Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (46)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Liu, Q.
Right arrow Articles by Sommer, S. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, Q.
Right arrow Articles by Sommer, S. S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics Pages 107-114
Bi-directional dideoxy fingerprinting (Bi-ddF): a rapid method for quantitative detection of mutations in genomic regions of 300-600 bp
Introduction
Results
   The dideoxy and SSCP components of Bi-ddF
   The Bi-ddF pattern is a simple combination of the patterns produced by the downstream and upstream primers
   Bi-ddF of two regions of the factor IX gene
   Blinded analysis
Discussion
   Bi-directionality of the fingerprint enhances both the SSCP and dideoxy components
   Potential for detection of heterozygotes
   Technical tips
Materials And Methods
   Oligonucleotides
   Mutations
   PCR and purification
   Bi-directional cycle sequencing reaction with Taq polymerase
   Electrophoresis
   Gel analysis
Acknowledgements
References

Bi-directional dideoxy fingerprinting (Bi-ddF): a rapid method for quantitative detection of mutations in genomic regions of 300-600 bp

Bi-directional dideoxy fingerprinting (Bi-ddF): a rapid method for quantitative detection of mutations in genomic regions of 300 -600 bp Qiang Liu, Jinong Feng and Steve S. Sommer*

Department of Biochemistry and Molecular Biology, Mayo Clinic/Foundation, Rochester, MN 55905, USA

Received September 1, 1995; Revised and Accepted October 23, 1995

There is a great need for rapid screening methods that detect essentially all mutations. Dideoxy fingerprinting (ddF) is a highly sensitive screening method that is performed by electrophoresing one lane of a Sanger dideoxy termination reaction through a nondenaturing gel. Mutations may produce an extra segment or eliminate a segment from the termination products (informative dideoxy component). In addition, mutations can be detected by the altered mobility of one or more termination segments (informative SSCP component). To screen larger segments with virtually 100% sensitivity, bi-directional ddF (Bi-ddF) was developed. Bi-ddF is a `second generation ddF' in which the dideoxy termination reaction is performed simultaneously with two opposing primers. Bi-ddF has two important advantages over ddF: (i) the dideoxy component can detect 10 of the 12 types of possible single-base substitutions; and (ii) the SSCP component is enhanced because alterations of mobility can be detected in either the downstream or upstream direction. As a result, Bi-ddF can screen larger regions of genomic DNA with virtually 100% sensitivity. Bi-ddF detected 100% of 28 single-base substitutions in a 494 bp segment containing exons B and C of the human factor IX gene and 100% of 42 single-base substitutions and one microdeletion present in a 577 bp region containing exon H. In a blinded analysis in which 39 wildtype samples were randomly mixed with 51 mutant samples, all mutations were detected with no false positives. Bi-ddF requires essentially the same effort as ddF, yet twofold more DNA sequence can be screened reliably per unit effort.

INTRODUCTION

Single-stranded conformation polymorphism (SSCP) (1 ) is a widely used DNA screening method in which single-base sequence changes can be detected by abnormal electrophoretic migration of one or both single strands on a nondenaturing gel. SSCP does not detect all sequence changes and its sensitivity is a complex function of sequence and size (2 -10 ). RNA single-stranded conformation polymorphism (rSSCP) may well have a higher sensitivity, but mutations may still be missed (2 ,11 ). The sensitivity of SSCP is highly variable (typically 60-95% for segments of 200 bp). It is a function of both the distribution of mobility differences due to single-base changes and the position of the wildtype sequence mobility within the distribution (Fig. 1 ).Thus, substantial redundancy is required to consistently detect virtually 100% of mutations with the SSCP phenomenon.


Figure 1. Schematic of how the sensitivity of SSCP can vary dramatically as a function of sequence. For a given segment, the sensitivity depends on the distribution of mobilities for segments with single-base changes and the location of the normal sequence within that mobility distribution. For a 200 bp segment, there are 600 possible variants that differ by a single-base substitution. If it were possible to generate all 600 possible variants and to plot the mobility in units of band widths, it is apparent that the sensitivity of SSCP will be less for a segment in which the mobility of the wildtype sequence is close to the mode (A). For the segment electrophoresed under a second set of conditions, or a region with a different sequence electrophoresed under the same set of conditions, the variance of the distribution may differ. If the variance of mobility is wider (B), SSCP sensitivity, on average, will be higher than with the first set of conditions. However, this is not necessarily so, because the location of the wildtype sequence within the distribution is also critical (see hypothetical wildtype sequence B in comparison to wildtype sequence A). For a hypothetical 200 bp segment, the mobility distribution of the 600 possible single bp substitutions are plotted (see Introduction). A Gaussian distribution of mobilities is shown for the sake of illustration, but other distributions could be substituted without changing the argument. The sensitivity of SSCP depends on the mobility of the wildtype sequence in comparison with the distribution of mobility of all possible mutated sequences. (A) The X-axis represents segment mobility in units of band thicknesses from the average of mobilities. The mobility of two segments can be distinguished if they differ by more than one-half band thickness. The Y-axis shows the relative frequencies of segments at a given mobility. The mobility of two hypothetical wildtype sequences are indicated by A and B respectively. (B) Electrophoresis of the same samples under a different set of hypothesized conditions in which the distribution is broader; in general the hypothetical wildtype sequences will be located at different positions within the distribution (A' and B').

ddF is a hybrid technique that detects the presence of single-base changes in a quantitative manner. In ddF, a Sanger dideoxy termination reaction is performed with only one of the dideoxy terminators, denatured, quick-chilled and electrophoresed through a nondenaturing gel. Mutations can be detected as a result of the gain or loss of a dideoxy termination segment (informative dideoxy component) and/or by an alteration in the mobility of at least one of the termination segments that contain the mutation (informative SSCP component). ddF can detect virtually all mutations in a 250-300 bp region of DNA. For ~25% of mutations, an extra dideoxy termination segment is created by the mutation and, for another 25% of the mutations, a termination segment is lost as a result of the mutation. For the SSCP component, the number of potential segments with altered mobilities can vary from a few to >50 depending on the location of the mutation. The redundancy of ddF facilitates detection of virtually all mutations, because one segment with altered migration is sufficient to detect the presence of the mutation.

Initial experiments with ddF showed that 84 of 84 different sequence changes were detected in the human factor IX gene (3 ). Thirty-six different mutations were detected in a blinded analysis in which the regions of likely significance of the factor IX gene (2.2 kb) were analyzed by ddF and compared to direct sequencing. Although almost 100% sensitivity can be achieved with polyacrylamide gels with glycerol, further analysis of mutations in the factor IX gene indicates that the use of MDEtm and GeneAmptm in place of polyacrylamide and/or electrophoresis at 8oC increase the ease with which mutations are detected by increasing the number of informative segments in the SSCP component (10 ). In a subsequent comparison of ddF and direct sequencing, all of 25 unique mutations were detected by ddF in a blinded analysis of the p53 gene in 73 primary breast cancers (12 ). ddF also has been applied to the genotypic detection of multi-drug resistant Mycobacterium tuberculosis (13 ). Most of these strains are resistant to rifampin and more than 95% of M.tuberculosis isolates exhibiting resistance to rifampin have mutations within a 69 bp segment of the [beta] subunit of RNA polymerase. The fingerprint produced by ddF is particularly helpful because each new mutation should produce a novel fingerprint. In addition, the common functional mutations present in a given region can be recognized without sequencing and distinguished readily from common neutral variants in the population.

Restriction endonuclease fingerprinting (REF) is another modification of SSCP that was developed to detect the presence of essentially all mutations in a 1-2 kb segment (14 ; Sommer et al., manuscript in preparation). REF typically involves five or six different restriction endonuclease digestions so that the mutant bases are contained within multiple different fragments. Thus, there are multiple chances to detect a mutation by alterations of electrophoretic mobility. REF initially was tested in a 1 kb segment of the human factor IX gene. The region was amplified with PCR, digested with each of five groups of restriction endonucleases, mixed, 5' end-labeled with T4 polynucleotide kinase, denatured and electrophoresed in one lane under nondenaturing conditions. Point mutations resulted in the gain or loss of a restriction site in 29% of 24 test mutations (informative restriction component). Mutations also were detected if any of the mutation containing restriction fragments (producing 10 single-stranded segments) displayed abnormal mobility (SSCP component). Blinded analyses indicate that REF detects virtually 100% of mutations in 1-2 kb segments (14 ; Sommer et al., manuscript in preparation).

Restriction endonuclease fingerprinting is ~2-fold more labor intensive per sample screened than ddF. Thus, REF is the more efficient method per basepair screened when 1-2 kb segments need to be screened with virtually 100% sensitivity, while ddF is more efficient for 250-300 base segments. However, neither method is unequivocally superior if regions from 300 to 600 bp require screening. Since exons and flanking splice junctions are often in this size range, it would be useful to have a method in which 600 bp could be screened in one lane with an effort equivalent to that of ddF. Herein we describe a modification of ddF with this property. The method is called Bi-directional ddF (Bi-ddF) because a Sanger dideoxy termination reaction is performed simultaneously with an upstream and downstream primer by utilizing a modified cycle sequencing protocol. This seemingly simple modification has some complex and beneficial consequences (see below). Bi-ddF is applied to the detection of 71 mutations in 494 and 577 bp segments of the human factor IX gene.

RESULTS

The dideoxy and SSCP components of Bi-ddF

Bi-ddF is simple to perform and to analyze. Figure 2 illustrates the principle of Bi-ddF and the multiple components that generate the sensitivity of the method. The mutation can be detected if 1 of 60+ segments within the fingerprint do not match precisely to the fingerprint of the wildtype sequence. Dideoxy and SSCP components from both strands contribute to the sensitivity of Bi-ddF. In Bi-ddF, termination segments can be gained or lost in either the sense or antisense direction (dideoxy component). If ddGTP is used in the termination reaction, a termination segment is gained when A, T or C in the wildtype sequence is mutated to G and a segment is lost if G is mutated to A, T or C. If the Sanger termination reactions in the sense and antisense directions are considered, ~50% of the mutations are associated with the gain of a termination segment and/or 50% with the loss of a termination segment (Table 1 ). A G -> C or C -> G transversion will be associated with both the gain of a termination segment at one site and the loss of termination segment at another site, while an A -> T or T -> A transversion will not have an informative dideoxy component. Thus, in the highly informative `window' of 60-400 bp (see Fig. 2 ) in which the dideoxy component displays maximal sensitivity, 83.5% (10 of 12) of the possible types of single-base substitutions will have an informative dideoxy component (Table 1 ). Since the mutation rate for transversions is 2.5-fold lower than for transitions (15 ), the dideoxy component within the central region of the gel is expected to be informative for >90% of the mutations in the mammalian genome. In regions of the gel in which termination segments from one direction are >400 bp, some resolution is lost, because the segments migrate too closely together to detect a fraction of the dideoxy component.

Table 1 . Predicted dideoxy component sensitivity for Bi-ddF with ddGTP: 10 of the 12 possible single-base substitutions can be detected
Wildtypea

 Mutationb
 

 -> A

 -> T

 -> G

 -> C
A ->

X

 0/0c

 +/0

 0/+
T ->

0/0c

 X

 +/0

 0/+
G ->

-/0

 -/0

 X

 -/+
C ->

0/-

 0/-

 +/-

 X
aBases in the wildtype sequence in the downstream direction are represented by `A -> ', `T -> ', `G -> ' and `C -> '.bPotential single-base changes in the downstream strand are represented by ` -> A', ` -> T', ` -> G' and ` -> C'.cIn Bi-ddF, the Sanger dideoxy termination reaction can add or eliminate the termination segment in either the downstream or the upstream direction. The dideoxy component in the downstream direction is listed first followed by the dideoxy component in the upstream direction. `+' signifies an extra termination segment and `-' signifies an absent termination segment and `0' signifies no change in the termination segment. Most of the possible single-base changes are associated with a + or - dideoxy component in either the downstream or upstream direction. With ddGTP as the terminator, G -> C and C -> G transversions are associated with an informative dideoxy component in both the sense and antisense direction. These two changes are seen in different positions in the gel.

The SSCP component provides a second way of detecting the presence of mutations. The SSCP component is composed of all the termination segments subsequent to the mutation. In Bi-ddF, the SSCP component can be informative for the downstream and/or the upstream Sanger dideoxy termination reactions (Fig. 2 ). If one or more segment displays an altered mobility, the SSCP component is considered informative. Bi-ddF can screen substantially longer segments of genomic DNA than ddF with virtually 100% sensitivity because the SSCP sensitivities from the downstream and upstream components complement one another (Fig. 2 ).

The Bi-ddF pattern is a simple combination of the patterns produced by the downstream and upstream primers

A 516 bp genomic segment was analyzed with Bi-ddF (Materials and Methods). Primers A and B were used to amplify a 1 kb segment and then one Sanger dideoxy termination reaction was performed with three combinations of primers: (i) in the downstream direction with only primer G; (ii) in the upstream direction with only primer I; or (iii) in both directions with primers G and I (Materials and Methods). The Bi-ddF pattern was found to be a simple combination of the upstream and the downstream patterns (data not shown). The same was true when the bi-directional Sanger reaction was performed with primers G/H and E/F (which produce 577 and 494 nucleotide segments respectively).


Figure 2. Schematic of Bi-ddF. (A) The Bi-ddF reactions. A region of genomic DNA is amplified by PCR with primers P1 and P2. Subsequently, a bi-directional Sanger dideoxy termination reaction is performed with nested primers (D and U). (B) Bi-ddF detects mutations in multiple overlapping ways. The downstream and upstream dideoxy and SSCP components of Bi-ddF are diagrammed. To improve separation at the upper part of the gel, products of the Sanger termination reaction are electrophoresed through a nondenaturing gel such that segments <60 bp are electrophoresed off of the gel. The downstream and upstream components of the bi-directional termination reaction are identified on the gel by comparison with control reactions in which only one of the primers used for the termination reaction are labeled. In this example, there is one mutation (X1) in the center of the region and another mutation (X2) at one end of the region. The dashed lines are small segments that are electrophoresed off the gel. Mutant X2 first appeared in such a segment in the downstream direction, so any informative dideoxy component is missed. However, all the segments above 60 bases contain the mutation so the SSCP component in the downstream direction is expected to be informative almost all of the time. The solid lines are segments within the highly informative window. The dotted lines represent segments of >400 bases which have low resolution. The total Bi-ddF dideoxy and SSCP components depend on the sum of the dideoxy and SSCP components from the downstream and upstream termination reaction. The schematic emphasizes how the dideoxy and SSCP components contribute to the sensitivity of Bi-ddF. The efficiency of the components (see Table 2 for the definition) are of technical and theoretical interest but, for practical purposes, sensitivity is the key measure since it reflects the percentage of mutations that are detected by the method. (C)Note that a control unidirectional ddF reaction allows the downstream and upstream components to be distinguished, thereby allowing mutations to be localized, typically within 30 nucleotides. Schematic of an autoradiograph resulting from Bi-ddF of the two hypothetical mutations shown in part B. The arrows indicate which of the segments derive from the downstream primer and which derive from the upstream primer. The X indicates the presence of a mutation within the segment. The highly informative window of 60-400 nucleotides is indicated. In this schematic, the centrally located lane contains a wildtype control (C) while mutations X1 and X2 from part B were in lanes 1 and 3 respectively. In mutation X1, the dideoxy component is informative in the downstream but not the upstream direction. This occurs for G -> A or T -> C transitions if ddGTP is used (see Table 1). The SSCP component is also informative in both directions. For mutation X2, the dideoxy component cannot be detected in the downstream direction because any gained or lost segment would be less than 60 bases and therefore electrophoresed out of the gel. For mutation X2, the SSCP component is informative in one direction.

Bi-ddF of two regions of the factor IX gene

Segments of 494 and 577 bp containing exons B/C and H of the factor IX gene were screened by Bi-ddF (Fig. 3 A,B). Twenty-one single-base substitutions were present in each of the regions (Materials and Methods). For exons B/C and H, the dideoxy component (within the segments of 60-400 bases) of Bi-ddF, was informative in 86 and 67% of the samples, respectively (Table 2 ). All the expected changes in the dideoxy component were detected except in two cases. In exon B/C, a mutation (#18) expected to eliminate a segment was missed because the 190 base sequence co-migrated with two additional segments. In exon H, a mutation (#5) expected to eliminate a segment was not detected because the 151 base segment co-migrated with a neighboring segment; in retrospect the absent segment could have been detected as a clearly discernable 2-fold decrease in the intensity of this segment relative to the termination segments just above and below.


Figure 3 (A) Autoradiogram of Bi-ddF of exons B/C. A 494 bp region was screened. The four segments near the top of the gel are utilized to determine the `run-off' SSCP component (see Table 3). Lane D: wildtype control in which only the downstream primer was labeled; lane U: wildtype control DNA in which only the upstream primer was labeled; lane C: wildtype DNA in which both upstream and downstream primers were labeled. Lanes 1-21: 21 samples from patients with mutations in exons B/C of the factor IX gene. Mutation numbers correspond to those listed below. The segments due to the termination reaction in the downstream direction versus the upstream direction are identified by the downstream (D) and the upstream (U) controls. For reference, the size of certain termination segments are given along the side of the gel. The sizes of the segments can be determined precisely because a mutation at that site resulted in the absence of the segment. The downstream segments are given from bases 55-407. The upstream segments are numbered from bases 52-397. It should be noted that an artifact of cycle sequencing had a minor detrimental affect. A control sequencing reaction electrophoresed on a denaturing gel indicated that certain expected termination segments were either very faint or not observed, that is in the downstream direction at bases 133, 187 and 211 and in the upstream direction at bases 398 and 407. (B) Bi-ddF of 21 mutations in exon H. Bi-ddF was performed with primers E and F as described in Materials and Methods. A 577 bp region was screened.

All the mutations in exon B/C and exons H had an informative SSCP component in at least one direction. Table 2 summarizes the data, stressing the comparison of uni-directional ddF and Bi-ddF. The dideoxy, SSCP and ddF sensitivities and the SSCP efficiency can be calculated for uni-directional ddF because each gel included a control in which only the termination segments due to one or the other primer was labeled (see Table 2 footnotes for definitions of the terms sensitivity and efficiency). This allows the Bi-ddF pattern for each mutation to be broken down into the downstream uni-directional pattern and the upstream uni-directional pattern. The data illustrate the idiosyncratic nature of the SSCP phenomenon. The SSCP component efficiencies vary substantially with the region and with the sense and antisense strands. For example, for Bi-ddF, only 53% of the 834 segments potentially informative for the SSCP component displayed altered mobility in exons B/C, while 77% of 676 potentially informative segments displayed altered mobilities in exon H.

All types of single-base substitutions are represented in the analyzed sample. No simple pattern was discerned when the average SSCP efficiency was analyzed by the type of mutation (data not shown). There were three instances in which there were two different mutations at one base. In these instances, the SSCP efficiencies varied substantially, highlighting the idiosyncratic nature of mobility changes.

Blinded analysis

To test the sensitivity of Bi-ddF under actual screening conditions, blinded analyses were performed. In the first analysis, 48 coded genomic samples, including 33 mutant alleles, were analyzed in exon H. Twenty-four of the 33 mutations were different from those analyzed previously. All 33 mutations were detected and there were no false positives when previously published rules were observed (10 ,12 ). A similar blinded analysis was done for exons B/C in which all 26 mutations were detected from 42 analyzed samples.

DISCUSSION

Bi-directionality of the fingerprint enhances both the SSCP and dideoxy components

Bi-ddF is a modification of ddF in which non-blinded and blinded analyses suggest that virtually 100% of mutations can be detected in segments of 600 bp. Bi-ddF is performed by modifying the cycle sequencing protocol to perform a downstream and an upstream ddF simultaneously. Bi-ddF involves the same effort as ddF. Both methods can detect the presence of a mutation if only one segment in the SSCP component displays altered mobility. With Bi-ddF, mutations can be detected reliably in a larger region because a mutation that may occur in large suboptimally resolved segments in one direction appears in small, optimally resolved segments in the opposite direction.

Mutations also can be detected by an abnormal dideoxy component which is more highly informative than in standard ddF. In Bi-ddF, the dideoxy component is informative if the dideoxy component of either the upstream or downstream primers are informative. In the 42 samples examined (see Fig. 3 ), the dideoxy component of Bi-ddF is predicted to be informative in 79% of the mutations within the highly informative window of 60-400 bases. Informative dideoxy components were observed for all these mutations except for two cases mentioned above. Since additional mutations had informative dideoxy components which could be detected outside of the 60-400 base window, an informative dideoxy component can be seen in >79% of the mutations.

Potential for detection of heterozygotes

The mutations analyzed in this study are in hemizygous males with the X-linked disease hemophilia B. However, previous data with standard ddF (12 ) strongly suggest that Bi-ddF will detect virtually all heterozygous mutations. The only component of mutation detection which is substantially compromised in heterozygotes is the loss of a segment in the dideoxy component. None of the mutations described herein depend critically on the loss of a segment in the dideoxy component for their detection. Thus, it is expected that all the mutations would have been detected in the heterozygous form. However, it will be necessary to test this hypothesis in the future.

Technical tips

Each of the above methods can produce excellent results in experienced hands. As with all methods, small procedural differences may make the difference between an optimal and suboptimal result. We have published helpful hints for achieving optimal results with ddF (10 ,12 ). These hints should be directly applicable to Bi-ddF. The following are additional hints which relate to the bi-directional Sanger dideoxy termination reaction:

(i) Occasionally, several extra segments appear because suboptimal extension reactions give shadow bands that can mimic altered patterns due to mutation. These can be recognized by determining that the intensity of the signal fades out as the segments become larger. This pattern is due to poor termination reactions producing multiple nonspecific termination products; these samples should be repeated and not scored as positive.

Table 2 . Sensitivity and efficiency of Bi-ddF for exons B/C and Ha
 

 Components

  

  
 

 Dideoxyb

 SSCPc

 `Run-off' SSCPd
Exons B/C

Downstream sensitivity

 13/21 (62%)

 15/21 (71%)

 5/21 (24%)
Upstream sensitivity

 7/21 (33%)

 8/21 (38%)

 1/21 (5%)
Bi-ddF sensitivity

 18/21 (86%)

 21/21 (100%)

 6/21 (29%)
Downstream efficiency

 -

 376/631 (60%)

 6/42 (14%)
Upstream efficiency

 -

 70/203 (34%)

 1/42 (2%)
Bi-ddF efficiency

 -

 446/834 (53%)

 7/84 (8%)
Exon H

  

  

  
Downstream sensitivity

 11/21 (52%)

 17/21 (81%)

 11/21 (52%)
Upstream sensitivity

 7/21 (33%)

 13/21 (62%)

 12/21/ (57%)
Bi-ddF sensitivity

 14/21 (67%)

 21/21 (100%)

 14/21 (67%)
Downstream efficiency

 -

 273/339 (81%)

 16/42 (38%)
Upstream efficiency

 -

 246/337 (73%)

 19/42 (45%)
Bi-ddF efficiency

 -

 519/676 (77%)

 35/84 (42%)
aThese data are compiled from 1510 mutation-containing segments from a total of 42 mutations. Sensitivity and efficiency are discussed further in subsequent footnotes. The sensitivity is defined as the ability of a particular component to detect the mutation. For the SSCP component, there will be many mutant segments that are potentially informative but only one must show altered mobility for the mutation to be detected. The efficiency of the SSCP component is the fraction of all mutant segments that actually do show mobility changes. The sensitivity and efficiency of the dideoxy and SSCP components are calculated for the highly informative window of termination segments 60-400 bases in length (see Fig. 2B). In contrast, the sensitivity and efficiency of the `run-off' SSCP component is determined from analysis of the large and intense segments produced by the Bi-ddF reaction (see footnote d). bThe dideoxy component sensitivities are underestimates because an informative dideoxy component is occasionally seen for segments larger than the 400 base maximum of the highly informative window. The Bi-ddF sensitivity is less than the sum of the downstream and upstream sensitivities because G -> C or C -> G transversions can produce an informative dideoxy component in both directions (see Table 1).cThe sensitivity of the SSCP component of Bi-ddF is 100% for both regions, but it is instructive to look at the downstream and upstream termination reactions individually. Despite the shorter length of the region encompassing exons B/C (494 bases versus 577 bases), the total Bi-ddF efficiency and the downstream and upstream efficiencies were substantially lower than for exon H. The total sensitivity of the SSCP component of Bi-ddF is 100% in both regions because a mutation that appears only in larger segments in one direction appears in smaller, generally more informative segments in the opposite direction and a mutation which has a lower efficiency in one direction may well have a higher efficiency in the other direction (see Fig. 2B). dThe run-off SSCP component is composed of four intensely staining segments at or near the top of the gels. These correspond to `run-off' of Taq polymerase extension products that did not terminate with the available dideoxy nucleotide (in these experiments, dideoxy G). The shorter two segments correspond to the downstream and upstream strand of the amplification product from primer D to primer U (Fig. 2) which results from the 20 or 30 rounds of cycle sequencing. These segments are separated by the SSCP effect. The larger two bands correspond to a segment derived from annealing of the Bi-ddF primers to the input template produced by amplification of genomic DNA with primers P1 and P2 (i.e. segments of length corresponding to the segment extending from primer D to primer P2 and the segment extending from primer U to primer P1. The number of downstream and upstream extension products with altered mobility are indicated. For exons B/C, the sizes of the `run-off' segments are: 494 and 607 bases in the downstream direction and 494 and 671 bases in the upstream direction. For exon H, the sizes of these segments are: 577 and 795 bases in the downstream direction and 577 and 782 bases in the upstream direction. Although these segments are large, they sometimes display altered mobilities on MDE gels.

(ii) The relative ratio of the downstream and upstream primer concentrations often needs to be adjusted so that the two directions in Bi-ddF are of similar intensity. There seems to be an approximately linear relationship between the primer concentration and signal intensity.

(iii) The ratio of ddGTP to dGTP can affect the relative intensities of segments seen in the lower and upper parts of the gel. A molecular ratio of 5:1 was utilized for the 577 bp region while a ratio of 6:1 was utilized for the 494 bp region.

(iv) Although good results can be achieved with [32P] radiolabeling, [33P] reduces the thickness of segments which is of particular benefit for Bi-ddF since these are more closely spaced segments in comparison to ddF.

MATERIALS AND METHODS

Oligonucleotides

All the oligonucleotides are specific for the human factor IX gene. The abbreviated informative names (see ref. 16 ) are listed below: A. I7(30646)-34D; B. E8(31645)-31U; C. I1(6094)-30D; D. I3(6878)-27U; E. I1(6272)-22D; F. I3(6764)-22U; G. I7(30851)-19D; H. E8(31429)-18U; I. E8-(31366)-20U.

As an example of the nomenclature, I7(30646)-34D describes an oligonucleotide where the 5' end begins in intron 7 at bp 30646 (numbering as described in ref. 17 ). The length of the oligonucleotide is 34 bases and the orientation of the oligonucleotide is `downstream' (D) (i.e. in the direction of transcription). The precise sizes and locations of the amplified segments and dideoxy termination reactions can be obtained from the informative names.

Mutations

For the present analysis, genomic DNA with the following mutations in the human factor IX gene were analyzed. The sample numbers and associated mutations correspond to the lane numbers for Figure 2 A. 1: C6364T; 2: G6365A; 3: G6365T; 4: G6374A; 5: G6375T; 6: G6376C; 7: A6398G; 8: T6401C; 9: G6428A; 10: T6442C; 11: G6451A; 12: C6460T; 13: G6461A; 14: G6461C; 15: G6463A; 16: G6463C; 17: C6488T; 18: C6575G; 19: G6677C; 20: A6690T; 21: T6696G.

The following are the samples utilized in Figure 3 B: 1: A30897G; 2: A30918C; 3: T30936G; 4: C30973A; 5: G31001T; 6: G31029A; 7: G31051C; 8: C31077A; 9: C31096A; 10: C31118G; 11: C31140G; 12: T31166A; 13: G31187T; 14: G31203C; 15: G31211T; 16: A31227G; 17: T31274A; 18: A31301G; 19a: A30972G; 19b: C31328T; 20: T31340C; 21: del31355-57.

PCR and purification

PCR was performed from human genomic DNA with the Perkin Elmer model 9600 thermal cycler. The following cycling parameters were used: 94oC for 1 min, 55oC for 2 min and 72oC for 3 min for 30 cycles followed by incubation at 72oC for 10 min. The PCR mixture contained a total volume of 50 [mu]l: 50 mM KCl, 10 mM Tris/HCl pH 8.3, 1.5 mM MgCl2, 200 [mu]M of each dNTP, 0.1 [mu]M of each primer, 1 U of Amplitaq (Perkin Elmer) and 500 ng of genomic DNA. PCR primers A and B amplify a 785 bp region which includes the exons B/C and primers C and D amplify a 1 kb region which includes the coding region of exon H. The PCR product was electrophoresed through 2.5% agarose to confirm that the expected band was the sole product.

The amplified sample was diluted to 2 ml with TE buffer (10 mM Tris/HCl, 0.5 mM EDTA pH 8.0) and then purified by ultrafiltration with a Centricon®-100 microconcentrator (Amicon) and centrifuged at 1000 g for 20 min. The amount of recovered DNA was determined by absorbance at 260 nm and then diluted to 10 ng/[mu]l.

Bi-directional cycle sequencing reaction with Taq polymerase

Bi-directional Taq cycle sequencing was performed by a modification of the protocol of Innis et al. (18 ). ddGTP was chosen as the terminator since it has the highest termination efficiency. For Bi-ddF, the Sanger termination reaction cycling parameters were 95oC for 15 s, 55oC for 30 s and 72oC for 1 min for a total of 30 cycles for primers E and F to screen a 494 bp region in the exons B/C, while the annealing was at 50oC for primers G and H for 20 cycles to screen a 577 bp region in exon H. The cycle sequencing mixture contained a total volume of 8 [mu]l: 50 mM KCl, 50 mM Tris/HCl pH 8.8, 2.5 mM MgCl2, 20 [mu]M of each dNTP, 100-120 [mu]M of ddGTP, 0.05 [mu]M of primer, 0.4 U of Amplitaq and 5-10 ng of purified PCR amplified DNA. After Taq cycling sequencing, 50-60 [mu]l of stop/loading buffer (7 M urea, 50% formamide, 2 mM EDTA) was added to each tube.

Uni-directional ddF was performed for comparison with Bi-ddF. Cycling parameters were 95oC for 15 s, 50oC for 30 s and 72oC for 1 min for a total of 30 cycles with primers E or I. The termination reaction was diluted with the stop buffer in a volume ratio of 1:2. For Bi-directional ddF with primers E and I, 20 cycles were performed followed by 10-fold dilution with stop buffer. The signal intensity was increased by 5-fold to 10-fold when both primers were utilized in the same reaction (Bi-ddF). The enhanced efficiency of bi-directional cycle sequencing may be due to the exponential generation of additional templates during the cycling reactions and/or inhibition of template reannealing by primer extension.

Electrophoresis

Electrophoresis of gels (45 cm * 37.5 cm * 0.35 mm) produced by a 4-fold dilution of MDEtm gel stock solution was performed using a Poker Face SE 1500 sequencing apparatus with 50 mM Tris-borate and 1 mM EDTA (pH 8.3) at 12 W constant power, at room temperature. Circulating tap water was used so that the temperature of the glass plate was 20oC when measured by a liquid crystal thermometer. After an initial electrophoresis for 30 min, samples of 1.5 [mu]l were loaded and electrophoresed for 6 h. The samples were electrophoresed longer than with standard ddF so that a 60 base segment migrates at the bottom of the gel. The gel was dried and subjected to autoradiography. Squaretooth combs give better results than sharkstooth combs.

Gel analysis

An informative dideoxy component was easily detected by a missing segment or an extra segment. All the segments of the SSCP component were scored by visual analysis for the presence of abnormal migration in comparison with a wildtype control. For these experiments, a wildtype control was loaded in every third lane so that a mutant sample always was immediately adjacent to at least one wildtype sample. In addition, a pair of unidirectional ddF controls were included to distinguish downstream from upstream ddF segments. Unequivocal mobility changes were scored. Typically, a migration change of 1/2 of a band width on the upper part of the gel or 1/4 of a band width on the lower part was the limit of resolution. Mutations also may be detected if two closely migrating segments are resolved into separate segments. Conversely, if a mutation caused compression of separate segments, the segments were scored as positive.

ACKNOWLEDGEMENTS

We thank Mary Johnson for expert secretarial assistance. This work was supported by MH44276.

REFERENCES

1 Orita, M., Iwahana, H., Kanazawa, H., Hayashi, K. and Sekiya, T. (1989) Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc. Natl. Acad. Sci. USA, 86, 2766-2770. MEDLINE Abstract

2 Sarkar, G., Yoon, H. and Sommer, S.S. (1992) Screening for mutations by RNA single-strand conformation polymorphism (rSSCP): comparison with DNA-SSCP. Nucleic Acids Res., 20, 871-878. MEDLINE Abstract

3 Sarkar, G., Yoon, H. and Sommer, S.S. (1992) Dideoxy fingerprinting (ddF): a rapid and efficient screen for the presence of mutations. Genomics, 13, 441-443. MEDLINE Abstract

4 Hayashi, K. and Yandell, D.W. (1993) How sensitive is PCR-SSCP? Hum. Mutat., 2, 338-346. MEDLINE Abstract

5 Michaud, J., Brody, L.C., Steel, G., Fontaine, G., Martin, L.S., Valle, D. and Mitchell, G. (1992) Strand-separating conformational polymorphism analysis: efficacy of detection of point mutations in the human ornithine [delta]-aminotransferase gene. Genomics, 13, 389-394. MEDLINE Abstract

6 Sheffield, V.C., Beck, J.S., Kwitek, A.E., Sandstrom, D.W. and Stone, E.M. (1993) The sensitivity of single-strand conformation polymorphism analysis for the detection of single base substitutions. Genomics, 16, 325-332. MEDLINE Abstract

7 Hongo, T., Buzard, G.S., Calvert, R.J. and Weghorst, C.M. (1993) `Cold SSCP': a simple, rapid and non-radioactive method for optimized single-strand conformation polymorphism analyses. Nucleic Acids Res., 21, 3637-3642. MEDLINE Abstract

8 Glavac, D. and Dean, M. (1993) Optimization of the single-strand conformation polymorphism (SSCP) technique for detection of point mutations. Hum. Mutat., 2, 404-414. MEDLINE Abstract

9 Takahashi-Fujii, A., Ishino, Y., Shimada, A. and Kato, I. (1993) Practical application of fluorescence-based image analyzer for PCR single-stranded conformation polymorphism analysis used in detection of multiple point mutations. PCR Methods Appl., 2, 323-327. MEDLINE Abstract

10 Liu, Q. and Sommer, S.S. (1994) Parameters affecting the sensitivities of dideoxy fingerprinting and SSCP. PCR Methods Appl., 4, 97-108. MEDLINE Abstract

11 Danenberg, P.V., Horikoshi, T., Volkenandt, M., Danenberg, K., Lenz, H., Shea, L.C.C., Dicker, A.P., Simoneau, A., Jones, P.A. and Bertino, J.R. (1992) Detection of point mutations in human DNA by analysis of RNA conformation polymorphism(s). Nucleic Acids Res., 20, 573-579. MEDLINE Abstract

12 Blaszyk, H., Hartmann, A., Schroeder, J.J., McGovern, R.M., Sommer, S.S. and Kovach, J.S. (1995) Rapid and efficient screening for p53 gene mutations by dideoxyfingerprinting (ddF). BioTechniques, 18, 256-260. MEDLINE Abstract

13 Felmlee, T.A., Liu, Q., Whelen Christian, A., Williams, D., Sommer, S.S. and Persing, D.H. (1995) Genotypic detection of Mycobacterium tuberculosis rifampin resistance: comparison of single-strand conformation polymorphism and dideoxy fingerprinting. J. Clin. Pathol., 33, 1617-1623.

14 Liu, Q. and Sommer, S.S. (1995) Restriction endonuclease fingerprinting (REF): a sensitive method for screening mutations in long, contiguous segments of DNA. BioTechniques, 18, 470-477. MEDLINE Abstract

15 Sommer, S.S. (1995) Recent human germ-line mutation: inferences from patients with hemophilia B. Trends Genet. 11, 141-147. MEDLINE Abstract

16 Sommer, S.S. and Vielhaber, E.L. (1994) Phage promoter-based methods for sequencing and screening for mutations. In Mullis, K., Ferre, F. and Gibbs, R.A. (eds), The Polymerase Chain Reaction. Birkhauser, Boston, pp. 214-221.

17 Yoshitake, S., Schach, B.G., Foster, D.C., Davie, E.W. and Kurachi, K. (1985) Nucleotide sequence of the gene for human factor IX (anti-hemophilic factor B). Biochemistry, 24, 3736-3750. MEDLINE Abstract

18 Innis, M.A., Myambo, K.B., Gelfand, D.H. and Brow, M.A.D. (1988) DNA sequencing with Thermus aquaticus DNA polymerase and direct sequencing of polymerase chain reaction-amplified DNA. Proc. Natl. Acad. Sci. USA, 85, 9436-9440. MEDLINE Abstract

*To whom correspondence should be addressed

Journal Home Page

Home

Journals Home Page

Journals

OUP Home Page

OUP Home

Search

Search

Help

Help

jnl.info{at}oup.co.uk

-->

This page is maintained by OUP admin. Last updated Thu Oct 31 15:08:29 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
M. I. Rees, K. Harvey, H. Ward, J. H. White, L. Evans, I. C. Duguid, C. C.-H. Hsu, S. L. Coleman, J. Miller, K. Baer, et al.
Isoform Heterogeneity of the Human Gephyrin Gene (GPHN), Binding Domains to the Glycine Receptor, and Mutation Analysis in Hyperekplexia
J. Biol. Chem., June 27, 2003; 278(27): 24688 - 24696.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
S. Beyzade, S. Zhang, Y.-k. Wong, I. N. M. Day, P. Eriksson, and S. Ye
Influences of matrix metalloproteinase-3 gene variation on extent of coronary atherosclerosis and risk of myocardial infarction
J. Am. Coll. Cardiol., June 18, 2003; 41(12): 2130 - 2137.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
M. I. Rees, T. M. Lewis, J. B. J. Kwok, G. R. Mortier, P. Govaert, R. G. Snell, P. R. Schofield, and M. J. Owen
Hyperekplexia associated with compound heterozygote mutations in the {beta}-subunit of the human inhibitory glycine receptor (GLRB)
Hum. Mol. Genet., April 1, 2002; 11(7): 853 - 860.
[Abstract] [Full Text] [PDF]

Home page
 CROBMHome page
R. H. Finnell, K. A. Greer, R. C. Barber, J. A. Piedrahita, G. M. Shaw, and E. J. Lammer
Neural Tube and Craniofacial Defects With Special Emphasis On Folate Pathway Genes
Critical Reviews in Oral Biology & Medicine, January 1, 1998; 9(1): 38 - 53.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (46)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Liu, Q.
Right arrow Articles by Sommer, S. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Liu, Q.
Right arrow Articles by Sommer, S. S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?