Proc. NatL Acad. Sci. USA
`Vol. 80, pp. 278-282, January 1983
`Medical Sciences
`
`Detection of sickle cell (38-globin allele by hybridization with
`synthetic oligonucleotides
`(sickle cell anemia/prenatal diagnosis/genetic disease)
`BRENDA J. CONNER*, ANTONIO A. REYEst, CHRISTOPHE MORIN+, KEIICHI ITAKURAt, R. L. TEPLITZ*, AND
`R. BRUCE W ALLACEt
`
`*Division of Cytogenetics and Cytology, City of Hope Medical Center, and tDivision of Biology, Molecular Genetics Section, City of Hope Research Institute,
`Duarte, California, 91010; and tLaboratorie de Chimie, Institut Pasteur, Paris, France
`
`Communicated by Norman H. Horowitz, September 23, 1982
`
`Two 19-base-long oligonucleotides were synthe(cid:173)
`ABSTRACT
`sized, one complementary to the normal human P-globin gene
`(/JA) and one complementary to the sickle cell fJ-globin gene (/Js).
`The nonadecanucleotides were radioactively labeled and used as
`probes in DNA hybridization. Under appropriate hybridization
`conditions, these probes can be used to distinguish the pA gene
`from the ps allele. The DNA from individuals homozygous for the
`normal fJ-globin gene (/JA pA) only hybridized with the pA specific
`probe; the DNA from those homozygous for the sickle cell fJ-globin
`gene (ps ps) only hybridized with the ps specific probe. The DNA
`from heterozygous individuals (f1Af18) hybridized with both
`probes. This allele-specific hybridization behavior of oligonucleo(cid:173)
`tides provides a general method for diagnosis of any genetic dis(cid:173)
`ease which involves a point mutation in the DNA sequence of a
`single-copy gene.
`
`Synthetic oligodeoxyribonucleotides have been shown to hy(cid:173)
`bridize specifically to complementary DNA sequences (1-3).
`Under appropriate hybridization conditions, only perfectly
`base-paired oligonucleotide· DNA duplexes will form; duplexes
`containing a single mismatched base pair will not be stable. This
`high degree of hybridization specificity has led to the devel(cid:173)
`opment of a general method for using synthetic oligonucleotides
`as specific probes to identify cloned DNAs coding for proteins
`of interest. Recently, this technique has been applied to the
`successful isolation of a human P2-microglobulin eDNA done
`(4) as well as a murine transplantation antigen eDNA clone (5).
`Because a mutation in a single base in the DNA sequence of
`a gene would affect the hybridization behavior of an oligonu(cid:173)
`cleotide complementary to the region of the mutation (2), oli(cid:173)
`gonucleotide hybridization has the potential to provide a
`method of detecting single-base changes within genomic DNA.
`Point mutations are the cause of a substantial number of human
`genetic diseases (6). Synthetic oligonucleotides, therefore,
`could be used as specific probes for determination of genotype
`and aid in diagnosis of genetic disease, even prenatally.
`We chose the ~globin gene as a model system to test the
`applicability of using synthetic oligonucleotides to detect a point
`mutation within a single copy gene. The ~globin gene is a
`member of the single-copy ~globin-like gene family which in(cid:173)
`cludes the e-, cy-, Ay-, l>-, and ~globin genes arranged 5' to
`3' in·order of expression during development (7). Recently,
`these.genes have been cloned, restriction maps have been de(cid:173)
`fined, and much ofthe DNA sequence has been determined
`for the coding and immediate flanking regions (8-11). Sickle
`cell anemia, a human genetic disease found predominantly in
`the Black population, is the result of a single base pair (bp)
`
`The publication costs of this article were.defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertise(cid:173)
`ment'' in accordance with 18 U.S. C. §1734 solely to. indicate this fact.
`
`change (adenine to thymine) in the ~globin gene, correspond(cid:173)
`ing to the sixth amino acid residue (changing glutamic acid to
`valine) in the ~globin protein (12, 13). The sickle cell disorders
`follow a single-gene Mendelian mode of inheritance. Instead
`of the normal ~globin genotype (pApA), individuals with sickle
`cell trait have one normal ~globin gene and one sickle cell allele
`(pAp5); those with sickle cell disease have two sickle cell alleles
`(p5p 5) and no pA gene.
`In this report, we show that nonadecanucleotides comple(cid:173)
`mentary to the ~globin gene (pA) or to the sickle cell allele
`(p5) in the region of the sickle cell point mutation can: (i) dis(cid:173)
`tinguish pA from p 5 as well as from other members of the /1-
`globin-like gene family, (ii) specifically detect the single-copy
`~globin gene in human genomic DNA, and (iii) allow the un(cid:173)
`ambiguous determination of ~globin genotype of individuals,
`confirming the hematological diagnosis of sickle cell trait or sic(cid:173)
`kle cell disease. Applied to prenatal diagnosis of sickle cell dis(cid:173)
`orders, this procedure offers several advantages over methods
`now available. More significantly, these techniques may be ap(cid:173)
`plied generally for diagnosis of genetic diseases that involve a
`specific change-such as a base substitution, insertion, or dele(cid:173)
`tion-in the DNA sequence of a single gene.
`
`MATERIALS AND METHODS
`Chemical Synthesis of Oligodeoxyribonucleotides of Unique
`Sequence. Nonadecanucleotides (Table 1) were synthesized on
`a solid support by the modified triester approach as described
`(14).
`32P-Labeling of Synthetic Oligonucleotides. Synthetic oli(cid:173)
`gonucleotides were labeled with adenosine 5' -[ y-32P]triphos(cid:173)
`phates (ICN, crude preparation, >7,000 Ci/mmol; 1 Ci = 3. 7
`X 1010 Bq) by a kinase reaction (1). Separation of labeled oli(cid:173)
`gonucleotide from unlabeled nonadecanucleotide and reaction
`products by homoehromatography (15) yielded oligonucleotide
`probes with specific activities of approximately 2 X 109 cpm/
`JLg.
`Source of Recombinant DNA and Transformants. A-HpG1
`and A-H yG5 and pBR322-HpPst clones were a generous gift
`ofT. Maniatis. The pBR322-HpPst plasmid contained a 4.4-
`kilobase (kb) Pst I fragment of pA subcloned in the Pst I site of
`pBR322 (16). DNA was isolated as described (3, 17). Recom-
`
`Abbreviations: bp, base pair{s); kb, kilobase(s); {3A, normal {3-globin
`gene; {3s, sickle cell (3-globin gene; f3.A {3A, normal {3-globin genotype;
`{3A~, sickle cell trait genotype; {3s{3s, sickle cell disease genotype; A(cid:173)
`H{3G1, recombinant bacteriophage constructed with DNA insert of
`human 8-globin and {3-globin genes; A-H-yG5 recombinant bacterio(cid:173)
`phage constructed with DNA insert of human c"Y- and A-y-globin genes;
`pBR322-H{3Pst, recombinant plasmid constructed with insert of human
`{3-globin gene.
`
`278
`
`GeneDX 1018, pg. 1
`
`

`

`Medical Sciences: Conner et aL
`
`Proc. NatL Acad. Sci. USA 80 (1983')
`
`279
`
`Table 1. DNA sequence of synthetic oligonucleotide probes for
`normal/3-globin gene (pA) and sickle cell/3-globin allele ({38 )
`Probe
`Gene
`DNA sequence
`H{319A
`5 ' CT CCT GAG GAG AAG TCT GC 3 '
`{3A
`H{319A'
`3 ' GA GGA CTC CTC TTC AGA CG5 '
`
`H{319S
`H{319S'
`
`5 ' CT CCT GTG GAG AAG TCT GC 3 '
`3 ' GA GGA CAC CTC TTC AGA CG 5 '
`
`binant DNA was handled in accordance with the National In(cid:173)
`stitutes of Health guidelines.
`Isolation of Human DNA. Peripheral blood samples (10-20
`ml) were collected from scientific personnel (normal for ~glo­
`bin) or from patients with sickle cell trait (flAf3s.) or sickle cell
`disease (13SfJS); all gave infOrmed consent. DNA was isolated
`from leukocytes as described (18) with some modifications. In(cid:173)
`stead of dialysis, the DNA was extracted three times with
`phenol/chloroform, 3:1 (vol/vol), and treated with RNase (19).
`The average yield was 50 f.Lg of DNA per ml of whole blood.
`Restriction Endonuclease Digestion of DNA. A-Hf3G 1 DNA
`and pBR322-Hf3Pst DNA were digested at 1 enzyme unit/ f.Lg
`of DNA at 37"C for 30 min. Human DNA (10 f.Lg) was digested
`at 5 enzyme units/ f.Lg of DNA at 37"C for 4 hr; additional en(cid:173)
`donuclease was then added at 5 units/ f.Lg of DNA and digestion
`was continued for 12 hr or overnight. The restriction endonu(cid:173)
`cleases used were BamHI, EcoRI, and Hpa I (Boehringer
`Mannheim). BamHI and Hpa I digestions were carried out in
`7 mM NaCI/7 mM Tris·HCl, pH 7.5/7 mM dithiothreitol/7
`mM magnesium acetate. The buffer for EcoRI was 100 mM Tris,
`pH 7.2/5 mM magnesium acetate/100 mM NaCI/0.02% Non(cid:173)
`idet P-40. When double digestions were required, the amounts
`of enzyme and times of digestion were as described above, ex(cid:173)
`cept the endonuclease requiring low-salt buffer was used first.
`The buffer was then adjusted to higher salt conditions for diges(cid:173)
`tion with EcoRI. Digestions were terminated by heating the
`sample at 65°C for 3 min.
`Electrophoresis and Hybridization of DNA. Digested DNAs
`were loaded into individual wells of a vertical (1 or 3 mm thick)
`1.0-1.2% agarose gel (SeaKem), electrophoresed at 1 V /em at
`50 V for 750 V hr in a Tris borate electrophoresis buffer, stained
`with ethidium bromide (Calbiochem) at 0.25 f.Lg/ml for 30 min,
`and photographed over UV light (3, 20). The DNA was dena-
`
`tured in situ and transferred to nitrocellulose paper (21) .. Al(cid:173)
`ternatively, the gel itself was dried for direct hybridization (22).
`After ethidium bromide staining, the gel was denatured with
`0.5 M NaOH and 0.15 M NaCl at room temperature for 30 min
`and neutralized in 0.5 M Tris, pH 8.0/0.15 M NaCl at 4°C for
`30 min. The gel was then dried under vacuum onto Whatman
`3MM paper with a Hoefer gel dryer. The dry gel was rinsed in
`0.15 M NaCl to remove backing paper. Nitrocellulose paper or
`the dried gel was sealed in plastic (Dazey Seal-A-Meal), hy(cid:173)
`bridized, and washed as specified in the figure legends. The
`nitrocellulose or gel was blotted dry with Whatman 3MM pa(cid:173)
`per, wrapped in Saran Wrap, and autoradiographed between
`·two Quanta III intensifier screens (DuPont) at -80°C for 4 hr
`or 3--5 days.
`
`RESULTS
`Rationale of Specificity of Oligonucleotide Hybridization.
`The DNA sequences of the synthetic oligonucleotides used in
`this study are given in Table 1. The position and length of the
`sequence of the oligonucleotides were based on several criteria.
`(i) The oligonucleotide was designed to be 19 nucleotides long
`in order that this sequence would have a high probability of
`recognizing a unique sequence (23). Hf319A and Hf319A' were
`specific for normal ~globin DNA (flA); H,819S and HJU9S'
`were specific for sickle cell ~globin DNA~). (ii) The sickle
`cell mutation was positioned near the center of the sequence
`to maximize thermal instability of mismatch hybridization. (iii)
`The sequences synthesized were not complementary to the
`e-, cy-, A'}'-, or 8-globin genes in the region of probe comple(cid:173)
`mentarity (Table 2).
`The HfJ19A' oligonucleotide (or Hf319A) should form a per(cid:173)
`fect hybrid with the fJA DNA (Table 3). If the sickle cell point
`mutation ~) is present, there would be one mismatched nu(cid:173)
`cleotide at the site of the point mutation in the oligonucleo(cid:173)
`tide· DNA duplex with H,819A' (T/T mismatch) or Hf319A (A/
`A mismatch) probe. Conversely, Hf319S' (or H,819S) oligonu(cid:173)
`cleotide probe should hybridize perfectly with ~ DNA. How(cid:173)
`ever, one mismatched nucleotide would be present in a duplex
`of ,BA DNA and H,819S' (A/A mismatch) or H,819S (T/T mis(cid:173)
`match) oligonucleotide.
`
`Table 2. Amino acid sequence and DNA sequence of human f3-globin-like genes in the region of
`oligonucleotide hybridization
`
`Gene
`
`li
`
`G'Y
`A'Y
`
`5
`
`{3A
`
`3'
`1
`5'
`10
`5
`... Met Val His Phe Thr Ala Glu Glu Lys Ala Ala Val ...
`... ATG GTG CAT TTT ACT GCT GAG GAG AAG GCT GCC GTC ...
`
`...
`... Met Gly His
`Phe Thr Glu Gly Asp Lys Ala Thr
`lle
`... ATG GTG CAT TTC ACA GAG GAG GAC AAG GCT ACT ATC ...
`
`... Met Val His Leu Thr Pro Glu Glu Lys Thr Ala Val ...
`... ATG GTG CAT CTG ACT CCT GAG GAG AAG ACT GCT GTC ...
`
`... Met Val His Leu Thr Pro Glu Glu Lys Ser Ala Val ...
`... ATG GTG CAC CTG ACT CCT GAG GAG AAG TCT GCC GTT ...
`
`~
`
`... Met Val His Leu Thr Pro Val Glu Lys Ser Ala Val ...
`... ATG GTG CAC CTG ACT CCT GTG GAG AAG TCT CCC GTT ...
`
`Mismatches, no.
`H{319A
`H{319S
`
`2
`
`7
`
`1
`
`0
`
`1
`
`3
`
`8
`
`2
`
`1
`
`0
`
`Italic nucleotides represent position of base-pair differences relative to the {3A. The connecting line represents the position
`of oligonucleotide hybridization.
`
`GeneDX 1018, pg. 2
`
`

`

`280
`
`Medical Sciences: Conner et aL
`
`Proc. Natl. Acad. Sci. USA 80 (1983)
`
`Table 3. Type of base-pair mismatch in duplexes of synthetic
`oligonucleotide probe and DNA of I3A and of /35
`J3-Globin allele
`
`HJ319A
`HJ319A'
`H,S19S
`HJ319S'
`
`Perfect match
`Perfect match
`T /T mismatch
`A/ A mismatch
`
`A/ A mismatch
`T /T mismatch
`Perfect match
`Perfect match
`
`Selectivity of Oligonucleotide Hybridization to Tempera(cid:173)
`ture. The effect of temperature on oligonucleotide hybridiza(cid:173)
`tion to human J3-globin DNA is shown in Fig. L pBR322-Hf3Pst
`DNA was digested with BamHI and subjected to electropho(cid:173)
`resis in agarose gels in eight identical lanes; the DNA was dena(cid:173)
`tured in situ and transferred to nitrocellulose by the standard
`Southern procedure (21). For each duplicate, one lane was hy(cid:173)
`bridized with 5' -32P-labeled Hf319A (perfect match) and the
`other with 5' -32P-labeled Hf319S (T/T mismatch) (Table 1). The
`temperature of the hybridization and the washes that followed
`were then varied as indicated. With ·hybridization at 45°C and
`wash at 0°C, hybridization to the l. 8-kb BamHlrestriction frag(cid:173)
`ment containing the 5' end ofthe J3-globin gene was evident
`with the HfU9A (perfect match) and, to a lesser extent, with
`the Hf319S (T/T mismatch) probe. Based on the length and
`DNA sequence of the oligonucleotide, it was calculated that
`55°C would be a temperature at which the nonadecanucleo(cid:173)
`tide·DNA complex would .be stable only if base pairing were
`perfect (3). Hybridization of the gel at 45°C followed by a wash
`at 55°C removed hybridization with the Hf319S probe (T/T
`mismatch), allowing stable hybridization only with the Hf319A
`probe (perfect match). At a hybridization temperature of55°C,
`
`only the Hf319A (perfect match) probe hybridized to the 1.8-kb
`restriction fragment. Under these stringent conditions, only
`perfectly matched oligonucleotide· DNA duplexes were stable.
`Hybridization at the higher temperature increased specificity
`but resulted in a slightly decreased signal (compare 45/55 with
`55/55 in Fig. 1).
`Oligonucleotide Probes Can Differentiate f3-Giobin Gene
`from Other f3-Giobin-Like Genes. The stringent hybridization
`conditions established above were used to determine if the oli(cid:173)
`gonucleotide probes could also distinguish the J3-globin gene
`sequence from those ofthe J3-globin-like genes. The recombi(cid:173)
`nant bacteriophage A-Hf3G1 contains both the 8- and J3-globin
`genes; A-H:yG5 contains the human c'Y_ and Ay-globin genes
`(17). There is one base-pair difference in the region of Hf319A
`hybridization between 8- and /3-globin genes, and seven nu(cid:173)
`cleotide changes in this region between c'Y_ or A'Y_ and J3-globin
`(Table 2). These two DNAs were digested with EcoRI and hy(cid:173)
`bridized with 5'-32P-labeled Hf319A (Fig. 2). At 45°C the la(cid:173)
`beled probe hybridized to the fragment containing 8-globin (T /
`T mismatch) as well as to the fragment containing the J3-globin
`gene in A-Hf3G1 DNA (perfect match). At 55°C, only hybrid(cid:173)
`ization to the fragment containing the J3-globin gene was evi(cid:173)
`dent. The A-HyG5 DNA, with seven noncomplementary bases
`in the region of hybridization, did not hybridize to the probe
`at either temperature.
`Detection of f3-Giobin Gene in Human Genomic DNA. Be(cid:173)
`cause the sequence of the J3-globin gene has been determined
`and extensive restriction maps have been established (11), it is
`possible to predict the fragment sizes expected to contain the
`5' region of the J3-globin gene if total human genomic DNA
`were digested with various restriction endonucleases. Fig. 3A
`shows the results for digestion of A-Hf3G1 DNA and human
`genomic DNA normal for the J3-globin gene (f3Af3A) with
`BamHI or double digestion with either BamHI/EcoRI or Hpa
`1/EcoRI. From the known restriction map of the J3-globin gene
`(ll), the sizes of restriction fragments predicted to contain the
`5' end of the J3-globin gene are: 1.8 kb, BamHI; 1.8 kb, BamHI/
`EcoRI; 2.2 kb, Hpa 1/EcoRI (Fig. 38). A Hf3G1 was similarly
`digested and included on the gelas a marker. In the BamHI
`
`2
`
`2
`
`2
`
`2
`
`FIG. 1. Effect of temperature on the hybridization of oligonucleo(cid:173)
`tides to globin DNA. Individual lanes of pBR322-HJ3Pst DNA (1 ng)
`digested with BamHI were hybridized for 16 hrin hybridization buffer
`[5x Denhardts (lx modified Denhardts is 0.02% bovine serum albu(cid:173)
`min/0.02% polyvinylpyrrolidone/0;02% NaDodS04/0.02% Ficoll)
`containing 10% dextran sulfate (Pharmacia or Sigma), 6 x NET (1 x
`NET is 0.15 M NaCl/0.03 M Tris·HCl, pH 8.0/1 mM EDTA), and 0.5%
`Nonidet P-40]. The buffer also contained (106 cpm/ml) of a labeled
`oligonucleotide probe [5'-32P]HJ319A (perfect match) (lanes 1) or[5'-
`3¥JHJ319S (T/T mismatch) (lanes 2). After hybridization at 45•c or
`55°C as indicated, each nitrocellulose paper strip was washed with
`three changes (15 min each) of 0.9 M NaCl/0.09 M sodium citrate at
`o•c. Half of the filters were then washed for 1 min at 55•c; as indicated
`(the numbers at the top of the figure indicate hybridization temper(cid:173)
`ature first and wash temperature second). The ethidium bromide(cid:173)
`stained gel is shown at the left. Arrow, position of the 5' end of the 13-
`globin gene.
`
`{3 -
`
`8 -
`
`Gy-
`
`- 7.2 Kb
`
`-
`
`- 5.2 Kb
`
`- 2.25
`
`Ay -
`
`- 2.7 Kb
`
`45°
`
`55°
`
`45• ss•
`
`FIG. 2. Hybridization of 32P-labeled HJ319A to G-y-, A-y-, 8-, and 13-
`globin genes. Duplicate samples of A-H/3G1 DNA (Left) or A-HIG5
`(Right) (0.25 p,g) were digested withEcoRI and hybridized with 5'- 2P(cid:173)
`labeled HJ319A probe at either 45•c or 55~C for 18 hr (as in Fig. 1). The
`positions of the EcoRI restriction fragments containing the 5' end of
`each globin gene are noted to the left of the ethidium bromide-stained
`gels.
`
`GeneDX 1018, pg. 3
`
`

`

`Medical Sciences: Conner et al
`
`Proc. Natl Acad. Sci. USA 80 (1983)
`
`281
`
`i
`E
`0
`QJ
`
`i
`0:
`0:
`-
`E ... o
`~+~
`til ~
`~ ~ ~
`
`I
`
`I
`
`I
`
`I
`
`I
`
`I
`
`A
`
`H.8 19A'
`
`B
`
`H.8 19S
`
`q
`
`(/)
`
`(/)
`
`G
`~ ~ ~ ~ ~ ~
`~ C!l.
`C!l.
`C!l.
`C!l.
`C!l.
`C!l.
`I
`I
`I
`I
`I
`I
`I
`
`q
`
`(/)
`
`Cl)
`
`•
`
`2.2Kb -
`1.8Kb -
`
`•
`
`••
`
`A.
`
`B.
`
`/3
`0
`it
`it
`it
`i
`ii_ i
`0:
`0:
`o
`o Eo
`EEo Eo
`o
`tij ~ tij
`tij ~~~ ~ tij
`tij
`._I ___ _.l--i£:11'--'-1 --'-1-'1.....1.....,1 li=Jir------ >.H,BG 1
`~
`~
`
`FIG. 3. Hybridization of 32p-labeled H,919A' to human genomic
`DNA digested with different restriction endonucleases. (A) A-H/3G1
`DNA (lanes a) or 10 ,...g of human genomic DNA (lanes b) (,SA ,SA) was
`subjected to single or double digestion as indicated. After electropho(cid:173)
`resis, the agarose gel was dried, hybridized with [5'-32P)HJU9A' at 107
`cpm/ml for 2 hr at 55°C, washed at OOC (Fig. 1), and then given a 1-min
`wash in 0.9 M NaCI/0.09 Msodium citrate at550C. A-H/3G1 was loaded
`at 150 pg as a control for intensity of hybridization of a single copy
`gene, except that in the BamlfljEcoRI double digestion 300 pg of DNA
`was loaded. (B) Localization of selected restriction enzyme sites for
`EcoRI, Bamlfl, and Hpa I (10). The expected sizes of the Bamlfl or
`BamlfljEcoRI fragment (1.8 kb) and the Hpa 1/EcoRI fragment (2.2
`kb) are indicated.
`
`and the Hpa I/EcoRI lanes, the amount of A-H,8G1 DNA
`loaded was equivalent to a single-copy gene in 10 J.Lg of human
`DNA (5 x 10-6 pmol). In each digestion, the hybridizing band
`in the genomic digests comigrated with the restriction frag(cid:173)
`ments containing the 5' end of the ,8-globin gene in the A-H,BG 1
`marker DNA (Fig. 3A). In addition, the level of hybridization
`of the hybridizing band in the genomic digests was as expected
`for a single-copy gene, based on the hybridization obtained for
`the A-H,8G1 DNA.
`Determination of Gene Dosage of fJA and fJA Allele in Hu(cid:173)
`man Genomic DNA. Genomic DNAs from patients previously
`diagnosed as normal for ,8-globin (,BA,BA) or having sickle cell
`trait (,BA ,as) or sickle cell disease (,BS ,as) were digested with
`BamHI endonuclease and subjected to electrophoresis in du(cid:173)
`plicate. For one half of the gel, the genomic DNAs with an ap(cid:173)
`propriate marker for the 5'-end of ,BA gene (A-H,8G1 digested
`with BamHI) were hybridized with H,819A' probe (,BA, perfect
`match; ,as, T/T mismatch) (Fig. 4A). The duplicate lanes were
`hybridized with H,819S probe (,8A, T /T mismatch; ,as perfect
`match) (Fig. 4B). In Fig. 4A, the 1.8-kb BamHI restriction frag(cid:173)
`ment hybridized with H,819A' for normal (,BA,BA) and sickle cell
`trait (,BA,Bs) genomic DNA. The level of hybridization to the 5'
`
`FIG. 4. Hybridization of [32p)nonadecanucleotide (H,919A' and
`H,919S) to human genomic DNA. A-HIJG1 DNA (150 pg) and duplicates
`of genomic DNAs (10 ,..g) ,sA ,sA, ,gA,g8 , and ,98,98 , were digested with
`Bamlfl and electrophoresed. The gel was dried and half was hybridized
`with [5'-32PJH,919A' (A); the other half was hybridized with [5'-
`32p)H,919S (B) at 107 cpm/ml for 2 hr at 550C. The gels were washed
`at OOC and twice, for 1 min each, in 0.9 M NaCl/0.9 M sodium citrate
`at 55°C. The gel pieces were realigned and autoradiographed. The 1.8-
`kb fragment containing the 5' end of the ,9-globin gene is indicated by
`the arrow.
`
`end of the globin gene fragment for normal DNA was noticeably
`greater than that for sickle cell trait DNA. Very little hybrid(cid:173)
`ization was seen with ~,as) DNA with the H,819A' probe (T/
`T mismatch). When the H,819S probe was hybridized in the
`duplicate lanes, the I>robe hybridized to the 1.8-kb fragment
`in the ,as ,as and ,BA(¥ DNAs but not to the ,BA,BA DNA (T/T
`mismatch) (Fig. 4B). Again, the level of hybridization with the
`H,819S probe was greater with ,as,as DNA than with ,BA,as
`DNA.
`
`DISCUSSION
`We have demonstrated in this report that synthetic oligonu(cid:173)
`cleotides recognizing a specific sequence of DNA can detect a
`single-copy gene in human genomic DNA. The nonadecanu(cid:173)
`cleotide probes can differentiate the ,8-globin gene from other
`members of the ,8-globin-like gene family and can distinguish
`the normal ,8-globin gene (,BA) from the ,as allele, a difference
`of a single nucleotide change.
`In this study, 1 bp mismatch out of 19 bp decreased the ther(cid:173)
`mal stability of the oligonucleotide-DNA duplex. Hybridization
`and wash conditions were established that would allow discrim(cid:173)
`ination of ,BA DNA from DNAs containing a single base change
`such as &.globin and ,as genes, as well as other genes in the ,8-
`globin-like gene family. The type of nucleotide mismatch(cid:173)
`e.g., A/A, T/T, etc.-could be expected to have an effect on
`the stability of the complex. However, our results showed that
`the hybridization patterns of nonadecanucleotide·DNA du(cid:173)
`plexes with aT /T bp mismatch [H,819A' with ,as DNA, H,819S
`
`GeneDX 1018, pg. 4
`
`

`

`282
`
`Medical Sciences: Conner et al
`
`Proc. Natl Acad. Sci. USA 80 (1983)
`
`with ~A DNA (Fig. 4)] were similar to those obtained with an
`A/A bp mismatch (H~l9A with~ DNA, H~l9S' with ~A
`DNA; data not shown).
`The ability of synthetic oligonucleotides to detect a point
`mutation within this sequence was tested by hybridization with
`genomic DNA from individuals normal for ~-globin (~A~A) or
`having the sickle cell allele (~) in the heterozygous or homo(cid:173)
`zygous state. The oligonucleotide probes H~l9A' and H~l9S
`hybridized to their respective genes in an allele-specific manner
`(Fig. 4). Significantly, the intensity oflabeling at the hybridized
`~globin gene fragment for each probe was proportional to gene
`dosage. The genotype of the ~globin gene could be determined
`from the hybridization pattern, confirming diagnoses made pre(cid:173)
`viously by hemoglobin typing (S. Rahbar, personal communi(cid:173)
`cation) (24). The reason for the hybridization in the upper region
`of the gel for genomic DNA samples (Figs. 3 and 4) is not known.
`However, it is not likely that this binding represents partial or
`incompletely digested fragments because of the experimental
`conditions used.
`The technique of hybridization with synthetic oligonucleo(cid:173)
`tides yields results equivalent to analysis with Mst II endonu(cid:173)
`clease (25-27). Both procedures offer direct analysis, and both
`utilize small amounts of DNA. A limitation of the Mst II analysis
`is that Mst II will not distinguish the ~c from the ~ allele be(cid:173)
`cause the~ mutation (GAG to AAG, glycine to lysine, sixth
`codon, ~globin) occurs at the N position (N = any) of the Mst
`II recognition sequence (C-C-T-N-A-G-G). Oligonucleotide
`probes, on the other hand, are specific for the ~A and ~s alleles.
`The most significant advantage this technique offers is that
`it has the potential to be applied to the diagnosis of any genetic
`disease in which a specific change in DNA sequence is involved,
`particularly in the case of base substitution but also for insertion
`or deletion not analyzable by any other methods. The sequence
`of the oligonucleotide can be designed precisely according to
`need. This eliminates dependence on restriction enzyme rec(cid:173)
`ognition site alteration, which has a low probability of occur(cid:173)
`rence for any given point mutation (23). Therefore, this tech(cid:173)
`nique can be applied not only to the diagnosis of the sickle cell
`allele~) but also to detection of the~ mutation and the single
`base changes recently reported for ~thalassemia (see ref. 28
`for review), a-thalassemia (29), and arantitrypsin deficiency
`disorders (30). These point mutations, which do not affect any
`known restriction endonuclease recognition site, could be de(cid:173)
`tected readily by an appropriate oligonucleotide probe as de(cid:173)
`scribed herein.
`
`We thank T. Maniatis for the ~globin clones, S. Rahbar for the gift
`of blood samples, and C. C. Impraim for assistance in the final stages
`of the work. R.B.W. and K.l. are members of the Cancer Research
`Center (CA16434) at the City of Hope Research Institute. The stay of
`C.M. was made possible in part by a grant from the North Atlantic
`Treaty Organization. This work was supported by National Institutes
`of Health Grant HL 29516 (R.B. W.) and Health and Welfare Agency,
`Genetic Disease Section, State of California Grant 81-77521-A1 (B.J.C.
`and R.L.T.).
`
`15.
`
`l. Wallace, R. B., Shaffer, J., Murphy, R. F., Bonner, J., Hirose,
`T. & Itakura, K. (1979) Nucleic Acids Res. 6, 3543-3557.
`2. Wallace, R. B., Johnson, M. J., Hirose, T., Miyake, T., Kawa(cid:173)
`shima, E. H. & Itakura, K. (1981) Nucleic Acids Res. 9, 87~94.
`3. Wallace, R. B., Schold, M., Johnson, M. J., Dembek, P. & Ita(cid:173)
`kura, K. (1981) Nucleic Acids Res. 9, 3647-3656.
`4. Suggs, S. V., Wallace, R. B., Hirose, T., Kawashima, E. H. &
`Itakura, K. (1981) Proc. NatL Acad. Sci. USA 18, 6613-6617.
`5. Reyes, A. A., Schold, M., Itakura, K. & Wallace, R. B. (1982)
`Proc. NatL Acad. Sci. USA 19, 3270-3274.
`6. Neel, J. V. (1978) Can.]. Genet. CytoL !0, 295-306.
`7. Maniatis, T., Fritsch, E. F., Lauer, J. & Lawn, R. M. (1980).
`Annu. Rev. Genet. 14, 145-178.
`8. Baralle, F. E., Shoulders, C. C. & Proudfoot, N. J. (1980) CeU
`21, 621-626.
`9. Slightom, J. L., Blechl, A. E. & Smithies, 0. (1980) CeU 21, 627-
`638.
`10. Spritz, R. A., DeRiel, J. K., Forget, B. G. & Weissman, S. M.
`(1980) CeU 21, 647-651.
`ll. Lawn, R. M., Efstradiatis, A., O'Connell, C. & Maniatis, T.
`(1981) CeU 21, 652-654.
`12. Pauling, L., Itano, H. A., Singer, S. J. & Wells, I. C. (1949) Sci(cid:173)
`ence no, 543-548.
`13. Marotta, C. A., Wilson, J. T., Forget, B. G. & Weissman, S. M.
`(1977)]. BioL Chem. 252, 5040-5053.
`14. Dembek, P., Miyoshi, K. & Itakura, K. (1981)]. Am. Chem. Soc.
`103, 706-708.
`Jay, E., Bambara, R., Padmanabhan, R. & Wu, R. (1974) Nucleic
`Acids Res. 1, 331-353.
`16. Fritsch, E. F., Lawn, R. M. & Maniatis, T. (1980) CeU 19, 959-
`972.
`17. Lawn, R. M., Fritsch, E. F., Parker, R. C., Black, G. & Man(cid:173)
`iatis, T. (1978) CeU 15, 1157-1174.
`18. Kan, Y. W., Dozy, A. M., Trecardin, R. & Todd, D. (1977) N.
`EngL]. Med. 297, 1081-1084.
`19. Goossens, M. & Kan, Y. W. (1981) Methods EnzymoL 16, 805-
`817.
`20. Peacock, A. C. & Dingman, C. W. (1968) Biochemistry 7, 668-
`674.
`21. Southern, E. M. (1975)]. MoL BioL 98,503-517.
`22. Shinnick, T. M., Lund, E., Smithies, 0. & Blattner, F. R. (1975)
`Nucleic Acids Res. 2, 1911-1929.
`23. Nei, M. & Li, W.-H. (1979) Proc. NatL Acad. Sci. USA 76, 5269-
`5273.
`24. Steinberg, M. H. & Adams, J. G., III (1978) South. Med.]. 71,
`413-416.
`25. Chang, J. C. & Kan, Y. W. (1982) N. EngL]. Med. 307,30-32.
`26. Orkin, S. H., Little, P. F. R., Kazazian, H. H. & Boehm, C. D.
`(1982) N. EngL]. Med. 307, 32-36.
`27. Wilson, J. T., Milner, P. F., Summer, M. E., Nallaseth, F. S.,
`Fadel, H. E., Reindollar, R. H., McDonough, P. G. & Wilson,
`L. B. (1982) Proc. NatL Acad. Sci. USA 79, 3628-3631.
`28. Orkin, S. H., Kazazian, H. H., Jr., Antonarakis, S. E., Goff, S.
`C., Boehm, C. D., Sexton, J. P., Weber, P. G. & Giardina, P.
`J. V. (1982) Nature (London) 296, 627-631.
`29. Goossens, M., Lee, K. Y., Liebhaber, S. A. & Kan, Y. W. (1982)
`Nature (London) 296, 864-865.
`30. Kurachi, K., Chandra, T., Degen, S. J. F., White, T. T., Mar(cid:173)
`chioro, T. L., Woo, S. L. C. & Davie, E. W. (1981) Proc. NatL
`Acad. Sci. USA 78, 6826-6830.
`
`GeneDX 1018, pg. 5
`
`