A new double choice model developed in laying hens reveals high preference for L-Alanine

S. Cho, J.M. Kim And E. Roura

54

It has been well established that nutrient deficiencies or imbalances increase the risk of feather pecking in layer hens (Kjaer and Bessei, 2013). Marginal imbalances may occur under commercial conditions due to individual variations in feeding behavior and digestive and metabolic efficiencies in nutrient utilization. These individual differences lead to nutrient specific appetites, which in turn, relate to the feather pecking habits. However, very little is known on nutrient preferences and appetites in commercial layer hens.

Furthermore, there is currently a lack of a practical methodology that allows assessing taste perception and preferences in chickens. Ninety-six 20 month-old laying hens were housed individually and assessed through two testing periods. Period one consisted of two days of training before being offered eight treatments defined as a factorial design with four potential taste active formulations (control with no tastants, 3.5% L-alanine (L-Ala), 10% calcium carbonate (Ca) or complete feed (positive control) where the active compounds were presented in two delivery matrices (starch or ground wheat –GW-). The second period involved the test of two additional tastants (2.3% NaCl and 6.8% MSG) administered using GW as the selected delivery matrix (based on results from period 1).

All the treatments were offered in a double choice set using two foil containers with one of the containers holding the control matrix (without tastant) and the other container one of the four treatments assigned following a complete block design.

Feed disappearance from each individual container was measured at 1, 2, 4, 8 and 24 hours. Preference values (test feed intake divided by total intake –i.e. the sum of the two containers) of each treatment were compared to the random choice value of 50% using SAS software. Differences were considered significant at P < 0.05.

The results of the first period showed that the GW was (P < 0.01) preferred as the delivery matrix over starch. In addition, after the first hour, a significant (P < 0.05) preference was found for L-Ala, whereas Ca addition was related to an aversion (consumption was significantly –P<0.05- lower than 50%).

These results are consistent with the findings of Baldwin et al. (2014) who reported an in vitro chicken taste receptor cell reporter system had a high affinity for L-Ala. The avoidance of the Ca in this experiment was related to a potential imbalance with P as calcium appetite in chickens is highly related to the ratio with phosphorus (P) (Wilkinson et al., 2014). We did not observe significant preferences for NaCl or MSG.

In conclusion, data supports the use of GW as a delivery system for taste active compounds in a double choice model in chickens. It appears that in our new model one hour is sufficient to detect nutrient specific appetites and that L-Ala and Ca have the potential to become reference treatments of preference and avoidance, respectively, for future evaluations.

References available on request.

Presented at the 2016 APSS