Larval Protein Quality of Six Species of Lepidoptera (Saturniidaeo Sphingidae, Noctuidae)

STEPHEN V. LANDRY,I GENE R. DEFOLIART,IANP MILTON L. SUNDE,

University of Wisconsin, Madison, Wisconsin 53706

J. Econ. Entomol. 79; 600-604 (I98€) ABSTRACT Six lepidopteran species representing three families were evaluated for their potential use as protein supplements for poultry. Proximate and amino acid analyses were conducted on larval powders of each species. Larvae ranged from 49.4 to 58.1% crude protein on a dry-weight basis. Amino acid analysis indicated deficiencies in arginine, me- thionine, cysteine, and possibly lysine, when larvae are used in chick rations. In a chick- feeding trial with three of the species, however, these deffciencies were not substantiated: the average weight gained by chicks fed the lepidopteran-supplemented dlet did not differ significantly from that of.chicks fed a conventional corn/soybean control diet.

LnprpoptBne ARE among the many species of in- by feeding trials on poultry (Ichhponani and Ma- sects that have played an important role in nutri- Iek 1971, Wijayasinghe and Rajaguru 1977). tion, especially in areas where human and domes- To determine and compare the protein quality tic animal populations are subject to chronic protein of a wider assortment of lepidopterous larvae, we deficiency (e.g., Bodenheimer 1951, Quin 1959, conducted proximate and amino acid analyses on Ruddle 1973, Conconi and Bourges 1977, Malaise larvae of six species representing three families. and Parent 1980, Conconi et al. 1984). Conconi et These included the cecropia moth, Hgalophora ce- al. (f9Sa), for example, listed 12 species in 8 fam- cropia (L.), and the promethea moth, Callosamia ilies that are gathered and consumed in Mexico. promethea (Drury), in the family Saturniidae; the Wider utilization of Lepidoptera as domestic ani- tobacco hornworm, Manduca serla (L.), in the mal feedstuffs has probably been precluded by the family Sphingidae; and the fall armyworm, Spo- difficulties inherent in gathering them in sufficient doptera frugiperda (I. E. Smith), southern army- quantities on a regular basis. Pupae of the silk- worm, S. eridania (Cramer), and the armyworm, worm, BombEx mori (L.), however, available as a Pseudaletia unipuncta (Haworth) in the family byproduct of the silk industry, have been used not Noctuidae. Chick feeding trials were conducted only as human food but as a high-protein supple- with dry powders of the sphingid and saturniid ment for poultry in parts of Asia. Ichhponani and larvae. Although our primary interest was in the Malek (1971) reported the annual production of noctuid species because of their gregarious nature de-oiled silkworm pupae in India alone at 20,000 and potential for opportunistic mass harvest of wild metric tons. populations, we were unable to accumulate enough Proximate analyses (percent protein, moisture, noctuid powder for a feeding trial. fat, ash, crude fiber, and carbohydrate) have been conducted on larvae or pupae of species in a num- ber of families of Lepidoptera (Quin 1959, Con- Materials and Methods 1977, Malaise and Parent 1980, coni and Bourges Souree and Rearing of Experimental Insects Conconi et al. 1984). Crude protein content was found to range from 7.8 to II.8% in samples of Saturniidae. H. cecropia pupae were obtained freshly killed larvae to between 50 and SOVo in from j. M. Scriber (Department of Entomology, larval samples subjected to various degrees of de- University of Wisconson-Madison). Adults hydration. By contrast, amino acid profiles are emerged and mated; eggs were collected and available only for pupae of B. mori (Bombycidae) placed in plastic crispers (30 by 12.7 by 6.4 cm) (Chopra et al. 1970, Wijayasinghe and Rajaguru provided with fresh'Prunus serotina (Ehrhart) 1977) and larvae of Xgleutes redtenbachi (Ham- (black cherry) foliage. A total of 800 g of dried H. merschmidt) (Cossidae), a species utilized as hu- cecropia was produced for a chick feeding trial. man food in Mexico (Conconi and Bourges 1977). C. promethea were also reared on P. serotina f.o- We are not aware of any laboratory studies in- liage. Totally, 350 g of dried C. promethea were volving Lepidoptera fed to rats to determine bio- prepared for a chick feeding trial. availability of amino acids; only in the case of B. Sphingidae. M. sexta larvae were obtained from mori pupae has amino acid analysis been followed the laboratory colony of W. G. Goodman (De- partment of Entomology, University of Wisconsin- of Entomology. Madison). Some of these larvae were reared on 'Dep. (Bio-Serv 'zDep. of Poultry Science. artiftcial diet Tobacco Hornworm diet 600 June 1986 L.lNpnv ET AL.: Nurnl]]loNer- Ver.ur oF LEPIDOPTERA 601

Table l. Proximate analysis of fifih-instar lepidopteran larval powders conpared with conventional protein suP- plements

Protein Moisture Fat Ash Crude ffber Carbohy- Calories/ Protein source (%) (v") (%) (%) (%) drates (%) 100 g

Fish meal (menhaden) 60.5 8.0 9.4 o.7 282 Soybean meal (dehulled) 48.5 r0.0 1.0 3.9 244 Meat and bone meal 50.4 7.0 8.6 2.8 196 H. cecropia 54.7 2.6 r0.2 * 14.7 t* 358 C. promethea 49.4 4.5 r0.0 6.9 I0.8 r8.4 361 M. sertaa 58. I 4.7 20.7 7.4 9.4 > 0.1 418 M- sextab 57.8 4.7 16.5 8.1 8.4 4.5 398 S. frugiperdaa 57.8 2.t 20.2 ,r 5.6 6.7 7.7 443 S. frugiperdab 57.2 3.6 r1.3 tL.2 t2.o 4.8 349 P. unipuncta 54.4 2.0 t4.s 6.9 5.0 15.7 424 S. erid.ania 54.7 4.5 I3.9 9.8 7.1 10.0 384

a Reared on artiffcial diet. b Reared on fresh plant material.

No. 9788); others were rear'ed on Solanum dul- Arnino acid proftles were obtained by acid hy- canlara (L.) (nightshade). drolysis of the powders and subsequent analysis on Noctuidae. S. frugiperde were obtained from an amino acid analyzer (Beckman). John Young (USDA Laboratory, Tifton, Ga.). They were fed a variation of the diet formulated by Shorey and Hale (1965). The colony ot S. eridania Method of Chick Bioassay was obtained from J. M. Scriber and fed the same artificial diet. P. unipuncta adult females were The chemical analysis was followed by a biolog- collected in the field. Eggs were laid, and emerg- ical assay. Three powders, If. cecropia. C. pro- ing individuals were fed artificial diet or Agro- rnethea, and M. sexta, were tested as the protein pAron repens (L.) (quack grass), supplement in a corn-based diet. AII diets were , based on the most recent National Academy of Science-National Research Council (NAS-NRC) Methods of Biochemical Analysis recommendations (National Research Council All insects were harvested in the final instar or 1977). They were isonitrogenous and isocaloric. prepupal stage. They were frozen at -23'C, and There was no information on the vitamin content lyophilized. The coarse-ground larvae were re- of the various powders; thus, all necessary vitamins duced to a powder in a Wiley Mill using a 20- were added as a supplement as prescribed by NAS- mesh screen. Proximate analyses were performed NRC. The availability of the minerals also was using procedures described by the Association of unknown and all required minerals were added. Official Analytical Chemists (Anonymous 1945). This supplementation allowed comparisons of the

Table 2. Amino acid profiles of lepidoprerous larvae compared with dietary requirements {or chickens and zMVo soybean meal

Amino acid Requirement for: soYbean H. cecro- c. oro- M. M. S. P. uni- S. mg/g ol meal i, frugi Broiler Laying Pn" metnea b sertaa sextab perdaa punctaa eridaniab protein chick hen (44%)

Arginine 63 53 75 53 42 53 42 67 54 40 Glycine-serine 65 108 208 184 105 90 100 102 r05 Histidine l5 l5 26 27 29 37 4r 3.8 29 3l Isoleucine 35 54 29 43 37 44 38 45 Leucine 59 80 80 51 44 68 62 77 64 72 Lysine 52 40 67 44 43 7t 81 77 180 74 Methionine 22 t8 l5 't4 l0 18 20 24 23 2t Methionine- cysteine 40 30 22 15 26 26 3l 36 28 Phenylalanine 3l 27 52 43 34 47 42 49 39 4l Phenylalanine- tyrosine 58 53 8r 140 Il4 lt3 93 t28 93 7S Threonine 33 27 4t 4I 39 41 36 5l 46 53 Tryptophan r0714 NA NA NA NA NA NA NA Valine 36 33 53 45 41 66 56 52 47 58

NA, not attempted. o Reared on artificial diet. D Reared on fresh plant material. 602 JounNel or EcoNontc ENrotrolocv Vol. 79, no. 3

Table 3. Composition of experimental diets fed to chicks

Ingredient Diet I Diet 2 Diet 3 Diet 4 (% of diet) (Control) (H. cecropia) (C. promethea) (M. sexta) Ground corn 55.75 62.95 52.80 64.50 Soybean meal (44Vo) 28.00 Wheai middlings 5.00 Alfalfa rdeal (I7%) 3.00 Meat meal 3.00 Insect meal 30.20 37.20 30.04 Iodized salt 0.50 0.40 0.40 0.40 Dicalcium phosphate o.75 2.57 3. l2 3.00 CaC03 1.00 0.80 0.50 0.60 MnSOa 0.03 0.03 0.03 ZnC03 0.01 0.01 0.01 Fe citrate 0.04 0.04 0.04 Se Premix: (1 mg Se/g NaCl) 0.0r 0.01 0.0I Corn oil 2.40 5.00 1.00 Choline (50%) 1,300 mglkg t,300 mglkg 1,300 mglkg Vitamin mix 3.004 o.50b 0.50 0.50 Vitamin E 20 Iu/kg 20 tU /ke 20 IU /kg Amino acid supplementation (mg/g protein) Arginine 38 58 Isoleucine IO l9 : Leucine 15 Lysine 42 46 46 Methionine 40 60 38

Only half the chicks fed diets 2 and 4 received the amino acid supplementation. d Viiamin mix diluted with wheat middlings used in the standard diet by the Department of Poultry Science, University of Wisconsin, Madison. l, Vitamin mix designed to fulfill requirements of NAS-NRC. diets that more accurately reflected protein qual- After 8 days, each treatment was divided. Half ity. the chicks continued to eat the same diet and half The chicken feeding trials were designed for the received the diet supplemented with amino acids Iimited amount of protein powder available. Use to meet NAS-NRC recommendations. All of the of New Hampshire x white leghorn chicks al- C. promethed diet was supplemented because of Iowed a more sensitive test due to their lower feed the limited supply. Mean weight gains for the consumption. Twenty-eight l-day-old chicks were chicks fed the various diets were compared by assigned to four treatments in a stratified random analysis of variance. block design. For each treatment, eight birds were placed banded, individually weighed, and in elec- Results and Discussion trically heated chick starter wire cages, Four chicks were used in the C. promethe& treatment because In the proximate analysis, the crude protein of the relatively small amount of C. promethea content of the lepidopteran powders was compa- powder available. Food and water were provided rable with that of the conventional protein feed ad lib. The birds were weighed on alternate days. supplements, fish meal, meat and bone meal, and

Table 4. Mean weights and feed/gain ratios of chicks fed experimental diets

18 days 2L days Feed wt Feed Wt wt , Protein source consumed gained Feed/gain 'consumed gained Feed/gain ( +sD) -= ,Yj^, f (g p.r ratio ."iii (g per biid) ,f,ifi ,f;,:ii bird) rJj*i?ir kU Soybean meal (control) r37 (r4.2) 99.9 2.32 167 (15.0) 326 129.3 2.52 Soybean meal (control) r3r (27.3) 240 93.9 2.56 r58 (32.7) 364 120.3 3.03 C. prometheaa 120 (1s.0) 196 82.0 2.39 NT NT NT NT M. sexta r27 (r4.8) 2to 89. r 2.36 r50 (r5.s) 285 112.6 2.53 M. sextaa r2r (9.4) 194 83. I 2.33 154 (6.6) 267 I 15.6 2.3r H. cecropia r22 (34.0) 225 84.6 2.66 r4s (44.6) 282 106.6 2.65 H. cecropiaa r4r (22.8) 233 I03. I 2.26 180 (2r.7) 292 142.6 2.O5

NT, not tested. d Diet with supplemental amino acids. Vnlun oF LEPIDoPTERA 603 June 1986 LaNonv ET AL.: NutnrtloNel

weights or feed/ sovbean meal. The six species had a range of from nificant differences, either in final et al' lg.+ to 58.1% crude protein on a dry-weight basis gain ratios (DeFoliart et al. 1982, Finke -1985)' with ca' 58% found in Th" htt"r investigators suggested that the high lerel (Table 1). This compares ' ih" Mor*ot cricket (Anabtus simplex Haldeman) of choline in thJ Mormon cricket (4,900 mg/kg) (DeFoliart et al. 1982), but is somewhat lower than may act as a methyl group donor in transmethyl- percentages reported for many insects, including ation reactions, thus reducing the need of methi- purified diets, that i-oine other species of Lepidoptera' A 7I% protein onine. They also found, using content was reported by Conconi and Bourges the lysine in ground crickets is readily a-vailable, (1977) for larvae of X. redtenbachi, and a protein and [hat methionine and arginine are co-limiting' there content of 63.3% has been reported for silkworm Only when both amino acids were added was weights and a signif- ourlae Q4-76% for de-oiled pupae) fed to poultry a significant increase in final in isia (Bora and Sharma 1965, Chopra et al' 1970, icant decrease in feed/gain ratios. Wiiavasinghe and Rajaguru 1977). 'ih" pto'*ittt"te analyses indicated that the fat Acknowledgment and, thus, energy content were higher in the insect was supported by the Coll. of Agricul- oowders than in the conventional supplements in This research and Life Sciencei,- and by the Graduate School, (Table The extremely high fat content tural ull 1). Univ. of Wisconsin, Madison. ol M.""r"r sexta and S. ftugiperda that were fed on the artificial diet probably reflebts the diet on which thev were ,""r"d and the fact that not all larvae References Cited har- weie able to clear the gut before they'were Anonymous. 1945. Official and tentative methods of vested. analysls of the Association of Official Analytical The amino acid profiles (Table 2) indicated de- Chemists, Washington, D.C. ficiencies in arginine, methionine, cysteine, and Bodenheirner, F. S. 1951. Insects as human food' possibly lysine, if lepidopteran powders are to be Junk, The Hague. rations. These profiles are consistent Bora, L. R', and P. K. Sharma. 1965. Assam Muga used in chick pro- on Lepidoptera by other investi- silk worm, Antharaca assamensis Ww, pupae as with findings chick ration. Indian Yet' 42: gators (Conconi and Bourges 1977, Wijayasinghe tein supplement in J' regard to the 354-359. ind Rajaguru I977), particularly,in A. K., N. S. Malek' G. S. Makker,-and-J' S' cys- Chopra, sulfur-coitaining amino acids methionine and I-chhponani. Lg7O. Evaluation of poultry feeds teine, which have appeared to be deficient by bio- avail-able in India. I. Proximate analysis, energy val- chemical methods. These apparent deficiencies in- ues, and basic amino acid contents of feed ingredi- dicated by biochemical analysis were not ents. J. Res. Ludhania 8:232-236. substantiated, however, in the feeding trial that Conconio J. R. E. de, and H- R. Bourges. 1977' Yalor y we conducted. nutritivo de cierto insectos comestibles de Mexico mundo Ann' The chicks used in the bioassay fed readily on Iista de algunos insectos comestibles del Univ. Nat. Auton. Mex. 48 Ser. Zool' I: the experimental diets shown in Table 3. There Inst. Biol. the average r65-186. *"r" .ro significant differences in R. E. de, J. M. Pino Morenoo C. M' May- : : 25; Conconio J. weight gain of chicks at 18 (F 0.57; df audon, F' R. Valdez, M. A. Perez, E. E. Pradoo and f iO.OOt)or at 21 (F:0.95;df :21; P-< 0'001) H. B. Rodriguez. 1984. Protein content of some days of age (Table 4), although chicks initially-grew edible insects in Mexico. J. Ethnobiol. 4:6I-72. ce- and M' L. Sunde' l9B2' *ot" t"pidly on amino acid-supplemented -H. DeFoliart, G. R., M. D. Finke, cropia diet.- Chicks fed the experimental diets Potential value of the Mormon cricket (Orthoptera: protein feed for tended to eat less, and feed-to-gain ratios were as Tettigoniidae) harvested as a high 75:848-852' good or better than those obtained on the controls. poultry. J. Econ. Entomol. and G. R' DeFoliart' 1985' as shown Finte, M. D., M. L. Sunde, Supplementing the diets with amino acids protein quality of Mormon gains, An evaluation of the in Table 4 hadno signilicant effect on weight (Anabrus simplex Haldeman) when used as used in crickets indicating that the lepidopteran protein a high protein feedstuff for poultry. Poult. Sci. 64: these testi was probably of higher quality than the 708-712. amino acid profiles indicateil. Mortality was-con- Ichhponani, J. S,, and N. S. Malek. 19?1. Evaluation fined to the control treatment, in which two birds oi de-oiled silkyorm pupae meal and corn-steep fluid died of unknown causes. as protein sources in chick rations. Br. Poult. Sci. 12: The results of this feeding trial indicate that the 23r-234. Les chenilles co- protein used is probably similar in its Malaise, F.' 5nd G. Parent. l'98O. lepidopteran (Zaire). Nat Belg. 61: acid availability to that of the Mormon mestibles du shaka meridional 2-24. "mitro- house cricket, Acheta domesticus cricket and the National Research Council. 1977. Nutrient require- (L.), two insects on which we have accumulated ments of poultry, 7th ed. National Academy of Sci- coniiderably more data. The amino acid profiles ences, Washington, D.C. of the lepidopterans were similar to those of the Ouin.- P. J. 1959. Foods and feeding habits of the crickets, and broiler chick feeding trials compar- Pedi. Witwatersrand Univ., Johannesburg. ing conventional corn/soybean diets with corn- Ruddle, K. f 9?3. The human use of insects: examples ground Mormon cricket diets have shown no sig- from the Yukpa. Biotropica 5:94-101. I

' 604 JounNeL or Ecouotlc Exrouorocv Vol.179, no. 3

Shorey, H. H., and R. L. Hale. 1965. Mass rearing tein supplement in poultry rations. J. Nat. Sci. Counc. of the larvae of nine noctuid species on a simple Sri Lanka 5: 95-104. artiffcial medium. J. Econ. Entomol. 58:522-524. Vijayasinghe, M. S., and A. S. B. Rajaguru. 1977. ' Recetoed fot publlcatlon 2 Jul,J 1985; accepted' 37 Use of silkworm (BornbEx rnori L.) pupae as a pro- December 7985.

i I i