved, our calculation based on our preliminary experiments shows that the efficiency of this method will be similar to that of the RNase cleavage method. In contrast to this, pure single-stranded RNA probes may be consistently obtained very easily employing the in vitro transcription system in which D N A fragments inserted into transcription vectors are used as templates (Melton et al., 1984; Myers et al., 1985d). Smith et al. (1986) demonstrated that single-base mutations in the N S gene of the influenza virus were detectable by D G G E of R N A : R N A duplexes or D N A : RNA duplexes which had been made by hybridization between virion RNA and RNA probes prepared by this method or by D N A probes. In the present study, we demonstrated that sequence differences present not only in cloned D N A but also in genomic D N A are detectable by D G G E of duplexes made by hybridization of RNA probes with DNAs. Because of the reduction in
the time needed to prepare R N A probes, the utility of this method for screening is markedly increased. Hybridization can be carried out in a solution containing 2 or more different probes selected so that the bands of resulting different duplexes will be distributed broadly over the gradient at the end of run. For example, let us assume that 500 bp per haploid genome will be able to be effectively examined by 1 probe and 5 probes will be used in 1 examination to produce 5 duplexes which can be separated on 1 lane. Based on this assumption, we estimate that this method is at least about 5-10 times more efficient than the D G G E of D N A : D N A heteroduplexes. The detection of mutations should be made by direct comparison of D N A sequences of appropriate genes isolated from each individual of the study group. However, it is very questionable whether present sequencing techniques have the necessary accuracy for screening for mutations. The error level may be still of the order of 10 -4 , even after improvement in accuracy is attained by using a better optical processor and a powerful computer (Wada, 1987). At present, moreover, the efficiency of D N A sequencing analysis is too low to detect mutations. By contrast, we could demonstrate the usefulness of D G G E of R N A : D N A duplexes in the model experiment in which sequence differences in genes of individuals were efficiently detected. The sensitivity of D G G E of R N A : D N A duplexes could be dramatically increased by amplifying the genomic D N A by PCR (Saiki et al., 1985). From our preliminary experiences (Takahashi et al., in preparation), the intensity of the bands of duplexes made from genomic DNAs amplified by PCR was the same as that of the band of duplexes from cloned DNAs. However, even if this improvement is incorporated, D G G E alone will not allow the detection of all possible single-base mutations, since our preliminary data showed that some single-base mismatches would not lead to a shift in the mobility of the duplex in DGGE. If the RNase cleavage method is employed in combination with D G G E , there is a high probability of detecting mutations which cannot be detected by D G G E alone. Thus, a very large fraction of all possible mutations should be detected by combining the 2 methods, although
69 t h e r e is s o m e o v e r l a p i n t h e m u t a t i o n s t h a t c a n b e detected by the 2 methods. The sequence differences caused by the deletion o f l a r g e f r a g m e n t s will b e d e t e c t a b l e b y D G G E , if t h e f o l l o w i n g t e c h n i q u e is i n c o r p o r a t e d . T h e u s e of either electronic autofluorography (Davidson and Case, 1982) or a 2-dimensional fl-scanner to impart a greater dynamic range to quantification of 32p-labeled bands should be especially effective for detection of the bands containing half of the expected diploid amount of DNA. Introduction of these techniques could be powerful in studies of radiation mutagenesis, where deletion mutations a p p e a r to b e p r e d o m i n a n t . At present, we think that the combination of D G G E a n d t h e R N a s e c l e a v a g e m e t h o d is e f f e c tive a n d p r a c t i c a b l e f o r s c r e e n i n g f o r m u t a t i o n s . U s e o f t h e s e m e t h o d s will a l s o b e c o n d u c t i v e t o the detection of a great number of polymorphisms a n d m a k e it p o s s i b l e t o p r e p a r e l i n k a g e m a p s i n great detail.
Acknowledgements W e a r e g r a t e f u l t o N . K o s a k a , H. O m i n e a n d Y. K i m u r a f o r t h e i r e x c e l l e n t t e c h n i q u e i n D G G E . W e a l s o t h a n k M s . J. K a n e k o f o r h e r t e c h n i c a l work throughout this work. We are much indebted t o D r . J.V. N e e l f o r h i s a s s i s t a n c e in p r e p a r i n g t h e manuscript.
References Davidson, J.B., and A.L. Case (1982) Rapid electronic autofluorography of labeled macromolecules on two-dimensional gels, Science, 215, 1398-1400. Delehanty, J., R.L. White and M.L Mendelsohn (1986) Approaches to determining mutation rates in human DNA, Mutation Res., 167, 215-232. Fischer, S.G., and L.S. Lerman (1979) Two-dimensional electrophoresis separation of restriction enzyme fragments of DNA, in: R. Wu (Ed.), Methods in Enzymology, Vol. 68, Academic Press, New York, pp. 183-191. Flavell, R.A., J.M. Kooter, E. De Boer, P.F. Little and R. Williamson (1978) Analysis of the beta-delta-globin gene loci in normal and Hb Lepore DNA: direct determination of gene linkage and intergene distance, Cell, 15, 25-41. Fukumaki, Y., E. Matsunaga, T. Taga, Y. Takihara and T. Nakamura (1987) Multiple origins of the fl-thalassemia gene with a four-nucleotide deletion in its second exon, Jpn. J. Hum. Genet., 32, 155. Geever, R.F., L.B. Wilson, F.S. Nallaseth, P.F. Milner, M.
Bittner and J.T. Wilson (1981) Direct identification of sickle cell anemia by blot hybridization, Proc. Natl. Acad. Sci. (U.S.A.), 78, 5081-5085. Hamilton, H.B. (1985) Genetics and the atomic bombs in Hiroshima and Nagasaki, Am. J. Med. Genet., 20, 541-546. Hiyama, K., M. Kodaira and C. Satoh (1988a) Detection of mismatches in RNA : DNA duplexes, Jpn. J. Hum. Genet., 33, 234. Hiyama, K., N. Takahashi, M. Kodaira and C. Satoh (1988b) Detection of mismatches in R N A : D N A duplexes. Report 2, Jpn. J. Hum. Genet., 34, 76. Hiyama, K., M. Kodaira and C. Satoh (1989) Detection of deletions, insertions and single nucleotide substitutions in cloned /3-globin genes and new polymorphic nucleotide substitutions in /3-globin genes in a Japanese population using ribonuclease cleavage at mismatches in R N A : D N A duplexes, submitted for publication. Kazazian Jr., H.H., S.H. Orkin, S.E. Antonarakis, J.P. Sexton, C.D. Boehm, S.C. Goff and P.G. Waber (1984) Molecular characterization of seven fl-thalassemia mutations in Asian Indians, EMBO J., 3, 593-596. Kidd, V.J., R.B. Wallace, K. ltakura and S.L.C. Woo (1983) ai-Antitrypsin deficiency detection by direct analysis of the mutation in the gene, Nature (London), 304, 230-234. Kimura, A., E. Matsunaga, Y. Takihara, T. Nakamura, Y. Takagi, S.-T. Lin and H.-T. Lee (1983) Structural analysis of a fl-thalassemia gene found in Taiwan, J. Biol. Chem., 258, 2748-2749. Lerman, L.S., S.G. Fischer, I. Hurley, K. Silverstein and N. Lumelsky (1984) Sequence-determined DNA separations, Annu. Rev. Biophys. Bioeng., 13, 399-423. Lerman, L.S., K. Silverstein and E. Grinfeld (1986) Searching for gene defects by denaturing gradient gel electrophoresis, Cold Spring Harbor Symp. Quant. Biol., 51,285-297. Maniatis, T., E.F. Fritsch and J. Sambrook (1982) Isolation of high-molecular-weight, eukaryotic DNA from cells grown in tissue culture, in: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 280-281. Melton, D.A., P.A. Krieg, M.R. Rebagliati, T. Maniatis, K. Zinn and M.R. Green (1984) Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter, Nucleic Acids Res., 12, 7035-7056. Myers, R.M., and T. Maniatis (1986) Recent advances in the development of methods for detecting single-base substitutions associated with human genetic diseases, Cold Spring Harbor Symp. Quant. Biol., 51, 275-284. Myers, R.M., L.S. Lerman and T. Maniatis (1985a) A general method for saturation mutagenesis of cloned DNA fragments, Science, 229, 242-247. Myers, R.M., S.G. Fischer, LS. Lerman and T. Maniatis (1985b) Nearly all single base substitutions in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis, Nucleic Acids Res., 13, 3131-3145. Myers, R.M., S.G. Fischer, T. Maniatis and L.S. Lerman (1985c) Modifications of the melting properties of duplex
70 DNA by attachment of a GC-rich DNA sequence as determined by denaturing gradient gel electrophoresis, Nucleic Acids Res., 13, 3111-3129. Myers, R.M., Z. Larin and T. Maniatis (1985d) Detection of single base substitutions by ribonuclease cleavage at mismatches in R N A : D N A duplexes, Science, 230, 1242-1246. Myers, R.M., N. Lumelsky, L.S. Lerman and T. Maniatis (1985e) Detection of single base substitutions in total genomic DNA, Nature (London), 313, 495-498. Myers, R.M., T. Maniatis and L.S. Lerman (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis, in: R. Wu (Ed.), Methods in Enzymology, Vol. 155, Academic Press, New York, pp. 501-527. Neel, J.V., C. Satoh, K. Goriki, J. Asakawa, M. Fujita, N. Takahashi, T. Kageoka and R. Hazama (1988) Search for mutations altering protein charge and/or function in children of atomic bomb survivors: final report, Am. J. Hum. Genet., 42, 663-676. Noll, W.W., and M. Collins (1987) Detection of human DNA polymorphisms with a simplified denaturing gradient gel electrophoresis technique, Proc. Natl. Acad. Sci. (U.S.A.), 84, 3339-3343. Orkin, S.H. (1984) The mutation and polymorphism of the human fl-globin gene and its surrounding DNA, Annu. Rev. Genet., 18, 131-171. Orkin, S.H., H.H. Kazazian Jr., S.E. Antonarakis, S.C. Goff, C.D. Boehm, J.P. Sexton, P.G. Waber and P.J.V. Giardina (1982) Linkage of fl-thalassemia mutations and fl-globin gene polymorphisms with DNA polymorphisms in human fl-globin gene cluster, Nature (London), 296, 627-631. Pirastu, M., Y.W. Kan, A. Cao, B.J. Conner, R.L. Teplitz and R.B. Wallace (1983) Prenatal diagnosis of fl-thalassemia: detection of a single nucleotide mutation in DNA, New Engl. J. Med., 309, 284-287. Saiki, R.K., S. Scharf, F. Faloona, K.B. Mullis, G.T. Horn,
H.A. Erlich and N. Arnheim (1985) Enzymatic amplification of fl-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia, Science, 230, 1350-1354. Satoh, C., and J.V. Neel (1988) Biochemical mutations in the children of atomic bomb survivors, in: H. Takebe and J. Utsunomiya (Eds.), Genetics of Human Tumors in Japan, Gann Monograph on Cancer Research No. 35, Japan Scientific Societies Press, Tokyo, and Taylor and Francis, London and Philadelphia, pp. 191-208. Satoh, C., K. Hiyama, N. Takahashi and M. Kodaira (1988) Detection of new variations in cloned fl-globin genes and a DNA polymorphism in a Japanese population with ribonuclease cleavage at mismatches in R N A : D N A duplexes, Genome, 30 (Suppl.), 1, 369. Paper presented at the 16th International Congress of Genetics, Toronto, 20-27 August. Smith, F.I., J.D. Parvin and P. Palese (1986) Detection of single base substitutions in influenza virus RNA molecules by denaturing gradient gel electrophoresis of RNA°RNA or D N A - R N A heteroduplexes, Virology, 150, 55-64. Takahashi, N., K. Hiyama, M. Kodaira, H. Omine and C. Satoh (1988a) Detection of mismatches in RNA:DNA duplexes. Report 3, Jpn. J. Hum. Genet., 34, 77. Takahashi, N., K. Hiyama, M. Kodaira and C. Satoh (1988b) Detection of mismatches in RNA : DNA duplexes. Report 4, Paper presented at l l t h Meeting of the Japan Molecular Biology Society, Tokyo, 20-23 December. Takihara, Y., E. Matsunaga, T. Nakamura, S.-T. Lin, H.-T. Lee, Y. Fukumaki and Y. Takagi (1984) One base substitution in IVS-2 caused a fl+-thalassemia phenotype in a Chinese patient, Biochem. Biophys. Res. Commun., 121, 324-330. Wada, A. (1987) Automated high-speed DNA sequencing, Nature (London), 325, 771-772.
61
Elsevier MUTENV 08749
An improved method for the detection of genetic variations in DNA with denaturing gradient gel electrophoresis Norio Takahashi, Keiko Hiyama, Mieko Kodaira and Chiyoko Satoh Department of Genetics, Radiation Effects Research Foundation, 5-2 Hijiyama Park, Minami-ku, Hiroshima 732 (Japan)
(Received 25 July 1989) (Revision received 5 October 1989) (Accepted 9 October 1989)
Keywords: Denaturing gradient gel electrophoresis; fl-Globin; DNA polymorphism; RNA : DNA duplexes
Summary We have examined the feasibility of denaturing gradient gel electrophoresis ( D G G E ) of R N A : D N A duplexes to detect variations in genomic and cloned DNAs. The result has demonstrated that employment of R N A : D N A duplexes makes D G G E much more practical for screening a large number of samples than that of D N A : D N A heteroduplexes originally developed by Lerman et al. (1986), because preparation of R N A probes is easier than that of D N A probes. Three different 32p-labeled R N A probes were produced. Genomic or cloned D N A s were digested with restriction enzymes and hybridized to labeled R N A probes, and resulting R N A : D N A duplexes were examined by D G G E . The presence of mismatch(es) was detected as a difference in mobility of bands on the gel. The experimental conditions were determined using D N A segments from cloned normal and 3 thalassemic human fl-globin genes. The results of the experiments on the cloned DNAs suggest that D G G E of R N A : D N A duplexes will detect nucleotide substitutions and deletions in DNA. In the course of these studies, a polymorphism due to a single-base substitution at position 666 of IVS2 (IVS2-666) of the human fl-globin gene was directly identified using genomic D N A samples. A study of 59 unrelated Japanese from Hiroshima was made in which the frequency of the allele with C at IVS2-666 was 0.48 and that of the allele with T was 0.52. This approach was found to be very effective for the detection of heritable variation and should be a powerful tool for the detection of fresh mutations in DNA, which occur outside the known restriction sites.
A study to detect mutations at the protein level to elucidate the genetic effects of atomic bomb radiation was started at the Radiation Effects Research Foundation (RERF) in 1976. The final report (Neel et al., 1988; Satoh and Neel, 1988) of
Correspondence: Dr. N. Takahashi, Department of Genetics, Radiation Effects Research Foundation, 5-2 Hijiyama Park, Minami-ku, Hiroshima 732 (Japan).
the study showed no demonstrable genetic effects of parental exposure. Four fresh mutations were detected in the exposed group and 3 in the control group. The mutation rates per locus per generation of 0 . 6 0 x 10 -s and 0 . 6 4 x 10 -s in the 2 groups, respectively, did not differ significantly. It was emphasized at the same time, however, that a still larger number of gene loci must be examined before this negative conclusion is ascertained with confidence. Recently, this laboratory has under-
0165-1161/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
62 taken pilot studies to detect mutations at the D N A level, in which not only exon regions coding the proteins but also non-coding regions are examined. Many methods have been developed for detecting sequence heterogeneities in D N A molecules. Each has provided important information for gene linkage studies and diagnosis of genetic diseases. Two recently developed methods have significantly increased the number of nucleotides which can be examined at one time for single-base substitutions or more complex sequence differences. One is based on the melting behavior of D N A and uses denaturing gradient gel electrophoresis ( D G G E ) (Lerman et al., 1986; Myers et al., 1985a,b,c, 1987). In this procedure, a singlestranded, radioactive D N A probe was hybridized to cloned or genomic D N A and the resulting D N A - D N A duplex was analyzed using D G G E . D G G E was also employed in order to detect variations in R N A using R N A : R N A duplexes or D N A : R N A duplexes (Smith et al., 1986). The other method, 'ribonuclease (RNase) A cleavage', is based on the susceptibility of single-base mismatches in R N A : D N A duplexes to RNase A cleavage (Myers et al., 1985d; Myers and Maniatis, 1986). Myers et al. (1985d,e) reported that these 2 methods could independently detect known single-base mutations in the /~-globin genes. We have examined whether these methods are suitable for the large-scale screening of D N A samples necessary to the projected new effort to study the genetic effects of the atomic bombs. Preliminary reports have been presented (Hiyama et al., 1988a,b; Takahashi et al., 1988a,b; Satoh et al., 1988). The results obtained from the ' R N a s e cleavage' method will be described elsewhere (Hiyama et al., 1989). The disadvantage of D G G E of D N A : D N A duplexes is that synthesis of radiolabeled singlestranded D N A probes requires many steps. Since synthesis of R N A probes, on the other hand, is much simpler, we developed D G G E of R N A : D N A duplexes. In this report we demonstrate the feasibility of D G G E with R N A probes and the ability of this method to detect a polymorphic substitution at IVS2-666 of the ~-globin gene in both cloned and genomic D N A samples which cannot be directly observed by restriction frag-
ment length polymorphism (RFLP) (review in Orkin, 1984). This method can also be applied for the detection of nucleotide variations such as insertions and deletions.
Materials and methods
High-molecular-weight D N A samples were extracted from cell lines established from EpsteinBarr virus-transformed B lymphocytes obtained from 59 unrelated Japanese from Hiroshima using a modification of the techniques described by Maniatis et al. (1982). The 59 individuals were 35 healthy volunteers, and 11 husband-and-wife pairs and 2 husbands from 13 families belonging to a group of approximately 1000 families of atomic b o m b (A-bomb) survivors, their spouses, children and their controls. Permanent lymphoblastoid cell lines from these families are being established in our laboratory for a future study to examine the potential genetic effects of the A-bombs ( H a m ilton et al., 1985). Genomic D N A s of 20 children of these 11 couples were also extracted from cell lines and examined in the family study. Although some of these husbands and wives had sustained significant radiation exposure at the time of the b o m b detonation, the variation encountered proved to be inherited, so that the question of a radiation effect does not enter into the results. A plasmid containing the normal fl-globin gene (YF204) and plasmids (YF226, 15H-3 and YF227) containing 3 thalassemia genes (Takihara et al., 1984; Fukumaki, personal communication) were obtained from Dr. Y. Fukumaki of Kyushu University. Cloned D N A fragments were obtained by digestion of YF204, YF226, 15H-3 and YF227 with E c o R I and RsaI. The known variations in these fragments are as follows (Fig. 1): thymine (T) at position 1161 relative to the m R N A cap site, which corresponds to IVS2-666 in YF204, is replaced by cytosine (C) in YF226. There is a C to T substitution in 15H-3 at position 1149 (IVS2654). YF227 has deletion of 1 of the 3 Gs between positions 1074 and 1976 (IVS2-579-IVS2-581) (Hiyama et al., 1989) in addition to the substitution noted in 15H-3. Therefore, 3 sequence differences are involved between YF226 and YF227, i.e., 2 base substitutions and 1 base deletion.
63
~MIIIIIH#478 496
1346 724
(229)
956 (461)
~((
1149 16041 1161
~
1394
666)
Exon2~lVS
2"
Number from
cap site (
):Number of lVS 2
Exon 3
B
I
V
i
h(
c
T
V
c
v
E;EcoRI.
YF204-111-917 Y F204-111-439
c
V v i\# B; Barn HI.
YF204
Y F226 K16 15H-3 YF227
1075 (580) a G delete R; Rsa I.
Fig. 1. Schematic diagram showing the location of sequence differences in DNA fragments cloned from the human fl-globin gene used in this study. Thick fines show the fragments which were figated into a transcription vector and the resulting plasmids were used for preparing the RNA probes.
Restriction enzymes were obtained from Takara Shuzo (Kyoto, Japan) or Toyobo (Osaka, Japan); SP6-polymerase, RNasin, p G E M - 3 and p G E M - 3 blue plasmids from Promega (Madison, WI, U.S.A.); [ct-32p]CTP (800 C i / m m o l e , 20 m C i / m l ; 1 Ci = 37 GBq) from Amersham International (Amersham, U.K.); RNase-free DNase I from Worthington (U.S.A.); proteinase K from Boehringer Mannheim (F.R.G.); RNase A from Sigma (St. Louis, MO, U.S.A.), and RNase T1 from Sankyo (Tokyo, Japan). Water used in this study was treated with diethylpyrocarbonate and autoclaved.
Preparation of RNA probes and RNA : DNA duplexes Preparation of uniformly labeled singlestranded R N A probes and R N A : D N A duplexes was carried out as described by Hiyama et al. (1989). Briefly, in order to obtain D N A templates for synthesizing R N A probes, 3 types of plasmid were prepared by recloning each of 3 types of fragments (Fig. 1) into p G E M - 3 or pGEM-3 blue. These plasmids were linearized by EcoRI and used as templates for the preparation of probes using SP6-polymerase. The resulting probes were designated sense YF204-III-917, sense YF204-III-
439 and sense K16, respectively, since all probes used in the present study are designated 'sense probes' which have the same nucleotide sequence (except uracil (U) to T) of the sense strands of the D N A fragments. Both sense YF204-III-917 and sense YF204-III-439 have U at position 1161, while sense K16 has C at the same position. To synthesize the probe for examination of cloned D N A samples, 1 ~g template was treated in the transcription mixture containing 500 /~M each of ATP, G T P and UTP, 20 ~Ci [a-32p]CTP (40 C i / m m o l e , 25/~M). For the R N A probe with high specific activity to examine genomic D N A samples, 100 ~tCi [et-32p]CTP (400 C i / m m o l e , 1 2 / I M ) was used. After transcription, the template D N A was digested with D N a s e I, and the single-stranded R N A probe was resuspended in an appropriate amount of buffer containing 10 m M tris[hydroxymethyl]aminomethane (Tris)-HC1 (pH 7.5), 1 m M ethylenediaminetetraacetic acid (EDTA) and 0.1% sodium dodecyl sulfate. Restriction enzyme-treated cloned D N A s (200 ng) or genomic D N A s (4.5 ~g) were hybridized with the probe (1-5 x 105 cpm of the probe made for cloned D N A samples, 1-2.5 x 10 6 cpm for genomic D N A samples) according to the method of Myers et al. (1985d). The length of sense YF204-III-917, sense YF204-III-439 and sense K16 were 960 nucleotides (nt), 489 nt and 694 nt, respectively, since each probe included the transcript of a polylinker part. All of these probes are longer than the RsaI-EcoRI fragments (439 base pairs (bp)) of examined D N A (Fig. 1). After the hybridization reaction, the single-stranded 'overhang' part of the probe and the unhybridized R N A probe were digested with RNase A and RNase T1. The quantities of R N a s e A and T1 are critical. The quantities must be minimized to the level which is enough for the completion of the above-mentioned reaction because too large an amount of RNases causes cleavage of the probe at mismatches and the resulting products may disturb the pattern of D G G E . After preliminary experiments, we determined the conditions as follows: RNase solution of 300/~1 should contain 0.2 /~g/ml RNase A and 5 n g / m l RNase T1 for examination of cloned D N A s and 5 / ~ g / m l A and 250 n g / m l T1 for duplexes derived from genomic DNAs.
64
Electrophoresis In this study, we used 2 types of gel. One is termed 'parallel gradient gel', since the concentration gradient of the denaturant in a polyacrylamide gel is parallel to the direction of electrophoresis. The other is the 'perpendicular gradient gel' in which the gradient is perpendicular to the direction of electrophoresis (Lerman et al., 1984). 'Parallel gradient gel' electrophoresis was performed using the apparatus (Model SE-520, Vertical Flat Gel) purchased from Hoefer Scientific Instruments (San Francisco, CA, U.S.A.). The gel, measuring 14 cm wide x 31 cm high x 0.75 m m thick, was prepared by the method of Fischer and Lerman (1979) with the following modifications: a 6.5% polyacrylamide gel in TAE buffer (40 mM Tris, 20 m M sodium acetate, 1 mM EDTA, and adjusted with acetic acid to p H 7.4) was prepared 2 cm up from the bottom of the gel to prevent leakage of the denaturant from the gel into the electrode solution during electrophoresis because our equipment had no bottom trough. The gels were composed of a fixed acrylamide concentration (acrylamide in ge1=6.5% (wt/vol); bisacrylamide in total acrylamide = 2.6% (wt/wt)) in T A E buffer with a linearly increasing concentration gradient of the denaturant. In most of the experiments, gels with a denaturant gradient of 0-100% were used (100% denaturant = 7 M urea and 40% formamide). Depending on the situation, gels with gradients of 0-50% or 0-25% were also used. Gradient gels were formed by mixing 2 solutions containing no denaturant and denaturant of the appropriate concentration in a 2-chamber gradient maker (LKB, 92-90-1886, Bromma, Sweden) and poured into the gel cassette from the top. These gels had a short stacking gel containing no denaturant. 'Perpendicular gradient gel electrophoresis' was performed with equipment made according to Fischer and Lerman (1979). Gel solution with 0-25% gradient was poured into the gel cassette and rotated 90 °. The sample was applied as a band across the top of the gel. In both types of electrophoresis, the gel plates were fixed onto an apparatus which had a cathodal chamber at the top, and the whole assembly was completely submerged in a buffer-filled tank that functioned as the anode. A peristaltic p u m p circulated buffer between the 2 chambers. The elec-
trophoresis chambers were maintained at 6 0 ° C with a circulating heater. The 'parallel gel' was run at 200 V for 17 h and the 'perpendicular gel' at 150 V for 10 h. After electrophoresis, the gels were dried on a Gel Drying Processor AE-370 (Atto, Tokyo, Japan). Autoradiography was performed at - 8 0 ° C with Fuji RX X-ray film (Kanagawa, Japan) and a Du Pont Gronex HI-Plus intensifying screen (Wilmington, DE, U.S.A.). Results
Detection of sequence differences in cloned DNA fragments In order to establish the optimum conditions for the detection of base mismatches, we used 4 types of duplexes, i.e., sense K 1 6 / Y F 2 0 4 , sense K 1 6 / Y F 2 2 6 , sense K 1 6 / 1 5 H - 3 and sense K I 6 / Y F 2 2 7 . Sense K 1 6 / Y F 2 0 4 denotes a duplex produced from the 32p-labeled sense K16 probe and a complementary strand of D N A fragment between EcoRI and Rsal of YF204. The general melting behavior of these duplexes was examined in the 'perpendicular gradient gel'. The pattern on the gel containing the gradient of 0-25% denaturant is shown in Fig. 2. Sense K I 6 / Y F 2 2 7 separated from 3 other duplexes in
2
e~
o IM
P
Increasing Denaturant Concentration Fig. 2. Melting profiles of 4 types of R N A : D N A duplexes. D N A s used for making duplexes with labeled sense K16 probe are (A) YF226, (B) YF204, (C) 15H-3, (D) YF227. Electrophoresis was carried out on acrylamide gel containing a denaturant (0-25%) gradient perpendicular to the electric field.
65 the gel where the denaturant concentration was 0-16%. The remaining 3 duplexes separated from each other between denaturant concentrations 9% and 16%. We therefore examined their separation by electrophoresis on 3 types of 'parallel gels' containing 0-25%, 0-50%, and 0-100% denaturant, respectively, for 17 h at 200 V. Under our conditions, the gel containing the gradient 0-100% denaturant provided the best separation. The duplexes were retarded between the denaturant concentrations 10% and 20%. We conducted an experiment to establish the optimum electrophoresis time to separate 3 types of duplexes (sense K16/YF204, sense K 1 6 / Y F 226, sense K 1 6 / Y F 2 2 7 ) on a parallel gel. At 11 h, the band of sense K 1 6 / Y F 2 2 7 was distinguishable from those of sense K 1 6 / Y F 2 0 4 and sense K 1 6 / YF226, but the latter 2 were still indistinguishable. After 15-19 h, these latter 2 bands were clearly separated. When electrophoresis was continued for 27 h, the 3 bands migrated to higher denaturant concentrations, but no marked change was observed in the differences between them. The mobility of the bands decreased rapidly as they migrated to higher concentrations of denaturant. Based on these results, we employed the following electrophoretic conditions: acrylamide concentration 6.5%, denaturant gradient 0-100%, electrophoresis time 17 h, temperature 6 0 ° C . Typical patterns obtained under these conditions are shown in Fig. 3. Sense K16/YF226, which contained no mismatch, migrated farthest. Sense K16/YF227, with 3 types of mismatch, showed the slowest mobility. Sense K16/YF204, with 1 mismatch, and sense K16/15H-3, with 2 mismatches, were intermediate between them. These results are consistent with the theory that duplexes with more mismatches are less stable and melt at lower denaturant concentrations. Sense YF204-III-439 and sense YF204-III-917 were hybridized with the RsaI-EcoRI fragments of YF204, YF226, 15H-3, and YF227, respectively, and electrophoresed on the denaturing gradient gels. The patterns are shown in Fig. 4. In these experiments, the number of mismatches in the duplexes made from fragments of YF204, YF226, 15H-3 and YF227 are 0, 1, 1, 2, respectively, since both probes have U at IVS2-666. In YF226, a T---, C substitution occurs at IVS2-666,
e, om ~a eO e-
eO ¢.9
2 u U.I
--I e-
¢-.
Q
O
e.m
t5 U e-
1 2 3 4 Fig. 3. The typical pattern of DGGE of duplexes between
sense K16 and cloned DNAs. DNAs used for making duplexes are (lane 1) YF226, (lane 2) YF204, (lane 3) 15H-3, (lane 4) YF227. Electrophoresis was carried out in a denaturant (0-100%) gradient gel.
m
1 X
2
3
m
4
5
/
~
Sense YF204-111-439 Sense
6
7
8 /
YF204-HI-917
Fig. 4 . T h e t y p i c a l p a t t e r n o f D G G E of duplexes between either sense YF204-III-439(lanes 1-4) or sense YF204-III-917 (lanes 5-8) and cloned DNAs. DNAs used for making duplexes are shown above the photograph. Electrophoresis was carried out in a denaturant (0-100%) gradient gel.
66
while in 15H-3 a C ~ T substitution occurs at IVS2-654. The different types of substitution at different positions reflected the mobilities of these duplexes in Fig. 4. Theoretically, the duplexes made from a D N A fragment with sense YF204III-439 and sense YF204-III-917, respectively, were the same when they were applied to a gradient gel since the 'overhang' part of the probes had been cleaved. However, in Fig. 4, bands obtained from the shorter probe (sense YF204-II-439) were broader than those from the longer probe (sense YF204-III-917). Since the band of sense YF204III-439/YF204 (lane 1 of Fig. 4) was not clearly separated from that of sense YF204-III-439/ YF226 (lane 2 of Fig. 4), the duplexes made of genomic D N A from an individual exhibiting 2 bands will be likely to be observed as a single broad band. For this reason, sense YF204-III-917 was used in addition to sense K16 in the screening of total genomic D N A s to be described below.
Screening of genomic DNA Total genomic D N A s obtained from 79 individuals (including 20 children) were examined by D G G E . D G G E patterns from 1 family using 2 probes are shown in Fig. 5. When the sense K16 probe was used, 2 clearly identified bands were observed (Fig. 5B). Mobilities of the faster band and the slower band were identical with those of sense K 1 6 / Y F 2 2 6 and sense K 1 6 / Y F 2 0 4 (Fig. 5A), respectively. Since the sense K16 probe, YF226 and YF204 have C, C and T, respectively, at IVS2-666 of the /3-globin gene, the former duplex has no haismatch but the latter has a mismatch that makes its migration slower. The genotype of an individual showing a single faster band was termed C / C while that of an individual showing a single slower band was termed T/T. Accordingly, the genotype of an individual exhibiting 2 bands was C / T . Screening of the same population was also conducted using sense YF204-III-917 which has U at IVS2-666. Two clearly identified bands were observed (Fig. 5D). In this case, duplexes made of genomic DNAs from individuals with genotype C / C migrated slower than those made of genomic D N A s obtained from individuals with genotype T / T . Thus, the genotype of each individual was independently determined using these 2 probes
TIT C/C
TIT
F10
[] ©
'--I--'
i
A
1
,
i
[]
©
C/T
C/T
C/C
~
[]
C/T
B
d
C/T
C
2
1
2
Fig. 5. D G G E patterns of duplexes from human genomic DNAs, together with those from cloned DNAs as comparison. (A) The duplexes with sense K16 and the cloned DNAs. Lane 1: sense K16/YF226 (YF226 has C at IVS2-666). Lane 2: sense K16/YF204 (YF204 has T at IVS2-666). (B) The duplexes with sense K16 and genomic DNAs from 1 nuclear family. The genotypes are shown with the pedigree. The mobilities of faster bands and slower bands are identical with those of sense K16/YF226 (lane 1 of (A)) and sense K16/YF204 (lane 2 of (A)), respectively. (C) The duplexes with sense YF204-III-917 and the cloned DNAs. Lane 1: sense YF204-III-917/YF204. Lane 2: sense YF204-III-917/YF226. (D) The duplexes with sense YF204-II1-917 and the genomic DNAs from the same family shown in (B). The mobilities of faster bands and slower bands are identical with those of sense YF204-III-917/YF204 (lane 1 of (C)) and sense YF204-III-917/YF226 (lane 2 of (C)), respectively.
and the results were in complete agreement. Table 1 presents the distribution of the genotypes and allele frequencies obtained from these experiments TABLE 1 DISTRIBUTION OF GENOTYPES A N D THEIR QUENCIES 1N 59 UNRELATED JAPANESE
FRE-
Genotype
Number observed
Number expected
Frequency
C/C C/T T/T
14 29 16
13.77 29.47 15.76
C = 0.4831 T = 0.5169 X2 = 0.0150, 0.950 > P > 0.900 (dr = 1)
Total
59
59.00
67 TABLE 2 DISTRIBUTION OF GENOTYPES AT IVS2-666 OF THE fl-GLOBIN GENE IN 11 FAMILIES Genotype of parents
Number Number Expected Observed of of genotype genotype couples children of of children children
TT x C/T
1
2
C/T x C/T
4
8
C/C x T/T C/C x C/T
1 4
2 6
C/C × C/C
1
2
C/T T/T T/T C/T C/C C/T C/T c/c C/C
2 x C/T 1 x T/T 4 x C/T 3 x C/C 2 x C/T 5 × C/T 1× c / c 2 x C/C
among 59 unrelated individuals. The frequencies of the alleles C and T were 0.48 and 0.52, respectively. The distribution of the observed genotypes is in good agreement with the expectation predicted from the allele frequencies on the assumption of the Hardy-Weinberg equilibrium. The distribution of genotypes in the children of 11 couples is shown in Table 2. There was no exception to the genotypes among 20 children of the 11 couples expected on the basis of a 2-allele system. Discussion
The results obtained with D G G E show that we detected the polymorphism based on a nucleotide substitution at IVS2-666 (see Results). The samples of the same population were also screened in our laboratory by the RNase cleavage method (Hiyama et al., 1989). The results obtained with this method, in which the location of the mismatch can be estimated from the size of the newly produced fragments from RNase A cleavage, also demonstrated that the polymorphic substitution occurred at IVS2-666. In addition, results of family studies employing these 2 methods demonstrated that the sequence difference that produced the polymorphism is inherited in a Mendelian fashion (Table 2) (Hiyama et al., 1989). This indicates that the alteration of melting behavior observed in this study was not caused by artifacts due to base methylation or partial digestion of examined D N A s with restriction enzymes. To confirm the
results obtained with D G G E and the RNase cleavage method, sequence analysis of this region was carried out (Hiyama et al., 1989). Direct sequencing of the products of polymerase chain reaction (PCR) (Saiki et al., 1985) from the family members shown in Fig. 5, except the daughter, revealed that the nucleotides at IVS2-666 in husband, wife and their son were T, C, and both C and T, respectively. Studies on the fl-globin genes clearly show that the allele in which T is replaced by C at IVS2-666 is not necessarily associated with the fl-thalassemia allele in which 4 bases within codons 41 and 42 of the fl-globin gene are deleted (Kimura et al., 1983; Takihara et al., 1984; Kazazian et al., 1984; Fukumaki et al., 1987). Thus, an individual having an allele with a T to C substitution at IVS2-666 is not always a patient with thalassemia (Orkin et al., 1982). Several methods have been developed for the detection of mutations in genomic DNA. R F L P is the method that was first introduced and has been used most widely (Flavell et al., 1978; Geever et al., 1981). Unfortunately, this approach has one significant limitation in that variations occurring outside the restriction enzyme recognition sites cannot be detected, and even large panels of restriction enzymes recognize only a fraction of the existing sequence variations. In this method, per diploid genome, each combination of a unique probe and a 6-cutter restriction enzyme examines 36 bp consisting of 24 bp for loss of a cutting site and on average 12 bp for gain of a cutting site (Delehanty et al., 1986). Base substitutions which do not alter the restriction enzyme cleavage sites can be detected by the differential hybridization of synthetic oligonucleotide probes, if the location and nature of the mutation and the surrounding nucleotides are known (Kidd et al., 1983; Pirastu et al., 1983). This method, therefore, is applicable only to genes that are well characterized prior to examination. Only 40 bp per diploid genome can be examined at one examination, since, for practical purpose, the maximum length of each probe should be less than 20 nucleotides. For these reasons, the abovementioned 2 methods are not efficient in screening for new mutations. The RNase A cleavage method has been developed to detect single-base substitutions in
68 cloned and genomic D N A (Myers et al., 1985d). Since several hundreds to 1000 bp can be examined at a time with 1 probe, the efficiency of this method is higher than that of the 2 methods described above. Moreover, the advantage of this method is that it not only detects mutations but also localizes them. The limitation of this method, however, is that all types of mismatch are not always cleaved by RNase A. Myers et al. (1985d; Myers and Maniatis, 1986) reported that approximately one-third of all possible single-base substitutions were not detectable, even if 2 types of probe were used, one of which was homologous to one strand of the test D N A and the other homologous to the opposite strand of the test DNA. Based on the results of our study (Hiyama et al., 1989) employing the RNase cleavage method, we are confident that this is a very effective method to detect variations in cloned DNAs or DNAs amplified by PCR (Saiki et al., 1985) although it has this limitation. D G G E of DNA : D N A heteroduplexes has been reported to be a very useful tool for identifying base substitutions in human genomic D N A (Myers et al., 1985e; Myers and Maniatis, 1986). By this method, approximately 600 bp can be analyzed with 1 probe (Noll and Collins, 1987), the limitations of this method, however, were described earlier and until they are resolved, our calculation based on our preliminary experiments shows that the efficiency of this method will be similar to that of the RNase cleavage method. In contrast to this, pure single-stranded RNA probes may be consistently obtained very easily employing the in vitro transcription system in which D N A fragments inserted into transcription vectors are used as templates (Melton et al., 1984; Myers et al., 1985d). Smith et al. (1986) demonstrated that single-base mutations in the N S gene of the influenza virus were detectable by D G G E of R N A : R N A duplexes or D N A : RNA duplexes which had been made by hybridization between virion RNA and RNA probes prepared by this method or by D N A probes. In the present study, we demonstrated that sequence differences present not only in cloned D N A but also in genomic D N A are detectable by D G G E of duplexes made by hybridization of RNA probes with DNAs. Because of the reduction in
the time needed to prepare R N A probes, the utility of this method for screening is markedly increased. Hybridization can be carried out in a solution containing 2 or more different probes selected so that the bands of resulting different duplexes will be distributed broadly over the gradient at the end of run. For example, let us assume that 500 bp per haploid genome will be able to be effectively examined by 1 probe and 5 probes will be used in 1 examination to produce 5 duplexes which can be separated on 1 lane. Based on this assumption, we estimate that this method is at least about 5-10 times more efficient than the D G G E of D N A : D N A heteroduplexes. The detection of mutations should be made by direct comparison of D N A sequences of appropriate genes isolated from each individual of the study group. However, it is very questionable whether present sequencing techniques have the necessary accuracy for screening for mutations. The error level may be still of the order of 10 -4 , even after improvement in accuracy is attained by using a better optical processor and a powerful computer (Wada, 1987). At present, moreover, the efficiency of D N A sequencing analysis is too low to detect mutations. By contrast, we could demonstrate the usefulness of D G G E of R N A : D N A duplexes in the model experiment in which sequence differences in genes of individuals were efficiently detected. The sensitivity of D G G E of R N A : D N A duplexes could be dramatically increased by amplifying the genomic D N A by PCR (Saiki et al., 1985). From our preliminary experiences (Takahashi et al., in preparation), the intensity of the bands of duplexes made from genomic DNAs amplified by PCR was the same as that of the band of duplexes from cloned DNAs. However, even if this improvement is incorporated, D G G E alone will not allow the detection of all possible single-base mutations, since our preliminary data showed that some single-base mismatches would not lead to a shift in the mobility of the duplex in DGGE. If the RNase cleavage method is employed in combination with D G G E , there is a high probability of detecting mutations which cannot be detected by D G G E alone. Thus, a very large fraction of all possible mutations should be detected by combining the 2 methods, although
69 t h e r e is s o m e o v e r l a p i n t h e m u t a t i o n s t h a t c a n b e detected by the 2 methods. The sequence differences caused by the deletion o f l a r g e f r a g m e n t s will b e d e t e c t a b l e b y D G G E , if t h e f o l l o w i n g t e c h n i q u e is i n c o r p o r a t e d . T h e u s e of either electronic autofluorography (Davidson and Case, 1982) or a 2-dimensional fl-scanner to impart a greater dynamic range to quantification of 32p-labeled bands should be especially effective for detection of the bands containing half of the expected diploid amount of DNA. Introduction of these techniques could be powerful in studies of radiation mutagenesis, where deletion mutations a p p e a r to b e p r e d o m i n a n t . At present, we think that the combination of D G G E a n d t h e R N a s e c l e a v a g e m e t h o d is e f f e c tive a n d p r a c t i c a b l e f o r s c r e e n i n g f o r m u t a t i o n s . U s e o f t h e s e m e t h o d s will a l s o b e c o n d u c t i v e t o the detection of a great number of polymorphisms a n d m a k e it p o s s i b l e t o p r e p a r e l i n k a g e m a p s i n great detail.
Acknowledgements W e a r e g r a t e f u l t o N . K o s a k a , H. O m i n e a n d Y. K i m u r a f o r t h e i r e x c e l l e n t t e c h n i q u e i n D G G E . W e a l s o t h a n k M s . J. K a n e k o f o r h e r t e c h n i c a l work throughout this work. We are much indebted t o D r . J.V. N e e l f o r h i s a s s i s t a n c e in p r e p a r i n g t h e manuscript.
References Davidson, J.B., and A.L. Case (1982) Rapid electronic autofluorography of labeled macromolecules on two-dimensional gels, Science, 215, 1398-1400. Delehanty, J., R.L. White and M.L Mendelsohn (1986) Approaches to determining mutation rates in human DNA, Mutation Res., 167, 215-232. Fischer, S.G., and L.S. Lerman (1979) Two-dimensional electrophoresis separation of restriction enzyme fragments of DNA, in: R. Wu (Ed.), Methods in Enzymology, Vol. 68, Academic Press, New York, pp. 183-191. Flavell, R.A., J.M. Kooter, E. De Boer, P.F. Little and R. Williamson (1978) Analysis of the beta-delta-globin gene loci in normal and Hb Lepore DNA: direct determination of gene linkage and intergene distance, Cell, 15, 25-41. Fukumaki, Y., E. Matsunaga, T. Taga, Y. Takihara and T. Nakamura (1987) Multiple origins of the fl-thalassemia gene with a four-nucleotide deletion in its second exon, Jpn. J. Hum. Genet., 32, 155. Geever, R.F., L.B. Wilson, F.S. Nallaseth, P.F. Milner, M.
Bittner and J.T. Wilson (1981) Direct identification of sickle cell anemia by blot hybridization, Proc. Natl. Acad. Sci. (U.S.A.), 78, 5081-5085. Hamilton, H.B. (1985) Genetics and the atomic bombs in Hiroshima and Nagasaki, Am. J. Med. Genet., 20, 541-546. Hiyama, K., M. Kodaira and C. Satoh (1988a) Detection of mismatches in RNA : DNA duplexes, Jpn. J. Hum. Genet., 33, 234. Hiyama, K., N. Takahashi, M. Kodaira and C. Satoh (1988b) Detection of mismatches in R N A : D N A duplexes. Report 2, Jpn. J. Hum. Genet., 34, 76. Hiyama, K., M. Kodaira and C. Satoh (1989) Detection of deletions, insertions and single nucleotide substitutions in cloned /3-globin genes and new polymorphic nucleotide substitutions in /3-globin genes in a Japanese population using ribonuclease cleavage at mismatches in R N A : D N A duplexes, submitted for publication. Kazazian Jr., H.H., S.H. Orkin, S.E. Antonarakis, J.P. Sexton, C.D. Boehm, S.C. Goff and P.G. Waber (1984) Molecular characterization of seven fl-thalassemia mutations in Asian Indians, EMBO J., 3, 593-596. Kidd, V.J., R.B. Wallace, K. ltakura and S.L.C. Woo (1983) ai-Antitrypsin deficiency detection by direct analysis of the mutation in the gene, Nature (London), 304, 230-234. Kimura, A., E. Matsunaga, Y. Takihara, T. Nakamura, Y. Takagi, S.-T. Lin and H.-T. Lee (1983) Structural analysis of a fl-thalassemia gene found in Taiwan, J. Biol. Chem., 258, 2748-2749. Lerman, L.S., S.G. Fischer, I. Hurley, K. Silverstein and N. Lumelsky (1984) Sequence-determined DNA separations, Annu. Rev. Biophys. Bioeng., 13, 399-423. Lerman, L.S., K. Silverstein and E. Grinfeld (1986) Searching for gene defects by denaturing gradient gel electrophoresis, Cold Spring Harbor Symp. Quant. Biol., 51,285-297. Maniatis, T., E.F. Fritsch and J. Sambrook (1982) Isolation of high-molecular-weight, eukaryotic DNA from cells grown in tissue culture, in: Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 280-281. Melton, D.A., P.A. Krieg, M.R. Rebagliati, T. Maniatis, K. Zinn and M.R. Green (1984) Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter, Nucleic Acids Res., 12, 7035-7056. Myers, R.M., and T. Maniatis (1986) Recent advances in the development of methods for detecting single-base substitutions associated with human genetic diseases, Cold Spring Harbor Symp. Quant. Biol., 51, 275-284. Myers, R.M., L.S. Lerman and T. Maniatis (1985a) A general method for saturation mutagenesis of cloned DNA fragments, Science, 229, 242-247. Myers, R.M., S.G. Fischer, LS. Lerman and T. Maniatis (1985b) Nearly all single base substitutions in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis, Nucleic Acids Res., 13, 3131-3145. Myers, R.M., S.G. Fischer, T. Maniatis and L.S. Lerman (1985c) Modifications of the melting properties of duplex
70 DNA by attachment of a GC-rich DNA sequence as determined by denaturing gradient gel electrophoresis, Nucleic Acids Res., 13, 3111-3129. Myers, R.M., Z. Larin and T. Maniatis (1985d) Detection of single base substitutions by ribonuclease cleavage at mismatches in R N A : D N A duplexes, Science, 230, 1242-1246. Myers, R.M., N. Lumelsky, L.S. Lerman and T. Maniatis (1985e) Detection of single base substitutions in total genomic DNA, Nature (London), 313, 495-498. Myers, R.M., T. Maniatis and L.S. Lerman (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis, in: R. Wu (Ed.), Methods in Enzymology, Vol. 155, Academic Press, New York, pp. 501-527. Neel, J.V., C. Satoh, K. Goriki, J. Asakawa, M. Fujita, N. Takahashi, T. Kageoka and R. Hazama (1988) Search for mutations altering protein charge and/or function in children of atomic bomb survivors: final report, Am. J. Hum. Genet., 42, 663-676. Noll, W.W., and M. Collins (1987) Detection of human DNA polymorphisms with a simplified denaturing gradient gel electrophoresis technique, Proc. Natl. Acad. Sci. (U.S.A.), 84, 3339-3343. Orkin, S.H. (1984) The mutation and polymorphism of the human fl-globin gene and its surrounding DNA, Annu. Rev. Genet., 18, 131-171. Orkin, S.H., H.H. Kazazian Jr., S.E. Antonarakis, S.C. Goff, C.D. Boehm, J.P. Sexton, P.G. Waber and P.J.V. Giardina (1982) Linkage of fl-thalassemia mutations and fl-globin gene polymorphisms with DNA polymorphisms in human fl-globin gene cluster, Nature (London), 296, 627-631. Pirastu, M., Y.W. Kan, A. Cao, B.J. Conner, R.L. Teplitz and R.B. Wallace (1983) Prenatal diagnosis of fl-thalassemia: detection of a single nucleotide mutation in DNA, New Engl. J. Med., 309, 284-287. Saiki, R.K., S. Scharf, F. Faloona, K.B. Mullis, G.T. Horn,
H.A. Erlich and N. Arnheim (1985) Enzymatic amplification of fl-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia, Science, 230, 1350-1354. Satoh, C., and J.V. Neel (1988) Biochemical mutations in the children of atomic bomb survivors, in: H. Takebe and J. Utsunomiya (Eds.), Genetics of Human Tumors in Japan, Gann Monograph on Cancer Research No. 35, Japan Scientific Societies Press, Tokyo, and Taylor and Francis, London and Philadelphia, pp. 191-208. Satoh, C., K. Hiyama, N. Takahashi and M. Kodaira (1988) Detection of new variations in cloned fl-globin genes and a DNA polymorphism in a Japanese population with ribonuclease cleavage at mismatches in R N A : D N A duplexes, Genome, 30 (Suppl.), 1, 369. Paper presented at the 16th International Congress of Genetics, Toronto, 20-27 August. Smith, F.I., J.D. Parvin and P. Palese (1986) Detection of single base substitutions in influenza virus RNA molecules by denaturing gradient gel electrophoresis of RNA°RNA or D N A - R N A heteroduplexes, Virology, 150, 55-64. Takahashi, N., K. Hiyama, M. Kodaira, H. Omine and C. Satoh (1988a) Detection of mismatches in RNA:DNA duplexes. Report 3, Jpn. J. Hum. Genet., 34, 77. Takahashi, N., K. Hiyama, M. Kodaira and C. Satoh (1988b) Detection of mismatches in RNA : DNA duplexes. Report 4, Paper presented at l l t h Meeting of the Japan Molecular Biology Society, Tokyo, 20-23 December. Takihara, Y., E. Matsunaga, T. Nakamura, S.-T. Lin, H.-T. Lee, Y. Fukumaki and Y. Takagi (1984) One base substitution in IVS-2 caused a fl+-thalassemia phenotype in a Chinese patient, Biochem. Biophys. Res. Commun., 121, 324-330. Wada, A. (1987) Automated high-speed DNA sequencing, Nature (London), 325, 771-772.
