The Nutritional Value and Physiological Properties of Diets with Raw and Candida utilis-Fermented Lupin Seeds in Rats

Functional foods have recently gained much att ention because of their nutrient composition and benefi cial impact on consumer health (1). One of the treatments used to increase health-promoting properties of the products is the fermentation of raw materials (2–4). Fermented products, in addition to their nutrient content (proteins, fi bre, minerals and vitamins) usually include live, dried or lyophilized cells of lactic acid bacteria or yeast, which can work probiotically or prebiotically on digestive tract microfl ora (5–7). Fermentation is a natural and cheap method of changing the physical, chemical and functional parameters of food and feed (8,9). Supplementation of fermented components to feed has a benefi cial eff ect on growth performance and intestinal microfl ora regulation in pigs and rats (10–12). Yeast fermentation of starch components such as peas, beans or crops enriches products with high-value proteins of microbial origin, improves the digestibility of the protein and amino acid profi le and reduces the concentration of antinutritional factors (9,13), but the eff ect of lupin fermentation has not been determined so far.


Introduction
Functional foods have recently gained much att ention because of their nutrient composition and benefi cial impact on consumer health (1).One of the treatments used to increase health-promoting properties of the products is the fermentation of raw materials (2)(3)(4).Fermented products, in addition to their nutrient content (proteins, fi bre, minerals and vitamins) usually include live, dried or lyophilized cells of lactic acid bacteria or yeast, which can work probiotically or prebiotically on digestive tract microfl ora (5)(6)(7).Fermentation is a natural and cheap method of changing the physical, chemical and functional parameters of food and feed (8,9).Supplementation of fermented components to feed has a benefi cial eff ect on growth performance and intestinal microfl ora regulation in pigs and rats (10)(11)(12).Yeast fermentation of starch components such as peas, beans or crops enriches products with high-value proteins of microbial origin, improves the digestibility of the protein and amino acid profi le and reduces the concentration of antinutritional factors (9,13), but the eff ect of lupin fermentation has not been determined so far.
Lupins are a protein-rich plant source for animals and people.The presence of antinutritive factors in lupin The Nutritional Value and Physiological Properties of Diets with Raw and Candida utilis-Fermented Lupin Seeds in Rats seeds has a negative eff ect on the availability and utilization of nutrients by monogastric organisms (14).Results of another study (15) indicate that raffi nose family oligosaccharides (RFOs) more than alkaloids can reduce the utilization of nutrients of low-alkaloid lupin seeds.The bacterial conversion of carbohydrates, proteins and antinutritive compounds present in raw lupin seeds leads to the formation of a large number of substances that may have benefi cial or adverse eff ects on human and animal health.RFOs are not hydrolyzed by native digestive enzymes of non-ruminant animals, but only by bacterial α-galactosidase in the large intestine to gases, which can interfere with nutrient absorption (15) and cause fl atulence and diarrhoea (16,17).On the other hand, oligosaccharides and polysaccharides are recognized as potential substrates for fermentation by positive bacteria which can stimulate their proliferation and activity (4).It must also be mentioned here that in many countries, where lupin has already been accepted by consumers, there are reports about allergic reactions following consumption of lupin-containing products.Allergenic potential of lupin derives from its infi ltration into organism through the respiratory system and aft er eating lupin-fortifi ed products.
The results suggest that lupin allergens are heat stable but their status aft er fermentation was not recognized (18).Lactofermentation of lupin seeds improved the gut health of rats as compared to raw seeds (4).Fermentation of Lupinus angustifolius seeds using yeast increased the concentrations of proteins and essential amino acids, and significantly reduced the concentrations of oligosaccharides and phytates (19).Moreover, the products are characterized by acidic pH with prevalence of lactic and propionic acids, which is benefi cial because of their positive impact on the gastrointestinal microfl ora (a natural acidifi er).There are only a few studies concerning the nutritional value and physiological properties of fermented lupin seeds used in animal nutrition (4).Candida utilis was selected for this study on the basis of previously conducted experiments, which used diff erent yeast strains for lupin seed fermentation, where it was recognized as the most efficient strain (data not published).
This study presents the eff ects of raw and Candida utilis-fermented yellow and blue lupin seeds in comparison with soya bean meal on the growth and conditions of intestinal ecosystems of rats and metabolic activity of their intestinal microfl ora, determined on the basis of activities of selected enzymes and concentrations of volatile products in caecum digesta.

Microorganisms and fermentation media
For the experiments and preparation of fermented products, seeds of Lupinus luteus (cv.Lord) and Lupinus angustifolius (cv.Graf) were selected.Seeds were obtained from the Plant Breeding Station Przebedowo (Przebedowo, Poland).Soya bean meal was obtained from the market.
For fermentation, a Candida utilis strain was acquired from the Pure Culture Collection of Industrial Microorganisms LOCK 105 at the Institute of Fermentation Tech-nology and Microbiology, Technical University of Lodz (Poland).Initially, yeast grew at 30 °C for 48 h in sterilized tubes fi lled with an enriched YPD base (glucose 20 g/L, yeast extract 10 g/L, bacteriological peptone 20 g/L; all Oxoid, Hampshire, UK).Then, the content of fi ve tubes was transplanted to the sterilized bott le fi lled with enriched YPD base, and incubated at 30 °C for the next 48 h.A mass of 2 kg of lupin meal was mixed with 8 L of distilled water in autoclaved plastic fermentation buckets and mixed vigorously.Mixtures were sterilized at 121 °C for 30 min in a Getinge autoclave (Getinge, Rastatt , Germany).Aft er cooling, the resulting suspension was inoculated with yeast (10 % inoculum, by volume) and mixed.Fermentation was conducted for 72 h at 30 °C under anaerobic conditions.Next, the yeast enzymes were deactivated for 30 min at 80 °C, and the material was dried at 55 °C.Fermented yellow and blue lupin seed products were obtained.

Animals and diets
The experimental animals were used in compliance with European Guidelines for the Care and Use of Laboratory Animals next to the approval of the Ethical Committ ee for Animal Experimentation in Northeastern Poland.The experiment was performed on 40 male Wistar rats aged approx.4 weeks.The experimental diets were administered for four weeks to eight rats per each group housed individually in Plexiglas cages.The administered diets had a similar content of proteins, fat, minerals and vitamins.Control diets contained 20.5 % of soya bean meal, 70.3 % of wheat, 5 % of soya bean oil, 3 % of AIN-93 mineral premix and 1 % of AIN-93 vitamin premix (American Society for Nutrition, Bethesda, MD, USA) and 0.2 % of choline chloride, whereas in the experimental diets soya bean meal was replaced by raw or fermented blue and yellow lupin seeds (at 33.8, 27.0, 28.5 and 19.3 %, respectively) and wheat levels were 57.0, 63.8, 61.9 and 71.5 %, respectively.The protein content in all the diets was about 18.5 %.Experimental diets and tap water were administered ad libitum.The animals were kept under standard conditions at a temperature of 21-22 °C, relative air humidity of 50-70 %, with intensive room ventilation and a 12-hour light regimen.Individual body mass and food intake were recorded.

Sample collection
In the last week of the experiment, the rats were placed in the balance cages (Tecniplast Spa, Buguggiate, Italy) and faeces and urine were quantitatively collected to a plastic bag to calculate the protein digestibility coeffi cients and nitrogen balance of the diets.Aft er four weeks of the experiment, the rats were anaesthetized using sodium pentobarbitone (14 mg per kg of body mass).Blood samples were collected from the caudal vena cava.Serum samples were prepared by centrifugation at 1500×g for 15 min at 4 °C, and then stored at -40 °C for further analyses.Aft er laparotomy, selected parts of the digestive tract (small intestine, caecum, colon) were removed and weighed.Directly aft er euthanasia (approx.10 min), ileal, caecal and colonic pH values were measured, tissue samples were collected to determine dry matt er, ammonia and short--chain fatt y acid contents, and the remaining material was frozen at -70 °C for the determination of protein content and microbial enzyme activity.The ileal, caecal and colonic walls were fl ushed clean with ice-cold saline, blott ed on fi lter paper, and weighed to determine tissue mass.

Chemical and microbial analysis
For chemical analysis, all the samples were ground to pass through a 0.5-mm sieve.Dry matt er, crude protein, ether extract, crude fi bre, crude ash, acid detergent fi bre and neutral detergent fi bre were analyzed in duplicate in raw seeds, fermented products and digesta samples using methods 934.01, 976.05, 920.39, 978.10, 942.05 and 973.18 respectively, according to AOAC (20).Lupin alkaloids were extracted from fl our by trichloroacetic acid and methylene chloride (Sigma-Aldrich, Munich, Germany).The determination of alkaloids was done using gas chromatography method (Shimadzu GC17A, Kyoto, Japan) with a capillary column (Phenomenex, Torrance, CA, USA).Raffi nose family oligosaccharides (RFOs) were extracted and analyzed by high-resolution gas chromatography as described previously by Zalewski et al. (21).Phytate content was analyzed according to AOAC method 986.11 (20).The amino acid content was determined with an AAA-339 Mikrotechna amino acid analyzer (Prague, Czech Republic), using ninhydrin for post-column derivatization.Before analysis, the samples were hydrolyzed with 6 M HCl for 24 h at 110 °C according to AOAC method 994.12 (20).
Samples for bacteriological analysis were prepared by adding 27 mL of buff ered peptone water (Oxoid) to 3 g of samples and homogenizing for 30 s in a laboratory stomacher.Microbial counts were determined using a decimal dilution series of homogenized samples.The total bacteria count and lactic acid bacteria count were determined by plate standard methods using plate count agar and MRS broth (Oxoid), respectively, aft er 72-hour incubation at 30 °C.The Salmonella count was determined using pre-supplemented dichloran Rose Bengal chloramphenicol (DRBC) and Salmonella Chromogen agar (Oxoid), aft er 18 and 24 h of incubation, respectively, at 37 °C.The yeast count was determined using pre-supplemented DRBC (Oxoid) aft er incubation at 25 °C for 3-5 days.Coliform bacteria were determined using Violet Red Bile lactose agar (Oxoid) aft er 24 h of incubation at 30 °C.
The caecal pH was measured using a microelectrode and a pH/Ion meter (model 301, Hanna Instruments, Vila do Conde, Portugal).Ammonia was extracted and trapped in a boric acid solution, and it was analyzed by direct titration with sulphuric acid (POCh, Gliwice, Poland).Short --chain fatt y acids in fresh caecal contents were determined by gas chromatography (Shimadzu GC-14A, Kyoto, Japan) with a 2.5 mm×2.6 mm glass column containing 10 % SP-1200/1 % H 3 PO 4 on 80/100 Chromosorb WAW (Shimadzu Corp.); column temperature 110 °C, fl ame ionization detector (FID) temperature 180 °C and injector temperature 195 °C.Aliquots of caecal digesta were mixed with 0.2 mL of formic acid (POCh), diluted with deionized water and centrifuged at 10 000×g for 5 min.Samples of the supernatant were subjected to gas chromatography analysis.Caecal short-chain fatt y acid pools were calculated as the acid concentration per caecal digesta mass.
The activity of microbial enzymes (α-and β-glu cosidase, α-and β-galactosidase, β-glucu ro ni dase and xylo sidase) was measured on the basis of the rate of p-ni trophenol and o-nitrophenol release from nitrophenyl glucosides, and it was expressed in micromoles of the product formed per hour per gram of caecal digesta at 10 000×g for 10 min.Glucose, cholesterol, triacylglycerol and cholesterol concentrations in the serum were determined using Alpha Diagnostics (Warsaw, Poland) and Pointe Scientifi c (Warsaw, Poland) commercial kits.

Statistical analysis
Each group was compared with the control treatment using student's t-test and then the data (except the control group) were analyzed by two-way ANOVA with two main factors: lupin type (blue and yellow) and fermentation eff ect (untreated and fermented with Candida utilis plant material).If signifi cance was observed (p<0.05),Duncan's multiple range test was used to identify diff erences in the eff ect of individual diets.Calc ulations were made with STATISTICA v. 10.0 soft ware (StatSoft Corporation, Kraków, Poland).

Chemical composition of raw and fermented lupin meal
Seeds of yellow lupin contained more crude protein than raw blue lupin seeds (37.3 vs. 31.5%, Table 1).Yellow lupin seeds were characterized by a higher mass fraction of true proteins and amino acids in dry matt er than blue lupin seeds.The content of crude, acid and neutral detergent fi bres was similar in the seeds of both lupin species.The total alkaloid level was twice as high in yellow lupin seeds.Mass fractions of phytate phosphorus and RFOs were also higher in yellow lupin seeds.
Fermentation by yeast culture increased the crude protein content for about 13.5 and 26 % in comparison with raw seeds of blue and yellow lupin, respectively.On the other hand, the level of true protein, and especially the methionine mass fraction, was lower in the dry matt er of fermented products than in raw seeds.The content of lysine, cystine and threonine in dry matt er of fermented seeds was higher than in the raw material.Fermentation by Candida utilis partly reduced the crude fi bre, acid detergent fi bre and neutral detergent fi bre content in comparison with unprocessed seeds.The level of alkaloids increased almost three times in fermented products and reached about 0.042 % in fermented yellow lupin seeds.Phytate phosphorus mass fraction decreased the most (about 67 and 29 % in blue and yellow lupin, respectively) aft er fermentation.Oligosaccharides of raffi nose family disappeared aft er fermentation.

Microbial status of raw and fermented lupin meal
Fermentation eff ectively reduced the pH of lupin products from about 5.5 in raw seeds to about 4.0.The total count of bacteria, lactic acid bacteria and coliform bacteria counts were similar in both lupin species (Table 2).The yeast count was slightly higher in blue lupin seeds.
Salmonella was not detected in all the material.Fermented products were characterized by a higher content of total bacteria than unprocessed seeds.The fermentation process increased lactic acid bacteria (about 10 1 to 10 2 ) and yeast (10 2 to 10 3 ) number but reduced the number of coliform bacteria.

Intake and utilization of diets with diff erent protein sources
The use of comparable high-protein components in the diets of the rats, in supplementation to wheat meal, soya bean oil and mineral and vitamin mixtures, diversifi ed the diet intake and some parameters of nutrient utilization (Table 3).Higher feed intake was observed in the group fed fermented yellow lupin (p=0.039)than in the control diet (soya bean meal).In this group, also a higher apparent protein digestibility (p=0.001) and in the group fed fermented blue lupin a higher nitrogen balance (p=0.031)were found, which did not result in a higher body mass gain of rats.Lower body mass gain in the rats fed blue lupin, yellow lupin and fermented blue lupin seeds (p=0.001) and lower protein effi ciency ratio in those fed blue lupin seeds were recorded than in the control.The results of two-way ANOVA showed that the inclusion of the blue lupin meal in the diets, compared to yellow lupin, reduced diet intake (p=0.043),apparent protein digestibility coeffi cient (p=0.004),fi nal body mass (p=0.005),body mass gain (p=0.002) and protein effi ciency ratio (p=0.035).The use of fermented lupin seeds increased the apparent protein digestibility (p=0.030),body mass gain (p=0.028) and protein effi ciency ratio (p=0.005),compared to unproces sed seeds.
The applied dietary treatments did not aff ect the mass fraction and the pH of the small intestine, but changed some of the parameters of the caecum and colon of rats (Table 4).Compared to the group consuming soya bean meal, a higher mass fraction of tissue (p=0.001) and digesta (p=0.001), and pH (p=0.002) as well as a lower content of ammonia (p=0.001) were detected in the caecum of rats fed blue lupin and fermented blue lupin seeds.A higher pH of the caecum was also found aft er the application of the other diets (yellow lupin and fermented yellow lupin).Compared to the soya bean meal diet, the use of all of the experimental diets resulted in increased colon mass fraction (p=0.001), and the fermented blue lupin and fermented yellow lupin diets increased the pH of colon digesta (p=0.001), while the diet with the blue lupin increased the colon mass fraction (p=0.001).The results of two-way ANOVA showed that the addition of blue lupin seeds to the diet, compared to the yellow lupin, resulted in an increase in the caecal digesta (about 39 %) and the caecal tissue mass fraction by about 27 % (both p=0.001) and the reduction of the ammonia content of the caecal digesta (about 69 %; p=0.001).Seeds of blue lupin in the diet also increased the mass fraction of the colon digesta and the colon tissue (p=0.003 and p=0.013, respectively), and the pH of colon digesta, in comparison with yellow lupin in the diet.Yeast fermentation of lupin seed meal caused a decrease in the mass fraction of colon digesta and digesta pH (p=0.021 and p=0.001, respectively), and reduction in colon tissue mass fraction (p=0.001).

Microbial enzyme activity and short-chain fatt y acids in the caecal digesta
The applied dietary treatments did not aff ect the activity of microbial α-and β-glucosidase in the caecal digesta of rats (Table 5).The microbial activity of α-galactosidase was reduced in the groups fed fermented blue and yellow lupin (about 31 and 43 %, respectively) (p= 0.001), but β-galactosidase activity was lower (p=0.001) in the group fed fermented yellow lupin only, in comparison with the control group.The activity of β-glucuronidase was lower in the rats fed blue lupin (p=0.023)than in the control.Xylosidase activity was higher in the group consuming fermented yellow lupin in comparison with the control (p=0.001).
The results of two-way ANOVA showed that the content of blue lupin seeds in the diet resulted in lower xylosidase activity in caecal digesta (p=0.002)than the diet with yellow lupin.Fermentation of lupin seeds decreased the activity of α-and β-galactosidase (both p=0.001) about 64-68 %, but increased the activity of β-glucuronidase (p=0.017) and xylosidase (p=0.005) about 35 % in comparison with unprocessed lupin seeds.
The content of acetate in the caecum of animals from the groups consuming blue lupin and fermented blue lupin seeds was lower (p=0.001)than in the group consuming soya bean meal (Table 6).The content of propionate in the digesta of the rats fed yellow lupin, fermented blue lupin and fermented yellow lupin seeds was signifi cantly lower than in the control group (p=0.014).In the control group, the valeriate content in the caecum was significantly higher (p=0.002)than in other groups.The content of isobutyrate, butyrate and isovaleriate was similar in all the groups (p>0.05).The propionate pool was lower in the groups fed yellow lupin and fermented yellow lupin (p=0.002), and butyrate and the total short-chain fatt y acid pool in that fed fermented blue lupin (both p=0.001) than in the control group.The mass fraction of acetate was lower and of butyrate was higher in the groups consuming blue lupin and fermented blue lupin (both p=0.001), compared to the control.The propionate mass fraction in the total short-chain fatt y acid profi le was lower (p=0.016) in the groups fed blue or yellow lupin than in the control group.
The results of two-way ANOVA showed that the addi tion of blue lupin seeds to the diet resulted in a lowered acetate but higher butyrate fraction in digesta, compared to yellow lupin seeds (p=0.001 and p=0.028, respectively).Moreover, diets containing blue lupin seeds result ed in a higher propionate, butyrate and total short--chain fatt y acid pool (for all, p=0.001) than yellow lupin diets.Blue lupin seeds in diets also increased the propionate and butyrate mass fractions (p=0.016 and p=0.001, respectively) but reduced the acetate mass fraction (p= 0.001), compared to diets based on yellow lupin seeds.No eff ect of the fermentation process on the short-chain fatt y acid content, their pool or profi le was found (p>0.05).

Blood parameters
There were no diff erences in the concentration of serum glucose (p=0.130)among the groups (Table 7).The content of triacylglycerols in blood serum of rats in all the experimental groups was signifi cantly lower (p=0.001)than in the group consuming soya bean meal.The blood of rats fed the diet containing blue lupin was characterized by the lower level of the total cholesterol and high-density lipoprotein cholesterol (HDL) than the control group, but the contribution of HDL cholesterol to the total cholesterol in the blood of rats of all the groups was similar.Two-way ANOVA showed that blue lupin in the diets decreased the total cholesterol (p=0.031) and HDL cholesterol (p=0.017) in the blood plasma of rats, compared to diets based on yellow lupin seeds.Fermentation increased the total cholesterol (p=0.009) and HDL cholesterol (p= 0.025) in the blood plasma of rats, compared to raw seeds.

Discussion
In the present study, yellow lupin seeds are characterized by a higher content of crude and true proteins, as well as essential amino acid content than blue lupin seeds.Yellow lupin seeds also contained more antinutritional compounds.These results are generally consistent with those obtained in studies by Kim et al. (12), Jezierny et al. (14) and Chilomer et al. (22).Kim et al. (12) also found more antinutritional substances in the seeds of yellow than of blue lupin as opposed to Sobotka et al. (17).It is commonly known that the chemical composition of lupin seeds is strongly related to the variety and soil properties and, especially, environmental conditions during seed germination and maturation (23).
Fermentation increased the protein content and improved the amino acid profi le of seeds.The mass fraction of methionine was lower but of cystine higher in the fermented material.A similar increase in the protein content aft er fermentation was observed by other authors (9,13).During fermentation, yeast uses the available sources of nitrogen such as free amino acids, dipeptides, proteins, ammonium ions and urea for growth, which results in an intensive increase of biomass production (24).Khatt ab et al. (9) and Yabaya et al. (13) found an increase in the concentration of essential amino acids, including methionine, aft er fermentation of legume seeds, but in the present research the methionine content was reduced in both fermented products.This diff erence could be explained by the specifi city of plant material and yeast used in the experiments.
In the fermented seeds, reduced content of crude fibre, acid detergent fi bre and neutral detergent fi bre as well as oligosaccharides and phytates were found.A similar eff ect of fermentation on carbohydrate structure was found by Canibe et al. (25).Structural carbohydrates are quite resistant to direct yeast digestion, but cellulose, hemicellulose and pectin can be metabolized by some non-yeast enzymes (24).Raffi nose family oligosaccharides (RFOs) were completely removed during fermentation.These results are consistent with the results of Vidal--Valverde et al. (26), who found no RFOs in lentil seeds aft er natural fermentation.Khatt ab and Arntfi eld (27) found about 70 % reduction of RFOs in diff erent legume seeds aft er 24 h of fermentation by Saccharomyces cerevisiae.During fermentation, yeast produces diff erent types of hydrolases, which degrade the oligosaccharides into simple sugars (28,29).Candida utilis also reduced phytate phosphorus mass fraction by about 67 % in blue and by 29 % in yellow lupin.Yeasts have the ability to hydrolyze phytic acid into inositol and phosphoric acid (30).This phenomenon was confi rmed by other authors.Khatt ab and Arntfi eld (27) found a reduction of phytate aft er S. cerevisiae fermentation of diff erent legume seeds by about 30-35 %, and Egounlety and Aworth (28) a 30 to 60 % reduction of phytate aft er fermentation by Rhizopus oligosporus.Among the antinutritional substances, only the lupin alkaloid content increased in fermented products in comparison with raw seeds.This was probably the result of the increase of the mass fraction of alkaloids in dry matt er, reduced by a loss of degraded oligo-and polysaccharides.In previous studies, it was found that lupin alkaloids were resistant to Saccharomyces cerevisiae fermentation (19).
Fermentation by Candida utilis reduced the pH value, which was accompanied by an increase in the number of yeasts and lactobacilli.The number of Enterobactericeae decreased aft er fermentation.Similar results were found by Yabaya et al. (13), who observed a pH reduction from 5.6 to 5.1, and a signifi cant increase in the number of yeasts during the fermentation of soya cake by S. cerevisiae for 24 h.Mbata et al. (31) found that the addition of bambara groundnut to fermented maize fl our reduced the pH and increased the number of yeasts and Lactobacillus spp.Canibe et al. (25) showed a similar increase in the number of yeasts and lactic acid bacteria and signifi cant reduction in the number of Enterobacteriaceae in a study on fermenting grains.The pH reduction is due to the activity of microorganisms which degrade carbohydrates into organic acids (19,32,33).The acidic conditions particularly promote the development of lactic acid bacteria and yeasts which can tolerate low pH.Enterobacteriaceae, especially coliform bacteria, are acid intolerant, so their population is reduced (31).
The use of lupin components as a replacement for soya bean meal did not have a negative eff ect on the diet intake, digestibility and nitrogen retention, but it reduced the growth of rats.This is consistent with the results obtained by Sobotka et al. (17), who also found that the replacement of casein with blue and yellow lupin seeds in the diets did not reduce the feed intake but limited the growth rate of rats.It is probably the result of a bett er amino acid profi le and lower content of antinutritional substances in the soya bean meal diet than in lupin-containing diets.On the other hand, in this research, the dietary intake, apparent protein digestibility, body mass gain and protein effi ciency ratio of rats that were administered yellow lupin products in the diet was signifi cantly higher than of those administered blue lupin.Sobotka et al. (17) did not fi nd diff erences in the feed intake and body mass gain of rats that were administered diets containing yellow or blue lupin seeds.This implies that the diet intake and mass gain of animals depend on the content of some antinutritional substances in the diets, which was also confi rmed by a comparison of the eff ects of feed containing raw and fermented lupin seeds.Fermented products contained low phytate and oligosaccharide levels but had higher alkaloid content.The use of fermented products increased the apparent protein digestibility, body mass gain and protein effi ciency ratio in rats in comparison with raw seeds.These results are consistent with those obtained in studies by Bartkiene et al. (4) of yellow and white lupin seeds fermented with Pediococcus acidilactici.Previous studies on protein digestibility of legumes have shown interactions between antinutritional compounds (RFOs, phytates) and the protein and carbohydrate structure, forming complexes less susceptible to enzymes.Enzymes produced by yeast do not degrade complex carbohydrates; however, the acidic conditions (pH=4.0) and fermentation time (3 days) can accelerate the degradation of complex components by enzymes from the fermentation medium (7).The results suggest that lower digestibility and utilization of lupin diets is due to the presence of RFOs rather than alkaloids, which is consistent with the results obtained by Zduńczyk et al. (15), who show that oligosaccharides may reduce the intestinal absorption, thus decreasing protein digestibility and growth effi ciency more than alkaloids.According to Seve et al. (16), the removal of α-galactosides from soya fl our had a small but benefi cial infl uence on digestibility of nutrients and nitrogen utilization from piglet diets.
Experimental diets did not aff ect the small intestine parameters, but impacted the caecum and colon environment.In the digestive tract of animals fed blue lupin (raw or fermented) diets, a higher caecal tissue and digesta mass fractions and reduced ammonia mass fraction, and a higher colon tissue mass fraction were observed than in the control group.All diets containing lupin increased colon digesta mass fraction as compared to the soya bean meal diet.This is in accordance with the work of Bartkiene et al. (4), who used raw or fermented yellow and white lupin seeds in the diets of rats.Diets containing lupins signifi cantly increased the mass fractions of caecal tissue and colon digesta in comparison with soya bean meal.This was also confi rmed by Sobotka et al. (17) in rat diets which contained yellow and blue lupin cultivars in comparison with casein diet, and by Stanek and Bogusz (34) for diets with yellow lupin seeds.Zduńczyk et al. (15) found that the amount of toxic ammonia produced during the bacterial degradation of proteins tends to decrease in rats fed lupin seeds.
Higher caecal and colon tissue mass fractions, digesta mass fraction and also higher pH of colon digesta were found in animals off ered blue lupin than yellow lupin diets.This indicates that blue lupin seeds are more eff ectively fermented by ileal microfl ora.The same tendency in the caecum was found by Sobotka et al. (17).The increase in the caecal wall mass fraction could be due to the presence of short-chain fatt y acid, produced during the bacterial degradation of carbohydrates from lupin seeds, which is consistent with a higher total short-chain fatt y acid pool in digesta of rats fed blue lupin diet.The increase in the caecal digesta mass fraction could result from a greater bacterial count in the caecum (35).Sobotka et al. (17) found that diets based on blue lupin seeds increased the acidity of caecal digesta more than yellow lupin, but Bartkiene et al. (4) found no impact of lupin cultivar on the pH of the colon and caecum.
Fermentation had a signifi cant eff ect only on the colon parameters.The use of fermented seeds in the diets reduced the colon tissue and digesta mass fractions in comparison with raw seeds.It indicated poorer microbial fermentation in this part of the ileum, which is probably the result of the lack of RFOs in the fermented products.Fermentation of lupin seeds did not lower the pH of digesta from the small intestine, caecum or colon.Bartkiene et al. (4) found a similar tendency in the pH value, digesta mass and ammonia production when the seeds of yellow and white lupin were fermented by Pediococcus acidilactici.Higher acidity levels could promote the development of benefi cial microfl ora (Bifi dobacterium or Lactobacillus) and prevent the growth of harmful bacteria (31).
In a digestive tract some microorganisms, especially the Enterobacteriaceae family, can produce hydrolytic enzymes which transform precarcinogens into active carcinogens.The activity of these enzymes (α-and β-ga lac tosidase, α-and β-glucosidase, β-glucuronidase, xylosidase) is oft en used as an indicator of the eff ect of additives on the intestinal microfl ora.In comparison with soya bean meal the activity of α-galactosidase was reduced by fermented blue and yellow lupin diet, of β-galactosidase by fermented blue lupin and of β-glucuronidase by the blue lupin diet.Only xylosidase activity increased when using the fermented yellow lupin diet in comparison with soya bean meal.Moreover, the inclusion of blue lupin seeds in the diets of rats reduced the activity of xylosidase in comparison with yellow lupin seeds.Fermented products in the diets reduced signifi cantly α-and β-galactosidase, but increased β-glucuronidase and xylosidase activities in comparison with raw seeds.This is consistent with research by Bartkiene et al. (4), who found the same tendency comparing diets containing raw blue lupin seeds and soya bean meal.On the contrary, Sobotka et al. (17) found that the use of diets with blue and yellow lupin seeds signifi cantly increased α-and β-galactosidase and β-glucosidase activity in digesta of rats.Similarly, no impact of the lupin seed variety in the diet on α-glucosidase and β-glucuronidase was observed.Comparing raw and lactofermented yellow lupin seeds, Bartkiene et al. (4) found a tendency which is similar to the one observed for α-, β-glucosidase and α-galactosidase.The activity of β-galac tosidase and β-glucuronidase was higher in fermented seeds than in the raw material.The α-and β-glucosidase are involved in the degradation of structural carbohydrates (cellulose, hemicellulose) in the caecum and what is important is that they break down oligosaccharides to simple sugars by cleaving the glycosidic bonds (36).This can explain the signifi cantly lower activity of these en-zymes in diets with fermented lupin seeds, which are characterized by a lower RFO content.On the other hand, an increase in the enzymatic activity in the caecal digesta can also indicate a rapid growth of benefi cial bacteria in the caecum.
The rate of carbohydrate hydrolysis in the caecum is the result of short-chain fatt y acid production.The total amount of short-chain fatt y acids in the caecal digesta of control animals reached 220 μmol/g, and it was similar to that noted in rats fed raw and fermented blue and yellow lupin seeds (193-208 μmol/g).In the caecal digesta of animals from all the groups off ered lupin seeds (raw or fermented), reduced content of acetate, propionate and valeriate were found.The level of total short-chain fatt y acids in the present work was higher than in the similar research provided by Bartkiene et (35) also found no diff erences in total short-chain fatt y acid production in digesta between rats fed soya bean meal diet and raw or lactofermented seeds of yellow or white lupin, and reduction of acetic, propionic and valeric acid content.Sobotka et al. (17) showed that short-chain fatt y acid production in caecal digesta of rats was similar when consuming diet containing blue lupin seeds but higher when consuming diet with yellow lupin seeds than in casein--based diet.
The lupin species did not aff ect the total short-chain fatt y acid content, but in digesta of rats off ered yellow lupin seeds, a signifi cantly higher acetate and lower butyrate content was found.Sobotka et al. (17) showed that the production of short-chain fatt y acids in rats off ered a diet containing yellow lupin seeds was signifi cantly higher than in the rats fed blue lupin, because of the diff erences in α-galactoside content between the two cultivars.According to Bartkiene et al. (4), there were no diff erences in the short-chain fatt y acid content of caecal digesta of rats off ered raw yellow or white lupin seeds in the diets.There were no diff erences in the short-chain fatt y acid composition between groups as a consequence of fermentation, which was generally consistent with the results of Bartkiene et al. (4).Klewicka et al. (2) found that the concentration of propionic, butyric and valeric acids in the caecal digesta of rats was higher (than in the casein--based diet) when they were fed a diet containing fermented beetroot juice.It is worth noting that actual products were off ered in a dry form, as opposed to Klewicka et al.
(2) and Bartkiene et al. (4).This means that the activity of yeasts and bacteria present in the fermented mass could be partly inhibited (enzyme deactivation) and biomass could be used as a protein source in the initial segments of the digestive tract, which eliminated their expected probiotic eff ect in the distal part of the digestive tract of rats.Champ et al. (37) proved that short-chain fatt y acid production originated in the ingestion of lupin meal (in comparison with isolated fi bres) and that 50 % of short--chain fatt y acids appears to be derived from non-starch polysaccharides, and 50 % from the α-galactosides, which are not present in the fermented seeds.Some researchers have suggested that the calculated short-chain fatt y acid pool produced in the caecum provides more precise information on the eff ect of the applied preparations on the intensity of the fermentation in the gastrointestinal tract.The propionate pool was lower in diets with raw and fermented yellow lupin seeds but the butyrate and total short-chain fatt y acid pools were higher in the diet containing fermented blue lupin than in the soya bean meal diet.In blue lupin diets signifi cantly higher pools of propionate, butyrate and the total short--chain fatt y acids were found than in yellow lupin seeds.
Generally, in digesta of rats, lower amounts of acetate and propionate but a higher content of butyrate were found than in rats fed soya bean meal.In the profi le of short-chain fatt y acids in digesta of animals off ered blue lupin a lower level of acetate but higher levels of propionate and butyrate were detected than in diets containing yellow lupin.The profi le of short-chain fatt y acids in actual results was diff erent from those presented by Bartkiene et al. (4), because of a lower propionate and a higher butyrate amount, probably because of diff erent diet formulations.
Acetic acid is the major by-product of Bifi dobacterium fermentation, which modifi es the caecal ecosystem (36).Sobotka et al. (17) found that the propionic acid content in caecal digesta was dependent on oligosaccharide concentrations in diets, which is in opposition to our observations.Propionic acid is almost entirely metabolized in the liver, and it is capable of inhibiting hepatic cholesterol synthesis from acetic acid (38).The level of butyric acid increased aft er the administration of diets containing blue lupin seeds.Butyric acid is an important source of energy for epithelial cells and it regulates cell growth and diff erentiation.In our study, similarly to Sobotka et al. (17), an increase in butyric acid content was accompanied by an increase in the cecal wall mass fraction in rats off ered blue lupin diets.
The replacement of soya bean meal with lupin seeds in the diets of rats decreased triacylglycerol concentrations in blood plasma.In animals fed a diet containing raw blue lupin seeds, a signifi cant reduction in the concentration of total cholesterol and HDL cholesterol was found in comparison with the soya bean meal and yellow lupin diet.Fermentation signifi cantly increased the total cholesterol and HDL cholesterol in blood plasma of rats.The results obtained are consistent with the results of other studies (39,40).Spielmann et al. (41) found that a protein isolated from L. albus is strongly hypotriglyceridemic in rats, which is in part due to downregulation of SREBP--1c in the liver, which, in turn, leads to a reduction in hepatic fatt y acid synthesis.Osman et al. (42) suggest that the hypocholesterolemic eff ect of lupin seed supplements might be due to their abilities to lower the plasma cholesterol level as well as to slow down the lipid peroxidation process and to enhance the antioxidant enzyme activity.

Conclusions
The presented research showed many diff erences among physiological responses of gastrointestinal tract to a diet supplemented with raw or fermented blue or yellow lupin seeds.The inclusion of yellow lupin seeds in the diets of rats had a positive eff ect on the feed intake and feed utilization in comparison with blue lupin, but blue lupin in the diet had a more benefi cial eff ect on gas-trointestinal fermentation processes.Lupin seeds treated with Candida utilis had a bett er chemical composition than unprocessed seeds, because of a higher protein and amino acid content, low pH and reduced antinutritional substance content.Fermentation had a positive infl uence on protein digestibility, body mass gain and protein effi ciency ratio of rats, but despite the acidic pH, it did not have a benefi cial eff ect on the gastrointestinal parameters of rats.On the other hand, it had a positive infl uence on some bacterial enzyme activity and it increased HDL cholesterol concentrations in the blood serum.Taking this into account, lupin seeds fermented with Candida utilis could be recommended as an alternative, protein-rich component of diets for animals and people.However, it should be remembered that lupin is a potential allergen and its status aft er fermentation has not yet been investigated.

Table 1 .
Chemical composition of raw and fermented lupin seeds

Table 2 .
Microbial status of raw and fermented lupin seeds

Table 3 .
Diet intake, apparent protein digestibility coeffi cient, nitrogen retention, initial and fi nal body mass, body mass gain and protein effi ciency ratio in rats fed diet with diff erent sources SEM=standard error of the mean; data with diff erent lett ers in superscripts in the same column are signifi cantly diff erent with each other at p<0.05; *data signifi cantly diff erent from the control group at p<0.05 (t-test)

Table 4 .
Parameters of gastrointestinal function of rats fed diet with diff erent sources of proteinSEM=standard error of the mean, bm=body mass; data with diff erent lett ers in superscripts in the same column are signifi cantly diff erent with each other at p<0.05; *data signifi cantly diff erent from the control group at p<0.05 (t-test)

Table 5 .
Microbial enzyme activity in the caecal digesta of rats SEM=standard error of the mean; data with diff erent lett ers in superscripts in the same column are signifi cantly diff erent with each other at p<0.05; *data signifi cantly diff erent from the control group at p<0.05 (t-test)

Table 6 .
Content of short-chain fatt y acids (SCFA), their pool and profi le in the caecal digesta of rats SEM=standard error of the mean; data with diff erent lett ers in superscripts in the same column are signifi cantly diff erent with each other at p<0.05; *data signifi cantly diff erent from the control group at p<0.05 (t-test)

Table 7 .
Biochemical indices of blood plasma of ratsSEM=standard error of the mean; data with diff erent lett ers in superscripts in the same column are signifi cantly diff erent with each other at p<0.05; *data signifi cantly diff erent from the control group at p<0.05 (t-test); * * high density lipoprotein (HDL) fraction in total cholesterol al. (4), Juśkiewicz et al. (35) and Sobotka et al. (17).Juśkiewicz et al.