Catalog  |  Cart  |  Log In

• Corporate
• Products
• Resources

Comparison of Short Tandem Repeat (STR) Detection Using Silver, Fluorescence and Matrix Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrophotometry (MALDITOF-MS)

Steven B. Lee¹, Martin Buoncristiani¹, James W. Schumm² and Dale Wingeleth^3
1 California State Department of Justice DNA Laboratory 626 Bancroft Way, Berkeley, CA 94710
² Promega Corporation 2800 Woods Hollow Road Madison WI 53711
³ ChemiTox Laboratory 5401 Western Boulder, CO 80301.

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

Amplification of CTT
Detection of a 16 bp dsDNA oligonucleotide using MALDITOF-MS

RESULTS AND DISCUSSION

Sensitivity of fluorescent dye detection using fluorescent scanning and real time detection
Relative precision of sizing STRs using different instruments
MALDITOF-MS precision sizing of DNA

ACKNOWLEDGMENTS

REFERENCES

TABLES and FIGURES

ABSTRACT

This report includes comparisons of the STR loci, HUMTHO1, HUMTPOX, and HUMCSF1PO (CTT). DNA labeled with different dyes (fluorescein, ROX, TAMRA, JOE and FAM) and the relative capabilities of different detection methods (using different instruments) to visualize these dyes were compared. Fluorescent detection methods compared in our study include laser scanning with the FlourImager 575 (Molecular Dynamics) and post-electrophoresis fluorescent dye staining, the FMBIO100 laser induced imaging system (Hitachi) using either post-electrophoresis fluorescent dye staining or the detection of STRs amplified with fluorescently labeled primers and automated fluorescent detection utilizing the ABD373. The precision of sizing was evaluated for the same samples run on different gels by different operators on different instruments. DNA fragment sizing of STRs was attempted using matrix assisted laser desorption ionization time-of-flight mass spectrophotometry (MALDITOF-MS).

The use of fluorescent end-labeled primers is superior to post-fluorescent and post-silver staining because it is faster and because only one of the two strands is visualized in this format, simplifying the interpretation of results. All instruments were capable of excitation and detection of fluorescein used to label the GenePrint™ CTTV products. DNA labeled with ROX, TAMRA, and JOE were detected on the Hitachi FMBIO100 laser scanner.

Sizing of PCR products using the ABD373 and Hitachi FMBIO was + 0.5 bp using the GeneScan ROX labeled size standards. Highly reproducible STR sizing was achieved using ABI 373 GeneScan of +0.27 bp for the same samples ran on different gels by different operators. Finally, highly reproducible molecular weight analysis of a 16 bp dsDNA was achieved using MALDITOF-MS. Initial attempts at sizing STRs using MALDITOF-MS were unsuccessful, but future studies will include further evaluation of STR sizing using MALDITOF-MS.

INTRODUCTION

Short tandem repeats (STRs) are head-to-tail arrangements of sequence units ranging in size from 3-7 bps. They are common in genomes, polymorphic, stably inherited and have been the topic of study in many fields including genome mapping, clinical diagnostics and forensic typing.

High levels of heterozygosity and discrimination, the ability to simultaneously amplify several loci in multiplex reactions and to analyze STRs from highly degraded and environmentally abused samples (due in part to their small size range: 120-330 bp) have made STRs the pre-eminent system for forensic casework and databanking.

Typing can be achieved by amplification using STR specific primers (individually or in multiplex reactions), separation of amplified alleles using polyacrylamide gel or capillary electrophoresis and detection using silver or fluorescent methods. Analysis of biomolecules is also possible by using a sensitive new technology, matrix assisted laser desorption/ionization time-of-flight mass spectrophotometry (MALDITOF-MS). MALDITOF-MS is reported to have femtomole sensitivity, and molecular weight accuracy on the order of +0.01% (Siuzdak, 1994).

The goals of this study were to evaluate detection strategies of STR systems for sensitivity and precision using automated real-time analysis (Applied Biosystems 373, fluorescent scanning (Hitachi FMBIO100, Molecular Dynamics FlourImager SI) of fluorescently tagged STR PCR products (PRISM-CTT and GenePrint™ CTTV) and compare the results to visual detection of silver stained products and analysis using matrix assisted laser desorption/ionization time-of-flight mass spectrophotometry (MALDITOF-MS).

We performed comparisons of the relative sensitivity of fluorescent dye detection using fluorescent scanning and real time detection methods of the STR loci (HUMTHO1, HUMTPOX, and HUMCSF1PO), precision of sizing STRs within and among methods and an evaluation of sizing reproducibility of DNA using MALDITOF-MS.

MATERIALS AND METHODS

Amplification of CTT

Samples were stored, extracted and quantitated as previously described (Lee et al. 1994.). STR loci examined in this study (HUMTHO1, HUMTPOX, HUMCSF1PO) have been previously described (GenePrint™ Product Catalogue, Promega; Crouse and Schumm 1995; Micka et al. 1995) and elsewhere in these proceedings. Primers used were labeled with fluorescein (CTTV: GenePrint™ Fluorescent STR Multiplex System) or with TAMRA (PRISM STR Primer set CTT). Amplification of the loci was performed as described in the GenePrint protocols (Promega) or using recommended conditions by ABD. Fluorescent dye evaluation was also performed for ABD dyes ROX, JOE and FAM. Detection of the dyes and sizing of alleles was performed using all of the following instruments: Molecular Dynamics FlourImager SI, Hitachi FMBIO and Applied Biosystems 373 and 377. Table 1 lists the characteristics of the fluorescent dyes evaluated in this study.

Detection of a 16 bp dsDNA oligonucleotide using MALDITOF-MS

For our precision determination of MALDITOF-MS we used a 16 bp oligonucleotide and a 6-aza-2-thiothymine matrix, ATT reported to provide inter-assay variation to 0.006% precision (Lecchi et al. 1995). The MALDITOF-MS we used is a Hewlett Packard 2025G model.

RESULTS AND DISCUSSION

Sensitivity of fluorescent dye detection using fluorescent scanning and real time detection

Relative sensitivity of detection of different fluorescent dyes on three different instruments is shown in Table 2 (ABD= ABD373 and ABD 377 real time analysis, Hitachi= Hitachi FMBIO and MD = Molecular Dynamics FlourImager). All of the instruments evaluated were capable of detecting fluorescein used in the GenePrint™ Fluorescent STRs. Both the ABI and Hitachi instruments are capable of detecting all of the dyes evaluated. Detection of ROX, TAMRA (Figure 1) and JOE matrix standards was achieved on the Hitachi FMBIO. Recently, we have also been able to detect FAM on the Hitachi FMBIO100 although with reduced sensitivity. We used the MD FlourImager to detect all dyes except ROX. Evaluations of dye sensitivity of FAM on the fluorescent scanners were done using the products amplified with the ABD AmpflSTR BLU system.

Relative precision of sizing STRs using different instruments

Precision of sizing of CTT alleles was performed for the same amplified samples (n=20) using different gel compositions, prepared by different analysts on different instruments. Sizing reproducibility results for representative samples are shown in Tables 3-5. The average size difference for all loci was 0.27 bp on the ABI 373 providing excellent precision for a comparison by two different analysts (MB and SL).

Precision of sizing of the same samples on two different gels using GeneScan 500 size standards on the Hitachi FMBIO is shown in Table 4. Sizing reproducibility was within one base pair using this detection strategy.

Precision of sizing the same samples on two different instruments is shown in Table 5. The average difference is sizing using the same samples for THO1 between the ABI373 and the Hitachi FMBIO was 0.5 bp.

A summary of the comparison of detection strategies is shown in Table 6. Results of fluorescent scanning using real-time detection and fluorescent end labeling of primers resulted in reduced time of analysis and increased precision but is more costly in set-up. The accurate, rapid determination of fragment sizes found for the real-time detection strategy in our report corroborates previous findings by the European DNA profiling group (EDNAP-Kimpton et al. 1995) and Canadian RCMP Laboratory (Fregeau and Fourney 1993), reporting that automated fluorescence detection was found to be "robust and reproducible, precise, accurate and sensitive."

MALDITOF-MS precision sizing of DNA

MALDITOF-MS stands for Matrix Assisted Laser Desorption/Ionization Time-of-flight Mass Spectrophotometry and is a rapid, simple, accurate method of obtaining molecular weight information on a wide range of compounds (Fig. 3; Siuzdak 1994; Fitzgerald and Smith 1995).

Briefly, MALDITOF-MS starts with ionization induced by a laser with the matrix assisting by transferring energy to the sample without artifactual modification. Ions, once formed can be directed electrostatically into a mass analyzer that differentiates ions according to their mass-to-charge ratio (m/z).

The MALDITOF-MS has multiple advantages.

• a wide range of different applications, including analysis of DNA, proteins, carbohydrates, drug metabolites and oligonucleotides
• rapid analysis time, ease of operation (no dedicated operator is required)
• sensitivity down to femtomoles
• tolerance to some buffers, salts and denaturants.
• high mass range to 500,000 D
• low cost per analysis

Some disadvantages are the instrument cost ($150,000-300,000), and a limited amount of precision in determining molecular weights of large DNA fragments (greater than 100 bp)

We evaluated the reproducibility of sizing DNA using the MALDITOF-MS as described in our methodology, MALDITOF-MS inter-assay variation of a 16 bp mass was 0.004% using the ATT matrix (10174.0 vs 10212.9 Da). Furthermore, MALDITOF-MS sizing of larger DNA fragments has been recently reported (Liu et al. 1995). In that particular study, fragments of PCR products up to 250 bp were resolved to 1 bp. They reported analysis of the short tandem repeat locus vWF in homozygous and heterozygous canids. These results suggest that MALDITOF-MS may be a rapid, precise, inexpensive means of analyzing fragment sizes of STRs in the future.

ACKNOWLEDGMENTS

The authors would like to acknowledge Nikki Duda, Michi Lee, Sherrie Post and John Tonkyn from the California Department of Justice DNA Laboratory for providing some of the data used in this paper; Warner Yuen from Hitachi for helping to implement the scanner in our laboratory; Nikki Fildes, Jim Robertson and Sean Walsh from ABD for providing CTT primers and protocols; Elaine Mansfield, Marilynn Munson, Dick Rubin, Jennifer Worley, and Deborah Smead from Molecular Dynamics for their help with fluorescent scanning applications; Ann Lins and Cindy Sprecher from Promega for providing the CTTV primers; Margaret Kuo from Orange County for information leading to our purchase of the Hitachi FMBIO and the CAC; A. Reed and the V. McGlaughlin Grant to Steven B. Lee for partial funding of the student research assistant and the supplies for this project.

REFERENCES

Crouse C.A. and Schumm J. (1995) Investigation of species specificity using nine PCR-based human STR systems. J. Forensic Sci. 40:952-56.

Fitzgerald M.C. and Smith L.M. (1995) Mass spectrometry of nucleic acids: the promise of matrix-assisted laser desorption-ionization (MALDI) mass spectrometry. Ann. Rev. Biophys. Biomol. Struct. 24:117-40.

Fregeau C.J. and Fourney R.M. (1993) DNA typing with fluorescently tagged short tandem repeats: a sensitive and accurate approach to human identification. Biotechniques 15:100-119.

Huang N.E., Schumm J. and Budowle B. (1995) Chinese population data on three tetrameric short tandem repeat loci--HUMTHO1, TPOX, and CSF1PO--derived using multiplex PCR and manual typing. Forensic Sci. Intl. 71:131-136.

Kimpton C., Gill P., D'Aloja E., Andersen J.F., Bar W., Holgersson S., Jacobsen S., Johnsson V., Kloosterman A.D., Lareu M.V., et al. (1993) Report on the second EDNAP collaborative STR exercise. European DNA Profiling Group. Forensic Sci. Intl. 71:137-152.

Lecchi P., Le H.M. and Pannell L.K. (1995) 6-Aza-2-thiothymine: a matrix for MALDI spectra of oligonucleotides. Nucl. Acids Res. 23:1276-1277.

Lee S.B., Ma M., Worley J.M., Sprecher C., Lins A.M., Schumm J.W. and Mansfield E.S. (1995) Microwave extraction, rapid DNA quantitation and Fluorescent detection of amplified short tandem repeats. Proceedings of the Fifth International Symposium on Human Identification. Promega Corporation, 1994:137-146.

Liu Y.H., Bai J., Liang X.O., Lubman D.M. and Venta P.J. (1995) Use of a nitocellulose film substrate in matrix-assisted lase desorrption ionization mass spectrometry forr DNA Mapping and Screening. Anal. Chem. 67:3482-3490.

Liu Y.H., Bai J., Zhu Y., Liang X., Siemieniak D., Venta P.J. and Lubman D.M. (1995) Rapid screening of genetic polymorphisms using buccal cell DNA with detection by matrix-assisted laser desorption/ionization mass spectrometry. Rapid Comm. Mass Spec. 9:735-743.

Micka K.A., Sprecher C.J., Lins A.M., Comey C.T., Koons B.W., Crouse C., Endean D., Zold K., Lee S.B., Duda N., Ma M. and Schumm J.W. (1995) Validation of Multiplex Polymorphic STR Amplification Sets Developed for Personal Identification Applications. J. Forensic Sci. (forthcoming).

Siuzdak G. (1994) The emergence of mass spectrometry in biochemical research. Proc. Natl. Acad. Sci. U.S.A. 91:11290-11297. (Erratum (1995) Proc. Natl. Acad. Sci. U.S.A. 92:646).

Table 1. Characteristics of Dyes Evaluated

Table 2. Relative sensitivity of detection of differnet fluorescent dyes on three different instruments

* Dye Detection can be extended on the MD SI by energy transfer strategies (see Sensabaugh, this vol.)

Table 3. Sizing reproducibility in PRISM CTT as detected on the ABI 373

Table 4. Size Reproducibility of PRISM CTT with GeneScan 500 size standards as detected on the Hitachi FMBIO 100

* Sizes are rounded off to the nearest bp by the software.

Table 5. Size difference in PRISM THO1 (TAMRA) using GeneScan 500 (ROX) standards as detected on the ABI 373 and Hitachi FMBIO

Table 6. Evaluation of detection strategies

NA = Not applicaable because allelic ladders were used and no sizing was performed.

* The dollar sign ($) represents relative cost.

** Precision of sizing using fluroescent scanners can be improved if in-lane standards and dual color analysis is utilized (see Fig. 2, lanes 8-10).

Figure 1. Detection of CTT TAMRA (PRISM ABD) and ROX (red color) GeneScan size standards on the Hitachi FMBIO.

Dual color detection of ROX GeneScan500 size standards (lanes 1, 3, 5, 7, 9, 11 and 13) and TAMRA (ABD PRISM) CTT products (lanes 2, 4, 6, 8 and 10) and TAMRA matrix standard (lane 12 ) on the Hitachi FMBIO 100. Laser scanning was performed using 80% laser power. Scans were performed using two different filters: one at 585nm and one at 605nm. The images were then overlaid using dual-color analysis on the Hitachi FMBIO.

Dual color analysis of Fluorescein labeled CTTV PCR products from 5, 10 and 25 ng of K562 DNA (GenePrint™ Fluorescent STR kit-Promega) and ROX labeled GeneScan 500 size standards in the same lanes (Lanes 8-10). CTTV products from 5, 10 and 25 ng K562 DNA alone in lanes 5-7. CTTV allelic ladders in lanes 1, 3 and 4. FFFL allelic ladder in lane 2 (LPL alleles not shown in figure). Laser scanning was performed using 80% laser power. Two scans were performed using different filters: 505nm and 605nm. Images were overlaid using dual-color analysis on the Hitachi FMBIO.

Figure 3. The Technique of MALDITOF-MS

Go to proceedings home page