Macrophage polarization: the link between inflammation and related diseases
- 1.8k Downloads
In the present review, we try to critically evaluate the two faces of the macrophages and their roles in relation to gene alteration in some inflammatory conditions. The pros- and cons of each type of macrophage in immunologic outcomes are discussed.
If “Diversity is the rule of nature”, macrophages have proven to be its obedient followers. A cell type that was classically considered to be activated by Interferon-γ, under the influence of TH-1 type of response and a well-accepted warrior of cellular immunity to the intracellular pathogens is not as simple as once considered. Past decade has revolutionized this notion with the advent of TH-2 influenced alternatively activated macrophages, now established as wound repairing and tissue regenerating.
Literature survey was done to present a detailed study on this macrophage dichotomy and its relevance to immune disorders via expression of some critical genes, nuclear factor kappa-light-chain-enhancer of activated B cells, peroxisome proliferator-activated receptors and SH2-containing inositol-5′-phosphatase 1, highly implicated in a myriad of immunological emergencies like inflammation, insulin resistance, wound healing, cancer, etc.
The evaluation of macrophage dichotomy in these disorders may prove to be the first step towards the formulation of innovative therapeutic approaches.
KeywordsCytokines Inflammation Insulin resistance Macrophage polarization Obesity Tumour
Low-grade inflammation is considered to be associated with disorders like metabolic syndrome, arthritis, atherosclerosis, etc. thus becoming a focal point for researchers throughout the world. As inflammation is a characteristic of innate immune response, much attention is laid on the cells controlling innate immune response like macrophages, dendritic cells, natural killer cells, etc. and their secreted products. Macrophages, derived from myeloid precursors of bone marrow are the most important culprits of inflammation. They primarily enter the blood as monocytes, and then traverse into the tissue as tissue macrophages, where they adapt to the local tissue and perform specific functions e.g. Kupffer cells in the liver, microglial cells in brain, mesangial cells in the kidney, etc. Macrophages have a quality of transcriptional probability, in response to an inducer, which is gene-autonomous, leading to the diversity of gene expression, by expressing random combinations of thousands of inducible defence genes. It has been reported earlier that many phenotypic differences are shown by recruited macrophages diverging them from the resident tissue macrophages. Although “macrophage activation” is the term commonly used in this scenario, however, the transcriptional profiling analysis shows that these differences depend highly on the recruiting stimulus and the macrophage location. The phenotypic heterogeneity of macrophages generated in vitro underlies the wide array of phenotypes that occur in vivo. Macrophages resident in tissues adapt to their locale, so that they perform functions specific to their local microenvironment  e.g. in obese tissue which is closely associated with inflammation, against an earlier perception that they represent activated or inactivated forms of the same macrophage. Fate-mapping studies have demonstrated that the macrophages residing in various tissues are established before birth and maintain themselves throughout the adulthood without replenishment by blood monocytes, however, an inducing stimulus recruits monocytes, the descendants of whom seem to integrate into the resident macrophage environment . Fate-mapping research needs to be carried further to analyze the equivalence (if any) of the progeny of these recruited monocytes to the resident cells of solid tissues.
Macrophages have been observed to occur in the form of two strains M1 (proinflammatory) and M2 (anti-inflammatory) . In a normal tissue, the ratio of M1:M2 macrophages is highly regulated . On inflammation this ratio is changed i.e. elevated. Although M2 macrophages are further subdivided into well-defined forms like the ones induced by IL-13 and IL-4 are classified as M2a, those exposed to agonists of TLR’s and IL-1R is believed to give rise to M2b form and the third form is generated by the exposure to glucocorticoids and IL-10. However, M1 and M2 are considered to be primary lineages with respect to inflammatory innate response and the present attempt is made to throw some light on the importance of macrophage dichotomy through which they regulate the immune response and the effect of this heterogeneity on some of the common immune disorders. Hence we discuss only two extremes M1 and M2 macrophages for trying to draw a differential line between pro and anti-inflammatory innate immune response.
M1 and M2 macrophages: different strains of the same ancestry
The macrophages present in the inflamed tissue in a high concentration are known as M1 classically activated macrophages, referring to macrophages having undergone cell activation in response to LPS or IFN-gamma [9, 10]. Gene expression profile of M1 macrophages exhibit high levels of pro-inflammatory cytokines like tumour necrosis factor-α (TNF-α), monocyte chemoattractant protein-1 (MCP-1), Interleukin-6 (IL-6) and inducible nitric oxide synthase (iNOS). Macrophages present in relatively high concentration in normal are known as alternatively activated M2 macrophages characteristically expressing high levels of IL-10, YM1, macrophage and granulocyte inducer-form 1(MgI1), arginase-I to name the important ones and are actively involved in tissue remodelling and repair [11, 12]. Macrophages show their effect mostly via their respective cytokines. Apart from these genotypic differences, phenotypically, M1 classically activated macrophages have been observed to be small, less vacuolated and express more MHC II than alternatively activated macrophages. Thus explaining in a lucid way macrophages can be “killer M1” or “healer M2” . M1 macrophages produce nitric oxide (NO)  in high concentration, which aids in inflammatory response as it is a potent vasodilator; on the other hand M2 macrophages not only do produce NO in low concentrations, but also secrete high levels of arginase-1, arginine being the common substrate for both arginase and iNOS. NO cause’s tissue damage and inhibits replication while as Arginase-I causes cell healing and proliferation via formation of ornithine [14, 15, 16].
Fundamental cytokine biology lays stress on the pathways followed by cytokines where some of them are upregulated during inflammation and others downregulated. A balance between cytokines is maintained by a balance between their respective macrophage strains which in turn is regulated by a balance between Th1 and Th2 lymphocytes via their respective lymphokines . Studies have shown a link between the cytokines of one macrophage strain to that of another e.g. when transforming growth factor-β1 (TGF-β1), a potent anti-inflammatory cytokine from alternatively activated (M2) macrophages is inhibited by its antibody, IFN-γ significantly stimulates nitric oxide (NO) production . Pro-inflammatory cytokines like IL-6, TNF-α and Interferon-γ (IFN-γ) from classically activated (M1) macrophages upregulate genes like cyclo-oxygenase-2 (COX-2), interstitial nitric oxide synthase (iNOS) and phospholipase A2 which in turn stimulate the synthesis of mediators of inflammation like PAF1, prostaglandin E2 (PGE2), NO, etc. triggering the inflammatory cascade. A balance is maintained when anti-inflammatory cytokines from alternatively activated (M2) macrophages inhibit the cascade keeping the process regulated.
Disorders of unregulated macrophage plasticity
Inflammation is an innate immune response elicited due to tissue injury or destruction and serves to destroy, dilute, or wall off both the injurious agent and the injured tissue. It is a highly regulated process which is under continuous check by most of the immune cells and genes, macrophages being the most important part of them. Macrophages are essential for the initiation and resolution of pathogen- or tissue damage-induced inflammation . However, loss of regulation in inflammation due to genetic or environmental interferences may cause a number of disorders.
Obesity-associated complications are the widely studied problem which is associated with a state of low-grade inflammation. Obesity alters the metabolism of adipose tissue with an elevation of hormones and fatty acids that lead to the complications. Obese individuals are predisposed to cardiovascular pathologies and type 2 diabetes. Studies on human subcutaneous adipose tissue have shown that the adipose tissue macrophages (ATM’s) differ in their composition from obese overweight individuals to the lean ones, with remodelling of subcutaneous adipose tissue  associated with infiltration of classically activated (M1) macrophages. Macrophages can thus be said to have a major role to play in obesity-associated inflammation. MCP-1/CCR-2 pathway is a very crucial step that signals macrophage recruitment into the obese adipose tissue. Macrophage chemoattractant protein-1 (MCP-1) secreted by classically activated (M1) macrophages has hence been associated with the macrophage infiltration into the obese or inflamed adipose tissue [19, 20, 21, 22], showing that the M1 macrophages have a role in the generation of low-grade inflammation in obesity.
Obesity-associated insulin resistance
One of the known complications of obesity is insulin resistance, a disease where the tissues of body become resistant to secreted insulin, which gradually proceeds to ‘the metabolic syndrome’ where insulin resistance is associated with hypertension, acanthosis nigercans, nephropathy, poly-cystic ovary syndrome, etc. Salicylates have been used for treatment of type II diabetes since early times; although at that time their anti-inflammatory properties were not known. Role of TNF-α in inducing insulin resistance has been widely studied, and link between inflammation and insulin resistance elucidated [23, 24]. This revolutionized the concept about the role of adipose tissue in inflammation, as adipose tissue is associated with high levels of IL-6, TNF-α, PAI-1 etc. . Studies have shown that deficiency of TNF-α and iNOS improves insulin sensitivity in obese wild-type mice . Anti-inflammatory therapies and the disruption of proinflammatory genes, such as IκB kinase-β (IKKβ) have shown to improve insulin sensitivity in obese animals  with high expressions of JNK and IKK-β in adipose tissue [28, 29]. Examination of the properties of adipose tissue macrophages (ATMs) in obese animals with a focus on the macrophages that are recruited to adipose tissue during high-fat diet exposure, have shown that these cells have unique inflammatory properties [30, 31].
An inflammatory component is also observed in a number of tumours for example there is a massive infiltration of immune cells including macrophages [18, 32]. The spleens of lymphoma bearing mice have been shown to have a subset of alternatively activated macrophages [33, 34]. Clinical and experimental evidence has been given that M1 macrophages mostly promote progression of tumours, including metastasis . In >80 % of the cases, the density of macrophages is directly proportional to poor prognosis . Macrophages isolated from tumours have been observed to have similarity to the macrophages found in developing and regenerating tissues , since, M1 macrophages are the ones found abundantly in regenerating tissues hence, it can be inferred that M1 macrophages have a progressive role to play in tumours.
Atherosclerosis is now regarded to be a modified type of inflammation where monocyte-derived macrophages have a central role to play . One of the characteristics of the atherosclerotic plaque is the infiltration of chronic inflammatory cells. There are many types of plaques but, inflammation is a part of all the forms [38, 39]. The role of macrophages is so immense that insulin like growth factor-1 (IGF-1), an important monocyte and macrophage chemotactic agent represents one of the most important mediators in the transformation of a stable lesion into an unstable one. Lumeng et al. observed that ApoE (Apoe) gene expression was decreased in the recruited ATMs compared with resident macrophages affecting cholesterol efflux in recruited macrophages, similar to macrophage foam cells formed over fatty streaks in atherosclerosis . Inflammation, insulin resistance and atherosclerosis are so much related that adipokines like leptin and resistin and inflammatory cytokines released by adipose tissue like TNF-α or the agents released in response to their release e.g. CRP are considered to be good markers of cardiovascular condition .
Wound healing is another phenomenon known to be effected by the inflammatory state of the body as the wound healing process proceeds via three steps: inflammation, new tissue formation and remodelling . Epidermis and connective tissue participate actively in wound healing. Several cytokines and growth factors enter the circulation following a wound, majority of which are secreted by neutrophils and macrophages . Redd and colleagues  observed that PU.1 null mice which are genetically incapable of raising inflammatory response showed less scaring properties than the normal ones. Thus in this scenario also regulation of inflammation can be a significant step towards prohibition of scar forming and quick healing of wounds. Macrophage skewing is an important factor in the process of wound healing. This can be observed as impaired wound healing in absence of M2 cytokines like TGF-β and vascular endothelial growth factor (VEGF) as was observed by Eming et al. , getting M2 macrophages to the central stage of the process of wound healing.
Macrophage plasticity and altered gene expression
Nuclear factor kappa-B (NF-κB)
Each strain of macrophages shows its effect via its respective cytokines which in turn signal the over- or under-expression of the required genes. NF-κB is a well-known, widely studied pro-inflammatory gene expressed highly by classically activated macrophages. The NF-κB pathway is an important mediator of immunity and inflammation, with short-lived responses characteristic of the canonical pathway, controlled by inhibitor of nuclear factor kappa-B kinase alpha (IKKa) activation, and long-lived responses characteristic of the alternative pathway controlled by nuclear factor kappa-B inducing kinase (NIK) activation . NF-κB activation is mainly associated with nuclear translocation of the p50/p65 heterodimer . It has been observed that the macrophages showing tolerance to the endotoxins, action against endotoxins being the important role of NF-κB, express M2 phenotype with impaired expression of M1 functions. P-50 homo dimers are associated with reduced nuclear translocation of NF-κB . Ablation of the NF-κB p50 subunit prevents development of tolerance, assessed by the selective restoration of M1 mediators (e.g. TNF-α and iNOS) and inhibition of M2 cytokines and chemokines, in response to LPS rechallenge [56, 57]. Thus, the activation or inhibition of NF-κB via selective targeting of its p50 subunit may be the important step for macrophage skewing.
NF-κB and insulin resistance
Inflammation is a predominant factor for insulin resistance where NF-κB has a huge role to play as described in this review. An increased M1 macrophage content with elevated nuclear localization of NF-κB and up regulated transcription of its target genes in the adipose tissue is a characteristic in obesity assisted insulin resistance [45, 58, 59]. Relation of NF-κB to the macrophage polarization came into light when it was observed that TNF-α, a well-known pro-inflammatory cytokine secreted by M1 macrophages, causes serine phosphorylation at two N-terminal domains of IKK, causing up regulation of inflammatory factors by nuclear translocation of NF-κB . Inhibition of NF-κB has been observed to reduce the expression of inflammatory markers like IL-6 and suppressor of cytokine signalling-3 (SOCS-3), in the in vitro inflammatory conditions of 3T3-L1 adipocytes exposed to TNF-α . Relatively weak inhibitors of IKKB have been observed to show anti-inflammatory effects  and it leads to inhibition of M1 macrophages on infection by negatively interacting with Stat1 pathway. Deletion of IKKb in macrophages leads to the resistance to infection with an increased expression of NOS2, IL-12, and major histocompatibility complex (MHC) class II by macrophages  making IKKb a potent insulin sensitization target in future. A number of studies have demonstrated that saturated fatty acids lead to insulin resistance  via NF-κB, induced pro-inflammatory cytokines like IL-6, TNF-α [46, 64, 65], thus, fatty acid-induced insulin resistance is reduced by blocking NF-κB activation . Thus by the discussion above, the role of NF-κB in insulin resistance can be summarized as the molecule which has a potential induce to insulin resistance via induction of macrophage polarization to M1 form.
NF-κB and tumour
The role of NF-κB in tumour is immense as its activity is modulated by the tumour suppressors like P53 and alternative reading frame (ARF) and its subunits suppress the expression of tumour suppressor genes . NF-κB came out to be a major tumorogenic transcription factor after analysis of NF-κB1 subunit (P50–P105) [68, 69] showed its homology to a potent transforming oncogene, V-rel . Another subunit, NF-κB2, is constitutively found in B and T cell lymphomas in rearranged and truncated form . Activation of NF-κB in Hodgkins lymphoma is one of its well-known characteristics . Many studies have shown NF-κB to give resistance to a number of cancers, this may partly be due to the varying causes of cancers where some of them are inflammatory in nature and some are not. The role of NF-κB in anti-cancerous activity can be derived from the fact that DNA damage, apart from activating P53 tumour suppressor induces enhanced DNA binding and transcriptional activity of NF-κB . Ryan and colleagues  observed that NFκB-P65 was required for P53-dependant apoptosis. Tumour suppressors like ARF can induce association of P65 with histone deacetylase-1 (HDAC1), causing repression of the gene expression . Thus, NF-κB has a central role whenever involved in tumours be it tumour progression or tumour suppression depending on the type of cancer.
SH2 containing inositol-5′-phosphatase 1 (SHIP-1)
SH2-containing inositol-5′-phosphatase 1 is a 145 kDa protein that has been observed to show the property of endotoxin tolerance . The stimuli for action of SHIP-1 are cytokines, growth factors, antibodies, chemokines, and the integrin ligand [77, 78] SHIP-1 is present in mature granulocytes, monocytes/macrophages, mast cells and platelets [79, 80]. It performs a wide variety of functions by reducing mast cell activation and adhesion to fibronectin by reducing extracellular calcium entry and activating NF-κB and certain PKC isoforms , activating B cells , thrombin- or collagen-induced activation and fibrinogen-induced spreading of platelets [77, 80], neutrophil surveillance  and early erythroid colony formation . It was observed by Rauh et al. that SHIP 1− mutant mice produced tenfold less nitric oxide in comparison to wild counterparts . A tolerance to endotoxins can be improved by upregulation of SHIP as it enhances the classical (M1) activation of bone marrow-derived macrophages. Peritoneal and alveolar macrophages from SHIP−/− mice are abundantly M2 skewed in comparison to their wild-type counterparts, possessing impaired LPS-induced NO production, constitutive arginase-I (ArgI) and YM1 levels . Hence, from above discussion it is evident that SHIP-1 has an important role in M1 macrophage function and change in its expression can be associated with the skewing of macrophages.
SHIP-1 and inflammation-associated insulin resistance
Src homology 2-containing inositol-5-phosphatase, SHIP, is a potent inhibitor of the phosphatidylinositol 3-kinase (PI3K) pathway which negatively regulates the lipopolysaccharide (LPS)-induced proinflammatory cytokine and nitric oxide (NO) production. The response of a particular gene to LPS is an important indicator of its role in inflammation and related disorders. Since, the strain of macrophages activated in response to LPS is M1 type, hence, SHIP may be an important gene marker of this phenotype. The production of pro-inflammatory cytokines likes TNF-α and IL-6 is found to be very low in case of SHIP−/− bone marrow macrophages. A 10-fold increase in SHIP protein is observed in response to LPS which shows the essentiality of SHIP-1in response to endotoxins  and it also shows the role of SHIP-1 in M1 macrophages as this is the macrophage type involved in combating endotoxins before the development of endotoxin tolerance which is the characteristic of M2 macrophages. PI3K has been shown to negatively regulate the stability of COX2 mRNA in LPS-stimulated human alveolar macrophages , the dampening of PI3 K may be the main cause of pulmonary inflammation. The above discussion leads towards a suspicion that SHIP-1 plays a substantial role in inducing inflammation. Although, no substantial data are available on the relationship between SHIP-1 and insulin resistance, however, a number of inflammatory markers like MCP-1  and TNF-α  have been associated with insulin resistance, and inflammation is considered to be one of the major pre-requisites for insulin resistance, hence, the role of SHIP-1 in insulin resistance should be analysed in future.
SHIP-1 in tumour
An overflow of alternatively activated macrophages associated with heavy deposits of YM1 crystals is a characteristic of SHIP−/− mice [86, 87]. The dampening of PI3K pathway by hydrolysis of PIP3 to PI2P is a well-studied effect of SHIP-1 [48, 88]. PI3K pathway is quite essential in proliferation, differentiation, survival activation and migration of haematopoietic pathway . The pathway is used by a number of oncogenes like HER 2/Neu, Ras [88, 89] etc. The mutated form of SHIP-1 has been found in a number of leukemias and lymphomas [90, 91, 92, 93, 94, 95]. The presence of a number of isoforms of SHIP adds to the complexity of the study as the isoforms may be playing completely different roles. Holding SHIP responsible for all the tumours can be a misinterpretation as tumours have also got a genetic and physiological variety and the same gene can be anti for one type of cancer and pro for another one.
Peroxisome proliferator-activated receptors (PPAR’s)
These are the nuclear receptors like steroid and thyroid hormone receptors present in endothelial cells [96, 97]. Following their discovery by Issemann and Green  PPAR’s opened the gateway to the study of a number of similar receptors which have types and subtypes generated due to differential RNA splicing and alternate promoter use  The nuclear hormone receptor peroxisome proliferator-activated receptor γ (PPARγ) is a central switch that regulates the inflammatory potential of macrophages in both adipose tissue and atherosclerotic plaques , though widely studied, role of PPAR-γ and its involvement in macrophage polarization remains surrounded by several questions and confusions. In the absence of ligand, nuclear co-repressors bind PPAR/retinoid X receptor heterodimers  and are thought to recruit histone deacetylases to downregulate a gene. Free fatty acids, oxidized lipids, eicosanoid and prostaglandin derivatives weakly activate PPARγ [102, 103, 104]. The receptor may respond to an integrated concentration of nutritionally derived fatty acid ligands and that no high-affinity regulatory ligand exists . PPAR-γ-dependent polarization to the M2 phenotype has been observed at the monocyte level, also, activation of macrophage PPAR-γ inhibits the expression of numerous inflammatory mediators . Lazar, reviewed that the expression of PPAR-γ in the vessels colocalizes with CD68-positive macrophages which is a marker for M2 macrophages . The above discussion supports the fact that PPARγ has a role in regulating macrophage plasticity to some extent and its expression is associated with M2 macrophages.
PPAR’s in inflammation-associated Insulin Resistance
PPAR-γ-advancing drug rosiglitazone is used as an Insulin-sensitizing drug worldwide, showing PPAR-γ as an important key to the doorway of treatment of Insulin Resistance. PPARγ inhibits the expression of a wide range of inflammatory mediators . Thiazolidinediones (TZDs) and other PPARγ ligands have been shown to inhibit the secretion of TNF-α and IL-6 (pro-inflammatory cytokines) in PPARγ-deficient macrophages . Some studies suggest that PPAR-γ does not inhibit inflammatory cytokines; however, it enhances the activation of anti-inflammatory ones. Activation of PPARγ induces transcription of several genes and causes effects such as weight gain, increase in low-density lipoprotein cholesterol and plasma volume expansion . Adipocyte-specific genes, such as adiponectin and perilipin, contain response elements for PPARγ in their promoter regions [50, 110, 111] Mice with a macrophage-specific deletion of PPAR-γ have shown impairment in the maturation of alternatively activated M2 macrophages and up regulation of diet-induced obesity, insulin resistance and glucose intolerance. These mice were found to be defective in the in vivo generation of M2 to a similar extent as macrophage-specific IL-4Rα−/− mice or STAT6 null mice which are associated with inflammatory disorders. As a consequence, PPAR-γ induces more resistance to Th2/M2-driven pathologies, such as cutaneous leishmaniasis . PPAR-γ also causes ligand-based transrepression of NFκB [51, 113] which is an important inflammatory gene as discussed above. Thus, the role of PPAR-γ in insulin resistance has been well elucidated in insulin resistant models and shown to reduce macrophage polarization associated insulin resistance i.e. M1 type of classically activated macrophages .
PPAR’s in tumour
A dysregulation of the cell cycle leading to cell proliferation is the representation of Cancer cells making the agents that modulate this phenomenon as important chemotherapeutic agents. Among PPAR’s, PPAR-α is a well-known anti-apoptotic receptor . Study conducted by Peters and colleagues  revealed that there is an up regulation of proteins like CDK-4 and C-myc in PPAR-α+ mice as compared to PPAR-α− mice, increased expression of Cyclin-dependent kinase 4 (CDK-4) and c-myc is a common characteristic of a number of cancers . It also causes the induction of a common factor of colon carcinoma, cyclo-oxygenase-2 (COX-2), in the epithelial cells of colon . The role of PPAR-γ as a tumour promoter or suppressor remains a matter of controversy as on one hand some studies demonstrate PPAR-γ agonists to cause tumour progression in mice  and on another hand PPAR-γ agonists improve differentiation in tumour cell lines .
Eicosapentaenoic acid (EPA), fatty acid, suspected to have a role in macrophage polarization has been observed to decrease pancreatic cancer via activation of PPAR-γ . A wide array of research has been done on the role of PPAR-γ in cancers [52, 119]. Thus, looking from the PPAR-γ front, it is an important precursor for macrophage polarization towards alternative M2 phenotype which is an anti-inflammatory and regenerative type of macrophage. Before jumping to any conclusions regarding the therapeutic role of PPAR’s in tumour, a wide array of research needs to be conducted on relationship of peroxisome proliferators or their receptors on promotion or attenuation of carcinogenesis and also the type of agonists that may be helpful for the same.
The role of SHIP-1 in macrophage polarization in case of tumours has not been well elucidated yet; it can be a therapeutic target of immense importance in cancer drug research.
The varying expression profile of several important signalling molecules of inflammatory response like STAT’s, SOCS, etc. in case of macrophage skewing is still far from known.
A direct correlation between macrophage polarization and several inflammatory diseases is not well understood.
Reviewing the complete information about the two forms of macrophages and drawing a separating line between them to characterise each one of them including their relative gene expression can lay a foundation for the development of nucleus-targeted drugs which may be useful in treatment and prognosis of cancers that have a root cause in inflammation although, an extensive research needs to be done before reaching to any therapeutic conclusion.
- 12.Fernando M, Gordon S. The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000prime Rep. 2014;6:13.Google Scholar
- 32.Mantovani Alberto. Inflammation and cancer: the macrophage connection. Medicina. 2007;67:32–4.Google Scholar
- 86.Huang XL, Wang YJ, Yan JW, et al. Role of anti-inflammatory cytokines IL-4 and IL-13 in systemicsclerosis Inflamm. Res. 2015;64:151–9.Google Scholar
- 90.Ho VW, Hamilton MJ, Kuroda E, et al. Tumour-associated macrophages. Biomed Life Sci. 2012;3:135–51.Google Scholar
- 99.Fajas L, Auboeuf D, Raspé E, et al. The organization, promoter analysis, and expression of the human PPAR gamma gene. J Biol Chem. 1997;272(18):779–89.Google Scholar
- 107.Barlic J, Zhang Y, Foley JF, Murphy PM. Oxidized lipid-driven chemokine receptor switch, CCR2 to CX3CR1, mediates adhesion of human macrophages to coronary artery smooth muscle cells through a peroxisome proliferator-activated receptor gamma-dependent pathway. Circulation. 2006;114:807–19.PubMedCrossRefGoogle Scholar
- 114.Yerly-Motta V, Contassot E, Pavy JJ, et al. Expression of cyclins and cdks throughout murine carcinogenesis Cell. Mol Biol. 1999;45:1217–28.Google Scholar