<<

ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 21, No. 4 Copyright © 1991, Institute for Clinical Science, Inc.

Applications of Restriction Fragment Length Polymorphism

SHESHADRI NARAYANAN, Ph .D.

Department of Pathology, New York Medical College/Metropolitan Hospital Center, New York, NY 10029

ABSTRACT The availability of a variety of restriction endonuclease enzymes that cleave deoxyribonucleic acid (DNA) at specific sites has made it possible to identify the presence of polymorphic regions in the isolated fragments. Such restriction fragment length polymorphism (RFLP) results owing to a variation in the number of tandem repeats (VNTR) of a short DNA segment. These VNTR sequences can uniquely specify an individual and, as such, are used in DNA fingerprinting and in paternity testing. Restriction frag­ ment length polymorphism may be found close to a disease gene, and, as such, can be used as a genetic disease marker. Certain criteria need to be fulfilled, however, for RFLP to be useful as a genetic disease marker, such as its closeness to the disease gene. Materials and methodology for detect­ ing RFLP are reviewed with the current emphasis on amplification proce­ dures utilizing the polymerase chain reaction (PCR).

Introduction nucleases can recognize specific DNA sequences four to six bases long. The evi­ The discovery of enzymes in bacteria dence for the existence of restriction that cleave the DNA of a foreign intruder enzymes in bacteria emerged in the is a mechanism by which bacteria pro­ 1960s and was followed up in 1970 by the tects itself from outside attack, such as isolation and purification of a restriction invasion by virus. While these enzymes, enzyme from the bacterium H aem o­ called restriction nucleases, can recog­ philus influenzae. Since then, restriction nize specific sites in foreign DNA and enzymes have been isolated from over cleave them, they are, however, pre­ 200 bacterial strains, and the specific vented from attacking the DNA of the base sequences cleaved by these bacterial cell. This is accomplished by enzymes have been identified. The vari­ another enzyme which methylates the ation in specificity of restriction enzymes specific bacterial DNA sequence that is illustrated in the example given in fig­ would otherwise have been recognized ure 1. by the , thereby pro­ While some restriction nucleases, such viding a mechanism for the restriction as Hind II isolated from the bacterium enzymes to distinguish between the bac­ Hemophilus influenzae, cleave at the terial and foreign DNA. These restriction center of their recognition sites produc- 291 0091-7370/91/0700-0291 $00.90 © Institute for Clinical Science, Inc. 292 NARAYANAN Restriction morphism is within the restriction Source (Microorganism) Enzvme Base Sequence enzyme site, short or long DNA frag­ ments are obtained after treatment with Escherichia coli Ry-13 EcoRI 51...G-1-AATTC...31 specific restriction endonucleases. These 31...CTTAa Tg ...51 restriction fragment-length polymor­ Providencia Stuartii Pstl 51...CTGCAÌG...3' phisms (RFLP) may be found close to a 31...GÎACGTC...51 disease gene, and, as such, family studies of RFLP can serve as a genetic dis­ Haemophilus Hind III 51...AÌAGCTT...31 ease marker. influenzae Rd 31...TTCGAÌ-A...51 Haemophilus Hhal 51...GCGÌC...31 Basis for Restriction Fragment Length haemoiyticus 31...CÎGCG...51 Polymorphisms (RFLP) ATCC10014 Moraxella species Mspl 51...C|CGG...31 Restriction fragment length polymor­ 31...GGCÎC...51 phism results owing to a variation in the FIGURE 1. Variation in specificity of restriction number of tandem repeats of a short DNA enzymes. Explanation of symbols. A = Adenine, T segment. This variable number of tan­ = thymine, G = Guanine, C = Cytosine. The dem repeats (VNTR) serves as the basis arrow indicates the position of cleavage by the restriction enzyme. of DNA fingerprinting.10 Thus, one indi­ vidual may have, for instance, 14 repeats of a DNA segment on one chromosome ing fragments that do not stick together, and eight repeats on the other, whereas other restriction enzymes, such as EcoRI another individual may have different isolated from Escherichia coli, cleave VNTR patterns. Tandemly repeated away from the center of its recognition DNA sequences can be very short sites producing fragments with sticky repeats of a simple sequence such as ends. The impetus for recombinant DNA (deoxy cytidine-deoxy adenine)n (deoxy technology came with the finding that guanine-deoxy thymine)n which are col­ these sticky fragments can be joined or lectively called (CA)n blocks.18 Consider­ annealed with another complementary ing that in the human genome there are sticky end produced by restriction 50,000 to 100,000 (CA)n blocks with n enzyme cleavage using an enzyme called being in the range of 15 to 30, a (CA)n DNA ligase.11 The availability of such block can be expected for every 30 to 60 restriction enzymes, also called restric­ kilobase segment of the human genome. tion endonucleases, has also resulted in It has been proposed that these blocks an explosion of research in uncovering may have a role in gene regulation polymorphic regions in DNA. or recombination. Thus, the use of restriction endonu­ clease enzymes that cleave DNA at spe­ cific sites has uncovered the presence of Usefulness and Criteria for RFLP as polymorphic regions through the length Genetic Disease Marker of the resulting fragment. In general, one in every 200 to 500 nucleotides in regions not coding for proteins is polymorphic. It The usefulness of RFLP as a genetic is known that many different individuals marker arises from the fact that RFLPs have a distinct number or sequence of are more abundant when compared to bases in a specific area of the genome. polymorphisms at the level of protein. Depending on whether or not the poly­ This is so since nucleotide diversity or APPLICATIONS OF RESTRICTION FRAGMENT LENGTH POLYMORPHISM 293 the extent of polymorphism at any locus able region (HVR).10 This hypervariable is found on an average of one in every 500 region can be characterized by using base pairs of DNA.2 multiple restriction enzymes, and the dif­ For RFLP to be useful as a genetic dis­ ferences in the lengths of the resulting ease marker, certain criteria need to be fragments can be used to establish the fulfilled. First, the RFLP has to be very presence of a specific genetic disease. close to the disease gene, since the closer Indeed, this approach is limited only by the RFLP to the disease gene, the less the ability of the electrophoretic gel to would be the likelihood of recombination resolve small differences in the sizes of to occur. Under such circumstances, the the restriction fragments. accuracy of the test as a specific disease marker would be very high. Indeed, as Materials and Typical RFLP Procedure the two loci get closer to each other, the recombination fraction, 0, approaches A variety of samples can be used for zero, making the RFLP very specific in testing of RFLP. These range from terms of its ability to screen for genetic peripheral blood, isolated lymphocytes, disease. Accuracy can, in fact, be to tissues such as cultured lymphocytes, increased further by using what are fibroblasts, or placenta. called flanking markers, which are actu­ A typical RFLP procedure would ally RFLPs very close to both sides of the involve extraction of DNA from approxi­ defective gene. The rationale for using mately 10 to 20 ml aliquots of heparin- such flanking markers is that even if ized whole blood. The extract is digested recombination occured between one overnight with a five to 10 fold excess of RFLP and the disease locus, it is highly restriction endonuclease. The resulting likely that the other RFLP would segre­ DNA fragments can be detected by a gate with the disease.3 variety of techniques which are summa­ rized in figure 2. Types of RFLP for Studying Genetic Disease Methods of Detecting RFLPs

A few genetic defects can be detected Two typical approaches merit discus­ directly by isolating an RFLP that results sion. from a difference in a single base pair. A well known example of such a single S o u t h e r n B l o t t i n g base pair mutation is the substitution of adenine (A) for thymine (T) in codon 6 of In this procedure, the DNA fragments the B-globin gene, resulting in valine that result from restriction endonuclease being substituted for glutamine, and giv­ digestion can be separated by either aga­ ing rise to sickle cell hemoglobin.8 This rose or polyacrylamide gel electrophore­ approach of detecting an RFLP that sis. The denatured DNA fragments are results from a point mutation is, however, transferred or blotted to nylon or nitrocel­ limited to only a few genetic diseases. lulose membranes. The fragments on the For the most part, RFLPs that result membrane are hybridized with radio­ owing to a variation in the number of tan­ labelled 32p complementary DNA (c- dem repeats (VNTR) at a locus are useful DNA) probe. The membrane is washed for studying genetic disease. These to remove unhybridized probe, and the VNTR DNA sequences are made up of a hybridized bands are detected by autora­ variable number of short repetitive DNA diography, which entails exposure of the sequences characterized as a hypervari­ hybridized bands to x-ray film at — 70°C. 294 NARAYANAN HEPARINIZED BLOOD I EXTRACT DEOXYRIBONUCLEIC ACID I DIGEST OVERNIGHT WITH RESTRICTION ENDONUCLEASE

I SOUTHERN BLOTTING AMPLIFY FRAGMENTS BY i POLYMERASE CHAIN REACTION SEPARATE DNA FRAGMENTS ON AGAROSE

4 BLOT OR TRANSFER TO NYLON SEPARATE AMPLIFIED FRAGMENT SPOT AMPLIFIED FRAGMENT MEMBRANES BY ELECTROPHORESIS. TO SOLID SUPPORT. VISUALIZE BANDS BY DETECTION BY HYBRIDIZATION I STAINING GEL. TO SPECIFIC PROBES. HYBRIDIZE TO RADIOLABELLED (32p) COMPLEMENTARY DNA (cDNA) PROBE 4 WASH TO REMOVE UNHYBRIDIZED PROBE 4 EXPOSE TO X-RAY FILM (AUTORADIOGRAPHY) TO VISUALIZE HYBRIDIZED BANDS F ig u r e 2. Approaches to detecting restriction fragment length polymorphism (RFLP).

The Southern blotting procedure17 is strands. The amplified fragments can be time consuming, since the autoradiogra­ separated by electrophoresis and visual­ phy step can take as long as 10 days. ized by staining the electrophoretic gel. Alternatively, the amplified fragment can be spotted on to a solid support and P o l y m e r a s e C h a i n R e a c t i o n (P C R ) detected by hybridization to specific probes. Analysis of PCR is rapid in that it The extremely high amplification can be completed in one day. power of PCR is increasingly being exploited in detecting RFLPs. In this Applications of RFLP technique,15 the fragments resulting from restriction endonuclease digestion are The power of the methodology of amplified a million to billion fold RFLP is illustrated by its use in diverse depending on the number of repetitive areas ranging from paternity testing to cycles of primer annealing, synthesis of forensic testing and to the detection of DNA in the presence of a heat stable genetic diseases. polymerase enzyme, and denaturation to Exclusion of paternity with a cumula­ separate the newly synthesized DNA tive probability greater than 99.9 percent APPLICATIONS OF RESTRICTION FRAGMENT LENGTH POLYMORPHISM 295 is possible with the use of as few as four amplification of DNA by PCR may help probes.6 Likewise, three to five probes to shed a light on why only 12 to 31 per­ can provide an individualized fingerprint cent of heavy drinkers develop cirrhosis in forensic testing.7 Sequences of VNTR of the liver.5 serve as the basis for fingerprinting. The Recently, the cystic fibrosis gene was (CA)n block markers that were referred to cloned and the defect traced to a 3-base earlier can be typed rapidly by amplifica­ pair deletion resulting in absence of phe­ tion by PCR followed by electrophoresis. nylalanine at position 508 (A F 508) in The procedure permits typing of hun­ the resulting protein. This finding makes dreds of (CA)„ block markers on a single it easier to screen for diseases where spe­ polyacrylamide gel using a small blood cific gene mutations have been identi­ sample, making it a powerful technique fied, by resorting to amplification of PCR to detect DNA polymorphisms.16 of the mutation product and subsequent Recently, DNA typing presented as identification by visualization of the evidence of the identity of a suspected amplified product on stained gel or murderer in a criminal investigation has hybridization to allele specific oligonu­ come under criticism for lack of inclusion cleotide probes.13 of internal controls in the typing study. The need for standardization of typing of Conclusion DNA is clear if the technique is to be routinely accepted by courts as evidence The exploitation of the versatility of presented as a result of forensic testing.9 RFLP in several areas ranging from In genetic disease testing, linkage paternity testing to forensic applications analysis using RFLF close to the genetic to genetic disease can be realized to its defect, such as the cystic fibrosis locus, fullest potential, with developments now permits detection.4 This is accomplished on the horizon. Specifically, automation by amplification by PCR of two different of PCR and other alternative amplifica­ regions very close to the cystic fibrosis tion based technology applied to testing locus which are detected with the use of of RFLP coupled with non-isotopic two DNA probes.4 Genetic testing, how­ probes for detection would bring testing ever, is involved, in that several family of RFLP into the realm of the clinical members have to be screened. laboratory.12 The availability of kits Polymerase chain reaction has been adapted to analysis of RFLP should fur­ used to study genetic diseases such as ther facilitate expansion of the scope of hemophilia A and B, Duchenne and testing of RFLP in the clinical laboratory. Becker muscular dystrophy and a-1-antitrypsin deficiency.14 References The use of PCR to study specific sequences of VNTR in the 3'-portion of 1. B o e r w i n k l e , E., X i o n g , W., F o u r e s t , E., and the apolipoprotein gene that led to the C h a n , L.: Rapid typing of tandemly repeated hypervariable loci by the polymerase chain demonstration of 12 different alleles at reaction: Application to apolipoprotein B 3' that locus within a single racial group is hypervariable region. Proc. Natl. Acad. Sei. an approach that is being extended to the USA 86:212-216, 1989. 2. C h a k r a v a r t i , A., E l b e i n , S. C ., an d P e r - study of other gene defects, such as the MUTT, M. A.: Evidence for increased recombi­ insulin gene in diabetics.1,2 nation near the human insulin gene. Implica­ Studies of RFLP at the alcohol dehy­ tion for disease association studies. Proc. Natl. Acad. Sci. USA 83:1045-1049, 1986. drogenase (ADH) loci (ADH 2 and ADH 3. D a w s o n , D . B.: Use of nucleic acid probes in 3) using a microsample of blood and genetic tests. Clin. Biochem. 23:279—285, 1990. 296 NARAYANAN

4. G a s p a r i n i , P ., N o v e l l i , G ., S a v o i a , A., D a l ­ 12. N a r a y a n a n , S.: Current status of D N A probe l a p i c c o l a , B ., a n d P i g n a t t i , P. R.: First- utilization in the clinical laboratory: A discus­ trimester prenatal diagnosis of cystic fibrosis sion of cost effectiveness and outlook for the using the polymerase chain reaction: Reports of future. Bull. Mol. Biol. Med. 14:157—166, 1989.

eight cases. Prenatal Diagnos. 9:349-355, 1989. 13. P r i o r , T. W ., H i g h s m i t h , E . W . Jr ., F r i e d ­ 5. G r o p p i , A., B e g u e r e t , J., a n d I r o n , A.: m a n , K. J., P e r r y , T. R ., S cheuerbrandt , G ., Improved methods for genotype determination and S il v e r m a n , L. K.: A m odel for m olecular of human alcohol dehydrogenase (ADH) at screening of new boms: Simultaneous detec­ ADH 2 and ADH 3 loci by using polymerase tion of Duchenne/Becker muscular dystro­ chain reaction—directed mutagenesis. Clin. phies and cystic fibrosis. Clin. Chem. 36:1756- Chem. 36:1765-1768, 1990. 1759, 1990. 6. J e f f r y s , A. J., B r o o k f i e l d , J. F. Y., and 14. R e i s s , J. and C o o p e r , D. N.: Application of the S e m e ONOFF, R.: Positive identification of an immigration test-case using human DNA fin­ polymerase chain reaction to the diagnosis of gerprints. Nature 3J 7:818—819, 1985. human genetic disease. Human Genet. 85:1-8, 1990. 7. J e f f r y s , A. J ., W i l s o n , V., T h e i n , S. L., W e a t h e r a l l , D. J. and P o n d e r , B. H . J.: 15. S a i k i , R. K., G e l f a n d , D . H ., S t o f f e l , S ., DNA “fingerprints” and segregation analysis of S c h a r f , S. J., H i g u c h i , R ., H o r n , G . T., M u l - multiple markers in human pedigrees. Am. J. LIS, K. B., and ERLICH, H. A.: Primer-directed Hum. Genet. 39:11-14, 1986. enzymatic amplification of DNA with a thermo­ 8. Ka n , Y. W. an d D o z y , A. M.: Polym orphism of stable DNA polymerase. Science 239:487—491, DNA sequence adjacent to human beta-globin 1988.

structural gene: Relationship to sickle mutation. 16. S k o l n ic k , M. H. and W a l l a c e , R. S. Sim ulta­ Proc. Natl. Acad. Sci. USA 75:5631-5635, 1978. neous analysis of multiple polymorphic loci 9. LEWIN, R.: DNA typing on the witness stand. using amplified sequence polymorphisms. Science 244:1033-1035, 1989. Genomics 2:273—279, 1988. 10. N a k a m u r a , Y., L e p p e r t , M ., O ’C o n n e l l , P ., 17. S o u t h e r n , E . M.: Detection of specific W o l f f , R., H o l m , T., C u l v e r , M ., M a r t i n , sequences among DNA fragments separated by C ., F u j i m o t o , E., H o f f , M ., Ku m l i n , E., and gel electrophoresis. J. Mol. Biol. 98:503—517, W HITE, R.: Variable number of tandem repeat (VNTR) markers for human gene mapping. Sci­ 1975. en ce 235:1616-1622, 1987. 18. W e b e r , J. L. and M a y , P. E .: Abundant class of 11. N a r a y a n a n , S.: DNA probes. In: P rinciples human DNA polymorphisms which can be and Applications of Laboratory Instrumenta­ typed using the polymerase chain reaction. Am. tion. Chicago, ASCP Press, 1989, pp. 185-191. J. Hum. Genet. 44:388-396, 1989.