Protein degradation in the rumen of sheep and cattle

Sheep and steers were given, at a maintenance level of feeding, four diets consisting of either poor quality dried grass, good quality dried grass or separate mixtures (63:35) of each of the dried grasses and barley. Ammonia and total N concentrations in rumen liquor were significantly higher in sheep than in steers whereas total volatile fatty acid concentration was significantly lower and protein N concentration, pH and rumen fluid dilution rate did not differ significantly between species. For all rumen measurements, except total volatile fatty acid concentration, there were significant differences between diets with the dietary responses being similar in both species. Protein degrading activity in the rumen was measured in vitro with casein as the substrate and in situ by measuring N disappearance when soyabean meal, cotton seed meal, groundnut meal, meat and bone meal, fish meal and dried grass were incubated in polyester bags in the rumen. Casein degrading activity of rumen liquor did not differ significantly between species, whereas rumen in-situ degradation of all feedstuffs, except fish meal, was significantly higher in sheep. In both species, ruminal casein degrading activity was higher when the good quality forage was given than when the poor quality forage was given and also increased when part of each forage was replaced by barley. In contrast, rumen in-situ degradability of feedstuffs did not differ when the two all-forage diets were given and the inclusion of barley in the diet reduced the rate of degradation. In both species and with all diets the rumen in-situ degradability ranking of the feedstuffs was similar.     

Evaluation of the effect of formic acid and level of formaldehyde application before ensiling on silage protein degradability

The effect of the application, before ensiling, of formic acid alone, and together with increasing levels of formaldehyde, on the degradability of the protein of ryegrass and red clover silages has been assessed on the basis of nitrogen solubility in mineral buffer; susceptibility of N to degradation during in-vitro incubation with either rumen microorganisms, acid pepsin or neutral protease; and N disappearance when the silages were incubated in situ in Dacron bags in the rumen of sheep receiving dried grass. The relative effects of the additives were generally consistent for both crops and with all procedures: formic acid either had no effect or reduced degradability by only a small amount, whereas a mixture of formic acid and formaldehyde was more effective than formic acid alone in protecting protein from degradation, and degradability decreased in a curvilinear manner with increasing levels of formaldehyde application. Absolute values for protein degradability based on buffer solubility and in-vitro degradation by rumen microorganisms were very similar but lower than those based on digestion with proteolytic enzymes which in turn were lower than those obtained with the rumen in situ procedure. Buffer solubility and in-vitro incubation with rumen microorganisms also showed much bigger differences between the formic acid-treated and the formaldehyde-treated silages than the other methods.     

Methylene Chloride Extraction of Gossypol from Cottonseed Products

Methylene chloride was used to reduce the amount of free and total (free plus bound) gossypol in hexane-defatted meal and the liquidcyclone-processed (LCP) underflow fraction of glanded cottonseeds from 2.6% and 3.4% to 0.013% and 0.15%, respectively (the accepted level of free gossypol in cottonseed products for food is 0.045%). The cottonseed meals were pretreated one of three ways to rupture the gossypol glands: (a) equilibrated with additional water; (b) suspended in various water-propylene glycol mixtures; or (c) mixed with an acetic acid-water-propylene glycol solution. The gossypol was then readily extracted from the meals with methylene chloride. Low levels of water and acetic acid in propylene glycol aided methylene chloride in the removal of free and total gossypol and did not greatly alter the proximate composition, solubility, and gel electrophoretic properties of proteins; amino acid content; and chemical scores of the treated meals. Success with this process should improve the potential of LCP-cottonseed by-product (underflow) as feed or food.     

挤压膨化温度对棉粕游离棉酚及营养成分的影响

本文研究了不同挤压膨化温度(90、100、110、120和130℃)对棉粕中游离棉酚和营养成分的影响。结果表明棉粕挤压膨化脱酚的最佳温度是120℃,超过这一温度后游离棉酚的降解速率不再随温度的升高而增加。随着挤压膨化温度升高棉粕蛋白氮溶解指数(NSI)迅速降低,棉粕中粗纤维、有效赖氨酸和总赖氨酸的含量也随温度升高而显著降低,但当温度达到120℃时,总氨基酸(AA)和必需氨基酸(EAA)的含量达到最大值。此外,120℃时,棉粕中支链氨基酸(BCAAs)的含量也最高。所以,挤压膨化温度在120℃时,既可以显著降低游离棉酚的含量又可较好的保持棉粕的营养价值。      

Effects of different protein supplements on milk production and nutrient utilization in lactating dairy cows

Sixteen (8 ruminally cannulated) multiparous and 8 primiparous lactating Holstein cows were used in 6 replicated 4 × 4 Latin squares to test the effects of feeding supplemental protein as urea, solvent soybean meal (SSBM), cottonseed meal (CSM), or canola meal (CM) on milk production, nutrient utilization, and ruminal metabolism. All diets contained (% of DM) 21% alfalfa silage and 35% corn silage plus 1) 2% urea plus 41% high-moisture shelled corn (HMSC), 2) 12% SSBM plus 31% HMSC, 3) 14% CSM plus 29% HMSC, or 4) 16% CM plus 27% HMSC. Crude protein was equal across diets, averaging 16.6%. Intake and production were substantially reduced, and milk urea, blood urea, and ruminal ammonia were increased on urea vs. the diets supplemented with true protein. Although intake was lower in cows fed SSBM compared with CM, no differences were observed for milk yield among SSBM, CSM, and CM. Yields of fat and protein both were lower on CSM than on CM, whereas SSBM was intermediate. Milk urea and milk protein contents also decreased when CSM replaced SSBM or CM. Diet did not affect ruminal volatile fatty acids except that isobutyrate concentration was lowest on urea, intermediate on CSM, and greatest on SSBM and CM. Urinary excretion of urea N and total N was greatest on urea, intermediate on SSBM and CM, and lowest on CSM. Apparent N efficiency (milk N/N intake) was lower on the CSM diet than on the SSBM diet. Overall, production and N utilization were compromised when the diets of high-yielding dairy cows were supplemented with urea rather than true protein and the value of the true proteins, from most to least effective, was in the order CM > SSBM > CSM.     

The effect of methanal (formaldehyde) treatment of casein on its digestion in vivo and in vitro

The effect of methanal (HCHO) treatment of casein on its in vitro and in vivo digestion was investigated. As the quantity of HCHO bound to casein increased, the in vitro rumen degradability decreased but no further reduction occurred above 10.6 g HCHO kg^?1 treated casein DM. The proportion of the original HCHO (25.4 g HCHO kg^?1 treated casein DM) remaining bound to casein at pH values 3.0 to 6.0 was approximately 0.5; the proportion remaining bound at pH 2.0 was significantly less (0.39). N solubility and in vitro digestion of casein by pepsin/HCl and/or intestinal proteolytic enzymes was significantly reduced by treatment with HCHO (10.6 g HCHO kg^?1 treated casein DM). When fed to rats the dry matter (DM) and true N digestibilities decreased significantly as the ratio HCHO;casein increased; the digestibility of the N-free component of the diet remained unchanged. The quantity of available lysine in the diets and their true biological value (BV) decreased as the level of bound HCHO increased; there was significant linear relationship between BV and available lysine. The apparent disappearances of amino acids from the small intestine of sheep, when untreated or HCHO treated caseins (12.3 g HCHO kg^?1 DM) were infused into the duodenum, were not affected by the pH of the infusate (pH 2.5 or 6.8) but were significantly reduced by HCHO treatment. It is concluded that exposure of HCHO-treated casein to low pH values does not completely release bound HCHO from casein and that the intestinal digestion of casein to which HCHO remains bound is reduced when compared to untreated casein. All amino acids determined showed similarly reduced apparent intestinal absorptions due to HCHO treatment. It is suggested that the lower BV of HCHO-treated caseins was due to reduced metabolic availability of lysine.     

棉粕蛋白及其枯草蛋白酶酶解产物的组分分析

通过常规测定、凝胶层析、高效液相色谱等3种方法，测定和分析了棉粕蛋白及其枯草蛋白酶酶解产物的组份，并进行比较研究。结果表明，棉粕经过枯草杆菌蛋白酶酶解之后，可溶性蛋白比例比原来增加了1倍，游离氨基酸的含量增加了2.9倍；棉粕原料和酶解后的蛋白质产物经过SephadexG-50凝胶层析分离后均得到2个主要蛋白峰，并且它们相对分子质量分别为16300u和7950u左右，而酶解物2个峰的蛋白质含量比棉粕原料中的少；通过高效液相色谱分析，棉粕蛋白含有较多大分子蛋白，酶解之后大分子蛋白含量下降了7.71％，小分子肽的种类与含量明显增加，含量由16.36％增加至63.27％，游离氨基酸的含量由4.9％增加至19.4％。     

棉酚降解菌的筛选、鉴定及棉粕固态发酵效果研究

目的:筛选分离出棉酚降解菌株,研究其生长降解特性和对棉粕中棉酚脱毒效果,为深入研究微生物降解棉酚的机制提供依据。方法:从山东棉花产地采集土壤样本,以醋酸棉酚为唯一碳源,稀释涂布法分离筛选出一株具有棉酚降解能力的菌株,通过16S rRNA基因序列系统发育分析以及生理生化特征对菌株进行初步鉴定,确定菌属。用高效液相色谱测定其对棉酚的降解和棉粕固态发酵脱毒效果。结果:筛选到一株可以降解棉酚的细菌,命名为YL01,初步鉴定为拉乌尔菌属Raoultella sp.。该菌株除利用棉酚外,还可以邻苯二酚、原儿茶酸和对苯醌为唯一碳源生长。YL01在以棉酚为唯一碳源的无机盐液体培养基中生长10 d后,棉酚的降解率为48%;接种该菌株到棉粕中,通过固态发酵,8 d内棉粕中棉酚含量下降43%,蛋白含量增加10%。结论:本研究分离筛选到一株棉酚降解菌Raoultella sp. YL01,为进一步深入研究棉酚的微生物降解机制奠定了基础,其对运用生物处理技术消除棉粕中棉酚具有潜在的应用前景。     

Condensed tannins from Lotus corniculatus and Lotus pedunculatus exert different effects on the in vitro rumen degradation of ribulose‐1,5‐bisphosphate carboxylase/oxygenase (Rubisco) protein

Condensed tannins (CT) or proanthocyanidins (PA), which occur in a restricted range of forages, have the ability to interact with proteins and enzymes and can influence the digestion of plant protein in the rumen. We compared the effects of CT extracts from Lotus corniculatus and pedunculatus on degradation of the principal leaf protein, ribulose‐1,5‐bisphosphate carboxylase/oxygenase (Rubisco), by rumen microorganisms. Total soluble leaf protein extracted from white clover (Trifolium repens ) was incubated with fresh rumen fluid from sheep and a range of concentrations of each CT extract. The rate of degradation of the large (LSU) and small subunit (SSU) of Rubisco was quantified by fractionating the proteins in samples taken from in vitro rumen incubations using sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS‐PAGE) and imaging densitometry. To deduce the effects of the CT extracts, experiments were performed in the presence (CT inactive) and absence (CT active) of polyethylene glycol (PEG; MW 3350). The two CT extracts differed markedly in their effects on the degradation of the LSU and SSU of Rubisco. At concentrations of 0.89 and 1.79 mg CT mg ⁻¹ total soluble leaf protein, the CT extract from L pedunculatus was more effective at preventing the degradation of the LSU and SSU by rumen microorganisms than the CT extract from L corniculatus. At a concentration of 1.79 mg CT mg ⁻¹ total soluble leaf protein, the CT extracts from L corniculatus and pedunculatus prevented about 0.75 and 0.83 of the LSU and about 0.69 and 0.86 of the SSU, respectively, from being degraded. Addition of PEG removed the inhibition and almost complete degradation of these proteins occurred, as was the case in incubations without CT extracts. The results of this study suggest that the concentration of CT in the diet and the chemical structure which affects the activity of the CT needs to be considered when assessing the effects of CT on protein metabolism in ruminants. © 1999 Society of Chemical Industry     

Effect of Masonex and Forms of Cottonseed Hulls on Dairy Cows

Thirty-six lactating Holstein cows were in a 5 × 3 factorial partially balanced incomplete block design with three missing categories to study effects of different forms of cottonseed hulls and liquid supplements on milk production and composition. Roughages were regular cottonseed hulls, pelleted cottonseed hulls, pelleted cottonseed hulls with 9% fat, pelleted undelinted cottonseed hulls, and pelleted undelinted cottonseed hulls with 9% fat. Liquid supplements were 8% Masonex² (hemicellulose extract) and 8% cane molasses. Control had no supplement. All rations were adjusted to contain 30% cottonseed hulls. Roughages with added fat gave total rations of 2.5% added fat (air dry).Least square means for daily intake of dry matter, milk yield, and fat percent were regular cottonseed hulls 21.6 kg, 20.3 kg, 3.37%; pelleted cottonseed hulls 20.3 kg, 21.4 kg, 3.06%; pelleted cottonseed hulls plus fat 20.1 kg, 21.2 kg, 2.51%; pelleted undelinted cottonseed plus fat 19.4 kg, 20.9 kg, 2.73%. Pelleted cottonseed hulls increased milk yield, decreased dry matter intake and milk fat percent, but did not affect milk fat yields. Rations with pelleted undelinted cottonseed hulls resulted in higher milk fat percent and body weight than pelleted cottonseed hulls. Added fat decreased milk fat percent and yield because of high degree of unsaturation. Liquid treatments produced no detectable effects on dry matter intake, milk yield, or fat percent.     

Absence of Effect after Introducing Bacillus thuringiensis Gene on Nutritional Composition in Cottonseed

ABSTRACT To assess the effect of nutritional composition of cottonseed after introducing Bacillus thuringiensis gene, compositional analyses of Bollgard II cotton event 15985, DPH37B, and 5415 were carried out. Cotton DPH37B contains cry1Ac gene, and Bollgard II 15985 contains cry2Ab and cry1Ac genes. Cotton 5415 was a control. Crude protein, crude fiber, fat, moisture, ash, amino acids level, and selected minerals were measured by chemical analysis. Anti‐nutrients including free gossypol, total gossypol, and tannin were also measured. The results showed that the compositional components of the transgenic cottonseed were comparable to those of the nontransgenic cottonseed. The results of numerous qualitative and quantitative analyses revealed that the transgenic cottonseed was compositionally equivalent to the conventional counterpart.     

Degradation of starch by incubation with rumen fluid. A comparison of different starch sources

The degradability of starch from various feedstuffs was investigated in vitro by incubation of 500-mg amounts in 50 ml of a 3:1 rumen fluid/buffer solution at 39°C for 6 h. The rumen fluid was obtained from one of three cows fed on hay or hay and concentrate. The degree of degradation after 6 h incubation varied strongly for the 23 feedstuffs investigated. The degradation of starch from the same feedstuff in rumen fluid from a hay-fed cow was significantly lower than in rumen fluid from a concentrate-fed cow. It seemed that differences in degradability between feedstuffs were not determined by the ration of the donor cow, but merely by the properties of the starch. Processed feedstuffs showed a higher degradation of their starch than the unprocessed feedstuffs, independent of the ration of the donor cow. Particle size influenced degradation, but not of the starch of tapioca meal. A fairly constant ranking in degradability between the various feedstuffs was found. Fermentation of mixtures of feedstuffs showed about the same rate of degradation as found for the single products. Only when great differences in the degree of degradability existed was the degradation of the total starch enhanced. The time of collection of rumen fluid strongly influenced the in-vitro degradation of starch.     

EVALUATION OF EXTRUSION COOKED COTTONSEED/SOYBEAN BLENDS OF DIFFERENT PROPORTIONS

Cottonseed kernel/full-fat soy flour blends of different proportions were cooked in a Brady low-cost extruder. Total and free gossypol, available lysine, proximate analysis, and PER and NPU values varied directly with composition. Reduction in free gossypol as a result of extrusion was constant, and independent of composition (ca. 90%). The 25 and 50% cottonseed blends exhibited low free gossypol (within limits permitted for human consumption), high available lysine, high PER and NPU values, and amino acid patterns equalling or exceeding the FAO/WHO (1973) children's pattern, except for moderate deficiencies in total sulfur amino acids and small to negligible deficiencies in valine. Both blends resembled full-fat soy products in proximate analysis. These results indicate that extension of soy with cottonseed products up to a level of 50%, utilizing the procedure described in this work, would be feasible.     

Nitrogen in browse species: ruminal degradability and post-ruminal digestibility measured by mobile nylon bag and in vitro techniques

This study determined nitrogen degradability and digestibility of rumen undegradable nitrogen using mobile nylon bag (MNB) and pepsin/pancreatin in vitro technique (IV) of 40 browse species. Thirty Ethiopian highland sheep fitted with rumen cannulae were used in nitrogen (N) degradability studies. Six steers fitted with rumen cannulae were used in preparation of 16-h and 24-h ruminal undegraded residues and four steers fitted with distal abomasal cannulae were used in MNB technique. The browses varied widely in nitrogen solubility (15–468 g kg⁻¹), potential degradability (223–976 g kg⁻¹), rate of degradation (0·13–24% h⁻¹) and effective degradability (135–821 g kg⁻¹). The apparent N digestibility (ND) of the rumen undegraded residues differed significantly (P<0·05) among browse species. No significant difference (P>0·05) was observed in ND of 16-h and 24-h residues. The ND of the 16-h residue varied from -218 to 759 g kg⁻¹ and 169 to 851 g kg⁻¹ for MNB and IV methods, respectively. Browse species with high tannin contents such as Acacia hockii, A horrida, A melanoxylon, A persiciflora, A salicina, A saligna and Flemingia macrophylla had high rumen by-pass and a low ND, while Sesbania spp and A nilotica with low tannin contents underwent rapid and extensive dry matter and nitrogen degradation in the rumen. Acacia sieberiana, Chamaecytisus palmensis, Erythrina spp, Gliricidia sepium, Samanea saman and Enterolobium cyclocarpum had high proportions of protein escaping rumen degradation (BP ) and with a high proportion of the by-pass protein digested in the intestine, therefore these browses had a high potential as protein supplements. The ND measured with the MNB were significantly lower (P<0·001) than by the IV method. The correlation between MNB and IV was high and significant (R²=0·89, P<0·0001) as also indicated by the regression equation (SE in parentheses): MNB=-22·8 (4·55)+1·0 (0·08)IV (RSD=10·56, R²=0·79, n=40, P<0·001). The intercept of the linear relationship obtained was different from zero while the slope was not different from unity. Multiple regression analysis suggested that some of the unexplained variation could be accounted for by either nitrogen, acid detergent fibre, total phenolics or neutral detergent fibre bound tannin levels in browses. The IV method is accurate for estimating digestibility of ruminally undegradable N, and hence its use would considerably reduce the need for delicate surgery and the elaborate procedures involving the MNB technique. © 1998 SCI.     

Nitrogen solubility in three solvents and in situ protein degradability of ruminant feedstuffs

Feedstuffs commonly fed to ruminants were assayed for nitrogen solubility by using McDougall's buffer, 0·02 M sodium hydroxide (NaOH), or 0·15 M sodium chloride (NaCl) as solvents. In addition, in situ dry matter and protein disappearance from the same feedstuffs were determined using the nylon bag technique in ruminally cannulated sheep for varying times of incubation. The mean effective nitrogen disappearance ranged from 235 for maize to 894 g kg⁻¹ of total N for lupins, and dry matter disappearance from 240 for meat and bone meal to 793 g kg⁻¹ for lupins. Protein solubility was lowest (<10% of total N) for oilseed by-products and animal and fish by-products, intermediate (15–30% of total N) for some cereals and highest (35–45% of total N) for wheat varieties and plant protein sources. Furthermore, solubility and degradability data for various feed proteins are presented which demonstrate the variability in solubility and degradability of ruminant feedstuffs due to protein type or processing. © 1997 SCI.     

Preparation of Low Free Gossypol and High Available Lysine Cottonseed/Soybean Blends

Cottonseed/soybean blends (50/50) were ground in an Alpine pin mill with or without previous cooking in a Brady low-cost extruder. Free gossypol decreased with increasing amount of water added prior to extrusion and number of passes through the extruder. Available lysine increased with decreasing amount of added water but was unaffected by number of passes through the extruder. Type of soybean raw material utilized had no effect on free gossypol. Blends prepared from defatted cottonseed meal, however, had lower free gossypol than those made from cottonseed kernels. A blend similar to full-fat soy flour but of lower cost was prepared from unextruded 50% defatted cotton-seed meal/50% extruded soybean flakes, without added water, and ground in the Alpine mill.     

Nitrogen metabolism in the rumen

Protein metabolism in the rumen is the result of metabolic activity of ruminal microorganisms. The structure of the protein is a key factor in determining its susceptibility to microbial proteases and, thus, its degradability. Ruminal protein degradation is affected by pH and the predominant species of microbial population. Ruminal proteolytic activity decreases as pH decreases with high-forage dairy cattle-type rations, but not in high-concentrate beef-type rations. Accumulation of amino acid (AA) N after feeding suggests that AA uptake by rumen microorganisms could be the limiting factor of protein degradation in the rumen. In addition, there are several AA, such as Phe, Leu, and Ile, that are synthesized by rumen microorganisms with greater difficulty than other AA. The most common assessment of efficiency of microbial protein synthesis (EMPS) is determination of grams of microbial N per unit of rumen available energy, typically expressed as true organic matter or carbohydrates fermented. However, EMPS is unable to estimate the efficiency at which bacteria capture available N in the rumen. An alternative and complementary measure of microbial protein synthesis is the efficiency of N use (ENU). In contrast to EMPS, ENU is a good measurement for describing efficiency of N capture by ruminal microbes. Using EMPS and ENU, it was concluded that optimum bacterial growth in the rumen occurs when EMPS is 29 g of bacterial N/kg of fermented organic matter, and ENU is 69%, implying that bacteria would require about 1.31 x rumen-available N per unit of bacterial N. Because the distribution of N within bacterial cells changes with rate of fermentation, AA N, rather than total bacterial N should be used to express microbial protein synthesis.     

Canola Meal Versus Cottonseed Meal as the Protein Supplement in Dairy Diets

Four trials were conducted to evaluate effects of incorporating cottonseed meal or canola meal in dairy diets. In Experiment 1, 36 Holstein cows, 12 first lactation and 24 second or more lactations, were in two groups fed complete mixed diets containing either cottonseed meal or canola meal as the protein supplement. Yields of milk and milk components and feed intake were not affected by protein supplement. In Experiment 2, 6 Holstein cows in first lactation were fed diets of Experiment 1 in two periods of 3 wk to evaluate effects of protein supplement on milk composition. Protein supplement did not affect fat, lactose, ash, total solids, total nitrogen, casein nitrogen, whey protein nitrogen, or nonprotein nitrogen content in milk. Diets of Experiment 1 were fed to four rumen-fistulated heifers (Experiment 3) for changes of ruminal fluid characteristics. Ruminal fluid samples were collected at –1, 2, 4, and 8 h postfeeding. Volatile fatty acid and ammonia concentrations of ruminal fluid were not different for protein supplement. In Experiment 4, in situ disappearance of nitrogen and dry matter of protein sources in the rumen of a Holstein cow was studied. Disappearance of nitrogen and dry matter of protein supplements from nytex nylon bags was similar.     

Effect of rumen protein degradability on milk yield of dairy cows in early lactation

Twenty-four mature Holstein cows were fed diets of 40% corn silage and 60% concentrate (dry matter) beginning at parturition through wk 16 of lactation. A control concentrate (corn, soybean meal, and barley) was fed through wk 4 followed by assignment of cows to either a concentrate of low or high rumen protein degradability. In situ trials with two fistulated cows fed similar diets yielded rumen protein degradabilities of 78.5, 70.3, 69.9, 67.3, 49.1, and 36.5% for barley, corn, corn gluten feed, soybean meal, brewer's grains, and cottonseed meal. The low degradability concentrate (corn, cottonseed meal, brewer's grains, and corn gluten feed) had an estimated rumen protein degradation of 52.9% and a total ration crude protein of 14.3%. The high degradability concentrate containing corn, barley, and soybean meal was 72.8% rumen degradable, and total ration protein for this treatment was 14.5%. Dry matter intakes were 21.0 and 22.0 kg/day for the low and high degradability diets. Milk yield, fat percent, and fat-corrected milk were not affected by treatment. Milk protein percent and protein yield decreased from 3.00 to 2.84% and 1.07 to .99 kg/day in the high and low degradability diets. Efficacy of use of degradability as a criterion for feed formulation is questioned until understanding of both feed protein breakdown and microbial synthesis is greater.     

Nutrient composition and the use of solubility to estimate rumen degradability of food proteins in cattle

The use of protein solubility to estimate protein degradability in cattle was tested using 12 by‐products representing agricultural and distillers' food raw materials (RM). Agricultural RM included cereals (rice bran, maize gluten feed), legumes (field beans, soybean hulls) and oil‐seeds (rape meal, sunflower meal). Distillers' RM involved dark grains (DG1, DG2, DG3) and malts (MT1, MT2, MT3). Soluble proteins, as proportions of total soluble N (TSN), were extracted from RM and their undegraded residues (RU) after 18 h of in sacco rumen incubation in cattle (dg₁₈) by using water (albumin), salt (globulin), acid (glutalin 1) and alkali (glutalin 2) in succession. RM and RU differed significantly for all soluble proteins (P < 0.001). While albumin was the major protein in legume RM (0.74TSN), glutalin 1 was the major protein in legume RU (0.54TSN). Cereals were the most variable group, where maize gluten contained five times more albumin and two times more TSN in RM and three times more glutalin 1 + 2 in RU than those in rice bran. Distillers' RM contained slightly less albumin (0.53 vs 0.56TSN) but over twice as much glutalin 1 (0.29 vs 0.12TSN) as agricultural RM. Albumins were the most (0.55TSN, RM; 0.16TSN, RU) and glutalin 1 the least (0.21TSN, RM; 0.48TSN, RU) degradable proteins. There were moderate to strong correlations between protein solubility, dg₁₈ and acid detergent‐insoluble N (ADIN). ADIN correlated satisfactorily with glutalin 1 + 2 (r = ?0.55, RM and ?0.70, RU), TSN (r = ?0.75, RM and RU) and slowly degradable N, (SDN; r = ?0.70, all). TSN in all RM related well with SDN (r = 0.65) and dUN (r = 0.72). TSN in distillers' RM correlated closely with SDN (r = 0.91) and dUN (r = 0.96). Glutalin 1 in distillers' foods correlated extremely well (r = 0.94, RM and 0.93, RU) with dUN. Globulins correlated most consistently with SDN (r = 0.82, all; r = 0.77, agricultural; r = 0.92, distillers' RM). It is possible to predict protein degradability from the presence or absence of soluble proteins in various foods. While ADIN may be more suitable to predict DUN, globulins showed more potential to predict SDN in various foods. However, it may be necessary to be selective in choosing solvents for various foods. Further studies must validate this method by using a greater range of foods before suggesting its routine use for food evaluation. © 2001 Society of Chemical Industry