Assuming that you mean in the macrophages in mice:

For M2 we use CD206 (Mannose receptor), and

For M1 we use MHC class II.

There are several other markers of course such as for M2: CD163, LYVE-1 (careful: also on lymphatic vessels), YM-1 and

for M1: CD86.

I can refer you to two of the best papers that tried answering the question (in my opinion).

#1

Martinez, Fernando O, Siamon Gordon, Massimo Locati, and Alberto Mantovani. 2006. “Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression..” Journal of immunology (Baltimore, Md. : 1950) 177(10):7303–7311.

#2

Kigerl, Kristina A et al. 2009. “Identification of two distinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord..” The Journal of neuroscience : the official journal of the Society for Neuroscience 29(43):13435–13444.

Best of luck in your experiments!

In our experience, it is very simplistic to use only one marker to make the differentiation between M1 and M2 macrophages. Macrophages are very plastic cells and you will need to use 2 antibodies, at least, to have sure that you are "seeing" macrophages and to check the polarization status (if you are working with tissue, as FFPE tissue, for example).

I do not particularly agree that CD163 been specific to M2 polarization. More details about this, you can see in our recent paper:

Tumor-associated macrophages in pediatric classical Hodgkin lymphoma: association with Epstein-Barr virus, lymphocyte subsets, and prognostic impact.

Barros MH, Hassan R, Niedobitek G.

Clin Cancer Res. 2012 Jul 15;18(14):3762-71. Epub 2012 May 29.

Good Luck!

I have three approaches to determine the phenotype: morphology (Quik-diff staining), Gene expression of iNOS/Arg1 ratio, cytokine IL-1beta, chemokine CCL2 and Ptgs2, YM-1, Fizz1, and Flowcytometric analysis. Need a standard way to decribe macrophage phenotype for the future study. I have read the JI paper, it is impossible to do all your experiments by chips, too expensive.

M1 and M2 are nice concepts but are unfortunately the extremes of a continuum of intermediate cells, which becomes quite clear when you look in vivo instead of in cells cultured in vitro with IFNg vs IL-4. You'll hardly find "real" M1's or M2's in your mouse but should rather focus on the question whether your macrophages are more or less "skewed towards the M1 or M2 direction". This means that you can for example detect an upregulation of CD163 and/or CD206 in your "M2-like" macrophages but you'll hardly detect a downregulation of CD86 or MHC-II. Moreover, while there are quite some useful markers for M2 differentiation (and the best markers to use really depend a lot on the disease model you're using and the organ where you isolate your macrophages from etc), there are almost no good M1 surface markers.

Therefore, it's always better to look at changes in ratios, with the most useful ones being the iNOS/arginase ratio (easy to test in vitro with a NO assay vs an arginase assay, difficult to test in vivo) and the IL-12/IL-10 ratio (possible to test on ex vivo material with intracellular FACS).

See one of the articles mentioned above or "Stöger, Goossens & de Winther, Curr Vasc Pharmacol 2010"

I agree with Pieter, M1 and M2 are definitely extremes of a continuum and generally only found in vitro. To be as accurate as possible in vivo, you should definitely use multiple markers for each population. In addition to some of the markers already mentioned, CD40 tends to be a good cell surface marker for M1-like macs and the Transferrin Receptor is a strong marker of M2-like macs (Becker et al, PLOS ONE, 2012).

I agree with Sean and Pieter regarding the continuum.

Even, in the exterme conditions the changes in surface markers are quantitative. You will still see some CD206 in M1 and definitely MHCII in M2s; the difference will be in the level of expression (MFI is a good way as Peter Suggested). The bottom line is, we need to use multiple markers and some "baseline" for comparison wether these markers are increased or decreased. Among the multiple markers co-stimulatory molecules (CD80 and CD86) may be usefull, we see more of them in M1s. We also use Galectin3 (or MAC2) as a readout of M2. Finally, there are intracellular antibodies for arginase and iNOS proteins and these coudl be used as well.

Chang-Jiang, I would recommend the following 2 papers to answer your question:

1 Gustafsson C et al PLoS One 2008

2 Chinetti-Gbaguidi 2011

Also, I personally would recommend including an assay for a functional outcome measure, otherwise you risk being criticized for only being descriptive. Thus, reporting what the physiological consequence of a skewed M1 population is, vs a skewed M2 population, may be important.

According to my understanding of the concept of M1/M2 polarization of macrophages, the essence is different way of arginine metabolism by these two phenotypes. M2 macrophages express high level of arginase, involved in wound healing functions of these cells, whereas high expression of iNOS in M1 macrophages contributes largely to their killing functions. Application of other criteria, without determining relative expression or activities of these two enzymes, would not be convincing for me.

Are mouse "M2" macrophages express 7/4 (Ly-6B.2) or is this marker common on all macrophages? Thanks in advance.

I think in view of all the prededing discussion, it is not only important to look at secreted products and surface markers. Some functional tests are very useful, particularly whe looking at M1/M2 ratios in vivo. M2 are more phagocytic/efferocytotic and M1 contribute more to antigen presentation

Sorry, incorrectly written: M2 are more effective at antigenpresentation

I would suggest intracellular staining for Relm-alpha it is only increased following IL-4/IL-13 and not under other inflammatory conditions.

Good luck!

The most important thing to understand about M1 and M2 macrophages is that these activities are Inhibit or Heal responses, respectively. M1 and M2 was originally described by the ability of macrophage populations to preferentially produce Nitric Oxide (Inhibit) or Ornithine (Heal). Fascinatingly, both of these polar opposite molecules result from the enzymatic cleavage of Arginine in slightly different ways. Innumerable subsequent studies have indicated that succinct cytokines and other products are closely associated with M1/Inhibit or M2/Heal dominant responses.

Products closely associated with M1/Inhibit activity are: iNOs/NO production; increased IL-12 production; increased MHC Class II expression; and IL-8 and CCL2. M2 activity, in contrast, is associated with: Arginase/Ornithine production; mannose receptor expression; EGF, VEGF, and TGF-β production; Fizz 1, and the production of certain matrix and matrix remodeling molecules, such as metalloproteinases. Certain cell surface markers are useful, but Class II and the mannose receptor are the best to distinguish M1 from M2.

Products such as TNF-1, IL-1/6, and IL-10 have been incorrectly ascribed by some investigators as M1 or M2. These cytokines have been observed in both M1 and M2 dominant circumstances. Inflammation is a commonly used term, but only means that macrophages are present. So, it is far better to describe what kind of inflammation is involved. Some of the confusion about M1/M2 has arisen because investigators (some listed above) have made so called "subsets" of macrophages by adding various T cell and other cytokines in vitro. None of these various Type II, Immunoregulatory, or M2 a-c cells created in vitro by D. Mosser, A. Mantovani, and others, have been shown to be real distinct subsets in vivo.

In this connection, it is useful to understand that M1 and M2 activities do not require T cells. And, that macrophages were renamed M1 and M2 because they direct naive T cells to make Th1 or Th2 like responses. Innate First; Adaptive Second. Thus, M1 and M2 are not synonymous with the older and misleading terms "activated" or "alternatively activated". For example, M2 like macrophages cause T cells to make a Th2 response, and IL-4 is produced. "Alternatively activated has it backward: that IL-4 first stimulates macrophages. That Innate controls Adaptive is a sea change in our understanding of how immune systems work. Papers in which I discovered M1/M2, and recently reviewed these fascinatingly different macrophage activities are attached. I hope you find the foregoing useful. Good luck. Charlie Mills

Review Paper About M1/M2

Seppe:

I would encourage you to re examine how you came to classify macrophages above. Most, read a review article, and go with that. The cytokine categories you have separated above are quite incorrect, be they from Mosser or Mantovani, etc. For example, TGF-b, and other so called "immunoregulatory" cytokines very strongly track with M2 like activity as found similarly in wounds or tumors. M2, by the way, is the phenotype of resident tissue macrophages, and does not required "activation. You also mention focusing on what macs do. I completely agree. The 2 most important functions of macrophages throughout the animal kingdom, including in humans, is to Inhibit or Heal. As mentioned, "inflammation" is not a function.

Chang - Jiang:

NO and Ornithine are quite easily measured in vivo, or in vitro. Perhaps not as easy as cell surface markers. But, surface markers, mostly, are not directly equated with functions. And, functions are what matter to hosts/disease outcomes. Charlie Mills

Hi there,

we look for CD163, CD80 and CD86. Surface marker CD163 is upregulated in M2 type macrophage populations, but its upregulation is depending on how you polarized macrophages (IL-4/IL-10; medium RPMI or ex-Vivo). We observe B7 upregulation (CD80/86) after infection in M2 macrophages. Intracellular markers like IL-10 (more in M2) and TNF-alpha (more in M1) are easier to interprete.

All the responses above are correct. There are many markers to chose from. I just wanted to add my "3 cents".

1) There are some differences between human and mouse.

For example CD206 is a good M2 marker in mice but CD200R seems to work better in human cells. 2) The differences between M1 and M2 markers are quantitative. For example both M1 and M2 will express MHCII, but M1 will express it with much greater intensity. The same applies to most if not all surface markers that are used for M1/M2 differentaiton, so it is always good to have some sort of "positive control".

3) There are some great intracellular stain reagents (kits) developed to study M1/M2 by flow: iNos (M1), Arg1 and RELM-alpha (M2) worked well in our lab (in mice).

I agree with Michal. The differences between M1 and M2 are quantitative. In my hands the more marked differences between M1 and M2 derived from THP-1 cells is CD80 (M1) and CD206 (M2), the rest is a pattern of low/high

M1 usually exhibit CD68+/CD80+/CD163- or CD163 low, whereas M2 CD68+/CD80-/CD163+.

Check this publication: Khramtsova G, Liao C, Khramtsov A, Li S, Gong C, Huo D, Nanda R (2009). The M2/Alternatively Activated Macrophage

Phenotype Correlates with Aggressive Histopathologic Features and Poor Clinical Outcome in Early Stage Breast Cancer. Cancer

Res, 69(24): Supplement 3, doi:10.1158/0008-5472

additional info in attached file...

I think PD-L2 is the best cell surface marker for M2a or alternatively activated macrophages, but only for macrophages derived from monocytes.

The M1 and M2 macrophage differs between mouse/human, between macrophage cells/peritoneal macrophages and also by different stimulli like patient samples or ex vivo stimulii. For example-

In human he best marker is CD206 or mannose receptor 1 or MRC1. Additionally you will find arginase 1 activity, decreased intramonocytic or intramacrophage NO, ROS if you stain them with fluorescent dyes like DAF-2DA, CMH2DCFDA. You should look for the increased expression of PPAR-gamma also (peroxisome proliferator activated receptor-gamma)

In mouse cells you should look for Fizz-1, RetnlB, arginase, and then mannose receptor.

Now, critical thing is when you perform ex vivo generation of different macrophages-

For M1 you can use IFN-g alone or with LPS or GM-CSF or a cockltail of all. You will get different result for different stimulii, so pls specify that when you use such protocols, otherwise it will be difficult to understand.

Similarly, for M2 macrophage generation you also have plenty of options-

1. IL-4 only

2. IL-13 only

3. IL-4 + IL-13

4. M-CSF

5. IL-4 + MCSF

6. IL-4+IL-13 +MCSF

7. IL-13 + MCSF

8. IL-10+TGF-b1

8 IL-4+ IL-13+TGF-b+IL-10

9. All the above in a cocktail

10. Immune complex + IL-10

For all these stimulli you can find M2 macrophage but with different transcriptome, so mentioning the stikulii is essential.

Finally, it is better to see the morphology under the microscope-

M1: egg shape or round

M2. Fried egg shape or with dendrites

Hope this will help, I have struggled for many years with these cells

A commonly accepted marker profile for M1-macrophages is CD68+/ CD80+, whereas M2-macrophages are characterized as CD68+/CD163+.

In vitro differentiation of monocytes in the presence of the PromoCell M1-Macrophage Generation Medium DXF (C-28055, contains GM-CSF) leads to macrophages exhibiting a CD68+/ CD80+/CD163-/low M1-like polarized phenotype, whilst the M2-Macrophage

Generation Medium DXF (C-28056, contains M-CSF) promotes M2-like polarized CD68+/CD80-/low/CD163+ macrophages.

All the "markers" that trascritomic or genomics can identify (which is essentially infinite) in macrophages is mechanistically interesting. But, for clinically utility, all that finally matters is how macrophages affect heath in vivo. M1-type kill (or direct T cells into Th1-type responses) or M2-type repair macrophages direct T cells into Th2-type responses. These types of responses affect health in profoundly different ways - such as killing or promoting the growth of tumors in vivo. I'd strongly recommend  moving on from determining how many functionless "subsets of macrophages (defined by such things as only cell surface markers) to defining macrophages by what they "do". Articles attached that may be helpful. Remember: macrophage functions are required for hosts survive; "phenotypes" of macrophages don't cure (or exacerbate) diseases. Heterogeneity is not a function of macrophages - its how they achieve functions. Good luck. Charlie

I absolutely agree to the opinion of my collegues regarding "poor" M1 markers....

However - in this context - you should consider to challenge macrophages  with bacteria. M1 macrophages shoud inactivate more bacteria and express more TNF-alpha compared to M2 macrophages. Again, it`s all about ratios....  

For some who have "perceptions" about M1/M2, I might add that M1/kill and M2/repair activities were first elucidated In Vivo in wounds and tumor. not In Vitro. For example, macrophages that inhibit proliferation or promote proliferation, respectively. A recent review is attached, or see 1990 and 1992 JI papers by Albina or Mills et al. Various other "subsets" were later proposed from in vitro studies (e.g., Mantovani). But, it is not clear that many of the in vitro described "subsets" have identifiable health-impacting functions in vivo. So, M1 and M2-type activities are well known to strongly (and differently) affect health outcomes. The molecules first identified that mediate these very different outcome are Nitric Oxide (kills, easily measured as NO2/NO3) or ornithine (proliferation, measured by amino acid analysis or by the expression of arginase gene or enzyme). As I said earlier, some have substituted other "markers" which can be easier to measure, but also might not give as accurate  a reading of M1/kill or M2/repair-type activity. Charlie

At a conference last year, I heard a medical doctor suggest that M1's have not yet phagocytosed material, and M2's have, which was an appealing description to me because it defines the subsets based on in vivo function, not surface markers. I think this description agrees with the many papers that have been referenced in this discussion, but this description is simple and intuitive and so I found it memorable. I haven't read them all, but it seems like most papers have indeed fallen into the trap of defining the cell types based on what is measurable; whether the phenotype coincides with an in vivo function has been an afterthought. I agree with Stephan then, that one would expect M1's to be able to "eat" more bacteria than M2's. Charlie, is this perspective (which is admittedly physiological, rather than flow-cytometric) consistent with what you're saying?  Chad

Hi Chad:

Thanks for your interesting post. Phagocytosis was the basis for Metchnikov naming them "Big Eaters" (macrophages), of course, in the late 1800's. Classical activation (as originally elucidated by Mackaness and colleagues in vivo at the Trudeau Institute) has long been associated with increased phagocytic activity, and increased killing ability through the production of reactive oxygen and nitrogen species. So, temporally, it would seem that "killer"/M1 macrophage activity is linked to phagocytic activity often. Of course, phagocytosis is not required to classically activate macs; they can get revved up by Toll and other receptors (e.g., binding LPS). So, in this sense I'm not sure I could support M1-type as non phagocytic versus M2-type as  a stand alone way to distinguish macs. Too, eating by macrophages is not really a "functional" activity because it is what happens after they eat that really matters. Do they digest/eliminate what they eat and/or do they kill it with O or N species.? Of note: M1/kill activity results in the production of Nitric Oxide which is a gas that freely diffuses in all directions killing everything nearby - it is non-specific. This includes killing the macs that make it (reviewed in the Anatomy of a Discovery paper I posted earlier). So, those that have posited bidirectional "plasticity" (e.g., Mosser), I feel do not fully understand the biology of macs. Likely, M1 activation normally results in  an end stage cell that "takes one for the team" (dies in battle). Thus, M2s (resident macs) can become M1s, but not normally vice versa. Actually, I coined the term "plasticity" in 2001 to indicate that macs can change, and that all populations of macs are evolving mixtures as inflammation proceeds in one way or the other. Getting back to phagocytosis, certainly M2s eat. Whether on a cell per cell basis they eat more than M1 activated, or that this is a valid way to distinguish M1 and M2 activity, I'm not so sure.  After all, what we agree on is that assessing macrophages by their health-impacting effects (be it wound healing or pathogen elimination) is the most valuable metric. Finally, Alex Carrel in the early 1990s proposed that macrophage could "transform" into fibroblasts. I believe this is true in some regard, though what "flavor" of fibroblast they become is unknown. This seems to occur in inflammation that is "sterile", like wound healing, or in vitro if there is no M1-type activation. So, if this is true, M2s are the ones that transform. Fibroblasts lose most ability to phagocytose, like macs. So in this circumstance, M2s evidently have less, not more phagocytic activity. Check out Figure 4 of my attached original M`/M2 paper. Mac morphology changes dramatically depending on the culture conditions. If interested, there are some references to macrophage conversion to fibroblast-like cells in my 2012 review attached. Anyway, those are some thoughts about the complex "lives" that macs lead. I hope helpful.  Charlie

Thanks for the great information, Charlie.  Chang-Jiang, I'm curious, since you initiated the discussion, have you assessed macs for the phenotypic markers, and found any correlation between markers and functional effects like wound healing or pathogen elimination? In the time since you originally posted the question, have you made any observations that you can share?

For everyone else in the discussion, I'm fascinated with immunological influences on aging skeletal muscle, so Charlie's mention of macrophage-to-fibroblast transformation makes me think that chronic transformation of macrophages to fibroblasts might explain why skeletal muscles stiffen as we age...and that "excessive" transformation (however you wish to define that) might be a factor explaining why some elderly people's muscles are stiffer than others'...any thoughts on that are welcomed.  Chad

Excellent question and posts. 

We struggled with the same issue of difficulty finding differentiating markers for M1 and M2 macrophages. We were expecting the continuum of macrophage phenotypes to impact in vivo phenotype but not highly defined M1 or M2 in vitro BMDM. However, we found that those were also difficult to distinguish by standard flow cytometry techniques, since CD206/Mannose receptor expression was very low and Arginase-1 protein expression was low (besides the fact that some M1 macrophages also up-regulated Arg-1). Therefore, we looked for exclusive M1 and M2 markers by the strategy of selecting genes up-regulated during M1 differentiation but down-regulated during M2 differentiation and vice versa. The results of this study and the new markers we have identified, including a new CD38/Egr2-based flow cytometry strategy to distinguish M1 and M2 macrophages, can be found in Jablonski et al. Plus One. 2015 (find link below). 

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0145342

It is important to point out that M1/kill and M2/repair-type macrophage responses were defined in vivo originally (not in vitro). And defined by the production of functional molecules that inhibit or promote proliferation of other cells (i.e., NO or ornithine). Specifically in sterile wounds and growing tumors macs only make ornithine whereas during tumor rejection there is prodigious production of NO (see attachment one). Thus, mac functions that affect health in vivo is what originally defined  M1 and M2 type responses. "Markers" came later (mostly from in vitro studies, like with BMDM you mentioned). But, markers are not necessarily functions; a potentially big pitfall that some in vitro-type researchers, like A. Mantovani, and some others don't understand. For example, NO production by macs decreases ornithine production (attachment two) because  N hydroxy l arginine (an intermediate in NO synthesis) inhibits the arginase enzyme. So, measuring only arginase activity in vitro, staining for arginase by IHC or looking at arginase gene expression can be misleading because the actual M2 functional product, ornithine, is not measured. If you wish to read more about how M1/M2 responses were discovered and came to be named such, you might enjoy attachment three. Good luck. Charlie Mills

What a fantastic Posting!

I am very new to this field and have found these posts most helpful for my research.

Coming from a cell biology background and entering the field of flow cytometry, I see the importance of linking markers and function. I'd like to thank the researchers on this page for posting such useful information.

Sorry I don't have an answer to contribute, but I just submitted an abstract to CYTO2016. I don't suppose any of you will be there?

Best regards,

Jenni

Hi, 

CD206/MR and Arginase-1 have been classically considered M2 and iNOS, IL-1, TNF, have been classically classically considered M1. However, CD206/Mannose receptor and Arginase-1 protein expression was low for ideal flow. In addition, some M1 macrophages also up-regulated Arg-1. Since reliable markers of mac phenotype would be a huge benefit to the field, we looked for exclusive M1 and M2 markers. Our strategy: select genes up-regulated during M1 differentiation but down-regulated during M2 differentiation and vice versa. The results of this study and the new markers we have identified, including a new CD38/Egr2-based flow cytometry strategy to distinguish M1 and M2 macrophages, can be found in Jablonski et al. Plus One. 2015 (find link below).The markers described in this paper allowed to perfectly distinguish in vitro M1 and M2 macrophages and also identified inflammatory populations in vivo.

Article Novel Markers to Delineate Murine M1 and M2 Macrophages

The production of NO by macrophages causes a decrease in ornithine production. These are the 2 best functional molecules that distinguish M1/kill from M2/repair populations. Genes are not functional molecules that affect health (for good or bad), they are 'markers'. Further, an intermediate in NO production is N-hydroxy-l-arginine which inhibits ornithine production by inhibiting arginase enzyme activity. Thus, measuring the products, not the genes or enzymes is the  most direct way to assess M1/M2 activity. Finally, what paper exactly (not a review) has ever shown that arginase activity is unregulated in macrophages when NO production increases. A few have written that in the literature without data, and it doesn't make sense biochemically. Charlie

I'll try to summarize the recent conversation--It sounds like the question has evolved to "What cell surface markers in macrophages most closely correlate with NO and ornithine production?"   There have been several great insights/references shared in previous posts.

A few thoughts about caveats we might take in answering this question:

Whether we are focusing in vivo or in vitro,

If in vivo, what species,

If in vitro, what are the culture conditions,

and strengths / limitations of the methods we use.

Good points Chad. Both NO and ornithine levels were first described in vivo in wounds and tumors. Summarized in attached paper in Figures 4 and 5. Original studies were in mice and rats. Studies in vivo in humans have verified the results. Both NO and ornithine are made in many species of animals. In vitro analyses has mirrored in vivo results. Strengths of measuring functional molecules are obvious, but more difficult because amino acid analysis is more involved (for urea/ornithine and citrulline). NO measurement (as NO2) is easy with Greiss (colorimetric). Other markers (cell surface, genes or enzymes) are easier, but may not track properly with M1/kill and M2/repair responses Best 'markers' are summarized in Figure 9 of attached paper, though some mentioned in this discussion are good as well.  Remember, 'inflammation' markers can be troublesome because whenever macs are present there is inflammation. Some like Mantovani have confused the issue by measuring such 'inflammatory' markers So, stay with kill or repair functions or markers. Finally, M1 or M2-type responses stimulate Th1 or Th2-type responses as another way of analyses. I hope helpful. Charlie

Hi,

Can these populations - M1, M2 switch back ? say M2-M1 to M2 back ? Any idea?

Hi, for macrophage switch,  I suggest you to read manuscript: 

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1809537/

Regards,

Denis

Hi Shilpa:

Understand that macrophages are not chameleons: able to change their 'colors' as necessary. If macrophages commit to kill, they make molecules, like NO, that kill everything nearby, including the macrophages themselves. Innate kill molecules are non-specific unlike adaptive responses, like CTL that can bind to and kill specifically without killing themselves or surrounding targets. So, the whole idea of bi directional 'plasticity' in macrophages welled from people who don't fully understand macrophage biology/biochemistry. Mosser et al is on source of thinking that macrophages are like a 'color wheel', akin to chameleons. M2/repair to M1/kill: yes. The other way, no. I hope helpful. Charlie Mills

Demis: Perhaps my thoughts above are helpful to you too. Charlie

This may help also

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0145342

Novel Markers to Delineate Murine M1 and M2 Macrophages

Kyle A. Jablonski, Stephanie A. Amici, Lindsay M. Webb, Juan de Dios Ruiz-Rosado, Phillip G. Popovich, Santiago Partida-Sanchez, Mireia Guerau-de-Arellano

Published: December 23, 2015http://dx.doi.org/10.1371/journal.pone.0145342

Good thoughts above. It is worth noting that M1 and M2 classifications arose from in vivo studies. For example, in sterile wounds, M2/repair dominates. In things like rejecting tumors, M1/kill dominates. M1/M2 measured inside the wounds or tumors by the levels of NO/citrulline or ornithine/urea. Adding IFN-g or IL-4 is a technically useful way to skew macrophages. One should note, however, that iL-4 is not present to any degree in sterile wounds or most tumors. M2/repair doesn't require T cell cytokines - its the default type of activity in resident macs. In fact, Martinez et al referenced above found this. IL-4 stimulated macs very closely resemble resident macs. IFN-g, in contrast had a Very different gene (and product) profile. So, the authors had the data, but didn't come to the right conclusion. M1/M2 propensities are found in T cell deficient  mice. Finally, I originally referred to M1/M2 as a possible continuum. But, M1 kills and M2 can promote proliferation of other cells (in vivo or in vitro). So, if you have something like cancer, 'continuums' are not what will reject your tumor. So, agree about the ratio way of judging the M1/M2 balance, but one needs functions to affect diseases (for better or worse). Relevant papers attached. I hope useful. Charlie Mills

From practical standpoint we use Class II, CD80 and CD86 upregulation as the marker of M1 or Classical Activation (they will be also present on M2 but they will increase the more M1 polarized is the cell) in contrast CD206 and Galectin 3 (MAC2) are good determinant of M2 skewing, again they will go up the more M2 are the macrophages. These antibodies are available in many colors and these stains are really robust. If you have 2 or more groups to compare they are good readouts which group is more M1 and which is more M2. Finally, at PLOS ONE, Jablonski et al (link) identified CD38 and Egr2 as highly selective markers of M1 and M2, respectively. We have not tested this yet in our hands, but their data looked quite convincing.

Regarding Dr. Mills comment, above, I agree that M1 and M2 balance is an oversimplification  as there are so many factors that concurrently affect macrophage polarization in tissues. I am in favor of the concept of modular spectrum model (see paper by Mosser and Edwards), rather than simple M1/M2 division. So we typically see some "flavor combination" of classical, wound healing, and regulatory phenotypes that are readout of all the factors influencing the macrophage status.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0145342

http://www.nature.com/nri/journal/v8/n12/full/nri2448.html

Fantastic thread. Really helped hearing all the opinions.

My question is from the idea that M1 cells drive Th1 immunity but also secrete NO by definition.

How can a cell that kills nearby cells indiscriminately also present antigen to T cells and not kill it?

Are there M1 type cells that inhibit without killing?

Hi Liam:

Good observation about M1/kill and M1/present antigen. It has long been a conundrum for me how both can happen at once. I've talked about this previously as in my 2012 review attached below. This issue has not yet been solved. A recent paper is attached that seems to be able to separate the iNOS and MHC II positive macs (though they call them DC). This may be beginning to provide an explanation. As for a couple of earlier comments about 'continuums' or color wheels of macrophages, some of that comes from analyzing macs from different inflammatory circumstances where there are of course mixtures of macs expressing different M1 or M2-type proclivities. But, what is missing from models like Mosser's and some others is Functions. Continuums, color wheels or heterogeneity doesn't cure or cause diseases, mac functions do. So, without stating what functions are in a color wheel, etc., the models are incomplete. An article is attached where we discuss this - SHIP functions of macs. Good luck. Charlie

Please see this paper where membrane-associated proteins have been identified that can discriminate M1 and M2 macrophages: Article Unique Proteomic Signatures Distinguish Macrophages and Dend...

Please also see our recent study that utilized this information in order to identify macrophage polarity: Article Nanomedicine-driven neuropathic pain relief in a rat model i...

I hope this helps.

Thank you!!