Cancer cell lines are used worldwide in biological research, and data interpretation depends on unambiguous attribution of the respective cell line to its original source. Short-tandem-repeat (STR) profiling (DNA fingerprinting) is the method of choice for this purpose; however, the genetic stability of cell lines under various experimental conditions is not well defined. We tested the effect of long-term culture, subcloning, and generation of drug-resistant subclones on fingerprinting profiles in four widely used leukemia cell lines. The DNA fingerprinting profile remained unaltered in two of them (U937 and K562) throughout 12 months in culture, and the vast majority of subclones derived therefrom by limiting dilution after long-term culture revealed the same profile, indicating a high degree of stability and clonotypic homogeneity. In contrast, two other cell lines (CCRF-CEM and Jurkat) showed marked alterations in DNA fingerprinting profiles during long-term culture. Limiting dilution subcloning revealed extensive clonotypic heterogeneity with subclones differing in up to eight STR loci from the parental culture. Similar heterogeneity was observed in subclones generated by selection culture for drug resistance where DNA fingerprinting proved useful in identifying possible resistance mechanisms. Thus, common tissue culture procedures may dramatically affect the fingerprinting profile of certain cell lines and thus render definition of their origin difficult.

Capillary ion-pair reversed-phase high-performance liquid chromatography (IP-RP-HPLC) was used to separate and purify DNA fragments amplified by the polymerase chain reaction (PCR) prior to their characterization by electrospray ionization mass spectrometry (ESI-MS). The investigation by ESI-MS of single- or double stranded species could be effortlessly selected by chromatography of the nucleic acids under either nondenaturing or denaturing conditions, which were realized by proper adjustment of the column temperature. ESI-MS detection sensitivity was improved by a factor of 10 upon replacement of 25 mM triethylammonium bicarbonate as ion-pair reagent by 25 mM butyldimethylammonium bicarbonate because of the applicability of higher acetonitrile concentrations to elute the DNA from the monolithic, poly(styrene/divinylbenzene)-based capillary columns. For fragments ranging in size from 67 to 84 base pairs, the mass accuracies and mass reproducibilities were typically better than 0.02 and 0.008%, respectively, which enabled the characterization and identification of the PCR products with high confidence. The hyphenated method was applied to the genotyping of polymorphic short tandem repeat (STR) loci from the human tyrosine hydroxylase gene (humTH01). The different alleles both in homo- and heterozygotes were identified on the basis of the masses of the single-stranded amplicons and were in full accordance with the alleles identified by conventional capillary electrophoretic sizing.

A multiplex PCR was designed for the loci D2S1338, D16S539, D18S51, TH01 and FGA using redesigned primers in order to reduce the lengths of the amplification products compared to the designs used in commercially available multiplex PCR kits, also including amelogenin. The new PCR primers were used to amplify highly degraded DNA from casework samples, which had shown no or only poor results for these loci in previous analyses with standard primer sets. The application of the new miniSTR-multiplex resulted in an increased overall typing success rate for degraded DNA samples. In a concordance study between the conventional and the newly designed primers, no genotype differences were revealed in 124 randomly selected individuals.

Unusually large variant alleles were observed in the short tandem repeat (STR) systems D3S1358 and D21S11, both of which are included in the international standard set of loci (ISSOL) and routinely typed in National DNA intelligence databases worldwide. The observed alleles fell within the size range of the adjacent STR marker, which could easily cause problems with respect to correct allele assignments for both loci concerned. We compared the amplification and potential interpretation with three different commercially available kits, which are frequently used in forensic work. PCR products were cloned and sequenced in order to determine the structure of these unusual allele variants and confirm their size and designation (D3S1358 allele 26, D21S11 allele 46). In the locus D21S11 we observed an as yet undescribed partial duplication of the constant region.

PCR primers for the locus D2S1338 were redesigned in order to reduce the length of the amplification product compared with the conventional design. The new amplification primers were applied to highly degraded casework samples, which gave no or only weak results for D2 in previous analyses. The application of the new primers resulted in an increased overall typing success rate. In a concordance study between the conventional and the newly designed primers, no genotype differences were revealed in 503 randomly selected individuals. This suggests robust amplification conditions with respect to mutation-based allele drop-out.