**1. Introduction**

Human brucellosis, a zoonotic disease mostly caused by *Brucella melitensis*, *B. suis* and *B. abortus*, affects over 500,000 people each year around the world [1]. The infection can be found in several domestic animals (cattle, sheep, goats, pigs, and dogs) and in some wild species. Transmission to humans usually occurs by contact with infected animal tissues and consumption of dairy products.

The clinical manifestations of human brucellosis are usually linked to inflammatory phenomena in the affected organs [2]. Involvement with the reproductive organs is common in animals, which frequently present problems such as abortion and perinatal death. Studies performed in animals have shown that placental *Brucella* infection is accompanied by the infiltration of inflammatory cells [3,4]. The fact that placental inflammatory responses are involved in infection-triggered abortion by several pathogens [5–7] suggests that placental inflammation may also have a role in *Brucella*-induced abortion.

Abortion due to *Brucella* infection has been also reported in humans, with an incidence that ranges from 7% to 40% according to different studies [8–10]. Among pregnan<sup>t</sup> women who presented with acute brucellosis at a Saudi Arabian hospital, 43% had spontaneous abortion during the first and second trimester, and 2% in the third trimester [11]. In spite of the importance of *Brucella*-related abortion, the pathophysiology of this complication in humans is largely unknown. Recent studies have shown that *Brucella* spp. can infect and replicate in human trophoblasts, and that the infection elicits a proinflammatory response [12,13]. These trophoblastic inflammatory responses may be relevant to the pathogenesis of abortion in human brucellosis. However, the potential of other placental cell populations to contribute to an inflammatory environment during *Brucella* infection has not been explored.

For several microorganisms that reach the placenta by the hematogenous route, including *Brucella abortus*, in vivo studies in animal models have indicated that the maternal decidua is the initial site of placental colonization [14,15]. Decidualization of the endometrium, a process essential for successful implantation and maintenance of pregnancy, involves progesterone-driven morphological and biochemical changes of fibroblast-like endometrial stromal cells (ESCs) to differentiate into decidual stromal cells (DSCs). These DSCs are characterized by the secretion of prolactin, insulin growth factor-binding protein and several cytokines that act as regulators of the innate immunity [16].

Given the relevance of the decidua as the initial site of placental colonization for several hematogenously spread infections, the ability of decidual cells to respond to pathogens is especially relevant. Primary DSC and ESC cell lines have been shown to express several Toll-like receptors (TLRs) and Nod-like receptors (NLRs), and respond to pathogen-associated molecular patterns (PAMPs) with an enhanced production of matrix metalloproteinases (MMPs) and proinflammatory cytokines including MCP-1, IL-6, IL-8, IL-1β, and CCL5 (RANTES) [17]. At least for group B streptococcal infection the cytokine response of endometrial stromal cells is modulated by decidualization, so that decidualized cells produce IL-6, TNF-<sup>α</sup>, IL-10, and TGF-β while non-decidualized cells do not [18].

In addition to decidual stromal cells, the decidua also contains significant proportions of immune cells, including macrophages, natural killer cells, dendritic cells, and T cells [19]. Early pregnancy is considered to resemble an open wound which requires a strong inflammatory response, thus the first trimester is considered a proinflammatory phase, which turns to an anti-inflammatory phase in the second trimester [20,21]. Although decidual macrophages exhibit an M2 phenotype and exert an immunosuppressive effect on local lymphocyte populations, in the context of local infection they may increase their production of proinflammatory cytokines and contribute to pregnancy disorders [19]. Of note, DSC or ESC have been shown to interact with macrophages in several ways [22,23]. In response to stimulation with lipopolysaccharide (LPS) from *Escherichia coli*, a coculture of ESC and PMA-differentiated THP-1 cells (human monocytes) produced enhanced levels of many cytokines (IL-1β, RANTES, MCP-1, IL-10, TGF-β, MIC-1, G-CSF) as compared to the respective monocultures [24]. Importantly, *B. abortus* is known to survive and replicate in macrophages from several animal species, inducing the secretion of proinflammatory cytokines [25–27].

The T-HESC cell line, derived from normal primary human ESC by telomerase immortalization, has been widely used to study several aspects of human ESC biology, including infection and cytokine production [23,24,28–31]. T-HESC are karyotypically, morphologically, and phenotypically similar to the primary parent cells, and after treatment with estradiol and medroxyprogesterone acetate (MPA) display the morphological and biochemical pattern of decidualization [32]. In the present study we evaluated the ability of *Brucella* spp. to infect and survive in decidualized T-HESC, and also assessed the cytokine production induced in these cells by the infection or by their interaction with infected macrophages.
