The periparturient relaxation of immunity (PPRI) to parasites in mammals is sensitive to both metabolisable protein (MP) supply and animal genotype (different reproductive outputs). We tested the hypothesis that the sensitivity of PPRI to MP scarcity would not differ between different levels of reproductive output when nutrient intake is adjusted for associated differences in MP demand; this hypothesis assumes that PPRI has a nutritional basis only. Scottish Blackface (BF) and the more productive Mule (MU) ewes were infected with the abomasal parasite Teladorsagia circumcincta, and from day− 21 to day32 (day0 is parturition), they were fed restrictedly at either 0·8 (low protein (LP)) or 1·3 (high protein (HP)) times their breed-specific estimated MP requirement (n 18 for each breed–feeding treatment combination). During late pregnancy, LP feeding reduced ewe body weight gain in both breeds, tended to increase faecal egg count (FEC), but it did not affect plasma pepsinogen. During lactation, LP feeding reduced litter growth rate and ewe plasma urea and plasma albumin concentrations compared with HP feeding in both breeds. However, breed and feeding treatment interacted for ewe FEC, worm egg excretion and plasma pepsinogen, which were higher for the LP-MU ewes compared with the HP-MU and BF ewes. The lower degree of PPRI of the BF ewes during lactation compared with the MU ewes at a similar degree of MP scarcity suggests that the effect of reproductive output on nutritional sensitivity of PPRI cannot be explained by associated differences in nutrient demand only.

The aim of this work was to test the robustness of the 0.68 estimate of the efficiency of conversion of metabolisable protein into true milk protein (Agriculture and Food Research Council (AFRC), 1993) for protein-limiting diets and to determine whether a different value is appropriate for practical rationing. Seventy-two multiparous cows were blocked on the basis of milk energy output per unit of dry matter intake (DMI), and allocated at random to one of four treatments. Treatments supplied metabolisable energy (ME) at a fixed level to individuals within a block, but varied metabolisable protein (MP) supply from 25% below the estimated requirements, through −12.5% and +12.5% up to 25% above requirements for the average performance of animals within blocks at the start of the study. Cows were offered diets to meet their predicted ME requirements for each 3-week period with measurements performed in the last week of each period. Milk protein output was regressed against the estimated MP available for production for each cow and the efficiency of conversion of MP into milk true protein was calculated, assuming a maintenance requirement according to the MP system. The efficiency of conversion of MP into milk true protein decreased with the increasing supply of MP from 0.77 to 0.50. Using an iterative approach to determine the best fit of the data when supply matched requirement resulted in a range of efficiency values between 0.62 and 0.64 g of true milk protein per g of MP.