MATERIALS AND METHODS

Experimental setup: Evaluation of cricket was performed in two experiments. In experiment 1, three forms of cricket were prepared, i.e., C, CER and CCE. These were subjected to chemical composition determination (proximate composition, amino acid composition and fatty acid profile) and in vitro rumen fermentation incubation as individual substrates. In experiment 2, C and CER were included in concentrate rations at different proportions to substitute soybean meal, either partially or totally. The concentrates were then evaluated in vitro for their rumen fermentation and digestibility characteristics. These experiments were conducted from March-June, 2017 at Laboratory of Dairy Nutrition, Faculty of Animal Science, Bogor Agricultural University, Indonesia.

Cricket preparation and treatment: To obtain whole cricket (C) meal, discarded cricket from a commercial cricket farm in Bogor, Indonesia, was oven-dried at 50°C for 48 h and ground by a hammer mill to pass a 1 mm screen. The CER was obtained by manually removing head, legs and wings of the oven-dried cricket. The CCE was obtained through delipidation, extraction and precipitation processes of cricket using chemical solvents according to procedures from Agboola et al.⁹ and De Souza et al.¹⁰. In brief, delipidation was performed by solubilizing cricket meal into n-hexane (1:4 w/v) and homogenized for 30 min. Sample was centrifuged thereafter at 5,000 rpm, maintenance temperature was 4°C for 10 min and supernatant was discarded. Pellet was solubilized by using NaOH 2 N for 60 min to shift pH from neutral to alkali (pH around 12). Precipitation was conducted at isoelectric pH by adding HCl 2 N and kept for 30 min. Precipitate was centrifuged twice at 5,000 rpm, maintenance temperature was 4°C for 10 min and dried at 50°C for 8 h to obtain CCE.

Chemical composition determination: Samples were determined for their proximate composition, i.e. organic matter (OM), ash, crude protein (CP), ether extract (EE) and crude fiber (CF) according to AOAC¹¹. Total digestible nutrient (TDN) content was estimated from proximate composition by using an equation from Wardeh¹². Procedure for chitin determination was based on No et al.¹³. Analyses of amino acid and fatty acid profiles in samples were conducted by employing high performance liquid chromatography and gas chromatography (Shimadzu Corp., Kyoto, Japan), respectively, according to AOAC¹¹. Chemical composition determination for each parameter was performed in duplicate.

In vitro rumen fermentation: Samples were incubated in vitro with buffered rumen fluid by following the procedure of Tilley and Terry¹⁴. An amount of 0.5 g sample was put into a glass tube and mixed with 10 mL rumen fluid and 40 mL McDougall buffer. The buffer was composed of (per liter solution) 9.8 g NaHCO₃, 2.44 g Na₂HPO₄, 0.57 g KCl, 0.47 NaCl, 0.12 g MgSO₄.7H₂O and 0.16 g CaCl₂.2H₂O. Rumen fluid was taken from three Ettawa crossbred goats approximately 4 h after morning feeding by employing a stomach-tube method. The tube was continuously flushed with CO₂ to ensure anaerobic environment, closed with a ventilated rubber and put into a shaking water bath (Memmert GmbH+Co. KG, Schwabach, Germany) at 39°C to start the incubation. After 48 h incubation period, the ventilated rubber was opened and added with 3 drops of HgCl₂ to terminate the first incubation step. The tube was centrifuged at 4,000 rpm for 10 min to separate residue and supernatant. Supernatant was discarded and the residue was further incubated for 48 h with 50 mL pepsin-HCl 0.2%. Residue obtained from the second in vitro incubation step was filtered with Whatman paper No. 41 (Sigma-Aldrich, St. Louis, MO, USA) under vacuum and subjected to DM and OM determination. Delta between initial amounts of DM and OM and their corresponding residues were used for, corrected with blank, calculating in vitro dry matter digestibility (IVDMD) and in vitro organic matter digestibility (IVOMD), respectively. Concentrations of total volatile fatty acid (VFA) and ammonia (NH₃) were measured after 4 h incubation period by using steam distillation and Conway micro-diffusion methods, respectively, as described in Jayanegara et al.¹⁵.

Another in vitro incubation procedure was performed to measure gas production during fermentation according to Menke and Steingass¹⁶. Samples (200 mg DM) were incubated with 10 mL rumen fluid and 20 mL buffer solution in glass syringes at 39°C for 24 h. Volume of fermentation gas was manually read from the calibrated scale printed on the glass syringes. Supernatant obtained after 24 h incubation was further determined for proteolytic bacteria and protozoa population and protease activity. Population of proteolytic bacteria was determined through roller tube method¹⁷ whereas, protozoa population was counted by using Fuchs-Rosenthal counting chamber (Brand GmbH+Co. KG, Wertheim, Germany). Protease activity was determined by following the procedure of Brock et al.¹⁸. In vitro incubations were performed in three runs and each run was represented by two incubation tubes or syringes.

Statistical analysis: Data obtained were subjected to two-way analysis of variance (ANOVA) based on a randomized complete block design with the following statistical model:

Y_ij = μ+α_i+β_j+ε_ij

where, Y_ij is observed value, μ is overall mean, α_i is treatment effect, β_j is block effect (replicate) and ε_ij is random residual error. Different in vitro runs were considered as block effect in the statistical model. Duncan’s multiple range test was employed for comparison among different treatment means when ANOVA result for a certain parameter significant at p<0.05. Statistical analysis was performed by using IBM SPSS Statistics software version 20 (IBM Corp., Armonk, NY, USA).

RESULTS AND DISCUSSION

Experiment 1: Cricket was high in CP, EE and chitin contents (Table 1). Removal of exoskeleton decreased CP and chitin contents of cricket. Chemical extraction of cricket increased its CP and TDN but decreased EE and CF contents and such procedure completely removed its chitin. With regard to amino acid composition, cricket was particularly high in glutamic acid and histidine proportions but low in methionine (Table 2). Exoskeleton removal and chemical extraction generally did not lead to major changes in amino acid profiles of cricket except for histidine that markedly reduced due to both treatments.

Table 1:	Proximate composition (in percent dry matter) of whole cricket (C), cricket after exoskeleton removal (CER) and cricket after chemical extraction (CCE)
TDN: Total digestible nutrient, nd: Not detected

Table 2:	Amino acid profile (in percent total amino acid) of whole cricket (C), cricket after exoskeleton removal (CER) and cricket after chemical extraction (CCE)

Table 3:	Fatty acid profile (in percent total fatty acid) of whole cricket (C), cricket after exoskeleton removal (CER) and cricket after chemical extraction (CCE)

Table 4:	In vitro rumen fermentation and digestibility of whole cricket (C), cricket after exoskeleton removal (CER) and cricket after chemical extraction (CCE)
Different superscripts within the same row are significantly different at p<0.05. NH3: Ammonia, VFA: Volatile fatty acid, IVDMD: In vitro dry matter digestibility, IVOMD, in vitro organic matter digestibility

Table 5:	Concentrate formula containing cricket meal or cricket after exoskeleton removal (CER) and their chemical composition
DM: Dry matter, NFE: Nitrogen free extract, TDN: Total digestible nutrient

Main fatty acids observed in cricket were (ordered from the highest proportion) linoleic acid, palmitic acid, oleic acid and stearic acid and the composition was unaltered due to exoskeleton removal or chemical extraction (Table 3). Cricket was relatively highly digestible, i.e., IVDMD and IVOMD values were higher than 70% (Table 4). Exoskeleton removal and chemical extraction did not further improve IVDMD and IVOMD of cricket. This was also the case with ruminal ammonia concentration. Exoskeleton removal increased total VFA concentration of cricket in the rumen in vitro than that of control (p<0.05).

Experiment 2: In this experiment, R1 was a concentrate formula containing 30% soybean meal and it was substituted by cricket or CER in R2-R4 (Table 5). In R2 and R3, soybean meal was replaced with cricket by 50 and 100%, respectively, whereas, in R4 soybean meal was completely substituted with CER. Crude protein contents of rations increased from R1-R4 but their TDN contents decreased. In vitro ruminal ammonia concentrations of cricket or CER containing rations were higher than that of control (p<0.05, Table 6). Total VFA concentration and protease activity were similar among treatments. Treatment R2 produced similar gas production than that of R1, but gas production in R3 or R4 incubation was lower in comparison to R1 (p<0.05). Populations of proteolytic bacteria and protozoa revealed similar patterns, R2 increased both populations (p<0.05) whereas, R3 and R4 decreased them (p<0.05). With regard to in vitro digestibility, R1 and R2 revealed similar IVDMD and IVOMD, but R3 and R4 resulted in lower values for both parameters than those of R1 and R2 (p<0.05).

Chemical composition of cricket: Cricket used in the present study contained considerable amounts of CP, EE and chitin. Such high CP and EE contents of cricket were in agreement with other studies^19,20, whereas, the high chitin was in agreement with that of Wang et al.⁵. Exoskeleton removal decreased chitin content of cricket by approximately half due to the fact that chitin in insect is mainly deposited in its exoskeleton⁶. This simple approach may, therefore, be used to decrease chitin content of cricket that limits its utilization as an animal feed. A decrease of CP content of cricket after exoskeleton removal is apparently related to chitin decrease since chitin is a nitrogenous molecule with N-acetylglucosamine monomer²¹. Lower EE content after chemical extraction was expected since extraction procedure involved delipidation of cricket by using an organic solvent n-hexane and thus partially removed the fat. Further chemical extraction step after delipidation was solubilization and precipitation by using NaOH and HCl, respectively, mainly aimed to separate cricket protein from other organic molecules particularly fiber; such procedure was successful in decreasing CF. Since chitin is recovered as fiber in the proximate analysis²², concentration of the compound seems to be negligible due to substantial CF reduction following the chemical extraction.

Table 6:	In vitro rumen fermentation and digestibility of concentrate containing cricket meal or cricket after exoskeleton removal (CER)
Different superscripts within the same row are significantly different at p<0.05, R1: Control ration containing 30% soybean meal, R2: Ration containing 15% cricket, R3: Ration containing 30% cricket; R4: Ration containing 30% CER, NH3: Ammonia, VFA: Volatile fatty acid, DM: Dry matter, CFU: Colony forming unit, IVDMD: In vitro dry matter digestibility, IVOMD: In vitro organic matter digestibility

In the previous study, chitin in some insect species was recovered in nitrogenous fiber fractions namely neutral detergent insoluble crude protein and acid detergent insoluble crude protein⁴.

Quality of protein depends very much on amino acid composition. High proportion of glutamic acid in cricket protein, which is a non-essential amino acid, was in agreement with other studies that characterized amino acid profiles of cricket^20,23,24. Low proportion of methionine in cricket was also observed in these studies^23,24. Jozefiak et al.²⁵ described that, in comparison to fish meal, insects including cricket generally contain lower concentration of methionine that need to be considered when formulating ration based on insect protein. In contrast to other studies, histidine was in high proportion although it decreased considerably after exoskeleton removal or chemical extraction. Such difference in chemical composition within an insect species might be caused by a number of factors such as developmental stage, rearing condition and composition of growth media for insect production^1,3,25. An appropriate balance between essential and non-essential amino acids is required for effective utilization of dietary protein. Ratio between essential and non-essential amino acids in C, CER and CCE was 1.11, 0.87 and 1.03, respectively. These cricket preparations thus meet high-value protein sources in which at least 40% of their amino acids should be essential²⁰.

High proportion of linoleic acid, palmitic acid and oleic acid in cricket was also observed by other authors^19,20,23,24. These fatty acids contributed to almost 90% of total fatty acid present in cricket. Supporting present study finding, in the study of Oonincx et al.¹⁹, linoleic acid, palmitic acid and oleic acid made up more than 75% of total fatty acid. Apparently such pattern is not only belong to cricket but also to some other insect species. House cricket, locust and mealworm larvae were reported to contain linoleic acid, palmitic acid and oleic acid 68.35-86.14% from total fatty acid²⁰. High proportion of linoleic acid found in cricket is apparently due to 12 desaturase activity in the insect that uses oleoyl-CoA as a substrate to produce linoleic acid, thus converting oleic acid to linoleic acid²⁶. Fatty acid profile of insects may vary greatly since such composition is influenced by dietary fatty acid profile although there are few exceptions like in yellow mealworm¹⁹. In the context of animal nutrition, insect lipid contributes to animal by supplying energy and essential fatty acids. Polyunsaturated fatty acid (PUFA) like linoleic acid is favorable since PUFA may be deposited in animal product, particularly in monogastric animal and contribute to human health. For ruminant, PUFA undergoes massive lipolysis and stepwise biohydrogenation processes to form various isomers of fatty acids with higher degree of saturation and therefore, dietary fatty acid profile could not directly be represented in the product²⁷.

In vitro fermentative and digestibility characteristics of cricket: Values of IVDMD and IVOMD of cricket observed in the present study were higher than those of researchers previous study⁴. In that study, IVDMD and IVOMD values of cricket were 64.2 and 64.7%, respectively. Such differences may be attributed to different chemical composition of cricket in the two studies. Content of CF of cricket in researchers previous study was 14.6% DM whereas, it was much lower, i.e., 6.9% DM in this study. Negative influence of fiber, particularly lignocellulose fraction on digestibility has been widely recognized²⁸. It is interesting to observe that exoskeleton removal and chemical extraction did not increase IVDMD and IVOMD of cricket although the treatments were successful in (partially) removing chitin. Apparently chitin is degraded in the rumen since some rumen microbial species possess chitinolytic activity such as endochitinase, exochitinase, N-acetyl-glucosaminidase, chitosanase and chitin deacetylase and these microbes included bacteria, anaerobic fungi and protozoa species^29-31.

An increase of ruminal ammonia concentration following substitution of soybean meal by cricket meal or CER was apparently due to higher CP content of R2-R4 in comparison to R1. Ammonia in the rumen is originated from protein after transformation through proteolysis and deamination processes by proteolytic microbes³² and this was confirmed by an increase of proteolytic bacteria population in R2. Ammonia itself is further utilized by rumen microbes for microbial protein synthesis. Type of protein present in feed material may also determine ammonia concentration in the rumen. Protein in soybean meal is considered as ruminally degradable¹⁵ and thus potentially contribute to high concentration of ammonia. It seems that cricket contains high proportion of rumen degradable protein as well based on its ruminal ammonia concentration, however, this requires further study either by employing in sacco method or by using protein fractionation method of Cornell³³. Partial replacement of soybean meal by cricket meal did not decrease in vitro gas production and digestibility. However, its full replacement led to lower values of both parameters. Despite cricket contains higher CP content than that of soybean meal, it also contains higher fiber thus limiting its use to completely replace soybean meal in ruminant diet. Another factor that may explain such decrease of in vitro gas production, IVDMD and IVOMD is an increase of dietary EE with cricket substitution. Dietary EE has been known to cause negative effect on carbohydrate degradation by rumen microbes, particularly fiber degradation³⁴.

Results from this study imply that exoskeleton removal and chemical extraction are effective methods to reduce chitin present in cricket without adversely affecting other nutrients such as amino acid and fatty acid profiles. However, chitin decrease is not accompanied with an increase in cricket’s digestibility in vitro. In application, therefore, such chitin removal may not be necessary when cricket is used in ruminant feed, but it would be useful for those of monogastrics like poultry and swine. With regard to the use of cricket in ruminant diet, we recommend that the insect may be used up to 50% in replacing soybean meal. Further studies involving in sacco and in vivo experiments are required in order to confirm the results obtained since this study is limited to in vitro rumen fermentation experiments.

CONCLUSION

Cricket contains high CP, EE and chitin contents and treatment with either exoskeleton removal or chemical extraction is effective in decreasing chitin content of the insect. Cricket contains high proportion of glutamic acid and histidine but low in methionine. Linoleic acid, palmitic acid, oleic acid and stearic acid are major fatty acids present in cricket. Exoskeleton removal or chemical extraction does not alter amino acid and fatty acid profiles of cricket. Cricket is relatively highly digestible and may be used to substitute soybean meal up to 50% in concentrate for ruminant.

SIGNIFICANCE STATEMENT

This study discovers that treatment with either exoskeleton removal or chemical extraction is effective in decreasing chitin content of cricket; this can be beneficial for improving nutritive value and nutrient utilization of cricket to be used as an animal feed. The present study also discovers that cricket meal may be used to substitute soybean meal up to 50% in concentrate for ruminant. This study will help researchers to uncover the critical areas of insect utilization for ruminant feed in which such study presently is still limited.

ACKNOWLEDGMENT

This study was funded by Indonesian Ministry of Research, Technology and Higher Education through "Penelitian Berbasis Kompetensi" research grant, contract number 011/SP2H/LT/DRPM/IV/2017.