Nutritional Modulation of Immunity

7. Nutritional Modulation of Immunity

Polyunsaturated Fatty Acids

Dietary polyunsaturated fatty acids (PUFA) can modulate immune responses through several mechanisms (Figure 13).

Figure 13. Mechanisms for the modulation of immunity by dietary polyunsaturated fatty acids.

Eicosanoid Production

The dietary content of polyunsaturated fatty acids determines the proportions of the 20 carbon n-6 fatty acids arachidonic acid (ARA), dihomo-g-linolenic acid (DGLA), and the n-3 fatty acid eicosapentaenoic acid (EPA) within the phospholipids cell membranes of leukocytes and other cell types. When ARA is used as the substrate, 2-series prostaglandins and thromboxane (e.g., PGE2, and TXA2), and 4-series leukotrienes (e.g., LTB4) are produced. Those derived from EPA are the 3-series prostaglandins and thromboxane (e.g., PGE3, and TXA3), and the 5-series leukotrienes (e.g., LTB5; Figure 6). EPA and ARA are competitive substrates for cycloxogenase (COX) and lipoxogenase (LOX). EPA is a less efficient substrate for COX, resulting in reduced prostaglandin production. In contrast, EPA is the preferred substrate for LOX, and when both ARA and EPA are available, the production of 5-series leukotrienes predominate.

Feeding diets that are enriched in the n-3 PUFA EPA can reduce ARA-derived eicosanoids by up to 75%. The conversion of the 18-carbon alpha linolenic acid (a-LA) into EPA does not occur to any significant degree in cats. Therefore the effect of enriching a diet in a-LA will likely have little effect on immunity in cats.

The EPA-derived thromboxane TXA3 is a much less potent platelet aggregate and vasoconstrictor than TXA2. In contrast, the efficacy of the prostacyclins PGI2 and PGI3 in inducing vasodilation and inhibiting platelet aggregation are equal. Thus diets enriched in n-3 PUFA will reduce thrombosis and improve microcirculation at sites of endothelial activation.

The EPA-derived leukotriene LTB5 is a much less potent vasoconstrictor and neutrophil chemotaxin than the ARA-derived LTB4. Similarly, PGE3 is less biologically active than PGE2, and is less effective in inducing fever, increasing vascular permeability, and vasodilation. However, PGE2 and PGE3 are similarly effective in decreasing Th1 cytokine production and shifting the Th1 - Th2 balance in favor of a Th2 response in human lymphocytes (Dooper et al., 2002).

Dietary EPA will therefore result in the production of eicosanoids that range from antagonistic to equipotent to those derived from ARA, and the overall effect of PUFA on immunity is not explained simply by the reduced efficacy of EPA-derived eicosanoids.

At present then, the effects and mechanisms of modulation of eicosanoids by dietary lipid is complex, and is very poorly described in cats, although there is some value to the generalization that diets enriched in n-3 PUFA will have an anti-inflammatory effect relative to diets enriched in n-6 PUFA. It is also not even known how significant alterations in eicosanoid production are in the modulation of immunity by n-3 PUFA, and it may be that other mechanisms are as, or even more important.

Gene Transcription

PUFA can directly affect gene transcription by interacting with nuclear receptors. The peroxisome proliferators-activated receptors (PPARs) are a family of cytosolic proteins that, once bound to an appropriate ligand, diffuse into the nucleus and either promote or inhibit gene transcription. PPARs are expressed by macrophages, T cells, B cells, dendritic cells, endothelial cells and other cell types (Glass & Ogawa, 2006). Both EPA and DHA are ligands for PPAR-α, PPAR-γ, and PPAR-δ (Kliewer et al., 1997). PPAR-α agonists have been shown to inhibit TNF-α, IL-6 and IL-1 production as well as inhibiting iNOS, matrix metalloprotease-9 and scavenger receptor expression by activated macrophages (Kostadinova et al., 2005). In T-lymphocytes, PPAR-α agonists inhibit IL-2 production and hence indirectly depress lymphocyte proliferation (Glass & Ogawa, 2006).

Long chain n-3 PUFAs reduce expression of COX-2, 5-LOX and 5-LOX activating protein in chondrocytes. Thus, PUFAs alter eicosanoid synthesis at the level of gene expression, as well as by providing the substrates from which they are produced.

Membrane Structure (Figure 14)

Incorporation of EPA in place of ARA in phospholipid membranes alters the physical and structural properties of the cell membranes in lymphocytes. In particular, the assembly of lipid rafts, within which most cell surface receptors are localized, is altered. In T-lymphocytes in vitro, this has the effect of decreasing signal transduction through the T cell receptor and thus depresses T-lymphocyte activation (Geyeregger et al., 2005).

Figure 14. Membrane structure.

Inhibition of LPS-signaling

Animals fed diets enriched in EPA and/or DHA produce decreased amounts of inflammatory cytokines, and experience less morbidity and mortality following challenge with gram negative sepsis or lipopolysaccharide. In addition, lipid emulsions administered to human patients with systemic sepsis results in reduced systemic inflammatory responses as a result of suppressing production of TNF-α, IL-1, IL-6, and IL-8 by LPS-stimulated macrophages (Mayer et al., 2003). DHA and EPA, inhibit TLR4 agonist (LPS)-induced up-regulation of the costimulatory molecules, MHC class II, COX-2 induction, and cytokine production through suppression of NF-κB activation. In contrast, COX-2 expression by TLR2 or TLR4 agonists was increased by the saturated fatty acid, lauric acid (Lee et al., 2004; Weatherill et al., 2005).

Dietary PUFA Content, Supplementation, and Ratios

The complexity of eicosanoid production and effects is added to by the complexity of dietary fatty acid interactions and metabolism. The prediction of an effect of a given diet has to take into account all of the following:
- Total fat content of the diet
- Relative proportions of 18-carbon n-3 and n-6 fatty acids (ALA, GLA, and LA)
- Relative proportions of 20 carbon n-3 and n-6 fatty acids (ARA, DGLA, and EPA)
- Absolute amounts of all individual n-3 and n-6 fatty acids
- Previous dietary history of the animal
- Duration of exposure to the diet in question.

The reduction of the description of the fat content of a diet to a simple ratio of n-3 to n-6 fatty acids provides very limited and potentially misleading information.

In addition, it can be seen that supplementation of a diet with a source of n-3 fatty acids (e.g., marine fish oil) will have greatly varying effects depending on the nature of the basal diet and patient. Most commercial diets are highly concentrated in linoleic acid, and the addition of a small amount of n-3 fatty acids will achieve little.

Recommendations

There is insufficient evidence to make firm recommendations for disease modulation in cats using dietary PUFA. Using a dietary fat content of approximately 70 g/kg DMB, Saker et al found that a total n-6 to n-3 ratio of 1.3:1 (using corn oil, animal fat, and menhaden fish oil) reduced platelet aggregation (Saker et al., 1998). Such a value provides a very rough estimate to the proportions required for modulating eicosanoid production, although the concentrations of EPA and ARA were not specifically assayed. In addition, the dietary concentrations required for the other effects of n-3 PUFA are unknown.

Genistein

Genistein is an isoflavone compound principally found in plants of the family Leguminosae including soy, clover, and alfalfa (Dixon & Ferreira, 2002). Genistein is structurally similar to 17 ‚ β-estradiol, as depicted in Figure 15.

Figure 15. The structural diagrams of genistein and 17 β-estradiol.

Genistein has been confirmed as a phytoestrogen in vivo through its ability to increase uterine weight, mammary gland development, and stimulation of prolactin secretion in ovariectomized rats and function as an estrogen in some estrogen-dependant cell lines (Santell et al., 1997; Morito et al., 2001). However, due to the complexity of estrogen signaling in different tissues, in differing cells, perhaps even at varying times, genistein can have estrogenic activity, no activity, or anti-estrogenic activity (Diel et al., 2001).

Tyrosine Kinase and Topoisomerase II Inhibition

In addition to genistein’s estrogenic activity is its ability to inhibit tyrosine kinases by competitively binding to their ATP-binding site and forming non-productive enzyme-substrate complexes (Akiyama et al., 1987). Inhibition of tyrosine kinases in turn inhibits numerous leukocyte signaling cascades involved in lymphocyte activation and proliferation, neutrophil activation and superoxide production, bacterial phagocytosis by macrophages, antibody responses, and delayed-type hypersensitivity responses (Trevillyan et al., 1990; Atluru et al., 1991; Atluru and Atluru, 1991; Atluru and Gudapaty, 1993; Yellayi et al., 2002; 2003). Genistein has also been found to inhibit DNA topoisomerase II, resulting in double strand breaks in DNA, and has been linked to efficacy as a cancer chemotherapeutic, and as a disrupter of lymphocyte proliferation (Markovits et al., 1989; Salti et al., 2000).

Genistein in Cat Food

Soy-based ingredients are common in commercial diets fed to cats; the soy plant provides a source of protein, fiber, and polyunsaturated oil. As a result, several commercial diets contain genistein concentrations that could be sufficient to affect immune responses in cats. The isoflavone content of several cat foods has been assayed and concentrations have been found that would result in a cat ingesting up to 8.13 mg/kg body weight (Court and Freeman, 2002; Bell et al., 2006).

Recently, it has been shown that once daily oral genistein treatment decreases circulating CD8+ cells, increases neutrophil respiratory burst, and decreases delayed-type hypersensitivity responses. Unexpected effects of genistein suggest that extrapolation from one species to other species may not be appropriate in regards to the effects of genistein on immunity.

Carotenoids

Cats are capable of absorbing dietary carotenoids, including b-carotene and lutein (Figure 16). Significant amounts of both compounds are incorporated into organelle membranes, especially in the mitochondria or lymphocytes (Chew et al, 2000; Chew and Park, 2004). It has been suggested that their efficiency in absorbing and stabilizing free radicals (Figure 17) and their ability to localize in the mitochondria combine to make them very effective antioxidants in protecting cells against endogenously derived oxidants. Their localization to organelle membranes makes them particularly effective in protecting mitochondrial proteins, lipid membranes, and DNA. In addition, since NF-κB can be activated in leukocytes in response to oxidative stress, antioxidants that concentrate in leukocytes might be expected to reduce NF-κB activation. One might question whether such effects would produce antiinflammatory or even immunosuppressive effects, or whether simple cellular preservation through the antioxidant effect might enhance immunity.

Figure 16. The structure of lutein.

Figure 17. Lutein effector sites.

In most studies performed to date, the supplementation of a diet with carotenoids with or without vitamin A activity (e.g., β-carotene vs. lutein) has produced enhanced responses in several different immunological assays (Chew and Park 2004).

The incorporation of lutein into the diet of cats has been shown to significantly affect immune responses (Kim et al., 2000). The DTH response to an intradermally administered vaccine was increased, as was the in vitro lymphocyte proliferation following activation. Finally, the total IgG response after vaccination was increased by lutein treatment (Kim et al., 2000). Overall then, carotenoids seem to enhance immunity independent of their vitamin-A activity. Whether this effect is solely, or even partially due to their ability to function as antioxidants is unsolved.

Arginine

Arginine is an essential amino acid for cats because of their inability to synthesize sufficient quantities in the fasting state. However, beyond its role as an essential intermediate in the ornithine cycle, dietary arginine has long been known to enhance certain aspects of immunity.

L-Arginine is oxidized to L-citrulline + •NO by nitric oxide synthetase (Figure 18). The inducible form within leukocytes (iNOS) produces much greater amounts of •NO than the constitutive endothelial (eNOS) or neuronal (nNOS) forms. The production of •NO after induction of iNOS in an activated phagocyte is limited mostly by the availability of free arginine. Therefore any increase in available arginine will increase the •NO produced by any given inflammatory stimulus (Eiserich et al., 1998).

Figure 18. Origin of nitric oxide.

Nitric oxide is a free radical. However, compared with other free radical species, in physiological conditions the molecule is relatively stable, reacting only with oxygen and its radical derivatives, transition metals, and other radicals. This low reactivity, combined with its lipophilicity, allows the molecule to diffuse away from its place of synthesis, and function as a signaling molecule on an intracellular, intercellular, and perhaps even systemic level.

Nitric oxide is required for normal intestinal epithelial maturation. It may be the principle inhibitory neurotransmitter in intestinal motility, and is essential for the maintenance of normal mucosal blood flow. In addition, •NO inhibits the expression of cellular adhesion molecules limiting unnecessary leukocyte entry, especially into the mucosal tissues. Nitric oxide inhibits T-cell proliferation, decreases NF-κB activation, and induces a Th-2 bias to local responses. However, in contrast to the paradigm that •NO inhibits the key pro-inflammatory transcription factor NF-κB, some studies have suggested that iNOS inhibition can increase pro-inflammatory cytokine production.

As mentioned, •NO is relatively unreactive with non-radical molecules. However, reaction with superoxide (O2•-) to form peroxynitrite (ONOO-) is diffusion limited. Peroxynitrite is not a free radical, though it is a powerful oxidant, and has been shown to elicit a wide array of toxic effects ranging from lipid peroxidation, protein oxidation and nitration leading to inactivation of enzymes and ion channels, DNA damage, and inhibition of mitochondrial respiration (Virag et al., 2003). The cellular effect of ONOO- oxidation is concentration dependant; for instance very low concentrations will be handled by protein and lipid turnover and DNA repair, higher concentrations induce apoptosis, whereas very high concentrations induce necrosis. Since both •NO and O2•- are produced in sites of inflammation, it is reasonable to propose that ONOO- might be involved in the pathogenesis of many cases.

In light of differences in the radius of effect of both O2•-and NO, co-localization of both molecules within the same cell would be expected to lead to disease. In this context, the finding that iNOS is capable of generating O2•- in conditions when L-arginine is unavailable is significant. This has been demonstrated in macrophages, where limiting L-arginine availability resulted in the simultaneous production of functionally significant amounts of O2•- and NO, and the immediate intracellular formation of ONOO- (Xia and Zweier 1997).

The large number of conflicting studies evaluating the role of •NO in inflammatory disease, has resulted in a polarization of view points between those that argue •NO is protective, and those that argue it contributes to the pathogenesis. This is unfortunate since both views are probably correct. The fate of any individual molecule of •NO is determined by multiple variables that determine its role in pathogenesis including:
- Site of production
- Timing within the local disease process that the molecule is produced
- Amount of •NO produced
- Redox status of the immediate environment
- Chronicity of the disease.

Overall it appears that supplemental arginine, either parenterally or orally administered, enhances the depressed immune response of individuals suffering from trauma, surgery, malnutrition, or infection. This action is presumably through its ability to augment the production of •NO by iNOS in activated neutrophils and macrophages.

However, in cases of severe sepsis (i.e. infection accompanied by a systemic inflammatory response), augmentation of •NO production might be detrimental because of its effect as a negative cardiac inoand chronotrope, its ability to inhibit coagulation and its potent venous and arterial dilator effects (Suchner et al., 2002).

Most commercial enteral nutritional formulas suitable for feeding to cats contain 1.5 to 2 times the minimum requirement of arginine for growth. However, supplementation of diets for intensive care nutrition has frequently been recommended, and is widely used in human medicine for enhancement of the immune system in critical care. Although clinical improvements in some studies have been reported, critically ill patients with SIRS, sepsis, or organ failure may actually deteriorate as the result of arginine supplementation (Stechmiller et al., 2004). Thus there may be cases where supplementation with arginine, beyond that provided by a conventional protein source may be beneficial, whilst in other cases it may be detrimental.

Lysine

As indicated in Figure 2, the diet ingested by the host can directly affect the pathogen. The interaction between lysine and herpes viruses is an example of such an interaction, rather than an interaction between diet and immunity (Figure 19).

Figure 19. Interest of L-lysine supplementation in herpes virus infection.

The genome of the feline herpes virus (FHV-1) is similar to the genomes of other alpha-herpes viruses, and several different viral proteins have been described (Mijnes et al., 1996). All of the 20 common amino acids are utilized, including L-lysine (Pellett et al., 1985). However, when herpes viruses are grown in cell cultures in vitro, there is no requirement for lysine to be added to the culture media, thus what little lysine is required for viral replication is derived from the labile pool of free intracellular amino acids (Maggs et al., 2000). In contrast, the omission of arginine or histidine from the culture media profoundly inhibits viral replication more so that other amino acid omissions (Tankersley, 1964). The addition of lysine to the culture media actually inhibits viral replication, but the breakpoint at which replication is inhibited has not be clearly defined. Tankersley (1964) demonstrated that normal replication occurs at 70 μg/mL, but profound inhibition occurs at 180 μg/mL. It is worth noting that, when cats are fed a diet containing the adequate intake of lysine recommended by the NRC for gestating queens (i.e., 1.1% in a 4000 kcal/kg diet), plasma concentrations of lysine are 14 ± 2.2 μg /mL (Fascetti et al., 2004).

It has been proposed that lysine may antagonize arginine availability in vitro by competing for arginine entry into cells (Figure 19) and in vivo by both competition, and by the induction of renal and hepatic arginase. However, Fascetti et al., have demonstrated that even very large dietary concentrations of lysine do not alter plasma arginine concentrations in cats within a two week period (Fascetti et al., 2004).

- When 500 mg of L-lysine monochloride was given twice daily to cats as a bolus, starting 6 hours prior to inoculation with FHV-1, a mean plasma concentration of 97 μg/mL was achieved. Clinical signs associated with acute FHV-1 infection were reduced, but there was no reduction in viral shedding (Stiles et al., 2002).

- A once daily bolus of 400 mg given to latently infected cats produced mean peak plasma concentrations of 65 μg / mL and reduced viral shedding, but had no significant effect on clinical signs (Maggs et al., 2003).

- Finally, when a diet containing 5.1% lysine (as fed, in a 4000 kcal/kg diet) was fed to cats, a mean plasma concentration of 44 μg/mL was achieved. This diet was fed to groups of spontaneously latently infected cats, having recently experienced an epizootic upper respiratory disease, no effect was seen on clinical signs or herpes viral shedding (Maggs et al., 2006). In fact, one group (the male cats) fed the lysine supplemented diet experienced worse clinical signs than any other, supplemented or not, and increased viral shedding. This observation was probably more due to stress or another pathogen (Mycoplasma felis, Bordetella bronchiseptica) than an effect of the diet, but did influence the results of the study.

Thus the efficacy of treatment with L-lysine on feline herpes viral infection remains to be challenged in cats with enzootic upper respiratory disease. To date, lysine supplementation has not show any toxic effects. Experimentally, the cat’s dietary consumption is lower with a 13% lysine-diet (as fed, in a 4000 kcal/kg diet) but this level largely exceeds practical cat food formulation (Fascetti et al., 2004).