The Journal of Nutrition Vol. 128 No. 6 June 1998, pp. 1042-1047

Nutritional Utilization by Rats of Chickpea (Cicer arietinum) Meal and Its Isolated Globulin Proteins Is Poorer than That of Defatted Soybean or Lactalbumin^1,2

Luis A. Rubio³, George Grant, Peter Dewey, David Brown, Maureen Annand,, Susan Bardocz, and Arpad Pusztai

The Rowett Research Institute, Bucksburn, Aberdeen AB2 9SB, UK

	ABSTRACT

Abstract Introduction Methods Results Discussion References

The effects on performance, digestibility, N utilization and plasma amino acid concentrations of dietary chickpea (Cicer arietinum, var. Kabuli) seed meal, globulin proteins or buffer-insoluble residue [starch + non-starch polysaccharides (NSP) + lignin] were studied in growing rats. Chickpea meal, defatted soybean meal, chickpea globulins and lactalbumin were each incorporated into diets as the sole source of dietary protein (100 g/kg). In addition, chickpea insoluble residue was included in a control diet in the same proportion found in the chickpea meal. Rats were killed while under halothane anesthesia after 10 d of consuming the diets, and ileal contents were washed out and freeze-dried for digestibility measurements. Weight gains and gain:feed ratios of rats fed chickpea diets for 10 d did not differ from those of rats fed defatted soybean but were significantly lower than those of rats given the control (lactalbumin) diet. However, ileal and fecal N digestibilities and N retention by rats fed the chickpea diet were significantly lower than those obtained with the lactalbumin or soybean diet. The inclusion of both chickpea meal or its globulin proteins in the diet significantly increased the amount of N excreted, primarily as urea, through the urine. However, although ileal N digestibility values for chickpea meal were significantly lower, those for its constituent globulins did not differ from control values. Urea levels in plasma in rats fed diets containing chickpea meal, globulins or soybean meal were significantly higher than in those fed lactalbumin. Furthermore, the concentrations of glycine, phenylalanine, histidine, arginine and ornithine in the plasma of rats fed chickpea meal, its globulins or defatted soybean were significantly higher, whereas those of threonine, leucine, lysine and tryptophan were significantly lower than lactalbumin-fed controls. The chickpea insoluble residue had no adverse effects on performance or N utilization by rats. We conclude that the low nutritional value of chickpea meal is likely to be due mainly to adverse effects of its globulin proteins on growth and N metabolism rather than to the action of any known antinutritional factor present in the diet.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Among legume crops, chickpeas (Cicer arietinum) have been studied only to a limited extent in animal nutrition. This is probably due to the fact that chickpeas have been and still are a staple food in many tropical and subtropical countries. However, the increasing shortage of protein for both human and animal consumption and the development of seed varieties potentially useful in practical animal feeding have led to an increased interest in the use of chickpeas as protein concentrates. Furthermore, chickpea seeds, particularly those from the kabuli varieties (light colored), have been reported to contain relatively low amounts of antinutritional factors (ANF)⁴ (Chavan et al. 1986, Grant et al. 1983 and 1995); in addition, they appear to be well tolerated by monogastric animals (Batterham et al. 1990, Savage and Thompson 1993).

Previous studies with other legumes such as faba beans and lupins suggest that their low nutritional value is due to the inefficient utilization of their proteins by test animals. This appears to be related to the poor nutritive quality of the major reserve proteins rather than to the presence of any known ANF (Rubio et al. 1991 and 1995). Recent work has also shown that the digestibility in vivo in the rat small intestine of isolated globulin preparations from faba bean, lupin and soybean (~90%) was in fact higher than that of heat-treated lactalbumin (Rubio et al. 1994), which is frequently used as a control protein. Therefore the low utilization of some legume seed proteins might be the result of putative harmful systemic effect(s) of the absorbed small amounts of undegraded legume proteins or their partial degradation products on N metabolism.

Accordingly, the aim of this work was to study the effects in rats of feeding diets based on chickpea meal or its main fractions (proteins, carbohydrates) to determine the nutritional value of the meal and identify the fraction(s) that might influence it. Additionally, because soybean is the legume crop broadly used at present as vegetable protein concentrate in feeds for monogastrics, a defatted soybean-based diet was also incorporated into the study and used for practical comparison.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Purification of fractions and chemical analysis.  Chickpea seeds (cv Kabuli) and defatted soybean were purchased locally. Lactalbumin, amino acids and heparin were obtained from Sigma (Poole, Dorset, UK). Globulins were purified from chickpea seed meal by extraction at pH 8.0 in 0.2 mol/L borate buffer and precipitation of the globulin proteins by lowering the pH to 4.5 with acetic acid (Danielsson 1949). The sediment after centrifugation (48,000 × g) was resuspended, dialyzed against distilled water (10,000 mol. weight cut-off membrane, 48 h, 4 changes) and freeze-dried. After buffer extraction, the insoluble residue containing the starch plus the insoluble fiber [non-starch polysaccharides (NSP) + lignin], was also recovered, dialyzed and freeze-dried (Rubio et al. 1994).

Amino acids were determined after hydrolysis of samples in 6 mol/L boiling HCl (2 mL/5 mg of protein) for 18 h, sulfur-containing amino acids after performic acid treatment, tryptophan after hydrolysis with 4 mol/L LiOH and free plasma amino acids after precipitation of proteins with 0.59 mol/L sulfosalicilic acid, with norleucine as an internal standard. Amino acids were determined by a Pharmacia LKB-Alpha Plus amino acid analyzer (Pharmacia, Herts, UK). Nitrogen was determined in a N analyzer (Foss Electric UK, York, UK), and in urine samples by a microKjeldhal method. Carbohydrate analyses were conducted by the method of Englyst and Cummings (1984). Amounts of free sugars in chickpea meal were estimated by HPLC after extraction with boiling 80% ethanol solution. Starch was determined as glucose after enzyme digestion (Aman and Hesselman 1984). Lignin was determined after hydrolysis in 7.35 mol/L H₂SO₄ of the acid detergent fiber fraction. Lipids in chickpea meal, dry carcasses and feces were extracted (1:100, wt/v) in chloroform methanol (2:1, v/v) for 24 h, and the difference in the sample weight before and after extraction was taken as the lipid content. Chromium in diets and ileal samples was determined by atomic absorption spectrophotometry after acid hydrolysis of the samples (Arthur 1970). Plasma urea was determined in a KONE Analyser (Espoo, Finland) and urinary urea in a Technicon Analyser (Technicon Instruments, Basingstoke, UK).

Animals, diets and feeding regimen.  Male Hooded-Lister rats (Rowett strain), reared and housed in the breeding and experimental small animal unit of The Rowett Research Institute, were used in the study. Rats were weaned at 19 d of age, after which they were fed stock diet (Biosure, Special Diets Services, Manea, Cambridgeshire, UK) for 7 d, followed by control diet (lactalbumin; Table 1) for 3 d to ensure their adaptation to experimental diets. Four rats, matched by weight (82 ± 1 g) were used per group. They were housed individually in Techniplast metabolism cages (Biotech, Alva, Clackmannanshire, UK). Food was withheld overnight before the start of the experiments. Rats were weighed daily. Water was freely available at all times. All management and experimental procedures conducted in this study were done in strict accordance with the requirements of UK Animals (Scientific Procedures) Act 1986 by staff personally licensed to conduct such procedures.

The diets (Table 1) were based on raw (heat-untreated) chickpea meal or legume fractions and contained the same amount of digestible energy (15.5 kJ/g) and protein (lactalbumin in controls or bean proteins in the experimental diets, 100 g/kg). Crude protein was calculated as N × 6.25 for lactalbumin and N × 5.5 for bean protein (Mossé 1990). Appropriate amounts of synthetic amino acids were added to legume- or legume protein-based diets taking into account their amino acid composition (Table 2) to equalize them to the levels in control (lactalbumin) diets, which met the requirement of young rapidly growing rats. The diets were supplemented with vitamins and minerals to target requirements (Grant et al. 1993). Chromium oxide (2 g/kg diet) was added to the diets as an indigestible marker. All rats were offered 12 g diet/d, which is ~80% of the normal intake of control rats with free access to the food. The chickpea residue diet contained the same amount of starch and insoluble material as the chickpea diet. N retention was calculated as [(N intake  N excreted)/N intake]. Net protein utilization (NPU) values were calculated by using previously determined basal N values obtained with rats of the same strain, weight and age, fed a protein-free diet for 10 d (Rubio et al. 1991).

Sampling procedures.  Animals were killed while under halothane anesthesia on d 10, exactly 2 h after receiving 2.5 g food. Blood was collected in preheparinized tubes (3 × 10⁴ IU heparin/L of blood). Ileal (0-25 cm up from the ileocecal junction) contents for digestibility measurements were washed out with ice-cold water, collected in plastic vials, immediately frozen and subsequently freeze-dried. Acidified urine and fecal samples were collected daily and stored at 20°C until required. Fecal samples were freeze-dried and ground in a mortar.

Statistical analysis.  The results were subjected to one-way ANOVA using the Minitab Statistical Software Package (Minitab, New York, NY). Differences between means were identified by Tukey's multiple comparison test with the use of the Instat software package (GraphPad, San Diego, CA).

	RESULTS

Abstract Introduction Methods Results Discussion References

Chemical analysis.  Protein (N × 5.5) constituted 190 g/kg of the total seed weight of chickpeas (Table 3). Most of the protein could be extracted in aqueous buffers at pH 8. However, a proportion (125 g/kg) of the total protein remained in the insoluble residue after extraction of the meal with buffer. Although chickpea was appreciably higher in starch than defatted soybean (405 and 46 g/kg, respectively), both contained similar amounts of total NSP (114-121 g/kg) and oligosaccharides (71-108 g/kg). However, NSP composition differed, with higher proportions of arabinose and lower galactose in chickpea NSP than in soybean NSP. Calculated from the composition of the buffer-insoluble residue, a considerable proportion (440 g/kg) of the NSP in chickpea meal was buffer soluble.

The essential amino acid composition of the globulin fraction was similar to that of the whole-seed protein (Table 2). Compared with lactalbumin, chickpea protein was particularly deficient in sulfur amino acids, tyrosine, leucine and tryptophan, but not lysine. The amount of arginine in chickpea meal and globulins was double that in lactalbumin. Defatted soybean was also lower than lactalbumin in sulfur amino acids and leucine, and higher in arginine.

Performance, digestibility and N utilization.  Performance indices (weight gains and gain:feed ratios) (Table 4) of rats fed diets containing whole chickpea meal as the only source of protein did not differ from those obtained with rats fed a soybean diet. However, they were inferior (P < 0.01) to those achieved by rats fed the control (lactalbumin) diet. Inclusion of the chickpea insoluble residue (starch + fiber) in the control diet did not affect any of these variables. Fecal excretion (dry weight and N) by rats fed chickpea or soybean was significantly (P < 0.01) greater than that by controls. Inclusion of the insoluble residue in control diets for rats slightly increased fecal output although the differences did not reach significance (P = 0.11). Rats fed chickpea meal or globulin proteins excreted high amounts of N (primarily urea) in their urine (Table 4). Consumption of soybean also increased urinary N output by rats but not to the same extent as did chickpea or its proteins. Rats fed chickpea residue- or lactalbumin-based diets had the lowest levels of urinary N excretion.

NPU values obtained for chickpea meal or globulins were significantly lower than those for control (lactalbumin) proteins (Table 4). In contrast, inclusion of the chickpea residue in control diet did not alter the NPU value obtained for the lactalbumin-control diet. With soybean, the NPU values fell between those for chickpea and the control diets. Ileal N digestibility values did not differ significantly among most of the test and control groups of rats (Table 4). Only in rats fed chickpea meal was ileal N digestibility lower than controls. However, fecal N digestibilities for chickpea meal, chickpea globulins and soybean meal were significantly lower than those for the lactalbumin-control diet.

N retention by rats was low when they were fed diets containing chickpea meal or globulins (Table 4). Soybean lowered N retention to a lesser extent, and inclusion of chickpea residue in diets had no significant effect of N retention.

Plasma urea and free amino acids.  The concentration of urea in plasma was significantly (P < 0.01) elevated in rats fed diets containing chickpea meal, globulins or soybean (Table 4). Among plasma amino acids (Table 5), concentrations of glycine, phenylalanine, ornithine, histidine and arginine were significantly (P < 0.01) higher in rats fed these diets. In contrast, valine and isoleucine concentrations were much lower in rats fed chickpea meal or globulins as were threonine, leucine, lysine and tryptophan levels in rats fed chickpea meal, globulin or soybean. Methionine was lowered only in rats fed soybean.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Performance, digestibility and N utilization.  The growth of rats was impaired when chickpea meal was included in diets as the only source of protein (Table 4). Thus, although this diet was equalized in energy and protein with the controls and supplemented with essential amino acids, weight gains and gain:feed ratios obtained with this diet were inferior to those with the control diet. This lower nutritional value of chickpea diets appeared to be due predominantly to interference with systemic protein metabolism. This was reflected in a higher excretion of N, particularly as urea through the urine. The lower efficiency of protein utilization was not due to the insoluble residue (starch + fiber) of the meal, because its inclusion in a well-balanced control diet had no detrimental effect on performance of rats or NPU values. Ileal N digestibility values for chickpea globulins did not differ significantly from those for the control diet. This appears to rule out the possibility that the reduced NPU values were due to lower net N absorption from the small intestine. These results fully agree with the previously reported high digestibilities of other isolated legume proteins in the rat small intestine (Rubio et al. 1994).

The digestibility of chickpea NSP was probably low, as reflected by the higher fecal outputs (dry weight) of rats fed diets containing chickpea meal or its insoluble residue. The higher excretion of N by rats fed diets containing chickpea meal or residue likely originated, at least in part, from microbial growth in the large intestine facilitated by the considerable amounts of NSP reaching the lower bowel (Mason 1984). Because the insoluble residue from chickpea could be included in the diet at levels comparable to that in the original meal without causing significant differences in nutritional performance of the rats, it can be concluded that these substances have little or no intrinsic detrimental effects. Similar findings were previously made for lupin and faba bean insoluble NSP and starch (Rubio et al. 1991 and 1995).

The inclusion of chickpea globulins in the diet had a profound effect on the nutritional performance of the animals (Table 4), with slightly lower feed intakes and condiderably reduced gain:feed ratios leading to poor weight gains compared with controls. The N retention and NPU values obtained for chickpea meal and chickpea globulins were very similar, suggesting that the lower nutritional value of chickpea meal was probably due largely to poor utilization of the globulin proteins in these seeds. Because ileal digestibilities for the globulins were not low but urine N excretion was high, it would appear that these proteins or their digestion products might have had a negative affect on the general protein metabolism of the animals. This agrees with previous observations with faba bean and sweet lupin seed meals (Rubio et al. 1991 and 1995), although the possibility still exists that the ileal absorption of one or several essential amino acid(s) is low, even under high total N digestibility (see, e.g., Sarwar and Peace 1986). However, information recently obtained by us with lupins (data not shown) and other authors in broiler chickens (Pérez et al. 1993) indicates that the true amino acid digestibilities of several protein sources (soybean, field pea, vetch and bitter vetch) in the lower ileum are not significantly different.

Among the known ANF in chickpea, only the trypsin inhibitors appear to be present in nutritionally important amounts (Chavan et al. 1986). This may explain the small reduction in ileal and fecal N digestibility found with chickpea-fed rats. It would also be consistent with the finding that these variables were unaltered on feeding the globulin proteins because the trypsin inhibitors would be mainly in the water-soluble (albumin) protein fraction (Sastray and Murray 1987). However, in contrast to the findings with soybean, pancreas relative weights (g/kg body weight) in chickpea-fed rats were not increased (data not shown). Therefore, as previously indicated (Batterham et al. 1990, Savage and Thompson 1993), these ANF appear to make only a limited contribution to the effects of chickpea on body metabolism.

Plasma urea and free amino acids.  There is an inverse correlation between the biological value of foodstuff and blood urea concentration in rats (Eggum 1970). Because urea is one of the main end products of protein catabolism in mammals, high plasma values are associated with disturbances in protein metabolism and increased protein degradation, which can finally result in a high loss of N through the urine. In this study, plasma urea concentrations were high in all groups of rats fed legume proteins. However, the total excretion of urea and N in urine differed according to the diet. Thus, it was very high with rats fed chickpea meal or globulins but much lower with rats fed soybean (Table 4). Das and Waterlow (1974) suggested that a poor-quality protein cannot support protein synthesis because it does not supply a balanced amino acid mixture. The diets used in these experiments were fully supplemented with essential amino acids up to the same values as in control diets, and the total N digestibility in the small intestine was high. Therefore, as previously suggested for faba bean and lupin globulins, the retention of N by rats fed chickpea globulins may be the result of differential rates of digestion of proteins or absorption of peptides (Rubio et al. 1995).

The diet containing chickpea residue fraction (Table 1), which still contained 12 g/kg of chickpea protein, also increased urea concentrations in plasma and urine (Table 4). Therefore it appears that chickpea proteins may be able to affect urea formation even when they are present in the diet only in small amounts. However, wasteful urea production did not appear to be the only mechanism involved in growth depression found with chickpea-fed rats because N digestibility at the terminal ileum was also reduced with this diet. Thus digestion and absorption of N as well as retention of absorbed N were adversely affected by inclusion of chickpea meal in the diet.

Disturbances in protein metabolism were also indicated by the changes in the free concentrations of some amino acids in plasma. Thus, some amino acids (glycine, phenylalanine, histidine, arginine and ornithine) were increased in animals fed legume diets, possibly as a result of higher dietary intakes (Table 2). In contrast, the concentrations of some other amino acids (particularly threonine, leucine, lysine and tryptophan) were substantially lower than those found in plasma from lactalbumin-fed controls even though the amino acid profiles of the diets had been equalized by supplementation with individual amino acids. Lysine was particularly affected; it was 15-51% lower in plasma from rats fed chickpea or soybean than in that from controls. Therefore, for reasons as yet unknown, dietary amino acids from legume seed meals seem to be absorbed from the gut but not subsequently utilized for protein synthesis in an efficient manner. As a result, a significant proportion are catabolized, leading to increased urea formation and excretion.

In conclusion, the nutritional performance of rats fed diets based on chickpea meal as the only source of protein was similar to that of rats fed soybean but inferior to that of controls fed a lactalbumin-based diet. This appeared to be due mainly to the adverse effects of the globulin proteins on N metabolism and growth rather than to the action of ANF present in the diet. It is suggested that the increased urea production and loss of N through the urine is due predominantly to increased protein catabolism. The mechanism(s) involved are unknown at present.

	FOOTNOTES

Manuscript received 21 October 1997. Initial reviews completed 1 December 1997. Revision accepted 9 February 1998.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Aman P., Hesselman K. Analysis of starch and other main constituents of cereal grains. Swed. J. Agric. Res. 1984; 14:135-139
• Arthur D. The determination of chromium in animal feed and excreta by atomic absorption spectrophotometry. Can. Spectrosc. 1970; 11:134-140
• Batterham E. S., Andersen L. M., Saini H. S., Baigent D. R. Tolerance of growing pigs to trypsin and chymotrypsin inhibitors in chickpea (C. arietinum) and pigeon pea (C. cajan) meals. Proc. Aust. Soc. Anim. Prod. 1990; 18:453
• Chavan J. K., Kadam S. S., Salunke D. K. Biochemistry and technology of chickpea (C. arietinum) seeds. CRC Crit. Rev. Food Sci. Nutr. 1986; 25:107-132
• Danielsson C. E. Globulins of the Gramineaceae and Leguminosae. Biochem. J. 1949; 44:387-400[Medline]
• Das T. K., Waterlow J. C. The rate of adaptation of urea cycle enzymes, aminotransferases and glutamic dehydrogenase to changes in dietary protein intake. Br. J. Nutr. 1974; 32:353-373[Medline]
• Eggum B. O. Blood urea measurement as a technique for assesing protein quality. Br. J. Nutr. 1970; 24:983-988[Medline]
• Englyst H. N., Cummings J. H. Simplified method for the measurement of total non-starch polysaccharides by GLC of constituent sugars and alditol acetates. Analyst 1984; 95:181-193
• Grant G., Dorward P., Pusztai A. Pancreatic enlargement is evident in rats fed diets containing soybean (G. max) or cowpea (V. ungiculata) for 800 days but not in those given diets based on kidney bean (P. vulgaris) or lupin seed (L. angustifolius). J. Nutr. 1993; 123:2207-2215
• Grant G., Edwards J. E., Pusztai A. -Amylase inhibitor levels in seeds generally available in Europe. J. Sci. Food Agric. 1995; 67:235-238
• Grant G., More L. J., McKenzie N. H., Stewart J. C., Pusztai A. A survey on the nutritional and haemagglutination properties of legume seeds generally available in the U.K. Br. J. Nutr. 1983; 50:207-214[Medline]
• Mason V. C. Metabolism of nitrogenous compounds in the large intestine. Proc. Nutr. Soc. 1984; 43:45-53[Medline]
• Mossé J. Nitrogen to protein conversion factor for ten cereals and six legumes or oilseeds. A reappraisal of its definition and determination. Variation according to species and to seed protein content. J. Agric. Food Chem. 1990; 38:18-24
• Pérez L., Fernández-Fígares I., Nieto R., Aguilera J. F., Prieto C. Amino acid ileal digestibility of some legume seeds in growing chickens. Anim. Prod. 1993; 56:261-267
• Rubio L. A., Grant G., Bardocz S., Dewey P., Pusztai A. Nutritional response of growing rats to faba beans (V. faba) and faba bean fractions. Br. J. Nutr. 1991; 66:533-542[Medline]
• Rubio L. A., Grant G., Cavallé C., Martinez-Aragón A., Pusztai A. High in vivo (rat) digestibility of faba bean (V. faba), lupin (L. angustifolius) and soybean (G. max) globulins. J. Sci. Food Agric. 1994; 66:289-292
• Rubio L. A., Grant G., Scislowsky P., Brown D., Annand M., Pusztai A. The utilization of lupin (L. angustifolius) and faba bean globulins by rats is poorer than of soybean globulins or lactalbumin but the nutritional value of lupin seed meal is lower only than that of lactalbumin. J. Nutr. 1995; 125:2145-2155
• Sarwar G., Peace R. W. Comparisons between true digestibility of total nitrogen and limiting amino acids in vegetable proteins fed to rats. J. Nutr. 1986; 116:1172-1184
• Sastry M.C.S., Murray D. R. The contribution of trypsin inhibitors to the nutritional value of chickpea seed protein. J. Sci. Food Agric. 1987; 40:253-261
• Savage, G. P. & Thompson, D. R. (1993) Effect of processing on the trypsin inhibitor content and nutritive value of chickpeas (Cicer arietinum). In: Recent Advances of Research in Antinutritional Factors in Legume Seeds (Van der Poel, A.F.B., Huisman, J. & Saini, A.S., eds.), pp. 435-440. Wageningen, The Netherlands.

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences