Hepatic and splenic immune response during acute vs. chronic Brucella melitensis infection using in situ microscopy.

Abstract

Brucella melitensis is an intracellular bacteria causing disease in humans as an incidental host. The infection initiates as acute flu-like symptoms and may transform into a chronic cyclic infection. This cyclic infection may be partly due to the bacteria’s ability to persist within antigen presenting cells and evade the CD8+ T cell response over long periods of time. This research aims to characterize the immune response of the acute and chronic forms of brucellosis in the murine liver and spleen. We also sought to determine if the exhaustion of the CD8+ T cells was a permanent or temporary change. This was accomplished by using adoptive transfer of acutely infected CD8+T cells and chronically infected CD8+ T cells into a naïve host followed by re-infection. The histological examination presented supports the concept that exhausted T-cells can regain function through evidence of granulomatous inflammation after virulent challenge in a new host environment.

1. Introduction

Gram-negative, intracellular Brucella melitensis causes human and animal disease in various parts of the world and has considerable economic impact in endemic areas [1]. Most species of Brucella have a preferred reservoir host species in which Brucella localizes to mononuclear phagocytic cells and trophoblasts[1]. Brucella melitensis prefers sheep and goats while humans, as incidental hosts, are infected from contact with blood or ingestion of milk products from the reservoir species. In humans, infection with Brucella causes acute flu-like symptoms accompanied by high fevers, which can transform into a chronic, cyclic infection[2].

Brucellosis can be approximately 1 to 16 weeks from inception of symptoms to diagnosis of human infection, with the average being 35 days[3]. This protracted kinetics until diagnosis and treatment provides the bacteria ample time to establish a successful replicative niche in the host. Approximately 10% of the treated brucellosis cases relapse in the first year post-diagnosis, which may or may not be associated with reinfection. Certain factors increase the risk of relapse: inadequate or noncompliant treatment, infection for less than 10 days before treatment starts, bacteremia, and thrombocytopenia[4]. An increased risk of relapse when treatment begins during the early acute phase of infection further suggests that timing is critical in establishing chronic infection and the potential for reactivation of disease.

The ideal vaccine needs to limit the acute phase but also prevent the chronic reactivating phase of infection, along with producing only a minimal local or systemic reaction. The body fights infection with Brucella melitensis by IFN-γ mediated activation of bactericidal macrophages, decreasing intracellular bacterial survival, and allowing cytotoxic CD8+ T cells kill the infected macrophages. A vaccine would need to increase the Th1 cellular response, which would increase ability of the immune system to directly kill infected cells[5]. Numerous human brucellosis vaccines have been developed in the past, including live attenuated strains and subcellular Brucella components. Unfortunately, live vaccines cause brucellosis in humans, making them unsafe to use. The Brucella subcellular component vaccines are highly immunogenic, but also highly reactive causing severe local and systemic reactions[6].

One enigma of chronic brucellosis is how the bacteria are able to persist within antigen presenting cells and evade the CD8+ T cell response over long periods of time. Understanding the chronic phase of infection has been challenging due to the lack of a reliable and cost-effective small animal model. Recent work has confirmed that this model may exist in the form of the Brucella-susceptible BALB/c mouse. In the BALB/c mouse model bacteremia can fall below the level of detection for considerable time before re-emerging in various anatomical locations[7]. This is an inconvenient finding for previous vaccine validation that has used clearance of Brucella from BALB/c mice at early time points (<30 days) as the endpoint and success determinant[8]. Better defining the BALB/c model of brucellosis, including reactivation of chronic disease, may encourage more successful translation of vaccine research from the bench to the clinic.

Brucella is an intracellular bacterium, therefore cytotoxic CD8+ T cells should be able to destroy infected cells. Th1 CD4+ T cells are responsible for producing cytokines, such as IFN- γ and IL-2, to activate cytotoxic CD8+ T cells. IFN- γ is necessary to eliminate Brucella during acute, active infection [8]. In some cases, Brucella has the ability to proceed to a chronic infection, yet the mechanism of IFN- γ inhibition is not completely understood in this setting[9]. Interestingly, Brucella can evade the immune system through CD8+ T cell suppression. Previous research has shown that Brucella can suppress the CD8+ T cells in acute infection and potentially disrupt conversion to effective memory T cells[7].

Further understanding the granuloma and cellular dynamics of the Brucella response would aid in the development of a successful vaccine. Bacteria induced granulomas have been more extensively studied in other infections, including tuberculosis, but there has been far less understanding of Brucella induced granulomas[10,11]. The macrophages that isolate/respond to infectious material within a granuloma can subsequently differentiate into multinucleated cells, foamy cells, and epithelioid cells[12]. The macrophages typically seen in Brucella infections transform into epithelioid cells[13, 14]. These granulomas are most prominent in the liver but can also be seen in the spleen along the cortex. This research aimed to characterize the immune response seen in the murine liver and spleen for both acute and chronic brucellosis. We also sought to determine via pathology if the Brucella induced exhaustion of the CD8+ T cells was a permanent or temporary change.

2. Materials and Methods

2.1. Immunization of Mice

Female BALB/c mice (6–8 weeks of age) were obtained from Harlan (Indianapolis, IN) and housed in AAALAC approved facilities under pathogen-free conditions using standard protocols. For initial immunization, groups of 16 mice were immunized via the intraperitoneal route (i.p.) with 10⁷ virulent B. melitensis strain GR023 6 months prior to sacrifice for chronic infections and 2 weeks prior for acute infections[15]. Groups of 4 uninfected age-matched mice were injected i.p. with phosphate buffered saline (PBS). For challenge experiments, groups of 16 recipient mice were challenged i.p. with 10⁶ B. melitensis strain GR023 immediately following adoptive transfer of splenocytes from previously infected animals. All animal experiments were conducted upon review and approval from the Institutional Animal Care and Use Committee of the University of Wisconsin-Madison in compliance with the U.S. Department of Health and Human Services Guide for the Care and Use of Laboratory Animals.

2.2. Cell Tracking and Adoptive Transfer

Spleens from Brucella infected or naive animals were homogenized, strained through a 70μm cell strainer, and immediately treated with RBC lysis buffer. The splenocytes were washed thoroughly and then stained with 10μm carboxyfluorescein diacetate succinimidyl ester (CFDA-SE/CFSE, Invitrogen) prior to transfer. CFSE stained cells were adoptively transferred (1 × 10⁷ total cells/mouse) via retro-orbital injection to anaesthetized 8-week-old female BALB/c mice within an average of 2 h of the donor splenocytes being collected.

2.3. Histology

Spleens and livers from Brucella melitensis infected and naïve animals were collected and fixed in 10% neutral buffered formalin for a period no shorter than 72 h. Upon fixation, tissues were transferred into phosphate buffered saline until processing which included paraffin embedding and sectioning at a thickness of 5μm. Tissue sections were then stained with hematoxylin and eosin (H&E). Briefly, paraffin-embedded sections were deparaffinized with xylene, then rehydrated with gradient percentages of ethanol. Sections were stained with hematoxylin for 5 minutes, followed by bluing in ammonia water, then eosin for 10 minutes. The stained sections were dehydrated through gradient ethanol, cleared in xylene, and then mounted with a coverslip.

2.4. Statistics

Granulomas were identified based on histologic aggregation of immune cells in circular formations including lymphocytes, neutrophils, and macrophages. Twelve fields at 4× magnification were evaluated across three H&E slides from acutely infected transfer, three H&E slides from chronically infected transfer, and one H&E slide from uninfected transfer. To determine statistical significance of granuloma size between uninfected cells to acutely infected adoptive transfer cells and chronically infected adoptive transfer cells, a One-way ANOVA test was performed. All tests were performed using GraphPad Prism v5.0f (GraphPad Software). P-values of < 0.05 were considered significant.

3. Results

3.1. Identification of liver granulomas during acute and chronic Brucella infection.

Acutely infected livers had areas of inflammation throughout the parenchyma characterized by an influx of neutrophils and macrophages, indicative of an acute infection (Figure 1A–C). The inflammatory areas were not associated with any particular structure as they are found surrounding portal tracts, central veins, and within the parenchyma in the liver. Each inflammatory area was discrete in relation to other nearby inflammatory areas, forming a pyogranulomatous inflammatory reaction. The spaces where the immune cells had infiltrated are poorly differentiated and dispersed between morphologically normal hepatocytes.

Figure 1:

Photomicrographs of acute and chronic brucellosis in liver with uninfected age-matched controls. A-C: Acute infection is characterized by poorly differentiated areas of inflammation (black arrow) dispersed between normal hepatocytes. G-I: Chronic infection inflammatory areas are well demarcated consisting of mainly fibrosis (red arrow) and lymphocytes (black star). D-F, J-L: Uninfected age matched controls demonstrate lack of granuloma formation as expected in normal tissue. H&E stain; 4x bar = 200μm, 10x bar = 50μm, and 40x bar = 20μm.

In contrast to acutely infected livers, livers from mice that were chronically infected displayed areas of inflammation with a slightly different immune cell collection (Fig. 1G–I). The inflammatory areas were typically found surrounding or associated with portal tracts. The areas had well demarcated borders of hepatocytes and macrophages. Some areas of inflammation appeared to merge together to form a larger expanse. The immune cell infiltration in chronic infection includes macrophages, lymphocytes, and neutrophils. The addition of lymphocytes to the areas of inflammation may indicate that the infection is chronic because lymphocytes are part of the adaptive immune response and take longer to activate. Further, the presence of neutrophils suggests Brucella is still present and active in the mouse because neutrophils typically respond in seconds to minutes to a cellular insult or infection and have a short half-life between 5 and 90 hours [16].

3.2. Splenic microarchitecture of animals with chronic infection

The spleen with acute infection revealed an increased number of secondary follicles in the white pulp with poorly differentiated, prominent germinal centers (Figure 2A–C). Increased cellular proliferation was observed in the mantle zone surrounding the follicles. Many of the B cells surrounding the periarteriolar lymphoid sheath were transitioned into plasma cells as made evident by the lack of differentiation of the germinal centers. These changes are suggestive of active infection.

Figure 2:

Photomicrographs of acute and chronic brucellosis in spleen with uninfected age-matched controls. A-C: Acutely infected spleen show an increased amount of lymphocyte activation around the germinal center (black arrow). G-I: Chronically infected spleens have increased amount of erythrocytes (white arrow) and fibrosis (red arrow). D-F, J-L: Uninfected age matched controls demonstrate normal white and red pulp distribution. H&E stain; 4x bar = 200μm, 10x bar = 50μm, and 40x bar = 20μm.

Conversely, the spleen with chronic infection presented a more normal distribution of white pulp containing both primary and secondary follicles with well-differentiated germinal centers (Figure 2G–I). The splenic follicles in the chronic infection had an increased amount of activated B cells due to the slightly pleomorphic shape. In uninfected spleens, the follicle had an even distribution of B cells, while the chronically infected spleens had thicker and thinner follicular areas, again suggestive of reactivating infection. There was an apparent increase in erythrocytes and reticulocytes in red pulp, suggesting increased extramedullary hematopoiesis.

Splenomegaly was observed for both the acute infection and chronic infection, but due to seemingly different cell types. The acute infection had an increase in the number of activated B cells centered around the periarteriolar lymphoid sheath, along with macrophages in the parenchyma (Figure 2A–C). The relative number of erythrocytes appears similar between the 2 week uninfected spleen and the acute infection, therefore hematopoietic changes seem distinct to chronic brucellosis.

Age related splenic changes may contribute to the difference in the spleen sizes between the uninfected and chronically infected mice. In addition to any age-related changes, there were indications of splenomegaly due to hematopoetic changes when comparing the acute infection and the 2-week uninfected spleen in age-matched mice. The physical space between the germinal centers was increased in infected mice, due to an apparent infiltration of erythrocytes and reticulocytes.

3.3. Increased host response upon infectious challenge in animals with adoptively transferred splenocytes

We adoptively transferred cells from acute and chronic Brucella infected mice into naïve recipient mice with subsequent virulent challenge (Figs. 3 and 4). Naïve mice, not having developed the Brucella specific immune environment, provided a “fresh” environment for the Brucella responding splenocytes including specific CD8+ T cells.

Figure 3:

Photomicrographs of Brucella melitensis virulent challenge in liver after receiving adoptive transferred uninfected, acutely infected, and chronically infected splenocytes. A-H: Uninfected adoptive transferred splenocytes produce a typical granulomatous inflammatory response (black arrow). A-D 4× magnification, E-H 10× magnification. I-P: Splenocytes from a previous acute brucellosis infection produce a granulomatous inflammatory response (black arrow) in a novel environment. I-L 4× magnification, M-P 10× magnification. Q-X: Chronically infected splenocytes, that have been previously exhausted, are capable of producing an inflammatory response (black arrow) in a novel environment. Q-T 4× magnification, U-X 10× magnification H&E stain; 4x bar = 200μm, 10x bar = 50μm.

Figure 4:

Photomicrographs of Brucella melitensis virulent challenge in spleen after receiving adoptive transferred acutely infected, and chronically infected splenocytes. A-H: Splenocytes from a previous acute brucellosis infection show lymphocytic activation around the germinal centers (black arrow) in a novel environment. A-D 4× magnification, E-H 10× magnification. I-P: Chronically infected splenocytes, that have been previously exhausted, are capable of producing lymphocytic activation (black arrow) in a novel environment similar to the acutely infection transferred splenocytes. I-L 4× magnification, M-P 10× magnification. H&E stain; 4x bar = 200μm, 10x bar = 50μm.

Using tissues stained with hematoxylin and eosin to show the tissue morphology, we identified morphological differences in liver and spleen. As Brucella is an intracellular bacterium, the main immunological response in the liver is to form granulomas. Macrophages phagocytize Brucella and form into epithelial like cells in an attempt to isolate the infection. The mice that received uninfected cells follow the liver granuloma pattern of a primary Brucella infection (Figure 3A–H). The infection starts with a lower number of small granulomas. By day 5, the number of granulomas increased rapidly with a small increase in granuloma size. As infection progressed, the average granuloma size increased but the total number of granulomas decreased. This could suggest a merging of numerous smaller granulomas into a larger granuloma. An alternative idea is that some areas of inflammation are taking longer to resolve than other, thus recruiting more inflammatory cells to that area.

The acute and chronic recipient livers follow a slightly different granuloma pattern (Figure 3I–P). The splenocytes transferred from the acutely infected spleen are already primed to respond to Brucella, which is observed in the granuloma response. Surprisingly, the cells transferred from the chronic infection mimic the acute transferred cells. The granuloma size (Figure 5) appeared to increase faster than the primary infection while the number of granulomas (Figure 6) stay more consistent throughout the infection. Even though the granuloma size appears to be larger at an earlier data point versus the uninfected cells, it is not statistically significant.

Figure 5:

Average size of areas of granulomatous inflammation in liver with adoptive transferred splenocytes over time. Data are presented as mean ± SEM.

Figure 6:

Number of areas of granulomatous inflammation in liver with adoptive transferred splenocytes over time. Data are presented as total count from all slides evaluated.

The next organ of interest in the Brucella virulent challenge of acutely (Figure 4A–H) and chronically (Figure 4I–P) infected splenocytes was the spleen. Spleen images were evaluated based on the size and distribution of the white pulp. The onset of Brucella infection showed a densely populated mantle zone of lymphocytes around the central artery in the white pulp. Germinal centers had a normal distribution. As the infection progressed, the germinal center becomes poorly differentiated with an increased amount of proliferating B cells surrounding the periarteriolar lymphoid sheath. The number of macrophages in the red pulp increased along with the Brucella infection. Interestingly, the spleens that received previously chronically infected splenocytes had lymphocyte activation in a similar pattern to the acutely infected splenocytes. The chronically infected splenocytes showed the largest amount of germinal center activation on day 5, similar to the acutely infected splenocytes, with progressively poor differentiation as the infection continues. The next step in the evaluation would be to include granulomas seen in the spleen. Splenic granulomas have been previously described [17], but were not visualized in the sections evaluated.

4. Discussion

Our aim in this work was to identify differences in the microarchitecture between acute and chronic brucellosis, and to determine if previously exhausted Brucella specific T-cells are able to regain function in a new host environment. The time frame for acute and chronic infections in this study was 2 weeks and 6 months, respectively. Frequently, chronic brucellosis studies using the mouse model are limited to 30 days. We have recently shown that splenocytes from BALB/c mice infected for 9-months still possess measurable Brucella melitensis CFUs[7]. This current work helps to more thoroughly established long-term chronic brucellosis in BALB/c mice, supporting the concept that mice are an acceptable research model for chronic brucellosis in humans.

The microarchitecture of the liver and spleen in the acute and chronic infections are consistent with previously described infections [14]. An acute infection has multiple areas of purulent granulomatous inflammatory aggregates, while chronic infection includes lymphocytic infiltration. The combination of neutrophils macrophages, and lymphocytes indicates the Brucella infection is active enough to induce an inflammatory response.

Acute brucellosis causes activation of the immune cells in the spleen, which is noted as increased size and decreased organization of the white pulp. The notable splenic changes seen in chronic brucellosis include an increase in erythrocytes and in reticulocytes. The absolute cause of the increased reticulocytes and erythrocytes in the infected spleens is not yet known. We postulate it could be due to sequestration of erythrocytes in the spleen, increased hematopoiesis in the bone marrow, or potentially both. Bone marrow biopsy would be needed to determine hematopoiesis. Additionally, testing blood levels of erythropoietin may be helpful in detecting systemically induced hematopoiesis. Brucella spp. have been identified in the marrow of epiphyseal and metaphyseal bone, supporting the concept that bone marrow erythropoiesis could be compromised supporting splenic erythropoiesis [18].

Interestingly, the adoptive transfer experiment suggests that exhausted T-cells may be able to regain function in a new host environment that lacks the chronic low-level antigen presentation of a chronic infection. We have previously shown that the CD8+ T-cells from chronically infected mice produced less IFN-γ then uninfected mice. This decrease in IFN-γ is thought to cause a delay in the immune response, which would be identified as a reduction in the number and size of granulomas at the earlier time points. A decreased immune response would allow the bacteria to evade the immune system, infecting a larger amount of cells before an appropriate response could be elicited. Yet, using flow cytometry, exhausted T-cells regained the ability to produce IFN-γ when placed in a novel environment[19]. Splenic CD8+ T cells from chronic infected mice were exhausted before adoptive transfer suggesting that they would not respond readily to Brucella infection, though their ability to regain production of IFN-γ suggested that a new host environment may revitalize these cells. Interestingly, we observed that in the new host environment cells from chronically infected animals mimicked cells from acutely infected animals by the early formation of lymphocytic, pyogranulomatous inflammation. The acutely infected adoptive transferred CD8+ cells are hypothesized to have been primed to identify infected cells, leading to a faster response of inflammatory areas. The liver samples that contained previously exhausted CD8+ T cells formed granulomatous inflammation more rapidly than the CD8+ T cells from uninfected mice after virulent challenge. The host receiving the adoptive transferred cells does maintain their natural immune system. If the host immune system provided the only defense against the bacterial challenge, then all the challenge mice would have the same immunological response time regardless of the adoptive transferred splenocytes. The histological examination presented here further supports the concept that exhausted T-cells can regain function through evidence of granulomatous inflammation after virulent challenge in a new host environment. Therefore, the exhausted CD8+ T cells may indeed have the ability to regain function once placed in a new host environment

Ongoing efforts are underway to identify all affected organs during both acute and chronic infections to fully portray the murine models of brucellosis[7,18]. Organs of interest based on brucellosis in other species include vertebrae, testes, ovaries, and uterus.

Highlights.

• Acute brucellosis includes area of purulent granulomatous inflammatory aggregates

• Chronic brucellosis inflammatory response dominated by lymphocytic infiltration

• Previously exhausted CD8+ T-cells regain function in new host environment