Food components play a role in influencing, either directly or indirectly (through hormonal regulation), the expression of genes encoding for proteins involved in energy metabolism, cell differentiation and growth and immune responses. More specifically, diet exerts diverse effects on the development of the immune system, even at the level of gene regulation 1 . It is known that polyunsaturated fatty acids (PUFAs) can modulate the expression of a variety of genes encoding for cytokines, adhesion molecules, and inflammatory proteins 2 3 . This fact seems to be very important during early life, since docosahexanoic and arachidonic acids were reported to participate in the development of the neonate immune system, although their proportion among total fatty acids in human breast milk is very low 4 . Human milk contains PUFAs such as conjugated linoleic acid (CLA), among others, that seem to contribute to immune development 5 6 7 8 .

CLA is a class of positional and geometric conjugated dienoic isomers of linoleic acid, among which, cis9,trans11 (c9,t11) predominate, accounting for 83% to 100% of total CLA present in milk 5 6 7 , whereas trans10,cis12 (t10,c12) exist in lower proportion 9 10 . However, it is the intake of food of ruminant origin which determines the total concentration of CLA in dam's milk 7 . Many other beneficial physiological effects have also been attributed to CLA, including reduced body fat, and inhibition of carcinogenesis, atherosclerosis, and diabetes 11 12 13 .

Existing data regarding the effects of CLA on the immune system show great variability, mainly due to differences in the animal species used, the length of the supplementation period, and the differences in the isomer mixtures used in the experimental approach. In this direction, recent studies in suckling animals showed that the immune function is enhanced after feeding with an 80:20 isomer mix of c9,t11 and t10,c12 CLA 14 15 . Specifically, sera IgG concentration and IgM in vitro production by splenocytes are increased after CLA supplementation during suckling. CLA downmodulatory effects on lymphoproliferation were only observed after an extra week of diet 16 . The immune effects of CLA were also described in adult rats receiving this mixture since pregnancy 17 . However, little work has been done on the effects of CLA on gene expression, and even less regarding the development of the immune system in early life.

Nutritional genomics is a result of the genetic revolution experienced over the past 10 years. Nutrigenomics deals with the interactions between dietary components and the genome and the resulting changes in proteins and other metabolites. On the other hand, nutrigenetics aims to understand the gene-based differences in response to dietary components and to develop nutraceuticals that are the most compatible with the health status of individuals based on their genetic makeup 18 . The number of successful examples of transcriptome, proteome, and metabolome profiling as tools for evaluating the cellular responses to nutrients and identifying their molecular targets, has grown significantly. The use of high-density microarrays is a useful approach to estimate correlations among genes, which in turn can become the basis of transcriptional networks. The availability of microarrays for a number of model systems allows the quantification of relative transcript abundances in a comprehensive fashion. Despite the relatively simplistic nature of correlation measurements, they reflect an integrative view of gene-gene interactions in any given system, pointing out general structure characteristics of transcriptional interaction networks 19 .

Additionally, nutrigenomic approaches have been undertaken to get further insight on the molecular understanding of mechanisms triggered by nutritional interventions. Diets enriched in different long-chain polyunsaturated fatty acids (LC-PUFAs) have been tested in rat nutritional intervention models. One report revealed steaoryl-CoA desaturase as an enzyme target for an arachidonate-enriched diet 20 . In another study, Berger et al. 21 looked at transcriptional effects of these LC-PUFA-enriched diets on murine hepatic and hippocampal gene expression. Additionally, the beneficial effect of LC-PUFAs was assessed by a nutrigenomics experiment designed to understand the mechanisms by which these lipids induce and control gene signalling involved in carcinogenesis 22 .

With all these concepts in mind and based on previous studies, we hypothesized that CLA intake during developmental phases would exert some influencing effect, among others, on genes involved in the regulation of the immune system. The aim of our study was the evaluation, by using whole genome microarrays, of the effects of dietary supplementation with an 80:20 isomer mix of c9,t11 and t10,c12 CLA, on mesenteric lymph nodes (MLN) gene expression, during gestation and/or suckling. The list of common genes differentially expressed in the three dietary interventions was used to construct a Biological Association Network (BAN). This approach allowed us to obtain a global view of gene expression in MLN, formed by a collection of nodes with different degrees of interrelationship, that could be used to explain the molecular mechanisms triggered by CLA.

In this work, we analyzed the gene expression profile of MLN cells from suckling rats receiving an 80:20 c9,t11:t10,c12 CLA mix during early life. This particular mixture enriched with the c9,t11 isomer was used due to its resemblance with that in human breast milk 5 . Previous studies have demonstrated that this mix modulates immune development in suckling rats receiving it during gestation/suckling 15 . The main effect of this type of diet was found regarding immunoglobulin production at systemic and intestinal level 14 15 . However, the precise mechanism of action of this PUFA was not established. Thus, we used microarrays to get further insight in the effects of CLA supplementation on MLN gene expression. A large number of genes were modulated when CLA was administered during gestation and suckling (groups A and B, 5 wk). However, the use of Venn Diagrams evidenced that many genes were differentially expressed in both groups B and C, and less common genes were found between group A and the others. This behavior could be due to the different nutritional intervention that was performed. In the group A, CLA was added to the diet of the dams during suckling period and therefore CLA was incorporated in breast milk and transferred to the suckling pups. However, in groups B and C, CLA was directly administered by oral gavage to the suckling pups. This fact influences the proportion between the two CLA isomers that get to pups, as the ratio of c9,t11 CLA/t10,c12 CLA in pups's plasma was 83:14 for group A, but 93:7 for group B and 94:6 for group C as previously described 15 .

The list of differentially expressed genes in the three conditions was used to generate a BAN. Several node genes, namely Ctgf, Timp1, Gal, Syt1, Actg2 and Acta2, were selected from this BAN, and changes in their mRNA expression were confirmed by RT-Real time PCR. Some of these node-genes may explain the behavior of CLA as immune modulator during early age.

Connective tissue growth factor (Ctgf) is a secreted extracellular matrix-associated protein that modulates many cellular functions, including proliferation, migration, adhesion, and extracellular matrix production, and it is involved in several biological and pathological processes 24 . Dietary supplementation with CLA reduced mRNA levels of Ctgf in MLN in our three experimental approaches, indicating that downregulation of Ctgf is independent of the length and via of CLA supplementation. Ctgf is mainly regulated at the transcriptional level by bioactive lipids, as well as TGFβ and downstream mitogen-activated protein (MAP) kinase signaling 25 .

We have previously reported that young (28-day-old) and adult animals fed with the same CLA mixture showed a reduction in the proliferation activity of spleen and MLN cells 16 17 . Given that Ctgf promotes DNA synthesis in chondrocytes, osteoblasts and fibroblasts 26 , the observed decrease in proliferation could be mediated through Ctgf downregulation.

Tissue inhibitors of metalloproteinases (Timp) are multifunctional proteins including four members (Timp1-4). Timp1 is expressed and secreted by a variety of cell types and is present in most tissues 27 28 . Timp1 is able to promote cell growth, and its expression levels inversely correlate with the susceptibility to induction of apoptosis (reviewed in 29 ).

Until now, two studies have reported different effects of CLA on Timp1 expression. On one hand, 1% CLA diet in a mice model of metastatic mammary tumor increased mRNA and protein levels of Timp1 30 , whereas an in vitro study using pure CLA isomers (c9,t11 CLA and t10,c12 CLA) on a macrophage cell line did not modify Timp1 gene expression 31 . Thus, the effect of CLA on Timp1 expression may depend on the tissue microenvironment and the experimental approach used. In our study, CLA administration reduced mRNA levels for Timp1 in all three groups of animals. As Timp1 may stimulate cell proliferation by using several distinct signalling pathways, such as MAPK, the inhibitory effect of CLA on this gene may reduce lymphoproliferation. Thus, both Timp1 and Ctgf would play a role in a same signaling pathway that could explain the downmodulation of the lymphoproliferative response observed in 28-day-old animals fed with CLA from birth 16 .

The neuropeptide Galanin (Gal) was also downregulated at the mRNA level by early CLA administration, independently of length, duration and via of supplementation. This peptide exerts a wide range of effects, not only in the central nervous system 32 but also in the enteric nervous system 33 . Although Gal is mainly produced by neurons, it has been described that lymphocytes and macrophages can also synthesize it at a lower level 33 . A clear effect of neuropeptides such as somatostatin (SOM) or vasoactive intestinal peptide (VIP), including gut-associated immune system (GALT), has been already reported on the immune system 34 . When these neuropeptides are found in the intestine, they seem to affect the proliferation, differentiation and function of immune cells 33 . Thus, the effect of CLA on Gal expression could explain the immunoenhancing effects of CLA on developing GALT 17 .

There is no data available on the effects of Gal on lymphoproliferation. SOM and VIP are mainly inhibitors of lymphocyte proliferation, apparently by reducing interleukin-2 production 34 , whereas calcitonin-gene-related-peptide (CGRP), substance P (SP) or met-enkephalin stimulate lymphoproliferation 35 36 . Gal-mediated effects through its receptor Galr1 led to antiproliferative effects, whereas those through Galr2 and Galr3 were proliferative. All three receptors are expressed in immune cells 37 . When the same CLA mix was administered to young rats supplemented during suckling 16 or until adult age 17 , a downmodulatory activity on proliferation in both spleen and MLN was observed, suggesting that this effect may be mediated in part by modulation of Gal expression. Thus, CLA is able to reduce Gal promoting activity on lymphoproliferation, indicating that Gal would be acting mainly through its Galr2 and Galr3 receptors in that environment.

During early life CLA is able to enhance the production of immunoglobulins 15 and more specifically IgA in the intestine 14 . Enhancement of intestinal IgA concentration by CLA could be due to the decrease of Gal expression in MLN, as it is the case for SOM in murine spleen, Peyer patches and MLN 38 . Moreover, as neuropeptides influence cytokine production, downregulation of Gal by CLA may be the responsible effector of the changes in cytokine concentrations found in previous studies 14 15 . However, the effect on Gal found here in early life, and their parallel immune effects, may not be present at older ages, since the maturation state of lymphocytes seem to affect the response to neuropeptides and cytokines due to their differential neuropeptidergic receptor expression 33 . This fact is in agreement with the effect described for CLA on developing immune system and immunoglobulin production 15 17 , which is not observed later in life.

Synaptotagmins (Syts) constitute a family of proteins that act as Ca^2+-sensors for regulated exocytosis and endocytosis 39 40 41 . Synaptotagmin proteins were first detected in the nervous system, but their expression has also been reported in other tissues. Thus, synaptotagmins have been proposed to be widespread regulatory proteins that confer Ca^2+-sensitivity in a wide range of fusion events 39 . Indeed, Syts have been described to play a role in the maintenance of plasma membrane integrity 40 , in cell resealing 42 , in chemotactic cell migration 43 , in insulin release from the pancreatic β-cells 44 45 , in FGF-1 secretion in response to heat shock 46 and in degranulation of mast cells 47 .

CLA supplementation caused downregulation of Syt1 by about 2.3 fold in the three dietary approaches. Syt1 is mainly localized in synaptic vesicles, where it functions as a Ca^2+-sensor for exocytosis 48 . Indeed, Syt1 has been proposed to play specific roles in synaptic vesicles trafficking, in facilitating docking 49 , and in vesicle fusion 50 . Although it has been traditionally proposed to be a neuron-specific isoform, Syt1 has been demonstrated to have important roles in membrane trafficking in non-neural and non-hormonal cells 51 52 .

Syt1 was previously associated with CLA supplementation, as its expression was significantly decreased in mice receiving a diet enriched in the t10,c12 CLA isomer compared with c9,t11 CLA 52 . Syt1 downregulation was associated with the presence of hepatic fat, suggesting that changes in expression of Syt1 may indicate that other genes than the traditional are involved in lipid handling in hepatic cells, and probably in MLN. Syt1 has also been shown to play a role in neurotransmitter release 53 54 . As described above, we found the neuropeptide Gal underexpressed upon CLA supplementation. Since it has been described that Gal is secreted by enteric nerves 55 , we propose that downregulation of Syt1 upon CLA supplementation would lead to a decrease in Gal secretion, thus enhancing the negative effects that downregulation of Gal would have on lymphoproliferation.

Actins are highly conserved proteins that are involved in various types of cell motility, and maintenance of the cytoskeleton. In vertebrates, three main groups of actin isoforms, alpha, beta and gamma have been identified. The alpha actins are found in muscle tissues and are a major constituent of the contractile apparatus. The beta and gamma actins co-exist in most cell types as components of the cytoskeleton, and as mediators of internal cell motility. There are three α-actins (skeletal, cardiac, and smooth muscle), one β-actin (β-nonmuscle), and two γ-actins (γ-smooth muscle and γ-nonmuscle) 56 . The cytoskeleton constitutes a network that serves to maintain cell shape and to regulate dynamic cellular functions. It provides essential scaffolding for the localization and activation or inhibition of diverse cytoplasmic signaling molecules, as well as anchors for motor proteins that are necessary for intracellular transport and cell division 57 58 . In our microarray experiments we found that expression of smooth muscle γ-actin (Actg2), found in enteric tissues, and α2-actin (Acta2) were decreased upon dietary supplementation with CLA.

Alpha-smooth muscle actin expression is regulated by TGF-β and TNF-α. Changes in alpha-smooth muscle actin, collagen, and fibronectin expression result in decreased contraction and stiffness of collagen matrices 59 . The effect of TGF-β has been related to a conserved TGF-β control element (TCE) within the 5'-region of alpha-smooth muscle actin, also known as smooth muscle cell differentiation marker. TCE could mediate both transcriptional activation and repression in cultured smooth muscle cells through interaction with members of the zinc finger Kruppel-like transcription factor family 60 . On the other hand, it has been reported that TGF-β1 is able to significantly increase Acta2 expression in both normal and cancer associated fibroblasts cultures 61 , which are believed to promote tumor growth and progression. Overall, the downregulation of actins upon CLA supplementation would affect both the contractile apparatus and the cytoskeleton with a final effect on cell motility. However its relationship with the development of the immune system remains unclear. Deficiencies in proteins that regulate the cortical actin cytoskeleton have increasingly been associated with immunodeficiency and autoimmune or autoinflammatory disease, indicating a critical role for these regulators in immune response and tolerance (reviewed in 56 ).

Besides the modulatory action of CLA found on the above node genes, other genes with importance on the immune system were differentially modulated and would require further studies. This is the case of for the receptors for interleukin-7 (Il7r) and chemokines (i.e. Cxcr7) which are involved in mucosal immune responses and chemotaxis, respectively 62 63 .

In summary, by using rat whole genome microarrays, we determined the changes in gene expression induced by an 80:20 c9,t11:t10,c12 CLA mix in MLN from suckling rats. The generation of a biological association network allowed the identification of specific node genes that might be involved in immune responses. We conclude that Ctgf, Timp1, Gal and Syt1, among others, are gene target candidates modulated by CLA which may explain the effect of this PUFA on mucosal immune responses in early life.

The authors declare that they have no competing interests.

CC, FJPC, MR, AF and VN designed the study and supervised the experimental work. ES, CRS, FJPC, CC, AF and VN performed the experimental work. ES, CJC, and FJPC analyzed the data. FJPC, AF, ES and VN wrote the manuscript, with input from all authors. All authors read and approved the final manuscript