Obesity is the nutritional disorder which has shown the greatest increase in prevalence, even in those countries in which deficiency diseases represent a severe public health problem. The goal of the present study was to analyse the anthropometric profile of a community living in the outskirts of Maceió, capital of Alagoas (northeastern Brazil), and to investigate the hypothesis of a coexistence of undernutrition and obesity in a very low-income population. The survey was conducted on 315 families (1247 individuals). Among the children (aged ≤10 years), the prevalence of wasting, stunting and wasting plus stunting was 3·8, 8·3 and 8·7 % respectively. Wasting (10·2 %) was the most prevalent form of undernutrition among adolescents; nonetheless, a higher frequency of stunting (11 %) and overweight–obesity (5·5 %) was seen specifically in girls, in agreement with trends found in other studies. Adults exhibited a high prevalence of overweight–obesity (25 %), but stunting was also present (22 %). Of the stunted individuals, 30 % were overweight–obese and 16·3 % were underweight. There were eighty-six families with at least one parent who was underweight (27 %) and 104 families with at least one parent who was overweight (33 %). Underweight and overweight–obesity were both present in ninety-six households (30 %). These results may indicate that better living conditions in urban areas in a population ‘adapted’ to chronic famine might increase the susceptibility to obesity. Considering the harm caused by the cumulative effect of these two conditions (undernutrition in childhood and obesity in adult life) there is a clear need for new studies to uncover the determinant factors so that preventive measures can be implemented.

Previous studies have indicated that the lactic acid bacterium Lactobacillus rhamnosus HN001 can enhance immune function in mice, following oral delivery. However, the influence of bacterial cell viability on immunoenhancement, and the optimum dose of HN001 required for this effect, have not been determined. In the present study, both live and heat-killed preparations of L. rhamnosus HN001 were shown to enhance the phagocytic activity of blood and peritoneal leucocytes in mice, at a dose of 109 micro-organisms daily. In contrast, only live HN001 enhanced gut mucosal antibody responses to cholera toxin vaccine. Feeding mice with 107 viable HN001/d for 14 d was shown to enhance the phagocytic capacity of blood leucocytes, with incremental enhancement observed at 109 and 1011 daily doses. In contrast, a minimum dose of 109 viable HN001/d was required to enhance the phagocytic activity of peritoneal leucocytes, and no further increment was observed with 1011 daily. This study demonstrates that L. rhamnosus HN001 exhibits dose-dependent effects on the phagocytic defence system of mice, and suggests that while the innate cellular immune system is responsive to killed forms of food-borne bacteria, specific gut mucosal immunity may only be stimulated by live forms.

Dietary incorporation of fermentable, indigestible fructans may be of benefit to gastrointestinal health by providing short-chain fatty acids, stimulating the proliferation of bifidobacteria or lactobacilli and suppressing potential pathogenic organisms in the gut. We tested the hypothesis that the effects of fructans on caecal, colonic and faecal short-chain fatty acid concentration and microflora composition depend on their chain length. Germ-free rats associated with a human faecal flora were randomly assigned to one of four treatments as follows: (1) commercial standard diet as a control (Con); (2) Con+50 g short-chain oligofructose/kg (OF); (3) C+50 g long-chain inulin/kg (lcIN); or (4) Con+50 g OF–lcIN/kg (Mix OF–lcIN). Changes in bacterial population groups in response to feeding these diets were investigated with 16S rRNA-targeted probes applied in in situ hybridization. Mix OF–lcIN- and lcIN-containing diets resulted in larger numbers of caecal, colonic and faecal bacteria of the Clostridium coccoides–Eubacterium rectale cluster than Con (10·6 and 10·3 v. 9·5 log10/g wet wt), whereas OF alone did not affect this bacterial group in caecum, colon or faeces. A bifidogenic effect was only observed in the colon and faeces of OF-treated rats. More lactobacilli were found in caecal and colonic contents of Mix OF–lcIN-fed rats and in faeces of OF-fed rats compared with Con. Mix OF–lcIN and OF led to significantly smaller numbers of caecal, colonic and faecal bacteria belonging to the Clostridium histolyticum and C. lituseburense groups than Con (6·8 and 6·9 v. 7·9 log10/g wet wt). Counts of total bacteria, Bacteroides–Prevotella and Enterobacteriaceae did not differ between the groups. OF and/or lcIN-containing diets significantly increased the caecal and colonic concentration of butyrate and its relative molar proportion. Only lcIN-containing diets resulted in a higher faecal concentration of butyrate than Con. Higher molar proportions of faecal butyrate were observed with all diets that had been supplemented with OF and/or lcIN. Stimulation of butyrate production could be of interest for the prevention of ulcerative colitis and colon cancer.

The aim of the present study was to explore whether mice fed a diet low in Zn (2·0 mg Zn/kg diet) for a relatively short period of time were more prone to severe Streptococcus pneumoniae infection than mice fed a normal diet (25 mg elemental Zn/kg). The Zn-deficient mice were compared with mice in two Zn-adequate control groups; one pair-fed and another with free access to the diet. After 2 weeks feeding, the mice were infected intranasally under anaesthesia with a suspension containing about 107 pneumococci. Clinical status was observed every day and blood samples were examined for S. pneumoniae every second day for a week. All infected mice examined carried the infecting strain intranasally. The survival time and time before positive blood culture were significantly shorter in the Zn-depleted group than in the pair-fed Zn-adequate group (hazard ratios 15·6 and 3·2, P<0·0001 and P=0·045 respectively). At the end of the observation period, ten of the twelve mice in the Zn-deficient group were dead while one of twelve and two of twelve were dead in the two Zn-adequate control groups. This study shows that even acutely-induced Zn deficiency dramatically increases the risk of serious pneumococcal infection in mice.