• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Simulated digestion status of insects and insect larvae: A spectroscopic investigation

Posted on: Tuesday, 29 July 2003, 06:00 CDT

Received: March 18, 2002

Accepted after revision: February 3, 2003

Abstract

In this study, we tested the hypothesis that puncturing the chitin exoskeleton of insect and insect larvae food sources aids the ingress of digestive fluids and increases the rate of digestion and energy uptake in insectivorous mammals. For this purpose 10 crickets (Acheta domesticus) and 10 mealworms (Tenebrio molitor larvae) were divided into two groups of 5; one group was punctured using a small blade to mimic the effect of a single bite, the remainder serving as controls. The insects were then individually immersed in 5 ml of a 1 x 10^sup -2^ mol[middot]dm^sup -3^ solution of hydrochloric acid (pH 2.0) for a period of 2 h in order to mimic digestion in the stomach. The matrix was then centrifuged and the supernatant fluid subjected to spectrophotometric and high-resolution proton (^sup 1^H) NMR analysis. Electronic absorption spectra of these supernatants revealed that puncturing the exoskeleton of mealworms and crickets gave rise to substantial elevations (up to 14-fold) in the concentrations of UV-absorbing biomolecules (p < 0.025 for both species). The 400-MHz ^sup 1^H NMR profiles of supernatants derived from mealworm and cricket specimens with punctured exoskeletons contained a wide variety of prominent biomolecule resonances, whereas those from unpunctured (control) insects contained signals of a much lower intensity, ascribable only to selected biomolecules. We conclude that puncturing the cuticle of insects and insect larvae prior to swallowing confers significant nutritional advantages over swallowing prey whole.

Key Words

Insects [middot] Digestion [middot] Spectroscopy [middot] Biochemistry

Copyright (C) 2003 S. Karger AG, Basel

Introduction

Herbivorous mammals devote considerable time to comminuting their food, whereas carnivores merely bite food into manageable portions prior to swallowing. The behavioural and dental adaptations to diet reflect the requirement for herbivores to maximise the surface area of their food and to fracture cell walls in order to aid digestion. In carnivores, the large pointed anterior teeth are adapted to cutting. Although some insectivorous amphibians and reptiles are toothless and hence swallow their prey whole (indicating that teeth are not necessary for an insectivorous diet per se), insectivorous mammals typically have sharp-pointed teeth, which may assist in grasping prey but do not seem well adapted to cutting or comminution. Exceptions are mymecophagic (ant-eating) animals such as the anteater, which is toothless, and the aardvark and armadillo, which have flattened posterior teeth. However, these animals are large in relation to the size of their prey [Popowics & Fortelius, 1997].

Small insectivorous mammals, for example the shrew, have very high metabolic rates, eating up to 3 times their own weight per day [Lewin, 1999]. It is therefore very important for these animals to optimise the rate of digestion of their food, in contrast to amphibians and reptiles, which have very low energetic requirements. The Western tarsier, a small insectivore, has been observed to take only 3-4 bites when consuming large insects. Consideration of the design of its teeth and the small number of chewing strokes involved indicates that it is unlikely the prey is comminuted to any significant degree [Jablonski & Crompton, 1994]. The energetic advantages of comminution for herbivores are well documented [Alexander, 1968; Janis & Fortelius, 1988], but in humans there appears to be no nutritive benefit from the comminution of meat [Alexander, 1991], although it may assist the swallowing process [Prinz & Lucas, 1997; Alexander, 1998]. However, the role of mastication in other carnivores and insectivores remains less clear.

Therefore, a multicomponent biochemical investigation of the digestive fluid-mediated release of biomolecules from insects and/ or their larvae both prior and subsequent to the biomimetic puncturing of their chitin exoskeletal matrices should, in principle, supply much useful information to this research area and hence further our understanding of the processes involved.

The recent development of high-field, high-resolution nuclear magnetic resonance (NMR) spectrometers with much increased sensitivity and dynamic range has permitted the rapid, simultaneous, multicomponent analysis of compounds of endogenous or, where appropriate, exogenous origin in biological samples. These include biofluids such as human blood plasma [Nicholson et al., 1995], knee- joint synovial fluid [Naughton et al., 1993], saliva [Lynch et al., 1999], urine [Nicholson et al., 1984], and foodstuffs [Grootveld et al., 1990], and generally requires only minimal sample preparation and little or no knowledge of sample composition prior to analysis. Indeed, chemical shifts, coupling patterns and coupling constants of signals present in NMR spectra of such complex, multicomponent samples offer much valuable information regarding the structures of molecules present. Hence, the technique has major advantages over alternative labour-intensive, time-consuming analytical methods. Moreover, the methodology and technology employed are virtually non- invasive and hence samples can be subsequently subjected to further, confirmatory methods of analysis if required.

High-resolution NMR spectroscopy is also a well-established and powerful tool for the investigation of cellular metabolism, providing detailed information regarding intracellular metabolite levels and fluxes [Neeman & Degani, 1989]. The technique can be performed either in vivo or by analysis of appropriate aqueous extracts [predominantly those involving prior macromolecule precipitation with perchloric acid (HClO^sub 4^)], in which all components present give sharp resonances that are readily identified and quantified.

In this investigation we test the hypothesis that puncturing the chitin-based exoskeleton of insects prior to swallowing aids the ingress of digestive fluids. For this purpose we have determined the chemical nature and concentrations of biomolecules released from such punctured insects in a simple in vitro model of insectivorous mammal digestion, involving dilute hydrochloric acid (corresponding unpunctured insects served as controls). 400-MHz ^sup 1^H NMR analysis of post-neutralised aqueous supernatants derived from the incubation of control (untreated) and exoskeletally punctured mealworms and crickets with 1 x 10^sup -2^ mol[middot]dm^sup -3^ HCl solution (2 h at 37[degrees]C) was conducted in order to ascertain the molecular nature and levels of components available for nutritional purposes. Moreover, electronic absorption spectra of these supernatants were acquired in order to provide a further quantitative measure of biomolecule release from these insects. The comminutional, digestive and nutritional significance of the results obtained are discussed in detail.

Materials and Methods

Reagents

Sodium 3-(trimethylsilyl)-[1,1,2,2-d^sub 4^]-propionate (TSP, internal chemical shift reference and quantitative NMR standard) was purchased from Sigma Chemical Co. (Poole, Dorset, UK), and deuterium oxide (^sup 2^H2O) was obtained from Goss Scientific Ltd. (Great Baddow, Essex, UK). All other reagents utilised were of the highest possible grade and purchased from commercially available sources.

Sample Preparation

Mealworms (Tenebrio molitor larvae) and crickets (Acheta domesticus) utilised in the experiments described here were purchased from a local retail outlet. Ten individuals were randomly selected from each insect/insect larvae group and accurately weighed subsequent to cooling to a temperature of 4[degrees]C in order to reduce their mobility (weight ranges were 40-200 mg for mealworms and 80-180 mg for crickets). These groups of 10 were then further divided into two subgroups (n = 5) for treatment and control. The first of these two subgroups (treatment subgroup) was subjected to a simulated bite by puncturing the insects/larvae with a single thrust from a 5-mm blade, the second subgroup (controls) remained unpunctured. Each specimen was then individually placed into separate sample tubes containing 5 ml of a 1 x 10^sup -2^ mol[middot]dm^sup -3^ solution of hydrochloric acid (HCl) in doubly distilled H2O, and the resulting matrices were incubated at 37[degrees]C for a period of 2 h.

Subsequently, the insects or insect larvae were carefully removed with forceps, each sample centrifuged (16,000 g for 30 min at ambient temperature) to remove debris. A 1-ml volume of the clear supernatant was removed and its pH value adjusted to about 7.4 by adding 10- to 20-[mu]l volumes of a 1 mol[middot]dm^sup -3^ NaOH solution. These solutions were then treated with a 0.25-ml volume of a buffer solution containing 5 x 10^sup -2^ mol[middot]dm^sup -3^ phosphate, 0.685 mol[middot]dm^sup -3^ NaCl and 0.0135 mol[middot]dm^sup -3^ KCl (pH 7.40) in order to ensure stabilisation of their pH value. After thorough rotamixing, 0.2 and 0.6 ml volumes of the resulting solutions were removed for spectrophotometric and ^sup 1^H NMR analysis, respectively.

Spectrophotometric Analysis of Samples

Aliquots (0.2 ml) of the neutralised, buffer-treated supernatants were diluted to a volume of 1.2 or 2.2 ml with doubly distilled H2O (finalvolume determined by the intensity of the sample's UV absorption bands), and electronic absorption spectra of these samples were recorded on a PC-controlled spectrophotometer (Unicam UV-2 spectrophotometer) in the 200 to 400 nm wavelength range.

^sup 1^H NMR Measurements

Proton NMR measurements were conducted on a Bruker AMX-400 (University of London Intercollegiate Research Services, London, UK) spectrometer operating at a frequency of 400.13 MHz for ^sup 1^H. All spectra were acquired at ambient temperature (22 + or - 1[degrees]C). Typically, 0.6 ml of each of the above solutions was placed in a 5 mm diameter NMR tube and 0.07 ml of ^sup 2^H^sub 2^O and 0.03 ml of a 5 x 10^sup -3^ mol[middot]dm^sup -3^ solution of TSP in ^sup 2^H^sub 2^O were added, the ^sup 2^H^sub 2^O providing a field frequency lock.

Pulsing conditions for one-dimensional (1-D) spectra acquired on samples were: 32 free induction decays (FIDS); 32,768 data points; 7 [mu]s pulses; 1 second pulse repetition rate; 2.031 seconds acquisition time. Line-broadening functions of 0.30 Hz were routinely utilised for the processing of experimental data. The intense water signal ([delta] = 4.80 ppm) was suppressed by presaturation via gated decoupling during the delay between pulses. Chemical shifts were referenced to internal TSP ([delta] = 0.00 ppm). Where present, the methyl group proton resonances of acetate (s, [delta] = 1.920 ppm), alanine (d, [delta] = 1.487 ppm) and lactate (d, [delta] = 1.330 ppm) served as secondary internal references for the samples examined.

The identities of biomolecule resonances present in the ^sup 1^H NMR spectra acquired were routinely assigned by a consideration of chemical shift values, coupling patterns and coupling constants. Where required, standard additions of authentic endogenous biomolecules were made to confirm assignments. Metabolite concentrations were determined by electronic integration of their ^sup 1^H resonances and expressing the intensities relative to that of the internal TSP standard. ^sup 1^H NMR measurements on standard solutions containing a range of physiologically relevant metabolite concentrations and internal TSP (added at the same concentration as that employed for the above test samples) were similarly conducted. The intensities of these resonances were also expressed relative to the TSP standard and a series of standard curves was generated.

Statistical Analysis of Experimental Data

Experimental ^sup 1^H NMR-determined weight-normalised concentration data were subjected to the transformation y = [1 + log^sub e^(normalised concentration)] in an attempt to homogenise intra-sample variances prior to the statistical analysis (predominantly those components where unpunctured specimens had a mean weight-normalised concentration and standard deviation of zero). However, these variances remained heterogeneous after applying this transformation, and therefore comparisons of the mean weight-normalised metabolite concentrations of the control (unpunctured) and treatment groups for these transformed data were made using an approximation of the Behrens-Fisher method [Cochran & Cox, 1950].

Fig. 1. Determination of total UV-absorbing biomolecules released from A. domesticus and T. molitor. Electronic absorption spectra of post-neutralised supernatants arising from the incubation of control and exoskeletally punctured mealworm and cricket specimens in 0.01 mol[middot]dm^sup -3^ HCl for 2 h at a temperature of 37[degrees]C. A 0.2 ml volume of each supernatant was diluted to 1.20 ml with further 0.01 mol[middot]dm^sup -3^ HCl prior to recording spectra. Typical spectra are shown. Spectra of control (a) and exoskeletally punctured mealworm (b) supernatants. Spectra of control (c) and exoskeletally-punctured cricket (d) supernatants.

Results

Spectrophotometric Analysis of Control and Exoskeletally Punctured Mealworm and Cricket Specimens

Typical electronic absorption spectra of supernatants derived from the incubation of control and exoskeletally punctured mealworm and cricket specimens in 0.01 mol[middot]dm^sup -3^ HCl for a period of 2 h at 37[degrees]C are shown in figure 1 (0.2 ml of each sample were diluted to a volume of 1.2 or 2.2 ml, with further 0.01 mol[middot]dm^sup -3^ HCl prior to the acquisition of spectra). Clearly, the aqueous supernatants arising from the insect larvae and insect specimens subjected to a simulated bite as outlined in Materials and Methods have a substantially greater absorbance in the 200-340 nm wavelength range than those serving as untreated (unpunctured) controls.

The mean + or - standard error (SE) values of the absorbances of these supernatant samples and corresponding log^sub e^-transformed data at the approximate [lambda]^sub max[middot]^ values of 224 and 274 nm (normalised for their initial sample weights, i.e. A^sub 224^ or A^sub 274[middot]^ [sample weight (g)]^sup -1^) are shown in table 1. These data reveal that puncturing the exoskeleton of mealworms increased the absorbance [sample weight (g)]^sup -1^ values at wavelengths of 224 and 274 nm by approximately 14 (p < 0.005) and 8 fold (p < 0.001), respectively, whereas for cricket specimens these treatment-mediated elevations in the normalised A^sub 224^ and A^sub 274^ values were lower, i.e. 4.2 and 4.4 fold, respectively (p < 0.025 in each case).

Table 1. Mean + or - standard error values of untransformed and log^sub e^-transformed absorbance data

Interestingly, the mean weight-normalised A^sub 224^ or A^sub 274^ values for the control (untreated) cricket specimen supernatants were higher than those of the mealworm samples, the latter being significant at the 5% level. This may reflect either greater quantities of 0.01 mol[middot]dm^sup -3^ HCl-soluble biomolecules on the exoskeletal surface of crickets than on mealworms, or, alternatively, minor levels of damage to these insects during the preparation procedure, a process giving rise to small exoskeletal fractures which afford some leakage of contents into the incubation medium. There were no significant differences detected between the mean weight-normalised absorbance values of punctured mealworm and cricket supernatants.

^sup 1^H NMR Evaluations of Metabolites Present in Supernatants Derived from the Simulated Digestion of Control and Exoskeletally Punctured Mealworms and Crickets

Figure 2 shows the high- and low-field region of typical 400 MHz ^sup 1^H NMR spectra of post-neutralised supernatants obtained after incubating control and exoskeletally punctured mealworms in the HCl medium (pH 2.0) for a period of 2 h at 37[degrees]C (cf. Materials and Methods). Clearly, the spectrum derived from the exoskeletally punctured larvae contain a wide range of prominent biomolecule resonances, whereas that arising from the unpunctured (control) specimen has only a very limited number of signals of a much lower intensity. These observations were highly reproducible in all samples obtained from experiments conducted with both punctured and control mealworms. Indeed, the high-field (aliphatic) region of spectra acquired on supernatants derived from the exoskeletally punctured larvae specimens contained well-resolved, sharp and intense resonances assignable to the biomolecules acetate, alanine, betaine, carnitine, choline, creatine, dimethylamine, free fatty acids (both saturated and unsaturated), glutamate, glutamine, glycine, isoleucine, lactate, leucine, lysine, N-acetylsugars such as N-acetylglucosamine (presumably arising from the H+-mediated hydrolysis of exoskeletal chitin), trimethylamine and valine, and occasionally methylamine, pyruvate and succinate (the latter three metabolites detectable in only 2, 1 and 1 samples, respectively). Moreover, the low-field (aromatic) region of the above spectra contained prominent resonances attributable to the amino acids histidine, phenylalanine and tyrosine, and the organic acid anion formate. Further features of the spectra acquired are signals assignable to the [alpha]-CH ^sup 1^H nuclei of aliphatic amino acids such as glycine (3.6 ppm), isoleucine, leucine and valine (3.6- 3.7 ppm), glutamate and glutamine (3.8 ppm), and aromatic amino acids such as tyrosine (about 4 ppm).

Fig. 2. ^sup 1^H NMR determinations of the identities and levels of biomolecules released from control and exoskeletally punctured insect larvae. High-field regions of 400-MHz ^sup 1^H NMR spectra acquired on post-neutralised 0.01 mol[middot]dm^sup -3^ HCl supernatants derived from control (a) and exoskeletally punctured mealworm (b) specimens. Corresponding low-field regions of these spectra from control (c) and punctured mealworm specimens (d). Typical spectra are shown. A = Acetate-CH^sub 3^; Ala = Alanine- CH^sub 3^; [beta]-HB = 3-D-hydroxybutyrate-CH^sub 3^; Bet-, Carn-, Chol-N+(CH^sub 3^) = betaine, carnitine and choline polar-N+(CH^sub 3^)^sub 3^ head groups, respectively; Bet-CH^sub 2^ = betaine- CH^sub 2^; Creat = creatine-CH^sub 3^; FA-CH^sub 3^ and -(CH^sub 2^)n- = free fatty-acid terminal-CH^sub 3^ and bulk acyl chain- (CH^sub 2^)n- groups, respectively; Form = formate-H; Gln = glutamine-CH^sub 2^; Glu = glutamate-CH^sub 2^; Gly = glycine- [alpha]-CH; Ile = isoleucine-CH^sub 3^; Leu = leucine-CH^sub 3^; Leu/ Ile/Val-[alpha]-CH = [alpha]-CH group protons of leucine, isoleucine and valine; Lys/Ala/Glu/Gln-[alpha]-CH = [alpha]-CH group protons of lysine, alanine, glutamate and glutamine; Lys-[epsilon]-CH^sub 2^ = lysine [epsilon]-CH^sub 2^; MA = methylamine-N(CH^sub 3^); N-Ac = acetamido-NHCOCH^sub 3^ group protons of N-acetylsugars (N- acetylglucosamine) and N-acetylglucosamine-containing oligosaccharides; Phe and Tyr = phenylalanine and tyrosine aromatic ring protons, respectively; Pyr = pyruvate-CH^sub 3^; Val = valine- CH^sub 3^.

Table 2. Mean + or - standard error weight-normalised supernatant metabolite concentration data for mealworm specimens

Weight-normalised concentrations of the ^sup 1^H NMR-detectable metabolites which were quantified (i.e., those with resonances clearly resolved from neighbouring signals and therefore electronically integratable) in supernatant specimens derived form the equilibration of both control (unpunctured) and exoskeletally punctured mealworm specimens are given in table 2. With the exception of acetate levels, the concentrations of all ^sup 1^H NMR- quantifiable metabolites in the punctured mealworm supernatants are clearly much greater than those of the unpunctured control samples, the differences observed between the mean levels of lactate, alanine, glycine, tyrosine, valine, betaine, carnitine, choline, dimethylamine, and N-acetylsugar levels being highly statistically significant (from p < 0.025 to p < 0.001).

The high- and low-field regions of typical 400 MHz ^sup 1^H NMR profiles of post-neutralised supernatants derived from the incubation of control and exoskeletally punctured cricket specimens in 0.01 mol[middot]dm^sup -3^ HCl (pH 2) at 37[degrees]C for a 2- hour period are shown in figure 3. As expected, the spectra acquired on the punctured sample supernatants contained many prominent resonances attributable to a wide range of low-molecular-mass biomolecules, whereas those of the control (unpunctured) specimens contained only resonances ascribable to betaine and N-acetylsugars, the former of very low intensity. Although a clear feature distinguishing the spectra of supernatants derived from exoskeletally punctured crickets from those of mealworms treated in the same manner was the presence of quite intense resonances arising from the [beta]-amino acid taurine (triplets located at [delta] = 3.24 and 3.43 ppm), the methylamine, dimethylamine, trimethylamine and tyrosine signals present in the spectra of the punctured mealworm sample supernatants were clearly absent from those of the corresponding cricket-derived supernatants, observations providing evidence for selective metabolic differences between these species.

Table 3. Mean + or - standard error weight-normalised supernatant metabolite concentration data for cricket specimens

The mean concentrations of the ^sup 1^H NMR-detectable metabolites acetate, formate, lactate, succinate, alanine, glycine, valine, taurine, betaine, carnitine, choline and creatine observed in the post-neutralised HCl supernatants of exoskeletally punctured cricket specimens (table 3) were significantly greater than those of the control samples, the elevations observed being biggest in those of acetate, formate, alanine and glycine (p < 0.001).

Fig. 3. ^sup 1^H NMR analysis of the nature and levels of components released from control and exoskeletally punctured insects. High-field regions of 400 MHz ^sup 1^H NMR spectra acquired on post-neutralised 0.01 mol [middot] dm^sup -3^ HCl supernatants derived from control (a) and exoskeletally punctured cricket (b) specimens respectively. Corresponding low-field regions of these spectra from control (c) and punctured cricket specimen (d). Lys = lysine [beta]-, [gamma]- and [delta]-CH^sub 2^ group protons; Tau = taurine-CH^sub 2^SO^sup -^^sub 3^. For all other abbreviations see legend to figure 2.

Discussion

The large increase in bioavailability of nutrients after the simulated digestion in 0.01 mol[middot]dm^sup -3^ hydrochloric acid clearly demonstrates the metabolic advantage of puncturing insect prey prior to swallowing. Although the digestive systems of insectivorous animals contain chitinases [Jeuniaux, 1961], the effect of these enzymes on degrading the exoskeletal matrices of mealworms and crickets was not modelled in our experiment. The puncture would also increase the area available for chitinases to act on via an enhanced level of ingress of digestive fluids into the prey body.

Jackson et al. [1992] found low absorption of chitin by-products in seabirds and suggested that chitinase was used to increase the rate of breakdown of prey exoskeletons to 'increase digesta transit' rather than serving a direct nutritive function. This view was supported by Smith et al. [1998], who investigated chitinase produced by the nine-banded armadillo (Dasypus novemcinctus) and in conclusion cast doubt on the value of chitinase to the animals' metabolic needs since the optimum operational pH (5.0) and temperature (50[degrees]C) of the enzyme differed substantially from those found in vivo.

Chitinases are also secreted by the pancreas in insectivorous mammals, so it is possible that digestion of chitin takes place further down the gut. However, some of the degradation of chitin may be ascribable to (1) the chitinolytic activity of enzymes present in the prey (fungi contain substantial amounts of chitin, so insects themselves probably contain chitinase) and (2) to commensal bacteria in the gut.

The mean low-molecular-mass N-acetylsugar concentrations of the control (unpunctured) prey specimen supernatants examined here suggest that the 0.01 mol[middot]dm^sup -3^ HCl medium employed can effect some hydrolysis of exoskeletal chitin at physiological temperature, with crickets yielding approximately 5 times as much N- acetylglucosamine than mealworms, an observation presumably reflecting the much greater surface area of a cricket. However, this phenomenon may also be explicable by a higher content of free N- acetylsugars and/or N-acetylglucosamine-containing oligosaccharides on the exoskeletal surface of crickets than on mealworms, and experiments to test this, together with the hydrolytic actions of chitinases, are currently in progress.

Intriguingly, puncturing the exoskeleton of mealworms gave rise to a substantial elevation in the mean concentration of N- acetylsugars in the acidic supernatant, whereas no significant differences between control and punctured cricket-derived supernatant levels were detectable. Hence, it would appear that the access of hydrogen ions to the interior surface of mealworms (including the internal chitinaceous surface of their exoskeleton) promoted the release of markedly higher levels of N-acetylsugars into the solution phase.

Although the exoskeleton of all insects is composed of chitin, the degree of cross-linking (tanning) differs between adults and larvae. The mechanical properties of the two types of prey differ: beetles have hard brittle shells whereas insect larvae typically have soft but tough exoskeletons [Alexander, 1968], two properties which determine the fracture behaviour. Agrawal et al. [1997] predicted that the fragmentation of a material is determined by the square root of Young's modulus divided by the square root of the toughness (toughness is the opposite of brittleness). Hence, it might be expected that a hard brittle beetle fragments more readily than a soft but tough mealworm. Thus, for comminuting hard brittle beetles, flat teeth might be more efficient than sharp pointed teeth; however, crushing the prey between two flat surfaces could cause some of its body contents to escape (i.e. 'spurt') from the predator's mouth.

Similar constraints operate for aquatic insects, which typically have piercing mouth parts rather than cutting mandibles that would allow food to float away before it could be swallowed.

Exceptions to the typical pattern of sharp pointed teeth are the aardvark (Oryteropodidae), anteater (Myrmecophagidae) and aardwolf (Carnivoral proteles cristalus), which have an exclusive diet of ants. These animals have long snouts, and hence there is little risk of loosing food items from the side of the mouth. The aardvark has two small peg-shaped molars on each side of each jaw, which wear to form a pestle and mortar (cusp/fossa) arrangement, implying that the aardvark grinds the ants prior to swallowing. The anteater, however, has no teeth and thus cannot grind its prey, although it should be noted that ants are brittle and may be fractured in the mouth. These small fractures in the exoskeleton may be as effective as a puncture. Although our study was performed with insects, it is possible that similar considerations apply to other prey animals, for which puncturing the skin may also assist in digestion.

Mastication by mammals delays both the ingestion of further prey items and chemical digestion by the gut [Alexander, 1994] and it is logical to suppose that an optimised digestive system will minimise oral transit time. The results given here suggest that exposure of the internal contents of insects can be achieved by just one or two chews with the tribosphenic molars typical of insectivorous mammals.

High-resolution ^sup 1^H NMR spectroscopy offers many valuable advantages over alternative analytical techniques in that it permits the simultaneous study of a wide range of biomolecular components in complex biological matrices. Moreover, the accumulation of such multicomponent information by the technique is rapid (spectral accumulation time is about 10 min per sample), and it also has the capacity to provide much useful molecular information regarding the identity of components not normally anticipated to be present in specimens examined. In view of these benefits, ^sup 1^H NMR analysis is also readily applicable to further nutrition-based research investigations, e.g., those involving determination of the identities and bioavailabilities of nutraceuticals, and we are currently conducting experiments in this research area.

In view of the results obtained, we conclude that dentition- mediated puncturing of the cuticles of insects and insect larvae by insectivorous mammals prior to swallowing offers major nutritional benefits over the swallowing of the prey whole.

Acknowledgments

We are very grateful to the University of London Intercollegiate Research Services for the provision of the NMR facilities and to Jane Hawkes for excellent technical assistance. We are also grateful to Mr. Barry Mills for helpful advice.

Folia Primatol 2003;74:126-140

DOI: 10.1159/000070646

KARGER

Fax +41 61 306 12 34

E-Mail karger@karger.ch

www.karger.com

(C)2003 S. Karger AG, Basel

0015-5713/03/0743-0126$19.50/0

Accessible online at: www.karger.com/fpr

References

Agrawal KR, Lucas PW, Prinz JF, Bruce IC (1997). Mechanical properties of foods responsible for resisting food breakdown in the human mouth. Archives of Oral Biology 42: 1-9.

Alexander RM (1968), Animal Mechanics. London, Sedgewick & Jackson.

Alexander RM (1991). Optimization of gut structure and diet for higher vertebrate herbivores. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences 333: 249-255.

Alexander RM (1994). Optimum gut structure for specified diets. In The Digestive System of Mammals (Chivers DJ, Langer P, eds.) pp 54-62. Cambridge, Cambridge University Press.

Alexander RM (1998). News of chews: The optimization of mastication. Nature 391: 329.

Cochran WG, Cox GM (1950). Experimental Designs. New York, John Wiley and Sons.

Grootveld M, Jain R, Claxson AWD, Naughton D, Blake DR (1990). The detection of irradiated food-stuffs. Trends in Food Science and Technology 1: 7-14.

Jablonski NG, Crompton RH (1994). Feeding behavior, mastication, and tooth wear in the western tarsier (Tarsius bancanus). International Journal of Primatology 15: 29-59.

Jackson SA, Place R, Seiderer LJ (1992). Chitin digestion and assimilation by seabirds. The Auk 109: 135-136.

Janis CM, Fortelius M (1988). On the means whereby mammals achieve increased functional durability of their dentitions, with special reference to limiting factors. Biological Reviews of the Cambridge Philosophical Society 63: 197-230.

Jeuniaux C (1961). Chitinase: An addition to the list of hydrolases in the digestive tract of vertebrates. Nature 192: 135- 136.

Lewin RA (1999). Merde. London, Arum Press.

Lynch E, Sheerin A, Silwood CJ, Grootveld M (1999). Multicomponent evaluations of the oxidising actions and status of a peroxoborate-containing tooth-whitening system in whole human saliva using high resolution proton NMR spectroscopy. Journal of Inorganic Biochemistry 73: 65-84.

Naughton DP, Haywood R, Blake DR, Edmonds S, Hawkes GE, Grootveld M (1993). A comparative evaluation of the metabolic profiles of normal and inflammatory knee-joint synovial fluids by high resolution proton NMR spectroscopy. FEBS Letters 332: 221-225.

Neeman M, Degani H (1989). Metabolic studies of estrogen-treated and tamoxifen-treated human-breast cancer-cells by nuclear magnetic- resonance spectroscopy. Cancer Research 49: 589-594.

Nicholson JK, O'Flynn MP, Sadler PJ, Macleod AP, Juul SM, Sonsken PH (1984). ^sup 1^H Nuclear magnetic resonance studies of serum, plasma and urine from fasting normal and diabetic subjects. Biochemical Journal 217: 365-375.

Nicholson JK, Foxall PJD, Spraul M, Farrant RD, Lindon JC (1995). 750-MHz ^sup 1^H and ^sup 1^H-^sup 13^C NMR spectroscopy of human blood plasma. Analytical Chemistry 67: 793-811.

Popowics TE, Fortelius M (1997). On the cutting edge: Tooth blade sharpness in herbivorous and faunivorous mammals. Annales Zoologici Fennici 34: 73-88.

Prinz JF, Lucas PW (1997). An optimization model for mastication and swallowing in mammals. Proceedings of the Royal Society of London Series B: Biological Sciences 264: 1715-1721.

Smith SA, Robbins LW, Steiert JG (1998). Isolation and characterisation of a chitinase from the nine-banded armadillo, Dasypus novemcinctus. Journal of Mammalogy 79: 486-491.

J.F. Prinz(a) C.J.L. Silwood(b) A.W.D. Claxson(b) M. Grootveld(b)

a Wageningen Centre for Food Science, Wageningen, The Netherlands;

b Department of Diabetes and Metabolic Medicine, Medical Unit, Barts and the Queen Mary's School of Medicine and Dentistry, London, UK

Dr. J.F. Prinz

Wageningen Centre for Food Science, PO Box 557

NL-6700 AN Wageningen (The Netherlands)

Tel. +31 317 485 383, Fax +31 317 485 384

E-Mail j.f.prinz@med.uu.nl

Copyright S. Karger AG May/Jun 2003

More News in this Category