A survey of nutritional—immunological interactions*

Memoranda A survey of nutritional-immunological interactions * There is some evidence to show that the immune response is suppressed in malnutrition but the mechanism is not entirely clear. A more complete understanding of nutritional- immunological relationships is important, especially for child health in developing countries. This memorandum discusses a general approach to the problem and proposes specific methods for investigating the effects of malnutrition on the immune response. Several field studies incorporating these proposals are now in progress. Throughout a large part of the world the two most important factors determining the health of young children are infection and malnutrition. In developing countries these factors often retard growth and pro- duce clinical features of protein-calorie malnutrition (PCM). The interaction of infection and malnutrition has been reviewed by Scrimshaw et al. (1968). This memorandum describes a general approach to the problem of the effects of malnutrition on the immune response. Epidemiological observations made in different countries have supported a view that infection fre- quently leads to malnutrition in young children. The mechanism is not completely understood, but three factors appear to be of importance: (1) many infec- tions, even asymptomatic ones, induce a negative nitrogen balance; (2) anorexia is a frequent concom- itant of infection; and (3) local customs in dealing with infections often lead to serious dietary restric- tions. The mechanisms by which PCM could predispose to infection are less well understood. One possibility is that the immune response to antigenic stimulation may be depressed or aberrant in children with PCM. Clinical investigations of the relationship ofimmune responses to PCM are difficult. Precise measure- ments of nutritional status are not always possible, and nutritional and immunological states may change significantly within short periods; the interpretation of results of research made over long periods may thus prove to be difficult. In areas where this problem will be studied, communicable diseases tend to be endemic and abnormal immune responses to defined * This memorandum was drafted by the signatories listed on page 544. Requests for reprints should be addressed to the Chief Medical Officer, Immunology, World Health Organization, 1211 Geneva, Switzerland. The signatories also drafted a working protocol for field studies of this problem, and copies may be obtained from the same address. antigens could result either from malnutrition or from concurrent infectious diseases. In spite of these difficulties, it was felt worthwhile for WHO to encourage efforts to decide if the immune responses of children are depressed or altered by malnutrition under circumstances that are relevant to health care programmes. The justification for making coordinated field studies arises from the following considerations. (1) Vaccines used in developing countries have usu- ally been tested in healthy individuals. It is necessary to determine if such vaccines are equally effective in conferring protective immunity in poorly nourished children in developing countries. If malnutrition is found to interfere with the development of protective immunity, it may be necessary to provide nutritional supplements in conjunction with vaccination pro- grammes or to plan vaccination campaigns in the light of local knowledge regarding the ages and seasons when the nutritional state of children is at its best. (2) A better understanding of nutritional inter- actions with immune responses may provide a more satisfactory basis for the clinical care of patients. (3) The possibility that malnutrition may have a lasting effect on immune responses has implications for both health services and clinical medicine. IMMUNOLOGICAL CONSIDERATIONS Arising in early life from a precursor cell found in the bone marrow, two separate populations of lymphoid cells differentiate and populate the lym- phoid organs. One population differentiates in the thymus (T-cells) and constitutes a substantial propor- tion, if not the major part, of the circulating pool of small lymphocytes in blood and lymph and in certain areas of lymph nodes and the spleen. Following contact with an antigen, these cells may proliferate, 2829 -537- 8 MEMORANDA but in any case they are responsible for the initiation of a variety of both humoral and cellular immune responses. The activation of phagocytic cells of the reticuloendothelial system and the initiation of in- flammation through chemical mediators released by antigen-stimulated T-cells also effect specific cellular responses. In children congenitally lacking this sys- tem of T-cells, the host-defence mechanism lacks the capacity to resist infection by many viruses, intra- cellular bacterial pathogens (Good et al., 1970), and fungi (group II of Table 1). These children also fail to reject foreign tissues. The second population of lymphoid cells (B-cells), differentiated in birds in the bursa Fabricii (Cooper et al., 1966a) and possibly in the gut-associated lymphoid tissue in mammals (Cooper et al., 1966b; Fichtelius, 1967; Faulk et al., 1970), represents a population of lymphatic cells responsible for the synthesis and secretion of antibody. These cells undergo proliferation and differentiation following contact with antigen, and produce antibodies that may include some or all of the 5 major immunoglo- bulin classes (IgG, IgA, IgM, IgD, IgE). A special form of IgA that appears in external secretions Table 1. Infections in immunodeficiency diseases Groupa Pathogen Primary bodily defencemechanism I Pneumococcus Haemophilus influenzae Streptococcus Meningococcus sp. humoral immune responses Pseudomonas aeruginosa hepatitis virus Pneumocystis carinil 11 rubella virus varicella virus vaccinia virus cytomegalic inclusion body cellular immune responses virus Mycobacterium tuberculosis Candida alb/cans Histoplasma capsulatum a Two groups of microorganism arranged according to the primary mechanism of host defence against each pathogen. These lists have been assembled by Good et al. (1970) mainly from clinical experience with children suffering from immunodeficiency diseases. (secretory IgA) may contribute to resistance to infec- tions along mucosal surfaces. The combination of antigen with some antibodies (IgG or IgM) activates the complement system to produce cell lysis, but often there is a subsequent attraction of phagocytes and the initiation of inflammation. The role of the complement proactivator system (Gotze & Muller- Eberhard, 1971) in this connexion is not yet clear. Children congenitally lacking the B-cell system show a major defect in the defence mechanism against the encapsulated pyogenic bacteria (group I of Table 1). The division of the immune system into T-cell and B-cell components is useful in the analysis of abnor- mal function, but interactions between both cell populations occur in many situations in both the induction and effector stages of immune responses. NUTRIMONAL EFFECTS ON IMMUNE RESPONSE Effects on cell replication In some specific nutritional deficiencies in man, e.g., iron, folic acid, or pyridoxine deficiencies, and in protein-deficient states in animals, the replication of cells of the erythroid series is disturbed, resulting in the appearance of anaemia (Seminars in Hemato- logy, 1970). It is possible that the kinetics of white cell replication are similarly disturbed. In kwashior- kor, failure of the leucocyte response has been re- ported (Trowell et al., 1954; Behar et al., 1956). However, slight increases of plasma cells in the bone marrow of malnourished children have also been noted (Kumate, 1967). Further investigations in man are needed to determine the extent to which malnutrition affects the number of circulating lym- phocytes and the size of the noncirculating reserves. Mice subjected to sustained post-weaning under- nutrition (marasmus) have a severe lymphopenia (Woodruff, 1970, 1971). Mice kept on a low-protein diet also tend to make fewer antibody-forming (i.e., plaque-forming) cells (Cooper et al., 1970). The molecular basis of these changes is not known, but pyridoxine-deprived rat liver and spleen cells synthe- size less messenger RNA and have a decreased num- ber of polysomes per unit weight of tissue (Montjar et al., 1965). Amino-acid-deprived Ehrlich ascites tumour cells have also been shown to shift rapidly single ribosomes into polyribosomes and to increase the protein synthesizing activity of the polyribosomes following addition of amino acids and glucose to the growth medium (Venrooij et al., 1970). These find- ings suggest that nutritional factors can modulate the rate of protein synthesis through mechanisms, in 538 SURVEY OF NUTRITIONAL-IMMUNOLOGICAL INTERACTIONS addition to substrate limitation, that operate at the transitional level. However, there is at present no evidence for or against similar mechanisms operating on immune mechanisms. Effects upon the thymus and lymph nodes It has been shown that the thymus in malnourished children is much smaller than it is in well-nourished children of equivalent age (Vint, 1930; Watts, 1969; Smythe et al., 1971; Marigo, unpublished data). Recent studies in South America have shown that marked changes in the morphology of the thymus are particularly common in severely malnourished children (Marigo, unpublished data). The changes include partial or complete depletion of cortical lymphoid cells, reduction in cortical thickness, and increased interstitial fibrosis with relatively few changes in the medulla. In mice, cortisone injections lead to very similar morphological changes in the thymus. Similar results have been reported from South Africa (Smythe et al., 1971). The South Amer- ican children also showed a diminished number of lymphocytes in the paracortical areas oflymph nodes, areas which are regarded as being thymus-dependent. In some cases, hypertrophy and hyperplasia ofbowel lymphoid tissue (Peyer's patches) was observed (Ma- rigo, unpublished data) while in others hypoplastic tonsils have been reported (Smythe et al., op. cit.). While a full evaluation of these findings is still awaited, they indicate that severe malnutrition in man may selectively affect certain populations of lymphoid cells. Further studies in this area are needed. Effects on cell-mediated immunity Knowledge of the effects of malnutrition on cell- mediated immunity is inadequate. The results of experiments in animals are conflicting and the data suggest that cellular immune responses are either decreased, unchanged, or increased in malnutrition. Several studies in man have shown a cutaneous anergy to tuberculin during acute malnutrition (Jayalakshmi & Gopalan, 1958; Lloyd, 1968). Positive reactions to tuberculin following BCG vaccination also develop less frequently in malnourished individ- uals (Harland, 1965). The blastogenic transformation of lymphocytes tested against tetanus and streptolysin 0 antigens is slightly depressed in children with nutritional dwarfing but is found to be increased when tested against streptococcal group-specific carbohydrate and in- fluenza A (Jose et al., 1970). A significantly impaired rate of transformation of phytomaemagglutinin- stimulated lymphocytes has also been reported in children with PCM (Smythe et al., 1971). Experi- mental work in animals suggests that the humoral antibody response may be depressed at a dietary protein level at which the cellular immune response remains normal, but both are depressed in extreme protein or calorie deprivation (Jose & Good, 1971). Cell-mediated immunity may appear to be more efficient in animals with protein deficiency on account of the absence of inhibition by blocking antibody in the serum (Jose et al., 1971). Effects on immunoglobulins In cases of kwashiorkor uncomplicated by infec- tion, it has been reported that rates of synthesis of albumin, but not of IgG, are greatly reduced (Gitlin et al., 1958; Cohen & Hansen, 1962). After treatment with a high-protein diet, rates of synthesis for albu- min increased greatly, but those for IgG remained virtually unchanged. In kwashiorkor complicated by infection, however, IgG synthesis was noted to be about three times greater than in cases of uncompli- cated kwashiorkor (Cohen & Hansen, op. cit.). These data suggest that available protein may be preferen- tially directed towards immunoglobulin synthesis in severe protein-calorie malnutrition. It is known that the metabolic rate of children with kwashiorkor is subnormal, and that after treatment it rises to the normal range (Ablett & McCance, 1971). Rapid rises in IgM, IgG, and IgA levels during the first year of life are reported in studies ofcommunities where malnutrition is frequent. IgM and IgG fre- quently approach adult levels by 12 months of age (Jose & Welch, 1970; Alvarado & Luthringer, 1971; Najjar et al., 1969). Data from these types of inves- tigation must incorporate normal values obtained from the local population, as in a study made in the Gambia (McGregor et al., 1970). In direct contrast, children in Egypt with severe kwashiorkor have been reported to have very low or undetectable levels of IgM and IgA during the first year of life, and a partial defect persisted after special feeding (Aref et al., 1970). There is evidence to suggest that, as far as physical and intellectual growth are concerned, malnutrition occurring during periods of rapid growth may have more permanent and long-term effects than malnutri- tion occurring during periods when the growth rate is slow (Winick, 1969). There is no evidence to show whether the immune response is similarly affected, but studies in this field are needed. 539 MEMORANDA Table 2. Summary of immunization results in malnutrition Antigen Nutritional Age group Response ReferenceAntigen ~status gr TAB mumps virus (killed) measles virus (killed and live) influenza kwashiorkor children depressed Reddy & Srikantia (1964) kwashiorkor 6 months- 3 years retarded growth 2-5 years nutritional dwarfing Corynebacterium kwashiorkor diphtheriae poliomyelitis kwashiorkor virus yellow fever virus kwashiorkor typhoid vaccine severe PCM tobacco mosiac PCM virus depressed Kumate Rodriguez, unpublished data normal Mata, unpublished data 3-6 years depressed Jose, unpublished data 2-4 years depressed Olarte et al. (1 956) 3-5 years normal McFarlane (1970) 3-5 years depressed Brown & Katz (1966) children depressed Budiansky & Da Silva (1957) adults depressed Gell (1948) Effects on specific antibody responses Measurements of antibody response to several antigens have been made in malnourished human subjects, and the results of some of these studies are given in Table 2. The subjects were in different nutritional states and showed either depression or no effect in the antibody titre. Some studies, moreover, have implicated specific nutritional deficiencies, e.g., of pantothenic acid and pyridoxine, in the suppres- sion of primary immune responses in man (Hodges et al., 1962a, 1962b). It should be emphasized that few of these studies have carefully defined or stan- dardized the antigens used; hence the observed variations in the immune response could actually be due to differences in the antigenicity of superficially similar preparations (Snyder et al., 1966). The immunizing effect of most vaccines in relation to age and nutritional status is not known. Some factors other than nutritional ones may also depress antibody responses in children; for instance, the antibody response to fluid tetanus toxoid has been shown to be depressed in the presence of malarial infections (McGregor & Barr, 1962). Salaman et al. (1969) have also shown a suppression of the immune response to sheep erythrocytes and tumour agents in mice infected with a malarial parasite. Since there is considerable evidence in both man and laboratory animals to suggest that antibody responses may be influenced by age, sex, pituitary growth hormone, etc., it is clear that any study to assess the immuno- logical role of a state such as " malnutrition " must be designed to control the many other variables that may influence the results. IMMUNOLOGICAL ASSESSMENT Measurements of humoral immunity The measurement of specific antibody production in response to antigenic stimulation may be the most practical method of assessing humoral immune mechanisms. Antibody titres in sera should be measured at specific intervals after inoculation of a standard antigen. The use of a standard vaccine approved for human use is recommended. The test selected to measure antibody will vary with the antigen used (Gill, 1970), but different tests may measure the level of various immunoglobulin classes of the antibody being studied. This has been shown, for example, with relative efficiency studies of the capacity of specific IgM and IgG antitoxins to neu- tralize diphtheria (Robbins, 1964) and tetanus (Edsall, unpublished data) toxins. The finding of lowered serum concentrations of individual immuno- globulin classes may indicate a gross deficiency in humoral function, while elevated levels may indicate stimulation by infectious or other antigens. Both specific antibody titres and immunoglobulin concen- trations may be influenced if major fluid shifts occur 540 SURVEY OF NUTRITIONAL-IMMUNOLOGICAL INTERACTIONS 541 between body compartments, as in kwashiorkor, for example. The most precise quantitative measurement of immunoglobulin production comes from studies using purified, radioactively labelled immunoglobu- lins, providing direct measurements of the synthesis, catabolism, and distribution of individual immuno- globulins, or their subclasses, among the body pools. Each immunoglobulin class has individual turnover rates and requires individual study (Waldmann, 1969). Measurements of antibody affinity following anti- genic stimulation (WHO Scientific Group on Factors Regulating the Immune Response, 1970) provide another index of immune responsiveness. Measurements of cell-mediated immunity The testing of delayed skin reactions to several antigens to which the population would be expected to react with high frequency is the most suitable method under field conditions. The antigens could include tuberculin, antigen extracts of Candida and Trichophyton, streptokinase-streptodornase, and killed mumps virus. A positive test is indicated by induration of the skin usually greater than 5 mm in diameter at the injection site occurring after 24-72 hours (delayed hypersensitivity). In addition to these tests for pre-existing delayed hypersensitivity, the measurement of sensitivity to tuberculin following BCG vaccination should provide a useful means of assessing ability to develop cellular immunity under field conditions (Guld, 1966). Interpretation of these tests may be obscured by changes in the skin, by the presence of an antibody-mediated (immediate) reac- tion, and by subjective variation in observations made by different observers (Nyboe, 1960). The in vitro testing of peripheral blood leucocytes provides further specific information on the function of the T-cell system (WHO Scientific Group on Cell- Mediated Immune Responses, 1969). Several useful tests are listed below. (1) Quantitative counts of the small lymphocyte population provide an indication of gross cellular deficits. (2) The blastogenic transformation of peripheral leucocytes by phytohaemagglutinin (PHA) as mea- sured by the incorporation of radioactive thymidine in 3-day cultures is thought to be a measure of T-cell replication on nonspecific stimulation (Nowell, 1960; WHO Scientific Group on Cell-Mediated Immune Responses, 1969). The quantitative PHA test is an index of T-cell function, and may be the most satisfactory test for field work (Jose, unpublished data). (3) The blastogenic transformation of peripheral leucocytes by optimum concentrations of specific antigens (Oppenheim, 1969; WHO Scientific Group on Cell-Mediated Immune Responses, 1969) mea- sured by the incorporation of radioactive thymidine in 5-day cultures is presumed to measure predomi- nantly the T-cell response to antigens. This test can be made after immunization with specific antigen along with measurements of antibody responses. Some degree of B-cell activation may occur in the transformation response to some antigens, and the results should be interpreted cautiously. (4) The one-way mixed leucocyte culture reaction (Bach & Voynow, 1966) measures the in vitro re- sponse of peripheral blood T-cells to transplantation antigens of leucocytes from a second donor. No prior sensitization is required, and the test is con- sidered to measure T-cell function alone. This test might find a particular application in the assessment of T-cell function in the acute phase of severe kwashiorkor or marasmus because it does not require the immunization of the patient and therefore does not involve the usual delay for the development of an immune response. (5) The inhibition of macrophage migration by cell-free supernatants from antigen-stimulated leu- cocytes is generally agreed to be a specific test for cell-mediated immunity (David, 1966). The test probably measures the ability of T-cells to respond to antigen by releasing of one of the chemical mediators of cellular immunity called a migration- inhibition factor (WHO Scientific Group on Cell- Mediated Immune Responses, 1969). Measurements of other factors in host defence Scrimshaw et al. (1968) have reviewed and analysed some of the factors that influence host defence in malnutrition. Several measurable factors of host resistance are given in the following list. (1) Determination of the opsonic capacity of serum to enhance phagocytosis (Biggar et al., 1971; Saba, 1970); (2) Neutrophil function and number: (a) nitro-blue tetrazolium 1 reduction test (Park et al., 1968; Park, 1971); (b) endotoxin stimulation of nitro-blue tetrazoliuni reduction; 1 3,3'-(3,3'-dimethoxy-4,4'-biphenylylene)bis[2-(p-nitrophe- nyl)-5-phenyl-2H-tetrazolium dichloride. MEMORANDA (c) peripheral blood counts and diurnal variation of the neutrophil cycle; (d) mobilization of the leucocyte pool following injections of endotoxin; (3) phagocytosis and intracellular killing (Alexander et al., 1968); this could be used particularly to test organisms endemic to the area; (4) chemotaxis: Boyden chamber techniques (Boyden, 1962; Ward, 1968); (5) inflammation: skin-window assessment (Rebuck, 1955); (6) serum complement and complement components (Fong et al., 1970); (7) interferon production (Merignan, 1967); (8) lysosymal activity and measurements of B-lysin (Donaldson et al., 1964; Glynn et al., 1970). These factors, in addition to the aspects of specific immunity that have been considered above, might be studied in malnourished subjects. Other aspects of host defence are less easy to measure. They relate to the integrity of the mucosal and skin surfaces with their various secretions and cleansing mechanisms-namely, the maintenance of endocrine balance, the stability of the bowel flora, the localization and walling-off of infectious process, the healing of host tissues, and granuloma formation. NUTRITIONAL ASSESSMENT The different and varying manifestations of PCM, ranging in its severe forms from nutritional marasmus to kwashiorkor, have in common a retardation in growth and development. Some effects of nutrition on body height are listed in Table 3. This table also gives the 5 categories of PCM proposed by a Joint FAO/WHO Expert Committee on Nutrition (1971) for the next revision of the International Classifica- tion of Diseases. For an assessment of the nutritional status of groups by means of anthropometry, measurements of 3 factors are necessary-namely, weight, height, and age. Other useful measurements are upper arm circumference and skinfold thickness over the triceps muscle. An accurate registration of age is important since in the young child both the rate of growth and the character of the immune response change with age. The classification of PCM by the degree of growth deficit compares actual growth with a refer- ence growth curve. One of the most commonly used references is the Harvard growth table (Nelson, 1964) and this has been recommended for use by the Table 3. Proposed classification of PCM a Categories Body weight of PCM as a Deficit in proposed percentage Body height Oedema weight for for the revision of the height d of the ICD b reference c (1) kwashiorkor 80-60 affected + + + (2) nutritional <60 affected 0 ++ marasmus (3) mixed forms <60 affected + ++ of severe PCM (4) moderate 80-60 affected 0 moderate forms of PCM (5) Lasting effects (a) retarded 80-60 pronounced 0 minimal growth deficit (b) dwarfing <60 very 0 minimal pronounced deficit a Joint FAO/WHO Expert Committee on Nutrition (1971). b International Classification of Diseases. c Reference taken as the 50th percentile of the Harvard values for age. ~~for ~~~~ weight of patient xl.d Weight height = ofptet . 1 00.reference weight for height of patient Joint FAO/WHO Expert Committee on Nutrition (1971). A number of apparently age-independent anthropometric ratios such as arm circumference/- head circumference (Kanawati & McLaren, 1970), and head circumference/chest circumference have been considered, but more field studies are needed to establish their relationships to other, more widely used, indices. More commonly used biochemical indices of pro- tein nutrition are the levels of protein and albumin in the serum, the hydroxyproline index, and the urinary creatinine/height index. In addition, it may often be useful to measure the levels of haemaglobin, iron, folate, and vitamin A in blood or serum. Assessment of the nutritional status was reviewed by Jelliffe (1966). OTHER CONSIDERATIONS IN THE NUTRITIONAL-IMMUNOLOGICAL RELATIONSHIP Role of age Most of the IgG in the newborn child is maternally derived (Brambell, 1970), but infants are certainly able to produce antibody responses when appro- priately stimulated (Stiehm et al., 1966). The pres- ence of maternal antibodies in infants may interfere 542 SURVEY OF NUTRITIONAL-IMMUNOLOGICAL INTERACIIONS with the infant's IgG response to natural infections or vaccines, and maternal IgG remains in the infant's circulation for approximately 3-6 months. The time at which endogenous biosynthesis begins and the length of time before adult levels are attained varies with each immunoglobulin class (Stiehm & Fuden- berg, 1966) and they also depend on environmental stimuli such as bacterial flora, infections, foods, etc. No information is available about the effects of malnutrition, without additional infection, on the maturation of immunoglobulin-producing cells. Some data suggest that there are higher levels of IgA in malnourished children than in well-nourished controls (Alvarado & Luthringer, 1971; Najjar et al., 1969; Keet & Thom, 1969; El-Gholmy et al., 1970; Lechtig et al., 1970). Observations made on prenatal infections reveal that immunologically competent cells stimulated in utero show an accelerated maturation (Silverstein & Lukes, 1962; Silverstein et al., 1963). Functional alterations in the immunological system can also occur as a result of virus infections (Notkins et al., 1970). The role of early antigenic stimulation, whether in utero or in the neonatal period, is not completely understood. In one study, for example, pertussis vaccine given to newborn infants induced a state of relative immunological unresponsive- ness to further stimulation with the same vaccine that lasted for more than a year in some cases (Provenzano et al., 1965). Also, deficient delayed hypersensitivity to 1-chloro-2,4-dinitrobenzene in premature human infants has been reported by Uhr et al. (1960). Role of breast feeding Physicians in areas where environmental sanitation is poor have long recognized that breast-fed infants are frequently less affected by gastrointestinal infec- tions and infections suffered by the parents than bottle-fed infants are, and it has been suggested that breast milk contains a barrier to infection (Mata, 1971). Maternal immunoglobulins are present in colostrum and also, in smaller amounts, in mature milk. While there is no evidence that these immuno- globulins are absorbed by the child, they may exert a localized protective function in the gut. IgA is probably of particular importance in this respect because it is known to be resistant to digestive enzymes. Specific antibodies to a number of viruses and bacteria have been found in colostrum. Human milk has a high content of lysozymes, phagocytes, antibody-forming cells, and bifidus fac- tor. Bifidus factor stimulates the development of Gram-positive anaerobic bacilli (bifida bacteria) that metabolize lactase, producing large amounts of acetic and lactic acids. These acids, together with the low pH, continue to create an environment within the gut that is antagonistic to infection with Shigella and other pathogenic organisms. The fact that the pre- valence of weanling diarrhoea varies from country to country according to the age of weaning is indirect evidence that breast milk offers some protection against intestinal infections. Role of other organs It is not known whether the small number of polymorphonuclear neutrophils (PMN) reported to appear in response to infection in malnutrition (Trowell et al., 1954) results from a decreased total body pool, faulty mobilization, or are the result of a metabolic defect in the leucocytes. Neutrophils from patients with PCM have been shown to have a depressed glycolytic pathway (Selvaraj & Bhat, 1970) and leucocytes from malnourished children have decreased pyruvate (Yoshida et al., 1968), low con- centrations of lactate (Yoshida et al., 1967), and high levels of alkaline phosphatase activity (Tejada et al., 1964). Exploration by the skin-window technique has shown that the PMN/mononuclear-cell ratio in malnourished children is higher than that in well- nourished matched controls (Kumate Rodriguez, unpublished data). Anatomical alterations in malnutrition affecting the skin and mucosal surfaces, particularly in the intestine (thinning of the wall, decrease in mucus secretion), render these organs more susceptible to invasion by infectious agents or to the action of their products. This increased liability to invasion may accelerate immunological maturation of lymphoid tissues and immunoglobulins. A similar thinning of the intestinal wall is also seen in germ-free mice in the absence of a bacterial flora. Whether the endocrine alterations that occur in malnutrition have any effect on the components of the immune response remains to be determined. Alterations in certain hormones do -affect the im- mune response (Wolstenholme & Knight, 1970). Sim- ilarly, the known sexual differences in immune re- sponsiveness, morbidity, and mortality have not yet been evaluated in relation to malnutrition. Role of infections There are many descriptions of the way in which concurrent malnutrition appears to enhance the fre- 543 544 MEMORANDA quency and severity of infections, as well as the incidence and nature of ensuing complications (Scrimshaw et al., 1968). Malnourished subjects are more susceptible to infection by organisms in their environment that are not normally pathogenic; this is seen also in animal models. Coxsackie virus B3 infection, which is tolerated in normal mice, produces severe disease in mice subjected to sustained post- weaning undernutrition (Woodruff, 1970). Disease in these animals can be prevented by the introduction of an optimum diet at the time of infection (Wood- ruff, op. cit.). These aspects of malnutrition require further study and assessment, involving healthy and malnourished individuals in the same ecosphere, as well as a careful preliminary evaluation of the pattems of infection that occur in the study area. There is some evidence of depression of delayed hypersensitivity during certain systemic viral infec- tions such as measles (Christensen et al., 1953). Whether infections by other viruses induce skin anergy, reinforcing a depressive effect of malnutrition on cell-mediated immunity, and whether this com- bined effect results in further consequences are problems that require further study. Role ofspecific nutrients While some evidence for general patterns in nutri- tion-infection interactions has emerged, there is still inadequate information on specific relationships between particular nutrients, infectious agents, and host-immune mechanisms (Arbeter et al., 1971). The literature provides clear evidence that deficiencies of two general groups of nutrients may generally be expected to depress immune response-namely proteins and specific vitamins, particularly panto- thenic acid, pyridoxine, and vitamin A. These aspects have been extensively reviewed by Scrimshaw et al. (1968). * * Z. L. Awdeh, Nutrition Research Programme, American University of Beirut, Beirut, Lebanon J. Bengoa, Chief Medical Officer, Nutrition, World Health Organization, Geneva, Switzerland E. M. Demaeyer, Medical Officer, Nutrition, World Health Organization, Geneva, Switzerland H. Dixon, Statistician, Health Statistical Methodology, World Health Organization, Geneva, Switzerland G. Edsall, Superintendent, State Laboratory Institute, Boston, Mass., USA W. P. Faulk, Medical Officer, Immunology, World Health Organization, Geneva, Switzerland H. C. Goodman, Chief Medical Officer, Immunology, World Health Organization, Geneva, Switzerland B. E. C. Hopwood, The Wellcome Trust, London, England D. G. Jose, Department of Pediatrics, University of Minnesota, Minneapolis, Minn., USA W. D. E. Keller, Nutrition Adviser, Regional Office for South-East Asia, World Health Organization, New Delhi, India J. Kumate Rodriguez, Hospital Infantil de Mexico, Mexico City, Mexico L. J. Mata, Chief, Division of Microbiology, Instituto de Nutricion de Centro America y Panama, Guatemala City, Guatemala I. A. McGregor, Medical Research Council Laboratories, Fajara, The Gambia P. A. Miescher, Division d'Hematologie, H6pital Can- tonal, Geneva, Switzerland D. S. Rowe, Head, WHO Immunology Research and Training Centre, Lausanne, Switzerland C. E. Taylor, Department of International Health, Johns Hopkins University, Baltimore, Md., USA G. Torrigiani, Medical Officer, Immunology, World Health Organization, Geneva, Switzerland ACKNOWLEDGEMENTS The special contributions made by Dr P. S. E. G. Harland, Medical Research Council Child Nutrition Unit, Kampala, and by Dr C. Marigo, Departamento de Patologia, Faculdade de Ciencias M6dicas da Santa Casa, Sao Paulo, Brazil, are acknowledged by the signatories. RtiSUMI ENQUETE SUR LES INTERACTIONS ENTRE L'ETAT NUTRITIONNEL ET LA RI-PONSE IMMUNITAIRE Le pr&sent document expose tout d'abord quelques notions fondamentales d'immunologie, comme par exemple l'origine et la fonction des sous-populations de lymphocytes, puis envisage certains des effets de la nutri- tion sur la reponse immunitaire: action sur la replication cellulaire, sur la morphologie du thymus et des ganglions lymphatiques, sur 1'immunite ai support cellulaire, sur la synthese des immunoglobulines et sur 1'elaboration des SURVEY OF NUTRITIONAL-IMMUNOLOGICAL INTERACTIONS 545 anticorps specifiques. La malnutrition a generalement pour effet d'entraver la reponse immunitaire, mais celle-ci peut etre normale dans certains cas. L'hypersensibilit6 retardee est habituellement affaiblie, comme le montre l'etude des reactions tuberculiniques apres vaccination par le BCG. De meme, dans les etats de malnutrition, les lymphocytes humains cultives in vitro paraissent moins sensibles a la stimulation par la phytohemagglutinine. Diverses methodes permettent d'apprecier le fonction- nement des mecanismes immunitaires: mesure de la pro- duction des anticorps specifiques, 6valuation de l'immu- nite a support cellulaire, 6tude d'une serie d'autres fac- teurs intervenant dans la defense de l'h6te. En presence de taux trop faibles d'immunoglobulines, on doit penser a une alteration globale de la fonction humorale et du metabolisme des immunoglobulines, alors que des concentrations elev&es peuvent resulter d'une stimulation par un agent infectieux ou tout autre antigene. Le moyen le plus sur pour evaluer l'immunite humorale est de mesurer la reponse immunitaire specifique a un antigene determine apres immunisation. L'immunite a support cellulaire peut etre etudiee in vivo en mesurant la reaction cutanee d'hypersensibilit6 retardee it l'injection de cer- tains antigenes comme le derive proteinique purifie (PPD), ou in vitro en utilisant l'epreuve de stimulation des lymphocytes par la phytohemagglutinine, la r6action de melanges de cultures de lymphocytes et l'epreuve d'inhibition de la migration des macrophages. D'autres facteurs mesurables intervenant dans la defense de l'h6te peuvent etre etudies chez des sujets en etat de malnutri- tion, notamment le degre d'activite phagocytaire et de destruction intracellulaire des bacteries. L'appreciation de 1'etat nutritionnel requiert la determination de divers indices anthropometriques et l'emploi d'epreuves bio- chimiques servant d'indicateurs. L'etude des interactions entre l'etat nutritionnel et la reponse immunitaire doit aussi tenir compte d'un certain nombre d'autres elements, comme l'age du malade, le mode d'alimentation (allaitement maternel ou artificiel), l'etat de divers organes, les infections associees et la carence en certains nutriments. Lorsqu'on veut apprecier le r6le immunologique d'un facteur unique, comme la nutrition, il est donc indispensable de concevoir les enquetes de maniere a prendre en consideration les nom- breuses autres variables qui peuvent influencer les resultats. Les etudes immunologiques portant sur des sujets dont l'etat nutritionnel est altere pr6sentent un grand interet car il ressort de nombreuses observations que les infec- tions et la malnutrition ont un effet synergique et que la reponse immunitaire represente l'une des principales defenses de l'organisme contre les infections. Plusieurs recherches dans ce domaine sont actuellement en cours. REFERENCES Ablett, J. G. & McCance, R. A. (1971) Lancet, 2, 517 Alexander, J. W. et al. (1968) J. Lab. clin. Med., 72, 136 Alvarado, J. & Luthringer, D. G. (1971) Clin. Pediat. (Phila.), 10, 174 Arbeter, A. et al. (1971) Fed. Proc., 30, 1421 Aref, G. H. et al. (1970) J. trop. Med. Hyg., 73, 186 Bach, F. H. & Voynow, N. K. (1966) Science, 153, 545 Behar, M. et al. (1956) Rev. Col. med. Guatemala, 7, 221 Biggar, W. D. et al. (1971) Proc. nat. Acad. Sci. (Wash.) 68, 8 Boyden, S. (1962) J. exp. Med., 115, 454 Brambell, F. W. R. (1970) The transmission of passive immunity from mother to young, Amsterdam, North- Holland Brown, R. F. & Katz, M. (1966) Trop geogr. Med., 18, 125 Budiansky, E. & Da Silva, N. N. (1957) Hospital (Rio de J.), 52, 251 Christensen, P. 0. et al. (1953) Acta med. scand., 144, 430 Cohen, S. & Hansen, J. D. L. (1962) Clin. Sci., 23, 351 Cooper, M. D. et al (1966a) Lancet, 1, 1388 Cooper, M. D. et al. (1966b) J. exp. Med., 123, 75 Cooper, W. C. et al. (1970) Fed. Proc., 29, 364 David, J. R. (1966) Proc. nat. Acad. Sci. (Wash.) 56, 72 Donaldson, D. M. et al. (1964) J. Immunol., 92, 896 El-Gholmy, A. et al. (1970) J. trop. Med. Hyg., 73, 192 Faulk, W. P. et al. (1970) Cell. Immunol., 1, 500 Fichtelius, K. E. (1967) Exp. Cell Res., 46, 231 Fong et al. (1970) J. Lab. clin. Med., 76, 836 Gell, P. G. H. (1948) Proc. roy. Soc. Med., 41, 323 Gill, T. J. (1970) Immunochemistry, 7, 977 Gitlin, D. et al. (1958) J. clin. Invest., 37, 682 Glynn, A. A. et al. (1970) Nature (Lond.), 225, 77 Good, R. A. et al. (1970) In: S. Mudd, ed., Infectious agents and host reactions, Philadelphia, Saunders Gotze, 0. & Muller-Eberhard. H. J. (1971) J. exp. Med., 134, 905 Guld, J. (1966) Bull. Wld Hlth Org., 35, 479 Harland, P. S. E. (1965) Lancet, 2, 719 Hodges, R. E. et al. (1962a) Amer. J. clin. Nutr. 11, 85 Hodges, R. E. et al. (1962b) Amer. J. clin. Nutr., 11, 187 Jayalakshmi, V. T. & Gopalan, C. (1958) Indian J. med. Res., 46, 87 Jelliffe, D. B. (1966) The assessment of the nutritional status of the community, Geneva, World Health Organization (WHO Monogr. Ser., No. 53) Joint FAO/WHO Expert Committee on Nutrition (1971) Wid Hith Org. techn. Rep. Ser., No. 477 Jose, D. G. & Good, R. A. (1971) Nature (Lond.), 231, 323 Jose, D. G. & Welch, J. S. (1970) Med. J. Aust., 1, 349 Jose, D. G. et al. (1970) Aust. paediat. J., 6, 192 Jose, D. G. et al. (1971) J. amer. Med. Ass., 218, 1428 Kanawati, A. A. & McLaren, D. S. (1970) Nature (Lond.), 228, 573 Keet, M. P. & Thom, H. (1969) Arch. Dis. Childh., 44, 600 546 MEMORANDA Kumate, J. (1967) Gac. med. Mix., 97, 1563 Lechtig, A. et al. (1970) Arch. Latinoamer. Nutr., 20, 321 Lloyd, A. V. C. (1968) Brit. med. J., 3, 529 Mata, L. J. (1971) Amer. J. clin. Nutr., 24, 249 McFarlane, H. (1970) Trop. geogr. Med., 22, 61 McGregor, I. A. & Barr, M. (1962) Trans. roy. Soc. trop. Med. Hyg., 56, 364 McGregor, I. A. et al. (1970) Clin. exp. Immunol. 7, 51 Merigan, T. C. (1967) New Engl. J. Med., 276, 913 Montjar, M. et al. (1965) J. Nutr., 85, 45 NajiJar, S. S. et al. (1969) Arch. Dis. Childh., 44, 120 Nelson, W. E., ed. (1964) Textbook ofpediatrics, Phila- delphia, Saunders, p. 48 Notkins, A. L. et al. (1970) Ann. Rev. Microbiol., 24, 525 Nowell, P. C. (1960) Cancer Res., 20, 462 Nyboe, J. (1960) Bull. Wld Hlth Org., 22, 5 Olarte, J. et al. (1956) Bol. mid. Hosp. infant. (Mix.), 13, 467 Oppenheim, J. J. (1969) Ann. Allergy, 27, 305 Park, B. H. (1971) J. Pediat., 78, 376 Park, B. H. et al. (1968) Lancet, 2, 532 Provenzano, R. W. et al. (1965) New Engl. J. Med., 273, 959 Rebuck, J. W. (1955) Ann. N.Y. Acad. Sci., 59, 757 Reddy, V. & Srikantia, S. G. (1964) Indian J. med. Res., 52, 1154 Robbins, J. B. (1964) In: J. Sterzl, ed., Molecular and cellular basis of antibody formation, Prague, Czecho- slovak Academy of Sciences, p. 241 Saba, T. M. (1970) Arch. intern. Med., 126, 103 Salaman, M. H. et al. (1969) J. gen. Microbiol., 59, 383 Scrimshaw, N. S. et al. (1968) Interactions of nutrition and infection, Geneva (World Health Organization: Monograph Series, No. 57) Selvaraj, R. J. & Bhat, K. S. (1970) In: Annual Report of the National Institute of Nutrition, Hyderabad, India, p. 63 Seminars in Hematology, 1970, 7 Silverstein, A. M. & Lukes, R. J. (1962) Lab. Invest., 11, 918 Silverstein, A. M. et al. (1963) J. exp. Med., 117, 799 Smythe, P. M. et al. (1971) Lancet, 2, 939 Snyder, H. E. et al (1966) J. Trauma, 4, 529 Stiehm, E. R. et al. (1966) New Eng. J. Med., 275, 971 Stiehm, E. R. & Fudenberg, H. H. (1966) Pediatrics, 37, 715 Tejad-, C. et al. (1964) J. Pediat., 64, 753 Trowell, H. C. et al. (1954) Kwashiorkor, London, Arnold Uhr, J. W. et al (1960) Nature (Lond.), 187, 1130 Venrooij, W. J. W. van et al. (1970) J. biol. Chem., 245, 5947 Vint, F. W. (1930) East. Afr. med. J., 13, 322 Waldmann, T. A. (1969) New Engl. J. Med., 281, 1170 Ward, P. A. (1968) J. exp. Med., 128, 1201 Watts, T. (1969) J. trop. Pediat., 15, 155 WHO Scientific Group on Cell-Mediated Immunity (1969) Wld Hlth Org., tech. Rep. Ser., No. 423 WHO Scientific Group on Factors Regulating the Immune Response (1970) Wld Hlth Org., tech. Rep. Ser., No. 448 Winick, M. (1969) J. Pediat., 74, 774 Wolstenholme, G. E. & Knight, J., ed. (1970) Control processes in multicellular organisms, London, Churchill Woodruff, J. F. (1970) J. Infect. Dis., 121, 164 Woodruff, J. F. & Woodruff, J. J. (1971) Proc. nat. Acad. Sci (Wash.), 68, 2108 Yoshida, T. et al. (1967) Nature (Lond.) 214, 525 Yoshida, T. et al. (1968) Amer. J. clin. Nutr., 21, 162