Selection potential for feed efficiency in farmed salmonids

Recording feed intake To be able to select directly for feed effi ciency, feed intake of individual fi sh should be recorded. Until recently, diffi culties in measuring individual feed intake on a large scale have prevented accurate genetic evaluation of feed utilization traits in farmed fish. To solve the recording challenges, we have applied the X-ray method to measure feed intake and feed effi ciency of thousands of individuals in pedigreed populations of farmed rainbow trout (Oncorhynchus mykiss) and European whitefi sh (Coregonus lavaretus).
To measure feed intake of individual fi sh using the X-ray method, all fi sh held in a tank are fi rst fed with feed containing small radio-opaque glass beads. Thereafter, the fi sh are X-rayed. The number of glass beads consumed can be counted from the X-ray fi lms. Because the glass bead content of the feed is known, it is possible to calculate the quantity of feed that each fi sh consumed on a specifi c meal.

The studies on rainbow trout and European whitefi sh3 show that direct selection for feed effi ciency is possible in farmed fi sh. Yet, genetic improvement of feed effi ciency is expected to be about three (rainbow trout) to eight-fold (European Whitefi sh) slower compared to the improvement of growth rate. This results because feed effi ciency seems to display modest amount of genetic variation for selection. For instance, in European whitefi sh, only 6 percent of the phenotypic variation in feed effi - ciency was explained by genetic effects. Simultaneously, feed intake and weight gain, the two component traits of feed effi ciency, exhibited moderate genetic variation (23-26% of variation explained by genetics).
Our studies have been conducted during three week to three month long trial periods at an exponential growth phase of the fi sh. During this time, most feed consumed is directed to growth, and there is only little variation in nutrient diversion to other body functions. It is possible that if feed effi ciency could be recorded during longer time periods (e.g. across the whole fi sh life), more genetic variation for feed effi ciency would be revealed.

Recording of individual feed intake from thousands of fi sh is challenging. Thus, it is of interest to assess whether more easily recorded traits that are genetically related to feed effi ciency could be used to indirectly select for feed effi ciency. Our study showed that feed efficiency can be indirectly improved by selecting on growth rate. Rapid growth is genetically related to improved feed efficiency.
This is good news because all fi sh breeding programmes select for rapid growth anyway. Moreover, selection against body lipid percentage can be used to indirectly select for lower feed intake, and thus to improve feed efficiency. This is logical because extensive feed intake is related to excess lipid deposition, and lipid deposition is energetically more expensive than deposition of muscle. Fish breeding programmes often control lipid deposition by selection to maintain high product quality. This has an additional benefit of maintaining high feed efficiency.

Feed effi ciency is economically a fundamental trait, and thus even small improvements are economically important. As breeding proceeds, the small genetic changes in feed effi ciency accumulate from generation to generation. This leads to moderate feed effi ciency changes in a longer term. For instance, during the last four generations of selection in the Finnish national rainbow trout breeding programme, growth rate has increased by ~28%5. Feed effi ciency is expected to have increased simultaneously by 8% as a correlated genetic response. When majority of fi sh farmers use the improved fi sh material, the practical impact of the selection work is extensive. Accordingly, all efforts to increase feed effi ciency will be of fundamental importance.