- Email: [email protected]
Analysis of the immune response in infections of the goldfish (Carassius auratus L.) with Mycobacterium marinum
Developmental and Comparative Immunology 38 (2012) 456–465
Contents lists available at SciVerse ScienceDirect
Developmental and Comparative Immunology journal homepage: www.elsevier.com/locate/dci
Analysis of the immune response in infections of the goldfish (Carassius auratus L.) with Mycobacterium marinum Jordan W. Hodgkinson a,1, Jun-Qing Ge a,c,1, Leon Grayfer a,1, James Stafford a, Miodrag Belosevic a,b,⇑ a
Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada School of Public Health, University of Alberta, Edmonton, Alberta, Canada c Institute of Biotechnology, Fujian Academy of Agricultural Sciences, Fuzhou, Fijian, China b
a r t i c l e
i n f o
Article history: Received 12 May 2012 Revised 9 July 2012 Accepted 11 July 2012 Available online 7 August 2012 Keywords: Mycobacteria Fish Macrophage Respiratory burst Nitric oxide NADPH iNOS Cytokines Cytokine receptors
a b s t r a c t The rapid doubling time and genetic relatedness of the fish pathogen Mycobacterium marinum to Mycobacterium tuberculosis has rendered the former an attractive model for investigating mycobacterial hostpathogen interactions. We employed the M. marinum-goldfish infection model to investigate the in vivo immune responses to this pathogen in the context of a natural host. Histological analysis revealed mycobacterial infiltrates in goldfish kidney and spleen tissues, peaking 28 days post infections (dpi). Quantitative gene expression analysis showed significant increases of mRNA levels of pro-inflammatory cytokines (IFNc, IL-12p40, IL-1b1) and cytokine receptors (IFNGR1-1, TNFR2) at 7 dpi. Conversely, the gene expression levels of key anti-inflammatory cytokines TGFb and IL-10 were elevated at 14 dpi. Furthermore, M. marinum infections markedly increased the cytokine-primed oxidative burst responses of isolated kidney phagocytes at 7 but not 56 dpi. We believe that the M. marinum-goldfish infection model will be invaluable in furthering the understanding of the mycobacterium host-pathogen interface. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Pathogenic mycobacteria are facultative intracellular pathogens that infect and reside inside host phagocytic cells. The currently recognized Mycobacterium genus consists of 130 species and 11 subspecies, including disease-causing human pathogens Mycobacterium tuberculosis, Mycobacterium leprae and Mycobacterium ulcerans (Gauthier and Rhodes, 2009). While the understanding of the interactions of these pathogens with their hosts’ immune systems is pivotal, very slow replication of mycobacteria causing human infections (such as M. tuberculosis), coupled with level 3-facility requirements for these pathogens have hindered the progress of research in this area (Gideon and Flynn, 2011; Russell et al., 2010; Sridhar et al., 2011). Mycobacterium marinum is an etiological agent of mycobacteriosis in fish, amphibians, and aquatic mammals (Ramakrishnan et al., 1997; Flowers, 1970; Swaim et al., 2006), and shares close genetic resemblance to M. tuberculosis (possessing 99.4% 16S RNA sequence identity). In
⇑ Corresponding author at: Department of Biological Sciences, CW-405 Biological Sciences Building, University of Alberta, Edmonton, AB, Canada T6G 2E9. Tel.: +1 780 492 1266; fax: +1 780 492 9234. E-mail address: [email protected] (M. Belosevic). 1 These authors contributed equally to this work. 0145-305X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dci.2012.07.006
addition, M. marinum exhibits relatively short doubling times (Rogal et al., 1990) and causes pathology similar to that caused by mammalian mycobacteria, including granulomatous inflammation and spherical lesions encased by layers of epithelioid cells (Gauthier and Rhodes, 2009). Accordingly, M. marinum represents an attractive comparative model for studying mycobacterium pathogenesis and the mechanisms of host defense. M. marinum infection studies using zebrafish embryos have underlined the importance of macrophages in the dissemination and clearance of this pathogen, where macrophage depletions produced more robust infections and bacterial growth (Clay et al., 2007). These embryo infections (prior to development of adaptive immune components) resulted in the formation of macrophage aggregates with hallmark granuloma structures even in the absence of lymphocytes, while the infecting M. marinum expressed newly identified, granuloma-specific genes (Davis et al., 2002). Thus, where classical granuloma formation has been described as an organization of macrophages and lymphocytes, the above suggests that the adaptive components were not absolutely required in the formation of these structures in zebrafish. Furthermore, zebrafish infections studies revealed that granuloma formation was actively induced and maintained by mycobacterium rather than the host immune system (Davis et al., 2002). This was documented by showing that M. marinum actively recruited macrophages to
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
primary granulomas, facilitated the infection of new host phagocytes, bacterial dissemination and seeding of secondary granulomas (Davis and Ramakrishnan, 2009). In fact, M. marinum actively redirected the embryonic macrophages from predetermined migration pathways dictated by development (Davis et al., 2002). M. marinum also effectively subverted the phagosome maturation of both mammalian (Barker et al., 1997) and fish (El-Etr et al., 2001) macrophages, by preventing phagosome acidification (Tan et al., 2006; Xu et al., 2007) and employed a number of strategies for evading and/or dampening the phagocyte oxidative burst responses after IFNc-stimulation (Subbian et al., 2007a,b; Pagán-Ramos et al., 2006). Our previous in vitro studies of M. marinum-infected goldfish phagocytes underlined the capability of this pathogen to significantly ablate pro-inflammatory cytokine-induced ROI and NO production by down-regulating gene expression of the NADPH oxidase components and up-regulating immunosuppressive genes such as IL-10, TGFb and SOCS-3 (Grayfer et al., 2011a). Indeed, pathogenic Mycobacterium spp. in general, possess a plethora of subversion strategies for overcoming phagocyte microbicidal responses and making the intraphagosomal environment hospitable for bacterial growth (reviewed by Flannagan et al., 2009 and Flynn and Chan, 2003). The use of the goldfish as an in vivo M. marinum infection model was first proposed by Talaat et al. (1998) for the identification of putative M. marinum pathogenicity determinants (Ruley et al., 2004). Here we report on the use of the goldfish M. marinum infection model to investigate the mechanisms of host defense elicited by this pathogen in the context of a natural host. 2. Materials and methods 2.1. Fish Goldfish (Carassius auratus L.) were purchased from Aquatic Imports (Calgary, AB) and maintained at the Aquatic Facility of the Department of Biological Sciences, University of Alberta. The fish were kept at 17 °C in a flow-through water system on a simulated natural photoperiod, and fed to satiation daily with trout pellets. The fish were acclimated to this environment for at least 3 weeks prior to use in the experiments. All of the fish ranged from 10 to 15 cm in length and whenever possible an equal number of both sexes were used. Prior to sacrifice, fish were sedated using TMS (tricaine methane sulfonate) at 40–50 mg/L. 2.2. Bacteria M. marinum, strain ATCC 927 (fish isolate) was provided by Dr. Lourens Robberts, School of Public Health, University of Alberta. M. marinum was grown with shaking at 30 °C as a dispersed culture in 7H9 (Difco, Detroit, MI) broth supplemented with 0.5% glycerol and 10% albumin-dextrose complex for 7 to 10 days. The colony forming units (cfu) per milliliter were determined by plating serial dilutions of cultures on Middlebrook 7H10 agar (Difco) supplemented with 0.5% glycerol and 10% albumin-dextrose. Prior to all experiments, bacterial cultures were dispersed by 10–15 passages through a 25-gauge needle. When required, enumerated bacterial cultures were heat killed by incubation in an 80 °C water bath for 30 min. Heat killing efficiency (loss of bacterial viability) was confirmed by plating heat killed M. marinum cultures on Middlebrook 7H10 agar (Difco), and failure of the bacteria to grow after 5 days of incubation. Fish were infected by intra-peritoneal injection of 106 CFU of M. marinum diluted in 1X PBS. CFU/fish were determined after 7, 14, 28 and 56 days post infection (dpi) by plating 10-fold dilutions of
457
tissue homogenates on Middlebrook 7H10 agar (Difco) supplemented with 0.5% glycerol and 10% albumin-dextrose. Colonies were confirmed as acid-fast staining bacteria using staining procedures as suggested by the manufacturer (BD Diagnostics, NJ). 2.3. Preparation of histological sections Spleen, kidney and liver tissues were isolated from goldfish 28 days following PBS or M. marinum injections. Tissues were fixed in 10% neutral buffered formalin for 48 h at room temperature and embedded in paraffin wax after processing using Tissue Processor TP1020 (Leica Biosystems, Wetzlar, Germany). Five 5 lm sections were prepared using a rotary microtome. Slides were de-waxed, stained with Carbol Fuchsin (5% phenol, 10% ethanol, 1% w/v Basic Fuchsin (BioRad, USA) and counterstained with Methylene Blue (0.5% Methylene Blue w/v, 0.5% glacial acetic acid) (Ellis and Zabrowarny, 1993) and acid-fast (Ziehl-Neelsen). Stained tissue sections were observed at 100X magnification. 2.4. Production and purification of goldfish rgTNF, rgIFNc, and rgIFNcrel The production of recombinant goldfish tumor necrosis factor alpha 2 (rgTNFa2), recombinant goldfish interferon gamma (rgIFNc) and recombinant goldfish interferon gamma related (rgIFNcrel) cytokines have been described previously (Grayfer and Belosevic, 2009a,b; Grayfer et al., 2008, 2010, 2011b). The recombinant proteins were concentrated using polyethylene glycol flakes (35 kDa), dialyzed overnight against 1X PBS and subsequently passed through EndoTrap Red endotoxin removal columns (Hyglos, Bernried, Germany). Purification of the recombinant cytokines was confirmed by Western blot, and the identity of the proteins confirmed by mass spectrometry. Protein concentrations were determined using a Micro BCA Protein Assay Kit (Pierce, Rockford, IL) and the proteins were stored at 4 °C, prior to use. 2.5. Analysis of M. marinum-induced gene expression in goldfish spleen, kidney, and kidney derived leukocytes Total RNA was isolated from spleen, kidney and isolated kidney leukocytes using TRIzol extraction and reverse transcribed into cDNA using the Superscript II cDNA synthesis kit according to manufacturer’s directions. The genes examined included: CCL1, CXCL8 (IL-8), IFNc, IFNcrel, IL-12 p35, IL-12 p40, IL-1b1, IL-1b2, TNFa1, TNFa2, TGFb, IL-10, SOCS-3, IFNGR1-1, IFNGR1-2, TNFR1, TNFR2, iNOS A, iNOS B, p22phox, p40phox, p47phox, p67phox and gp91phox. Expression analysis of all genes was performed using the 2^-deltaCT method relative to endogenous control gene, elongation factor -1 alpha (EF-1a) and the derived RQ values were normalized against the respective PBS-injected controls. All primer sequences used in this study are shown in Supplementary Table 1. 2.6. Respiratory burst assay Kidney-derived monocytes isolated from M. marinum infected or PBS-injected goldfish (7 or 56 dpi) were seeded into 96 well plates at a density of 5 105 cells per well. Cells were treated with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNc and incubated for an additional 16 h. The nitroblue tetrazolium (NBT) assay was performed as described previously (Grayfer and Belosevic, 2009b; Grayfer et al., 2008). Absorbances (630 nm) from cells alone (no PMA) were subtracted from those of treatment groups to account for background NBT reduction. Medium only-treated, PMAtriggered cells were negative controls. Leukocytes were obtained from cultures established from 5 individual fish for each treatment group (n = 5).
458
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
2.7. Nitric oxide assay Kidney-derived macrophages isolated from M. marinum infected or PBS-injected goldfish (7 or 56 dpi), were seeded into 96 well plates at a density of 1 106 cells per well and incubated for 3 h to allow cell adherence. The media containing the nonadherent cells were carefully removed from individual wells. The adherent cells were treated with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNcrel in a final volume of 100 lL and incubated for 72 h. Nitrite production in cell supernatants was determined using the Griess reaction (Stafford et al., 2004). Absorbance was measured at 540 nm and nitrite concentration determined using a standard curve. Leukocytes were obtained from cultures established from five individual fish for each treatment group (n = 5). 2.8. Flow cytometry analysis of goldfish kidney leukocytes Kidney leukocytes isolated from goldfish 7 or 56 days subsequent to PBS or M. marinum injections were assessed by flow cytometry using previously established parameters (Neumann et al., 2000a). Cells were discriminated based on forward and side scatter against pre-established FACS gates known to describe progenitor/lymphocyte, monocyte or mature macrophage populations (R1, R3, and R2, respectively). Flow cytometry was performed using FACSCalibur apparatus (Beton/Dickinson) and data analysis was performed using CellQuest software (Beton/Dickinson). 2.9. Statistical analysis Statistical analysis was performed using one-way analysis of variance (ANOVA). P < 0.05 was considered statistically significant. 3. Results 3.1. Enumeration of bacterial loads and histology Following goldfish infections with M. marinum, bacterial burdens were determined over the course of 56 days. The magnitude of M. marinum infected-fish derived colony forming units (CFUs) peaked at 28 dpi while the CFUs per fish were lower on 56 dpi, as compared to 28 dpi (Fig. 1A). The acid-fast (Ziehl-Neelsen) staining of M. marinum infected fish showed mycobacterial granuloma structures in the spleen, the kidney (Fig. 1B), as well as liver tissue (data not shown). Purple staining, indicative of acid-fast bacteria, was observed in the centers of these granulomas (Fig. 1B). 3.2. M. marinum infections elicit changes in pro-inflammatory gene expression in goldfish tissues In order evaluate the goldfish inflammatory responses to M. marinum infections, we performed gene expression studies of key pro-inflammatory genes in kidney and spleen tissues of PBSinjected and M. marinum infected goldfish (Fig. 2). Infected fish exhibited increased mRNA levels of IFNc, IL-12 (p35 and p40 subunits), and IL-1b1 in the kidney tissue at 7 dpi (Fig. 2A), with a return to baseline mRNA levels by 14 dpi (Fig. 2B). On both 28 and 56 dpi, the gene expressions of pro-inflammatory cytokines were substantially reduced compared to controls (Fig. 2C and D). Interestingly, at 7, 28, and 56 dpi, IL-1b1 mRNA levels were significantly higher in M. marinum infected fish as compared to PBS-injected controls (Fig. 2A, C, and D). The pattern of pro-inflammatory gene expressions in the spleen were similar to those observed in the kidney (Fig. 3A–D), where short-lived increases in mRNA levels (7 dpi) were followed by
returns to baseline expression levels by 14 dpi (Fig. 3A and B). Additionally, by 56 dpi pro-inflammatory cytokine transcript levels were significantly down-regulated in infected fish, with the exception of IL-1b1, which remained elevated at both 28 and 56 dpi (Fig. 3C and D). 3.3. M. marinum infections elicit changes in the anti-inflammatory gene expression in goldfish tissues The expression of anti-inflammatory genes, transforming factor beta (TGFb), interleukin-10 (IL-10), and suppressor of cytokine signaling-3 (SOCS-3) were measured in the kidney (Fig. 4) and spleen (Fig. 5) during the course of M. marinum infections. While only SOCS-3 mRNA levels were substantially elevated in the kidney at 7 dpi, the expressions of all three anti-inflammatory genes were significantly higher compared to those of PBS-injected fish at 14 dpi (Fig. 4A and B). By 28 and 56 days dpi, the mRNA levels of all three anti-inflammatory genes decreased compared to PBS-injected controls (Fig. 4C and D). In contrast, the spleen mRNA levels of TGFb and IL-10 of infected fish were significantly lower at 7 dpi compared to PBS-injected controls, and by 14 and 28 dpi, mRNA levels were similar between the infected and control fish (Fig. 5A and B). As with the kidney gene expression patterns, the mRNA levels of TGFb, IL-10 and SOCS-3 were also significantly diminished at 56 dpi compared to those of PBS-injected fish (Fig. 5D). 3.4. M. marinum infections alter goldfish kidney and spleen cytokine receptor gene expression patterns In addition to assessing the gene expression of pro-inflammatory cytokines, we also examined the transcriptional changes of cognate cytokine receptors (IFNGR1-1, IFNGR1-2, TNFR1, and TNFR2) in the kidney (Fig. 6A and B) and the spleen of M. marinum-infected goldfish (Fig. 6C and D) after 7 and 56 dpi. The mRNA levels of IFNGR1-1 significantly increased in the kidney and the spleen at 7 dpi (Fig. 6A and C), but were down-regulated by 56 dpi (Fig. 6B and D). The mRNA levels of TNFR2 were significantly elevated in the kidneys, but not spleens at 7 dpi (Fig. 6A and C) while the expression of this receptor was significantly downregulated in both the kidney and the spleen at 56 dpi (Fig. 6B and D). Additionally, the kidney gene expression of TNFR1 was also significantly increased at 56 dpi (Fig. 6B). 3.5. M. marinum-infected goldfish-derived kidney monocytes and macrophages produce greater levels of reactive oxygen and reactive nitrogen intermediates We examined the ability of kidney-derived monocytes/macrophages from M. marinum or PBS injected fish to generate reactive oxygen or reactive nitrogen (ROI and NO, respectively) antimicrobial responses after 7 or 56 dpi. In our previous studies, we observed that while rgTNFa2 was a potent elicitor of both goldfish phagocyte ROI and NO, rgIFNc was effective at priming ROI but not the NO responses. In contrast, rgIFNcrel induced robust NO production but not ROI responses of goldfish macrophages and monocytes, respoctively. Accordingly, in the present work, rgTNFa2 and rgIFNc were employed for eliciting phagocyte ROI responses and rgTNFa2 and rgIFNc were used for studying the NO production. Monocyte PMA-triggered ROI production was measured subsequent to priming with medium alone, 100 ng/mL of rgTNFa2 or 100 ng/mL of rgIFNc (Fig. 7). Cytokine-primed (rgTNFa2 or rgIFNc) cells from M. marinum-injected fish, isolated 7 dpi, exhibited significantly enhanced ROI responses compared to the respective PBS injected control fish (Fig. 7A). Interestingly, while the baseline ROI production of monocytes obtained from PBS-injected andM.
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
459
Fig. 1. The course of infection with Mycobacterium marinum in the goldfish. (A) The kidney and spleen were homogenized and the colony forming units (cfu) per organ were determined by plating serial dilutions of cultures on Middlebrook 7H10 agar. The data are the mean ± SEM CFU/fish of five fish (n = 5). () indicate significantly different between 28 days post infection (dpi) compared to 7 or 14 dpi. (B) The kidney and spleen were removed 28 dpi. Mycobacterium infected fish had granulomas with red-stained centers (arrows), indicative of infection. Tissue sections were stained using acid-fast (Ziehl-Neelsen) stain and visualized at 100X.
marinum-infected fish was elevated at 56 dpi, no differences were seen in the ROI production between the two experimental groups (Fig. 7B). We also measured the mRNA levels of NADPH oxidase subunits, the enzyme responsible for ROI production. While the expressions of phox genes in kidney tissues of infected and control fish were similar at 7 dpi (Fig. 8A), there was a significant reduction in the expression of p22phox, p40phox and p67phox at 56 dpi (Fig. 8B). Kidney macrophages from M. marinum infected and PBSinjected goldfish were assessed for their ability to mount a nitric oxide response following stimulation with medium alone, 100 ng/mL of rgTNFa2 or rgIFNcrel (Fig. 9). Medium-treated macrophages derived from fish at 7 dpi produced significantly more NO than cells from PBS injected controls (Fig. 9A). In addition, NO
responses of kidney leukocytes obtained from infected fish compared to PBS-injected controls, was further enhanced after treatment with rgTNFa2 or rgIFNcrel (Fig. 9A). In contrast, macrophages isolated from fish at 56 dpi and treated with medium alone or with rgTNFa2 produced similar nitrite levels compared to the controls (Fig. 9B). However, at 56 dpi, rgIFNcrel-treated cells obtained from M. marinum-infected fish had a significantly higher NO responses compared to those obtained from PBS-injected control fish (Fig. 9B). We also measured the expression of iNOS A and iNOS B in the kidneys of M. marinum infected and PBS-injected fish (Fig. 10). The mRNA levels of both iNOS A and iNOS B increased significantly at 7 dpi (Fig. 10A). In contrast, by 56 dpi, while the iNOS A mRNA
460
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 2. Mycobacterium marinum infection alters the expression of pro-inflammatory genes in the kidney determined using quantitative PCR. Kidney tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 3. Mycobacterium marinum infection alters gene expression of pro-inflammatory genes in the spleen determined using quantitative PCR. Spleen tissues were harvested (A) 7, (B) 14, (C) 28 and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
461
Fig. 4. Mycobacterium marinum infection alters gene expression of anti-inflammatory genes in the kidney determined using quantitative PCR. Kidney tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 5. Mycobacterium marinum infection alters gene expression of anti-inflammatory genes in the spleen determined using quantitative PCR. Spleen tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
levels remained significantly elevated in cells from infected fish, the gene expression of iNOS B returned to levels comparable to PBS-injected controls. In order to elucidate whether the enhanced ROI production by cells from acutely infected fish stemmed from changes in immune cell populations, kidney leukocytes isolated from PBS and M. marinum injected fish were assessed by flow cytometry using previ-
ously established parameters (Neumann et al., 2000a, Suppl. Fig. 1). Based on our prior work, we have established that the cells in the R1 gate consist of lymphocytes and progenitor cells. As goldfish kidney monocytes mature, they shift into the R3 gate and upon further differentiation into macrophages, these cells enter the R2 gate (Neumann et al., 2000a,b). Kidney leukocytes isolated from fish 7 days subsequent to PBS injections, consisted primarily of
462
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 6. Mycobacterium marinum infection alters gene expression of TNF and IFN receptors in the kidney and the spleen tissues determined using quantitative PCR. Kidney tissues were harvested (A) 7 and (B) 56 days post infection (dpi), and in the spleen (C) 7 and (D) 56 dpi. The data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
immature monocytic cells (gate R3) and lymphocytes/myeloid precursor cells (gate R1) and relatively few mature macrophages (Suppl. Fig. 1). In contrast, kidney leukocytes isolated from fish infected for 7 days with M. marinum comprised of further differentiated monocytic populations and of substantially more mature macrophages, compared to the PBS controls (Suppl. Fig. 1). Interestingly, the kidney leukocyte isolated from M. marinum-infected fish 56 dpi exhibited cell compositions similar to those seen in PBS-injected fish at 7 and 56 dpi (Suppl. Fig. 1). The results presented in Supplementary Fig. 1 are representative of those observed using five individual fish for each experimental group.
4. Discussion In the present study, we describe the use of an in vivo goldfish model system to characterize the host innate immune defenses during the course of M. marinum infections. The parallel analysis of expression patterns of immune genes and antimicrobial capacities of goldfish phagocytes offered interesting insights of this complex host-pathogen association. Our findings indicate that there is a relationship between the dissemination of bacteria and the resulting infection-associated pathology, and the pro- and anti-inflammatory gene expression profiles, and the capacity of goldfish phagocytes to mount appropriate antimicrobial responses. Ruley et al. (2002) have shown that in fish exposed to M. marinum, the viable bacteria can be detected in the tissues as early as 1 week after infection, while substantially greater mycobacterial loads and infection-related pathology occur several weeks after exposure. Our findings support these observations, where we also observed greatest mycobacterial loads and significant granuloma formation in goldfish organs at 28 dpi. It should be noted that the bacterial loads in the goldfish remained
high at 56 dpi, attesting to the chronicity of this infection, and failure of the fish to eliminate the pathogen. Innate immune responses against pathogenic mycobacteria are critical in determining the fate of these infections. Using both mammalian and zebrafish infection models, it has been demonstrated that phagocyte recruitment to the sites of infection and the uptake of pathogens by the phagocytes may serve as a means of mycobacterial dissemination (Clay et al., 2007; Wolf et al., 2007; Davis and Ramakrishnan, 2009). During this initial uptake of mycobacteria by phagocytes, microbial components are recognized by pattern recognition receptors and this interaction results in the induction of pro-inflammatory genes (Chan et al., 2001; Zhang et al., 1999; Grayfer et al., 2011a). Our gene expression results indicate that M. marinum infection induced an increase in the expression of pro-inflammatory cytokines during the first week of infection. This modulation is presumably important for the induction of antimicrobial defenses, but paradoxically may also aid mycobacteria to establish a foothold during the infection process (Clay et al., 2007). Indeed, as demonstrated in the zebrafishM. marinum infection model, the pro-inflammatory cytokine TNFa is crucial to thwarting the infection as well as maintaining the granuloma structures (Clay et al., 2008), the latter in turn facilitating bacterial survival and eventual dissemination (Davis et al., 2002). A return to the baseline gene expression of pro-inflammatory mediators was observed in conjunction with increased expression of anti-inflammatory cytokines at 14 dpi. Interestingly, a consistent trend of down-regulation of the expression of pro-inflammatory cytokines was observed at 56 dpi. After phagocytosis, mycobacteria that reside in the phagosomes have been shown interfere with mechanisms of phagolysosomal fusion (Armstrong and Hart, 1971; Mwandumba et al., 2004; Nguyen and Pieters, 2005). Furthermore, intraphagosomal mycobacteria also interfere with host
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
463
Fig. 8. Mycobacterium marinum infections alter the expression of NADPH oxidase components in primary kidney leukocytes. Quantitative gene expression analysis of p22phox, p40phox, p47phox, p67phox and gp91phox NADPH components were assessed in kidneys leukocytes isolated 7(A) and 56(B) days post M. marinum infections (dpi) or PBS injected controls. The data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 7. Mycobacterium marinum infection induces goldfish kidney monocyte reactive oxygen production. Kidney monocytes were isolated from M. marinum infected or PBS-injected goldfish 7 (A) or 56 (B) days post infection and primed with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNc for 16 h. The ROI response was then triggered using PMA (100 ng/mL). Kidney leukocytes were obtained from four to five individual fish for each treatment group (n = 4–5). Statistical analysis was performed using one-way ANOVA. () indicate significantly different (P < 0.05) from the PMA only treated controls, and (+) above lines denotes significant difference (P < 0.05) between different experimental groups.
cell activation states. They have been shown to induce the production of anti-inflammatory mediators, which effectively dampen the antimicrobial functions of phagocytes, creating an environment suitable for chronic infection (Van Crevel et al., 2002; Marino et al., 2010; Grayfer et al., 2011a). A persistent increase in the expression of IL-1b1 was observed during the course of the M. marinum infections. IL-1b has been shown to decrease the host susceptibility to mycobacteria (Mayer-Barber et al., 2010; Michelini-Norris et al., 1992). Although 20-fold and 10-fold increases in the expression of IL-1b1 were measured during the acute and chronic phases of the infection respectively, bacterial infection still persisted in the goldfish. The consistent IL-1b response was clearly not sufficient to contribute to the elimination of M. marinum. It is likely that other pro-inflammatory cytokines, whose expression decreased during the chronic phase of the infection, synergistically act with IL-1b1 for efficient clearance of mycobacteria. It is also possible that downstream interference by M. marinum through an unknown mechanism(s) (i.e. inflammasome activation, IL-1R engagement) prevented its clearance from the goldfish. We previously reported that in vitro M. marinum infected goldfish monocytes and macrophages exhibit bacterial load-dependent abrogation of antimicrobial capabilities and increased expression of anti-inflammatory genes (Grayfer et al., 2011a). Presumably the initial pro-inflammatory conditions seen at 7 dpi, coincide with a general stimulation of the goldfish immune systems by the extremely immunogenic mycobacterial components. Conversely, the
Fig. 9. Mycobacterium marinum infection induces goldfish kidney macrophage nitric oxide production. Kidney macrophages, isolated from M. marinum infected or PBS-injected goldfish (A) 7 or (B) 56 days post infection (dpi) were incubated with medium alone, or with 100 ng/mL of rgTNFa2 or rgIFNcrel for 72 h. Nitrite production was determined by Griess reaction and nitrite concentrations determined using a nitrite standard curve. Leukocytes were obtained from four to five individual fish for each treatment group (n = 4 or 5). Statistical analysis was performed using one-way ANOVA. () indicate significantly different (P < 0.05) from medium only treated controls, and (+) above lines denotes significant difference (P < 0.05) between experimental groups.
464
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 10. Mycobacterium marinum infection increases the expression of iNOS A and iNOS B in primary kidney monocytes/macrophages. Cells were harvested (A) 7 and (B) 56 days post infection (dpi). The expression data were normalized against PBSinjected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
general suppression of immune gene expression later in the infection may be due to the induction of anti-inflammatory cytokine production and increased bacterial load in infected fish. Kidney-derived monocytes and macrophages isolated from infected goldfish at 7 dpi exhibited increased ROI and NO production, respectively, the magnitudes of which were further enhanced by stimulation with rgTNFa2 and rgIFNc or rgIFNcrel. These increases in ROI and NO responses coincided with increases in the expression of select phox and iNOS genes, respectively. Interestingly, the ROI production by monocytes isolated from fish at 56 dpi was similar to that of monocytes from PBS-injected control fish. Our preliminary flow cytometry data suggests that the isolated kidney leukocyte cultures from goldfish at 7 dpi, comprised of more differentiated subpopulations of monocytes and mature macrophages, compared to those isolated from PBS-injected fish (both 7 and 56 dpi) and after 56 dpi. Additionally, cells isolated from fish 7 dpi also exhibited increased expression of the IFNGR1-1 and the TNF receptors, presumably contributing to the increased responsiveness of these cells to the cognate recombinant ligands (rgIFNc, rgIFNcrel and TNFa2). Together these corroborating lines of evidence suggest that during acute but not chronic mycobacterium infections, goldfish myelopoiesis may be altered towards the production of more effective microbicidal phagocytes. Concomitantly, at 7 dpi there was a significant up-regulation in the expression of macrophage activating cytokines, IFNc and IL-12. It is well established that IL-12 and IFNc operate in a positive feedback manner, where the production of either cytokine confers the synthesis of the other by distinct immune cell populations (Bliss et al., 1996; Ma et al., 1996; Windhagen et al., 1996; Yoshida et al., 1994). This cross-regulation would result in an enhanced production of both cytokines and may also account for the marked increases in antimicrobial capacities of kidney monocytes and macrophages obtained from fish at 7 dpi. In fact, it is well established that IFNc stimulation of myeloid cells enhances their reac-
tive oxygen (Cassatella et al., 1990) and reactive nitrogen (Martin et al., 1994) producing capabilities. On 14 dpi, the kidney gene expression of the pro-inflammatory cytokines returned to baseline transcription, and by 56 dpi their mRNA levels were significantly lower than those in control fish. It is possible that the increased gene expression of the anti-inflammatory cytokine IL-10 in the kidney at 14 dpi, may have contributed to the deactivation of the monocyte/macrophage antimicrobial functions. In addition, the dissemination of bacteria and the resulting infection associated pathology are likely reflected in the gene expression profiles and antimicrobial response changes reported here. In a series of in vivo M. marinum-goldfish infection studies, it has been demonstrated that although bacteria were detected after a week of infection, substantially greater mycobacterial loads and infection-related pathology were seen at later times in the course of infection (Ruley et al., 2002). These results are corroborated here, where we observed expansive granuloma formation in the spleens of 30-day infected fish. Undoubtedly, these factors play critical roles in the balance between the pathogen colonization and the host immunity. Also, we have previously observed that in vitro M. marinum infected goldfish monocytes and macrophages exhibit bacterial load-dependent abrogation of antimicrobial capabilities and increased expression of anti-inflammatory genes (Grayfer et al., 2011a). Presumably the initial pro-inflammatory conditions seen after 1 week of M. marinum infections may coincide with a general stimulation of the goldfish immune systems by the extremely immunogenic mycobacterial components. Conversely, the return to baseline and the general suppression of immune gene expression later on in the infection might be suggestive of increased bacterial proliferation, dissemination and immunomodulation. Our past and present work underlines not only the efficacy of using the goldfish as a natural host model for mycobacterial infections but also provides an additional perspective of the natural host-pathogen immune interactions. We believe that the goldfish represents an ideal platform for furthering our understanding of the mycobacterium-host interface, providing both in vitro and in vivo infection approaches with which to bridge the current gaps in our understanding of mycobacteri immune evasion strategies and the mechanisms of host defense. Acknowledgments This work was supported by the Natural Sciences and Engineering Council of Canada (NSERC) to MB. JWH was supported by an NSERC CGS-M and Alberta Innovates top-up award. LG was supported by an NSERC PGS-D scholarship. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.dci.2012.07.006. References Armstrong, J.A., Hart, P.D., 1971. Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J. Exp. Med. 134, 713. Barker, L.P., George, K.M., Falkow, S., Small, P.L., 1997. Differential trafficking of live and dead Mycobacterium marinum organisms in macrophages. Infect. Immun. 65, 1497–1504. Bliss, J., Van Cleave, V., Murray, K., Wiencis, A., Ketchum, M., Maylor, R., Haire, T., Resmini, C., Abbas, A.K., Wolf, S.F., 1996. IL-12, as an adjuvant, promotes a T helper 1 cell, but does not suppress a T helper 2 cell recall response. J. Immunol. 156, 887–894. Cassatella, M.A., Bazzoni, F., Flynn, R.M., Dusi, S., Trinchieri, G., Rossi, F., 1990. Molecular basis of interferon-gamma and lipopolysaccharide enhancement of phagocyte respiratory burst capability. Studies on the gene expression of several NADPH oxidase components. J. Biol. Chem. 265, 20241–20246.
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465 Chan, E.D., Morris, K.R., Belisle, J.T., Hill, P., Remigio, L.K., Brennan, P.J., Riches, D.W.H., 2001. Induction of inducible nitric oxide synthase-NO by lipoarabinomannan of Mycobacterium tuberculosis is mediated by MEK1-ERK, MKK7-JNK, and NF-jB signaling pathways. Infect. Immun. 69, 2001–2010. Clay, H., Davis, J.M., Beery, D., Huttenlocher, A., Lyons, S.E., Ramakrishnan, L., 2007. Dichotomous role of the macrophage in early Mycobacterium marinum infection of the zebrafish. Cell Host Microbiol. 2, 29–39. Clay, H., Volkman, H.E., Ramakrishnan, L., 2008. Tumor necrosis factor signaling mediates resistance to mycobacteria by inhibiting bacterial growth and macrophage death. Immunity 29, 283–294. Davis, J.M., Clay, H., Lewis, J.L., Ghori, N., Herbomel, P., Ramakrishnan, L., 2002. Realtime visualization of Mycobacterium-macrophage interactions leading to initiation of granuloma formation in zebrafish embryos. Immunity 17, 693–702. Davis, J.M., Ramakrishnan, L., 2009. The role of the granuloma in expansion and dissemination of early tuberculous infection. Cell 136, 37–49. El-Etr, S.H., Yan, L., Cirillo, J.D., 2001. Fish monocytes as a model for mycobacterial host-pathogen interactions. Infect. Immun. 69, 7310–7317. Ellis, R.C., Zabrowarny, L.A., 1993. Safer staining method for acid fast bacilli. J. Clin. Pathol. 46, 559–560. Flannagan, R.S., Cosío, G., Grinstein, S., 2009. Antimicrobial mechanisms of phagocytes and bacterial evasion strategies. Nat. Rev. Microbiol. 7, 355–366. Flowers, D.J., 1970. Human infection due to Mycobacterium marinum after a dolphin bite. J. Clin. Pathol. 23, 475–477. Flynn, J.L., Chan, J., 2003. Immune evasion by Mycobacterium tuberculosis: living with the enemy. Curr. Opin. Immunol. 15, 450–455. Gauthier, D.T., Rhodes, M.W., 2009. Mycobacteriosis in fishes: a review. Vet. J. 180, 33–47. Gideon, H., Flynn, J., 2011. Latent tuberculosis: what the host ‘‘sees’’? Immunol. Res. 50, 202–212. Grayfer, L., Belosevic, M., 2009a. Molecular characterization, expression and functional analysis of goldfish (Carassius aurutus L.) interferon gamma. Dev. Comp. Immunol. 33, 235–246. Grayfer, L., Belosevic, M., 2009b. Molecular characterization of novel interferon gamma receptor 1 isoforms in zebrafish (Danio rerio) and goldfish (Carassius auratus L.). Mol. Immunol. 46, 3050–3059. Grayfer, L., Garcia, E.G., Belosevic, M., 2010. Comparison of macrophage antimicrobial responses induced by type II interferons of the goldfish (Carassius auratus L.). J. Biol. Chem. 285, 23537–23547. Grayfer, L., Hodgkinson, J.W., Belosevic, M., 2011a. Analysis of the antimicrobial responses of primary phagocytes of the goldfish (Carassius auratus L.) against Mycobacterium marinum. Dev. Comp. Immunol. 35, 1146–1158. Grayfer, L., Hodgkinson, J.W., Hitchen, S.J., Belosevic, M., 2011b. Characterization and functional analysis of goldfish (Carassius auratus L.) interleukin-10. Mol. Immunol. 48, 563–571. Grayfer, L., Walsh, J.G., Belosevic, M., 2008. Characterization and functional analysis of goldfish (Carassius auratus L.) tumor necrosis factor-alpha. Dev. Comp. Immunol. 32, 532–543. Ma, X., Chow, J.M., Gri, G., Carra, G., Gerosa, F., Wolf, S.F., Dzialo, R., Trinchieri, G., 1996. The interleukin 12 p40 gene promoter is primed by interferon gamma in monocytic cells. J. Exp. Med. 183, 147–157. Marino, S., Myers, A., Flynn, J.L., Kirschner, D.E., 2010. TNF and IL-10 are major factors in modulation of the phagocytic cell environment in lung and lymph node in tuberculosis: a next-generation two-compartmental model. J. Theor. Biol. 265, 586–598. Martin, E., Nathan, C., Xie, Q.W., 1994. Role of interferon regulatory factor 1 in induction of nitric oxide synthase. J. Exp. Med. 180, 977–984. Mayer-Barber, K.D., Barber, D.L., Shenderov, K., White, S.D., Wilson, M.S., Cheever, A., Kugler, D., Hieny, S., Caspar, P., Núñez, G., Schlueter, D., Flavell, R.A., Sutterwala, F.S., Sher, A., 2010. Cutting edge: caspase-1 independent IL-1b production is critical for host resistance to Mycobacterium tuberculosis and does not require TLR signaling in vivo. J. Immunol. 184, 3326–3330. Michelini-Norris, M.B., Blanchard, D.K., Pearson, C.A., Djeu, J.Y., 1992. Differential release of interleukin (IL)-1a, IL-1b, and IL-6 from normal human monocytes stimulated with an virulent and an avirulent isogenic variant of Mycobacterium avium-intracellular complex. J. Infect. Dis. 165, 702–709. Mwandumba, H.C., Russell, D.G., Nyirenda, M.H., Anderson, J., White, S.A., Molyneux, M.E., Squire, S.B., 2004. Mycobacterium tuberculosis resides in non acidified vacuoles in endocytically competent alveolar macrophages from patients with tuberculosis and HIV infection. J. Immunol. 172, 4592–4598.
465
Neumann, N.F., Barreda, D.R., Belosevic, M., 2000a. Generation and functional analysis of distinct macrophage sub-populations from goldfish (Carassius auratus L.) kidney leukocyte cultures. Fish Shellfish Immunol. 10, 1–20. Neumann, N.F., Stafford, J.L., Belosevic, M., 2000b. Biochemical and functional characterisation of macrophage stimulating factors secreted by mitogen-induced goldfish kidney leucocytes. Fish Shellfish Immunol. 10, 167–186. Nguyen, L., Pieters, J., 2005. The Trojan horse: survival tactics of pathogenic mycobacteria in macrophages. Trends Cell Biol. 15, 269–276. Pagán-Ramos, E., Master, S.S., Pritchett, C.L., Reimschuessel, R., Trucksis, M., Timmins, G.S., Deretic, V., 2006. Molecular and physiological effects of Mycobacterial oxyR inactivation. J. Bacteriol. 188, 2674–2680. Ramakrishnan, L., Valdivia, R.H., McKerrow, J.H., Falkow, S., 1997. Mycobacterium marinum causes both long-term subclinical infection and acute disease in the leopard frog (Rana Pipiens). Infect. Immun. 65, 767–773. Rogal, T., Wolters, J., Fohr, T., Bottger, E., 1990. Toward a phylogeny and definition of species at the molecular level within the genus Mycobacteria. Int. J. Syst. Bacteriol. 40, 323–330. Ruley, K.M., Ansede, J.H., Pritchett, C.L., Talaat, A.M., Reimschuessel, R., Trucksis, M., 2004. Identification of Mycobacterium marinum virulence genes using signaturetagged mutagenesis and the goldfish model of mycobacterial pathogenesis. FEMS Microbiol. Lett. 232, 75–81. Ruley, K.M., Reimschuessel, R., Trucksis, M., 2002. Goldfish as an animal model system for mycobacterial infection. In: Bacterial Pathogenesis Part C: Identification, Regulation, and Function of Virulence Factors. Academic Press, pp. 29–39. Russell, D.G., Barry, C.E., Flynn, J.L., 2010. Tuberculosis: what we don’t know can, and does, hurt us. Science 328, 852–856. Sridhar, S., Pollock, K., Lalvani, A., 2011. Redefining latent tuberculosis. Future Microbiol. 6, 1021–1035. Stafford, J.L., Wilson, E.C., Belosevic, M., 2004. Recombinant transferrin induces nitric oxide response in goldfish and murine macrophages. Fish Shellfish Immunol. 17, 171–185. Subbian, S., Mehta, P.K., Cirillo, S.L., Cirillo, J.D., 2007a. The Mycobacterium marinum mel2 locus displays similarity to bacterial bioluminescence systems and plays a role in defense against reactive oxygen and nitrogen species. BMC Microbiol. 7, 4. Subbian, S., Mehta, P.K., Cirillo, S.L.G., Bermudez, L.E., Cirillo, J.D., 2007b. A Mycobacterium marinum mel2 mutant is defective for growth in macrophages that produce reactive oxygen and reactive nitrogen species. Infect. Immun. 75, 127–134. Swaim, L.E., Connolly, L.E., Volkman, H.E., Humbert, O., Born, D.E., Ramakrishnan, L., 2006. Mycobacterium marinum infection of adult zebrafish causes caseating granulomatous tuberculosis and is moderated by adaptive immunity. Infect. Immun. 74, 6108–6117. Talaat, A.M., Reimschuessel, R., Wasserman, S.S., Trucksis, M., 1998. Goldfish, Carassius auratus, a novel animal model for the study of Mycobacterium marinum pathogenesis. Infect. Immun. 66, 2938–2942. Tan, T., Lee, W.L., Alexander, D.C., Grinstein, S., Liu, J., 2006. The ESAT-6/CFP-10 secretion system of Mycobacterium marinum modulates phagosome maturation. Cell. Microbiol. 8, 1417–1429. Van Crevel, R., Ottenhoff, T.H.M., Van Der Meer, J.W.M., 2002. Innate immunity to Mycobacterium tuberculosis. Clin. Microbiol. Rev. 15, 294–309. Windhagen, A., Anderson, D.E., Carrizosa, A., Williams, R.E., Hafler, D.A., 1996. IL-12 induces human T cells secreting IL-10 with IFN-gamma. J. Immunol. 157, 1127– 1131. Wolf, A.J., Linas, B., Trevejo-Nuñez, G.J., Kincaid, E., Tamura, T., Takatsu, K., Ernst, J.D., 2007. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J. Immunol. 179, 2509–2519. Xu, J., Laine, O., Masciocchi, M., Manoranjan, J., Smith, J., Du, S.J., Edwards, N., Zhu, X., Fenselau, C., Gao, L.-Y., 2007. A unique Mycobacterium ESX-1 protein co-secretes with CFP-10/ESAT-6 and is necessary for inhibiting phagosome maturation. Mol. Microbiol. 66, 787–800. Yoshida, A., Koide, Y., Uchijima, M., Yoshida, T.O., 1994. IFN-c induces IL-12 mRNA expression by a murine macrophage cell line, J774. Biochem. Biophys. Res. Commun. 198, 857–861. Zhang, M., Gong, J., Yang, Z., Samten, B., Cave, M.D., Barnes, P.F., 1999. Enhanced capacity of a widespread strain of Mycobacterium tuberculosis to grow in human macrophages. J. Infect. Dis. 179, 1213–1217.
Contents lists available at SciVerse ScienceDirect
Developmental and Comparative Immunology journal homepage: www.elsevier.com/locate/dci
Analysis of the immune response in infections of the goldfish (Carassius auratus L.) with Mycobacterium marinum Jordan W. Hodgkinson a,1, Jun-Qing Ge a,c,1, Leon Grayfer a,1, James Stafford a, Miodrag Belosevic a,b,⇑ a
Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada School of Public Health, University of Alberta, Edmonton, Alberta, Canada c Institute of Biotechnology, Fujian Academy of Agricultural Sciences, Fuzhou, Fijian, China b
a r t i c l e
i n f o
Article history: Received 12 May 2012 Revised 9 July 2012 Accepted 11 July 2012 Available online 7 August 2012 Keywords: Mycobacteria Fish Macrophage Respiratory burst Nitric oxide NADPH iNOS Cytokines Cytokine receptors
a b s t r a c t The rapid doubling time and genetic relatedness of the fish pathogen Mycobacterium marinum to Mycobacterium tuberculosis has rendered the former an attractive model for investigating mycobacterial hostpathogen interactions. We employed the M. marinum-goldfish infection model to investigate the in vivo immune responses to this pathogen in the context of a natural host. Histological analysis revealed mycobacterial infiltrates in goldfish kidney and spleen tissues, peaking 28 days post infections (dpi). Quantitative gene expression analysis showed significant increases of mRNA levels of pro-inflammatory cytokines (IFNc, IL-12p40, IL-1b1) and cytokine receptors (IFNGR1-1, TNFR2) at 7 dpi. Conversely, the gene expression levels of key anti-inflammatory cytokines TGFb and IL-10 were elevated at 14 dpi. Furthermore, M. marinum infections markedly increased the cytokine-primed oxidative burst responses of isolated kidney phagocytes at 7 but not 56 dpi. We believe that the M. marinum-goldfish infection model will be invaluable in furthering the understanding of the mycobacterium host-pathogen interface. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Pathogenic mycobacteria are facultative intracellular pathogens that infect and reside inside host phagocytic cells. The currently recognized Mycobacterium genus consists of 130 species and 11 subspecies, including disease-causing human pathogens Mycobacterium tuberculosis, Mycobacterium leprae and Mycobacterium ulcerans (Gauthier and Rhodes, 2009). While the understanding of the interactions of these pathogens with their hosts’ immune systems is pivotal, very slow replication of mycobacteria causing human infections (such as M. tuberculosis), coupled with level 3-facility requirements for these pathogens have hindered the progress of research in this area (Gideon and Flynn, 2011; Russell et al., 2010; Sridhar et al., 2011). Mycobacterium marinum is an etiological agent of mycobacteriosis in fish, amphibians, and aquatic mammals (Ramakrishnan et al., 1997; Flowers, 1970; Swaim et al., 2006), and shares close genetic resemblance to M. tuberculosis (possessing 99.4% 16S RNA sequence identity). In
⇑ Corresponding author at: Department of Biological Sciences, CW-405 Biological Sciences Building, University of Alberta, Edmonton, AB, Canada T6G 2E9. Tel.: +1 780 492 1266; fax: +1 780 492 9234. E-mail address: [email protected] (M. Belosevic). 1 These authors contributed equally to this work. 0145-305X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dci.2012.07.006
addition, M. marinum exhibits relatively short doubling times (Rogal et al., 1990) and causes pathology similar to that caused by mammalian mycobacteria, including granulomatous inflammation and spherical lesions encased by layers of epithelioid cells (Gauthier and Rhodes, 2009). Accordingly, M. marinum represents an attractive comparative model for studying mycobacterium pathogenesis and the mechanisms of host defense. M. marinum infection studies using zebrafish embryos have underlined the importance of macrophages in the dissemination and clearance of this pathogen, where macrophage depletions produced more robust infections and bacterial growth (Clay et al., 2007). These embryo infections (prior to development of adaptive immune components) resulted in the formation of macrophage aggregates with hallmark granuloma structures even in the absence of lymphocytes, while the infecting M. marinum expressed newly identified, granuloma-specific genes (Davis et al., 2002). Thus, where classical granuloma formation has been described as an organization of macrophages and lymphocytes, the above suggests that the adaptive components were not absolutely required in the formation of these structures in zebrafish. Furthermore, zebrafish infections studies revealed that granuloma formation was actively induced and maintained by mycobacterium rather than the host immune system (Davis et al., 2002). This was documented by showing that M. marinum actively recruited macrophages to
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
primary granulomas, facilitated the infection of new host phagocytes, bacterial dissemination and seeding of secondary granulomas (Davis and Ramakrishnan, 2009). In fact, M. marinum actively redirected the embryonic macrophages from predetermined migration pathways dictated by development (Davis et al., 2002). M. marinum also effectively subverted the phagosome maturation of both mammalian (Barker et al., 1997) and fish (El-Etr et al., 2001) macrophages, by preventing phagosome acidification (Tan et al., 2006; Xu et al., 2007) and employed a number of strategies for evading and/or dampening the phagocyte oxidative burst responses after IFNc-stimulation (Subbian et al., 2007a,b; Pagán-Ramos et al., 2006). Our previous in vitro studies of M. marinum-infected goldfish phagocytes underlined the capability of this pathogen to significantly ablate pro-inflammatory cytokine-induced ROI and NO production by down-regulating gene expression of the NADPH oxidase components and up-regulating immunosuppressive genes such as IL-10, TGFb and SOCS-3 (Grayfer et al., 2011a). Indeed, pathogenic Mycobacterium spp. in general, possess a plethora of subversion strategies for overcoming phagocyte microbicidal responses and making the intraphagosomal environment hospitable for bacterial growth (reviewed by Flannagan et al., 2009 and Flynn and Chan, 2003). The use of the goldfish as an in vivo M. marinum infection model was first proposed by Talaat et al. (1998) for the identification of putative M. marinum pathogenicity determinants (Ruley et al., 2004). Here we report on the use of the goldfish M. marinum infection model to investigate the mechanisms of host defense elicited by this pathogen in the context of a natural host. 2. Materials and methods 2.1. Fish Goldfish (Carassius auratus L.) were purchased from Aquatic Imports (Calgary, AB) and maintained at the Aquatic Facility of the Department of Biological Sciences, University of Alberta. The fish were kept at 17 °C in a flow-through water system on a simulated natural photoperiod, and fed to satiation daily with trout pellets. The fish were acclimated to this environment for at least 3 weeks prior to use in the experiments. All of the fish ranged from 10 to 15 cm in length and whenever possible an equal number of both sexes were used. Prior to sacrifice, fish were sedated using TMS (tricaine methane sulfonate) at 40–50 mg/L. 2.2. Bacteria M. marinum, strain ATCC 927 (fish isolate) was provided by Dr. Lourens Robberts, School of Public Health, University of Alberta. M. marinum was grown with shaking at 30 °C as a dispersed culture in 7H9 (Difco, Detroit, MI) broth supplemented with 0.5% glycerol and 10% albumin-dextrose complex for 7 to 10 days. The colony forming units (cfu) per milliliter were determined by plating serial dilutions of cultures on Middlebrook 7H10 agar (Difco) supplemented with 0.5% glycerol and 10% albumin-dextrose. Prior to all experiments, bacterial cultures were dispersed by 10–15 passages through a 25-gauge needle. When required, enumerated bacterial cultures were heat killed by incubation in an 80 °C water bath for 30 min. Heat killing efficiency (loss of bacterial viability) was confirmed by plating heat killed M. marinum cultures on Middlebrook 7H10 agar (Difco), and failure of the bacteria to grow after 5 days of incubation. Fish were infected by intra-peritoneal injection of 106 CFU of M. marinum diluted in 1X PBS. CFU/fish were determined after 7, 14, 28 and 56 days post infection (dpi) by plating 10-fold dilutions of
457
tissue homogenates on Middlebrook 7H10 agar (Difco) supplemented with 0.5% glycerol and 10% albumin-dextrose. Colonies were confirmed as acid-fast staining bacteria using staining procedures as suggested by the manufacturer (BD Diagnostics, NJ). 2.3. Preparation of histological sections Spleen, kidney and liver tissues were isolated from goldfish 28 days following PBS or M. marinum injections. Tissues were fixed in 10% neutral buffered formalin for 48 h at room temperature and embedded in paraffin wax after processing using Tissue Processor TP1020 (Leica Biosystems, Wetzlar, Germany). Five 5 lm sections were prepared using a rotary microtome. Slides were de-waxed, stained with Carbol Fuchsin (5% phenol, 10% ethanol, 1% w/v Basic Fuchsin (BioRad, USA) and counterstained with Methylene Blue (0.5% Methylene Blue w/v, 0.5% glacial acetic acid) (Ellis and Zabrowarny, 1993) and acid-fast (Ziehl-Neelsen). Stained tissue sections were observed at 100X magnification. 2.4. Production and purification of goldfish rgTNF, rgIFNc, and rgIFNcrel The production of recombinant goldfish tumor necrosis factor alpha 2 (rgTNFa2), recombinant goldfish interferon gamma (rgIFNc) and recombinant goldfish interferon gamma related (rgIFNcrel) cytokines have been described previously (Grayfer and Belosevic, 2009a,b; Grayfer et al., 2008, 2010, 2011b). The recombinant proteins were concentrated using polyethylene glycol flakes (35 kDa), dialyzed overnight against 1X PBS and subsequently passed through EndoTrap Red endotoxin removal columns (Hyglos, Bernried, Germany). Purification of the recombinant cytokines was confirmed by Western blot, and the identity of the proteins confirmed by mass spectrometry. Protein concentrations were determined using a Micro BCA Protein Assay Kit (Pierce, Rockford, IL) and the proteins were stored at 4 °C, prior to use. 2.5. Analysis of M. marinum-induced gene expression in goldfish spleen, kidney, and kidney derived leukocytes Total RNA was isolated from spleen, kidney and isolated kidney leukocytes using TRIzol extraction and reverse transcribed into cDNA using the Superscript II cDNA synthesis kit according to manufacturer’s directions. The genes examined included: CCL1, CXCL8 (IL-8), IFNc, IFNcrel, IL-12 p35, IL-12 p40, IL-1b1, IL-1b2, TNFa1, TNFa2, TGFb, IL-10, SOCS-3, IFNGR1-1, IFNGR1-2, TNFR1, TNFR2, iNOS A, iNOS B, p22phox, p40phox, p47phox, p67phox and gp91phox. Expression analysis of all genes was performed using the 2^-deltaCT method relative to endogenous control gene, elongation factor -1 alpha (EF-1a) and the derived RQ values were normalized against the respective PBS-injected controls. All primer sequences used in this study are shown in Supplementary Table 1. 2.6. Respiratory burst assay Kidney-derived monocytes isolated from M. marinum infected or PBS-injected goldfish (7 or 56 dpi) were seeded into 96 well plates at a density of 5 105 cells per well. Cells were treated with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNc and incubated for an additional 16 h. The nitroblue tetrazolium (NBT) assay was performed as described previously (Grayfer and Belosevic, 2009b; Grayfer et al., 2008). Absorbances (630 nm) from cells alone (no PMA) were subtracted from those of treatment groups to account for background NBT reduction. Medium only-treated, PMAtriggered cells were negative controls. Leukocytes were obtained from cultures established from 5 individual fish for each treatment group (n = 5).
458
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
2.7. Nitric oxide assay Kidney-derived macrophages isolated from M. marinum infected or PBS-injected goldfish (7 or 56 dpi), were seeded into 96 well plates at a density of 1 106 cells per well and incubated for 3 h to allow cell adherence. The media containing the nonadherent cells were carefully removed from individual wells. The adherent cells were treated with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNcrel in a final volume of 100 lL and incubated for 72 h. Nitrite production in cell supernatants was determined using the Griess reaction (Stafford et al., 2004). Absorbance was measured at 540 nm and nitrite concentration determined using a standard curve. Leukocytes were obtained from cultures established from five individual fish for each treatment group (n = 5). 2.8. Flow cytometry analysis of goldfish kidney leukocytes Kidney leukocytes isolated from goldfish 7 or 56 days subsequent to PBS or M. marinum injections were assessed by flow cytometry using previously established parameters (Neumann et al., 2000a). Cells were discriminated based on forward and side scatter against pre-established FACS gates known to describe progenitor/lymphocyte, monocyte or mature macrophage populations (R1, R3, and R2, respectively). Flow cytometry was performed using FACSCalibur apparatus (Beton/Dickinson) and data analysis was performed using CellQuest software (Beton/Dickinson). 2.9. Statistical analysis Statistical analysis was performed using one-way analysis of variance (ANOVA). P < 0.05 was considered statistically significant. 3. Results 3.1. Enumeration of bacterial loads and histology Following goldfish infections with M. marinum, bacterial burdens were determined over the course of 56 days. The magnitude of M. marinum infected-fish derived colony forming units (CFUs) peaked at 28 dpi while the CFUs per fish were lower on 56 dpi, as compared to 28 dpi (Fig. 1A). The acid-fast (Ziehl-Neelsen) staining of M. marinum infected fish showed mycobacterial granuloma structures in the spleen, the kidney (Fig. 1B), as well as liver tissue (data not shown). Purple staining, indicative of acid-fast bacteria, was observed in the centers of these granulomas (Fig. 1B). 3.2. M. marinum infections elicit changes in pro-inflammatory gene expression in goldfish tissues In order evaluate the goldfish inflammatory responses to M. marinum infections, we performed gene expression studies of key pro-inflammatory genes in kidney and spleen tissues of PBSinjected and M. marinum infected goldfish (Fig. 2). Infected fish exhibited increased mRNA levels of IFNc, IL-12 (p35 and p40 subunits), and IL-1b1 in the kidney tissue at 7 dpi (Fig. 2A), with a return to baseline mRNA levels by 14 dpi (Fig. 2B). On both 28 and 56 dpi, the gene expressions of pro-inflammatory cytokines were substantially reduced compared to controls (Fig. 2C and D). Interestingly, at 7, 28, and 56 dpi, IL-1b1 mRNA levels were significantly higher in M. marinum infected fish as compared to PBS-injected controls (Fig. 2A, C, and D). The pattern of pro-inflammatory gene expressions in the spleen were similar to those observed in the kidney (Fig. 3A–D), where short-lived increases in mRNA levels (7 dpi) were followed by
returns to baseline expression levels by 14 dpi (Fig. 3A and B). Additionally, by 56 dpi pro-inflammatory cytokine transcript levels were significantly down-regulated in infected fish, with the exception of IL-1b1, which remained elevated at both 28 and 56 dpi (Fig. 3C and D). 3.3. M. marinum infections elicit changes in the anti-inflammatory gene expression in goldfish tissues The expression of anti-inflammatory genes, transforming factor beta (TGFb), interleukin-10 (IL-10), and suppressor of cytokine signaling-3 (SOCS-3) were measured in the kidney (Fig. 4) and spleen (Fig. 5) during the course of M. marinum infections. While only SOCS-3 mRNA levels were substantially elevated in the kidney at 7 dpi, the expressions of all three anti-inflammatory genes were significantly higher compared to those of PBS-injected fish at 14 dpi (Fig. 4A and B). By 28 and 56 days dpi, the mRNA levels of all three anti-inflammatory genes decreased compared to PBS-injected controls (Fig. 4C and D). In contrast, the spleen mRNA levels of TGFb and IL-10 of infected fish were significantly lower at 7 dpi compared to PBS-injected controls, and by 14 and 28 dpi, mRNA levels were similar between the infected and control fish (Fig. 5A and B). As with the kidney gene expression patterns, the mRNA levels of TGFb, IL-10 and SOCS-3 were also significantly diminished at 56 dpi compared to those of PBS-injected fish (Fig. 5D). 3.4. M. marinum infections alter goldfish kidney and spleen cytokine receptor gene expression patterns In addition to assessing the gene expression of pro-inflammatory cytokines, we also examined the transcriptional changes of cognate cytokine receptors (IFNGR1-1, IFNGR1-2, TNFR1, and TNFR2) in the kidney (Fig. 6A and B) and the spleen of M. marinum-infected goldfish (Fig. 6C and D) after 7 and 56 dpi. The mRNA levels of IFNGR1-1 significantly increased in the kidney and the spleen at 7 dpi (Fig. 6A and C), but were down-regulated by 56 dpi (Fig. 6B and D). The mRNA levels of TNFR2 were significantly elevated in the kidneys, but not spleens at 7 dpi (Fig. 6A and C) while the expression of this receptor was significantly downregulated in both the kidney and the spleen at 56 dpi (Fig. 6B and D). Additionally, the kidney gene expression of TNFR1 was also significantly increased at 56 dpi (Fig. 6B). 3.5. M. marinum-infected goldfish-derived kidney monocytes and macrophages produce greater levels of reactive oxygen and reactive nitrogen intermediates We examined the ability of kidney-derived monocytes/macrophages from M. marinum or PBS injected fish to generate reactive oxygen or reactive nitrogen (ROI and NO, respectively) antimicrobial responses after 7 or 56 dpi. In our previous studies, we observed that while rgTNFa2 was a potent elicitor of both goldfish phagocyte ROI and NO, rgIFNc was effective at priming ROI but not the NO responses. In contrast, rgIFNcrel induced robust NO production but not ROI responses of goldfish macrophages and monocytes, respoctively. Accordingly, in the present work, rgTNFa2 and rgIFNc were employed for eliciting phagocyte ROI responses and rgTNFa2 and rgIFNc were used for studying the NO production. Monocyte PMA-triggered ROI production was measured subsequent to priming with medium alone, 100 ng/mL of rgTNFa2 or 100 ng/mL of rgIFNc (Fig. 7). Cytokine-primed (rgTNFa2 or rgIFNc) cells from M. marinum-injected fish, isolated 7 dpi, exhibited significantly enhanced ROI responses compared to the respective PBS injected control fish (Fig. 7A). Interestingly, while the baseline ROI production of monocytes obtained from PBS-injected andM.
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
459
Fig. 1. The course of infection with Mycobacterium marinum in the goldfish. (A) The kidney and spleen were homogenized and the colony forming units (cfu) per organ were determined by plating serial dilutions of cultures on Middlebrook 7H10 agar. The data are the mean ± SEM CFU/fish of five fish (n = 5). () indicate significantly different between 28 days post infection (dpi) compared to 7 or 14 dpi. (B) The kidney and spleen were removed 28 dpi. Mycobacterium infected fish had granulomas with red-stained centers (arrows), indicative of infection. Tissue sections were stained using acid-fast (Ziehl-Neelsen) stain and visualized at 100X.
marinum-infected fish was elevated at 56 dpi, no differences were seen in the ROI production between the two experimental groups (Fig. 7B). We also measured the mRNA levels of NADPH oxidase subunits, the enzyme responsible for ROI production. While the expressions of phox genes in kidney tissues of infected and control fish were similar at 7 dpi (Fig. 8A), there was a significant reduction in the expression of p22phox, p40phox and p67phox at 56 dpi (Fig. 8B). Kidney macrophages from M. marinum infected and PBSinjected goldfish were assessed for their ability to mount a nitric oxide response following stimulation with medium alone, 100 ng/mL of rgTNFa2 or rgIFNcrel (Fig. 9). Medium-treated macrophages derived from fish at 7 dpi produced significantly more NO than cells from PBS injected controls (Fig. 9A). In addition, NO
responses of kidney leukocytes obtained from infected fish compared to PBS-injected controls, was further enhanced after treatment with rgTNFa2 or rgIFNcrel (Fig. 9A). In contrast, macrophages isolated from fish at 56 dpi and treated with medium alone or with rgTNFa2 produced similar nitrite levels compared to the controls (Fig. 9B). However, at 56 dpi, rgIFNcrel-treated cells obtained from M. marinum-infected fish had a significantly higher NO responses compared to those obtained from PBS-injected control fish (Fig. 9B). We also measured the expression of iNOS A and iNOS B in the kidneys of M. marinum infected and PBS-injected fish (Fig. 10). The mRNA levels of both iNOS A and iNOS B increased significantly at 7 dpi (Fig. 10A). In contrast, by 56 dpi, while the iNOS A mRNA
460
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 2. Mycobacterium marinum infection alters the expression of pro-inflammatory genes in the kidney determined using quantitative PCR. Kidney tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 3. Mycobacterium marinum infection alters gene expression of pro-inflammatory genes in the spleen determined using quantitative PCR. Spleen tissues were harvested (A) 7, (B) 14, (C) 28 and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
461
Fig. 4. Mycobacterium marinum infection alters gene expression of anti-inflammatory genes in the kidney determined using quantitative PCR. Kidney tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 5. Mycobacterium marinum infection alters gene expression of anti-inflammatory genes in the spleen determined using quantitative PCR. Spleen tissues were harvested (A) 7, (B) 14, (C) 28, and (D) 56 days post infection (dpi). The expression data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
levels remained significantly elevated in cells from infected fish, the gene expression of iNOS B returned to levels comparable to PBS-injected controls. In order to elucidate whether the enhanced ROI production by cells from acutely infected fish stemmed from changes in immune cell populations, kidney leukocytes isolated from PBS and M. marinum injected fish were assessed by flow cytometry using previ-
ously established parameters (Neumann et al., 2000a, Suppl. Fig. 1). Based on our prior work, we have established that the cells in the R1 gate consist of lymphocytes and progenitor cells. As goldfish kidney monocytes mature, they shift into the R3 gate and upon further differentiation into macrophages, these cells enter the R2 gate (Neumann et al., 2000a,b). Kidney leukocytes isolated from fish 7 days subsequent to PBS injections, consisted primarily of
462
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 6. Mycobacterium marinum infection alters gene expression of TNF and IFN receptors in the kidney and the spleen tissues determined using quantitative PCR. Kidney tissues were harvested (A) 7 and (B) 56 days post infection (dpi), and in the spleen (C) 7 and (D) 56 dpi. The data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
immature monocytic cells (gate R3) and lymphocytes/myeloid precursor cells (gate R1) and relatively few mature macrophages (Suppl. Fig. 1). In contrast, kidney leukocytes isolated from fish infected for 7 days with M. marinum comprised of further differentiated monocytic populations and of substantially more mature macrophages, compared to the PBS controls (Suppl. Fig. 1). Interestingly, the kidney leukocyte isolated from M. marinum-infected fish 56 dpi exhibited cell compositions similar to those seen in PBS-injected fish at 7 and 56 dpi (Suppl. Fig. 1). The results presented in Supplementary Fig. 1 are representative of those observed using five individual fish for each experimental group.
4. Discussion In the present study, we describe the use of an in vivo goldfish model system to characterize the host innate immune defenses during the course of M. marinum infections. The parallel analysis of expression patterns of immune genes and antimicrobial capacities of goldfish phagocytes offered interesting insights of this complex host-pathogen association. Our findings indicate that there is a relationship between the dissemination of bacteria and the resulting infection-associated pathology, and the pro- and anti-inflammatory gene expression profiles, and the capacity of goldfish phagocytes to mount appropriate antimicrobial responses. Ruley et al. (2002) have shown that in fish exposed to M. marinum, the viable bacteria can be detected in the tissues as early as 1 week after infection, while substantially greater mycobacterial loads and infection-related pathology occur several weeks after exposure. Our findings support these observations, where we also observed greatest mycobacterial loads and significant granuloma formation in goldfish organs at 28 dpi. It should be noted that the bacterial loads in the goldfish remained
high at 56 dpi, attesting to the chronicity of this infection, and failure of the fish to eliminate the pathogen. Innate immune responses against pathogenic mycobacteria are critical in determining the fate of these infections. Using both mammalian and zebrafish infection models, it has been demonstrated that phagocyte recruitment to the sites of infection and the uptake of pathogens by the phagocytes may serve as a means of mycobacterial dissemination (Clay et al., 2007; Wolf et al., 2007; Davis and Ramakrishnan, 2009). During this initial uptake of mycobacteria by phagocytes, microbial components are recognized by pattern recognition receptors and this interaction results in the induction of pro-inflammatory genes (Chan et al., 2001; Zhang et al., 1999; Grayfer et al., 2011a). Our gene expression results indicate that M. marinum infection induced an increase in the expression of pro-inflammatory cytokines during the first week of infection. This modulation is presumably important for the induction of antimicrobial defenses, but paradoxically may also aid mycobacteria to establish a foothold during the infection process (Clay et al., 2007). Indeed, as demonstrated in the zebrafishM. marinum infection model, the pro-inflammatory cytokine TNFa is crucial to thwarting the infection as well as maintaining the granuloma structures (Clay et al., 2008), the latter in turn facilitating bacterial survival and eventual dissemination (Davis et al., 2002). A return to the baseline gene expression of pro-inflammatory mediators was observed in conjunction with increased expression of anti-inflammatory cytokines at 14 dpi. Interestingly, a consistent trend of down-regulation of the expression of pro-inflammatory cytokines was observed at 56 dpi. After phagocytosis, mycobacteria that reside in the phagosomes have been shown interfere with mechanisms of phagolysosomal fusion (Armstrong and Hart, 1971; Mwandumba et al., 2004; Nguyen and Pieters, 2005). Furthermore, intraphagosomal mycobacteria also interfere with host
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
463
Fig. 8. Mycobacterium marinum infections alter the expression of NADPH oxidase components in primary kidney leukocytes. Quantitative gene expression analysis of p22phox, p40phox, p47phox, p67phox and gp91phox NADPH components were assessed in kidneys leukocytes isolated 7(A) and 56(B) days post M. marinum infections (dpi) or PBS injected controls. The data were normalized against PBS-injected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
Fig. 7. Mycobacterium marinum infection induces goldfish kidney monocyte reactive oxygen production. Kidney monocytes were isolated from M. marinum infected or PBS-injected goldfish 7 (A) or 56 (B) days post infection and primed with medium alone or with 100 ng/mL of rgTNFa2 or rgIFNc for 16 h. The ROI response was then triggered using PMA (100 ng/mL). Kidney leukocytes were obtained from four to five individual fish for each treatment group (n = 4–5). Statistical analysis was performed using one-way ANOVA. () indicate significantly different (P < 0.05) from the PMA only treated controls, and (+) above lines denotes significant difference (P < 0.05) between different experimental groups.
cell activation states. They have been shown to induce the production of anti-inflammatory mediators, which effectively dampen the antimicrobial functions of phagocytes, creating an environment suitable for chronic infection (Van Crevel et al., 2002; Marino et al., 2010; Grayfer et al., 2011a). A persistent increase in the expression of IL-1b1 was observed during the course of the M. marinum infections. IL-1b has been shown to decrease the host susceptibility to mycobacteria (Mayer-Barber et al., 2010; Michelini-Norris et al., 1992). Although 20-fold and 10-fold increases in the expression of IL-1b1 were measured during the acute and chronic phases of the infection respectively, bacterial infection still persisted in the goldfish. The consistent IL-1b response was clearly not sufficient to contribute to the elimination of M. marinum. It is likely that other pro-inflammatory cytokines, whose expression decreased during the chronic phase of the infection, synergistically act with IL-1b1 for efficient clearance of mycobacteria. It is also possible that downstream interference by M. marinum through an unknown mechanism(s) (i.e. inflammasome activation, IL-1R engagement) prevented its clearance from the goldfish. We previously reported that in vitro M. marinum infected goldfish monocytes and macrophages exhibit bacterial load-dependent abrogation of antimicrobial capabilities and increased expression of anti-inflammatory genes (Grayfer et al., 2011a). Presumably the initial pro-inflammatory conditions seen at 7 dpi, coincide with a general stimulation of the goldfish immune systems by the extremely immunogenic mycobacterial components. Conversely, the
Fig. 9. Mycobacterium marinum infection induces goldfish kidney macrophage nitric oxide production. Kidney macrophages, isolated from M. marinum infected or PBS-injected goldfish (A) 7 or (B) 56 days post infection (dpi) were incubated with medium alone, or with 100 ng/mL of rgTNFa2 or rgIFNcrel for 72 h. Nitrite production was determined by Griess reaction and nitrite concentrations determined using a nitrite standard curve. Leukocytes were obtained from four to five individual fish for each treatment group (n = 4 or 5). Statistical analysis was performed using one-way ANOVA. () indicate significantly different (P < 0.05) from medium only treated controls, and (+) above lines denotes significant difference (P < 0.05) between experimental groups.
464
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465
Fig. 10. Mycobacterium marinum infection increases the expression of iNOS A and iNOS B in primary kidney monocytes/macrophages. Cells were harvested (A) 7 and (B) 56 days post infection (dpi). The expression data were normalized against PBSinjected controls. Results are means ± SEM of RQ values from five fish (n = 5). () indicate significantly different from PBS-injected control (P < 0.05).
general suppression of immune gene expression later in the infection may be due to the induction of anti-inflammatory cytokine production and increased bacterial load in infected fish. Kidney-derived monocytes and macrophages isolated from infected goldfish at 7 dpi exhibited increased ROI and NO production, respectively, the magnitudes of which were further enhanced by stimulation with rgTNFa2 and rgIFNc or rgIFNcrel. These increases in ROI and NO responses coincided with increases in the expression of select phox and iNOS genes, respectively. Interestingly, the ROI production by monocytes isolated from fish at 56 dpi was similar to that of monocytes from PBS-injected control fish. Our preliminary flow cytometry data suggests that the isolated kidney leukocyte cultures from goldfish at 7 dpi, comprised of more differentiated subpopulations of monocytes and mature macrophages, compared to those isolated from PBS-injected fish (both 7 and 56 dpi) and after 56 dpi. Additionally, cells isolated from fish 7 dpi also exhibited increased expression of the IFNGR1-1 and the TNF receptors, presumably contributing to the increased responsiveness of these cells to the cognate recombinant ligands (rgIFNc, rgIFNcrel and TNFa2). Together these corroborating lines of evidence suggest that during acute but not chronic mycobacterium infections, goldfish myelopoiesis may be altered towards the production of more effective microbicidal phagocytes. Concomitantly, at 7 dpi there was a significant up-regulation in the expression of macrophage activating cytokines, IFNc and IL-12. It is well established that IL-12 and IFNc operate in a positive feedback manner, where the production of either cytokine confers the synthesis of the other by distinct immune cell populations (Bliss et al., 1996; Ma et al., 1996; Windhagen et al., 1996; Yoshida et al., 1994). This cross-regulation would result in an enhanced production of both cytokines and may also account for the marked increases in antimicrobial capacities of kidney monocytes and macrophages obtained from fish at 7 dpi. In fact, it is well established that IFNc stimulation of myeloid cells enhances their reac-
tive oxygen (Cassatella et al., 1990) and reactive nitrogen (Martin et al., 1994) producing capabilities. On 14 dpi, the kidney gene expression of the pro-inflammatory cytokines returned to baseline transcription, and by 56 dpi their mRNA levels were significantly lower than those in control fish. It is possible that the increased gene expression of the anti-inflammatory cytokine IL-10 in the kidney at 14 dpi, may have contributed to the deactivation of the monocyte/macrophage antimicrobial functions. In addition, the dissemination of bacteria and the resulting infection associated pathology are likely reflected in the gene expression profiles and antimicrobial response changes reported here. In a series of in vivo M. marinum-goldfish infection studies, it has been demonstrated that although bacteria were detected after a week of infection, substantially greater mycobacterial loads and infection-related pathology were seen at later times in the course of infection (Ruley et al., 2002). These results are corroborated here, where we observed expansive granuloma formation in the spleens of 30-day infected fish. Undoubtedly, these factors play critical roles in the balance between the pathogen colonization and the host immunity. Also, we have previously observed that in vitro M. marinum infected goldfish monocytes and macrophages exhibit bacterial load-dependent abrogation of antimicrobial capabilities and increased expression of anti-inflammatory genes (Grayfer et al., 2011a). Presumably the initial pro-inflammatory conditions seen after 1 week of M. marinum infections may coincide with a general stimulation of the goldfish immune systems by the extremely immunogenic mycobacterial components. Conversely, the return to baseline and the general suppression of immune gene expression later on in the infection might be suggestive of increased bacterial proliferation, dissemination and immunomodulation. Our past and present work underlines not only the efficacy of using the goldfish as a natural host model for mycobacterial infections but also provides an additional perspective of the natural host-pathogen immune interactions. We believe that the goldfish represents an ideal platform for furthering our understanding of the mycobacterium-host interface, providing both in vitro and in vivo infection approaches with which to bridge the current gaps in our understanding of mycobacteri immune evasion strategies and the mechanisms of host defense. Acknowledgments This work was supported by the Natural Sciences and Engineering Council of Canada (NSERC) to MB. JWH was supported by an NSERC CGS-M and Alberta Innovates top-up award. LG was supported by an NSERC PGS-D scholarship. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.dci.2012.07.006. References Armstrong, J.A., Hart, P.D., 1971. Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J. Exp. Med. 134, 713. Barker, L.P., George, K.M., Falkow, S., Small, P.L., 1997. Differential trafficking of live and dead Mycobacterium marinum organisms in macrophages. Infect. Immun. 65, 1497–1504. Bliss, J., Van Cleave, V., Murray, K., Wiencis, A., Ketchum, M., Maylor, R., Haire, T., Resmini, C., Abbas, A.K., Wolf, S.F., 1996. IL-12, as an adjuvant, promotes a T helper 1 cell, but does not suppress a T helper 2 cell recall response. J. Immunol. 156, 887–894. Cassatella, M.A., Bazzoni, F., Flynn, R.M., Dusi, S., Trinchieri, G., Rossi, F., 1990. Molecular basis of interferon-gamma and lipopolysaccharide enhancement of phagocyte respiratory burst capability. Studies on the gene expression of several NADPH oxidase components. J. Biol. Chem. 265, 20241–20246.
J.W. Hodgkinson et al. / Developmental and Comparative Immunology 38 (2012) 456–465 Chan, E.D., Morris, K.R., Belisle, J.T., Hill, P., Remigio, L.K., Brennan, P.J., Riches, D.W.H., 2001. Induction of inducible nitric oxide synthase-NO by lipoarabinomannan of Mycobacterium tuberculosis is mediated by MEK1-ERK, MKK7-JNK, and NF-jB signaling pathways. Infect. Immun. 69, 2001–2010. Clay, H., Davis, J.M., Beery, D., Huttenlocher, A., Lyons, S.E., Ramakrishnan, L., 2007. Dichotomous role of the macrophage in early Mycobacterium marinum infection of the zebrafish. Cell Host Microbiol. 2, 29–39. Clay, H., Volkman, H.E., Ramakrishnan, L., 2008. Tumor necrosis factor signaling mediates resistance to mycobacteria by inhibiting bacterial growth and macrophage death. Immunity 29, 283–294. Davis, J.M., Clay, H., Lewis, J.L., Ghori, N., Herbomel, P., Ramakrishnan, L., 2002. Realtime visualization of Mycobacterium-macrophage interactions leading to initiation of granuloma formation in zebrafish embryos. Immunity 17, 693–702. Davis, J.M., Ramakrishnan, L., 2009. The role of the granuloma in expansion and dissemination of early tuberculous infection. Cell 136, 37–49. El-Etr, S.H., Yan, L., Cirillo, J.D., 2001. Fish monocytes as a model for mycobacterial host-pathogen interactions. Infect. Immun. 69, 7310–7317. Ellis, R.C., Zabrowarny, L.A., 1993. Safer staining method for acid fast bacilli. J. Clin. Pathol. 46, 559–560. Flannagan, R.S., Cosío, G., Grinstein, S., 2009. Antimicrobial mechanisms of phagocytes and bacterial evasion strategies. Nat. Rev. Microbiol. 7, 355–366. Flowers, D.J., 1970. Human infection due to Mycobacterium marinum after a dolphin bite. J. Clin. Pathol. 23, 475–477. Flynn, J.L., Chan, J., 2003. Immune evasion by Mycobacterium tuberculosis: living with the enemy. Curr. Opin. Immunol. 15, 450–455. Gauthier, D.T., Rhodes, M.W., 2009. Mycobacteriosis in fishes: a review. Vet. J. 180, 33–47. Gideon, H., Flynn, J., 2011. Latent tuberculosis: what the host ‘‘sees’’? Immunol. Res. 50, 202–212. Grayfer, L., Belosevic, M., 2009a. Molecular characterization, expression and functional analysis of goldfish (Carassius aurutus L.) interferon gamma. Dev. Comp. Immunol. 33, 235–246. Grayfer, L., Belosevic, M., 2009b. Molecular characterization of novel interferon gamma receptor 1 isoforms in zebrafish (Danio rerio) and goldfish (Carassius auratus L.). Mol. Immunol. 46, 3050–3059. Grayfer, L., Garcia, E.G., Belosevic, M., 2010. Comparison of macrophage antimicrobial responses induced by type II interferons of the goldfish (Carassius auratus L.). J. Biol. Chem. 285, 23537–23547. Grayfer, L., Hodgkinson, J.W., Belosevic, M., 2011a. Analysis of the antimicrobial responses of primary phagocytes of the goldfish (Carassius auratus L.) against Mycobacterium marinum. Dev. Comp. Immunol. 35, 1146–1158. Grayfer, L., Hodgkinson, J.W., Hitchen, S.J., Belosevic, M., 2011b. Characterization and functional analysis of goldfish (Carassius auratus L.) interleukin-10. Mol. Immunol. 48, 563–571. Grayfer, L., Walsh, J.G., Belosevic, M., 2008. Characterization and functional analysis of goldfish (Carassius auratus L.) tumor necrosis factor-alpha. Dev. Comp. Immunol. 32, 532–543. Ma, X., Chow, J.M., Gri, G., Carra, G., Gerosa, F., Wolf, S.F., Dzialo, R., Trinchieri, G., 1996. The interleukin 12 p40 gene promoter is primed by interferon gamma in monocytic cells. J. Exp. Med. 183, 147–157. Marino, S., Myers, A., Flynn, J.L., Kirschner, D.E., 2010. TNF and IL-10 are major factors in modulation of the phagocytic cell environment in lung and lymph node in tuberculosis: a next-generation two-compartmental model. J. Theor. Biol. 265, 586–598. Martin, E., Nathan, C., Xie, Q.W., 1994. Role of interferon regulatory factor 1 in induction of nitric oxide synthase. J. Exp. Med. 180, 977–984. Mayer-Barber, K.D., Barber, D.L., Shenderov, K., White, S.D., Wilson, M.S., Cheever, A., Kugler, D., Hieny, S., Caspar, P., Núñez, G., Schlueter, D., Flavell, R.A., Sutterwala, F.S., Sher, A., 2010. Cutting edge: caspase-1 independent IL-1b production is critical for host resistance to Mycobacterium tuberculosis and does not require TLR signaling in vivo. J. Immunol. 184, 3326–3330. Michelini-Norris, M.B., Blanchard, D.K., Pearson, C.A., Djeu, J.Y., 1992. Differential release of interleukin (IL)-1a, IL-1b, and IL-6 from normal human monocytes stimulated with an virulent and an avirulent isogenic variant of Mycobacterium avium-intracellular complex. J. Infect. Dis. 165, 702–709. Mwandumba, H.C., Russell, D.G., Nyirenda, M.H., Anderson, J., White, S.A., Molyneux, M.E., Squire, S.B., 2004. Mycobacterium tuberculosis resides in non acidified vacuoles in endocytically competent alveolar macrophages from patients with tuberculosis and HIV infection. J. Immunol. 172, 4592–4598.
465
Neumann, N.F., Barreda, D.R., Belosevic, M., 2000a. Generation and functional analysis of distinct macrophage sub-populations from goldfish (Carassius auratus L.) kidney leukocyte cultures. Fish Shellfish Immunol. 10, 1–20. Neumann, N.F., Stafford, J.L., Belosevic, M., 2000b. Biochemical and functional characterisation of macrophage stimulating factors secreted by mitogen-induced goldfish kidney leucocytes. Fish Shellfish Immunol. 10, 167–186. Nguyen, L., Pieters, J., 2005. The Trojan horse: survival tactics of pathogenic mycobacteria in macrophages. Trends Cell Biol. 15, 269–276. Pagán-Ramos, E., Master, S.S., Pritchett, C.L., Reimschuessel, R., Trucksis, M., Timmins, G.S., Deretic, V., 2006. Molecular and physiological effects of Mycobacterial oxyR inactivation. J. Bacteriol. 188, 2674–2680. Ramakrishnan, L., Valdivia, R.H., McKerrow, J.H., Falkow, S., 1997. Mycobacterium marinum causes both long-term subclinical infection and acute disease in the leopard frog (Rana Pipiens). Infect. Immun. 65, 767–773. Rogal, T., Wolters, J., Fohr, T., Bottger, E., 1990. Toward a phylogeny and definition of species at the molecular level within the genus Mycobacteria. Int. J. Syst. Bacteriol. 40, 323–330. Ruley, K.M., Ansede, J.H., Pritchett, C.L., Talaat, A.M., Reimschuessel, R., Trucksis, M., 2004. Identification of Mycobacterium marinum virulence genes using signaturetagged mutagenesis and the goldfish model of mycobacterial pathogenesis. FEMS Microbiol. Lett. 232, 75–81. Ruley, K.M., Reimschuessel, R., Trucksis, M., 2002. Goldfish as an animal model system for mycobacterial infection. In: Bacterial Pathogenesis Part C: Identification, Regulation, and Function of Virulence Factors. Academic Press, pp. 29–39. Russell, D.G., Barry, C.E., Flynn, J.L., 2010. Tuberculosis: what we don’t know can, and does, hurt us. Science 328, 852–856. Sridhar, S., Pollock, K., Lalvani, A., 2011. Redefining latent tuberculosis. Future Microbiol. 6, 1021–1035. Stafford, J.L., Wilson, E.C., Belosevic, M., 2004. Recombinant transferrin induces nitric oxide response in goldfish and murine macrophages. Fish Shellfish Immunol. 17, 171–185. Subbian, S., Mehta, P.K., Cirillo, S.L., Cirillo, J.D., 2007a. The Mycobacterium marinum mel2 locus displays similarity to bacterial bioluminescence systems and plays a role in defense against reactive oxygen and nitrogen species. BMC Microbiol. 7, 4. Subbian, S., Mehta, P.K., Cirillo, S.L.G., Bermudez, L.E., Cirillo, J.D., 2007b. A Mycobacterium marinum mel2 mutant is defective for growth in macrophages that produce reactive oxygen and reactive nitrogen species. Infect. Immun. 75, 127–134. Swaim, L.E., Connolly, L.E., Volkman, H.E., Humbert, O., Born, D.E., Ramakrishnan, L., 2006. Mycobacterium marinum infection of adult zebrafish causes caseating granulomatous tuberculosis and is moderated by adaptive immunity. Infect. Immun. 74, 6108–6117. Talaat, A.M., Reimschuessel, R., Wasserman, S.S., Trucksis, M., 1998. Goldfish, Carassius auratus, a novel animal model for the study of Mycobacterium marinum pathogenesis. Infect. Immun. 66, 2938–2942. Tan, T., Lee, W.L., Alexander, D.C., Grinstein, S., Liu, J., 2006. The ESAT-6/CFP-10 secretion system of Mycobacterium marinum modulates phagosome maturation. Cell. Microbiol. 8, 1417–1429. Van Crevel, R., Ottenhoff, T.H.M., Van Der Meer, J.W.M., 2002. Innate immunity to Mycobacterium tuberculosis. Clin. Microbiol. Rev. 15, 294–309. Windhagen, A., Anderson, D.E., Carrizosa, A., Williams, R.E., Hafler, D.A., 1996. IL-12 induces human T cells secreting IL-10 with IFN-gamma. J. Immunol. 157, 1127– 1131. Wolf, A.J., Linas, B., Trevejo-Nuñez, G.J., Kincaid, E., Tamura, T., Takatsu, K., Ernst, J.D., 2007. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo. J. Immunol. 179, 2509–2519. Xu, J., Laine, O., Masciocchi, M., Manoranjan, J., Smith, J., Du, S.J., Edwards, N., Zhu, X., Fenselau, C., Gao, L.-Y., 2007. A unique Mycobacterium ESX-1 protein co-secretes with CFP-10/ESAT-6 and is necessary for inhibiting phagosome maturation. Mol. Microbiol. 66, 787–800. Yoshida, A., Koide, Y., Uchijima, M., Yoshida, T.O., 1994. IFN-c induces IL-12 mRNA expression by a murine macrophage cell line, J774. Biochem. Biophys. Res. Commun. 198, 857–861. Zhang, M., Gong, J., Yang, Z., Samten, B., Cave, M.D., Barnes, P.F., 1999. Enhanced capacity of a widespread strain of Mycobacterium tuberculosis to grow in human macrophages. J. Infect. Dis. 179, 1213–1217.