b-Defensin Gene Expression during the Course of Experimental ...

2 downloads 0 Views 1MB Size Report
The kinetics of gene expression and the cellular source of murine b-defensin–3 (mBD3) and murine b-defensin–4. (mBD4) were determined in mouse models of ...
BRIEF REPORT

b-Defensin Gene Expression during the Course of Experimental Tuberculosis Infection Bruno Rivas-Santiago,1,2 Eduardo Sada,1 Vı´ctor Tsutsumi,2 Diana Aguilar-Leo´n,3 Juan Leon Contreras,3 and Rogelio Herna´ndez-Pando3 1

Department of Microbiology Research, National Institute for Respiratory Diseases, 2Department of Experimental Pathology, Center for Research and Advanced Studies, National Polytechnic Institute, and 3Experimental Pathology Section, Department of Pathology, National Institute of Medical Sciences and Nutrition Salvador Zubira´n, Mexico City, Mexico

The kinetics of gene expression and the cellular source of murine b-defensin–3 (mBD3) and murine b-defensin–4 (mBD4) were determined in mouse models of progressive pulmonary tuberculosis and latent infection induced by high or low infecting doses, respectively. During progressive disease, there was an initial rapid expression of both defensins by respiratory epithelial cells that correlated with temporary control of bacillary proliferation, but expression decreased during the later progressive phase of the disease. In latent infection, both defensins were expressed continuously, but they were suppressed after reactivation of the disease. Thus, mycobacterial infection induces the expression of mBD3 and mBD4, and both might participate in the control of mycobacterial growth. Mycobacterium tuberculosis is a pathogen capable of producing both progressive disease and latent infection [1]. The initial infection usually occurs in the lungs, and, in most cases, it is controlled by the immune system. Only 10% of these infections lead to progressive disease [1]. Even after the successful control of primary tuberculosis infection, some bacilli remain in tissues Received 2 February 2006; accepted 20 April 2006; electronically published 28 July 2006. Presented in part: 6th International Conference on the Pathogenesis of Mycobacterial Infections, Stockholm, 30 June–1 July 2005 (abstract 33). Potential conflicts of interest: none reported. Financial support: National Council of Science and Technology in Mexico (grant SEP-2004C01-47745); European Community (International Scientific Cooperation Projects grant ICA4-CT2002-10063). Reprints or correspondence: Dr. Rogelio Herna´ndez-Pando, Experimental Pathology Section, Dept. of Pathology, National Institute of Medical Sciences and Nutrition Salvador Zubira´n, Vasco de Quiroga 15, cp-14000, Tlalpan, Mexico City, Mexico ([email protected] or rhpando@quetzal .innsz.mx). The Journal of Infectious Diseases 2006; 194:697–701  2006 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2006/19405-0023$15.00

in a nonreplicating or slowly replicating dormant state for the rest of the individual’s life. This infectious state is called “latent tuberculosis infection”; it is clinically asymptomatic, and most cases of active tuberculosis infection arise as a result of the reactivation of dormant bacilli [1, 2]. Immunological studies in animal models of latent tuberculosis have demonstrated that cytokines such as tumor necrosis factor (TNF)–a and interferon (IFN)–g, as well as nitric oxide (NO), contribute significantly to keeping infection in the latent state [2]. This type of immune response is also crucial for protection during active human and murine tuberculosis. Hypothetically, an efficient innate immune response could be a significant factor in natural resistance against mycobacterial infection and in the induction of longterm control of bacillary growth during latent infection, but limited information is available. Alveolar macrophages and lung epithelial cells are the first cells that encounter M. tuberculosis during primary infection. Epithelial cells can produce relevant molecules of the innate immune system, such as b-defensins, a prototype cationic antimicrobial small peptide. Defensins directly contribute to the killing of microbes, and some of them have chemotactic activities on immune cells [3]. Mammalian defensins are divided into 3 subfamilies: a- and b-defensins, which differ in the placement and connectivity of their 6 conserved cysteine residues, and v-defensins, which have a unique circular structure [3]. Four human b-defensins (hBD1–4) have been well characterized [4–7]. However, many more b-defensins have been reported by computer gene analysis [4]. Defensins are expressed in epithelial tissues [5–7], and hBD2–4 are inducible, whereas the expression of hBD1 is constitutive [3]. Thus, hBD2–4 are likely to play a crucial role against bacterial infection as inducible components of the epithelial barrier [5, 6]. Murine bdefensin–3 (mBD3) is the analogue of hBD2, and murine bdefensin–4 (mBD4) is the analogue of hBD3. We previously reported that in vitro infection of the lung epithelial-cell line A549 with M. tuberculosis induced the production of hBD2, and immunoelectronmicroscopic results revealed that hBD2 was associated with disrupted mycobacteria, which suggested that this molecule might participate in mycobacterial lysis [8]. The aim of the present study was to extend this in vitro observation by the analysis of mBD gene expression in vivo using murine models of latent infection [9] and progressive pulmonary tuberculosis [10]. Material and methods. The virulent M. tuberculosis strain H37Rv was cultured as described elsewhere [9, 10]. To induce latent infection, 8-week-old B6D2F1 mice were anesthetized, BRIEF REPORT • JID 2006:194 (1 September) • 697

tracheas were exposed, and 4 ⫻ 10 3 bacilli were injected [9]. Groups of 6 mice in 2 different experiments were killed at 1, 3, 5, and 7 months after infection. To induce reactivation, 10 mice with stable latent infection were treated with corticosterone (3 mg/L dissolved in drinking water) after 7 months and then were killed 2 and 4 weeks later. To induce progressive pulmonary tuberculosis, male BALB/ c mice were anesthetized, tracheas were exposed, and 2 ⫻ 10 5 viable bacilli were injected [10]. Groups of 6 mice were killed at 1, 3, 7, 14, 21, 28, and 60 days after infection. All animal work was performed in accordance with guidelines of the institutional Ethics Committee for Experimentation in Animals. Lungs from 3 different mice in 2 different experiments from both models were perfused with 100% ethanol via the trachea and were embedded in paraffin [9, 10]. Lung sections were incubated for 18 h with goat anti-mBD3 and anti-mBD4 antibodies (Santa Cruz Biotechnology). Bound antibodies were detected with avidin-biotin peroxidase (Vector). For quantification, we evaluated 3 different mouse lungs per time point in 2 different experiments. Five random fields were analyzed at ⫻400 magnification. At least 400 negative or positive cells per field were counted using an image analyzer (QWin Leica). In the bronchial epithelium, the morphological characteristics of each positive cell were evaluated, and the percentage of immunostained ciliated or goblet cells was determined. Three lungs, each from a different mouse, were processed with Trizol (Gibco BRL) at each time point during progressive disease, latent infection, and reactivation for the isolation of RNA [9, 10]. Reverse mRNA transcription was performed using oligo-dT and the Omniscript kit (Qiagen). Real-time polymerase chain reaction (PCR) was performed using the 7500 real-time PCR system (Applied Biosystems). Standard curves of quantified and diluted PCR products, as well as negative controls, were included in each PCR run. Primers were designed with Primer Express software (version 2.0; Applied Biosystems) as follows: for mBD3, forward (5-TCTGTTTGCATTTCTCCTGGTG-3) and reverse (5-TAAACTTCCAACAGCTGGAGTGG-3); and for mBD4, forward (5-TCTGTTTGCATTTCTCCTGGTG-3) and reverse (5-TTTGCTAAAAGCTGCAGGTGG-3). Results. According to the results of immunohistochemical analysis, noninfected control mice did not show mBD3 immunostaining, although occasional bronchial epithelial cells tested positive for mBD4. In contrast, lungs from Balb/c mice showed strong immunostaining 1 day after infection with a high dose of bacilli to produce progressive disease for both bdefensins in 80%–85% of bronchi and bronchioles, where a high percentage of epithelial cells showed positive immunostaining (70%–80% [SD, 10%]). This percentage was even higher at 3 days after infection (95% [SD, 3]), followed by a slight decrease at days 7 and 14 after infection (50%–60% [SD, 7%]). At day 21, when the maximum activity of pro698 • JID 2006:194 (1 September) • BRIEF REPORT

tective immunity was achieved in this model, 95% of bronchial epithelial cells showed mBD3 and mBD4 immunoreactivity, whereas, in bronchiolar ducts, 70% of epithelial cells were positive. The ciliated cells were the most frequently immunostained cell type (mean  SD, 87%  5% and 96%  3%, respectively), followed by goblet cells (mean  SD, 11%  3% and 3%  2%, respectively). Occasional pneumocytes and macrophages showed immunoreactivity (!2%). At day 28, when the progressive phase of the infection started, the bronchiolar epithelium maintained a high percentage of positive cells, whereas, in the bronchial epithelium, a 3-fold decrease was seen. At day 60 after infection, when the progressive phase was well established, as manifested by extensive pneumonia and high bacillary loads [10], the percentage of immunostained bronchial epithelial cells was !2%, whereas, in the bronchiolar epithelium, it was 25%  5% (figure 1). Compared with the progressive model, in which mBD production was strikingly reduced during the advanced phase of the disease, lungs from mice with latent infection showed a very high and stable percentage (80%–90%) of immunostained bronchial epithelial cells during the course of the whole experiment (7 months) (figure 1). Again, ciliated cells were the most frequently positive cell type (mBD3, 87%; mBD4, 78%), followed by goblet cells (mBD3, 8.5%; mBD4, 21%) and, occasionally, type II pneumocytes (3%) and macrophages (1%). Interestingly, latently infected mice in which disease was reactivated by treatment with corticosterone showed a striking disappearance of mBD3 and mBD4 immunostaining and developed extensive pneumonic areas and high mortality, denoting real reactivation of disease. The quantitative analysis of mBD gene expression correlated with immunohistochemical results. During the early phase of progressive disease (the first month), high mBD3 and mBD4 gene expression was found, whereas, during the progressive phase, a significant decrease of 1 log was seen. In contrast, constant high gene expression of both mBDs was observed during latent infection. The level of gene expression was 5-fold higher than that in progressive disease. This high expression of both mBDs in latently infected mice decreased strikingly after reactivation (figure 2). Discussion. Innate immunity has a wide repertoire of molecules that constitute the first line of defense against invading microorganisms [11]. Defensins are antimicrobial peptides that are considered to be a prototype family of innate immunity molecules widely distributed among various species, and they exhibit a broad spectrum of microbicidal activities affecting viruses, fungi, and bacteria [3]. Mammalian defensins are divided into 3 subfamilies: a-defensins, which are produced mainly by neutrophils; b-defensins, which are produced mainly by epithelial cells; and v-defensins, which were originally isolated from macaque leukocytes [3]. Defensins can kill microbes

Figure 1. Kinetics and representative micrographs of murine b-defensin–3 (mBD3)– and murine b-defensin–4 (mBD4)–immunostained bronchiolar epithelium during experimental pulmonary tuberculosis. A, Kinetics of mBD3 (black bars) and mBD4 (white bars) during progressive pulmonary tuberculosis. Data are the mean  SD of results from 6 mice at ⫻400 magnification. B, Bronchial epithelium from the lung of a noninfected control Balb/c mouse with no mBD3 immunostaining. C, Bronchial epithelial cells with positive mBD3 immunostaining at 21 days after infection in the Balb/c model of progressive disease. D, A striking decrease in mBD3 immunostaining in the bronchial epithelium of Balb/c mice 60 days after infection. E, Kinetics of mBD3 (black bars) and mBD4 (white bars) during experimental latent pulmonary tuberculosis in B6D2F1 mice. F, Bronchial epithelium from the lung of a noninfected control B6D2F1 mouse with no mBD3 immunostaining. G and H, Positive mBD3 immunostaining in the bronchial epithelium from latently infected B6D2F1 mice after 1 (G) and 7 (H) months. Staining was stronger at the later time point (magnification: noninfected control mice, ⫻200; infected mice, ⫻400).

Figure 2. Kinetics of murine b-defensin–3 (mBD3) (A and B) and murine b-defensin–4 (mBD4) (C and D) gene transcription determined by quantitative real-time polymerase chain reaction in the lungs of mice with progressive pulmonary tuberculosis (A and C) and those with latent infection and reactivation induced by corticosterone administration in drinking water (B and D). Three mice per time point from 2 experiments were analyzed. The no. of mRNA copies for each defensin was determined in relation to 1 ⫻ 106 copies of mRNA encoding the glyceraldehyde-3-phosphate dehydrogenase (G3PDH) gene.

directly, and, in the case of b-defensins, they may attract specialized immune cells, constituting a bridge between innate and adaptive immunity [11]. Scarce information is available on the role that defensins play in the pathogenesis of tuberculosis. A high concentration of adefensins has been found in pleural fluid, bronchoalveolar lavage samples, and plasma from patients with tuberculosis, and they have demonstrated some therapeutic benefit when administered during experimental tuberculosis [12, 13]. Regarding b-defensins, our results showed hBD2 production after infection of the lung epithelial A549 cell line with M. tuberculosis or incubation of the line with lipoarabinomannan, and our immunoelectronmicroscopic results showed hBD2 to be associated with disrupted intracellular bacilli, which suggests that hBD2 could contribute to intracellular mycobacterial lysis [8]. In the present study, we determined, to our knowledge for the first time, the kinetics of b-defensin gene expression and their cellular source in vivo, using well-characterized murine models of latent and progressive tuberculosis. We observed high production of mBDs by different cells during early stages of M. tuberculosis infection, predominantly bronchial and bronchiolar epithelial cells. We do not know the significance of this high mBD production at the beginning of 700 • JID 2006:194 (1 September) • BRIEF REPORT

the infection; however, considering that the usual human infecting dose is very low and that bacteria can be endocytosed by epithelial cells [14], in which they induce the production of b-defensins that efficiently kill bacteria, it is plausible to propose that b-defensins could participate in the early elimination of bacilli. This bactericidal activity might be more efficient in persons who apparently do not have infection and could also participate in the establishment of latent infection. Our latent infection model was characterized by low and stable bacillary counts, with few granulomas and without mortality [9]. In this model, as in others [15], there was a high expression of IFNg, TNF-a, and inducible NO synthase (iNOS). This immunological profile should contribute to the induction and maintenance of dormancy. Our results suggest that mBD3 and mBD4 could also be factors in the maintenance of latent infection. Pulmonary epithelial cells and macrophages are the first line of defense against inhaled pathogens, and interplay between them could be important during the early and long-term control of bacillary growth. After mycobacterial infection, airway epithelial cells rapidly produce mBD3 and mBD4, which probably contribute to the killing of mycobacteria. This strong induction of b-defensins could be induced by M. tuberculosis itself and also by TNF-a produced by macrophages. Indeed, TNF-

a is one of the most significant inducers of b-defensins [3]. It is also important to consider that both defensins elicit monocyte chemotaxis, and mBD3 possesses chemotactic activity for immature dendritic cells and memory T cells, which are major IFN-g–producing cells [11]. Thus, the high and stable production of mBD3 and mBD4 during latent infection could contribute to the long-term control of bacillary growth. This hypothesis is supported by our observation that disease reactivation provoked a striking suppression of mBD production. When Balb/c mice are infected intratracheally with a high dose of M. tuberculosis to induce progressive disease, an early phase is dominated by high production of Th1 cell cytokines, which, along with high levels of TNF-a and iNOS, temporarily control the infection. After the fourth week of infection, there is a decrease in levels of interleukin-2, TNF-a, and iNOS. Gradually, pneumonic areas prevail over granulomas. Pneumonia and a high burden of bacteria cause death [10]. Our results showed that, during the early phase of this model, which corresponded to the time of efficient control of bacillary growth, there was high production of both b-defensins. In contrast, during the progressive phase, there was a striking decrease in the production of both b-defensins, which again suggests that these antimicrobial peptides participate in the arrest of bacillary growth. The cellular localization of both mBDs in this model was very similar to that found in the latent infection model. In conclusion, mycobacterial infection induces the expression of mBD3 and mBD4 in mice, and both could participate in the control of mycobacterial growth. References 1. Parrish NM, Dick JD, Bishai WR. Mechanisms of latency in Mycobacterium tuberculosis. Trends Microbiol 1998; 6:107–12.

2. Tufariello JM, Chan J, Flynn JL. Latent tuberculosis: mechanism of the host and bacillus that contribute to persistent infection. Lancet Infect Dis 2003; 3:578–90. 3. Ganz T. Defensins: antimicrobial peptides of innate immunity. Immunology 2003; 3:710–20. 4. Schutte BC, Mitros JP, Bartlett JA, et al. Discovery of five conserved beta -defensin gene clusters using a computational search strategy. Proc Natl Acad Sci USA 2002; 99:2129–33. 5. Goldman MJ, Anderson GM, Stolzenberg ED, Kari UP, Zassloff M. Human b-defensin-1 is a salt-sensitive antibiotic in the lung that is inactivated in cystic fibrosis. Cell 1997; 88:553–60. 6. Harder J, Bartels J, Christophers E. Isolation and characterization of human b-defensin-3, a novel human inducible peptide antibiotic. J Biol Chem 2001; 276:5707–13. 7. Garcı´a J-R, Krause A, Schulz S, et al. Human b-defensin 4: a novel inducible peptide with a specific salt-sensitive spectrum of antimicrobial activity. FASEB J 2001; 10:1819–21. 8. Rivas-Santiago B, Schwander SK, Sarabia C, et al. Human b-defensin 2 is expressed and associated with Mycobacterium tuberculosis during infection of human alveolar epithelial cells. Infect Immun 2005; 73: 4505–11. 9. Arriaga AK, Orozco HE, Aguilar LD, Rook GAW, Hernandez-Pando R. Immunological and pathological comparative analysis between experimental latent tuberculous infection and progressive pulmonary tuberculosis. Clin Exp Immunol 2002; 128:229–37. 10. Herna´ndez Pando R, Orozco EH, Sampieri A, et al. Correlation between the kinetics of Th1/Th2 cells and pathology in a murine model of experimental pulmonary tuberculosis. Immunology 1996; 89:26–33. 11. Yang D, Chertov O, Bykovskaia SN, et al. b-defensins: linking innate and adaptive immunity through dendritic and T cell CCR6. Science 1999; 286:525–8. 12. Ashitani J, Mukae I, Hiratsuka T, Nakazato M, Kumamoto K, Matsakura S. Elevated levels of a-defensins in plasma and BAL fluid of patients with active pulmonary tuberculosis. Chest 2002; 121:519–26. 13. Sharma S, Verma I, Khuller GK. Therapeutic potential of human neutrophil peptide 1 against experimental tuberculosis. Antimicrob Agents Chemother 2001; 45:639–40. 14. Castro-Garza J, King CH, Swords WE, Quinn FD. Demonstration of spread by Mycobacterium tuberculosis bacilli in A549 epithelial cell monolayers. FEMS Microbiol Lett 2002; 212:145–9. 15. Flynn JA, Scanga Ch, Tanaka K, Chan J. Effects of aminoguanidine on latent murine tuberculosis. J Immunol 1998; 160:1796–803.

BRIEF REPORT • JID 2006:194 (1 September) • 701