The effects of selecting growing pigs for a high growth rate and low backfat on sow characteristics

The effects of selecting for finisher traits on the development of breeding sows is not well understood (Lewis and Bunter 2013). It has been estimated by Hermesch et al. (2010) that for a 1 g increase in estimated breeding value (EBV) for average daily gain (ADG) of the growing pig, the sow weight (SWT) before farrowing increased on average by 0.32 kg. For a 1 mm decrease in backfat (BF) EBV of the growing pig, sow backfat (SBF) decreased on average by 1.56 mm. The aim of this study was to evaluate the effects of historical selection for a high ADG and low BF in growing pigs on the sows’ weight, fatness and haemoglobin levels (Hb). It was hypothesised that sows with a higher genetic potential for lean meat growth as growing pigs are both heavier and leaner in each parity.

In 2012 and 2013 data were collected on 171 Landrace and 216 Large White sows with 581 litters. Sow weight, SBF, and Hb were measured ~5 days before farrowing, upon transfer to the farrowing facilities in Parity 1 to 5+. There were between 164 (Parity 1) and 75 (Parity 5+) observations per parity. PIGBLUP was used to obtain EBVs for ADG and BF on the growing pig, using own and relatives’ records. Pedigree was available for 50 431 animals from 260 sires and 2055 dams born between 2004 and 2013. There were 563, 573 and 568 observations for SWT, SBF and Hb, with a mean (standard deviation; s.d.) of 266 (35.2) kg, 18.6 (3.6) mm, and 112 (12.8) g/L. The EBV for ADG ranged from –47.06 to 82.81 and the EBV for BF ranged from –2.63 to 0.19. The average SWT was 230, 253, 280, 297 and 308 kg from Parity 1 to Parity 5+. Observations that deviated more than 3 s.d. from the mean were excluded. Measurement date (65 levels), parity (five levels) and breed (only for SBF, two levels) were fitted as fixed effects in linear models for SWT, SBF and Hb using R (R v3.3.2, R Foundation, Vienna, Austria). In addition, litter weight and ADG EBV were fitted as linear covariables for SWT, ADG EBV was fitted as linear covariable for Hb, and BF EBV was fitted as linear covariable for SBF. Regressions within parity group were also estimated to evaluate the effect of EBV for ADG or BF on sow characteristics within parity.

Across parities, regression coefficients were only significant for BF EBV (Table 1). With a 1 mm decrease in BF EBV, sow fatness reduced by 0.6 mm. With a 1 g increase in ADG EBV in the growing pig, SWT increased by 0.32 kg and Hb decreased by 0.15 g/L in Parity 1. The effect of ADG EBV on SWT and Hb was not significantly different from zero in later parities. Lewis and Bunter (2013) found genetic correlations between weight at selection and weight across parities ranging from 0.54 in Parity 1 to 0.32 in Parity 5.

Table 1.  The regression coefficients (SE) of EBV for ADG or BF on sow characteristics across (additive model) and within (interaction model) parity, significant effects in bold (P < 0.05)

This study found that the EBV for ADG and BF affected SWT, SBF and Hb, although the magnitude of effects changed over parities. Until the first farrowing, gilts with higher genetic merit for growth had a higher SWT and lower Hb. Due to feed restrictions and other management strategies, sows might not be able to express their genetic potential for growth. The EBV for BF had a positive effect on SBF, but was lower than found by Hermesch et al. (2010). Feeding strategies might affect the observed relationship between growing animals and the mature sow herd. This relationship should be explored further for the development of selection strategies to improve growth of growing pigs while limiting mature size of sows.