Despite a growing interest in microalgae and cyanobacteria as potential sources of nutrients in aquafeeds, little
information is presently available on their nutritive value for carnivorous fish species. The aim of this study was
to evaluate chemical composition and nutrient digestibility of a panel of microalgae and cyanobacteria dried
biomasses (MACB), using rainbow trout (Oncorhynchus mykiss W.) as a fish model. Nine test diets were obtained
by mixing 80 parts of a reference diet, added with 20 g/kg of acid insoluble ash as an indigestible marker, to 20
parts of each of the following dried whole-cell biomass: Arthrospira platensis, Nostoc sphaeroides, two strains of
Chlorella sorokiniana, Nannochloropsis oceanica, Tisochrysis lutea, Phaeodactylum tricornutum, Porphyridium purpureum
and Tetraselmis suecica. The digestibility measurements were conducted with rainbow trout (52.4 ± 1.5 g)
kept in six tank units each including three 60-L vessels singularly stocked with 12 fish and fitted with a settling
column for faecal recovery. Per each diet, faeces were collected over three independent 10-day periods. Apparent
digestibility coefficients (ADCs) of dry matter, crude protein (CP), organic matter and gross energy (GE) of single
MACB were calculated by difference relative to those of the reference diet. The MACBs had heterogeneous
chemical composition (CP, from 20 to 69%; Lipid, 5–27%; GE, 12.5-–22.6 MJ/kg dry matter basis) reflecting
their overall biodiversity. Most of them can be considered as virtually good sources of minerals and trace elements
and exhibit an essential amino acid profile comparable or even better than that of soybean meal commonly
used in fish feeds with P. purpureum showing the best protein profile. The digestibility results obtained with
rainbow trout allowed ranking the MACBs into two major groups. A first one, including C. sorokiniana,
N. oceanica and T. suecica, resulted in markedly lower (P < 0.05) crude protein and energy ADC (64–73%;
51–59%, respectively) compared to a second group including P. purpureum, T. lutea and cyanobacteria (CP-ADC,
83–88%; GE-ADC, 74–90%) while P. tricornutum resulted in intermediate values. Overall, the present study
confirms the consistently reported role of cell-wall structure/composition in affecting accessibility of nutrients to
digestive enzyme. Based on the overall outcomes, only T. lutea and cyanobacteria actually meet the requirements
for being used as protein sources in aquafeeds provided their mass production becomes more feasible and costeffective,
hence attractive for the feed-mill industry in the near future.
