- Email: [email protected]
A novel technique for rapid automated genotyping of DNA polymorphisms in the mouse
Molecular and Cellular Probes (1999) 13, 239–242 Article No. mcpr.1999.0239, available online at http://www.idealibrary.com on
A novel technique for rapid automated genotyping of DNA polymorphisms in the mouse A. Kuklin,1 A. P. Davis,2† K. H. Hecker,1∗ D. T. Gjerde1 and P. D. Taylor1 1
Transgenomic, Inc., 2032 Concourse Drive, San Jose, CA 95131, USA and 2 Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA (Received 4 June 1998, Accepted 12 March 1999)
The ability to rapidly and reliably genotype mice is an important concern. Traditional methods employ labour intensive and time consuming techniques such as test crossing, gel electrophoresis or nucleic acid hybridization. Here we show that a new molecular biology workstation, the WAVETM DNA Fragment Analysis System, can easily resolve polymerase chain reaction (PCR) products that have small differences in their lengths. Analysis is fully automated and takes less than 7 min per sample. Approximately 200 samples can be analysed per day with only minutes of hands-on time after completion of the PCR. Genotyping with the WAVETM DNA Fragment Analysis System is a fast and efficient method with minimal manual intervention. 1999 Academic Press
KEYWORDS: mice, genotyping, DNA polymorphisms.
INTRODUCTION The mouse is a powerful model organism for studying mammalian gene function. In addition to its numerous genetic tools, the mouse has a detailed comparative linkage map with humans. Thus, genetic information learned in the mouse can quickly and easily be addressed in humans, allowing for rapid breakthroughs in disease trait mapping. As a genetic reagent, the mouse provides a wealth of variants in the form of spontaneous, gene targeted, radiation induced and chemically induced mutations. However, an orderly and cost efficient mouse colony requires the ability to quickly and reliably genotype large numbers of animals in a high-throughput procedure. Chemically induced or spontaneous mutations must first be mapped to specific regions on mouse chromosomes by outbreeding and scanning for distinguishing DNA polymorphisms of simple marker loci that co-segregate
with the mutation. This can be instrumental for cost effectiveness when dealing with recessive mutations, since heterozygotes and wild-type homozygotes are phenotypically indistinguishable. Classically, the only way to genotype such animals was by a time consuming test cross. DNA polymorphism detection obviates this need. Once a closely linked DNA polymorphism is identified, mutations of interest can be followed by merely typing for the DNA locus. For example, a chemically-induced recessive mutation in the BALB/c mouse can be followed by crossing it with an FRCH wild-type mouse and using the closely linked D7Mit352 marker to provide a DNA length polymorphism between the two strains. The BALB/c allele, which is linked to the mutation of interest, yields a D7Mit352 polymerase chain reaction (PCR) product of 120 base pairs (bp) while the FRCH wild-type allele is only 116 bp (Fig. 1a, b). Conventional methods for genotyping make use
∗ Author to whom all Correspondence should be addressed at: Transgenomic, Inc., 2032 Concourse Drive, San Jose, CA 95131, USA. Tel: +14 08 432 3230; Fax: +14 08 432 3231; E-mail: [email protected] † Current address: The Jackson Laboratory, Mouse Genome Informatics, 600 Main Street, Bar Harbor, Maine 04609, USA.
0890–8508/99/030239+04 $30.00/0
1999 Academic Press
A. Kuklin et al.
240 (a)
1
2
3
4
5
6
7
bp 120 116
(b) 7 6 5 4 3 2 1 102 bp pUC18 0
2
4
6
8
10
Time (min)
Fig. 1. (a) Conventional identification of DNA polymorphisms in the mouse using gel electrophoresis to distinguish the D7Mit352 polymerase chain reaction (PCR) products used to genotype a litter of mice. The BALB/c allele yields a 120 bp product while the FRCH allele yields a 116 bp product. Heterozygotes display two bands corresponding to the 120 and 116 bp fragments. (b) Automated identification of DNA polymorphisms with the WAVETM System using PCR products identical to those used in (a). The FRCH and BALB/c alleles are shown in chromatograms 6 and 7. Embryo samples from heterozygous mice are presented by chromatograms 1–5. Heterozygotes are characterized by the presence of three peaks, 120 and 116 bp homoduplexes and a 120/116 bp heteroduplex eluting with the shortest retention time. The size marker used is a HaeIII digested pUC18 plasmid. Shown here are fragments of lengths: 298, 267, 257, 174, 102 and 80 base pairs; the 102 bp fragment is identified.
of agarose or polyacrylamide gel electrophoresis or Southern blotting to analyse PCR products. These methods are time consuming and labour intensive. Resolution of a 120 bp fragment from a 116 bp fragment requires high concentrations of agarose and typically requires longer electrophoresis times to achieve a resolution of 4 bp. Polyacrylamide gels require appropriate electrophoretic equipment and the handling of acrylamide, a neurotoxin. Southern blotting requires almost a whole workday and may take several days for the development of autoradiographs. Here, we report a rapid method for automated genotyping illustrated on a murine system using the WAVETM DNA Fragment Analysis System, a molecular biology workstation most widely used for polymorphism detection.1
MATERIALS Genomic DNA was isolated from individual 8·5-day mouse embryos. Primers specific to the marker locus D7Mit352 (Gibco BRL, Gaithersburg, MD, USA) were used to amplify a PCR product of either 120 bp (the BALB/c allele linked to the mutation of interest) or 116 bp (the FRCH wild-type allele). The forward primer has the sequence 5′-AGC CAA TTG CAA CCA AAA TTT-3′ and the reverse primer’s sequence is 5′AGC ATG GAA AAT TGA CAA TTC C-3′. Polymerase chain reaction was performed in 25-ll reactions amplified for 30 cycles at 94°C (30 s), 55°C (2 min) and 72°C (2 min). For electrophoretic analysis, aliquots of the PCR were run on a 3·2% MetaPhor (FMC BioProducts, Rockland, ME, USA) agarose gel.
Automated genotyping in mice
For automated analysis on the WAVETM DNA Fragment Analysis System (Transgenomic Inc., San Jose, CA, USA), PCR reactions were loaded in a 96-well format into the autosampler and 5 ll were automatically injected per analysis. Analysis conditions were as follows: column temperature was 50°C and flow rate 0·75 ml min−1. Buffer A: 0·1 triethylammonium acetate (TEAA) (Transgenomic Inc., San Jose, CA, USA) in water, Buffer B: 0·1 TEAA and 25% acetonitrile in water. Gradient: 0 min: 55% A, 45% B; 4·4 min: 40% A, 60% B; 7·5 min: 30% A, 70% B; 8 min: 0% A, 100% B; 10 min: 0% A, 100% B, 11–13 min: equilibration at 55% A and 45% B.2
DISCUSSION The WAVETM DNA Fragment Analysis System separates amplified DNA fragments on a DNASep column containing hydrophobic non-porous particles.1, 3 The stationary phase is converted into a dynamic anion-exchanger by the ion-pairing reagent TEAA. The positively charged ammonium ions interact with the negatively charged phosphate ions of the DNA. The number of negatively charged phosphates is directly proportional to the length of the DNA. Therefore, more TEAA molecules will interact with a longer DNA fragment. At the same time the alkyl chains of the TEAA molecule interact with the hydrophobic surface of the DNASep column. Longer DNA fragments will therefore exhibit a higher affinity to the column matrix than shorter fragments,3–5 effectively resulting in a size-based separation of DNA molecules. DNA fragments are released from the DNASep matrix with increasing amounts of organic solvent in the mobile phase. Acetonitrile interferes with the hydrophobic interaction between the stationary phase and the alkyl chains of the bridging TEAA molecules. Shorter fragments, which bind fewer TEAA bridging molecules, are released sooner than longer fragments. The latter leads to size-dependent separation of DNA fragments.5 Gradient conditions for the separation of DNA fragments (fragment sizing) are accurately predicted by the system’s software obviating the need for optimization of running conditions. The elution of DNA fragments is recorded by a u.v. detector at 254 nm. The BALB/c allele for D7Mit 352 yields a PCR product of 120 bp with a retention time of 6·14 min (Fig. 1b, trace 7). The FRCH allele for D7Mit 352 yields a 116 bp PCR fragment, which elutes under the given conditions after 5·85 min (Fig. 1b, trace 6). Shoulders observed at shorter retention times relative to the 116 and 120 bp homozygous peaks are caused
241
by PCR fragments that contain polymerase-induced errors. The presence of these errors leads the formation of heteroduplexes, which are partially melted under the analysis conditions used here. Partially singlestranded heteroduplexes are retained on the column less than fully natured homoduplexes. Thus, partially denatured heteroduplexes elute earlier than their corresponding homoduplexes. Unlike agarose gel electrophoresis, where only two bands are observed, a third peak is observed with a retention time of 5·15 min during analysis on the WAVETM DNA Fragment Analysis System. This peak corresponds to the heteroduplex produced by the annealing of the different-sized PCR products of the two alleles. Heteroduplexes are retained differentially and they can be separated from the homoduplex fragments.1, 6 The typical three peak pattern is consistently present in all heterozygous mice and differs from the corresponding homozygous mice (Fig. 1b, traces 6 and 7). Visual scoring is easy, one peak or three peaks. Chromatograms are recorded by the WAVETM System and retention times are documented. Analysis time for each sample is less than 8 min, which has not been optimized and could be further shortened. This allows for processing of approximately 200 samples, or two 96-well plates, within a 24-h time period. The only hands-on time consists of loading of samples into the autosampler of the WAVETM System and choosing the detection method. The latter is facilitated by the system’s software. This whole process requires only a few minutes of manual manipulation. Compared to the current gel-based methods, genotyping on the WAVETM DNA Fragment Analysis System omits sample preparation with gel-loading dyes, gel preparation, pipetting, electrophoresis, visualization of DNA and photographic recording. In many cases, scoring of transgenic animals is performed by Southern blotting, which further delays data acquisition by several days and typically involves the use of radioactive isotopes. The procedure that is reported in this paper therefore offers a convenient and viable solution for high-throughput genotyping.
REFERENCES 1. Kuklin, A., Munson, K., Gjerde, D. & Taylor, P. (1998). Detection of single-nucleotide polymorphisms with the WAVETM DNA Fragment Analysis System. Genetic Testing 4, 201–6. 2. Kuklin, A., Davis, A. P., Gjerde, D. & Taylor, P. (1998). Automated and rapid genotyping of mouse colonies
242
A. Kuklin et al.
with the WAVETM DNA Fragment Analysis System. Application Note (Transgenomic, San Jose, CA, USA) 106, 1–4. 3. Huber, C. G. & Berti, G. N. (1996). Detection of partial denaturation in AT-rich DNA fragments by ion-pair reversed-phase chromatography. Analytical Chemistry 68, 2959–65. 4. Huber, C. G. (1998). Micropellicular stationary phases for high-performance liquid chromatography of doublestranded DNA. Journal of chromatography 806, 1–28.
5. Huber, C. G., Oefner, P. J. & Bonn, G. K. (1995). Rapid and accurate sizing of DNA fragments by ion-pair chromatography on alkylated nonporous poly(styrenedivinylbenzene) particles. Analytical Chemistry 67, 578–85. 6. Marino, M. A., Devaney, J. M., Smith, J. K. & Girard, J. E. (1998). Sequencing using capillary electrophoresis of short tandem repeats alleles separated and purified by high performance liquid chromatography. Electrophoresis 19, 108–18.
A novel technique for rapid automated genotyping of DNA polymorphisms in the mouse A. Kuklin,1 A. P. Davis,2† K. H. Hecker,1∗ D. T. Gjerde1 and P. D. Taylor1 1
Transgenomic, Inc., 2032 Concourse Drive, San Jose, CA 95131, USA and 2 Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA (Received 4 June 1998, Accepted 12 March 1999)
The ability to rapidly and reliably genotype mice is an important concern. Traditional methods employ labour intensive and time consuming techniques such as test crossing, gel electrophoresis or nucleic acid hybridization. Here we show that a new molecular biology workstation, the WAVETM DNA Fragment Analysis System, can easily resolve polymerase chain reaction (PCR) products that have small differences in their lengths. Analysis is fully automated and takes less than 7 min per sample. Approximately 200 samples can be analysed per day with only minutes of hands-on time after completion of the PCR. Genotyping with the WAVETM DNA Fragment Analysis System is a fast and efficient method with minimal manual intervention. 1999 Academic Press
KEYWORDS: mice, genotyping, DNA polymorphisms.
INTRODUCTION The mouse is a powerful model organism for studying mammalian gene function. In addition to its numerous genetic tools, the mouse has a detailed comparative linkage map with humans. Thus, genetic information learned in the mouse can quickly and easily be addressed in humans, allowing for rapid breakthroughs in disease trait mapping. As a genetic reagent, the mouse provides a wealth of variants in the form of spontaneous, gene targeted, radiation induced and chemically induced mutations. However, an orderly and cost efficient mouse colony requires the ability to quickly and reliably genotype large numbers of animals in a high-throughput procedure. Chemically induced or spontaneous mutations must first be mapped to specific regions on mouse chromosomes by outbreeding and scanning for distinguishing DNA polymorphisms of simple marker loci that co-segregate
with the mutation. This can be instrumental for cost effectiveness when dealing with recessive mutations, since heterozygotes and wild-type homozygotes are phenotypically indistinguishable. Classically, the only way to genotype such animals was by a time consuming test cross. DNA polymorphism detection obviates this need. Once a closely linked DNA polymorphism is identified, mutations of interest can be followed by merely typing for the DNA locus. For example, a chemically-induced recessive mutation in the BALB/c mouse can be followed by crossing it with an FRCH wild-type mouse and using the closely linked D7Mit352 marker to provide a DNA length polymorphism between the two strains. The BALB/c allele, which is linked to the mutation of interest, yields a D7Mit352 polymerase chain reaction (PCR) product of 120 base pairs (bp) while the FRCH wild-type allele is only 116 bp (Fig. 1a, b). Conventional methods for genotyping make use
∗ Author to whom all Correspondence should be addressed at: Transgenomic, Inc., 2032 Concourse Drive, San Jose, CA 95131, USA. Tel: +14 08 432 3230; Fax: +14 08 432 3231; E-mail: [email protected] † Current address: The Jackson Laboratory, Mouse Genome Informatics, 600 Main Street, Bar Harbor, Maine 04609, USA.
0890–8508/99/030239+04 $30.00/0
1999 Academic Press
A. Kuklin et al.
240 (a)
1
2
3
4
5
6
7
bp 120 116
(b) 7 6 5 4 3 2 1 102 bp pUC18 0
2
4
6
8
10
Time (min)
Fig. 1. (a) Conventional identification of DNA polymorphisms in the mouse using gel electrophoresis to distinguish the D7Mit352 polymerase chain reaction (PCR) products used to genotype a litter of mice. The BALB/c allele yields a 120 bp product while the FRCH allele yields a 116 bp product. Heterozygotes display two bands corresponding to the 120 and 116 bp fragments. (b) Automated identification of DNA polymorphisms with the WAVETM System using PCR products identical to those used in (a). The FRCH and BALB/c alleles are shown in chromatograms 6 and 7. Embryo samples from heterozygous mice are presented by chromatograms 1–5. Heterozygotes are characterized by the presence of three peaks, 120 and 116 bp homoduplexes and a 120/116 bp heteroduplex eluting with the shortest retention time. The size marker used is a HaeIII digested pUC18 plasmid. Shown here are fragments of lengths: 298, 267, 257, 174, 102 and 80 base pairs; the 102 bp fragment is identified.
of agarose or polyacrylamide gel electrophoresis or Southern blotting to analyse PCR products. These methods are time consuming and labour intensive. Resolution of a 120 bp fragment from a 116 bp fragment requires high concentrations of agarose and typically requires longer electrophoresis times to achieve a resolution of 4 bp. Polyacrylamide gels require appropriate electrophoretic equipment and the handling of acrylamide, a neurotoxin. Southern blotting requires almost a whole workday and may take several days for the development of autoradiographs. Here, we report a rapid method for automated genotyping illustrated on a murine system using the WAVETM DNA Fragment Analysis System, a molecular biology workstation most widely used for polymorphism detection.1
MATERIALS Genomic DNA was isolated from individual 8·5-day mouse embryos. Primers specific to the marker locus D7Mit352 (Gibco BRL, Gaithersburg, MD, USA) were used to amplify a PCR product of either 120 bp (the BALB/c allele linked to the mutation of interest) or 116 bp (the FRCH wild-type allele). The forward primer has the sequence 5′-AGC CAA TTG CAA CCA AAA TTT-3′ and the reverse primer’s sequence is 5′AGC ATG GAA AAT TGA CAA TTC C-3′. Polymerase chain reaction was performed in 25-ll reactions amplified for 30 cycles at 94°C (30 s), 55°C (2 min) and 72°C (2 min). For electrophoretic analysis, aliquots of the PCR were run on a 3·2% MetaPhor (FMC BioProducts, Rockland, ME, USA) agarose gel.
Automated genotyping in mice
For automated analysis on the WAVETM DNA Fragment Analysis System (Transgenomic Inc., San Jose, CA, USA), PCR reactions were loaded in a 96-well format into the autosampler and 5 ll were automatically injected per analysis. Analysis conditions were as follows: column temperature was 50°C and flow rate 0·75 ml min−1. Buffer A: 0·1 triethylammonium acetate (TEAA) (Transgenomic Inc., San Jose, CA, USA) in water, Buffer B: 0·1 TEAA and 25% acetonitrile in water. Gradient: 0 min: 55% A, 45% B; 4·4 min: 40% A, 60% B; 7·5 min: 30% A, 70% B; 8 min: 0% A, 100% B; 10 min: 0% A, 100% B, 11–13 min: equilibration at 55% A and 45% B.2
DISCUSSION The WAVETM DNA Fragment Analysis System separates amplified DNA fragments on a DNASep column containing hydrophobic non-porous particles.1, 3 The stationary phase is converted into a dynamic anion-exchanger by the ion-pairing reagent TEAA. The positively charged ammonium ions interact with the negatively charged phosphate ions of the DNA. The number of negatively charged phosphates is directly proportional to the length of the DNA. Therefore, more TEAA molecules will interact with a longer DNA fragment. At the same time the alkyl chains of the TEAA molecule interact with the hydrophobic surface of the DNASep column. Longer DNA fragments will therefore exhibit a higher affinity to the column matrix than shorter fragments,3–5 effectively resulting in a size-based separation of DNA molecules. DNA fragments are released from the DNASep matrix with increasing amounts of organic solvent in the mobile phase. Acetonitrile interferes with the hydrophobic interaction between the stationary phase and the alkyl chains of the bridging TEAA molecules. Shorter fragments, which bind fewer TEAA bridging molecules, are released sooner than longer fragments. The latter leads to size-dependent separation of DNA fragments.5 Gradient conditions for the separation of DNA fragments (fragment sizing) are accurately predicted by the system’s software obviating the need for optimization of running conditions. The elution of DNA fragments is recorded by a u.v. detector at 254 nm. The BALB/c allele for D7Mit 352 yields a PCR product of 120 bp with a retention time of 6·14 min (Fig. 1b, trace 7). The FRCH allele for D7Mit 352 yields a 116 bp PCR fragment, which elutes under the given conditions after 5·85 min (Fig. 1b, trace 6). Shoulders observed at shorter retention times relative to the 116 and 120 bp homozygous peaks are caused
241
by PCR fragments that contain polymerase-induced errors. The presence of these errors leads the formation of heteroduplexes, which are partially melted under the analysis conditions used here. Partially singlestranded heteroduplexes are retained on the column less than fully natured homoduplexes. Thus, partially denatured heteroduplexes elute earlier than their corresponding homoduplexes. Unlike agarose gel electrophoresis, where only two bands are observed, a third peak is observed with a retention time of 5·15 min during analysis on the WAVETM DNA Fragment Analysis System. This peak corresponds to the heteroduplex produced by the annealing of the different-sized PCR products of the two alleles. Heteroduplexes are retained differentially and they can be separated from the homoduplex fragments.1, 6 The typical three peak pattern is consistently present in all heterozygous mice and differs from the corresponding homozygous mice (Fig. 1b, traces 6 and 7). Visual scoring is easy, one peak or three peaks. Chromatograms are recorded by the WAVETM System and retention times are documented. Analysis time for each sample is less than 8 min, which has not been optimized and could be further shortened. This allows for processing of approximately 200 samples, or two 96-well plates, within a 24-h time period. The only hands-on time consists of loading of samples into the autosampler of the WAVETM System and choosing the detection method. The latter is facilitated by the system’s software. This whole process requires only a few minutes of manual manipulation. Compared to the current gel-based methods, genotyping on the WAVETM DNA Fragment Analysis System omits sample preparation with gel-loading dyes, gel preparation, pipetting, electrophoresis, visualization of DNA and photographic recording. In many cases, scoring of transgenic animals is performed by Southern blotting, which further delays data acquisition by several days and typically involves the use of radioactive isotopes. The procedure that is reported in this paper therefore offers a convenient and viable solution for high-throughput genotyping.
REFERENCES 1. Kuklin, A., Munson, K., Gjerde, D. & Taylor, P. (1998). Detection of single-nucleotide polymorphisms with the WAVETM DNA Fragment Analysis System. Genetic Testing 4, 201–6. 2. Kuklin, A., Davis, A. P., Gjerde, D. & Taylor, P. (1998). Automated and rapid genotyping of mouse colonies
242
A. Kuklin et al.
with the WAVETM DNA Fragment Analysis System. Application Note (Transgenomic, San Jose, CA, USA) 106, 1–4. 3. Huber, C. G. & Berti, G. N. (1996). Detection of partial denaturation in AT-rich DNA fragments by ion-pair reversed-phase chromatography. Analytical Chemistry 68, 2959–65. 4. Huber, C. G. (1998). Micropellicular stationary phases for high-performance liquid chromatography of doublestranded DNA. Journal of chromatography 806, 1–28.
5. Huber, C. G., Oefner, P. J. & Bonn, G. K. (1995). Rapid and accurate sizing of DNA fragments by ion-pair chromatography on alkylated nonporous poly(styrenedivinylbenzene) particles. Analytical Chemistry 67, 578–85. 6. Marino, M. A., Devaney, J. M., Smith, J. K. & Girard, J. E. (1998). Sequencing using capillary electrophoresis of short tandem repeats alleles separated and purified by high performance liquid chromatography. Electrophoresis 19, 108–18.