Selective inhibition and augmentation of alternative macrophage activation by progesterone

Abstract

Progesterone is the female sex hormone necessary for the maintenance of pregnancy, and is known to modulate macrophage activation. However, studies have concentrated exclusively on the ability of progesterone to negatively regulate the innate and classical pathways of activation, associated with nitric oxide (NO) and interleukin (IL)-12 production. Our aim was to examine the ability of progesterone to modulate alternative macrophage activation. Bone marrow cells were isolated and differentiated from male BALB/c mice, exposed to varying concentrations of progesterone and stimulated with lipopolysaccharide (LPS) (innate activation), IL-4 (alternative activation) or LPS in combination with IL-4. Our present study demonstrates that progesterone not only down-regulates inducible nitric oxide synthase 2 (iNOS) activity in macrophages but also arginase activity, in a dose-dependent manner, independent of the stimuli, whether it is induced by LPS (innate activation), IL-4 (alternative activation) or LPS in combination with IL-4. The ability of progesterone to down-modulate IL-4-induced cell surface expression of the mannose receptor further suggested a negative regulation of alternative macrophage activation by this hormone. Analysis of mRNA expression, by quantitative reverse transcription–polymerase chain reaction (qRT–PCR), of genes associated with innate and alternative macrophage activation revealed that progesterone down-regulated LPS-induced macrophage nos2, argI and p40 (IL-12/IL-23) expression and IL-4-induced argI, mrc-1 and fizz1 expression. However, progesterone up-regulated IL-4-induced macrophage expression of ym1, while dectin-1 expression remained unaltered. Following treatment of macrophages with LPS and IL-4 in combination a similar pattern was observed, with the exception that progesterone up-regulated macrophage expression of fizz1 as well as ym1 and did not modify mrc-1 expression. Our data demonstrate for the first time that a hormone has the ability to regulate selectively the expression of different genes associated with alternative macrophage activation.

Introduction

Macrophages have been described as having a spectrum of activation states,¹ and the plasticity and versatility of macrophages is demonstrated by their ability to adapt their functions as the cytokine environment to which they are exposed changes.² Nevertheless, it is generally agreed that macrophage activation can be generally described as innate, classical or alternative.³

The innate activation of macrophages is the early response to a microbial stimulus [generally through Toll-like receptor (TLR) ligation] and leads to the production of a range of proinflammatory cytokines and mediators such as tumour necrosis factor (TNF)-α, interleukin (IL)-6, IL-12 and nitric oxide (NO). By comparison, classical activation involves the priming of macrophages with interferon (IFN)-γ prior to a second TNF-α-inducing signal such as lipopolysacchride (LPS).⁴ Classical activation of macrophages (M1 macrophages) is associated with driving the type-1 immune mechanisms, necessary in the defence against a range of pathogens such as Toxoplasma gondii⁵ and Listeria monocytogenes.⁶ Alternative activation was first described by Stein et al. (1992)⁷ and describes the macrophage phenotype (M2 macrophages) induced upon stimulation with T helper type 2 (Th2)-associated cytokines, such as IL-4 and IL-13. Such activation has been associated with wound-healing-type responses through its ability to induce activity of the arginase enzyme.^8–10 However, arginase is now known to be also induced following innate stimulation, but inhibited by IFN-γ.¹¹ Many markers of murine alternative activation have now been defined that are not induced by innate stimulation and include the mannose receptor (MR, encoded for by mrc-1),⁷ dectin-1,¹² fizz1 (found in inflammatory zone 1/RELM-α)¹³ and ym1.^3,14,15

MR and dectin-1 are macrophage receptors associated with innate immunity. However, these receptors have also been described as markers of alternative macrophage activation, preferentially up-regulated by exposure to IL-4,^7,12,16 and are involved in the innate recognition of fungal and microbial pathogens. In mice, the ym1 gene encodes a 45 000 molecular weight (MW) single-chain polypeptide that is a member of the chitinase family, thought to interact with components of the extracellular matrix.¹⁷ Fizz1 is a 9400 MV resistin-like secreted protein that was first identified in an experimentally induced asthma model, and was shown to be able to inhibit nerve growth factor.¹⁸ Fizz1 and ym1 have been identified as good markers of alternative activation, with their expression up-regulated by IL-4 and antagonized by IFN-γ/LPS.^3,11,15

Macrophages are present within the female reproductive tract and are found in high numbers at the maternal–fetal interface,¹⁹ where they may play an important role in the clearance of apoptotic cells.²⁰ The Th2 dominance of the microenvironment surrounding the maternal–fetal interface,²¹ thought to be established by the constitutive secretion of IL-3, IL-4, IL-5 and IL-10 by cells of the fetal–placental unit during all three trimesters of pregnancy, would be expected to lead to the alternative activation of local macrophages. Indeed, decidual macrophages in human pregnancy exhibit a suppressive phenotype^22–24 as well as characteristics associated with alternatively activated macrophages, such as increased expression of the mannose receptor^25,26 and stabilin-1 (MS-1), which has been suggested to be involved in the induction of anti-inflammatory pathways.²⁷

Progesterone, the female sex hormone essential for the maintenance of pregnancy, has been demonstrated previously to down-regulate the inflammatory processes associated with innate and classical immunity, and in particular nitric oxide synthase 2 (NOS2) activity.^28–30 However, the effect of progesterone on alternative macrophage activation and associated arginase activity,^3,31–34 which uses the same substrate (l-arginine) as NOS2, has yet to be investigated. Taking this information into consideration, we hypothesized that progesterone would promote the differentiation of macrophages towards an alternatively activated phenotype, and undertook studies to address this.

Contrary to our hypothesis, our studies demonstrate the ability of progesterone to not only down-modulate NOS2, but also for the first time down-modulate arginase enzyme activity in murine macrophages in a dose-dependent manner. In addition, we present the novel finding that progesterone selectively reduces transcription of a number of genes associated with alternative macrophage activation such as mrc1, but increases transcription of others such as ym1. Consequently, these results demonstrate plasticity in alternative macrophage activation under the influence of progesterone with potential significant consequences for pregnancy.

Materials and methods

Culture of bone marrow-derived (BMD) macrophages

BMD macrophages were prepared from male mice in order to prevent cells being influenced by the cyclical hormone environment present in female mice. Bone marrow cells were flushed from the femurs of 8-week-old male BALB/c mice (maintained at the University of Strathclyde), as described previously.³⁰ The BMD precursor cells were grown for 8 days in Petri dishes containing Dulbecco's modified Eagle's medium (DMEM) (Gibco-BRL, Paisley, UK) that had been supplemented with 30% (v/v) l-cell conditioned medium, 20% (v/v) heat-inactivated fetal calf serum (FCS; Sigma-Aldrich, St. Louis, MO), 2 mm l-glutamine (Cambrex BioScience, Veniers, Belgium), 100 U/ml penicillin (Cambrex BioScience) and 100 μg/ml streptomycin (Cambrex BioScience). l-cell conditioned medium provides a source of macrophage colony-stimulating factor (M-CSF) and granulocyte–macrophage (GM)-CSF and was derived from the supernatants of confluent L929 cells.

Treatment of cells

Cells were then harvested and plated out at 10⁶ cell/well in 24-well plates in RPMI-1640 medium (Gibco-BRL) supplemented with 10% (v/v) heat inactivated FCS (Sigma-Aldrich), 2 mm l-glutamine (Cambrex BioScience), 100 U/ml penicillin (Cambrex BioScience) and 100 μg/ml streptomycin (Cambrex BioScience). Progesterone (Sigma, Poole, UK) was dissolved in chloroform (Sigma) at a concentration of 50 mg/ml (160 mm). Dilutions were performed in RPMI-1640 complete medium. Solvent vehicle controls using chloroform was prepared by the same method. Solutions were sterilized using a 0·2 μm MilltexTM filter (Millipore Corporation, Billerca, MA). Cells were stimulated with mouse recombinant IL-4 (100 U/ml; BD PharmingenTM, San Diego, CA) or LPS from Escherichia coli 055:B5 (200 ng/ml; Sigma) for a defined period of time. Cells were then harvested for quantitative real-time polymerase chain reaction (PCR) analysis or for use in arginase assays. Supernatants were collected for analysis of nitrite levels.

Preparation of cDNA

Cells were harvested in 1 ml of Trizol™ reagent (Invitrogen, Paisley, UK) and total RNA isolated according to the manufacturer's instructions. Two μg of total RNA and 300 ng random primers (Promega, Madison, WI) were incubated for 5 min at 65°. The mixture was then incubated for 10 min at room temperature before the addition of 2 μl of AffinityScript™ reverse transcriptase (RT) buffer (Stratagene, Agilent Technologies, Santa Clara, CA), 10 mm dithiothreitol (DTT), 4 mm dNTP mix (Promega) and 1 μl of AffinityScript™ multiple RT (Stratagene). Samples were then incubated for 10 min at 25°, 1 hr at 50° and finally 15 min at 70°.

Quantitative reverse transcription real-time PCR (qRT–PCR) analysis

qRT–PCR experiments were performed and analysed using the Stratagene Mx3000p system. Each reaction contained 10 μl of SYBR® Green JumpStartTM Taq ReadyMixTM (Sigma), 25 pmol of both forward and reverse primers, 8 μl of molecular grade water (Sigma) and 1 μl of cDNA template. Primer nucleotide sequences and associated optimal annealing temperatures are shown in Table 1. Reactions were performed under the following conditions: one cycle at 95° for 10 min and 40 cycles of 1 min at 95°, 1 min at the appropriate annealing temperature and 1 min at 72°. A dissociation curve was generated to ensure amplification of a specific product. This was achieved by subjecting samples to 1 min at 95°, 30 seconds at 55° and 30 seconds at 95°. Relative gene mRNA expression levels were calculated by using an included standard curve for each individual gene and normalized to the housekeeping gene. The maximum gene expression for each experiment was assigned as 100% and all other treatments calculated in comparison to this, similar to previous studies.¹¹ Data are presented as the mean expression ± standard error (SE) for n = 3.

Table 1.

Oligonucleotide sequences of the primers used for analysis of gene expression by real-time polymerase chain reaction (PCR)¹¹

Primer name	Sequence (5′ to 3′)	Optimal annealing temperature (°)
argI	Forward TGACATCAACACTCCCCTGACAAC	61
	Reverse GCCTTTTCTTCCTTCCCAGCAG
dectin-1	Forward GGAATCCTGTGCTTTGTGGTAGTAG	64
	Reverse GGAAGGCAAGACTGAGAAAAACCTC
fizz1	Forward ACCTTTCCTGAGATTCTGCCCC	64
	Reverse CAGTGGTCCAGTCAACGAGTAAGC
p40(IL-12/IL-23)	Forward CCTGGTTTGCCATCGTTTTG	62
	Reverse TCAGAGTCTCGCCTCCTTTGTG
mrc-1	Forward TCTTTTACGAGAAGTTGGGGTCAG	64
	Reverse ATCATTCCGTTCACCAGAGGG
nos2	Forward GGTCTTTGACGCTCGGAACTGTAG	64
	Reverse CACAACTGGGTGAACTCCAAGGTG
Tbp	Forward AACAGCAGCAGCAACAACAGCAGG	64
	Reverse TGATAGGGGTCATAGGAGTCATTGG
ym1	Forward GGCTACACTGGAGAAAATAGTCCCC	64
	Reverse CCAACCCACTCATTACCCTGATAG

Arginase assays

Murine BMD macrophage arginase activity was determined using an assay based on a reaction with α-isonitrosopropiophenon, as described previously.³⁵ Arginase activity was calculated using an internal standard curve generated from known quantities of urea. One unit (U) of arginase activity was defined as the enzyme activity that catalyses the production of 1 μmol urea/min.

Greiss assays for determination of NO production

Quantification of nitrite accumulation, using a method similar to that described previously,³⁶ was used as a measure of NOS2 activity. Greiss reagent, consisting of equal volumes of 2% sulphanilamide in 5% H₃PO₄ and 0·2% naphylene diamine HCl in water, was added as a volume of 50 μl to a well along with 50 μl of cell supernatant. After incubation for 10 min at room temperature in darkness, absorbance was read at 540 nm. Nitrite production was determined by comparison to a standard curve generated using NaNO2.

Flow cytometry to analyse surface MR expression

After incubation, cells were washed in fluorescence activated cell sorter (FACS) buffer [0·375 g ethylenediamine tetraacetic acid (EDTA), 2·5 g bovine serum albumin (BSA), 500 ml of 1 × concentration phosphate-buffered saline (PBS)] and resuspended in Fc block [purified anti-mouse CD16/CD32 (FcγIII/IIreceptor) IgG2b from BD Pharmingen™]. Samples were incubated at 4° for 20 min. Cells were stained with the pan macrophage marker F4/80 [phycoerythrin (PE)-conjugated rat anti-mouse F4/80 IgG2a (eBioscience, San Diego, CA)] as described previously³⁷ and rat anti-mouse mannose receptor (MR) immunoglobulin G2a (IgG2a) (MR5D3). This primary antibody was developed and kindly provided by Professor Siamon Gordon (University of Oxford).³⁸ Samples were incubated at 4° in the dark for 20 min before washing and incubation with secondary antibody [fluorescein isothiocyanate (FITC)-conjugated goat anti-rat IgG2a] (STAR113F; AbDSerotec, Kidlington, UK). Cells were centrifuged for 5 min at 200 g, supernatant removed and pellet resuspended in 300 μl of FACS flow (BD Biosciences, San Diego, CA). Isotype controls were set up to gate for background staining using the appropriate isotype-matched antibody [rat IgG2a negative control (MCA1212) from AbDSerotec]. The FACSCanto™ flow cytometer from BD Biosciences was used to perform fluorescence analysis. Compensation controls were set up using CaliBRITE Beads from BD Biosciences. F4/80⁺ cells were gated on dot-plots and data analysed to show mean fluorescence intensity of these cells using facsdiva software (BD Biosciences).

Statistical analysis

All values are mean ± standard error (SE) of three separate experiments. Statistically significant of differences were calculated using the Mann–Whitney U-test with GraphPad Prism® version 5.0 software. P < 0·05 was accepted as significant.

Results

Progesterone down-regulates arginase activity and nitrite production in a dose-dependent manner

Bone marrow-derived macrophages were treated with a range of concentrations of progesterone and then stimulated with IL-4, LPS or a combination of both, and arginase and nitrite activity measured (Fig. 1). As shown in a previous study from our laboratory,¹¹ IL-4 stimulation induced a small but significant increase in arginase activity (Fig. 1a). Treatment of cells with > 15·625 μm progesterone prior to IL-4 stimulation significantly reduced arginase activity (P < 0·05). Nitrite levels within culture supernatants were undetectable (data not shown). LPS induced a twofold rise in macrophage arginase activity (Fig. 1b), which was significantly (P < 0·05) reduced by the presence of 15·625 μm progesterone. Addition of 31·25 μm progesterone resulted in a further small, but significant, reduction in macrophage arginase activity. Treatment of BMD macrophages with both IL-4 and LPS for 48 hr resulted in a significant increase in arginase activity (Fig. 1c) to a much greater extent than with either stimulant alone, as described previously.¹¹ IL-4 + LPS-induced arginase activity was reduced in a dose-dependent manner by preincubation with progesterone. Culture supernatants from experiments reported in Fig. 1b,c were also analysed for nitrite (Fig. 1d,e). Progesterone induced a significant dose-dependent reduction in macrophage nitrite production induced by either LPS (Fig. 1d) or IL-4 + LPS (Fig. 1e). Arginase activity and nitrite production in stimulated cells were not altered significantly by treatment with the hormone solvent chloroform (data not shown): data were comparable to cells treated with medium alone. The observed effects of the hormone progesterone, or its solvent chloroform, were not due to reductions in cell viability, as determined by measurements of mitochondrial activity by alamarBlue™ (data not shown).

Figure 1.

Progesterone down-modulates arginase activity and nitrite production by bone marrow-derived (BMD) macrophages. Arginase activity (a, b, c) and nitric oxide (NO) production (d, e) in progesterone-treated BMD macrophages that have been subsequently stimulated with interleukin (IL)-4 (100 U/ml) for 48 hr (a), lipopolysaccharide (LPS) (200 ng/ml) for 48 hr (b, d) or both IL-4 (100 U/ml) and LPS (200 ng/ml) for 48 hr (c, e). Open bars represent samples that were treated with progesterone at the stated concentration only. Closed bars represent those samples that were preincubated with progesterone for 16 hr prior to the addition of IL-4, LPS or IL-4 + LPS. Solvent vehicle (0·01–0·04% chloroform) did not result in any significant differences from the medium control (data not shown). Results show mean ± standard error (SE) of n = 3 and are representative of three separate experiments. *P < 0·05 in comparison with stimulated control.

Progesterone down-regulates argI and nos2 gene expression in BMD macrophages

BMD macrophages were treated with progesterone and stimulated with IL-4, LPS or both IL-4 and LPS and real-time PCR carried out to analyse any changes in argI gene expression after 6 or 24 hr of stimulation (Fig. 2). In unstimulated cells, argI and nos2 mRNA levels did not change over the 24-hr period, and both progesterone and chloroform had no significant effect on these levels. IL-4 stimulation induced an increase in macrophage argI expression after 6 hr (Fig. 2a), which was maintained at 24 hr (Fig. 2b). By comparison, LPS induced the maximum levels of argI expression only after 24 hr of stimulation (Fig. 2c,d). Stimulation with both IL-4 and LPS induced arg I expression after 6 hr, but to a lesser extent than with IL-4 alone (Fig. 2e). Progesterone significantly reduced expression of argI mRNA transcripts in IL-4, LPS and IL-4 and LPS-treated macrophages at the 24-hr time point (Fig. 2b,d,f).

Figure 2.

The effect of progesterone on argI and nos2 mRNA expression in bone marrow-derived (BMD) macrophages. Cells were left untreated, incubated with 0·04% chloroform (solvent control) or 62·5 μm progesterone for 16 hr prior to the addition of 100 U/ml interleukin 4 (IL-4) (a, b), 200 ng/ml lipopolysaccharide (LPS) (c, d, g, h) or 100 U/ml IL-4 and 200 ng/ml LPS (e, f, i, j) for 6 hr (a, c, e, g, i) or 24 hr (b, d, f, h, j). Control cultures were not treated with IL-4 and/or LPS. Expression levels of argI and nos2 mRNA transcripts were normalized to the housekeeping gene Tbp for each experimental run. The maximum gene expression for each run was designated 100% and all experiments calculated in comparison to this. Data are represented as the mean expression ± standard error (SE) of n = 3. *P < 0·05 in comparison with stimulated controls not treated with hormone or solvent for the same length of time.

IL-4 does not induce nos2 or p40 (IL-23/IL-23) mRNA transcripts, and treatment with progesterone (or the solvent chloroform) did not modulate the expression of these genes (data not shown). LPS-induced nos2 expression peaked at 6 hr (Fig. 2g) and was significantly reduced by 24 hr (Fig. 2h). Progesterone significantly reduced LPS-induced macrophage nos2 expression at both 6 and 24 hr. Progesterone had no effect on the nos2 mRNA levels after 6 hr of treatment with both IL-4 and LPS (Fig. 2i). However, the solvent control induced a significant increase. By the 24-hr time-point progesterone had down-regulated nos2 gene expression and chloroform had no effect (Fig. 2j).

Progesterone down-regulates mrc-1 and p40 gene expression

Studies in our laboratory have shown that arginase is induced by both innate (LPS) and alternative (IL-4) stimulation in a time-dependent manner.¹¹ Consequently, we examined the ability of progesterone to modulate mrc-1- and p40-specific markers of alternative and innate immunity, respectively. Progesterone caused a reduction in the expression of the alternative macrophage activation-associated marker mrc-1 after 24 hr of stimulation with IL-4 (Fig. 3b). After stimulation with both IL-4 and LPS for 6 hr, levels of mrc-1 (Fig. 3c) mRNA transcripts were significantly increased. These levels were maintained after 24 hr (Fig. 3d). Progesterone had no effect on the expression of IL-4–LPS-induced mrc-1 at either time-point.

Figure 3.

The effect of progesterone on mrc-1 and p40 [interleukin 12 (IL-12)/IL-23] mRNA expression in bone marrow-derived (BMD) macrophages. Cells were left untreated, incubated with 0·04% chloroform (solvent control) or 62·5 μm progesterone for 16 hr prior to the addition of 100 U/ml IL-4 (a, b), 200 ng/ml lipopolysaccharide (LPS) (e, f) or 100 U/ml IL-4 and 200 ng/ml LPS (c, d, g, h) for 6 hr (a, c, e, g) or 24 hr (b, d, f, h). Control cultures were not treated with IL-4, LPS or LPS and IL-4. Expression levels of mrc-1 and p40 (IL-12/IL-23) mRNA transcripts were normalized to the housekeeping gene Tbp for each experimental run. The maximum gene expression for each run was designated 100% and all experiments calculated in comparison to this. Data are represented as the mean expression ± standard error (SE) of n = 3. *P < 0·05 in comparison with stimulated controls not treated with hormone or solvent for the same length of time.

Progesterone and chloroform (used as a solvent) had no effect on the expression of p40 (IL-12/IL-23) mRNA transcripts in unstimulated cells (Fig. 3e–h). After 6 hr of LPS stimulation, p40 (IL-12/IL-23) transcripts were significantly up-regulated (Fig. 3e), but reduced by the 24-hr time-point (Fig. 3f). Progesterone reduced the ability of p40 (IL-12/IL-23) to be induced after 6 hr of LPS exposure, and this was reduced further by the 24-hr time-point. By comparison, progesterone had no effect on p40 (IL-12/IL23) transcripts in IL-4 and LPS-stimulated cells after 6 hr of treatment (Fig. 3g); however, by 24 hr of treatment with progesterone, transcript levels were significantly reduced (Fig. 3h).

Progesterone can modulate surface expression of the mannose receptor

Flow cytometry was carried out to determine the ability of progesterone to modulate the surface expression of the MR (Fig. 4). IL-4 was used to induce the alternative macrophage activation state and induced a significant increase in surface MR expression. Progesterone also induced an increase in surface MR expression in unstimulated cells. However, progesterone significantly reduced IL-4 induced MR expression to levels that were comparable with the unstimulated controls and in agreement with reduced mrc-1 mRNA expression (Fig. 3b).

Figure 4.

Influence of progesterone on interleukin 4 (IL-4) induced expression of mannose receptor (MR) in bone marrow-derived (BMD) macrophages. Cells were preincubated with medium (open bar), 62·5 μm progesterone (black bar), 0·04% chloroform (grey bar) for 16 hr followed by a 48-hr stimulation with IL-4 (100 U/ml). Unstimulated cells received medium. Results are the mean fluorescence intensity ± standard error (SE) of n = 3 samples and are representative of at least two experimental runs. *P < 0·05 compared with medium controls (open bar).

Progesterone differentially modulates other markers of alternative macrophage activation

Progesterone had no effect on IL-4-induced dectin-1 expression at either the 6-hr (Fig. 5a) or 24-hr (Fig. 5b) time-points. After stimulation with both IL-4 and LPS for 6 hr, dectin-1 (Fig. 5c) mRNA transcripts were significantly increased. These levels were maintained after 24 hr (Fig. 5d). Progesterone had no effect on the expression of this gene at either time-point.

Figure 5.

The effect of progesterone on dectin-1 mRNA expression in bone marrow-derived (BMD) macrophages. Cells were left untreated, incubated with 0·04% chloroform (solvent control) or 62·5 μm progesterone for 16 hr prior to the addition of 100 U/ml interleukin 4 (IL-4) (a, b) or 100 U/ml IL-4 and 200 ng/ml lipopolysaccharide (LPS) (c, d) for 6 hr (a, c) or 24 hr (b, d). Control cultures were not treated with IL-4 or LPS and IL-4. Expression levels of dectin-1 mRNA transcripts were normalized to the housekeeping gene Tbp for each experimental run. The maximum gene expression for each run was designated 100% and all experiments calculated in comparison to this. Data are represented as the mean expression ± standard error (SE) of n = 3. *P < 0·05 in comparison with stimulated controls not treated with hormone or solvent for the same length of time.

Macrophage fizz1 (Fig. 6) and ym1 (Fig. 7) mRNA transcripts were also analysed as markers of alternative macrophage activation. While progesterone alone had no effect on background levels of fizz1 mRNA at either time-point (Fig. 6a,b), a significant increase in ym1 mRNA levels after 6 hr and 24 hr was observed (Fig. 7a,b). While preincubation of the cells with progesterone significantly reduced the expression of fizz1 induced after 24 hr of treatment with IL-4 (Fig. 6b) progesterone significantly enhanced macrophage fizz1 mRNA expression induced by stimulation with both IL-4 and LPS for 24 hr (Fig. 6d). Progesterone significantly enhanced early (6 hr) but not late (24 hr) expression of ym1 mRNA transcripts after stimulation with IL-4 (Fig. 7a,b) or with both IL-4 and LPS (Fig. 7c,d).

Figure 6.

The effect of progesterone on fizz1 mRNA expression in bone marrow-derived (BMD) macrophages. Cells were left untreated, incubated with 0·04% chloroform (solvent control) or 62·5 μm progesterone for 16 hr prior to the addition of 100 U/ml interleukin 4 (IL-4) (a, b) or 100 U/ml IL-4 and 200 ng/ml lipopolysaccharide (LPS) (c, d) for 6 hr (a, c) or 24 hr (b, d). Control cultures were not treated with IL-4 or LPS and IL-4. Expression levels of fizz1 mRNA transcripts were normalized to the housekeeping gene Tbp for each experimental run. The maximum gene expression for each run was designated 100% and all experiments calculated in comparison to this. Data are represented as the mean expression ± standard error (SE) of n = 3. *P < 0·05 in comparison with stimulated controls not treated with hormone or solvent for the same length of time.

Figure 7.

The effect of progesterone on ym1 mRNA expression in bone marrow-derived (BMD) macrophages. Cells were left untreated, incubated with 0·04% chloroform (solvent control) or 62·5 μm progesterone for 16 hr prior to the addition of 100 U/ml interleukin 4 (IL-4) (a, b) or 100 U/ml IL-4 and 200 ng/ml lipopolysaccharide (LPS) (c, d) for 6 hr (a, c) or 24 hr (b, d). Control cultures were not treated with IL-4 or LPS and IL-4. Expression levels of ym1 mRNA transcripts were normalised to the housekeeping gene Tbp for each experimental run. The maximum gene expression for each run was designated 100% and all experiments calculated in comparison to this. Data are represented as the mean expression ± standard error (SE) of n = 3. *P < 0·05 in comparison with stimulated controls not treated with hormone or solvent for the same length of time.

Discussion

Macrophages are now recognized as being a versatile and plastic population of cells,^1,11,39–42 able to respond to the local microenvironment through altering receptor expression and cytokine production to adapt their function. Despite classification of macrophage phenotypes into distinct categories, such as classically activated (M1), innately activated or alternatively activated (M2), it is now recognized that it is difficult to strictly compartmentalize the different functions of these cells. Our present studies reinforce further this concept of plasticity, with the pregnancy-associated hormone progesterone demonstrated to be able to modulate differentially the so-called defined ‘markers’ of alternatively activated macrophages. Not only does progesterone down-modulate NOS2, but also arginase I enzyme activity in murine macrophages in a dose-dependent manner. In addition, we demonstrate that progesterone can selectively reduce the expression of transcript levels for some genes associated with alternative macrophage activation such as mrc1, but increase expression levels of transcripts for others, such as ym1. The consequences of these progesterone influences for a healthy pregnancy need to be addressed.

Macrophages represent a significant population of immune cells within the non-pregnant and pregnant uterus of both mice and humans.^43–53 In addition, macrophages are localized to areas of apoptosis and are implicated in the remodelling of tissue at the maternal–fetal interface.²⁰ The Th2 environment associated with pregnancy would favour the development of macrophages towards the alternatively activated phenotype. Indeed, Cupurdija and colleagues²⁵ have demonstrated that decidual macrophages of early human pregnancy display characteristics associated with the alternative activation phenotype. However, as monocytes are recruited from the circulation into the uterus during the later stages of pregnancy in humans under the apparent influence of trophoblast cells,⁵⁴ they are programmed towards a decidual phenotype⁵⁵ that are refractory to IFN-γ influences, but at the same time display characteristics of classically activated, alternatively activated and immunosuppressive/deactivated macrophages. Our data would suggest that progesterone could be playing a significant role in influencing the expression of such a mixture of phenotypes. However, the influence of other pregnancy hormones, such as oestrogen and prolactin, will also be taken into consideration.

Down-regulation of the classical pathway of macrophage activation during pregnancy ensures that proinflammatory cytokines and NO, that would be detrimental to the successful continuation of pregnancy, are not released.⁵⁶ This is supported by our data and previously published work,³⁰ showing that progesterone acts to down-modulate nitrite production in BMD macrophages as well as down-regulating the expression of nos2 mRNA transcripts and p40 (IL-12/IL-23) mRNA transcripts. To the best of our knowledge, no in vitro studies have considered effects of progesterone on the activity of macrophage arginase, which competes with NOS2 for the common substrate l-arginine. Previously, our group has demonstrated that at the concentrations used, IL-4 is a weaker stimulus of arginase I mRNA expression and activity than stimulation with LPS or a combination of IL-4 and LPS11. Surprisingly, our data demonstrate that progesterone not only suppresses classical activation, but also argI mRNA transcript levels and consequently arginase activity, irrespective of the method of stimulation.

l-arginine metabolism and arginase have been shown to affect the success of pregnancy and this provides a possible evolutionary advantageous reason for progesterone-mediated arginase modulation. It has been shown, for example, that depletion of l-arginine by arginase down-regulates the CD3ζ chain of the T cell receptor (TCR) resulting in the induction of T cell hyporesponsiveness favouring fetal survival.⁵⁷ Significantly, cervical ripening occurs immediately after a decline in progesterone synthesis which allows the recruitment of immune cells, including macrophages, to promote remodelling.⁵⁸ During this time, alternatively activated macrophage populations have been found to predominate which would serve to promote post-partum tissue repair via arginase and ym1-associated activities as noted in other tissues.^59–63 Our data would indicate that progesterone is one possible candidate for the negative regulation of arginase activity pre-labour.

In addition to elevated arginase I mRNA expression and activity, alternative macrophage activation has been associated with high expression of the pattern recognition receptors MR7 and dectin-112 and mRNA for the ym1^3,14,15 and fizz1¹³ genes. We have found that in the presence of high levels of progesterone, IL-4 induced mrc-1 and fizz1 expression is reduced, while ym1 mRNA expression is increased further and dectin-1 mRNA expression levels remain unchanged. The ability of progesterone to modulate expression of each of these markers may be related to their individual functions.

MR⁺ decidual macrophages have been identified at the human maternal–fetal interface²⁶ and microarray analysis of human decidual macrophages showed that mrc-1 was up-regulated in comparison with blood-derived mononuclear cells, along with alternative macrophage activation-associated CC chemokine 1 (AMAC-1) (CCL18),⁶⁴ a human marker of alternative macrophage activation. In addition, studies using human decidual macrophages have shown that tumour-associated glycoprotein-72 (TAG-72) is a progesterone-dependent molecule, which may be the natural ligand for the MR within the human decidua.²⁶ Ligand binding of the MR leads to an inhibition of Th1 polari zed immune responses⁶⁵ and so would favour the successful continuation of pregnancy. In the absence of IL-4, progesterone significantly increased the surface expression of the MR on macrophages, and would thus promote the probability of binding of TAG-72 to the MR to inhibit Th1 responses, which are widely viewed as detrimental to pregnancy. By comparison, when stimulated with IL-4, both mrc-1 expression and surface MR expression were reduced by progesterone. Under these circumstances, progesterone may function to regulate the levels of macrophage surface MR in order to limit the extent of Th1 inhibition. In contrast, the expression of dectin-1 remained unaltered in the presence of progesterone, regardless of whether the stimulus was simply IL-4 or IL-4 in combination with LPS. The dectin-1 receptor binds fungal-derived β-glucans^66,67 on fungal pathogens, meaning that this aspect of immunity to these pathogens should be unaffected by the presence of high levels of progesterone.

An interesting finding in these studies is the ability of macrophage fizz1 expression to be regulated differentially in the presence of progesterone, dependent upon the type of stimulus; progesterone reduced IL-4 induced fizz1 expression, but increased IL-4⁺ LPS-induced fizz1 expression; fizz1 is a resistin-like secreted protein also known as Relmα. Despite being classified as a marker of alternative macrophage activation¹³ and induced by the Th2-associated cytokines IL-4 and IL-13, recent studies employing the use of fizz1-deficient mice (Retnla^–/– mice) have demonstrated that this molecule acts to negatively regulate Th2-dependent inflammatory responses.^68,69 In the presence of both IL-4 and LPS, which would equate to a type-2 inflammatory response, fizz1 expression by BM macrophages was elevated by progesterone. Generally, Th1 responses are considered detrimental to pregnancy and Th2 responses beneficial (reviewed in ⁷⁰). However, our data pose the intriguing question as to whether type-2 inflammation (IL-4 plus additional inflammatory signals) could also be damaging through pregnancy. Consequently, progesterone could serve to down-regulate excessive type-2 inflammation via increased macrophage fizz1 production.

Macrophage ym1 mRNA transcripts were up-regulated by progesterone, both in the presence and absence of IL-4. Two major functions have been associated with ym1 activities, namely tissue remodelling and repair,⁷¹ and induction of Th2 cytokine production.⁷² Significantly, both functions would, by general consensus, be important in the maintenance of a healthy pregnancy. Recently, Routely and Ashcroft¹⁰ have shown in a murine model of wound repair in the skin that progesterone acts to promote an alternatively activated phenotype, as indicated by increasing levels of ym1 protein expression although not ym1 mRNA expression. Unsurprisingly, therefore, alternative macrophage activation and concomitant ym1 expression are related closely to tissue remodelling and repair during pregnancy in mice.^25,59

The ability of progesterone to modulate various aspects of macrophage activity has been well studied (reviewed in ²⁸). However, during the course of the present studies, we have demonstrated for the first time the ability of progesterone to alter expression of markers associated with alternative macrophage activation in vitro. In addition, the novel finding that progesterone regulates differentially markers of alternative macrophage activation, as demonstrated by its ability to down-regulate IL-4-induced argI and mrc-1, but up-regulate ym1 expression, has important implications both for the general study of alternative macrophage activation and for the study of macrophages in pregnancy and remodelling.

Disclosures

The authors declare no financial or commercial conflict of interest.