Surplus pig meat production and growing public concern on animal welfare and soil pollution problems have resulted in the encouragement of the production of native pig breeds such as the Iberian pig which are well adapted to local feed resources and provide high quality products [1] [2] . Iberian pig breeding is peculiar in that in order to achieve higher organoleptic quality, and its final fattening stage (from 100 to about 160 kg body weight) is carried out in free range conditions. During that period, pigs are fed basically on acorns and grass, which are the predominant feedstuffs available in the Mediterranean woodland prairies. In these conditions, the use of local feed resources is of paramount importance for the sustainability of the system. However, this regime has been shown to be lower in protein, and particularly in lysine, than that required by the pig [2] . As a consequence, other protein supplements such as legume seed meals, which are grown in the same areas where this breed is produced, might be used to equilibrate its diet and achieve better productive results, even though this practice might alter the final product quality [3] if used in the final fattening period in extensive conditions. Locally grown legume seed meals have traditionally been, and at present in some cases are, incorporated into formulas for Iberian pigs up to 100 kg body weight. In addition, cereals and legumes are ingredients among those legally authorized in Spain in the fattening period of Iberian pig production stages [4] . Therefore, it is important to determine not only the digestibility of nutrients in these feedstuffs, but also the potential effects on the intestinal physiology to establish their practical usefulness in Iberian pig dietary formulas. The amount of information available on the digestive physiology of the Iberian pig is scarce, particularly if compared to lean pig breeds.

Problems associated with intestinal disorders in different production stages are among those of greatest economic impact at present for the pig production industry, and it holds true not only for lean breeds but also for Iberian pig production systems, where costs derived from gastrointestinal disorders sometimes exceed market prices (Sánchez Romero Carvajal Jabugo S.A., personal communication). As a consequence, there is at present the need to look for alternatives that could enhance the natural defense mechanisms of animals and reduce the massive use of antibiotics. In pigs, the distal small intestine is known to harbor more than 10⁸ bacteria/g or ml of gut contents [5] . As a consequence, one way to improve productive parameters and health in the pig industry is to use specific feed additives and/or dietary raw materials able to modulate the gut microbiota which plays a critical role in maintaining host health [6] . One of the main putative effects of certain additives or feedstuffs might be the improved resistance to potentially pathogenic bacteria colonization and enhanced host mucosa immunity, thus resulting in a reduced pathogen load, an improved health status of the animals and a reduced risk of food-borne pathogens in foods [7] .

Accordingly, the present work was designed to study the effects of legume feeding on ileal microbiota composition of Iberian pigs. Pigs in the final fattening period (aprox. 100 kg body weight) were used because it is the period when the animals are fed with the ingredients which naturally grow in the Mediterranean prairies, and could more easily benefit from vegetable local ingredients. Lupins (Lupinus angustifolius) and chickpeas (Cicer arietinum) were chosen in this work due to their low content in antinutritional factors, and to differences in the composition of their constituent carbohydrates [8] [9] .

As shown in Table 2, higher (P < 0.01) lactobacilli log₁₀ number of copies was determined in the ileal contents of pigs fed on lupin or chickpea-based diets with respect to those fed on the soybean-based diet. Bacteroides log₁₀ number of copies was lower (P < 0.01) in ileal contents of chickpea-fed pigs. Enterobacteria and Escherichia/ Shigella log₁₀ number of copies was lower (P < 0.01) than that of soybean in pigs fed on diets based on lupin or chickpea. The log₁₀ number of copies of the Clostridium coccoides/Eubacterium rectale group was lower than that of soybean in pigs fed on the chickpea diet. Bifidobacteria and Clostridium leptum number of copies was not affected by the legume seed meal incorporated to the diet. The number of copies of the different bacterial groups analysed in the animals fed on the casein-based diet was lower (P < 0.01) than that of soybean for lactobacilli and bacteroides, while they were higher than that of soybean for bifidobacteria, enterobacteria and the Escherichia/Shigella group.

While information on the use of lupin seed meal for pigs is more abundant [14] , works with chickpea meal are scarce in the literature, and even less has been reported in Iberian pig feeding. In a previous work by our group, it was concluded that true ileal nitrogen and amino acid digestibilities of lupin and chickpea meals are comparable to those of defatted soybean in Iberian pigs [7] , and higher total amounts of lupin NSP and/or lupin and chickpea oligosaccharides were digested up to the terminal ileum compared to defatted soybean [9] . Also, previous works in rats and pigs [15] -[17] indicate that the main reason for the higher faecal N excretion found in legume fed animals is that intestinal, and particularly lower gut microflora, increases its growth at the expense of undigested carbohydrates coming from the small intestine and urea supplied mainly through the large intestinal wall. It is known that the digestibility and fermentation of legume carbohydrates (mainly starch, NSP and oligosaccharides) in the gut have both healthy and productive implications [18] .

In normal conditions, the main role of the intestinal microbiota is to provide a barrier against the invasion of potentially harmful bacteria [18] . Accordingly, a lot of research has been devoted in recent decades to the study of the potential therapeutic and productive uses of dietary additives/supplements able to limit the growth of potentially harmful bacteria such as enterobacteria, clostridia and E. coli [6] . Lactobacilli and bifidobacteria are among the most studied species out of those with alleged probiotic effects. The mechanism by which those species may play a beneficial role is related to the production of specific antibiotics, but particularly acetic and lactic acids, which lower the pH of the intestinal contents make it less favorable for the growth of coliforms [6] [18] . Also, the C. coccoides/E. rectale cluster includes bacteria producing butyric acid, a metabolite seen to be beneficial for gut functionality in pigs through regulation of epithelial cell growth, induction, differentiation and apoptosis in the small intestine, increase of intestinal cell proliferation and improvement of digestive and absorptive capacities of the small intestine [18] .

In the present work, higher log₁₀ counts of lactobacilli, and lower enterobacteria and Escherichia/Shigella counts, were determined with respect to soybean and casein-fed pigs in the intestinal contents of pigs fed on either lupin or chickpea diets (Table 2). Previous work in rats [19] and broiler chickens [20] fed on purified lupin fractions indicated that the NSP fraction was mainly responsible for similar effects on intestinal microbiota composition. As already mentioned, lupin and chickpea meals were chosen for the experiments described here, among other reasons, because of the different chemical composition of their constituent carbohydrates. Thus,

^a,b,cMeans (n = 5) in the same row with different superscript letters differ (P < 0.01).

while the main carbohydrate fraction in chickpea meal is starch, lupin meal contains only very low amounts of it (405 and 16 g·kg⁻¹, respectively), and the reverse situation occurs with respect to NSP contents (86 and 319 g·kg⁻¹, respectively). These differences in NSP contents are mainly due to high concentrations of galactose, glucose and uronic acids in lupin meal compared with other legumes. Soybean concentrate NSP values (168 g·kg⁻¹) were intermediate between lupin and chickpea meals [9] . Amounts of lupin NSP digested (94.1 g·kg⁻¹ feed) within the small intestine were higher than those for soybean and chickpea, which were not different from each other (24.3 and 27.1 g·kg⁻¹ feed, respectively). However, not only NSP sugars, but also oligosaccharides have been associated with prebiotic effects, as legumes are known to contain substantial amounts of these compounds. Higher proportions of total oligosaccharides in lupin and chickpea diets were also digested compared with the soybean diet within the small intestine (10.4, 21.7 and 23.7 g·kg⁻¹ diet ingested for soybean, lupin and chickpea diets, respectively) [9] . Therefore, since only whole seed meals were fed in the present investigation, and if not α-galactosides or NSP in separated diets, it is not possible from the results reported here to assign the effects on microbial growth to either NSP and/or α-galactosides in lupin or chickpea meals. The increase in bifidobacteria log₁₀ counts in pigs fed on the casein-based diet might be related to the bifidogenic effect of casein fractions [21] , while the increase in enterobacteria and the Escherichia/Shigella group is probably related to the low amount of dietary fibre, as higher enterobacteria and coliforms counts have been previously found in pigs fed on low fibre diets [22] .

It is concluded that, in addition to economic and legal considerations, lupin and/or chickpea meal feeding might induce a more protective intestinal microbiota composition of Iberian pigs in their final productive stages.

The authors are indebted to the Spanish Junta de Andalucía (project AGR395) for financial support. This research has been also partially supported by the FEDER and FSE funds from the European Union. MJP is a recipient of a JAE-predoc fellowship from CSIC.