Abstract
Early embryonic mortality accounts for a substantial portion of reproductive failure in agriculturally important livestock, including the dairy cow. The maintenance of early pregnancy requires a fully functional corpus luteum (CL) that is not susceptible to regression following fertilization, yet the cellular mechanisms of luteal regression are not clearly understood. Immune-cell accumulation within the CL at the time of regression is a well-documented phenomenon in a variety of species. In the dairy cow, immune-cell accumulation precedes luteal regression by several days and coincides with an increase in expression of the chemokine monocyte chemoattractant protein 1 (CCL2), suggesting that immune-mediated events promote tissue destruction. Recent studies indicate that endothelial cells comprising the CL are a primary source of CCL2 secretion. Moreover, although uterine-derived prostaglandin F2α (PGF) initiates luteal regression in the cow, PGF does not directly provoke CCL2 secretion by luteal endothelial cells. Instead, PGF-induced luteal regression is thought to require cooperative interaction among immune cells, endothelial cells, and steroidogenic cells of the CL to further promote CCL2 secretion, enhance immune-cell recruitment, and eliminate luteal tissue. This brief review focuses on putative interactions between immune cells and endothelial cells derived from the bovine CL that result in enhanced CCL2 expression and the elaboration of other inflammatory mediators (for example, cytokines), which perpetuate luteal regression. Fundamental knowledge of immune-endocrine interactions within the reproductive system of cows has relevance to other CL-bearing mammals, including humans and endangered animals, particularly in the development of methods to control and/or improve fertility. Thus, it is a timely topic for this symposium concerning ecological immunology and public health.
Introduction
Although the general public remains largely unaware of it, the dairy industry in the United States is currently facing a crisis. The decline of the family farm and continued pressure to relinquish farm land for suburban development have emerged as challenges to the agricultural economy, but perhaps more troubling, and certainly less obvious, is the knowledge that fertility in dairy cows is declining. Estimates of pregnancy rate at first insemination in lactating dairy cows currently ranges from 35 to 40% (Lucy 2001), suggesting that, much like the public concern surrounding infertility in humans and endangered animals, infertility in dairy cows has emerged as an important concern to the dairy industry.
According to reports by the United States Department of Agriculture (United States Department of Agriculture 1996, 2002), infertility costs the dairy industry more than $350 million per year and accounts for more than a quarter of the cows culled annually from milk-producing herds. Reasons for the increased incidence of reproductive failure in cows include environmental factors, metabolic factors (for example, nutritional stress), and factors pertaining to reproductive management that lead to irregular ovarian function, such as ovulation failure and inadequate hormonal support for pregnancy. Maintenance of early pregnancy in the dairy cow, as with most mammals, is dependent upon the hormonal support provided by the corpus luteum (CL). The CL is a transient endocrine structure within the ovary that forms immediately following ovulation and produces progesterone to sustain embryonic development. In the absence of pregnancy, however, the CL regresses in response to uterine-derived prostaglandin F2α (PGF), ensuring the onset of the next reproductive cycle and ovulation. This process of CL destruction, known as luteal regression, is remarkably efficient. In the cow, the mature CL typically consists of 8–12 g of tissue that is virtually eliminated within 4–6 days following the onset of luteal regression. The process of luteal regression is also highly organized, yet the cellular and molecular events involved and the critical aspects of regulation are not clearly understood. Knowing that premature luteal regression and inadequate progesterone support inevitably result in loss of pregnancy and reproductive failure in cattle (Inskeep 2004), further insight concerning the process and regulation of luteal regression is needed.
The cow is an ideal animal model in which to investigate the process of luteal regression. The bovine CL has sufficient mass and cell numbers to permit extensive experimental study using both in vivo and in vitro methods. It also consists of the same cell types (that is, endothelial cells, steroidogenic cells, and immune cells) in relatively the same proportions as CL of most other mammals. Lastly, and of relevance to this symposium, luteal regression in the cow involves immune-mediated mechanisms, specifically cell–cell interactions between immune cells (especially monocytes and macrophages) and endothelial cells of the CL, which are thought to promote destruction of the tissue. In this review, interactions between immune cells and endothelial cells are postulated to influence luteal regression and thus may impact fertility in the cow.
Immune-mediated aspects of luteal regression
Regression of the bovine CL is initiated by uterine secretion of PGF (Niswender and others 2000), yet it is also clear that this process is accompanied by immune-mediated events. Specifically, expression of chemokines, increased accumulation of immune cells, and augmented expression of major histocompatibility complex (MHC) molecules within the CL occur as regression becomes imminent. Monocyte chemoattractant protein 1 (chemokine ligand 2 or CCL2) is one of the chemokines that attracts monocytes and T lymphocytes to the CL during luteal regression. CCL2 is produced by a variety of cells, particularly endothelial cells, and is known to activate immune cells by stimulating their differentiation (Matsushima and others 1992). In the bovine CL, endothelial cells are a potent source of CCL2 (Goede and others 1999; Cavicchio and others 2002; Liptak and others 2005), and CCL2 expression increases prior to luteal regression (Townson and others 2002). Concomitantly, increases in monocyte/macrophage accumulation occur within the CL (Penny and others 1999; Townson and others 2002) and these effects are associated with an upregulation of MHC-class-II-positive cells within the bovine CL (Benyo and others 1991). Currently, it is not clear whether the increase in MHC-class-II expression is attributable to endothelial cells, immune cells, steroidogenic cells, or possibly all 3 cell types (Cannon and others 2006). Nevertheless, the increase in MHC-class-II expression is thought to enhance T-lymphocyte proliferation (Petroff and others 1997) and possibly influence immune-cell activation (Cannon and others 2006). Collectively, these observations implicate CCL2 expression in enhanced immune-cell recruitment and activation within the CL and suggest that mechanisms of immune response are in place prior to the onset of luteal regression.
PGF-induced luteal regression and interactions of immune cells and endothelial cells
It is noteworthy that although PGF-induced luteal regression is accompanied by increased CCL2 expression (Tsai and others 1997; Haworth and others 1998; Penny and others 1998; Tsai and Wiltbank 2001), there is no evidence that PGF directly stimulates CCL2 synthesis and secretion. Endothelial cells are a rich source of CCL2 within the CL (Goede and others 1999; Townson and others 2002), but there is disagreement as to whether these cells possess PGF receptor. Several studies report that endothelial cells of the bovine CL have PGF receptors, respond directly to PGF by secreting the vasoconstrictor endothelin 1, and thus mediate PGF-induced effects (Girsh and others 1996; Levy and others 2001; Meidan and others 2005). Others, however, indicate that endothelial cells of the bovine CL lack PGF receptors and do not produce CCL2 in direct response to PGF (Cavicchio and others 2002; Liptak and others 2005; Pru and others 2003). The inconsistency among these studies regarding PGF receptors and direct PGF effects may be explained in part by the phenotypic diversity of the endothelial cells comprising the bovine CL, in which some endothelial cells possess PGF receptors and others do not (Spanel-Borowski and van der Bosch 1990; Spanel-Borowski 1991; Spanel-Borowski and Fenyves 1994; Davis and others 2003). Still, it is clear that immune-cell-derived cytokines, particularly tumor necrosis factor alpha (TNF) and interferon gamma (IFNG), have direct effects on endothelial cells of the CL, implicating interactions of immune cells and endothelial cells in luteal regression. Questions remain, however, as to how an initial stimulus by PGF leads to immune-response events, including immune cell–endothelial cell interactions, to then hasten the process of tissue destruction. A conceptual model is depicted illustrating the scope of the problem (Fig. 1).
Initial conceptual model depicting distinct regulatory mechanisms observed (solid arrows) and postulated (dashed arrows) for chemokine expression (CCL2) within the bovine CL during the process of luteal regression. Encircled areas indicate topics of most recent experimental study. See text for further description. PGF = prostaglandin F2α; IFNG = interferon gamma; TNF = tumor necrosis factor alpha; CCL2 = chemokine ligand 2 or monocyte chemoattractant protein 1.
It is conceivable that resident immune cells within the CL mediate PGF-induced events, including CCL2 expression by endothelial cells, which can lead to further immune-cell recruitment, the continuation of an inflammatory response, and ultimately luteal regression (Fig. 1). The cytokines TNF and IFNG directly provoke CCL2 secretion by, and apoptosis of, endothelial cells (Fig. 1) (Cavicchio and others 2002; Pru and others 2003), suggesting that interactions of immune cells and endothelial cells account for increases in CCL2 and cell death during PGF-induced luteal regression. Until recently, however, this potential relationship between immune cells, PGF, and the regulation of endothelial cell expression of CCL2 within the bovine CL had not been examined. Moreover, interactions between immune cells and luteal steroidogenic cells could also be significant for luteal function (Fig. 1) but have not been extensively investigated.
Activated immune cells potentiate endothelial cell secretion of CCL2
In a series of coculture experiments involving combinations of endothelial cells, immune cells, and steroidogenic cells derived from bovine CL (that is, luteal steroidogenic cells), we recently determined that endothelial cells are the major source of CCL2 production within the bovine CL (Liptak and others 2005). Immune cells (specifically peripheral blood mononuclear cells or PBMCs) provoked further secretion of CCL2 by the endothelial cells, especially when concanavalin A-activated PBMCs were introduced into the coculture. Activation of PBMCs was verified by detection of IFNG in the study. CCL2 secretion in PBMC-endothelial cell cocultures was 5- to 12-fold greater than cultures of endothelial cells or PBMCs alone. Importantly, PGF did not influence these effects; the endothelial cells and PBMCs were not directly responsive to PGF nor did they express mRNA for the PGF receptor (Liptak and others 2005). Production of CCL2 was greatest in cocultures that permitted physical contact between endothelial cells of the CL and PBMCs, although further study is needed to determine which aspects of physical contact (for example, adhesion molecules, MHC molecules) are most critical. Others have observed similar increases in CCL2 following coculture of immune cells with various epithelial cell types (Andjelkovic and others 2000; Hao and others 2003; Hisada and others 2000). Hence, our results indicate that physical interactions between activated PBMCs and endothelial cells of the bovine CL enhance CCL2 expression, but that PGF does not influence these interactions. IFNG and/or soluble products of activated PBMCs may be required to elicit this response.
Activated immune cells impair luteal steroidogenic cell secretion of progesterone
Coculture experiments similar to those described above were also conducted utilizing PBMCs and luteal steroidogenic cells. Notably, progesterone secretion by the steroidogenic cells was inhibited in cultures containing activated PBMCs (Liptak and others 2005). Acute treatment with PGF enhanced progesterone production in all steroidogenic cell cultures but did not alter the inhibitory effect of activated PBMCs in the cocultures. These findings are consistent with those of others who found that cytokines such as TNF and IFNG inhibit luteal progesterone production in vitro (Fairchild and Pate 1991; Benyo and Pate 1992). Here, the identity of the anti-steroidogenic factor(s) secreted by the activated PBMCs was not determined. IFNG, however, is a strong candidate molecule because its synthesis and secretion is increased in conconavalin-A-activated PBMCs (Liptak and others 2005). In any event, the discovery that activated PBMCs impair luteal progesterone production provides further incentive to identify immune-cell-derived products that regulate steroidogenesis within the CL in future studies. These results also reveal that luteal steroidogenic cells serve a direct role in PGF reception and possibly an intermediary role in immune response during luteal regression. Identifying the key mechanism(s) by which luteal steroidogenic cells respond to PGF to then influence immune cells, and possibly endothelial cells, within the CL will be critical for understanding PGF-induced luteal regression (Fig. 2).
Current conceptual model depicting regulatory mechanisms of chemokine expression (CCL2) within the bovine CL during the process of luteal regression based upon recent evidence (Liptak and others 2005). Dashed arrows indicate areas that require further investigation. See text for additional description. PGF = prostaglandin F2α; IFNG = interferon gamma; TNF = tumor necrosis factor alpha; CCL2 = chemokine ligand 2 or monocyte chemoattractant protein 1.
Activated immune cells and endothelial cells of bovine CL can interact cooperatively to enhance CCL2 expression and promote inflammatory events that lead to luteal regression (Fig. 2). These events include but are not limited to apoptosis of cells (that is, endothelial cells, luteal steroidogenic cells), recruitment of additional immune cells, and secretion of proinflammatory molecules (for example, TNF, IFNG, and CCL2). Activated immune cells also interact with luteal steroidogenic cells to inhibit progesterone production (Fig. 2) and potentially trigger cell death (Benyo and Pate 1992; Petroff and others 2001). It appears that luteal steroidogenic cells play the pivotal role in directly mediating PGF-induced events, including those augmenting immune-cell activation (Cannon and others 2003) and influencing endothelial cell function within the CL (Fig. 2).
In conclusion, our findings reinforce the complexity of cell–cell interactions among immune, endothelial, and steroidogenic cells within the bovine CL and raise questions concerning the direct actions of PGF on luteal steroidogenic cells to promote inflammatory events during luteal regression. These observations also support the concept that immune–endocrine interactions within the CL impact ovarian function and fertility in cows, which has relevance and may serve as a model to other CL-bearing mammals including humans and endangered animals. Understanding these immune-response mechanisms and their effects on ovarian function will be important for developing methods to control and/or improve fertility in the future.
This work was supported by USDA grant 2002-35203-12257 and is scientific contribution 2300 from the New Hampshire Agricultural Experiment Station. The author wishes to thank Jessica A. Cherry, Amy R. Liptak, and Brian T. Sullivan for their contributions to this manuscript.
Conflict of interest: None declared.
References
Author notes
From the symposium “Ecological Immunology: Recent Advances and Applications for Conservation and Public Health” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 4–8, 2006, at Orlando, Florida.
