35 25. Retribuciones al Consejo de Administración

After weaning, 32 piglets per batch were transported (for about 1.5 hours) to their new housing facilities. During transport, the 16 MS piglets were penned in 1 group, whereas the 16 FC piglets were penned in groups of 4 litter-mates. After arrival, MS and FC litters were equally divided over 2 adjacent rooms, where the 4 pens in 1 room were filled alternatingly with piglets from MS and FC litters. Each pen (2.7 x 3.6 m) housed 4 litter-mates and was bedded with 550 L of sawdust, 10 kg of straw, and 90 L of peat on a solid floor. A physically enriched post-weaning environment was provided, as experiencing a reduction in the level of environmental enrichment may be detrimental (see e.g. Bolhuis et al., 2006; Munsterhjelm et al., 2009), which would occur if the MS piglets would be housed under more conventional conditions. On a daily basis, soiled bedding was removed from the pens and 70 L of fresh sawdust and 1 kg of straw was added. Fresh peat was added on a weekly basis (45 L). A hessian sack, a rope and a chain with bolts were available as further enrichment. The same weaner feed that was provided in the late pre-weaning phase was offered ad libitum in a feeder with 4 eating places. Water was continuously available from a drinking nipple. Animal health was checked twice daily.

102

4.3.5. Measurements

Piglets were weighed on days -6, -1, 2, 5, and 13 post-weaning. Post-weaning feed intake at pen level was determined for days 0 to 2, 2 to 5, and 5 to 13. A faecal consistency score was given to all piglets on days -1, 2, 5, 7, and 9 post- weaning by means of visual inspection of the anal area of each piglet. A score of 0 was given if no faeces was present, score 1 if pasty faeces was present, and score 2 if watery faeces was present. Furthermore, piglets were subjected to a sugar absorption test using a protocol based on Turpin et al. (2016) on day 5 before weaning and day 5 after weaning. Piglets were fasted for 3 hours, during which they had access to drinking water but not to solid feed or milk. Before weaning, piglets were fasted by relocating the 4 litter-mates to an empty pen in a different unit. After weaning, piglets were fasted by removing the feeders from the pens. After the fasting period, each piglet was fixated in a vertical position with the nose pointing upward. Then, a nasogastric tube lubricated with paraffin oil was inserted in the stomach via the nose, after which physiological saline was administered. Thereafter, a sugar solution was administered. On day 5 before weaning, all piglets received 5 ml/kg body weight of 10% mannitol. On day 5 after weaning, half of the piglets (balanced for pre-weaning housing system, weight class, and sex) received 5 ml/kg body weight of 10% galactose and the other half received 5 ml/kg body weight 10% mannitol. Galactose was administered only after weaning because of the presence of galactose in milk. Thereafter, the piglet was placed back with its litter-mates. Twenty minutes after administration of the sugar solution, the piglet was fixated in a supine position by 2 persons and a 4 ml blood sample was collected in a heparin tube by puncture of the jugular vein. The sample was stored on ice (max. 1 hour) until further processing. When the 4 piglets of 1 litter had been blood sampled, they were placed back in their home pen (before weaning) or the feeder was placed back in their pen (after weaning). Samples were centrifuged for 10 minutes at 1,300 g at 4°C and plasma was stored at -20°C until analysis. Mannitol concentrations were determined using a commercial colorimetric assay kit (Abcam, ab155890). D-mannitol was converted to D-fructose by mannitol dehydrogenase in the presence of NAD to form NADH, which reduces a colourless probe to a chromogen with strong absorbance at 450 nm. Galactose concentrations were determined using a commercial assay kit (Abcam, ab83382) in which galactose is specifically oxidised generating a product that produces colour (OD570 nm). For both

4

Peri-weaning performance and intestinal function

103 sugars, concentrations were calculated using their corresponding standard curve, in accordance with the manufacturer’s instructions.

4.3.6. Statistical analysis

Data were analysed with SAS 9.2 (SAS Inst. Inc., Cary, NC). Residuals were checked for normality and variables were transformed with a square root or logarithm if needed. Body weight gain, sugar concentrations, and the number of days piglets had pasty or watery faeces were analysed with mixed models including pre-weaning housing system, weight class, housing x weight class interaction, and batch. Pen within housing treatment was included as a random effect, as measures on pigs from the same litter and pen may not be independent.

Feed intake was determined at pen level and was analysed with mixed models including pre-weaning housing system and batch. One FC pen in the second batch was excluded from analysis of feed intake because piglets in this pen had a poor start after weaning with severe diarrhoea and therefore received liquid feed supplementation. Body weight was analysed with a mixed model including pre-weaning housing system, day post-weaning, weight class, their 2- way and 3-way interactions, and batch. Pen (within housing treatment and batch) was included as a random effect. Repeated observations on the same animals were taken into account by including a repeated effect of piglet (within pen, housing treatment, and weight class). The proportion of piglets with pasty or watery faeces was analysed with generalised linear mixed models using a binomial distribution and logit link function, including pre-weaning housing system, weight class, housing x weight class interaction, and batch. Pen within housing treatment was included as a random effect. Average faecal consistency scores were analysed with mixed models including pre-weaning housing system, day post-weaning, weight class, their 2-way and 3-way interactions, and batch. Pen (within housing treatment and batch) was included as a random effect. Repeated observations on the same animals were taken into account by including a repeated effect of piglet (within pen, housing treatment, and weight class). Sex did not affect any of the variables tested and was therefore removed from the final models. Results are presented as means ± SEM. P-values below 0.05 were considered statistically significant.

104

4.4. Results