Substitution of the fish meal by the earthworm and maggot meal in the feed of Nile tilapia Oreochromis niloticus reared in freshwater

The present experiment was conducted to evaluate the efficacy of dietary protein from maggot meal and earthworm meal to replace fish meal protein in Oreochromis niloticus juvenile feed. The fish used for this experiment come from the experience of the larval phase during which fish were fed with the formulated diets [fish diet (FD), maggot diet (MD) and earthworm diet (ED)] content 40% crude protein and the commercial diet (CD) content 34.5% crude protein. At the end of this larval phase, the average weights were 0.75 ± 1.93, 0.71 ± 3.55, 0.55 ± 2.52 and 0.62 ± 2.52 g for FD, MD, ED and CD respectively. These different weights constitute the initials weights for this experience. In this experimental, the formulated diets (FD, MD and ED) content 30% crude protein and 29.5% crude protein for CD. Fish were fed four times daily to triplicate groups at 10.7% body weight for consecutive 90 days. After these days, fish fed with FD gave significantly higher average daily gain (ADG, 0.3 ± 0.03 g.day -1 ) than MD (0.22 ± 0.18 g.day -1 ). Fish fed with ED had the lowest ADG (0.14 ± 0.12 g.day -1 ). The specific growth rate (SGR) were higher in fish group fed with FD, MD and CD and shows no significant difference (p > 0.05) compared to ED. Feed conversion ratio (FCR), protein efficiency ratio (PER) and survival rate (SR) show no significant difference (p > 0.05) in all the test. The production cost of 1 kg of fish (PC) and the carcass chemicals composition were evaluated. In conclusion, these results of this study indicate the possibility of completely using maggot meal as a source protein in the diet of O. niloticus juvenile to increase the growth of fish and reduce the cost of 1 kg of fish produced.


INTRODUCTION
Fishmeal is so far the main source of protein for composing fish feed.However, fishmeal is increasingly expensive (current prices on the market, 300 to 700 F.CFA/kg) and fish feed represents more than 50% *Corresponding author.E-mail: aridian181@gmail.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License of the cost of production on farms (FAO, 2014;Coyle et al., 2004).Its inaccessibility is a source of abandonment of fish farming activity by small-scale fish farmers, most of them in developing countries.Nevertheless, aquaculture has become the fastest growing food industry in the world.That is why, the introduction of others protein sources as alternatives to replace fishmeal in fish feed is necessary to enhance the efficient production of aquaculture (Suarez et al., 2013;Katya et al., 2014).Numerous studies have been conducted to replace or reduce the level of inclusion of fishmeal and to identify promising alternative sources of protein in aquaculture feeds.These protein sources include plant proteins (Koumi et al., 2009;Bamba, 2017) and animal protein sources (Achi et al., 2017).However, less attention has been paid to the use of unconventional protein sources, which are promising feeds such as insect larva (maggot) meal and earthworm meal in fish feed formulation.In recent years, considerable research has shed light on the effectiveness of these ingredients in the diet of several aquaculture species, including tilapia, Oreochromis niloticus, Lates calcarifer and Litopenaeus vannamei (Ogunji et al., 2008;Katya et al., 2017;Cummins et al., 2017).The crude protein, crude lipid, and essential amino acid content of insect larval flours (maggot) resemble that of fishmeal.The crude protein content of maggots is around 40 to 64% (John, 2015).Earthworms have similar amino acid profiles to fish (Dedeke et al., 2010) and have been used as additional protein in fish feed (Sogbesan et al., 2007;Monebi and Ugwumba, 2013).However, recent research has shown that growth stimulation in some cases is observed at less than 50% substitution for earthworms (Sogbesan and Madu, 2008) and maggots (Ezewudo et al., 2015).
According to these same authors, the total substitution of fishmeal by these protein sources has not yet given satisfactory results in terms of growth performance of fish.Therefore, it is important in this study to evaluate growth, feed use, economic value, and carcass composition of juvenile tilapia O. niloticus fed with feeds formulated with maggot and earthworm proteins without the addition of fishmeal.

Description of the study area
The place of the experiments is the Blondey fish farm, 25 km from Abidjan (Ivory Coast) in the sub-prefecture of Azaguié (Ivory Coast).It is located between 5° and 6° North Latitude and between 4° and 5° West longitude.This farm has an area of three hectares whose feed water comes from a 3 ha dam also (Figure 1).According to Eldin (1971), this region is home to a warm equatorial climate composed of four seasons.First there is a succession of a great rainy season (from April to July) and a short dry season (from August to September).Then, we have a small rainy season (October and November) and finally a long dry season that runs from December to March.Concerning the hydrographic network, this zone abounds several rivers and in particular a river (Comoé).Examples include the Agnéby River, secondary courses (Mé, Nieky, and Betté) that are in flood during the long rainy season (Girard et al., 1971).

Experimental condition and fish feeding
The fish used for this experiment come from the experience of the  The commercial diet used as the reference at the same larval stage contained 34.5% crude protein.At the end of this larval phase, 4800 fish of averaging weight 0.75 ± 1.93, 0.70 ± 3.55, 0.55 ± 2.52, and 0.62 ± 2.52 g respectively for FD, MD, ED and CD were randomly distributed in 12 hapas for this present experience.The stocking density used was 400 juveniles per hapa (2 × 1 × 0.75 cm 3 ).The feeding experiment was for a period of 90 days.Three hapas installed into pond were then randomly assigned to each of the four experimental diets.Fish were fed the experimental diets four times daily (08:00, 11:00; 14:00 and 17:00 h) at 10% of wet body weight/day at the beginning and 7% of wet body weight/day at the end of the feeding trial for 1 month.At 1 month intervals, 25% of the fish population in each hapa were randomly sampled, batch weighed.The average weight of the fish sampled in each hapa was determined and the amount of feed provided to the fish was adjusted accordingly.Wet weight was measured on an electronic digital balance SARTORIUS L 6200 S (accuracy of ± 0.001 mg).At the end of the feeding period, all experimental hapas were emptied and fish in each hapas counted to determine fish survival.Additionally, ninety (90) fish were randomly sampled per diet (thirty fish per hapa) to evaluate the chemical composition of fish body carcass.

Analytical methods
The feed ingredients, experimental diets and fish samples were analyzed according to AOAC (1990) for moisture, crude protein, crude lipid, crude fiber, nitrogen free extract (NFE) and ash.The gross energy contents of the diets and fish samples were calculated using factors of 22.22, 38.9 and 17.2 kJ.g -1 of protein, lipid, and nitrogen free extract, respectively (Luquet and Moreau, 1989).

Measurement of growth performance, feeds utilization parameters and economic values
Weight

Water quality parameters
Water quality parameters were monitored during rearing period.Water temperature, dissolved oxygen, and pH were measured daily 08:00 h using YSI 6920 V2.Nitrate, nitrite, ammonium and phosphorus were measured once twice in month using HACH DR/2000 spectrophotometer by the method of Golterman et al. (1978).The mean data of physicochemical parameters of water measured in the hapas are shown in Table 3.

Statistical analysis
Results were presented as mean ± standard deviation (SD) for three replicates.The statistical analyses were carried out using one-way analysis of variance (ANOVA).The Tukey's multiple range test and Duncan's multiple-range test were used to compare

Physicochemical parameters of water
Water quality characteristics monitored throughout the study period are summarized in Table 3.There were no significant differences (p > 0.05) in the water temperature, pH and dissolved oxygen among the treatment during the whole experimental period.The water temperature ranged from 28.8 ± 0.5 (CD) to 29.4 ± 0.8°C (ED), pH from 6.5 ± 0.04 (ED) to 6.9 ± 0.05 (CD).Dissolved oxygen ranged from 7.14 ± 0.32 (ED) to 7.77 ± 0.37 mg.L-1 (CD).Nitrate nitrogen (NO-3) values were 0.40 ± 0.02 (CD), 0.42 ± 0.04 (MD) and 0.46 ± 0.01 mg.L-1 (FD).These highest values of nitrate nitrogen were not significantly different (p > 0.05) in the water content of fish fed with FD, MD and CD.The water containing the fish fed with ED obtained the low values (0.31 ± 0.03 mg.L -1 ) of nitrate nitrogen.Nitrite nitrogen ranged from (NO -2 ) 0.04 ± 0.05 (CD) to 0.052 ± 0.03 mg.L -1 (FD), ammonium from (NH 4 + ) 0.04 ± 0.02 (MD) to 0.05 ± 0.01 mg.L -1 (FD, ED and CD).There were no significant differences (p > 0.05) in the water NO -2 and NH 4 + among the treatment during the whole experimental period.The highest values of phosphorus were obtained in the water content of fish fed with FD and CD.These ranged from (PO 3-4 ) 0.39 ± 0.01 (CD) to 0.42 ± 0.03 mg.L -1 (FD) and these values were not significantly different (p > 0.05).Therefore, the low values of phosphorus were obtained in the water content of fish fed with MD (0.32 ± 0.02 mg.L -1 ) and ED (0.33 ± 0.02 mg.L -1 ).

Nutrient profile of protein ingredients
The proximate compositions of fishmeal, maggot meal, earthworm meal, soybean meal, cotton meal, copra meal, corn flour and wheat bran meal used as the major protein ingredients in this study are shown in Table 4.The crude protein content was found to be the highest for fishmeal followed by soybean meal, cotton meal, earthworm meal and maggot meal, respectively.On the other hand, crude lipid was recorded to be the highest for maggot meal followed by earthworm meal, copra meal, fishmeal and soybean meal, respectively.Whereas, copra meal and cotton meal exhibited higher fiber content as compared to others ingredients meal.As for the ash concentration of the ingredients, the high values were observed in earthworm meal followed by fishmeal and maggot meal compared to other ingredients.

Growth performance, survival rate, feed utilization and production time
Significant effects (p < 0.05) of the dietary total replacement of fish meal with maggot meal and earthworm meal on the growth performance of O. niloticus juveniles were observed (Table 5).Use of fishmeal in the fish feed gave higher final weight (FW) and average daily gain (ADG) than maggot diet and commercial diet.These values were 27.64 ± 7.02 g and 0.3 ± 0.03 g.day -1 for FW and ADG respectively.Interestingly, O. niloticus juveniles fed with MD and CD showed no significant differences (p > 0.05) between the FW and ADG.These values ranged from 17.28 ± 3.56 (CD) to 20.82 ± 4.71 g (MD) for FW and ADG from 0.19 ± 0.12 (CD) to 0.22 ± 0.18 g.day -1 (MD).The lowest values (FW, ADG) were obtained in fish group fed on diet containing earthworm meal (13.38 ± 3.72 g and 0.14 ± 0.12 g.day -1 for FW and ADG, respectively).Survival rate (SR), feed conversion ratio (FCR) and protein efficiency ratio (PER) were similar for all diets.Use of earthworm meal as protein source increased the production time (PT) of kg fish produced (6.81 ± 0.28 days.kg - of fish produced) as compared to the others group.In contrast, the use of maggot meal as protein source in O. niloticus juvenile feed resulted in decrease PT (3.12 ± 0.09 days.kg - of fish produced).

Carcass chemical composition of O. niloticus juveniles
The chemical composition of O. niloticus larvae at the end of feeding experiments are shown in Table 6.No significant differences were found in the carcass moisture content and ash of fish fed different experimental diets.In contrast, crude protein content was significantly affected by experimental treatment (p < 0.05).The carcass protein content was higher in fish fed FD (49.43 ± 0.19%) followed by fish group fed MD and CD (48.4 ± 0.22 and 48.1 ± 0.20%, respectively).In contrast, the lowest values were obtained in the fish group fed with ED (45.75±0.17%).No significant differences (p > 0.05) were found in carcass lipid and gross energy content of fish from one diet to another.This values ranged from 17.34 ± 0.12 (CD) to 18.22 ± 0.15% (ED) for crude lipid and from 17.24 ± 0.11 (ED) to 17.83 ± 0.13 kJ.g -1 (FD) for gross energy.

Cost-benefit analysis
The data on the kilogram costs of the feeds used and the rates of reduction of these costs were evaluated (Table 7).
The costs per kilogram of feed (CF) were 309, 207, 214 and 300 F.CFA, respectively for FD, MD, ED and CD.Relatively to the cost linked to the total quantity of different feeds used to produce the kilogram of juveniles, the recorded values were 455, 334, 362, and 493 F.CFA, respectively for FD, MD, ED, and CD feeds.The analysis of financial profitability shows that the use of maggot meal and earthworm meal as a source of protein in the feed of O. niloticus juveniles resulted in a decrease in the cost of kg of feed (cost/kg of feed) as compared to the cost of kg of fishmeal-based feed and commercial diet.Regarding the production cost of kilogram (PC) of juveniles, it was lower in fish group fed with maggot diet followed by fish fed with earthworm diet.In addition, the use of maggot meal and earthworm meal as a source of protein in the feed of O. niloticus juveniles also helped reduce the production cost of kilogram of fish by 26.59% (MD) and 20.44% (ED) as compared to FD.This production cost per kilogram of fish also was reduced by 32.25% (MD) and 26.57% (ED) as compared to CD.

DISCUSSION
The average values of physicochemical water parameters recorded during the period of the experiment did not differ from one hapa to another.They are between 28.8 ± 0.5 and 29.4 ± 0.8°C for temperature and 6.5 ± 0.04 and 6.9 ± 0.05 for pH.These values corroborated the results of Bamba et al. (2014) in the same station which found 23.6 to 31.6°C for temperature and 6.32 to 9.11 for pH.These values were within the tolerant range of O. niloticus.As for the dissolved oxygen during the experiment, the values obtained are between 7.14 ± 0.32 and 7.77 ± 0.37 mg.L -1 . These mean values of dissolved oxygen in the hapa are consistent with the recommended limits for tilapia breeding.They can survive an oxygen concentration of 1.2 mg.L -1 (Coche, 1978).The highest values of nitrate nitrogen in the water containing the fish fed with FD, MD and CD could be explained by the actual use of dietary nitrogen by fish as reported by Chakraborty et al. (1992).Fishmeal contains a high level of phosphorus (NRC, 1993).The greater part of this phosphorus is present in the feed in the form of tri calcium phosphate.Because of this, it cannot be absorbed by certain species of fish, particularly Cichlidae (Yone and Tochima, 1979) and is rejected in the breeding environment.This would explain the high phosphorus content of pond water when the O. niloticus species is fed FD and CD feeds.Indeed, FD and CD have been formulated based on fishmeal.According to Boyd (1998), the nitrite concentration must be less than 0.3 mg.L -1 . The concentration of ammonium must be between 0.2 and 2 mg.L -1 and the concentration of phosphorus must vary between 0.005 and 0.2 mg.L -1 for good survival of fish.After 90 days of fish monitoring, juveniles fed with the fish diet obtained the higher final weight (FW) followed by fish fed with maggot diet when compared with those of other group.On the other hand, the growth of O. niloticus juvenile fed with earthworm diet is low when compared with that obtained in any breeding.However, the juveniles fed on the commercial diet registered similar growth with those fish fed MD.The similarities observed between SGR when the fish were fed FD and MD show that the feed formulated with maggot meal is of interest for aquaculture.Nearly, similar values of growth performance obtained in fish fed with MD and fish fed FD were in agreement with those obtained by Samuel and Nyambi (2013).For these authors, growth is proportional when fish meal is totally replaced by that of maggot.However, these obtained values are in contradiction with the results of Ezewudo et al. (2015).For these authors, the growth performance of fish is better when the fishmeal is replaced by maggot meal at a rate of 50% in a diet based on fishmeal.The low growth rates of juvenile fed with earthworm diet corroborated the results of Nandeesha et al. (1988).For these authors, in common carp without free access to abundant natural food resources, the replacement of fishmeal by dried earthworm meal (E.eugeniae) resulted in reduced growth.In addition, some authors indicate that the replacement of fishmeal with earthworm meal at an inclusion level of 50 to 75% is appropriate for optimal growth and optimum utilization of nutrients in fish (Monebi and Ugwumba, 2013).In the present study, the survival rate values obtained 85.33 ± 6.9% (ED) and 89.41 ± 4.7% (CD) were low.These survival rate values in this study are also inferior to those obtained (100%) by Ogunji et al. (2008) when maggot meal is used in fish feed.
The results of the current study showed that feed utilization (feed conversion ratio, protein efficiency ratio) was similar in all groups of juvenile, one diet to another.However, feed conversion ratio (FCR) was the lowest in fish fed FD and MD, indicating that they were most efficient in converting their food to body growth.Protein efficiency ratio (PER) were the highest in fish fed fish meal and maggot meal diet, an indication of good protein digestibility and bioavailability for optimum body protein increase and growth.In this study, the values of FCR and PER observed with fish fed ED show no significant difference with those obtained in all livestock.These results show that feed based-earthworm meal was also used by these fish.These results agree with those obtained by Dedeke et al. (2013) who concluded that fish meal can be substituted with earthworm meal up to 25% in the diet of Clarias gariepinus fry without adverse effects on growth and nutrient utilization.These results corroborated with those of Ali et al. (2015).These authors indicate that the used maggot meal as ingredients in O. niloticus feed did not affect the palatability of the diets.The crude carcass lipids are not different in all fish groups.In addition, the crude protein content of the carcasses is low in fish fed with earthworm diet, unlike other groups.This indicates that the juvenile of O. niloticus is effectively used the crude lipid supplied by maggot meal and fishmeal in the diets.These results also indicate that the formulated feed based on the earthworm meal would not be efficiently used for the growth of the O. niloticus juvenile, but increased the lipid deposits.In this study, the cost (cost/kg of feed) of fish diet as those of CD were overprice.Fishmeal and fish oil is overprice in international market (FAO, 2014).The production of maggot meal and earthworm meal is less costly resulting in the reduction of the cost/kg of feed formulated and the cost/kg of fish produced.These results agree with those obtained by Ali et al. (2015) who concluded that 100% maggot meal can be included in the diet of O. niloticus nilotcus to reduce cost and maximize profit.However, the time required to produce 1 kg of fish increased in fish fed with ED.It is 6.81 ± 0.28 days.kg - of fish produced for ED.The increase in production time (PT) observed with the ED would be a consequence of the reduction in growth observed.

Conclusion
The present study concludes that maggot meal positively improved growth performance and feed efficiency of O. niloticus juveniles as well.It reduced the cost of kg of fish produced and the time/kg of fish produced.The earthworm meal slows down the growth of the juveniles and increases the production time of kg of fish produced.This study revealed that maggot meal can completely replace fish meal in the diets of O. niloticus juvenile without adverse effects on fish growth and carcass composition.And most importantly, reduce the cost of producing 1 kg of fish and promote semi-intensive and intensive fish farming in developing countries.

Figure 1 .
Figure 1.Map of the study area.
the mean of three readings ± Standard deviation.Means has the different letters in the same row are significantly different at p < 0.05.NO - 2, = Nitrite, NO - 3 = Nitrate, PO 3- 4 = Phosphorus, NH4 + = Ammonium.larval phase which lasted 30 days at the Blondey Aquaculture Station (5°6, N, 4°5, W), Ivory Coast.During this larval phase, the fish received feed containing 40% crude protein.These are the feed based on maggot meal, earthworm meal and fishmeal (control diet).

Table 1 .
Proportion (%) of ingredients used in formulated diets (FD, MD and ED).

Table 2 .
Proximate composition (% Dry matter) and cost of experimental diets.

Table 3 .
Physicochemical parameters of water in pond.

Table 5 .
Growth performance, survival rate, feed conversion ratio, protein efficiency ratio, production time/kg of fish produced.Each value is the mean of three readings ± Standard deviation.Means has the different letters in the same row are significantly different at p < 0.05. c
a Each value is the mean of three readings ± Standard deviation.Means has the different letters in the same row are significantly different at p < 0.05.

Table 7 .
Cost-benefit analysis of Oreochromis niloticus juveniles.Initial number of fish, FNF: Final number of fish, IB: Initial biomass, FB: Final biomass, NB: Net biomass, QFU: Quantity of feed used, CF: Cost of 1 kg of feed, CFU: Cost of feed used, PC: Production cost of 1 kg of fish, R×R CF/FD: Reduction rate of CF compared to fish diet (FD), R×R PC/FD: Reduction rate of PC compared to fish diet (FD), RxR CF/CD: Reduction rate of CF compared to commercial diet (CD), R×R PC/CD: Reduction rate of PC compared to commercial diet (CD).Price in CFA pound: 100 CFA= 0.18 $ based on 2017 exchange prices in Ivory Coast.-= Absents values. INF: