Influence of androgens on the innate immune system

2.1 Function

Testosterone and dihydrotestosterone (DHT) are cholesterol-derived hormones that have masculinizing effects.⁸ Testosterone levels increase at three time-periods in the male: prenatal, neonatal, and pubertal stages.⁹ The peak during embryonic and neonatal life is required for the development and differentiation of components of the male reproductive tract, including the Wolffian ducts, urogenital sinus, and external genitalia.⁹ During this period, androgens also shape the central nervous system.¹⁰ During puberty, androgens act to promote the appearance of secondary sexual characteristics, such as the distribution of hair patterns, voice deepening, and increased muscle mass.¹¹ Androgens are also responsible for the initiation and maintenance of spermatogenesis.¹¹ Aside from playing an important role in reproduction, androgens also elicit a wide range of effects in non-reproductive tissues, including the liver, skin, muscle, bone, adipose tissue, and the nervous system.¹² As men age, testosterone levels decline approximately 0.5%–2% per year.⁸

2.2 Production and metabolism

Testosterone synthesis is regulated by a negative feedback loop within the hypothalamic–pituitary–gonadal axis.¹³ The hypothalamus secretes gonadotropin-releasing hormone (GnRH) that stimulates the secretion of luteinizing hormone (LH) and follicle-stimulating hormone.^{13, 14} These gonadotropic hormones travel via the blood and act on their respective receptors in the gonads. LH acts directly on Leydig cells in the testis to produce testosterone.¹⁵ When circulating levels of testosterone rise, this signals the hypothalamic–pituitary complex to reduce gonadotropin secretion.

The metabolism of androgens is multifaceted. Conversion of cholesterol into testosterone primarily occurs in the Leydig cells of the testis.¹⁶ Testosterone can also be converted into estrogen by aromatase in a context- and tissue-specific manner; for example, in adipose and hematopoietic cells, as well as in the central nervous system.¹⁷ Testosterone can also be converted into DHT by 5α-reductases and unlike testosterone, DHT is not able to be aromatized into estrogen.¹⁸ Therefore, the expression levels of either enzyme determines the balance of androgen and estrogen syntheses.⁴ The male and female placentae can synthesize and metabolize steroids, including testosterone, from the precursor cholesterol.¹⁹ Sex hormone binding globulin (SHBG) binds to sex hormones with high affinity; approximately half of total testosterone is bound to SHBG with most of the remaining testosterone bound with low affinity to serum albumin; free testosterone is only 1%–2% of total testosterone.²⁰

2.3 Androgen receptor

Both testosterone and DHT mediate their actions via the androgen receptor (AR). The AR is a member of the steroid hormone nuclear receptor family, with DHT binding with greater affinity than testosterone.²¹ There are three main functional domains of the AR, including the N-terminal domain, the DNA-binding domain, and the C-terminal ligand–binding domain.²² The AR, located on the X chromosome, is expressed in a variety of different cells and tissues; thus, it is not surprising that the effects of androgens are biologically diverse, including on the reproductive, immune, and musculoskeletal system and on organs, such as the liver, heart, and placenta. Classical AR signaling elicits genomic effects, with binding occurring in a DNA-binding-dependent manner. Androgens also possess the ability to bind and activate membrane ARs, a group of G protein–coupled receptors.²³ Unlike classic cytoplasmic binding that elicits genomic effects, cell surface receptors alter cell signaling via rapid alterations in intracellular signaling cascades, such as mitogen-activated protein kinases.⁸

4.1 Neutrophils

Polymorphonuclear cells (PMNs) or neutrophils are key components of the innate immune system. They are the most abundant leukocyte in peripheral blood and are rapidly recruited thanks to specific chemokines and alarmins to sites of infection.³⁶ PMNs have three important antimicrobial functions: phagocytosis, degranulation, and the release of material via neutrophil extracellular traps composed mainly of DNA from neutrophils trapping pathogens.^{37, 38} Aside from their role in eliminating pathogens, they also signal a range of cytokines, chemokines, free radicals, and alarmins to effectively generate an immune response.

Multiple studies have shown that androgens impact PMN maturation and function.^{4, 39} In a mouse melanoma model, flow cytometry analysis revealed that total neutrophil counts in orchiectomized mice were reduced to the levels found in female mice.⁴⁰ Interestingly, the PMN from orchiectomized mice also showed a trend toward decreased cytotoxicity and the number of mature, segmented PMN compared to those from control male mice.⁴⁰ On the other hand, PMN from control mice showed the enhanced phagocytosis and generation of reactive oxygen species upon stimulation when compared to those of orchiectomized mice.⁴⁰ These results were able to be reversed with testosterone replacement, clearly highlighting that abnormal phenotypic changes occur in PMNs that express the AR after orchiectomy. In a study by Chuang et al., AR-knockout male mice were neutropenic and more susceptible to acute bacterial infection.⁴¹ This highlights that the AR is important for the development of PMN precursor cells and that it impacts their proliferative activity. Similarly, another study using a bacterial model of prostate inflammation showed that PMN recruitment was higher in testosterone-treated rats.⁴² However, the PMN exposed to the high level of androgen (10 mg/kg/day) were less efficient ex vivo in terms of bactericidal ability and released a higher amount of anti-inflammatory cytokines compared to controls.⁴² The results from this study support the concept that androgens result in increased numbers of PMN that have impaired bactericidal ability generating an unbalanced pro- versus anti-inflammatory response; however, a limitation is that these results were not confirmed in in vivo models as well as in human. Tang et al. showed that androgen/AR signaling stimulates PMN development in the bone marrow and their infiltration in the liver and found that this was responsible for male-biased sexual dimorphism in liver metastasis.⁴³ Specifically, androgens were found to boost hepatic seeding of tumor cells and alter PMN recruitment and function.⁴³ In a pre-clinical model of chorioamnionitis due to group B Streptococcus, androgens were responsible for an increase in placental pro-inflammatory cytokines and infiltration of PMN.⁴⁴

Across the aforementioned studies, it is evident that the AR and/or androgens stimulate the production of PMN. Mechanistically, this could be explained through AR enhancing granulocyte colony-stimulating factor signaling as found in a study by Chuang et al.⁴¹ This mechanism works through the activation of extracellular signal-regulated kinase 1/2 and through sustaining signal transducer and activator of transcription (STAT)3 activity.³⁹ Many knockout mice studies have shown that when STAT3 activity is maintained, the result is a hyperproduction of neutrophils.^{45, 46} In sum, the reason why males are at a higher risk of developing sepsis and other innate immune driven-conditions could in part be explained by a skewed activation of PMN by testosterone, which contributes to a weaker ability to clear the infection than female, or, on the other hand, due to a testosterone-induced cytokine storm from innate immune cells, or both.

4.2 Macrophages

Monocytes are derived from the bone marrow and circulate in the blood and spleen until they are recruited to tissues, where they differentiate into macrophages and phagocytose pathogens and secrete chemokines and cytokines.⁴⁷ Monocytic growth factors are released, in particular granulocyte-macrophage and macrophage colony–stimulating factors (M-CSF) in order to induce such differentiation.⁴⁸ Tissue macrophages include, for instance, alveolar macrophages (lung), Langerhans cells (skin), microglial cells (central nervous system), and Kupffer cells (liver).⁴⁷

Similar to neutrophils, evidence exists regarding both the pro- and anti-inflammatory effects of AR and/or androgens on macrophages. Sex differences regarding monocyte development have been reported, for example, M-CSF expression is increased in the bone marrow, and monocytic differentiation is higher in male compared to female mice.⁴⁹ Although not directly tested in this study, androgens could play a role in the proliferation, differentiation, and survival of macrophages. When stimulated in vitro with LPS, peripheral blood monocytes from female patients produced less TNF-α and IL-1β than males.⁵⁰ Similarly, along with a high fat diet, male mice exhibited increased weight gain compared to female, and an amplified innate immune response as determined through adipose tissue macrophages and inflammatory gene expression compared to female.⁵¹ In a clinical setting, cutaneous wound healing takes longer in elderly males compared to females and comes with a higher inflammatory response compared to elderly females.³¹ In AR knockout mice, it was shown that wound healing was accelerated and accompanied by a reduced immune response, as indicated by lowered mRNA and protein expression of TNF-α.³¹ Using in vitro and in vivo models, it was shown that androgen increased the adhesion of monocyte on endothelial cell via hyperexpression of vascular cell adhesion molecule-1 on the surface of monocytes.⁵² Experimental autoimmune myocarditis is described as an inflammation of the myocardium in the presence of cardiac necrosis and fibrosis, as characterized by excessive macrophage infiltration.^{52, 53} In this model of myocarditis, where the AR was suppressed, there was a decrease in TNF-α expression by macrophages.⁵³ In turn, this attenuates heart inflammation and promotes tissue repair.⁵³ Because these sex biases have been demonstrated mostly in a post-pubertal context, they point to the influence of sex hormones in shaping the development and level of inflammatory responses of monocytes.

On the other hand, there is also evidence regarding the anti-inflammatory effects of androgens on macrophages. In one study, orchiectomized mice had significantly increased TLR4 cell surface expression on macrophages compared to sham orchiectomized mice.⁵⁴ This implies that androgens may reduce macrophage cell responsiveness to pathogens, thereby contributing to the generation of an immunosuppressive environment. In-line with these findings, macrophages in females display higher TLR4 expression.⁵⁵ In another in vitro study, testosterone decreased expression and secretion of TNF-α and IL-1β in human monocyte-derived macrophages, both key pro-inflammatory cytokines.⁵⁶ After trauma-hemorrhage, macrophage immune response was lowered in male than female.⁵⁷ Specifically in this study, the administration of DHT pellets to castrated mice resulted in a reduction in IL-1β and IL-6 in splenic and peritoneal macrophages following trauma-hemorrhage.⁵⁷ Altogether, it is difficult to draw conclusions about pro- versus anti-inflammatory effect of testosterone on macrophages; the effects appear to vary depending on the age, type of disease, acute or chronic pathophysiological process, titer and balance between the different androgens, nature of the different resident macrophages, and other factors that remain to be determined.

4.3 Dendritic cells

Dendritic cells act as an interface between the innate and adaptive arms of immunity. They are professional antigen-presenting cells originating from hematopoietic stem cells.^{4, 58} Dendritic cells are critical for immune surveillance and for later activation of naïve T lymphocytes.^{4, 58}

The impact of sex hormones on dendritic cells is much less studied in comparison to neutrophils and macrophages. Some studies have reported that androgens have an immunosuppressive effect on dendritic cells.⁴ In men with hypogonadism, the most widespread dendritic cell subset was increased, and this increase was subsequently reversed with testosterone treatment.⁵⁹ Ex vivo dendritic cell differentiation was promoted by estrogen but remained unchanged with DHT addition.⁶⁰ However, it is unlikely that androgens directly influence dendritic cells as one study by Paharkova-Vatchkova et al. highlighted that bone marrow–derived dendritic cells express the estrogen receptor but not the AR.⁶⁰

4.4 Natural killer cells

Natural killer (NK) cells also make up the first line of defense due to their direct cytotoxic activity on infected cells.⁶¹ Upon treating healthy men with a GnRH antagonist, simulating medical castration, there was an increase in the percentage of NK cells.⁶² In a study by Solerte et al., dehydroepiandrosterone increased the cytotoxic activity of NK cells on tumor cells in vitro through the increasing synthesis and secretion of insulin-like growth factor-I, thereby positively modulating immunity.⁶³

4.5 Mast cells

Mast cells play a key role in the allergic inflammatory response. Mast cell–associated diseases display a sex bias, with females at higher risk, thus bringing into question the role of sex steroids.⁶⁴ A study by Chen et al. investigated the expression of AR in human skin mast cells but found that there was no change in mast cell degranulation upon testosterone treatment.⁶⁵ However, a preclinical study by Mackey et al. showed that high perinatal androgens in males were responsible for reduced histamine levels and severity of anaphylaxis.⁶⁴ The same findings were able to be repeated in females through perinatal exposure to testosterone.⁶⁴

4.6 Eosinophils and basophils

Eosinophils and basophils, albeit rare, share many similar features with mast cells. A study that investigated the adhesion of eosinophils to human mucosal microvascular endothelial cells found that testosterone reduced eosinophil adhesion properties.⁶⁶ It was also determined that testosterone administration to castrated male mice infected with a parasite, Brugia pahangi, resulted in a downregulation of peripheral and bone marrow eosinophils in the earlier stages of infection.⁶⁷ It is not currently known whether basophils express the AR. In a study where exogenous testosterone treatment was administered to males, there was no change in circulating basophils.⁶⁸

4.7 Signaling molecules

There are various signaling molecules involved in the effect of androgen on the inflammatory response. TLRs are proteins that play a crucial role in innate immunity by recognizing PAMPs. In a study by Rettew et al., testosterone induced a reduction in TLR4 expression in a macrophage cell line.⁵³ NLR protein (NLRP) 3 inflammasome is an innate immune signaling complex present in innate immune cells and playing a major role in active IL-1 synthesis.⁶⁹ NLRP is mainly expressed in monocytes, macrophages, and dendritic cells.⁶⁸ In a liver injury mouse model, NLRP3 activation induced by androgens contributed to liver fibrosis.⁶⁹ As expected, NLRP3 activation, in turn, resulted in increased concentrations of cytokines: IL-1β, TNF-α, IL-6, IL-18 in the liver.⁶⁹ Another study showed the possible role of epidermal growth factor receptor/phosphoinositide-3 kinase/protein kinase B and nuclear factor kappa B activation pathways on the impact of androgen as inducers of some inflammatory mediators, namely, prostaglandins F2 (PGF2).⁷⁰ It would be helpful to integrate omics approaches in the future to further identify the changes induced by exposure to androgen at the molecular level.