Plg-RKT facilitates migration of proinflammatory monocyte and macrophage subsets.
Membrane-bound plasmin is used by immune cells to degrade extracellular matrices, which facilitates migration. The plasminogen receptor Plg-RKT is expressed by immune cells, including monocytes and macrophages. Among monocytes and macrophages, distinct subsets can be distinguished based on cell surface markers and pathophysiological function. We investigated expression of Plg-RKT by monocyte and macrophage subsets and whether potential differential expression might have functional consequences for cell migration. Proinflammatory CD14++CD16+ human monocytes and Ly6Chigh mouse monocytes expressed the highest levels of Plg-RKT and bound significantly more plasminogen compared with the other respective subsets. Proinflammatory human macrophages, generated by polarization with lipopolysaccharide and interferon-γ, showed significantly higher expression of Plg-RKT compared with alternatively activated macrophages, polarized with interleukin-4 and interleukin-13. Directional migration of proinflammatory monocytes was plasmin dependent and was abolished by anti–Plg-RKT monoclonal antibody, ε-amino-caproic acid, aprotinin, and the aminoterminal fragment of urokinase-type plasminogen activator. In an in vivo peritonitis model, significantly less Ly6Chigh monocyte recruitment was observed in Plg-RKT−/− compared with Plg-RKT+/+ mice. Immunohistochemical analysis of human carotid plaques and adipose tissue showed that proinflammatory macrophages also exhibited high levels of Plg-RKT in vivo. Our data demonstrate higher expression of Plg-RKT on proinflammatory monocyte and macrophage subsets that impacts their migratory capacity.
The zymogen plasminogen is activated by urokinase-type plasminogen activator (uPA) or tissue-type PA to the broad-spectrum protease, plasmin. Plasmin is the major enzyme responsible for fibrin degradation in vivo.1,2 In addition, plasmin regulates extracellular proteolysis, by degrading extracellular matrix components and by activating promatrix metalloproteinases.3 Consequently, plasmin is critically involved in physiological and pathological processes requiring cell migration.4
Plasminogen activation is markedly enhanced when plasminogen is bound to cell surfaces, and plasmin is then protected from inactivation by α2-antiplasmin.5 The novel plasminogen receptor, Plg-RKT, is expressed in membranes of a variety of cells and is colocalized with the uPA receptor, uPAR.6 Monocytes and macrophages express Plg-RKT, and this receptor plays a critical role in plasmin-mediated cell migration, invasion, and recruitment during inflammation.7,8
Human monocytes can be divided into functionally distinct subsets, based on expression of CD14 and CD16, into classical monocytes (CMs; CD14++CD16−), intermediate monocytes (IMs; CD14++CD16+), and nonclassical monocytes (NCMs; CD14+CD16++).9 The CD16+ subsets are associated with pathologies characterized by a chronic inflammatory state, including coronary artery disease and obesity.10-12 Macrophages also exhibit distinct functional heterogeneity and plasticity.13 Exposure to interferon-γ (IFN-γ) and lipopolysaccharide (LPS) primes macrophages toward a proinflammatory phenotype, whereas macrophages exposed to interleukin-4 (IL-4) and IL-13 are linked to tissue repair processes.14
We here investigated whether monocyte and macrophage subsets differentially express Plg-RKT and if potential differential expression would have functional consequences for their plasmin-mediated migratory capacity.
IL-6 and IL-10 were determined by specific enzyme-linked immunosorbent assays according to the manufacturer’s instructions (Abcam, Cambridge, UK).
Studies were approved by the institutions’ ethics committee and performed according to the Declaration of Helsinki. Informed consent was obtained from all participants.
Results and discussion
Three subsets of human monocytes can be distinguished, namely CD14++CD16− CMs, CD14++CD16+ IMs, and CD14+CD16++ NCMs (Figure 1A).9 CD16+ subsets are associated with pathologies characterized by a chronic inflammatory state, such as coronary artery disease, obesity, arthritis, inflammatory diseases of the intestinal tract, or systemic lupus erythematosus.10-12 We have shown that in humans with mild inflammation, the CD16+ subsets exhibit the highest inflammatory potential.15 Here, we found that IMs expressed significantly higher levels of cell surface Plg-RKT compared with CMs and NCMs (Figure 1B). This was paralleled in quantitative polymerase chain reaction (Figure 1C). There was a trend for higher Plg-RKT expression on female monocyte subsets (data not shown). IMs specifically bound more fluorescein isothiocyanate (FITC)-plasminogen than either CMs or NCMs. Plasminogen binding was blocked by ε-aminocaproic acid (EACA) and by anti–Plg-RKT monoclonal antibody (mAb)6,7 on IMs (Figure 1D). Furthermore, IM exhibited plasminogen- and plasmin-dependent directed cell migration that was abolished in the presence of either EACA, aprotinin, the aminoterminal fragment of uPA,16 or anti–Plg-RKT mAb, but not by isotype control antibody. The anti–Plg-RKT mAb had no effect on migration in the absence of plasminogen (Figure 1E). IMs expressed significantly higher levels of uPAR (Figure 1F), consistent with uPAR colocalization with Plg-RKT,6 and with the published key role of uPAR in cell migration.17
In murine blood, the relative amounts of Ly6Chigh and of Ly6Clow cells were 32% ± 9% and 68% ± 9%, respectively (Figure 1G), which contrasts to some published data.18,19 However, according to a recent review, in mice the 2 populations each account for about half of the total monocytes and an increase in Ly6Clow cells mobilized from the marginal pool could be due to blood drawn under stress, such as cardiac puncture under terminal anesthesia, as used by us.20 In murine whole blood samples, the proinflammatory Ly6Chigh subset of monocytes21 expressed significantly higher levels of Plg-RKT than the Ly6Clow subset (Figure 1H). Correspondingly, the Plg-RKT+/+ Ly6Chigh subset bound significantly more plasminogen than the Plg-RKT+/+ Ly6Clow monocyte subset (Figure 1I). Furthermore, both subsets of Plg-RKT+/+ monocytes bound significantly more plasminogen than the respective Plg-RKT−/− monocyte subsets. Taken together, these results also suggest specific binding of plasminogen to Plg-RKT (Figure 1I). Notably, female murine monocytes expressed significantly higher levels of Plg-RKT than male murine monocytes (Figure 1J). Interestingly, circulating plasminogen levels are significantly higher in female compared with male mice8 and in female humans compared with male humans, suggesting the presence of a more efficient cell-associated plasminogen activation activity in female patients.22
To evaluate the in vivo relevance of our results, we used a sterile peritonitis model and determined the number of peritoneal cells in Plg-RKT+/+ and Plg-RKT−/− mice either untreated or 72 hours following thioglycollate injection.7,8 In untreated mice, no significant difference in the number of Ly6Chigh cells in the peritoneal lavage between Plg-RKT+/+ and Plg-RKT−/− mice was seen. When Plg-RKT+/+ mice were injected intraperitoneally with thioglycollate, a 24- and a 26-fold increase in Ly6Chigh cells in female and male Plg-RKT+/+ mice was observed, respectively, after thioglycollate injection compared with cell numbers without thioglycollate treatment. In contrast, this increase in numbers of Ly6Chigh cells after thioglycollate injection was more than halved to only 11- and 10-fold in Plg-RKT−/− female and male mice, respectively (ie, significantly less), namely, 46% and 57% less Ly6Chigh cells were present in the peritoneum of Plg-RKT−/− female and male mice, respectively, compared with the numbers of Ly6Chigh cells in Plg-RKT+/+ mice (Figure 1K-L). Notably, there were no effects of genotype on recruitment of Ly6Clow cells (Figure 1K-L insets), consistent with lower expression of Plg-RKT and lower plasminogen binding capacity of Ly6Clow cells (Figure 1H-I). Peritoneal IL-6 levels did not differ between untreated Plg-RKT+/+ and Plg-RKT−/− mice: 1.3 ± 0.1 pg/mg and 1.3 ± 0.1 pg/mg of peritoneal fluid protein for Plg-RKT+/+ and Plg-RKT−/− mice, respectively (P = n.s., n = 6 per group). IL-6 values were substantially increased in thioglycollate-treated animals, but the response was significantly attenuated in Plg-RKT−/− mice: 9.3 ± 1.7 pg/mg and 2.4 ± 0.8 pg/mg of peritoneal fluid protein for Plg-RKT+/+ and Plg-RKT−/− mice, respectively (P = .0019, n = 10 per group). Peritoneal IL-10 levels were below the detection limit of the assay used in untreated mice irrespective of genotype, but were significantly lower in thioglycollate-treated in Plg-RKT−/− mice as compared with Plg-RKT+/+ mice: 76.6 ± 11.6 pg/mg and 24.2 ± 6.7 pg/mg for thioglycollate-treated Plg-RKT+/+ and Plg-RKT−/− mice, respectively (P = .001, n = 10 per group).
The relative percentages of Ly6Chigh cells and of Ly6Clow cells positive for the macrophage marker, F4/80, were 89% and 92%, respectively (Figure 1M). Transcriptional profiling has shown that Ly6Clow monocytes extravasated into the peritoneum differentiate toward alternatively activated M2-like macrophages, whereas Ly6Chigh monocytes differentiate into inflammatory M1-like macrophages.23 This suggests that this differentiation process is a continuum from monocytes transitioning into the respective macrophage subsets and also suggests that the cells present in the peritoneal lavage following thioglycollate injection are in this transitioning process from monocytes to macrophages.
Monocytes are key effectors in the innate immune system and are differentiated to macrophages that are crucial for host defense and wound healing. Macrophages also play a key role in the pathogenesis of chronic inflammatory pathologies,24,25 where they are recruited to sites of inflammation and are capable of polarizing into functional subsets depending on environmental cues. We showed recently that matrix degradation depends on membrane-bound proteases and the expression of uPAR on proinflammatory macrophages and that alternatively activated macrophages are rendered proteolytically quiescent through their high expression of PAI-1.26 Therefore, we investigated whether differences in the expression of Plg-RKT might also exist in differentially polarized human macrophages and if such differential expression might impact the migratory capacity of these cells.
Proinflammatory macrophages expressed significantly more cell surface Plg-RKT and Plg-RKT messenger RNA (mRNA) compared with alternatively activated macrophages or unpolarized macrophages (Figure 2A-B).
Furthermore, proinflammatory macrophages exhibited plasminogen-dependent cell migration that was impaired by treatment with anti–Plg-RKT mAb, whereas significant plasminogen-dependent migration was not exhibited by alternatively activated or unpolarized macrophages (Figure 2C).
To further evaluate the relevance of our in vitro findings, under in vivo inflammatory conditions, sections of human carotid artery plaques obtained from patients undergoing atherectomy and sections of adipose tissue obtained from morbidly obese patients undergoing gastric bypass surgery were stained with antibodies against the inflammatory macrophage marker CD80, Plg-RKT, and plasminogen. CD80high proinflammatory macrophages expressed significantly more Plg-RKT that was colocalized with plasminogen compared with CD80low macrophages in human carotid artery plaques and obese adipose tissue (Figure 2D-E). When these sections were stained with the markers for alternatively activated macrophages CD206 or CD209, no association between these markers and Plg-RKT expression was found (data not shown).
We here provide the first evidence that Plg-RKT is differentially expressed on monocyte and macrophage subsets, with inflammatory IMs and proinflammatory macrophages expressing the highest levels of this plasminogen receptor. Differential expression of Plg-RKT impacted the plasminogen- and plasmin-dependent migratory capacity of these cells. Proinflammatory macrophages expressed significantly more Plg-RKT in human atherosclerotic plaques and obese adipose tissue than macrophages expressing low levels of the inflammatory marker CD80. Finally, recruitment of Ly6Chigh monocytes was impaired in Plg-RKT–deficient mice, concomitantly with impairment in levels of the representative cytokines, the proinflammatory cytokine IL-6, and IL-10, which has anti-inflammatory properties, but is induced frequently in inflammatory situations. 27,28 Our results suggest that higher expression of Plg-RKT on proinflammatory monocyte and macrophage subsets contributes significantly to their migration and recruitment into areas of inflammation and their regulation of the inflammatory response in pathologies characterized by an inflammatory state.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.
The authors thank the Core Facilities for Flow Cytometry at Scripps Research Institute, La Jolla, CA, and the Core Facilities of the Medical University of Vienna, Vienna, Austria for their expertise in cell sorting.
This project was funded by the Austrian Science Fund (FWF; SFB-54) (J.W. and W.S.S.), by National Institutes of Health, National Heart, Lung, and Blood Institute grant HL 081046 (L.A.M.), and by Merit Review Award 5I01BX003933 from the US Department of Veterans Affairs (R.J.P.). B.T. received a Travel Grant from the International Society of Fibrinolysis and Proteolysis.
Contribution: B.T. performed the majority of the experiments; J.B. and N.B. assisted in cell culture experiments; A.P. scanned the tissue slides and helped with the analysis; G.R.-K., C.K., and M.P. provided and characterized the adipose tissue samples; S.S., S.D., and I.H. provided and characterized the carotid artery plaques; M.B.F. provided the leukapheresis chambers and critically revised the manuscript; the experiments were designed and analyzed by B.T., P.J.H., W.S.S., L.A.M., R.J.P., and J.W.; and the manuscript was written by B.T. and critically revised and modified by P.J.H., K.H., R.J.P., W.S.S., J.W., and L.A.M.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Lindsey A. Miles, Department of Molecular Medicine, The Scripps Research Institute, 10550 N. Torrey Pines Rd, SP30-3020, La Jolla, CA 92037; e-mail: email@example.com.
For original data please contact firstname.lastname@example.org.
L.A.M. and J.W. contributed equally to this study.