Christopher Becker, Jia Li, Thomas A. Shaler, Joanna M. Hunter, Hua Lin, and Joseph A. Monforte
GeneTrace Systems, Inc., 333 Ravenswood Avenue, Menlo Park, CA 94025

× Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø × Ø

An alternative approach to gel electrophoresis for size separation and detection of short tandem repeats (STRs) and other DNA is time-of-flight mass spectrometry. In this method, DNA molecules are separated by an electric field, causing transport through vacuum instead of through a gel. Because in vacuum there is nothing to slow down the molecules, characteristic separation times are on the order of microseconds, although signal averaging by summing over multiple spectra extends typical measurement times to a few seconds. Size separation begins when the DNA molecules are liberated from a host organic matrix by low intensity, pulsed, ultraviolet laser light. No radioactive or fluorescent labels are used and the technology has been automated. We have successfully employed matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) technology to detect polymerase chain reaction (PCR) amplified tetranucleotide STR loci. Four loci were examined, CSF1PO, F13A01, FESFPS, and TH01. PCR products were amplified from human genomic DNA, captured and released from magnetic beads, and analyzed by mass spectrometry. All the alleles of the 4 loci were well resolved and we were able to discriminate 1 nucleotide difference for TH01 allele 9.3 and 10. The results from MALDI-TOF MS analysis are much more accurate than those obtained from gel electrophoresis and the time for detection is orders of magnitude faster. This technology can be used for human identification, genome mapping, and medical diagnostics.

Short tandem repeat (STR) loci are composed of tandemly repeated sequences of 3 to 7 base pairs in length (1,2). Their abundance in human genomic DNA (one every 15 kb) (3), high polymorphism in the number of repeats, and small size amenable to amplification, make them ideal genetic markers for individual identification, genome mapping, and detection of genetic disorders and cancers (1, 4-6). Alleles of these loci are differentiated by the numbers of copies of the STR sequence within the amplified region.

For most procedures that detect alleles of these loci, gel electrophoresis followed by radioactive, silver stain or fluorescent detection is now commonly used to separate and measure the sizes of PCR products. This method, however, is time-consuming and laborious, often prohibiting large-scale testing. Large-scale STR loci detection needs the development of new techniques which are reliable, fast, accurate and easily automated. An alternative approach to gel electrophoresis for size separation and detection of DNA is time-of-flight mass spectrometry. In this method, DNA molecules are separated by an electric field causing transport through vacuum instead of through a gel. Because in vacuum there is nothing to interact with the molecules, separation is obtained within microseconds. No radioactive or fluorescent labels are used.

In this work we used optimized matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) technology (7,8) with proprietary reagents to examine four STR loci which are commonly used for human identification (CSF1PO, human c-fms proto-oncogene for CSF-1 receptor gene; F13A01, human coagulation factor XIII a subunit gene; FESEPS, human c-fes/fps proto-oncogene; and TH01, human tyrosine hydroyxylase gene).

Allele ladders and control DNA K562 were purchased from Promega Corporation (Madison, Wisconsin, USA). Other human genomic DNAs were provided anonymously (see Acknowledgments). All primers were synthesized by Biosource International (Menlo Park, California, USA), one of each set of PCR primers was biotinylated. STR loci examined in this study (CSF1PO, F13A01, FESFPS, and TH01) have been previously described (9-11) and their characteristic information is listed in Table 1. PCR amplification was performed essentially as described by Kimpton et al. (13) and PCR products were captured and released from streptavidin-coated magnetic beads (Patent Application No. 08/639, 363, 1995). The size separation and detection of PCR products were done by MALDI-TOF MS as previously described (8). All sample analyses were repeated at least three times.

To develop a new, fast and accurate method to analyze STR loci, we employed MALDI-TOF MS to examine the four loci already well characterized (9,11). The allele ladders of the four loci were well resolved, all the peaks were sharp, which assist in the precise assignment of the mass. The base accuracy is approximately 0.03. Figure 1 shows the TH01 allele ladder which consists of alleles 5, 6, 7, 8, 9, 10, and 11. The measured mass difference between each repeat is 1209.7, correlates well with the expected mass difference 1210.8. For control DNA K562 and unknown sample #1, all 4 loci were analyzed by MALDI-TOF MS. Figure 2 shows the mass spectra of sample #1 for the 4 loci studied, alleles of each locus were clearly determined: TH01: 8, 9.3; FESFPS: 10, 12; CSF1PO: 10, 11; and F13A01: 5, 6. All the expected alleles of the four loci for K562 were observed (data not shown). The sample #1 TH01 allele 9.3 was further confirmed by direct nucleotide sequencing of PCR products by MALDI-TOF MS (data not shown). For the TH01 locus, additional samples #2 and #3 were examined. The 9.3 and 10 alleles for sample #2 and 7 and 8 alleles for sample #3 were observed, which match exactly with allele information obtained for these two samples by gel electrophoresis (George Sensabaugh, personal communication). The one nucleotide difference for allele 9.3 and 10 was resolved clearly in the mass spectrum (see Figure 3). The little weak peak present on the high mass side of the main peak in some of the mass spectra is a matrix adduct.

In conclusion, we have successfully employed MALDI-TOF MS to determine polymorphic STR loci. The results obtained from these optimized MALDI-TOF mass spectra are superior to those obtained from gel electrophoresis in the sense that the detection speed and accuracy are greatly improved. This method can be easily automated for precise analysis of STR loci for human identification, genetic mapping, and detection of genetic disorders and cancers.

Anonymous DNA samples for the STR loci used in this study were kindly provided by Dr. Yi-wen Zhu of Lawrence Berkeley Laboratory, Berkeley, CA and Dr. George Sensabaugh of University of California, Berkeley, CA. We thank Yuping Tan and Charles L. Muir for technical assistance. This research was supported in part by a Cooperative Agreement from the Advanced Technology Program, National Institute of Standards and Technology (70NANB5H1029).

REFERENCES

Figure 2. Time-of-flight mass spectra of the products of four different short tandem repeat loci for DNA isolated from a single individual.

Figure 3. Time-of-flight mass spectrum of the TH01 locus for a sample containing a known 9.3, 10 heterozygous mixture demonstrating single-base resolving capability.