Community Research and Development Information Service - CORDIS

QTL for nematode resistance in the Sarda x Lacaune population

For the backcross ewes, the basic measure of resistance to nematodes was faecal egg count (FEC) under natural conditions of infection. Out of the very dry season in July-August when grass was missing, all the ewes grazed every day some hours on irrigated annual pastures of ryegrass and berseem clover. Supplementary feed was freely offered in each of the 4 opened houses, composed of alfalfa hay, maize silage and concentrates.

In Sardinia, the most common internal parasites are gastrointestinal nematodes and Protostrongyles in flocks using communal pastures.

Individual samples were taken in appropriate times along the year, on 8 occasions (sampling 1 to 8). FEC were measured before first calving (sampling 1) and seven times during their first four lactations: from June 2000 to June 2003. At each sampling date all individuals were sampled in 4 consecutive days. Samples were processed at the Veterinary Faculty in Sassari the same day by floatation in saturated salt solution in a McMaster slide (70ml for 5g of faeces or the same proportion if less than 5 g were available) and the eggs counted, those of Nematodirus genus separately. Presence of the tapeworm Moniezia was noted. To identify the genus distribution, common faecal cultures were done, one per sampling day for each of the seven first sampling, with the remaining faeces having more than 100 eggs per gram to get higher number of L3 for genus identification.

Average FEC varied highly from one sampling to the other, with particularly low values (and high proportion of zero values) in seventh and eight sampling series, and highest values of FEC for September sampling (sampling series 3 and 5).As the proportion of ewes positive for Nematodirus or Moniezia were under 3%, only the total of the gastrointestinal nematode species were analysed. FEC were transformed to lbpFEC, using the Bosseray Plommet transformation: Ln[(FEC+7)/Ln(FEC+7)].The log transformation achieved better adjustment to Gaussian distribution than row data although distribution of the two last FEC samplings showed high percentage of zero values.

Nine FEC traits were defined: FEC1 to FEC8, corresponding to eight single traits for each sampling date, and repeated FEC measures from first to sixth sampling. The best model for each FEC trait was defined according to significance of environmental effects using general lineal models (proc GLM, SAS®) for FEC1 to FEC6 continuous measures. Given the high proportion of null values for FEC7 and FEC8, the latter variables were considered as binary and analysed with logistic regression linear models (proc logistic, SAS®). Repeated continuous FEC1 to FEC6 measures were analysed using mixed model (proc MIXED, SAS®).

Fixed effects included in models were age, physiological status, and weight within each sampling. Additionally, a group effect (4 to 5 levels) was defined to account for the different physical groups of ewes defined after each lambing for management purpose. All four environmental effects were categorised into 3 to 6 levels within sampling date.

At the genome wise significance threshold with a suggestive linkage (p=0.038), a total of 20 QTLs were found for the 9 FEC traits. Out of these results, five QTL with a high confidence level (several traits; genome wise significance threshold; position and LRTmax profiles similar across traits) were detected on OAR2 (closest markers LSCV022), OAR3 (closest markers BMS0772, MRM154 and INRA131), OAR6 (closest markers INRA133 and BMS0360), OAR12 (closest markers HUJ625), OAR13 (closest markers BMC1222 and ILST059). Effects of QTL varied from 0.10 to1.50 standard deviation.

Related information


Rachel RUPP, (Researcher)
Tel.: +33-5-61285180
Fax: +33-5-61285353
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top