C57BL/6J and C57BL/Wlds mice were obtained from Harlan UK and, where required were cross-bred to obtain F1 mice, heterozygous for Wlds. Ear punches were taken from mice between 1 and 4 months of age (N = 36 of confirmed genotype, N = 91 of unknown genotype). Tail tips were taken from transgenic rat lines Tg23 (N = 10) and Tg79 (N = 14). The investigator carrying out the QPCR screen was always blinded to the genotype of each tissue sample.

Genomic DNA was extracted using a modified proteinase K digestion/isopropanol precipitation protocol 21. Tissue was incubated in tail lysis buffer (100 mM Tris-HCl, pH 8.5, 5 mM EDTA, 0.2% SDS, 200 mM NaCl, 100 μg/ml Proteinase K). Samples were incubated overnight at 55°C. Lysed tissue was vortexed for 1 minute and spun at 15800 g for 10 minutes to remove hair and bone. The supernatant was decanted into a new tube and isopropanol was added (equal volume to the tail lysis buffer used earlier). The solutions were inverted until a white precipitate was seen. The solution was centrifuged and the pelleted DNA was washed with 70% ethanol and air dried before it was resuspended in an appropriate volume of 1 × TE buffer. For example, a 5 mm length of tail tip was digested and the DNA was finally resuspended in 50 μl of TE buffer and then stored at -20°C as a stock solution.

The primers and probes (TaqMan Probes) were designed using Primer Express software as described in the ABI Primer Express User Bulletin (P/N 4317594). Wld^s primers were based on Genbank entry AF260927 and recognise and amplify bases 810–885 of the Wld^s chimeric gene (Figure 1). The Wld^s probe was tagged with a FAM fluorophore and a TAMRA quencher. The product of the Wld^s reaction is a 75 bp amplicon. β-Tubulin primers were based on Genbank entry M28739. The probe was tagged with a VIC fluorophore and a TAMRA quencher. The product of the β-Tubulin reaction is an 81 bp amplicon. The sequences (5'-3') used were as follows:

Wld^s forward GGCAGTGACGCTCAGAAATTC

Wld^s reverse GTTCACCAGGTGGATGTTGCT

Wld^s probe TCTACGAGTCCGATGTGCTGTGGAGACA

β-Tubulin forward GCCAGAGTGGTGCAGGAAATA

β-Tubulin reverse TCACCACGTCCAGGACAGAGT

β-Tubulin probe CTGGGCAAAGGGCCACTACACAGAGG

A serial dilution of genomic DNA from wild-type C57Bl/6J mice was used as a template for PCR to test the efficiency of amplification of the two primer pairs. The genomic DNA was diluted 1 in 10 each time and correlation coefficients were obtained for each primer pair (Figure 2). PCR efficiency was calculated by plotting the Threshold cycle (Ct) as a function of Log10 concentration of the template used (Figure 2; x-axis plotted as -Log μg DNA. See Applied Biosystems website for user bulletin #2 22). The slope of the trend line produced is a function of PCR efficiency, with a slope of -3.32 indicating close to 100% efficiency 22. Standard curves suggested efficient amplification of both primer sets, as indicated by correlation coefficients and linear regression slopes. Primers against a region of Wld^s and β-tubulin gave correlation coefficients of 0.9989 and 0.9928 respectively, with slopes of -3.455 for β-tubulin and -3.3647 for Wld^s (Figure 2). The reactions were found to proceed similarly when primers were used both separately and together, indicating that the two separate primer pairs do not interact with each other. Once this had been demonstrated, genotyping reactions were reliably performed containing both sets of primers and probes in the same well.

The optimal primer concentrations used were 900 nM and the optimal probe concentration was found to be 250 nM. The PCR mix (TaqMan universal PCR master mix, no amperase UNG from ABI part # 4364341) was prepared according to the manufacturer's instructions. 1 μl of DNA was used at a concentration of 1:200 dilution (in TE buffer) of the original stock sample for each reaction. All reactions were carried out in triplicate on an ABI Prism 7000 sequence detection system using Applied Biosystems' standard thermal cycling parameters.

The product of a PCR for the Wld^s amplicon alone run on an ethidium bromide stained gel showed that bands from each genotype could not be distinguished from each other (data not shown). However, the 2^-ΔΔCT method (for more detail see Applied Biosystems website for user bulletin #2), using β-tubulin as an endogenous control and Wld^s as a calibrator, allowed determination of copy number. Amplification plots for wild-type, heterozygous-Wld^s and homozygous-Wld^s showed clear differences for the ΔC_T between each group (Figure 3 &4). The difference between β-tubulin and Wld^s amplicon product at the set threshold can be used to determine Wld^s genotype. A wild-type mouse gives a mean cycle difference of -0.40 (± 0.26 SD) with 95% confidence limits of -1.04 & +0.24. A mouse heterozygous for Wld^s gives a mean cycle difference of 1.07 (± 0.32 SD) with 95% confidence limits of +0.97& +1.17. A mouse homozygous for Wld^s gives mean cycle difference of 2.07 (± 0.30 SD) with 95% confidence limits of +01.995 & +2.158 (Figure 4).

Plotting the ΔC_t against animals of known genotype in a box and whisker plot (N = 36) allows the illustration of 95% confidence limits for each genotype (Figure 4). When plotting a scatter of the results for all of the animals of unknown genotype (N = 91) there was a clear banding into the 3 genotypes as demonstrated by the box and whisker plot of animals of known genotype (Figure 4). Only one sample fell outwith the 95% confidence limits for all of the plotted genotypes shown (Figure 4). In such circumstances, the genotyping for this particular sample should be considered invalid. The Wld^s status of a representative set of unknown genotype animals whose identity was determined by QPCR was successfully confirmed using breeding records, Southern blotting techniques and subsequent experiments where the neuroprotective phenotype was revealed by nerve lesion and the use of morphological techniques (data not shown).

Transgenic animals (mice and rats) have been generated which express the Wld^s chimeric gene 1318. These animals show a strong neuroprotective phenotype. As it has previously been shown that the Wld^s neuroprotective phenotype is dose-dependent (see above), it is therefore of interest to determine the number of copies which have been integrated into the genome of these transgenic animals.

We examined 2 rat models of Wld^s: the transgenic line 23 and transgenic line 79 13. Through the use of QPCR on genomic DNA (utilising the 2^-ΔΔCT method described above) it is possible to calculate the copy number of insertions into the rat genome. The results examining the number of inserts in two of the Wld^s transgenic lines can be seen in Figure 5. There was additional variability in the amplification process at higher copy number with QPCR, as previously described when using this technology 22. However, the data show that transgenic rat lines 23 and 79 contain 12.34 ± 2.05 (N = 10) and 19.49 ± 1.52 (N = 14) copies of the Wld^s gene respectively, significantly more than homozygous Wld^s mice, as would be expected in transgenic animals generated using non-targeted insertion. It is also important to note that the data do not give any indication of transgene orientation or functionality. Despite this, the increase Wld^s copy number in the transgenic rats tallies with previously reported increases in phenotypic strength compared to homozygous Wld^s mice 13.

It is especially important to measure Wld^S copy number and expression level before making conclusions about its effects when attempting to transfer protective benefits to models of disease. For instance, two reports suggest little or no mitigation of disease onset or progression after crossbreeding Wld^S mice with mouse models of amyotrophic lateral sclerosis (8,15); but it is presently unclear whether these reports fully accounted for the known, weak synaptic-protective effects of the Wld^S gene in heterozygous and aged mice (3,18).