Using Immune Cell/Adipocyte Co-Culture Models to Identify Inflammatory Paracrine Signaling Mechanisms: A Process Attenuated by Long-Chain N-3 Polyunsaturated Fatty Acids

This invited Commentary is on the methods paper entitled “Studying adipocyte and immune cell cross talk using a co-culture system” in Immunometabolism: Methods and Protocols [1]. Co-culturing individual immune cell populations (as primary cells or cell lines) with adipocytes represents a model system to study the paracrine interactions (or cross-talk) between cell types that can impact adipose tissue (AT) function. This is particularly relevant in obese AT, wherein paracrine interactions between cell types promotes the secretion of inflammatory mediators that contribute to increased local (i.e. within the AT) and systemic low-grade inflammation and metabolic dysfunction, including insulin resistance (IR) [2-5].


Introduction
This invited Commentary is on the methods paper entitled "Studying adipocyte and immune cell cross talk using a co-culture system" in Immunometabolism: Methods and Protocols [1]. Co-culturing individual immune cell populations (as primary cells or cell lines) with adipocytes represents a model system to study the paracrine interactions (or cross-talk) between cell types that can impact adipose tissue (AT) function. This is particularly relevant in obese AT, wherein paracrine interactions between cell types promotes the secretion of inflammatory mediators that contribute to increased local (i.e. within the AT) and systemic low-grade inflammation and metabolic dysfunction, including insulin resistance (IR) [2][3][4][5].
Our research group [26][27][28][29][30][31][32] and others [33][34][35][36] have demonstrated the anti-inflammatory mechanisms through which long-chain (LC) omega-3 (n-3) polyunsaturated fatty acids (PUFA), eicosapentaenoic acid (20:5:n-3, EPA) and docosahexaenoic acid (22:6n-3, DHA), serve as an intervention strategy to improve obesity-associated AT inflammation and metabolic dysfunction. As such, high fat (HF) diet supplementation with fish oil derived LC n-3 PUFA can attenuate the severity of AT and systemic inflammation and associated metabolic dysfunction [26], which can be attributed, at least in part, to the paracrine interactions (or cross-talk) between immune cell subsets and adipocytes, as reviewed elsewhere [2]. Since research on adipocyteimmune cell cross-talk has centred on macrophages and T cell subsets, these immune cell populations are the focus of this commentary. We will discuss the development of the adipocyte-immune cell co-culture models and highlight our research findings demonstrating the ability of n-3 PUFA to mitigate paracrine signaling between co-cultured adipocytes and T cells (CD8 + and CD4 + ) or macrophages with an emphasis on the secretory profile (inflammatory and chemotactic mediators), however, it is worth noting that we have also shown a beneficial impact of n-3 PUFA on the NLRP3 inflammasome and/or macrophage M1/M2 polarization status in these models [29,31,32,37].
To further increase the translational relevance of the coculture model, we have utilized a physiologically relevant cellular ratio of the immune cell population co-cultured with adipocytes to recapitulate their abundance within obese AT, which is in contrast to co-culture models utilizing equal numbers of adipocytes and macrophages (1:1 cellular ratio) [43][44][45][46][47]. In our model, macrophages are co-cultured with adipocytes at 17% of cells in culture [28,38], which is reflective of the level of macrophage cellular infiltration in epididymal AT of db/db mice and recapitulates the in vivo cellular ratio of macrophages:adipocytes [49]. Similarly, T cells are co-cultured with adipocytes to recapitulate the murine obese AT cellular ratio of adipocytes to T cells [9], with co-cultures comprised of 10% CD8 + T cells [31,37,39] and 5% CD4 + T cells [29,32].
Our research group has shown that cellular co-cultures can be established for 12 hr [38], 24 hr [28,31,32,37,39] or 48 hr [29] and can be utilized to discern the difference in outcomes resultant from i) direct cell contact (i.e., contactdependent, Figure 1A and 1B), a combined outcome of both physical cellular interactions and paracrine signaling mechanisms, and ii) trans-well in direct cell contact (i.e., contact-independent, Figure 1C), wherein cells are separated by a semi-permeable 0.4 μM polyester membrane trans-well insert that precludes physical contact between cell types but permits the movement of To highlight a variation of the adipocyte/immune cell co-culture models, our research group has also utilized conditioned media (contact-independent) generated from i) one cultured cell line to another (Figure 2A), ii) adipocyte/CD8 + T cell co-culture conditioned media added to RAW264.7 macrophage cultures ( Figure 2B) [31,37,39], and iii) intact primary AT conditioned media added to adipocyte/CD4 + T cell co-cultures [32] or macrophages [27] ( Figure 2C).

CD8 + T cell and Adipocyte Co-Culture
CD8 + T cells have been shown to accumulate in obese AT prior to the accumulation of macrophages [9,50,51] and co-localize to crown-like structures in the AT [9]. In vivo depletion of CD8 + T cells via neutralizing antibody injections reduces the expression of macrophage chemotactic signals in AT and cellular trafficking, thereby reducing the magnitude of inflammation and systemic IR [9]. Similarly, the severity of the obese phenotype in CD8a -/mice is attenuated but can be reversed through adoptive transfer of CD8 + T cells resulting in inflammatory mediator production and IR [9], thus, demonstrating the essential role of this cell type in the development and maintenance of obesity-associated AT dysfunction.
To highlight the utility of our co-culture models to study the paracrine signaling between adipocytes and immune cells, our initial contact-dependent co-culture studies used 3T3-L1 mature adipocytes cultured with purified splenic CD8 + T cells isolated from lean mice fed either a fish oil [37] or flaxseed oil [39] supplemented diet as the source of either marine-or plant-derived n-3 PUFA, respectively. Despite the difference in marine versus plant sources of n-3 PUFA that enriched the CD8 + T cells, in contact-dependent co-culture with adipocytes stimulated with LPS, the resultant anti-inflammatory and anti-chemotactic effect was consistent and characterized by reduced TNFα, IL-6, MCP-1, MCP-3 and MIP-1β secreted protein into the culture media compared to control [37,39]. Subsequently, the CD8 + T cell/adipocyte co-culture conditioned media was used in a follow-up experiment with RAW264.7 macrophages in a chemotaxis assay to demonstrate that the number of macrophages migrating towards the chemotactic signals in the conditioned media was reduced compared to control [37,39], indicating a functional outcome of reduced macrophage trafficking as a result of n-3 PUFA attenuating the paracrine interactions between CD8 + T cells and adipocytes. The anti-inflammatory and anti-macrophage chemotactic secretory profile of n-3 PUFA-enriched CD8 + T cells/adipocyte co-cultures was later confirmed in both the cell-contact dependent and cell contact-independent (trans-well) co-culture models, which was mechanistically attributed, in part, to the effects of TNFα [31]. The anti-inflammatory and anti-chemotactic secretory profile in n-3 PUFA co-cultures could be reproduced in control n-6 PUFA-enriched CD8 + T cell/ adipocytes co-cultures treated with a TNFα neutralizing antibody [31].

CD4 + T cell and Adipocyte Co-Culture
CD4 + T cells subsets have been shown to change in obese AT prior to the infiltration and accumulation of macrophages [14], wherein IFNγ-secreting Th1 cells have been shown to promote macrophage M1 polarization and contribute to AT metabolic dysfunction [14,18,50,52]. Additionally, IL-17-secreting Th17 cells have also been shown to contribute to obese AT inflammation [53].
In our co-culture model (using a physiologically relevant cellular ratio and LPS concentration) adipocytes were co-cultured in direct cell contact with primary splenicderived purified CD4 + T cells from lean mice fed isocaloric diets enriched with either n-3 or n-6 PUFA. Using this approach, n-3 PUFA increased mRNA expression and/ or secreted protein of Th2 polarization markers (GATA3, IL-4) and reduced expression of Th1 polarization markers (Tbet, IFNγ), in addition to reducing the secretion of other inflammatory and macrophage chemotactic mediators (IL-1β, IL-6, MCP-1, MCP-3 and MIP-1α) [32]. These effects were reproduced in contact-dependent co-cultures containing adipocytes and splenic CD4 + T cells from HF diet-induced obese mice consuming either an n-3 or n-6 PUFA-supplemented isocaloric diet (containing equal amounts of lard/saturated fatty acids) [29]. Collectively, our findings demonstrate the ability of adipocyte/CD4 + T cell cross-talk to influence the local inflammatory and chemotactic microenvironment, which can impact subsequent macrophage chemotaxis and ultimately contribute to AT metabolic dysfunction. Future studies utilizing contact-independent conditioned media and the trans-well (cell contact-independent) models can help discern the inflammatory and chemotactic mediator cellular source to mechanistically identify the contribution of each cell type in co-culture to help direct targeted interventions to improve AT function.

Macrophages and Adipocyte Co-Culture
As obesity progresses, the infiltration of macrophages into the AT increases from 3% to approximately 20% of total non-adipocyte cells [54], wherein they exhibit an inflammatory M1 phenotype and increase inflammatory mediator production contributing substantially to obesityassociated inflammation and IR [2, 9,12,14,55]. We [28,38] and others [43][44][45][46][47] have used 3T3-L1 adipocyte and RAW264.7 macrophage co-culture models [i.e., in cell contact-dependent and cell contact-independent (trans-well)] to study the paracrine signaling mechanisms that underlie AT dysfunction using different dietary bioactive interventions. Our initial co-culture work utilizing cell lines (3T3-L1 adipocytes and RAW264.7 macrophages) demonstrated that n-3 PUFA (EPA and DHA) could attenuate the intensity of the inflammatory cross-talk between cell types (e.g. reduced secretion of IL-6 and MCP-1) in both the cell contact-dependent and cell contact-independent (i.e., trans-well) co-culture models, which was associated with increased expression of non-inflammatory M2 macrophage markers [38]. Co-culture of 3T3-L1 adipocytes with primary splenic CD11b + macrophages from obese mice consuming a HF diet supplemented with or without n-3 PUFA confirmed the attenuated inflammatory mediator secretory profile and reduced expression of M1 macrophage markers in n-3 PUFA cultures [28]. This outcome was partially attributable to the effects of adiponectin [28], wherein n-3 PUFA have been shown to stimulate adiponectin secretion from adipocytes [56]. Finally, using a conditioned media contact-independent model in which RAW264.7 macrophage cultures were treated with conditioned media derived from intact primary AT from n-3 PUFA-fed mice, we demonstrated that macrophage-derived inflammatory and chemotactic mediator secretion (IL-6, MCP-1, MCP-3 and RANTES) and M1 macrophage markers were reduced, again in part, through an adiponectin-dependent mechanism [27].

Skeletal Muscle and Macrophage Co-Culture
Moving beyond co-culture studies focused on AT in obesity is relevant given that other tissues also contribute to systemic inflammation and whole-body IR in obesity, in part, through inflammatory immune cell-tissue crosstalk, thereby, providing other targets for the systemic impacts of n-3 PUFA on the obese phenotype [26]. Skeletal muscle is the primary site for insulin-simulated glucose uptake; however, in obesity, increased circulating free fatty acids and inflammatory cytokines interfere with insulin signaling in skeletal muscle to promote development of whole-body IR [57][58][59]. Similar to AT, obese skeletal muscle is characterized by the infiltration and accumulation of immune cells, particularly M1 macrophages, which contribute to local inflammation and IR through immune cell-myocyte cross-talk [60][61][62][63]. Therefore, our research group [64][65][66] and others [67][68][69], have utilized a contact-independent conditioned media coculture model to examine the inflammatory macrophagemyocyte cross-talk that contributes to obese metabolic dysfunction. Importantly, attenuating macrophage-muscle inflammatory cross-talk represents another potential target for n-3 PUFA to improve obesity-associated insulin sensitivity. In this connection, RAW264.7 macrophages were stimulated with fatty acids [DHA (n-3 PUFA) versus palmitic acid (saturated fatty acid control)] and a physiologically relevant LPS dose (described above) to generate macrophage conditioned media (MCM), which contained secreted cytokines and chemokines that could impact muscle cell function in a cell contact-independent  [64]. A later co-culture study using MCM generated from RAW264.7 macrophages and L6 myotubes demonstrated that n-3 PUFA-mediated attenuation of inflammatory macrophage-myocyte crosstalk is attributable, in part, to a PPAR-γ-dependent mechanism [65].
Most recently, to increase the translational relevance of the macrophage-myocyte co-culture model to recapitulate more accurately the obese skeletal muscle microenvironment, rats were fed a HF diet enriched with either n-3 or n-6 PUFA and primary purified splenic CD11b + macrophages were isolated and co-cultured in direct contact with L6 myotubes stimulated with LPS, wherein n-6 PUFA increased inflammatory cytokine production compared to n-3 PUFA co-cultures [66]. Subsequently, in a contact-independent experiment to determine the response of macrophages to mediators secreted from myocytes, purified CD11b + cells from obese rats (consuming n-3 PUFA and n-6 PUFA-enriched HF diets) were cultured alone in conditioned media collected from LPS-stimulated L6 myocytes. This resulted in n-3 PUFA-enriched macrophages reducing expression of inflammatory cytokines and M1 polarization markers compared to n-6 PUFA [66]. Collectively, skeletal muscle cell (myotube)/macrophage co-culture models represent a relevant future direction to elucidate the underlying mechanisms contributing to obesity-associated IR.

Conclusion
Through the use of appropriately crafted adipocyte/ immune cell (CD8 + and CD4 + T cell, macrophage) coculture models [1], the critical features of obese AT can be recapitulated by including physiologically relevant cellular ratios of adipocytes to immune cell populations [9,49] and LPS stimulation conditions [40][41][42]. Using this approach, we have shown that n-3 PUFA can attenuate the severity of the inflammatory and chemotactic paracrine interactions between adipocytes and macrophages [27,28], CD8 + T cells [31,37,39] and CD4 + T cells [29,32], which collectively contribute to obese AT dysfunction [2]. Moreover, we have expanded this co-culture model to study paracrine interactions in another metabolically active tissue, namely skeletal muscle/immune cell cross-talk [64][65][66] to better understand the contribution of immune cells in various tissues towards the severity of the obese phenotype and identify immune-centric intervention strategies to improve obesity-associated metabolic dysfunction.

Conflicts of Interest
The authors state that there are no conflicts of interest.