**2. Results**

### *2.1. Interaction with Platelets Enhances the Activation of Endothelial Cells in the Context of B. abortus Infection*

We decided to evaluate the capacity of *B. abortus* to activate human brain microvascular endothelial cells (HBMECs) in presence of platelets. For this, HBMECs were co-cultured with platelets (cell:platelet ratio, 1:100) and infected with *B. abortus* (MOI of 100) for 24 h. As control, HBMECs were cultured with platelets alone or they were infected in the absence of platelets. We measured ICAM-1 (intercellular adhesion molecule 1, also known as CD54) expression to determine the level of activation of endothelial cells. Endothelial ICAM-1 plays a critical role at different steps of neutrophil migration into inflamed tissues [22]. As we have previously reported, infection with *B. abortus*in the absence of platelets induced a slight activation of HBMECs, measured as the upregulation of surface ICAM-1 [9]. The presence of platelets in the absence of infection also induced a slight activation of HBMECs. However, these effects were not significant. On the other hand, the presence of platelets during the infection of HBMECs induced a significant (*p* < 0.0005) upregulation of ICAM-1 (Figure 1A). These results demonstrate that the presence of platelets enhances *B. abortus*-induced activation of microvascular brain endothelial cells.

**Figure 1.** Interaction with platelets enhances the activation of endothelial cells in the context of *Brucella abortus* infection. Endothelial cells lines human brain microvascular endothelial cell (HBMEC) (**A**), human microvascular endothelial cells (HMEC-1) (**B**), and human umbilical vein endothelial cell (HUVEC) primary culture (**C**) were incubated with *B. abortus* (B.a.), in the presence or the absence of platelets (PTL), or with PTL alone for 24 h. ICAM-1 (intercellular adhesion molecule 1) expression on the cell surface was assessed by flow cytometry. Bars represent the mean ± SEM of duplicates. Data shown are from a representative experiment out of at least three performed. \*\* *p* < 0.005, \*\*\* *p* < 0.0005 vs. untreated cells (UT).

To investigate whether the effect of platelets during *B. abortus* infection also occurs in other endothelia, human microvascular endothelial cells (HMEC-1) and human umbilical vein endothelial cell (HUVEC) were infected with *B. abortus* in the presence or absence of platelets. The presence of platelets during the infection of both types of endothelial cells induced a significant upregulation of ICAM-1 expression compared to infected cells or cells incubated with platelets alone (*p* < 0.0005) (Figure 1B,C). These data demonstrate that the presence of platelets enhances *B. abortus*-activation of different endothelial cell types.

### *2.2. Supernatants from B. abortus-Stimulated Platelets Activate Brain Microvascular Endothelial Cells*

To investigate whether this effect was due to direct interaction between platelets and endothelial cells or factors released by *B. abortus*-activated platelets, we performed experiments using conditioned media. First, platelets were stimulated with or without *B. abortus* (platelets: *B. abortus* ratio, 1:1) for 24 h. Then, culture supernatants were collected and filtered to eliminate platelets and bacteria. Finally, cell-free supernatants were used to stimulate HBMECs for an additional 24 h. Stimulation of HBMECs with supernatants from *B. abortus*-stimulated platelets induced a significant (*p* < 0.005) upregulation of ICAM-1 surface expression (Figure 2A). These results demonstrate that supernatants from *B. abortus*-stimulated platelets are able to activate microvascular brain endothelial cells. Furthermore, in order to expand our results, we investigated whether these supernatants could also activate other endothelial cell types. For this, HMEC-1 and HUVEC were stimulated with supernatants collected from *B. abortus*-stimulated platelets. We observed an upregulation of ICAM-1 surface expression on both cell types (Figure 2B,C). Collectively, these data demonstrated that supernatants from *B. abortus*-stimulated platelet are able to activate several types of endothelial cells.

**Figure 2.** Supernatants from *B. abortus*-stimulated platelets activate endothelial cells. Supernatants collected from *B. abortus*-stimulated platelets (PTL + B.a.) or platelets alone (PTL) were used to stimulate the endothelial cell lines HBMEC (**A**), HMEC-1 (**B**), and HUVEC primary culture (**C**) for 24 h (dilution 1:2). ICAM-1 expression was measured on the cell surface by flow cytometry. \*\* *p* < 0.005, \*\*\* *p* < 0.0001 vs. untreated cells (UT).

Next, we studied in depth the HBMEC activation induced by supernatants from *B. abortus*stimulated platelets. Activation of HBMECs induced by supernatants from *B. abortus*- stimulated platelets was dose dependent (Figure 3A) and, more importantly, it was achieved by using supernatants from different platelet donors (Figure 3B). CD40 expression in endothelial cells has been implicated in several pathologic conditions of the CNS including Alzheimer's disease and human immunodeficiency virus 1 (HIV-1) encephalitis, where an important role of CD40 has been demonstrated in BBB disruption [23]. Besides the upregulation of ICAM-1, the activated phenotype induced by *B. abortus*infected platelet supernatants also included the significant upregulation of CD40 surface expression (*p* < 0.05) (Figure 3C) and the secretion of significant (*p* < 0.0005) amounts of IL-6, IL-8, and CCL-2 (Figure 3D–F) when compared with untreated HBMECs or HBMECs treated with supernatants from unstimulated platelets. Levels of activation were comparable to those obtained when HBMECs were stimulated with culture supernatants from *Brucella*-infected astrocytes [9] or IL-1β used as positive controls. Importantly, the concentrations of IL-6, IL-8, and CCL-2 measured in supernatants from *B. abortus*-stimulated platelets used for stimulation were negligible (<200 pg/mL in all cases, data not shown). These results demonstrate that supernatants from *B. abortus*-stimulated platelets induce an activated phenotype in microvascular brain endothelial cells, characterized by the upregulation of surface molecules such as ICAM-1 and CD40, and the secretion of both cytokines and chemokines.

**Figure 3.** *Cont*.

β β

β

β

**Figure 3.** Supernatants from *B. abortus*-infected platelets induce an activated phenotype in HBMECs. Platelets were incubated with *B. abortus* (PTL + B.a.) or alone (PTL) for 24 h. Culture supernatants from *B. abortus*-infected astrocytes and interleukin (IL)-1β were used as control. Platelets supernatants were collected, filtered, and used to stimulate HBMECs for 24 h at the indicated dilutions. ( **A**) ICAM-1 expression was measured by flow cytometry on HBMEC surface. (**B**) ICAM-1 expression obtained by stimulating HBMECs with supernatants obtained from five independent PTL donors. ( **C**) CD40 expression was measured on HBMEC surface by flow cytometry. The secretion of IL-6 ( **D**), IL-8 (**E**), and C-C motif chemokine ligand 2 (CCL-2) (**F**) was determined from HBMEC treated culture supernatants by ELISA. Bars represent the mean ± SEM of duplicates. Data shown are from a representative experiment out of at least three performed, except B. \* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0005 vs. untreated cells (UT). NI: non infected.

### *2.3. Secreted Factors from B. abortus-Activated Platelets Activate HBMECs*

β

Previous studies have shown that *Brucella* spp. release outer-membrane vesicles (OMVs, also known as blebs) containing lipopolysaccharide (LPS), outer membrane proteins, and other bacterial components [24]. To rule out the possibility that OMVs were implicated in HBMEC activation, they were removed from the supernatants by ultracentrifugation, as previously described [24]. HBMECs were then incubated with OMVs-free supernatants for 24 h and the activation of HBMECs was evaluated. There were no significant di fferences (*p* > 0.05) between non-depleted and OMVs-free supernatants regarding HBMEC activation, measured as ICAM-1 expression (Figure 4A) and IL-6, IL-8, and CCL-2 secretion (Figure 4B–D, respectively). To discard any putative participation of *Brucella*-secreted factors on HBMEC activation, *B. abortus* was incubated alone in the same culture conditions for 24 h. Then, culture supernatants were filtered and ultracentrifuged as described above. These platelet-free *B. abortus* culture supernatants were unable to activate HBMECs (Figure 4). Altogether, these results indicate that secreted factors from *B. abortus*-activated platelets are responsible for HBMEC activation.

**Figure 4.** *Cont*.

**Figure 4.** Secreted factors from *B. abortus*-stimulated platelets activate HBMECs. Supernatants from *B. abortus*-activated platelets (PLT + B.a.) or from *B. abortus* alone were ultracentrifuged (outer-membrane vesicle (OMV)-free supernatants) or not and used to stimulate HBMECs for 24 h. Supernatants from non-ultracentrifuged PTL alone were used as control. (**A**) ICAM-1 surface expression on HBMECs was assessed by flow cytometry. HBMEC secretion of IL-6 (**B**), IL-8 (**C**), and CCL-2 (**D**) was quantified by ELISA. Bars represent the mean ± SEM of duplicates from a representative experiment out of at least three performed. \* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0001 vs. untreated cells (UT).

### *2.4. B. abortus Lipoprotein-Stimulated Platelets Activate Brain Microvascular Endothelial Cells*

To test whether bacterial viability was necessary to induce the activation of platelets and consequently HBMEC activation, platelets were incubated for 24 h with heat-killed *B. abortus* (HKBA). Supernatants were then filtered and used as stimuli on HBMECs. HKBA supernatants were used to treat HBMECs as a negative control. As positive control, platelets were activated with thrombin. Supernatants from HKBA-stimulated platelets were able to activate HBMECs, inducing the upregulation of ICAM-1 (Figure 5A), and increasing the secretion of IL-6, IL-8, and CCL-2 (Figure 5B–D, respectively). We have previously demonstrated that different cell types can be activated by *B. abortus* lipoproteins [8,25,26]. Therefore, we further evaluated the contribution of lipoproteins in the induction of HBMEC activation by platelets. For this, platelets were incubated with *B. abortus* lipidated- or unlipidated-outer-membrane protein 19 (L-Omp19 or U-Omp19, respectively), used as a *Brucella* lipoprotein model [25]. Then, HBMECs were stimulated with the filtered supernatants for an additional 24 h, and the expression of ICAM-1 and secretion levels of IL-6, IL-8, and CCL-2 were evaluated. L-Omp19-activated platelets recapitulated HBMEC activation induced by supernatants from *B. abortus*-stimulated platelets. Furthermore, this activation was dependent on the lipidation of Omp19, as U-Omp19-stimulated platelets failed to induce HBMEC activation (Figure 5A–D). Culture supernatants from thrombin-activated platelets also induced partial activation of HBMECs. Neither HKBA-, L-Omp19-, nor U-Omp19-stimulated supernatants were able to activate HBMECs, demonstrating the presence of a platelet-secreted factor involved in the activation of HBMECs. Altogether, these results demonstrated that the presence of supernatants from platelets stimulated by structural components of *Brucella* (such as L-Omp19), independently of bacterial viability, are involved in the activation of HBMECs.

**Figure 5.** *B. abortus* lipoprotein-stimulated platelets activate HBMECs. Platelets were stimulated for 24 h with heat killed *B. abortus* (HKBA, 10<sup>8</sup> bacteria/mL), lipidated-outer-membrane protein 19 (L-Omp19), unlipidated-outer-membrane protein 19 (U-Omp19) (500 ng/mL), or thrombin (0.1 U/mL). Then, supernatants were collected, filtered. and used to stimulate HBMECs for an additional 24 h. Supernatants from PTL and stimuli incubated alone at the same conditions were used as control. ICAM-1 (**A**) was determined on HBMEC surface by flow cytometry. The secretion of IL-6 (**B**), IL-8 (**C**), and CCL-2 (**D**) was determined by ELISA. Bars represent the mean ± SEM of duplicates from a representative experiment out of at least three performed. \* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0001 vs. untreated cells (UT).

### *2.5. Platelet-Stimulated HBMECs Induce Transendothelial Migration of Neutrophils and Monocytes*

The presence of leukocytes in the cerebrospinal fluid and cerebral parenchyma has been described during neurobrucellosis [4]. This phenomenon, named pleocytosis, could be a consequence of the activation induced by *Brucella*-activated platelets on the blood–brain barrier [9]. To test this possibility, we used a previously established assay of transendothelial migration [9]. Briefly, HBMECs were seeded in the upper chamber of a Transwell plate, and they were cultured for 5 days to establish a monolayer. Then, HBMEC monolayers were treated for 24 h with supernatants from *B. abortus*-stimulated platelets. Culture supernatants from *Brucella*-infected astrocytes and human IL-1β were used as control. Finally, neutrophils or monocytes were seeded in the upper chamber and incubated for 3 h and the number of transmigrated cells to the lower chamber was quantified.

Monocyte as well as neutrophil migration increased when the HBMEC monolayer was treated with supernatants from *B. abortus*-stimulated platelets (Figure 6A,B), but not when HBMECs were stimulated with supernatants from platelets alone. Cellular transmigration was comparable to that obtained when HBMECs where stimulated with culture supernatants from *Brucella*-infected astrocytes or IL-1β used as positive controls. These results indicate that activation of brain endothelial cells by supernatants from *B. abortus*-stimulated platelets could induce transmigration of immune cells through a polarized brain endothelial cell monolayer. Taken together, these results sugges<sup>t</sup> that activated platelets could be responsible for the induction of pleocytosis in the context of *B. abortus* CNS infection.

**Figure 6.** *B. abortus*-infected platelets activate HBMEC, promoting transendothelial migration of neutrophils and monocytes. HBMEC monolayers were established on the membrane of Transwell plates and then treated with supernatants from *B. abortus*-stimulated platelets for additional 24 h. Culture supernatants from *Brucella*-infected astrocytes and human IL-1β were used as control. Next, neutrophils (**A**) or monocytes (**B**) were seeded in the upper chamber and incubated for 3 h. Finally, media from the lower chamber were harvested and the number of migrated cells was quantified. Bars represent the mean ± SEM of duplicates from a representative experiment of at least three performed. \*\* *p* < 0.005, \*\*\* *p* < 0.0005 vs. untreated cells (UT) or indicated treatment. NI: non infected.

### *2.6. The Erk1*/*2 Pathway Is Involved in HMBEC Activation Induced by B. abortus-Stimulated Platelets and It Is Implicated in Transendothelial Migration of Neutrophils*

We decided to further investigate the molecular mechanisms involved in endothelial activation and cellular transcytosis. It was previously reported that the extracellular signal-regulated kinase (Erk)1/2 pathway is involved in the activation of brain endothelial cells [27]. Taking this into account, we investigated the participation of the Erk1/2 signaling pathway in the activation of HBMECs by supernatants from *B. abortus*-stimulated platelets. For this, we used the Erk1/2-specific inhibitor PD98059 to treat HBMEC cells. Inhibition of the Erk1/2 pathway partially reduced the upregulation of ICAM-1 induced by supernatants from *B. abortus*-stimulated platelets (*p* < 0.005) (Figure 7A). Moreover, our results showed that cytokine and chemokine secretion of activated HBMECs is also regulated by the Erk1/2 pathway, since the inhibition with PD98059 also partially diminished the secretion of IL-6, IL-8, and CCL-2, compared to non-treated cells (Figure 7B–D). These results indicate that the Erk1/2 pathway is involved in HBMEC activation by supernatants from *B. abortus*-activated platelets.

The involvement of the Erk1/2 pathway in ICAM-1 upregulation is particularly interesting since ICAM-1 is one of the immunoglobulin-like cell adhesion molecules implicated in the transendothelial migration of immune cells [22]. Thus, we investigated the involvement of the Erk1/2 pathway on the increasing transendothelial migration throughout HBMECs activated by supernatants from *B. abortus*stimulated platelets. For this, HBMEC monolayers on Transwells were pre-treated with PD98059 2 h before and throughout treatment with *B. abortus*-stimulated platelet supernatants. 24 h later, neutrophils were seeded in the upper chamber for 3 h and the number of migrated cells to the lower chamber was quantified. Neutrophil migration was totally inhibited in HBMECs pre-treated with PD98059, demonstrating the implication of the Erk1/2 pathway on the increased transmigration of immune cells (Figure 7E).

**Figure 7.** Extracellular signal-regulated kinase (Erk)1/2 pathway is involved in HBMEC activation induced by *B. abortus*-stimulated platelets and it is implicated in transendothelial migration of neutrophils. HBMEC were pre-incubated with the Erk1/2 inhibitor (PD98059) for 2 h before platelets–supernatants stimulation and kept throughout. ICAM-1 was determined on HBMECs surface by flow cytometry (**A**). The secretion of IL-6 (**B**), IL-8 (**C**), and CCL-2 (**D**) was determined by ELISA. HBMEC monolayers were established on the membrane of Transwell plates. HBMEC activation was inhibited by PD98059 and then treated with supernatants from *B. abortus*-stimulated platelets for additional 24 h. Next, neutrophils were seeded in the upper chamber and incubated for 3 h. Finally, media from the lower chamber was harvested and the number of migrated cells was quantified (**E**). Bars represent the mean ± SEM of duplicates from a representative experiment of at least three performed \* *p* < 0.05, \*\* *p* < 0.005, \*\*\* *p* < 0.0005 vs. untreated cells (UT) or indicated treatment.
