Summary In trout farmed in freshwater, the use of low phosphorus diet and other concurrent changes in food formulations could affect the normal growth of the skeleton and enhance the occurrence of vertebral abnormalities. Hence, it is important to refine diagnostic of malformations to quickly detect signs of P deficiency during production. Two practical diets consisting of P‐deficient (0.5% total P) and P‐sufficient (0.9% total P) were formulated. Experiments were conducted on all‐female juvenile triploid rainbow trout (initial mass ~60 g) until fish fed continuously with P‐sufficient diet reached <850 g (241 days; 12°C; 18:6 D:L). The choice of using practical diets as well as triploid trout were made in order to adapt to the context of production. Early P‐deficient and P‐sufficient trout were fed continuously (starting at day 0 for 241 days) with the respective diets while late P‐deficient trout only started feeding on P‐deficient diet at day 134 of the experiment. Growth performance were used to assess P deficiency and induced vertebral abnormalities were count and described using the classification suggested by Witten et al. () for Atlantic salmon. Our results showed that P deficient diet impaired vertebral mineralization and growth performance. The P deficient vertebrae showed increasing evidence with time of compression and reduction of intervertebral spaces (vertebrae showing pronounced biconcave end plates ‐ type 11 have evolved into platyspondyly; type 1, 2, 3 and 4). As P deficiency seems to affect all vertebrae along the vertebral column, examination of the caudal region alone could enable to increase the resolution of x‐rays and minimize time‐consuming efforts for analyses. The dimensions of rearing tanks might be responsible of high frequency of post‐cranial vertebrae precluding conclusion on the intensity of the effect of P deficiency.

Summary Extended phosphorus (P) deficiency has been linked to an increase of skeletal deformities in farmed salmonids. To date, X‐ray examination still is the most practical way to detect abnormalities that are not externally visible. The aim of this study was to develop a standardized method based on empirical measurements that provides an early radiograph‐based detection of deformities. Triploid all‐female rainbow trout (60.8 ± 1.6 g) were fed with phosphorus (P)‐deficient and P‐sufficient diets (total P = 0.5% and 0.92%, respectively). Experiments were carried out in 12 circular re‐circulating tanks (n = 140/tank; 8.5–52.7 kg m−3; 12 ± 0.3°C; 18L: 6D). All fish were x‐rayed at week 5, 15 and 24 to monitor vertebral abnormalities. Fish were assigned to a vertebral phenotype (normal, biconcave or compressed) when all vertebrae of the caudal region (V31‐44) showed the same type of abnormalities. At week 9, 18 and 27, ten fish per tank were randomly sampled according to their vertebral phenotype. Vertebrae V38–V39 were collected and a second X‐ray was performed to confirm the above‐mentioned phenotypes. Morphometrics (vertebral length and height; intervertebral distances and vertebral body angles) were measured directly on radiographs. For all parameters measured, comparisons were made between normal P‐sufficient fish and different scenarios of vertebral development (normal, restored, stable and aggravated) in P‐deficient fish. The two parameters showing the most sensitivity were the median intervertebral distance and dorsal/ventral vertebral endplate angles. When analysed together, these two parameters allow a quantitative distinction of the different vertebral phenotypes. Measuring dorsal/ventral vertebral endplate angles seems a promising approach to detect aberrant vertebral development at early stages as slight changes were observed in P‐deficient vertebrae that were assessed as normal by visual examination. Further studies including histological examinations and comparison with other radiological imaging techniques are required to confirm these results.

SummaryRainbow trout (Oncorhynchus mykiss) is an important production species as well as one of the most studied fish models, particularly regarding nutritional physiology. Due to negative environmental impacts linked with rainbow trout farm effluents, significant restrictions have been established in numerous regions to reduce dietary phosphorus (P) outputs. However, questions have arisen regarding the link between abnormal skeletal development and mineralization and insufficient dietary P availability during rapid fish growth (juvenile fish). Despite significant work to understand the dynamics of P-deficiency and the resulting impact on tissue mineralization, the extent of the early responses in rainbow trout fed low-P diets is not well described. The aim of this study was to explore the early-responses of scales, vertebrae and carcass P and ash in rainbow trout fed low-P vs sufficient-P controls. Two practical diets (sufficient: 0.45% available P and deficient: 0.29% available P) were fed over a 5-week period to triploid rainbow trout (initial mass 60.8 AE 1.6 g). Ash and P contents were used to assess mineral status of the different tissues. The relative loss of mineral and coefficient of variation were also calculated to compare the relative response and the inter-individual variability. After 4 weeks of P deprivation, no detectable effects were observed on growth performance, mortality or feed intake. However, as early as the second week onward, ash and P levels in scales and carcasses were significantly lower (p < 0.05) in fish fed a P-deficient diet compared to those fed a P-sufficient diet. At week 4, the reduction in mineralization reached 19.3 and 18.4% for P and ash contents in scales, respectively. By the same week, P and ash contents in carcasses reached 15.1 and 12.8%, respectively. Interestingly, the absence of significant mineral loss in vertebrae did not prevent the emergence of a high incidence of vertebral deformities in P-deficient fish (45.6 AE 11.0%) when compared to P-sufficient fish (1.5 AE 2.1%). The main vertebral deformities observed in Pdeficient trout were specific to impede bone matrix mineralization such as pronounced biconcave (35.5%) and undersized and widely spaced vertebral bodies (9.3%). For the scales compartment (ash and P status) and mineral parameter, the coefficients of variation showed a higher inter-individual variability in P-deficient vs P-sufficient fish, while no such effect was observed in carcass or vertebrae compartment. This study provides a useful comparison of various commonly-used indicators of bone mineral status as well as an approach to assess variability of fish response under low-dietary P level.

Prolonged dietary phosphorus (P) deficiency induces a generalized demineralization of bone tissues; however, the differential responsiveness of the diverse tissues to P deficiency remains unknown. Triploid all‐female juvenile rainbow trout (initial mass ~60 g) were fed two practical diets (0.26% and 0.52% digestible P for P‐deficient and P‐sufficient diets) until P‐sufficient fish reached 850 g (241 days; 12°C; 18:6 D:L). Early P‐deficient fish were fed from day 0 the low P‐diet, while late P‐deficient only started feeding at day 134. Growth performance was assessed monthly and carcass samplings were performed at day 0, 134 and 241 to assess mineralization characteristics of different tissues. Ash and content of phosphorus, calcium and magnesium of scales, operculae, ribs and vertebrae as well as the occurrence of abnormalities in ribs and vertebrae were assessed. The tissues most affected by P deficiency were: vertebrae > scales > opercula > ribs. P deficiency affects bone tissues differentially based on the physiological and biomechanical roles of the tissues. As scales were found to be a better indicator to predict abnormalities, understanding of scale metabolism could facilitate P status estimation in fish.

No abstract