Baixe Impact of Macronutrients and Micronutrients on Immune Function: An In-depth Analysis e outras Notas de estudo em PDF para zootecnia, somente na Docsity!
2 Evaluation of the Effects of
Nutrients on Immune Function
SUSANNA CUNNINGHAM-RUNDLES
Immunology Research Laboratory, Division of Hematology and Oncology, Department of Pediatrics, New York Presbyterian Hospital, Cornell University Weill Medical College, 1300 York Avenue, New York, NY 10021, USA
Introduction and Overview
Nutrients are primary factors in the regulation of the human immune response. Both macronutrients and micronutrients derived from the diet affect immune-system function through actions at several levels in the gastrointesti- nal tract, thymus, spleen, regional lymph nodes and immune cells of the circu- lating blood (Chandra, 1997; Cunningham-Rundles and Lin, 1998; Wallace et al. , 2000; Cunningham-Rundles, 2001). Effects at one level may be opposed or modified at another level. Thus, the development of an experimental approach capable of revealing critical interactions requires study of more than one aspect of immune function (Cunningham-Rundles, 1993; Muga and Grider, 1999; Beisel, 2000). The effect of any single nutrient is dependent upon concentration, interactions with other key nutrients, host genetic expres- sion and internal environmental conditions. In situations of nutrient imbal- ance, duration of the altered condition and age of the host are also often critical factors (Cunningham-Rundles and Cervia, 1996; Hirve and Ganatra, 1997; Miles et al. , 2001). Nutrients affect specific immune–cell types differently through influencing intrinsic cell function and by influencing cell–cell interactions. Much of the critical action appears to occur in the local microenvironment during the response to antigen. Classically, the immune system has been considered as an operational duality divided into an innate system, mediating immune reactions that do not functionally change with re-exposure to signal, and an adaptive immune system, which is capable of developing the response to antigen encounter and evolving with re-exposure. Adaptive immunity has been further characterized according to cell type, as the response of bone- marrow-derived B-cells of the humoral immune system and thymus-derived T-cells of the cellular immune system. This rather static picture of compart- mentalized function is changing. Now, it is increasingly clear that significant T-
© CAB International 2002. Nutrition and Immune Function (eds P.C. Calder, C.J. Field and H.S. Gill) 21
cell differentiation does occur independently of the thymus – for example, in the gastrointestinal tract. Current studies also show that the innate immune system, mediated by such cells as natural killer (NK) and NK T-cells, mono- cytes and dendritic cells, influences the nature of cytokine production by the adaptive immune system. This occurs through secretion of cytokines by innate immune cells into the microenvironment (Doherty et al. , 1999; Garcia et al. , 1999; see also Devereux, Chapter 1, this volume). The effect of the microenvironment is to drive the immune response towards either a T-helper type 1 (Th1) or a T-helper type 2 (Th2) response (see Devereux, Chapter 1, this volume). Micronutrients, such as trace elements and vitamins, are present in the local environment and have important regulatory effects on adaptive immune-cell function. For example, the trace element zinc supports a Th response, whereas vitamin A appears to produce a Th2 response (Frankenburg et al. , 1998; Shankar and Prasad, 1998). Thus the new immunology provides a more fluid representation of a potentially evolving process that presents as a defined pattern according to an environmental dynamic rather than a static programme that is derived from fixed cellular characteristics. The basic elements are shown in Fig. 2.1.
22 S. Cunningham-Rundles
Innate immune system IFN- IL-12 Th
Th
Th
APC
IL-
IL-
Adaptive immune system
Activated APC
Classical analysis Current analysis Innate immune system Phagocytes NK cells
Microenvironment Antigen-presenting cells NK and NK T-cells Cytokine milieu Antigen processing MHC class I MHC class II Super antigen CD1-restricted
Adaptive immune system Humoral: B-cells Cellular: T-cells ▫ CD ▫ CD Response pathways Th Th
Fig. 2.1. Microenvironment of immune response. APC, antigen-presenting cell; IFN-, interferon-; IL, interleukin; MHC, major histocompatibility complex.
of the immune system has been affected. Immune studies are often based on limited studies of immune-cell subsets, serum or plasma concentrations of cytokines or the functional response of mononuclear cells cultured in highly standardized systems, using a chosen stimulus and often a single end-point. Newer methods have made it possible to assess differentiation in antigen expression on peripheral-blood mononuclear cells in response to activation, to study early events in the activation pathway and to analyse gene activation. The development of cytokine biology has provided a critical means of clar- ifying the fundamental impact of nutrients on immune response. In general, nutrients appear to affect the immune system most profoundly through regula- tory mechanisms affecting the expression and production of cytokines (e.g. Savendahl and Underwood, 1997; Rink and Kirchner, 2000). Since the type of cytokine pattern produced is crucial for the response to infectious pathogens, serious nutrient imbalance will ultimately compromise the development of the future immune response. However, while malnutrition promotes susceptibility to pathogens, even subclinical infections directly affect nutrient intake and metabolism. Severe, acute infection will have a very strong impact. The fact that cytokine production during the acute-phase response to generalized sepsis can lead to loss of lean tissue and body fat is well known (Lin et al. , 1998). Interestingly, this cascade of events can be altered by nutritional intervention (Jeevanandam et al. , 1999). Immune deficiency and susceptibility to infection are often directly linked with malnutrition, which was the leading cause of acquired immune deficiency before the appearance of the human immunodefi- ciency virus (HIV). Malnutrition is also a major factor contributing to the pro- gression of HIV infection, especially in less developed countries. Since malnutrition and HIV affect the host in similar ways, the combination is particu- larly devastating. Many of the infections observed in human protein–energy malnutrition (PEM), such as tuberculosis, herpes, Pneumocystis carinii pneu- monia and measles, are caused by intracellular pathogens, indicating that the cellular immune system is particularly affected (Keusch, 1993; see Chandra, Chapter 3, this volume). While the effects of infection and malnutrition on the immune response are interactive, the effects of each upon immune response are also independent. A recent examination by Mishra et al. (1998) of graded PEM in children in rela- tionship to tuberculosis infection and response to a skin-test anergy panel, including purified protein derivative of M. tuberculosis (PPD), has shown that impaired cellular immunity was observable in all grades of malnutrition, except for response to PPD in grade I, and that infection did not affect this. Differentiation of lymphocyte subpopulations is also directly affected by malnutrition. Studies show that T-cells from children with severe PEM are immature, compared with those from well-nourished children, and that the degree of immaturity is directly associated with thymic involution, as mea- sured by echo radiography (Parent et al. , 1994). While nutritional repletion affected anthropometric measures within 1 month, regrowth of the thymus took longer (Chevalier et al. , 1996, 1998). The long-term consequences of slow thymic regrowth are unknown. These studies underscore the importance of longitudinal studies.
24 S. Cunningham-Rundles
Response to certain pathogens may actually be enhanced in some states of malnutrition. Genton et al. (1998) assessed the incidence of malaria in children in Papua New Guinea, and found that increased height-for-weight at baseline (an indicator of a better nutritional state) predicted susceptibility to malaria during the year of study and that the lymphocyte response to malarial antigens was lower among the less wasted children. Furthermore, cytokine production towards malarial antigens was greater among malnourished children, suggesting that a favourable cytokine regulatory shift might be the basis of improved response among stunted, but not wasted, children. Stunting has often been considered as an adaptive and partially protective host response to prolonged nutrient depriva- tion. Rikimaru et al. (1998) evaluated lymphocyte subpopulations and immunoglobulins among healthy children and children with kwashiorkor, maras- mus and marasmic kwashiorkor in Ghana. Interestingly, immunoglobulin A (IgA) and C4 were higher, whereas C3 and relative B-cell percentage were lower, in the severely malnourished groups. These studies demonstrate the advantages of using linked measurements to develop a full immunological profile. In summary, the study of nutrient immune interaction requires considera- tion of the setting and a design that includes evaluation of possible comple- mentary effects at more than one level. Longitudinal studies are often useful and permit assessment of the evolution of the immune response and character- ization of downstream effects, which may modulate outcome.
Evaluation of Human Immune Response
Until recently, methods for evaluating the human immune system were derived largely from experimental approaches designed to analyse deficits in host defence in specific clinical settings. With the advent of molecular approaches, immune function has been studied more directly and has led to clarification of specific pathways. As a result, the molecular basis of primary and acquired immune deficiency syndromes is better understood. In addition, the develop- ment of vaccines and the study of the natural response to infectious exposure have expanded exponentially in the wake of the HIV crisis, leading to the development of increasingly targeted methods of measuring the immune response. While assessment of the humoral immune response at the level of specific antibody is now well standardized and often routine, evaluation of the complex interactions that are needed to produce specific antibody and the idio- typic interactions that govern this remains a specialized research endeavour. The study of the cellular immune response as a whole continues to remain largely a research activity, although this is beginning to change. This discussion will focus on methods that have been applied to the study of nutrients, and will include approaches that have led to new discoveries in other areas. The most widely applied methods of evaluating T lymphocyte activation have used peripheral-blood mononuclear cells, isolated by density-gradient centrifugation and cultured with plant lectins (mitogens), or bacterial or viral activators, or antigens, which elicit a secondary response that depends upon prior priming or natural exposure in vitro (Paxton et al. , 2001). The typical
Nutrients and Immune Function 25
Evidence that nutrients have direct effects on human host defence has come mainly from clinical observations and field studies in settings of severe or chronic nutrient deficiency. These investigations are often complicated by host environmental factors or by exposure to toxins, carcinogens, pathogens or endemic infection (Blot et al. , 1993; Zhang et al. , 1995; Giuliano et al. , 1997; Dai and Walker, 1999). While many studies have described interesting and potentially critical associations, few have identified causal relationships. No sin- gle investigational design is necessarily capable of revealing the causal links that govern these intricate relationships. The choice of study population is fundamental and this directly affects the kinds of controls that are needed. Laboratory controls are highly informative for internal technical quality if run in parallel with subject studies. In some cases, this can be achieved by using aliquots of frozen cells from the same donor, but this has the disadvantage of not providing information concerning the normal range. Parallel controls should include fresh samples from subjects matched for age, sex and clinical status. Longitudinal studies may be crucial and, in some cases, may enable the use of each subject as his/her own control. When the study design is observational and a nutrient or immune abnor- mality is known or suspected, study of other potentially related immune-func- tion variables becomes critical. For example, both Th1 and Th2 cytokines should be measured when a Th1 deficiency is suspected. In the context of intervention studies, reliable data can be obtained using different designs, such as placebo-controlled, double-blind and crossover. Inferences may also be drawn from some single-arm studies with unambiguous and quantifiable end- points. In some cases, it has been possible to use experimental depletion and repletion of the same study group. In other cases, lingering effects have blurred distinctions. For greater stringency, it may be necessary to include several reple- tion arms at graded doses and to follow changes for a length of time, since the immune system often shows a transient rebound effect that is not seen at later time points. It is also essential to measure other nutrient levels that are posi- tively or negatively regulated by the nutrient under study.
Experimental approach
Immune activators Immune activation requires a signal when circulating blood is used as the cell source, since the peripheral-blood lymphocyte is a resting cell. This signal is often a plant lectin, or another signal, such as certain divalent cations, calcium ionophores or surface-reactive molecules, including monoclonal antibodies to CD3, which provide a non-antigenic stimulus that activates T lymphocytes independently of antigenic history. Impaired response to mitogens in human settings of PEM may or may not be accompanied by loss of response to pathogens. Examples include the study of response to PPD in malnourished children at risk of tuberculosis and the effect of stunting on the response to malarial antigens (discussed above). It is well known that infections with even
Nutrients and Immune Function 27
relatively non-pathogenic viruses, such as measles, are often fatal in children with PEM, because measles-virus infection causes a serious but usually tran- sient suppression of the cellular immune response (Schlender et al. , 1996; Ito et al. , 1997), which, in the malnourished host, may continue to prevent immune clearance. Longitudinal studies are often essential to demonstrate long-term effects, such as the lingering effect of vitamin A deficiency, which increases mortality from infections (West et al. , 1999). Current studies suggest that the specificity of the response, defined as a Th1 or Th2 cytokine-pattern, to a specific microbe is critically associated with host defence. Study designs that incorporate antigens that are actually being encountered at the time of study or that focus on the type of cytokine production may therefore provide important and unique information.
Selection of methodology Good study design is based on the formulation of a clear question that addresses a critical issue. Table 2.1 illustrates how the integration of the study design with a well-chosen methodology can lead to informative results in differ- ent areas of research. A balance of human and experimental animal-model studies is presented, since development of this field has depended upon both. The study of whole foods, fats and certain micronutrients and how these could influence immune function is currently under development. Fundamental observation of human PEM has shown that generalized malnu- trition leads to impaired immune response and susceptibility to infection (see Chandra, Chapter 3, this volume). However, direct examination of how dietary intake of any particular nutrient affects the immune response is a complex undertaking. Table 2.1 includes four studies on dietary intake. Labeta et al. (2000) addressed the fundamental question of how human milk might activate the neonatal immune system by molecular mimicry through the isolation and sequencing of a relevant polypeptide. Fawzi et al. (2000) focused on how a whole food, specifically tomatoes, may protect against morbidity and mortality, an idea that has come from studies implicating antioxidants as improving immune function (see Hughes, Chapter 9, this vol- ume). The relationship held true even with correction for total vitamin A level (Fawzi et al. , 2000). The strength of this study is derived from the large scale
- more than 28,000 infants studied – and the careful surveys conducted, combined with excellent statistical analysis. When stress is added to the equation, the nutrient requirements for immune response are further altered, and development of hypotheses often requires more than one approach. For example, there are extensive observa- tions showing that total parenteral nutrition suppresses immune response in the surgical patient and related studies showing that glutamine becomes a condi- tional essential amino acid during metabolic stress (Calder and Yaqoob, 1999; O’Flaherty and Bouchier-Hayes, 1999; see also Calder and Newsholme, Chapter 6, this volume). These observations have led to the discovery that nutrients provide an essential stimulus for the induction, differentiation and maintenance of the mucosal immune system. Lack of enteral dietary intake
28 S. Cunningham-Rundles
impairs mucosal IgA and secretory-component production, the number of IgA- containing cells and the level of IgG and promotes mucosal growth (Heel et al. , 1998; Kudsk et al. , 2000). Even foods such as indigestible saccharides can have a stimulating effect upon the immune system (Kudoh et al. , 1998). The study of Kudsk et al. (2000), included in Table 2.1, has added significantly to this field, showing specific differences among animals fed on laboratory food, by total parenteral nutrition and by parenteral nutrition supplemented with glut- amine on the pattern of cytokine and IgA production. Loss of nutrient stimula- tion led to loss of total lymphocyte number in Peyer’s patches, in the intraepithelial layer and in the lamina propria, a reduced CD4+ T-cell to CD8+ T-cell ratio and a reduced intestinal level of IgA (Kudsk et al. , 2000). Furthermore, lack of enteral nutrition may signal increased neutrophil recruit- ment through up-regulation of the intercellular adhesion molecule 1 (ICAM-1), causing increased leucocyte binding in the intestine (Fukatsu et al. , 1999). These studies indicate how the immune response during stress may be modu- lated experimentally by specific amino acids in the diet. The study of lipids provides great challenges for study design, since incor- poration into membranes, as well as direct effects on metabolic pathways, must be considered. There is increasing evidence that increase in fat intake may impair immune function, as well as leading to obesity (Nieman et al. , 1996). A relationship between fat intake and cancer risk has been indicated (Risch et al. , 1994), but the mechanisms remain unclear. Recent data demonstrate that the fatty-acid composition of cellular membranes can cause immune perturbation. Mechanisms of action include modulation of adhesion-molecule expression (Miles et al. , 2000) and are apparently related to specific fatty-acid composi- tion. The activation state of the cell is a determining factor in how fatty acids affect the immune response (Wallace et al. , 2000). This topic has been addressed by Wallace et al. (2001) in a thorough study in which mice were fed low-fat diets or high-fat diets, containing either saturated or unsaturated fats. Both n-3 and n-6 polyunsaturated fatty acids were used, permitting distinction of their effects on cytokine production. Data showed that n-3 fatty acids were strongly suppressive of Th1 cytokines (see also Calder and Field, Chapter 4, this volume). This classic feeding study included measurement of fatty-acid incorporation, cytokine secretion and cytokine mRNA production. Other work has shown how emerging information about the human genome may be used to study basic mechanisms. For example, the discovery of the gene HFE has revealed that the molecular basis of hereditary haemochromatosis, which involves increased iron uptake from the gastroin- testinal tract, can be attributed to homozygous inheritance of one mutation (Feder et al. , 1998; Gross et al. , 1998). HFE regulates the metabolism and dis- tribution of iron by affecting the binding of iron to transferrin, is a major histo- compatibility complex (MHC) class I protein and is also non-covalently associated with 2 -microglobulin ( 2 m). The significance of this physical associ- ation is unclear. Excess iron has been observed in association with loss of CD8+ T lymphocytes in the 2 m-knockout mouse. CD8+ T-cells are also reduced in a subgroup of haemochromatosis patients who show an increased rate of iron loading (Porto et al. , 1997). The low-CD8 phenotype is also
30 S. Cunningham-Rundles
observed in a subset of patients with transfusion-related iron overload (Cunningham-Rundles et al. , 2000). Interestingly, studies in compound mutant mice lacking both HFE and 2 m have shown that more iron was deposited in various tissues than was observed in mice with either mutation alone (Levy et al. , 2000). However, studies in genetic-deletion models (e.g. the work of Bahram et al., 1999) indicate that the basis of a putative link between immune function and iron handling remains unresolved. Good study design is critically important for studies in complex settings, such as HIV infection, where nutrient imbalance is fundamentally linked to infection but hard to study in a clear-cut manner. Weight loss is a common occurrence in general chronic viral illness and, in the case of HIV infection, can evolve into a wasting condition, which may become intractable with failure of antiretroviral therapy. Infection-induced malnutrition, as discussed above, is pri- marily cytokine-mediated and is associated with the acute-phase response. This is accompanied by multiple effects on metabolism, such as altered fluxes of iron and zinc and loss of nitrogen, potassium, magnesium, phosphate, zinc and vita- mins. This process is accompanied by retention of salt and water. Malnutrition may also present during the asymptomatic phase of HIV infection (Niyongabo et al. , 1997; Peters et al. , 1998). Many studies have shown that micronutrient status is profoundly affected in HIV infection, but the aetiological significance of these changes has been difficult to demonstrate (Cunningham-Rundles, 2000). Therefore, the work of Campa et al. (1999), included in Table 2.1, has pro- vided an important advance. Using careful longitudinal studies and good statis- tical design, this group was able to establish that selenium deficiency in children with acquired immune deficiency syndrome (AIDS) was independently associ- ated with mortality.
Immune assessment New assay methods have enabled the design of experiments addressing differ- ent stages involved in immune-cell activation and the study of effects on sig- nalling pathways, which may then lead to the characterization of causal relationships. Table 2.2 outlines some of the types of methods currently in use. Most investigations begin with a general assessment of how a nutrient or altered nutritional state affects the general parameters of the immune system, immune-cell subsets and function. Measurement of changes in frequency and number of circulating lymphocyte subpopulations in the course of observation or dietary intervention is now accepted as a useful and widely comparable pro- cedure, but attention must be given to the issue of controls This analysis should include standardized performance of immunophenotyping, using correction for purity of the gating region, quantitative recovery of the cell type and positive identification of cellular subsets. For human studies, a complete blood count and differential are needed to quantify effects on absolute numbers of cells. Although there is frequently a limitation on blood to be drawn for nutritional studies, it is essential that the baseline evaluation includes parallel studies pro- viding a complete blood count, haematological analysis of haemoglobin, haematocrit, etc., on an aliquot of the same specimen of blood.
Nutrients and Immune Function 31
In addition to assessment of relative percentages of T-cells, B-cells and NK cells, immunophenotyping for activation-antigen expression (e.g. CD69), coex- pression of critical molecules involved in cell–cell interaction (e.g. CD28), T-cell receptor (TCR) changes and percentages of naive and memory cells may be informative. Functional studies should be carried out on fresh anticoagulated blood whenever possible (or blood stored at room temperature in the dark for under 24 h) before mononuclear cells are isolated. When blood is being sent by air or transported to a distant laboratory, it is extremely important to include a control specimen drawn in parallel to serve as an internal standard for the ship- ping process. In addition, the type of tube chosen to draw the blood is impor- tant. Lithium heparin- or ethylenediamine tetra-acetic acid (EDTA)-containing tubes cannot be used for functional studies. Sodium heparin (preservative-free) or acid citrate dextrose (ACD) tubes should be used and consistency of tube type is important. There may be differences between venous and arterial blood. The question of when the blood should be drawn is important. In gen- eral, most data have been obtained with blood drawn in the morning and there are circadian effects on hormones and immune-cell phenotypes that may influ- ence results. When this cannot be done, it is helpful to continue to maintain a uniformity of drawing time for an individual subject or group. Concurrent con- trol blood must be drawn to ensure that technical performance standards are met. It is important that positive and negative (normal range and abnormal range) controls be included. Double-baseline studies – as a minimum, before and after intervention is undertaken – are recommended. Studies of immune function usually start with a general assessment of mononuclear response in vitro to a mitogen, to another non-specific activator or to antigen, as discussed above. These methods are generally based on assay of cell division at the peak of response following microtitre plate culture for several days. Culture methods profoundly affect results, and conditions need to be opti- mized according to the kinetics of the response. Responses measured under most conditions favour T-cell proliferation, as the T-cell is the most prevalent lympho- cyte in peripheral blood. The elicited composite response is highly quantitative when radioactive tracers – usually thymidine – are used. Recently, whole-blood methodology has been introduced as an alternative ex vivo method that can reflect potential response in vivo (Sottong et al. , 2000); this method correlates with the level of DNA synthesis found when isolated mononuclear cells are cul- tured under optimal standard conditions. Comparative studies have also shown that there is a significant correlation between the whole-blood method and iso- lated mononuclear cells for cytokine production (Yaqoob et al. , 1999). Some lab- oratories have replaced thymidine incorporation assays with a combination of cell-surface marker-induction assays and a measurement of the percentage of cells in various phases of the cell cycle following activation. Dyes have been developed that stably integrate into the membranes of live lymphocytes, such that, with each successive division, the amount of dye per cell is decreased. Fluorescence can be used to measure the number of cell divisions. Other assays based on whole blood measure early responses of cells selected through adher- ence to magnetic beads to which monoclonal antibodies recognizing cells of par- ticular interest are attached. Assessment is achieved by an assay of adenosine
Nutrients and Immune Function 33
triphosphate (ATP) production by the luciferin/luciferase reaction (Sottong et al. , 2000). Assays such as this may provide accurate assessment of in vivo response in vitro. This method may be combined with a quantitative measure of specific lymphocyte subsets by flow cytometry for examination of response per cell. Other approaches use measurement of cytokine response, receptor up-regu- lation or activation antigen to assess initial immune response, rather than the secondary response of cells recruited in the amplified reaction. Also, in vivo reg- ulation of the immune response can be assessed through evaluation of unstimu- lated levels of secreted products when the producer-cell source of these products and normal levels are known. Methods measuring early events in T lymphocyte activation may or may not correlate with cell division, since cell division is only one aspect of the immune response. One of the earliest events that occurs fol- lowing T-cell activation is the rapid increase in intracellular free calcium. This is followed by a change in pH and changes in the membrane potential. All of these effects can be measured by flow cytometry, using functional probes. Following T- cell activation via CD3/TCR or via CD2 (the alternate T-cell activation pathway), the first measurable surface marker induced is CD69. This marker is a disul- phide-linked homodimer that is present on 20–30% of normal thymocytes, but which is not expressed on resting peripheral-blood lymphocytes. CD69 reaches peak levels within 18–24 h and declines if the stimulus is removed. Using flow cytometry, it is possible to measure increase in CD69 expression on specific lym- phocyte subsets. It is apparent that CD69 induction is not part of the pathway leading to cell division, as induction of CD69 can occur without subsequent cell proliferation. A good way to measure CD69 expression is to consider the relative expression of this marker on the subpopulation of interest, as this removes the confounding effect of subpopulation size. Other cell-surface markers appear on activated T-cells at variable times following activation, including CD25 (the chain of the interleukin-2 (IL-2) receptor) and the transferrin receptor CD (both within 24–48 h) and human leucocyte antigen (HLA)-DR (after 48 h). Finally, statistical evaluation is crucial to all of the studies described here. This includes evaluation of both the internal and the study-group controls. Studies of certain types may be suitable for the collection and banking of specimens prior to assay, such as cytokine supernatants. This may be helpful in giving a homoge- neous data set with a low coefficient of variation, as long as controls and experi- mental specimens are run simultaneously. Good design is often based on internal cross-checks, which can be developed from the working hypothesis and which allow for different elements in the same pathway to be considered. In summary, the emerging field of nutritional immunology has benefited from the evolution of cellular and molecular immunology. New approaches have provided a strong foundation for experimental design and offer a choice of analytical methods for approaching hypothesis testing. The key to any specific investigation is the identification of clear questions and the choice of relevant and practical methods. These methods then need to be tested in a pilot study, before launching the investigation. The use of an integrated design, including biostatistical considerations and complementary assays, is important in the development of meaningful data and of critical knowledge.
34 S. Cunningham-Rundles
Cunningham-Rundles, S. and Lin, D.H. (1998) Nutrition and the immune system of the gut. Journal of Nutrition 14, 573–579. Cunningham-Rundles, S., Giardina, P., Grady, R., Califano, C., McKenzie, P. and DeSousa, M. (2000) Immune response in iron overload: implications for host defence. Journal of Infectious Disease 182, s115–s121. Cunningham-Rundles, S., McNeely, D. and Ananworanich, J. (2002) Immune response in malnutrition. In: Stiehm, E.R., Ochs, H.D. and Winkelsten, J.A. (eds) Immunologic Disorders in Infants and Children , 5th edn. W.B. Saunders, Philadelphia (in press). Dai, D. and Walker, W.A. (1999) Protective nutrients and bacterial colonization in the immature human gut. Advances in Pediatrics 46, 353–382. Doherty, D.G., Norris, S., Madrigal-Estebas, L., McEntee, G., Traynor, O., Hegarty, J.E. and O’Farrelly, C. (1999) The human liver contains multiple populations of NK cells, T-cells, and CD3+CD56+ natural T-cells with distinct cytotoxic activities and Th1, Th2, and Th0 cytokine secretion patterns. Journal of Immunology 15 (163), 2314–2321. Fawzi, W., Herrera, M.G. and Nestel, P. (2000) Tomato intake in relation to mortality and morbidity among Sudanese children. Journal of Nutrition 130, 2537–2542. Feder, J.N., Penny, D.M., Irrinki, A., Lee, V.K., Lebron, J.A., Watson, N., Tsuchihashi, Z., Sigal, E., Bjorkman, P.J. and Schatzman, R.C. (1998) The hemochromatosis gene product complexes with the transferrin receptor and lowers its affinity for ligand bind- ing. Proceedings of the National Academy of Sciences of the USA 95, 1472–1477. Frankenburg, S., Wang, X. and Milner, Y. (1998) Vitamin A inhibits cytokines produced by type 1 lymphocytes in vitro. Cellular Immunology 185, 75–81. Fukatsu, K., Lundberg, A.H., Hanna, M.K., Wu, Y., Wilcox, H.G., Granger, D.N., Gaber, A.O. and Kudsk, K.A. (1999) Route of nutrition influences intercellular adhesion molecule-1 expression and neutrophil accumulation in intestine. Archives of Surgery 134, 1055–1060. Garcia, V.E., Uyemura, K., Sieling, P.A., Ochoa, M.T., Morita, C.T., Okamura, H., Kurimoto, M., Rea, T.H. and Modlin, R.L. (1999) IL-18 promotes type 1 cytokine production from NK cells and T-cells in human intracellular infection. Journal of Immunology 15 (162), 6114–6121. Genton, B., Al-Yaman, F., Ginny, M., Taraika, J. and Alpers, M.P. (1998) Relation of anthropometry to malaria morbidity and immunity in Papua New Guinean chil- dren. American Journal of Clinical Nutrition 68, 734–741. Giuliano, A.R., Papenfuss, M., Nour, M., Canfield, L.M., Schneider, A. and Hatch, K. (1997) Antioxidant nutrients: associations with persistent human papillomavirus infection. Cancer Epidemiology Biomarkers and Prevention 6, 917–923. Gross, C.N., Irrinki, A., Feder, J.N. and Enns, C.A. (1998) Co-trafficking of HFE, a non- classical major histocompatibility complex class I protein, with the transferrin recep- tor implies a role in intracellular iron regulation. Journal of Biological Chemistry 21, 22068–22074. Heel, K.A., Kong, S.E., McCauley, R.D., Erber, W.N. and Hall, J. (1998) The effect of minimum luminal nutrition on mucosal cellularity and immunity of the gut. Journal of Gastroenterology and Hepatology 10, 1015–1019. Hirve, S. and Ganatra, B. (1997) A prospective cohort study on the survival experience of under five children in rural western India. Indian Pediatrics 34, 995–1001. Insoft, R.M., Sanderson, I.R. and Walker, W.A. (1996) Development of immune function in the intestine and its role in neonatal diseases. Pediatric Clinics of North America 43, 551–571.
36 S. Cunningham-Rundles
Ito, M., Watanabe, M., Kamiya, H. and Sakurai, M. (1997) Changes in intracellular cytokine levels in lymphocytes induced by measles virus. Clinical Immunology and Immunopathology 83, 281–286. Jaye, A., Magnusen, A.F., Sadiq, A.D., Corrah, T. and Whittle, H.C. (1998) Ex vivo analysis of cytotoxic T lymphocytes to measles antigens during infection and after vaccination in Gambian children. Journal of Clinical Investigation 102, 1969–1977. Jeevanandam, M., Shahbazian, L.M. and Petersen, S.R. (1999) Proinflammatory cytokine production by mitogen-stimulated peripheral blood mononuclear cells (PBMCs) in trauma patients fed immune-enhancing enteral diets. Nutrition 15, 842–847. Keusch, G.T. (1993) Malnutrition and the thymus gland. In: Cunningham-Rundles, S. (ed.) Nutrient Modulation of Immune Response. Marcel Dekker, New York, pp. 283–299. Kramer, T.R. and Burri, B.J. (1997) Modulated mitogenic proliferative responsiveness of lymphocytes in whole-blood cultures after a low-carotene diet and mixed-carotenoid supplementation in women. American Journal of Clinical Nutrition 65, 871–875. Kudoh, K., Shimizu, J., Wada, M., Takita, T., Kanke, Y. and Innami, S. (1998) Effect of indigestible saccharides on B lymphocyte response of intestinal mucosa and cecal fermentation in rats. Journal of Nutritional Science and Vitaminology 44, 103–112. Kudsk, K.A., Wu, Y., Fukatsu, K., Zarzaur, B.L., Johnson, C.D., Wang, R. and Hanna, M.K. (2000) Glutamine-enriched total parenteral nutrition maintains intestinal inter- leukin-4 and mucosal immunoglobulin A levels. Journal of Parenteral and Enteral Nutrition 24, 270–274. Labeta, M.O., Vidal, K., Nores, J.E., Arias, M., Vita, N., Morgan, B.P., Guillemot, J.C., Loyaux, D., Ferrara, P., Schmid, D., Affolter, M., Borysiewicz, L.K., Donnet- Hughes, A. and Schiffrin, E.J. (2000) Innate recognition of bacteria in human milk is mediated by a milk-derived highly expressed pattern recognition receptor, soluble CD14. Journal of Experimental Medicine 191, 1807–1812. Lesourd, B.M. (1997) Nutrition and immunity in the elderly: modification of immune responses with nutritional treatments. American Journal of Clinical Nutrition 66, 478S-484S. Levy, J.E., Montross, L.K. and Andrews, N.C. (2000) Genes that modify the hemochro- matosis phenotype in mice. Journal of Clinical Investigation 105, 1209–1216. Lin, E., Kotani, J.G. and Lowry, S.F. (1998) Nutritional modulation of immunity and the inflammatory response. Nutrition 14, 545–550. Miles, E.A., Wallace, F.A. and Calder, P.C. (2000) Dietary fish oil reduces intercellular adhesion molecule 1 and scavenger receptor expression on murine macrophages. Atherosclerosis 152, 43–50. Miles, E.A., Thies, F., Wallace, F.A., Powell, J.R., Hurst, T.L., Newsholme, E.A. and Calder, P.C. (2001) Influence of age and dietary fish oil on plasma soluble adhesion molecule concentrations. Clinical Science 100, 91–100. Mishra, O.P., Agrawal, S., Ali, Z. and Usha (1998) Adenosine deaminase activity in pro- tein–energy malnutrition. Acta Paediatrica 87, 1116–1119. Muga, S.J. and Grider, A. (1999) Partial characterization of a human zinc-deficiency syndrome by differential display. Biological Trace Element Research 68, 1–12. Nieman, D.C., Nehlsen-Cannarella, S.I., Henson, D.A., Butterworth, D.E., Fagoaga, O.R., Warren, B.J. and Rainwater, M.K. (1996) Immune response to obesity and moderate weight loss. International Journal of Obesity and Related Metabolic Disorders 20, 353–360. Niyongabo, T., Bouchaud, O., Henzel, D., Melchior, J.C., Samb, B., Dazza, M.C., Ruggeri, C., Begue, J.C., Coulaud, J.P. and Larouze, B. (1997) Nutritional status of
Nutrients and Immune Function 37
Wallace, F.A., Miles, E.A., Evans, C., Stock, T.E., Yaqoob, P. and Calder, P.C. (2001) Dietary fatty acids influence the production of Th1- but not Th2-type cytokines. Journal of Leukocyte Biology 69, 449–457. West, K.P., Jr, Katz, J., Khatry, S.K., Le Clerq, S.C., Pradhan, E.K., Shrestha, S.R., Connor, P.B., Dali, S.M., Christian, P., Pokhrel, R.P. and Sommer, A. (1999) Double blind, cluster randomised trial of low dose supplementation with vitamin A or beta carotene on mortality related to pregnancy in Nepal. The NNIPS-2 Study Group. British Medical Journal 318, 570–575. Yaqoob, P., Newsholme, E.A. and Calder, P.C. (1999) Comparison of cytokine produc- tion in cultures of whole human blood and purified mononuclear cells. Cytokine 11, 600–605. Zhang, Y.H., Kramer, T.R., Taylor, P.R., Li, J.Y., Blot, W.J., Brown, C.C., Guo, W., Dawsey, S.M. and Li, B. (1995) Possible immunologic involvement of antioxidants in cancer prevention. American Journal of Clinical Nutrition 62, 1477S-1482S.
Nutrients and Immune Function 39
