Immune and vascular contributions to organogenesis of the testis and ovary

Abstract

Gonad development is a highly regulated process that coordinates cell specification and morphogenesis to produce sex-specific organ structures that are required for fertility, such as testicular seminiferous tubules and ovarian follicles. While sex determination occurs within specialized gonadal supporting cells, sexual differentiation is evident throughout the entire organ, including within the interstitial compartment, which contains immune cells and vasculature. While immune and vascular cells have been traditionally appreciated for their supporting roles during tissue growth and homeostasis, an increasing body of evidence supports the idea that these cell types are critical drivers of sexually dimorphic morphogenesis of the gonad. Myeloid immune cells, such as macrophages, are essential for multiple aspects of gonadogenesis and fertility, including for forming and maintaining gonadal vasculature in both sexes at varying stages of life. While vasculature is long known for supporting organ growth and serving as an export mechanism for gonadal sex steroids in utero, it is also an important component of fetal testicular morphogenesis and differentiation; additionally, it is vital for ovarian corpus luteal function and maintenance of pregnancy. These findings point toward a new paradigm in which immune cells and blood vessels are integral components of sexual differentiation and organogenesis. In this review we discuss the state of the field regarding the diverse roles of immune and vascular cells during organogenesis of the testis and ovary and highlight outstanding questions in the field that could stimulate new research into these previously underappreciated constituents of the gonad.

Graphical Abstract

Introduction

Sexually dimorphic development of the gonad is essential for successful gametogenesis and fertility. While a great deal of research has gone into studying the mechanisms of sex determination, i.e., identifying the genetic, molecular, or environmental trigger that launches testis- or ovary-specific developmental programs, we know less about sexual differentiation, the process by which male- and female-specific morphology is brought about. Arguably the process of sexual differentiation and its outputs, such as sex-specific gonadal architecture, are more vital to understand, since they are highly evolutionarily conserved among vertebrates (and to some extent invertebrates), despite divergent mechanisms of sex determination among different vertebrate classes [1].

Mammalian gonads (and generally most vertebrate gonads) consist of several major cell types (Fig. 1). Supporting cells, in which sex determination genes such as Sry are specifically expressed [2], comprise the first sex-specific cell type specified in the gonad; these cells are termed Sertoli cells in the testis and granulosa (or pre-granulosa) cells in the ovary. Supporting cells, as the name suggests, nurture gametogenesis upon primordial germ cell colonization of the fetal gonad until the final stages of gametogenesis in adulthood. These cells, in conjunction with germ cells and male-specific peritubular cells or female-specific theca cells, make up the basic functional units of the gonad, which are testicular seminiferous tubules (also called testis cords in fetal stages) and ovarian follicles.

Fig. 1. Architecture of the fetal and adult mammalian gonad.

Cartoon depicting the distribution of different cell types in the mouse E10.5 undifferentiated gonad (top), E13.5 fetal testis (middle left), E13.5 fetal ovary (middle right), adulttestis (bottom left) and adult ovary (bottom right). Cartoons also show fetal gonad morphology and adult gonadal architecture, focusing on seminiferous tubules in the testis and antral follicles and corpus luteum in the ovary, as well as their associated interstitial compartments. Note that structures and cell types are not drawn to scale.

Outside of these gametogenic compartments, there is also an interstitial compartment, whose major responsibility is the production of sex steroids, such as androgens (e.g., testosterone) in the male and estrogen and progesterone in the female. The interstitial compartment contains a wide variety of cell types in both sexes, including steroidogenic cells, smooth muscle, vascular cells, mesenchymal cells such as fibroblasts, and immune cells [3]. Recent studies have shown that the interstitial compartment is an active player in testicular spermatogenesis [4], suggesting that interstitial cells are not merely passive participants in gonadal development and homeostasis. While the interstitial compartment is traditionally appreciated for its steroidogenic function, testis-specific interstitial cell types, such as peritubular cells (also called peritubular myoid cells), vascular endothelial cells, and macrophages, may directly influence spermatogenesis within the seminiferous tubule [5–7]. Steroidogenesis in the testis and ovary is largely performed by Leydig cells in the testicular interstitium and theca cells surrounding ovarian follicles, respectively, although granulosa cells and corpus luteal cells (which are granulosa cells that luteinize after ovulation and produce progesterone) also contribute to sex steroid production in the ovary [8–10].

As compared to the supporting cell gametogenic compartment, less is known about how the interstitial compartment is sexually dimorphic and how the interstitium actively contributes to sex-specific morphogenesis of the gonad. In particular, the interactions between vasculature, immune cells, and the developing gonad are poorly understood, but are likely a fruitful area of research to understand more deeply the mechanisms driving testicular and ovarian differentiation. Here we will discuss the role of immune cells and vasculature in the development and function of the testis and ovary, highlighting previously underappreciated roles for these cell types in development and organogenesis.

Gonadal immune cells

The first immune cells to arise in the embryo are primitive macrophages and are detected in the vicinity of the nascent fetal gonad as early as E10.5 in mice, shortly after the initial specification of the gonadal primordium [11]. Therefore, during sex determination and the initial establishment of sexual dimorphism in the gonad, macrophages are the main immune cell population that can play a role in gonad morphogenesis and differentiation. Given their early appearance within the fetal gonad, together with their dominant presence as the most populous immune cell type in the adult testicular interstitium [12, 13], most research into gonadal immunology has focused on macrophages. However, over the course of fetal development until birth, other myeloid cells, such as monocytes and a small number of eosinophils, but not dendritic cells or neutrophils, are detected in the fetal gonad [14]. It is possible other myeloid cell types such as mast cells, and potentially also lymphoid cell types, play roles in later aspects of postnatal and adult gonad organogenesis. In general, relatively little is known about the role of these other (i.e., non-macrophage) immune cells in testicular and ovarian organogenesis and function.

Macrophages

Macrophage phenotype

Macrophages are the most versatile cells of the immune system, playing critical roles in tissue homeostasis by removing cellular debris and foreign pathogens via phagocytic mechanisms [15, 16]. Macrophages are also involved in innate and adaptive immune responses during organ developmental processes and chronic inflammatory diseases by interacting with other immune cells such as T cells [17–19]. Although macrophage phenotypes are heterogeneous and plastic, characterized by specific cellular surface markers and distinct cytokine secretion, they can be generally classified into two subsets: classically activated, or M1, macrophages (a pro-inflammatory profile) and alternatively activated, or M2, macrophages (an anti-inflammatory profile) [15, 16, 20], although recent trends are moving away from using such a binary system and moving more towards a spectrum of polarization phenotypes. In response to local microenvironmental stimuli and signals, macrophages can rapidly switch their functional phenotype between M1 and M2. The process is defined as macrophage polarization in which the dynamic change of M1/M2 ratio in the reproductive system is highly associated with physiological homeostasis of various tissues, such as the establishment of maternal-fetal tolerance and testicular immune privilege [21–23], as well as pathological changes such as pregnancy complications and ovarian cancer [21, 23–25].

M1 macrophages

M1 macrophages activated by Toll-like receptor (TLR) ligands from pathogens (e.g., lipopolysaccharide (LPS) or Th1 cell products such as interferon gamma, IFN-γ, and tumor necrosis factor alpha, TNFA), express CD80, CD86, inducible nitric oxide synthase (iNOS) and major histocompatibility complex class II receptor (MHCII) surface markers. They can induce the secretion of pro-inflammatory cytokines (interleukin-1α [IL1A], IL1B, IL6, IL12, and TNFA) and chemokines (chemokine [C-X-C motif] ligand 10 [CXCL10], and CXCL11) to further promote M1 polarization [20, 26–28]. In addition, granulocyte-macrophage colony-stimulating factor (GM-CSF), also known as colony-stimulating factor 2 (CSF2), increases the expression of monocyte and macrophage pro-inflammatory cytokines and induces macrophages towards an M1 state [29, 30]. LPS can stimulate CSF2 expression and, in turn, CSF2 priming also promotes LPS-induced release of proinflammatory mediators [31–33].

M2 macrophages

M2 macrophages induced by IL4, IL10, IL13, IL33, transforming growth factor beta (TGF-β) and macrophage colony-stimulating factor (CSF1, also called mCSF) can be identified by the expression of specific markers, such as mannose receptor (CD206/MRC1), CD163, CD36, found in inflammatory zone 1 (FIZZ1), arginase 1 (ARG1), and MAF [16, 23, 28, 34]. IL4 or IL13, as major inducers of M2 polarization, are produced by Th2 lymphocytes, eosinophils, basophils, and mast cells [35]. Activated M2 macrophages regulate immune and inflammatory reactions by secreting many cytokines and chemokines, such as IL10, TGF-β, CCL1, CCL17, CCL18, CCL22, and CCL24 [16, 23], as well as angiogenesis and lymphangiogenesis by producing vascular endothelial growth factor A/C/D (VEGFA/C/D), insulin-like growth factor 1/2 (IGF 1/2), platelet derived growth factor subunit B (PDGFB), and matrix metalloproteinase 2/9 (MMP2/9) [19, 36].

Given the modest overlap of some cell surface markers or secreted cytokines and transient phenotype shift in the process of M1/M2 polarization, it is difficult to discriminate between M1 or M2 macrophages in some cases. Therefore, in the future studies, it is important to find more differentiated macrophage markers in the development of monocytes under physiological and inflammatory conditions by epigenetic, transcriptomic, and proteomic analysis, as well as by using genetically modified mice. In addition, M1/M2 polarization may not fully correspond between in vitro experiments and in vivo situations due to the microenvironmental complexity of organogenesis and tissue homeostasis.

Testicular macrophage populations, from fetus to adult

Tissue-resident macrophages have complex and heterogeneous phenotypes in different tissues where they have both immune and non-immune activities [37]. In general, testicular macrophages (TM) are the most abundant leukocyte type in the testis and represent a large proportion of the interstitial cell population from embryogenesis to postnatal development [11, 13, 38]. In rats, TM were first reported at 19-20 days gestation, but in a subsequent study, they were detected in fetal testes 3 days earlier [39, 40]. A recent study in mice showed that macrophages arise in the urogenital ridge at embryonic day (E) 10.5, upon gonad specification, and increase in number near the gonad-mesonephros border and occasionally within the gonad at E11.5 and E12.5. Most fetal gonadal macrophages have a stellate morphology and strongly express M2 macrophage markers, such as CD206 (MRC1), MAF, and Arg1, suggesting that gonadal macrophages are M2-like in their polarization status [11]. Why fetal gonadal macrophages exhibit an M2-like phenotype is currently unclear. However, there are some possible clues: early gonadal macrophages are derived from primitive yolk sac macrophage progenitors [11], which progressively mature and develop an increased number of dendrites during their migration through bloodstream to infiltrate tissue and are often M2-like within developing organs [41]. Also, bone morphogenetic protein 4 (BMP4) and BMP7, TGF-β superfamily members, which can induce M2 macrophage differentiation [42, 43], are expressed in the fetal gonad before sexual differentiation [44, 45]. These gene expression patterns in the fetal gonad may contribute to a tissue microenvironment conducive to M2 macrophage activation. In addition, there are two macrophage populations from E14.5 until birth found in the testis: F4/80^Hi macrophages and F4/80^Int macrophages that highly express the M2 marker CD206 and monocyte marker LY6C, respectively [14]. M2 macrophages are generally considered to be immunosuppressive and are associated with vascular and tissue remodeling during development and cancer [37, 46]; therefore, M2 macrophages are well-suited to play an important role in fetal testis vascularization and development.

In the adult testis, most TM express the common tissue-resident macrophage markers F4/80, AIF1 (IBA1), and (C-X3-C motif) receptor 1 (CX3CR1). The heterogeneity of TM in adult mice has been demonstrated based on distinct surface marker expression patterns, anatomical localization, gene expression profiles, and lineage of origin. There are at least 2 TM populations: 1) macrophages located in the interstitium (also called interstitial TM), which highly express CD64 and CSF1R; and 2) macrophages located on the surface of the seminiferous tubules (also called peritubular TM), which express MHCII [7, 14, 47]. A recent study using high-dimensional mass cytometric analyses revealed the presence of a CD206⁻MHCII⁻ macrophage population, which declines in number during postnatal testis development, with the concurrent emergence of a CD206⁻MHCII⁺ macrophage population [14], indicating that CD206⁻MHCII⁻ macrophages may be the precursor of peritubular TM. TM are implicated in the regulation of spermatogonial development and Leydig cell differentiation and steroidogenesis [7, 11, 48–50]. They also contribute to the establishment of an immunoprivileged niche by exhibiting a M2 macrophage phenotype, which is characterized by the absence of proinflammatory gene expression and high expression levels of many immunosuppressive and M2-type activation factors, such as CD163 in both populations [47]. In the human testis, CD14⁺CD163⁺ resident macrophages represent the largest myeloid cell subset and also highly express the M2 macrophage marker CD163 [51]. Although it is well-known that the phenotypes of TM are dynamic and diverse throughout the entire lifecycle, more details of macrophage heterogeneity likely need to be examined.

M2 phenotype of testicular macrophages

Once macrophages colonize organs, specific signals from the local tissue microenvironment control the functional polarization of resident macrophages. One potential major source of organ-specific signals is the testicular interstitial fluid. Testicular interstitial fluid contains numerous immunomodulatory molecules, such as prostaglandins (PGs), steroid hormones (testosterone and corticosterone), and activin and inhibin (which belong to the TGF-β protein superfamily), all of which may contribute to TM phenotype and function [52–54]. Recent data show that prostaglandins PGE2 and PGI2, testosterone, and corticosterone present in rat testicular interstitial fluid can shift the polarization of blood-derived or CSF2-derived M1 macrophages: interstitial-fluid-exposed macrophages polarize toward the M2 macrophage phenotype, likely to perform immunosuppressive functions by inducing CD163 expression and secretion of the anti-inflammatory cytokine IL10, with concomitant low production of TNFA [53, 54]. The production of activin and inhibin by the testis also could be a potent inducer of M2 macrophage activation [16]. Testes from adult mice with elevated activin A levels exhibit an increased number of CD206⁺MHCII⁻ cells, indicating that activin A does have an effect on M2 macrophage activation [55]. In addition, IL13 produced in the testis induces the expression of the M2 macrophage marker YM1 (official name CHIL3), dependent on IL13R signaling [56]. However, whether the testis can secrete other molecules to govern TM phenotype and function needs to be further explored.

Macrophage populations in the ovary

During mouse embryonic development, ovarian macrophages at E14.5 were comprised of a dominant CD11b^IntF4/80^Hi population [57], consistent with yolk-sac-derived primitive macrophages that are also observed in the fetal testis and rest of the embryo at that stage [11, 58]. In contrast, CD11b^HiF4/80^Int cells, a phenotype consistent with definitive macrophages, were dramatically increased in number in the E16.5 ovary to similar levels as CD11b^IntF4/80^Hi cells [57], indicating a dramatic shift in macrophage populations as was reported in the fetal testis [59]. In addition, CD11b^HiF4/80^Int macrophages strongly expressed monocyte markers, such LY6C and CCR2, as compared to CD11b^IntF4/80^Hi macrophages that expressed the M2 macrophage marker CD206. But MHCII, a M1 marker, was not expressed in either macrophage population [57], indicating that fetal ovarian macrophages exhibit an M2-like phenotype. In addition to immunophenotyping, single-cell RNA sequencing has become a widely used technique to uncover cell identity and heterogeneity and to elucidate dynamic processes during tissue development. Single-cell RNA sequencing has revealed clear macrophage populations in fetal mouse (E14.5), monkey (E84 and E116), and human (20-26 weeks) ovaries [60–62]. CD 14+ primitive monocytes and activated MHCII+ monocyte-derived cells were present in human fetal ovaries at 24 weeks of gestation [63], supporting reports of ovarian macrophages in the fetal ovary. However, the immune phenotype of fetal ovarian macrophages in mammals is not well-defined, and their roles in ovarian developmental processes including vascularization, germ cell meiosis, and pre-granulosa cell differentiation and maturation are also unclear.

In the adult mouse ovary, macrophages are also the most abundant immune cell population and widely distributed in the thecal, interstitial, and luteal regions, as well as in the granulosa cell layer of atretic follicles [64–66]. In human ovaries, CD68-positive macrophages are mainly localized near vascular cells and the theca-lutein cell layer of the corpus luteum, while MHCII-positive macrophages reside within corpora lutea [63, 67, 68]. Several studies showed that MHCII^HiCD11b⁺F4/80⁺ macrophages are the most common cells among ovarian phagocytes and commonly display flattened cell bodies with dendritic morphology, which is consistent with an activated macrophage phenotype [57, 66]. A small MHCII^LoCD206⁺ macrophage population arises at birth and maintains a stable frequency during postnatal development [57]. However, there is still no direct evidence addressing whether the ovarian macrophage population can be classified as M1 or M2. In a recent study, monocyte recruitment and macrophage alternative activation (M2) have a distinct increase in aged ovaries as compared to young ovaries, suggesting that aging-associated chronic inflammation may cause a shift in ovarian macrophage ontogeny and polarization [69].

Macrophage origins

Tissue-resident macrophages arise embryonically from distinct and successive hematopoietic waves, including yolk sac (YS)-derived primitive macrophages at E7.0-7.5 (in mice), erythro-myeloid progenitors (EMPs) at E8.0-8.5, and aorta-gonad-mesonephros (AGM)-derived early hematopoietic stem cells (HSCs) at E10.5 that migrate to the fetal liver (FL) and generate myeloid progenitors and monocyte-like cells [37, 70–73]. Starting from E16.5, the presumptive bone marrow (BM) gives rise to all lineages of definitive hematopoiesis, including blood monocytes contributing to tissue macrophages [74]. For example, brain microglia and a subset of liver Langerhans cells are exclusively derived from early YS progenitors [70, 75], while large proportions of resident macrophages in the liver, lung, skin, and mammary gland have embryonic origins either from EMPs or FL-derived monocytes [76–79]. Macrophages in the intestine, dermis, and pancreas are maintained by progressive replacement of embryonic macrophages by BM-derived monocytes [80–82]. Therefore, each organ appears to have distinct ontogeny (or ontogenies) for their specific tissue-resident macrophages and have unique kinetics of how macrophages colonize tissues over the lifespan.

Likewise, the origins of resident macrophages in the reproductive organs have been investigated (Fig. 2), albeit in less detail than other organs. In the fetal testis, primitive YS hematopoietic progenitors give rise to the first macrophages that colonize the gonadal-mesonephric region [11], although recent single-cell analyses have revealed that multiple myeloid cell types, likely of different ontogeny, reside in the fetal testis [14]. In the adult testis, interstitial macrophages, in close proximity to Leydig cells, were thought to arise from embryonic progenitors, whereas peritubular macrophages were proposed to be exclusively seeded in the testis starting from postnatal stages, seemingly from BM-derived progenitors [47]. However, two recent studies showed that postnatal and adult BM-derived cells likely contribute minimally to adult testicular macrophage populations [14, 59]; specifically, FL-derived progenitors, rather than YS-derived or BM-derived cells, likely have a dominant contribution to both interstitial macrophages and peritubular macrophages [14]. In adult ovaries, both embryonic and BM-derived macrophages have the capacity to give rise to distinct macrophage subtypes [57]. However, the exact contribution of each of these hematopoietic waves to tissue-resident macrophages in adult life is still debated due to a lack of clarity from lineage-tracing mouse models, such as lack of specific tools to label FL-derived monocytes.

Fig. 2. Model of testicular macrophage origin.

Primitive macrophages generated in the yolk sac (E7.5) give rise to the majority of gonadal–mesonephric macrophages localized near vasculature. EMPs emerge in the yolk sac (E8.5) and exclusively contribute to a small subset of adult interstitial macrophages. Fetal liver-derived monocytes at mid-late fetal stages (E11.5-E14.5) can give rise to two testicular macrophage populations including interstitial macrophages, which are localized in the interstitial compartment in close contact with Leydig cells and blood vessels, and peritubular macrophages, which are localized to the peritubular myoid cell layer. However, adult BM-derived monocytes do not substantially contribute to testicular macrophages. Solid arrows indicate a positive contribution and dotted arrows indicate a currently unclear relationship. BM, bone marrow; EMPs, erythro-myeloid progenitors; HSCs, hematopoietic stem cells; M⌽, macrophage; CV, coelomic vessel; TC, testis cord; VP, vascular plexus.

The roles of macrophages during gonadal angiogenesis and lymphangiogenesis

The complex vascular system composed of blood vessels and lymphatic vessels play essential roles in the transport of oxygen, metabolites, fluids, and circulating leukocytes between tissues and organs [83]. Angiogenesis and lymphangiogenesis are not only associated with many pathological processes such as tissue repair, inflammation, and tumor growth, but are also involved in organogenesis and tissue homeostasis [83, 84]. In these microenvironments, recruited or tissue-resident macrophages are an important source of angiogenic/lymphangiogenic mediators to regulate vessel sprouting and function [19, 85]. Here, we will highlight the roles of macrophages in angiogenesis and lymphangiogenesis during testis and ovarian development.

Role of macrophages during fetal testis vascularization

Macrophages derived from primitive yolk-sac hematopoietic progenitors circulate in the blood and colonize the developing mouse embryo by as early as E9.5 in mice [58, 86], and yolk-sac-derived primitive macrophages are detected in the fetal gonad primordium during initial gonad formation as early as E10.5. Macrophages are localized near developing vasculature in the nascent testis, gradually accumulating from E10.5 to E13.5, specifically near the gonad-mesonephric vascular plexus region at the peak of de novo vascularization and vascular remodeling of the testis (E11.5-E13.5), as well as along the coelomic surface artery and its associated side branches [11]. Fetal gonadal macrophages are in an M2-like polarization status, a state that is linked to tissue remodeling and angiogenesis [11, 87].

Testis vasculature depletion in a whole-organ culture system results in a robust reduction of fetal gonadal macrophages, suggesting that vascularization of the fetal testis may be critical for macrophage migration and/or maintenance. In addition, the depletion of macrophages using a genetic cell-ablation technique leads to a disorganized vasculature and aberrant testis cord formation [11]. The mRNA expression of the proangiogenic factor Vegfa within sorted gonadal macrophages was significantly lower than that of the non-macrophage cell populations of the gonad/mesonephros, suggesting that macrophage-produced VEGFA is unlikely a major source of VEGFA in the gonad [11]. Similarly, during renal organogenesis most F4/80⁺CD206⁺ kidney macrophages are perivascular. Macrophage-depleted kidneys at E12.5 exhibit higher numbers of isolated and unconnected endothelial structures and a reduction in the average size of CD31+ (PECAM1+) structures, indicating that macrophages can positively shape the structure of blood vessels in fetal kidneys [88]. However, whether fetal renal macrophages produce proangiogenic factors, such as VEGFA, has not been investigated. Conversely, the CD64^HiCD206⁺ population of cardiac tissue macrophages at E14 and E17 expressing the highest levels of Vegfa mRNA seems to have a potential function in regulating cardiac angiogenesis [89]. Macrophages are also involved in fetal brain angiogenesis independently of VEGF [90]. Together, these studies indicate that macrophages, especially those that exhibit an M2 phenotype during embryogenesis, can regulate the remodeling of vasculature during organogenesis in a VEGFA-dependent or -independent fashion. In terms of fetal gonad development, macrophages play an important role in fetal testis neovascularization.

Effect of macrophages on blood vessels of the adult testis

Although adult interstitial TM are clearly associated with CD31+ interstitial vasculature, the morphology and number of endothelial cells are normal after short-term macrophage depletion [7, 14], suggesting that TM do not seem to have any direct effects on the structure of adult testis vasculature. In other organs, for example, long-term depletion of resident macrophages in the young choroid results in choroidal vascular atrophy and is accompanied by the reduction of Vegfa mRNA expression [91]. Therefore, we speculate that short-term and long-term macrophage depletion may have different outcomes for vascular development. Under pathological conditions, such as in germ cell tumors and Klinefelter syndrome, infiltrating TM play an important role in mediating neovascularization during tumor growth [92–95]. However, whether resident TM in the adult testis have a specific function for testis vasculature needs to be further addressed, and sustained TM depletion is likely necessary to assess their role in vascular development and homeostasis. In addition, resident TM express high levels of Il10 and Il1b [47], which are thought to promote vascular permeability and angiogenesis [96, 97], Therefore, whether IL10 and IL1B are involved in vascular permeability of the adult testis is also unclear.

Role of macrophages on angiogenesis of the ovarian corpus luteum

Angiogenesis is highly associated with ovarian follicular growth and development, as endothelial cells are recruited to the theca layer during latter stages of folliculogenesis [98]. However, other aspects of female reproduction are also reliant on angiogenesis, such as the corpus luteum, which is a transient endocrine organ in the mammalian adult ovary required for the establishment and maintenance of pregnancy by producing progesterone [99]. Similar to developing tumors, the sprouting of preexisting vessels from the follicular theca layer is accompanied by leukocytes. The disruption of angiogenesis inhibits corpus luteum formation, steroidogenesis, and ovulation [98, 100]. Thus, angiogenesis has critical roles in normal luteal function.

Macrophages are the most abundant immune cells within the ovary and significantly increase in number in the developing corpus luteum [101, 102]. Progressive macrophage ablation increases the amount of ovarian hemorrhage in luteal and thecal tissue due to significant endothelial cell depletion and erythrocyte accumulation [103]. Conditional depletion of macrophages during embryo implantation leads to the breakdown of the luteal vasculature and the decrease of ovarian expression of genes that encode vascular endothelial growth factors [104]. Therefore, macrophages play essential roles in the formation and maintenance of luteal vasculature. In epithelial ovarian cancer, tumor-associated macrophages promote angiogenesis and metastasis by inducing the production of IL8 or IGF [105–107]. However, the mechanism underlying how macrophages regulate angiogenesis in normal corpora lutea is still not well-defined. The VEGF, EGF (epidermal growth factor), IGF, or MMP signaling pathways may be considered a bridge between luteal macrophages and angiogenesis. Earlier studies have shown that EGF and IGF can be produced by ovarian macrophages [108, 109]; therefore, macrophages could be considered another potential source of these growth factors in the ovary.

Role of macrophages in testicular and ovarian lymphangiogenesis during development

During embryonic development, lymphatic endothelial cells, a distinct endothelial cell lineage, bud off the embryonic cardinal vein at E9.5-E10.5 in mice [110]. Lymphatic vascular structures are present within many organs after E14.5 through the processes of primary lymphangiogenesis, sprouting, maturation, and terminal differentiation [111, 112]. Lymphatic endothelial cells are characterized by expression of FMS-like tyrosine kinase 4 (FLT4; also called VEGFR3), lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1), prospero homeobox 1 (PROX1), and the membrane glycoprotein podoplanin (PDPN), which distinguishes them from blood endothelial cells [113]. A recent study using a lymphatic-specific Prox1-EGFP reporter mouse model showed that lymphangiogenesis in the testes begins during late gestation. However, in contrast to other organs, lymphatic vessels are restricted to the testis sheath, or tunica albuginea, from fetus to adult, and unlike blood vessels, do not invade the testicular interstitium [114]. Conversely, ovarian lymphangiogenesis is rare through fetal stages until well after birth. Adult ovaries develop a rich lymphatic network, mainly located in the medulla and in the theca layer of developing follicles [114]. In addition, lymphatic vessels are also abundant in the corpora lutea of mice, rats, and primates [115–117].

VEGFR3 binding its secreted ligand, namely VEGFC or VEGFD, is the main driver of lymphangiogenesis in developmental and pathological contexts [118, 119]. The numbers of LYVE1+ lymphatic endothelial cells in the primate corpus luteum increase from early to mid-luteal phase, associated with the increase of both VEGFC and VEGFR3 mRNA levels [115]. The similar expression trends of VEGFC, VEGFD, VEGFR3, and LYVE1 are also observed in the corpus luteum of buffalo and cow [120, 121]. These data suggest that VEGFC/VEGFD–VEGFR3 interactions may regulate the function of the corpus luteum via lymphangiogenesis. VEGFR3 neutralization inhibits ovarian lymphangiogenesis, leading to the decrease of normal follicular number and disruption of hormone levels during pregnancy [122]. However, the expression and function of lymphangiogenic-related genes within the testis have not been thoroughly investigated to date. It is possible that lymphangiogenesis may be not important for testis development because of the limited lymphatic vessels in testis throughout life. Therefore, fewer studies in the field have focused on testicular lymphangiogenesis.

In addition to angiogenesis, macrophages may produce additional important signals to stimulate developmental and inflammation-induced lymphangiogenesis [19, 123]. Activated macrophages regulate the proliferation of pre-existing lymphatic endothelial cells by secreting lymphangiogenic factors, such as VEGFC and VEGFD [36]. For example, embryonic dermal macrophages expressing VEGFC and VEGFD have the capacity to mediate developmental lymphangiogenesis and lymphatic remodeling [124]. In macrophage-deficient embryos, dermal lymphatic vessels are hyperplastic, but the phenotype in jugular lymph sacs is hypoplastic [125]. Csf1^op/op mice, which exhibit a severe reduction in macrophages, exhibit reduced lymphatic branching in postnatal tracheas and ears; however, embryonic lymphangiogenesis seems to be normal [126]. These findings indicate that macrophages may play tissue- or time-dependent roles during developmental lymphangiogenesis. In reproductive organs, the relationship between macrophages and lymphangiogenesis has not yet been explored in depth during development, especially in the ovary, which has extensive lymphatic vessels from birth throughout adulthood. Macrophage depletion or conditional knockout of macrophage-derived genes that are associated with lymphangiogenesis in the ovary will help address this question.

Other immune cell types

Evidence for CD11b+ myeloid cell differentiation into vascular cells

Emerging evidence suggests that macrophages contribute to growing vasculature by transdifferentiating into pericytes (a vascular wall cell type) and vascular and lymphatic endothelial cells. A recent study showed that 30% of pericytes were labeled in E15.5 skin by using a pan-hematopoietic Vav-iCre driver to permanently label hematopoietic-derived cells with a fluorescent reporter. When using a monocyte/macrophage myeloid-specific CD11b-Cre driver to lineage-trace CD11b-derived cells, 15% of pericytes were labeled in E15.5 skin, suggesting that some pericytes are derived from tissue-localized myeloid progenitors in embryonic skin vasculature [127]. Myeloid angiogenic cells (MACs), a subset of monocytes expressing endothelial markers CD31, VEGFR2, and Tie-2 (also called TEK, an angiopoietin receptor), are widely described as endothelial progenitor cells [19, 128]. Although MACs do not differentiate into endothelial cells, they can significantly induce angiogenesis through IL8 secretion [129]. However, recent studies have provided indirect evidence that a large number of endothelial cells in adult blood vessels originate from YS-derived EMPs, which are also a source of tissue-resident macrophages [130], suggesting that endothelial cells of blood vessels have an embryonic hematopoietic origin. In many inflammatory models and tumor environments, macrophages also can transdifferentiate into lymphatic endothelial cells to promote lymphangiogenesis [131–134]. CD11b+ monocytes are considered the main source of lymphatic endothelial cell progenitors in the majority of studies [36]. Although whether macrophages differentiate or transdifferentiate into endothelial cells has not been clearly defined until now, it remains an interesting question. In the ovary and testis, it is also worthy to study whether CD11b+ myeloid cells contribute to pericytes and vascular endothelial cells.

Monocytes

Other than serving as the precursor cell type for definitive macrophages, the exact function of monocytes themselves in normal biology is somewhat unclear. Given their relative abundance in circulating blood, they are proposed to be important sentinels in both pathological and homeostatic states by patrolling vascular and tissue microenvironments [135, 136]. Monocytes have been implicated in regulating angiogenic processes during cancer [137], but in normal contexts, in particular within reproductive tissues, very little is known about monocytes. Monocytes have only recently been identified as a definitive cell population in the fetal and adult testis [14, 59]. Specifically, it was shown that self-renewing monocyte populations, derived during fetal stages independently of circulating blood progenitors, are critical precursors for testicular macrophages [59]; additionally, two recent studies have demonstrated that monocytes from definitive bone-marrow-derived hematopoiesis do not appreciably contribute to testicular macrophages [14, 59], in contrast to a previous study that claimed peritubular macrophages were exclusively derived from bone-marrow-derived circulating progenitors (i.e., monocytes) [47]. Despite classical dogma proposing that monocytes are short-lived circulating cells, there is a precedent for long-lived monocytes in organogenesis: in the fetal and perinatal lung, tissue-resident alveolar macrophages are derived from a stable fetal monocyte population that is long-lived and self-maintains throughout life [77]. In light of the recent discovery of testicular monocytes, more studies should be undertaken to determine any specific functions of monocytes versus macrophages in the testis and the ovary.

Mast cells

Another myeloid cell that has been implicated in testicular function, and more specifically male infertility, is the mast cell. Historically, mast cells are known for their roles in allergy and inflammation [138], but their function within the testis is unclear. Human studies have revealed an increase of mast cells within the testis of infertility patients and a shift of mast cell localization from the interstitium to the seminiferous tubule peritubular wall [139–141], suggesting that inflammation mediated by mast cells, as well as promotion of testicular fibrosis [142], may be an important underlying factor in male infertility. Few studies have focused on the role of mast cells in normal gonadal function, but it has been proposed that specific mast-cell-derived secreted factors, such as tryptase, chymase, and histamine, play roles in multiple aspects of testicular and ovarian function [143–147]. In particular, mast cells have been implicated in testicular steroidogenesis in mice and hamsters [148, 149]; however, molecular mechanisms of this process remain to be elucidated. Further complicating dissection of mast cell function in the testis is the finding that there is great variation in testicular mast cell number among mammalian species [150].

Much less is known about mast cells in the ovary [145]. However, several pieces of evidence indicate that there is a link between hormones and ovarian mast cells, such as hormone-mediated mast cell activation and recruitment, dynamic mast cell number in the ovary at various points of estrus cycle in diverse animal species, and a loss of mast cells in the postmenopausal ovary [151–155]. Of note is the observation that rodent ovaries rarely contain mast cells, unlike human and bovine ovaries, and instead mast cells are mostly limited to the hilum of the rodent ovary [154], indicating potential species-specific functions for ovarian mast cells. Overall, future studies are needed to uncover the specific roles of mast cells in reproductive biology, both in normal and pathological conditions.

Lymphocytes

Lymphocytes are comprised of T cells, B cells, and natural killer (NK) cells, which play vital roles in immune function mainly within the adaptive immune response. While hematopoietic progenitors with lymphoid potential are detected in the mouse yolk sac and embryo even before sex determination takes place [156, 157], definitively committed or differentiated lymphoid cells are not appreciably present in the embryo until around E14.5 [158, 159]. Relatively little is known about the origins and dynamics of lymphocyte colonization in the gonad. Consistent with the kinetics and timing of their differentiation in the embryo, lymphocytes were not detected in the E13.5 fetal mouse gonad [11], and it is not as of yet clear when, or even if, lymphocytes colonize the fetal gonad. Testicular studies of human and other species have reported that, under normal conditions, there are relatively few T cells (as compared to macrophages) and virtually no B cells in the adult mammalian testis [160, 161], likely a result of the immunosuppressed testicular environment.

The function of lymphoid populations in gonad morphogenesis and differentiation has not been definitively addressed. Most research into gonadal lymphocytes has focused on the role of T cells in adult testicular autoimmune pathologies [162, 163] and in ovarian cancer [164]. However, lymphocyte-deficient mice, such as Rag1-mutant males and females that lack mature B and T cells, were reported to be fertile [165, 166], indicating that mature lymphocytes, under normal conditions, are not required for fertility or gonad organogenesis. One caveat is that in-depth studies of reproductive physiology and development of lymphocyte-deficient mice have not been performed, so subtle defects in gonad development or fertility may occur but have not yet been reported.

Roles of vasculature and perivascular mesenchymal stem cells in gonadogenesis

Sex-specific development of vasculature during gonadogenesis

Testes and ovaries arise from a common precursor, the bipotential gonad, which possesses similar primitive vasculature before sex determination in both XX and XY gonads. Despite this common beginning, gonadal vascular formation occurs in a sex-specific pattern. At E11.0, blood vessels are apparent at the gonad-mesonephric border of both XX and XY gonads, but only a few endothelial cells have crossed the border into the gonad [167]. At E11.5, shortly after Sry is expressed in Sertoli cells in the XY gonad, individual endothelial cells released from the breakdown of mesonephric vasculature migrate into the XY gonad and surround developing testis cords [167–171]. Between E11.5-E12.5, mesonephros-derived endothelial cells reaggregate and remodel to form the new testis-specific coelomic artery, concurrent with the formation of testis cords. The establishment of a new testicular arterial network is critical for the export of androgens produced by Leydig cells, as well as for partitioning Sertoli cells into distinct cord structures (see below). At the same time that new testicular surface vasculature is forming, no remnant vascular plexus or large vessels can be observed near the gonad-mesonephric border [169–171]. In contrast, mesonephric vasculature in the XX gonad remains intact, undergoing minimal local reorganization, and mesonephric cells do not migrate into the gonad as observed in the testis. These less dramatic vascular events do not result in any obvious vascular pattern in the early fetal ovary as compared to the fetal testis [171].

Fetal testis vascularization

Expression of Sry in pre-Sertoli cells triggers organogenesis of the testis, which includes de novo vascularization resulting from endothelial cells released from breakdown of mesonephric vasculature and remodeling of these cells into a testis-specific vascular network. There are a number of molecular pathways and cellular players that have been implicated in vascular development of the testis (Fig. 3):

Fig. 3. Different pathways associated with fetal testis vascularization.

This diagram represents the six key players involved in vascularization during development of the fetal testis. Dashed line indicates areas needing further study. Arrows indicate a positive influence and T-shaped lines indicate an inhibitory effect. WNT, wingless-type MMTV integration site family, member 4; FST, follistatin; SRY, sex determining region on Y chromosome; FGF9; fibroblast growth factor 9; SOX9, SRY (sex determining region Y)-box 9; AMH, anti-Mullerian hormone; VEGFA, vascular endothelial growth factor a; FLT1, VEGFR1; KDR, VEGFR2; PDGFB, platelet-derived growth factor b; PDGFRA/B, platelet-derived growth factor receptor a/b; PROKR2, Prokineticin receptor 2; CDH5, cadherin 5/VE-cadherin.

Vascular endothelial growth factor (VEGF)

VEGF is a likely candidate for the regulation of angiogenesis by stimulation of endothelial cell proliferation and migration through the activation of its receptors, tyrosine kinases FLT1 (VEGFR1) and KDR (FLK/VEGFR2) [172]. In XY gonads, Vegfa, encoding a VEGF ligand, is expressed throughout the XY gonad, especially in the coelomic surface domain, whereas it is absent in the coelomic domain of XX gonads. Both VEGF receptors, FLT1 and KDR, are expressed within microvasculature and the male-specific coelomic artery. Systemic inhibition of VEGF signaling is embryonic lethal, as systemic mutations of either VEGF receptor are lethal prior to testis differentiation (i.e., well before E11.5 in mice) [173, 174]. BV13, a monoclonal blocking antibody directed to the extracellular region of CDH5 [175], or VEGF-Trap (aflibercept), a specific reagent that contains critical domains of FLT1 and KDR that bind and inhibit secreted VEGFA [176, 177], are used in ex vivo organ culture methods to overcome the early embryonic lethality and/or rapidly induced lethality in vivo of systemic or conditional mutations in genes encoding VEGF receptors. A Vegfa-lacZ reporter allele and cell-type-specific gene expression profiling indicated that VEGFA is likely mostly secreted from the interstitial mesenchyme in the fetal testis, suggesting an indirect influence of vasculature on cord formation. Inhibiting VEGFA activity (via VEGF-Trap) blocks the migration of mesonephros-derived endothelial cells, resulting in a strong blockade of male-specific vascular development. Real-time live imaging of whole XY gonads showed that vascular migration triggers the expansion of interstitial mesenchymal cells at the coelomic surface. Following BV13 treatment in XY gonads at E11.5, interstitial cells failed to proliferate, which is required for the expansion of the coelomic domain [176]. It is now clear that endothelial cells influence interstitial mesenchymal cell proliferation, though the molecular mechanism requires further clarification.

Platelet-derived growth factor (PDGF)

Vessel-derived PDGFs and their receptors PDGFRA and PGFRB (encoded by Pdgfra and Pdgfrb) have important roles in migration, proliferation, and differentiation of cells during embryonic development [178]. Generally, PDGFA, PDGFB, and PDGFC, are expressed in Sertoli cells, endothelial cells, and mesonephros/gonad boundary in XY gonads, respectively [179]. Although PDGFA is strongly expressed in Sertoli cells within testis cords, Pdgfa-null embryos show normal fetal testis morphology [180], indicating that Pdgfa is either dispensable or is redundant with other PDGF ligands. Both Pdgfra and Pdgfrb are weakly expressed in the mesenchyme of the mesonephros and are expressed in the coelomic region of E11.5 XY gonads. Later, they are strongly expressed in the mesenchyme of the gonadal interstitium [179]. Pdgfra null mutation leads to defects in testis cord formation and vascular development in XY gonads, whereas knockdown of Pdgfrb does not result in overt defects in early XY or XX gonad development [179]. BV13 treatment in XY gonads results in a slight reduction of Pdgfb expression instead of Pdgfra. The addition of recombinant PDGF-BB rescued mesenchymal cell proliferation after inhibition of VEGFA or blockade of endothelial cells, although testis cord aggregation was not evident [176]. Taken together, these reports demonstrate the failure of interstitial proliferation in the absence of vasculature might be due to defects in PDGF signaling.

Macrophages

Macrophages are among the most prevalent cells within the testicular interstitial compartment. Various cell ablation models targeting macrophages have revealed two important functions of testicular macrophages: testicular Leydig cell steroidogenesis [181] and the self-renewal or differentiation of spermatogonial stem/progenitor cells (SSCs) [7]. Initial studies revealing critical roles for macrophages utilized Csf1 mutant mice, which have reduced macrophage number in most tissues. Csf1 mutant mice showed reductions in mating capability, low sperm count, and 90% lower serum testosterone levels, suggesting that testicular macrophages are vitally important for male reproductive function [182, 183]; additionally, CSF1 has been implicated as an extrinsic factor that promotes self-renewal of spermatogonial stem cells in vitro [184]. However, given the widespread loss of macrophages throughout the body in most targeting strategies, it is still somewhat unclear how much of the deleterious effects of macrophage depletion (and Csf1 loss) upon reproductive function is due to local (i.e., testis-specific) versus systemic loss of macrophages.

Macrophages are initially localized near the developing vascular plexus in the bipotential gonad and have been implicated in vascular remodeling. Depletion of fetal testis macrophages via a macrophage-specific Cx3cr1-mediated cell ablation method, instead of Csf1 mutation, thus avoiding compensation of maternal circulating CSF1 [185], resulted in disorganized surface vasculature and aberrant testis cord architecture (see also above in “Role of macrophages during fetal testis vascularization” section) [11]. These findings indicate that macrophages play an important role in testis-specific vascularization and morphogenesis.

Activins

Activins, consisting of either homodimers or heterodimers of inhibin beta A (INHB A) and beta B (INHBB) subunits (activin A=beta A:beta A, activin B=beta B:beta B, and activin AB=beta A:beta B), are involved in the patterning and organogenesis of vertebrate embryos, including within the reproductive organs. Inhbb expression is similar in XY and XX gonads at E11.5, at the onset of testis-specific vascularization, but is down-regulated by the WNT/CTNNB1 (β-catenin)/follistatin (FST) pathway in the XX gonad at E12.5 [186, 187], Knockout of Inhbb shows defects of testis-specific vasculature in the XY gonad, with varying degrees of vascular malformation and decreased vascular branching. Inhbb knockout in XX gonads, in the context of XX Wnt4- and Fst-knockout gonads that normally exhibit ectopic male-like vasculature, leads to reduced ectopic vasculature, indicating that Inhbb function is required in XX Wnt4- and Fst-knockout gonads for the ectopic formation of testis-like vasculature. Furthermore, exogenously administered activin B is able to induce ectopic testis-like vasculature in the ex vivo cultured wild-type fetal ovary, indicating that activin B contributes to coelomic vessel formation independent of the SRY pathway [187]. Ctnnb1 conditional knockout or Wnt4 mutant XX gonads, as well as fibroblast growth factor 9 (FGF9)-treated cultured XX gonads (which stabilizes SOX9 expression in XX gonads), also show ectopic male-like vasculature, indicating that vascular remodeling is part of normal testicular development and must be actively repressed by the ovarian developmental program [188–191].

Prokr2

Prokineticin receptor 2 (Prokr2), which represents 1 of the 2 receptors of the prokineticin signaling pathway [192], is associated with gonadotropin releasing hormone (GnRH)-neuron migration [193] and endothelial vessel integrity and permeability [194]. Prokr2 displays male-specific expression in interstitial cells of the developing testis, but not in ovaries. Prokr2 is significantly up-regulated starting at E11.5, and its expression peaks at E12.5, corresponding with the critical time of vascular remodeling in fetal testis. Loss of Prokr2 specifically compromises the integrity of the testis vasculature at E16.5 [195], suggesting that Prokr2 promotes fetal testis vascularization.

Significance of vasculature for testis development

Between E11.5-12.5, Sertoli cells surround germs cells to form testis cords [196]. Several studies revealed that vascular endothelial cells drive testis cord formation [171, 175, 176]. Time-lapse organ culture experiments demonstrated that mesonephros-derived endothelial cells migrate into the gonad within restricted tracts, thus subdividing Sertoli cells into presumptive testis cords before they are morphologically distinct [171]. Vegfa is expressed specifically by the undifferentiated mesenchyme, and its receptor tyrosine kinases, KDR and neuropilin 1 (NRP1), are expressed in the male-specific coelomic vessel and microvasculature of the gonad [176]. VEGF pathway inhibitor (VEGF-Trap) and CDH5 (VE-cadherin) blocking antibody (BV13) experiments, which disrupted initial vascular remodeling starting at E11.5, revealed that vascular-mesenchymal interactions initiate testis cord morphogenesis [175, 176]. However, later vascular ablation at E12.5 did not affect existing testis cord structure [197]. Taken together, these results indicate that vasculature is essential for the initiation of testis cord formation but is dispensable for the maintenance of testis cord architecture (Fig. 4).

Fig. 4. Roles of vasculature in fetal testis morphogenesis and differentiation.

Model of sex-specific vascular development during gonadogenesis and illustration of the mechanisms through which vasculature influences fetal testis development. Note that little is known about: vasculature’s role in fetal ovarian organogenesis; the role of perivascular cells in the fetal ovary; and whether active Notch signaling is involved in ovarian organogenesis, tc, testis cord.

Even though vascular disruption in ex vivo culture experiments did not induce hypoxia and did not affect testis cord morphology and Sertoli cell differentiation, it did result in excessive Leydig cell differentiation within the E12.5 XY gonad [197]. These interactions are likely mediated by cell-cell contact between endothelium and mesenchymal cells, which are required for Notch signaling. In mammals, the Notch signaling pathway mainly consists of Notch receptors (NOTCH1, NOTCH2, NOTCH3, and NOTCH4), Notch ligands (Delta like (DLL) 1 (DLL1), DLL3, DLL4, Jagged (JAG) 1 (JAG1), and JAG2) and intracellular effector molecule (CBF1, officially named RBPJ (recombination signal binding protein-j)) [198, 199]. Notch family member genes normally associated with angiogenesis [198], such as Dll4, Jagged2, Notch1, and Notch4, are specifically enriched in endothelial cells of the XY gonad [200, 201], Overexpression of the NOTCH1 intracellular domain in the XY gonad results in a loss of Leydig cells but does not affect Sertoli cells [202]. Notch2 is enriched in Sertoli cells and interstitial cells at E12.5 and E13.5, while Notch3 is only slightly expressed in interstitial cells. In addition, Jagged1 is specifically expressed in vasculature-associated cells in the interstitium of the fetal and postnatal testis. However, Dll1 and Dll3 are not detectable in the fetal gonad [200].

Active Notch signaling (via the receptor NOTCH2) in perivascular mesenchymal cells, driven by endothelial Notch ligands (likely DLL4), maintains interstitial mesenchymal progenitor cells, and restricts their differentiation into fetal Leydig cells [197, 202, 203]. Blocking Notch signaling via the γ-secretase inhibitor DAPT results in a failure of progenitors to either self-renew or maintain themselves in an undifferentiated state in XY gonads [197, 202]. Disruption of vasculature results in loss of active Notch signaling in the interstitium but does not affect active Notch signaling in Sertoli cells [197], indicating that Notch signaling between vasculature and mesenchyme regulates the Leydig progenitor niche. While Notch signaling has been well-characterized as a regulator of endothelial cell differentiation and angiogenesis [198], studies of ex vivo inhibition of Notch signaling in the fetal testis have not reported any defects in testis-specific vascular development [197, 202]. These findings indicate that during early testicular differentiation, communication via Notch is largely unidirectional from the endothelium to the mesenchyme. However, it is possible, and perhaps likely, that Notch signaling is required for gonadal vascular development prior to or after stages examined (E11.5-E13.5) in previous studies.

Nestin, an intermediate filament protein that marks many stem/progenitor populations [204], is expressed adjacent to endothelial cells in both fetal and adult testes [197, 205], indicating an interaction between vasculature and perivascular cells. A recent study revealed an established function of fetal testicular vasculature in providing a specialized niche that controls the maintenance and differentiation of perivascular Nestin+ stem cells, which in the fetal testis are a multipotent progenitor population that gives rise to Leydig cells, smooth muscle cells, and pericytes. Following vascular disruption in E12.5 XY gonads, which does not have an influence on testis cord morphogenesis at that stage, Nestin+ cells, which were regulated via active Notch signaling, failed to be maintained and excessively differentiated into Leydig cells [197]. In all, blood vessels play a critical role in regulating specialized microenvironments that are directly associated with testis development. Therefore, it appears that vasculature regulates the balance between maintenance and differentiation of Leydig cell progenitors in the fetal testis (Fig. 4).

Perspectives

Blood vessels and immune cells have been classically defined by their supportive roles during organ development and homeostasis. However, an increasing body of evidence supports the emerging idea that blood vessels and immune cells are important developmental regulators of tissue differentiation and cell specification. Here we have outlined some examples of how these constituents of the gonadal interstitial compartment contribute to sex-specific organ morphogenesis and function. To gain a deeper understanding of the role of these cells in organogenesis of the reproductive organs, several areas of investigation will likely lead to new discoveries revealing the diverse functions of vascular and immune cells.

With regard to immune cells, it is critical to address the mechanism underlying how macrophages promote vascular remodeling during fetal testis morphogenesis: is it due to a co-opting of angiogenic pathways, similar to what it is seen during tumorigenesis? Do macrophages degrade existing vasculature via their well-known phagocytic capacity? Is the extracellular matrix modified in a sex-specific manner by macrophages either physically or biochemically? The diversity of macrophage phenotypes uncovered by developmental biologists and immunologists suggest that macrophages can act in a multitude of ways. This diversity in macrophage phenotype may be due to their various potential cellular ontogenies, such as yolk sac versus fetal liver versus bone-marrow-derived origins. These distinct origins may impart unique characteristics to cells, but more likely the tissue microenvironment may drive specific macrophage identities. For example, do different adult testicular macrophage populations have distinct origins, and how are their cellular functions unique, for example, regulating other interstitial cells (e.g., vasculature, Leydig cells) versus spermatogenic cells?

Another outstanding question that needs to be addressed is the sex-specific nature of immune cells in the gonad. Do they have similar origins and kinetics of tissue colonization and turnover in the testis versus the ovary? What cytokines and chemokines do testis and ovarian macrophages release (or respond to?) to influence angiogenesis and lymphangiogenesis in the gonad? Do macrophages regulate ovarian blood and lymphatic vessels to support follicular growth and development; given the synchronous increase of all three cell types from birth to adulthood, how are they linked to one another? Addressing all these questions will likely require in-depth high-throughput techniques, such as single-cell studies of the transcriptome, epigenome, and proteome. Overall, this review has highlighted a particular lack of knowledge regarding fetal ovarian macrophages and their contributions to ovarian organogenesis and function relative to their testicular counterparts.

Regarding vasculature’s role in the fetal testis and its contributions to testis morphogenesis and differentiation, there are several outstanding questions in the field have not been addressed, e.g., identifying disrupted genes and pathways in mesenchymal and other cell types after vascular depletion in the fetal testis, or the role of endothelial-derived cytokines and growth factors in these processes. In contrast to the testis, our knowledge of the role of vasculature and perivascular cells in ovarian development is relatively limited, representing a major knowledge gap in the field. There are several areas of future research that will likely be productive, including examining the molecular mechanism of vascularization in the fetal ovary, elucidating interactions between endothelium and mesenchymal cells during ovarian organogenesis, and identifying how vasculature influences breakdown of oocyte cysts and follicle formation during ovarian development.

All of the above questions will help us understand the previously underappreciated roles of immune and vascular cells in gonadogenesis and fertility. Given that no organ exists in a vacuum, complex cellular interactions in the gonad likely exist on both local and systemic axes. Knowledge that we can gain about various gonadal cell types will likely prove useful to understand and treat medical conditions of the reproductive system such as gonad dysgenesis and infertility.

Abbreviations:

AGM

aorta-gonad-mesonephros region

ARG1

arginase 1

BM

bone marrow

BMP

bone morphogenetic protein

CD

cluster of differentiation

CDH5

cadherin 5 (VE-cadherin)

CSF1

macrophage colony-stimulating factor

CTNNB1

catenin (cadherin associated protein), beta 1 (β-catenin)

CX3CR1

chemokine (C-X3-C motif) receptor 1

CXCL

chemokine (C-X-C motif) ligand

DLL

delta like canonical Notch ligand

EDS

ethane dimethanesulphonate

EGF

epidermal growth factor

EMPs

erythro-myeloid progenitors

FGF9

fibroblast growth factor 9

FL

fetal liver

FLT1

FMS-like tyrosine kinase 1

FST

follistatin

GM-CSF

granulocyte-macrophage colony-stimulating factor (CSF2)

GnRH

gonadotropin releasing hormone

HSCs

hematopoietic stem cells

IGF

insulin-like growth factor

IL

interleukin

INHBA

inhibin beta A

INHBB

inhibin beta B

KDR

kinase insert domain protein receptor (VEGFR2)

LPS

lipopolysaccharide

LYVE1

lymphatic vessel endothelial hyaluronan receptor 1

MACs

myeloid angiogenic cells

M1

classically activated macrophage

M2

alternatively activated macrophage

MHCII

major histocompatibility complex class II receptor

MMP

matrix metalloproteinase

PDGF

platelet-derived growth factor

PG

prostaglandin

PROKR2

prokineticin receptor 2

PROX1

prospero homeobox 1

RBPJ

recombination signal binding protein for immunoglobulin kappa J region

SOX9

SRY (sex determining region Y)-box 9

SRY

sex determining region of chromosome Y

SSCs

spermatogonial stem cells

TGF-β

transforming growth factor beta

TLR

Toll-like receptor

TM

testicular macrophages

TNFA

tumor necrosis factor alpha

VEGF

vascular endothelial growth factor

WNT4

wingless-type MMTV integration site family, member 4

YS

yolk sac