Med Microbiol Immunol DOI 10.1007/s00430-015-0393-2

ORIGINAL INVESTIGATION

Increased viability but decreased culturability of Mycobacterium avium subsp. paratuberculosis in macrophages from inflammatory bowel disease patients under Infliximab treatment Nair Nazareth · Fernando Magro · Rui Appelberg · Jani Silva · Daniela Gracio · Rosa Coelho · José Miguel Cabral · Candida Abreu · Guilherme Macedo · Tim J. Bull · Amélia Sarmento 

Received: 24 September 2014 / Accepted: 10 February 2015 © Springer-Verlag Berlin Heidelberg 2015

Abstract  Mycobacterium avium subsp. paratuberculosis (MAP) has long been implicated as a triggering agent in Crohn’s disease (CD). In this study, we investigated the growth/persistence of both M. avium subsp. hominissuis (MAH) and MAP, in macrophages from healthy controls (HC), CD and ulcerative colitis patients. For viability assessment, both CFU counts and a pre16SrRNA RNA/ DNA ratio assay (for MAP) were used. Phagolysosome fusion was evaluated by immunofluorescence, through analysis of LAMP-1 colocalization with MAP. IBD macrophages were more permissive to MAP survival than HC macrophages (a finding not evident with MAH), but did not support MAP active growth. The lower MAP CFU counts in macrophage cultures associated with Infliximab treatment were not due to increased killing, but possibly

Nair Nazareth and Fernando Magro have contributed equally to this work. N. Nazareth · J. Silva · A. Sarmento (*)  FP‑ENAS (UFP Energy, Environment and Health Research Unit), CEBIMED (Biomedical Research Centre), University Fernando Pessoa, Rua Carlos da Maia, 296, 4200‑150 Porto, Portugal e-mail: [email protected] F. Magro · D. Gracio · J. M. Cabral  Institute of Pharmacology and Therapeutics, Faculdade de Medicina, Universidade do Porto, Alameda Professor Hernâni Monteiro, 4200‑319 Porto, Portugal F. Magro · R. Coelho · G. Macedo  Gastroenterology Department, Centro Hospitalar S. João, Alameda Professor Hernâni Monteiro, 4200‑319 Porto, Portugal F. Magro · D. Gracio · J. M. Cabral  MedInUP ‑ Center for Drug Discovery and Innovative Medicines, Universidade do Porto, Alameda Professor Hernâni Monteiro, 4200‑319 Porto, Portugal

to elevation in the proportion of intracellular dormant nonculturable MAP forms, as MAP showed higher viability in those macrophages. Increased MAP viability was not related to lack of phagolysosome maturation. The predominant induction of MAP dormant forms by Infliximab treatment may explain the lack of MAP reactivation during antiTNF therapy of CD but does not exclude the possibility of MAP recrudescence after termination of therapy. Keywords  Inflammatory bowel disease · Mycobacterium avium subsp. paratuberculosis · Macrophages · Phagosomal maturation

Introduction Mycobacterium avium complex (MAC) organisms are widely distributed in the environment and cause disease in R. Appelberg · A. Sarmento  Infection and Immunity Unit, IBMC – Instituto de Biologia Molecular e Celular, Universidade do Porto, Rua do Campo Alegre, 823, 4150‑180 Porto, Portugal R. Appelberg · C. Abreu  Department of Infectious Diseases, Centro Hospitalar S. João, Alameda Professor Hernâni Monteiro, 4200‑319 Porto, Portugal C. Abreu  Nephrology Research and Development Unit, Faculdade de Medicina, Universidade do Porto, Alameda Professor Hernâni Monteiro, 4200‑319 Porto, Portugal T. J. Bull  Infection and Immunity Research Institute, St George’s University of London, Cranmer Terrace, London SW17 0RE, UK

13



both animals and humans [1, 2]. MAC includes M. intracellulare (MI) and M. avium (MA), which includes four subspecies: M. avium subspecies avium (MAA), M. avium subspecies silvaticum (MAS), M. avium subspecies hominissuis (MAH) and M. avium subspecies paratuberculosis (MAP). The majority of MAC strains are opportunistic, infecting immunocompromised or debilitated hosts [3, 4]. MAH virulent isolates have the ability to grow well in human monocytes [5] and have been implicated in chronic pulmonary pathologies including disseminated disease in AIDS patients [6]. MAP is the etiological agent of Johne’s disease, a chronic intestinal granulomatous inflammation that affects mainly wild and domestic ruminants, but also a wide range of other animals, including primates [7–9]. Pathological similarities between Johne’s disease and Crohn’s disease (CD), a human chronic inflammatory bowel disease (IBD), have suggested that MAP could play a role in the etiology of CD [2]. Indeed, the presence of MAP DNA has been found to be more frequent in CD patients [10] when compared to normal patients [11] or those with other intestinal inflammatory diseases [12]. Even so, controversy still persists regarding a role of MAP in CD etiology. MAP is capable of penetrating the human intestinal wall either via M cells [13] or goblet cells [14], allowing access to the lamina propria where they are phagocytosed by macrophages and/or dendritic cells. Macrophages are key cells in the regulation of the intestinal innate immunity responses and also contribute to adaptive immunity through antigen presentation and cytokine production [15–17]. Although mycobacteria are able to infect other cell types, macrophages appear to be the primary niche for persistence and proliferation [18, 19]. MAP persistence in bovine macrophages seems to relate to the organism’s ability to suppress host cell apoptosis and inhibit phagolysosome fusion [20, 21]. Once established, MAP resides in cholesterol-rich immature vacuoles in human and murine monocytes, as well as in human neutrophils, and exhibits intracellular survival rates comparable to M. tuberculosis (MTB) [22, 23]. Indeed, MAP infection of murine and bovine macrophages have shown that MAP-containing phagosomes are poorly acidified with reduced colocalization with lysosome-associated membrane protein-1 (LAMP-1), a marker of mature endosomes and lysosomes, suggesting phagosomal maturation arrest [24–26]. Infection of activated human monocytes (THP-1 cell line) with a MAP strain of bovine origin resulted in decreased CFU counts over a 7-day period, but not complete elimination [27]. In these cells, a MAP-specific viability assay (pre16SrRNA gene RNA/DNA ratio) suggested that loss of MAP culturability was not completely a consequence of killing but included the presence and possible induction of non-culturable viable phenotypes.

13

Med Microbiol Immunol

The investigation of the IBD macrophage response to mycobacterial infection, particularly with MAP, is an essential step for establishing a link with CD causation. Until now, most human in vitro studies of MAP persistence have been performed in established cell lines. Little is known about mycobacterial persistence/growth in macrophages from IBD patients and how these compare with macrophages from healthy controls. If CD macrophages are defective in the handling of bacteria, as has been suggested [28, 29], it is possible they may be more permissive to MAP than macrophages from normal healthy patients. In this study, we describe the differential persistence/growth profile of both MAH and MAP in macrophages from healthy controls and IBD patients. We further demonstrate that Infliximab therapy (a monoclonal anti-TNF-α antibody), commonly used to induce CD remission [30], affects the MAP persistence/growth profile of macrophages from IBD patients, drives MAP latency and may also influence host phagosomal maturation processing.

Materials and methods Patients and samples This study included a total of 112 individuals including: 21 healthy controls (HC), 22 CD patients and 20 ulcerative colitis (UC) patients under Infliximab therapy (CD-IFX and UC-IFX, respectively), 31 CD patients and 18 UC patients not treated with Infliximab. Age and gender of the subjects enrolled in this study are represented in Table 1, and therapeutic regimens are included in Table 2. Informed consent was obtained in accordance with the institutional review board regulations at IBMC, University Fernando Pessoa and Hospital de São João. Patients were recruited from the Gastroenterology Department at Hospital de São João. Diagnosis was based on standard clinical, endoscopic, histologic and radiographic criteria [31, 32]. HC were recruited from the academic community of the Health Sciences Faculty, University Fernando Pessoa. Cell culture Whole blood (13.5 ml) collected from each individual was diluted 1:2 in phosphate-buffered saline (PBS) and layered onto Histopaque®-1077 (Sigma, St. Louis, MO, USA) in 15 ml sterile tubes. After centrifugation for 30 min at 400×g (room temperature), the mononuclear cells present in the monolayer were collected, washed 3× in Hank’s balanced salt solution (HBSS, Sigma) and resuspended in

Med Microbiol Immunol Table 1  Age and gender of CD patients enrolled in the studies HC

CD

CD-IFX

UC

UC-IFX

p valuea

n Male/female

21

31

22

18

20

–

9/12

15/16

13/9

9/9

13/7

ns

Mean age

34 (21–67)

43 (22–75)

37 (22–72)

47 (29–68)

45 (22–66)

0.0026

a

  ns differences not significant; p value for gender was calculated using Chi square test; p value for age was calculated using Student’s t test

Table 2  Therapeutic regimen of CD and CD-IFX patients enrolled in the studies CD

CD-IFX

UC

UC-IFX

5-ASA AZA CORT 5-ASA + AZA

21 5 0 0

0 0 0 0

9 3 0 0

0 0 0 0

IFX IFX + 5-ASA IFX + AZA

0 0 0

16 0 6

0 0 0

8 1 9

IFX + AZA + CORT

0

0

0

2

5-ASA, 5-aminosalicylate; AZA, azathioprine; PRED, prednisolone; AB, antibiotics (one of the following: ciprofloxacin, metronidazole or ceftriaxone); IFX, Infliximab

RPMI 1640 with GlutaMAX™ (Invitrogen, Carlsbad, CA, USA) supplemented with Antibiotic–Antimycotic 100× solution (Gibco, Life Technologies, MA, USA) (RPMI without FBS). Cells were cultured for 2 h in 10 cm-diameter tissue culture plates (Nunc, Thermo Scientific, MA, USA), and non-adherent cells were washed off with HBSS. Medium was replaced with 10 ml RPMI with GlutaMAX™ plus 10 % heat inactivated fetal bovine serum (FBS, Sigma) and supplemented with Antibiotic–Antimycotic 100× solution (Gibco) (complete RPMI). After 5-day incubation at 37 °C and 5 % CO2, adherent cells (monocyte-enriched population) were detached using 2.5 ml trypsin–EDTA solution (Sigma) and incubated for 9 min at 37 °C and 5 % CO2. After incubation, 10 ml of complete RPMI was added, plates were placed on ice and cells were detached by pipetting. Cell density was adjusted to 3 × 105 cells/ ml and seeded into a 24-well multidish culture plate, for colony counts and nucleic acid extraction (1 ml/well), or in chambered slides (µ-Slides 8 well, IBIDI, Martinsried, Germany) at a density of 300 µl/well for confocal microscopy analysis. Cultures were incubated at 37 °C and 5 % CO2, for an additional period of 2 days to complete monocyte differentiation into macrophages. After 7-day incubation, the cell monolayers obtained showed typical macrophage morphology. Medium was replaced with RPMI 1640 plus GlutaMAX™ and 10 % FBS (infection RPMI), and cultures were infected.

Infection and uptake of MAH and MAP by monocyte‑derived macrophages MAP (ATCC 43015, an isolate from a Crohn’s patient) was grown in Middlebrook 7H9 (BD Diagnostics, Sparks, MD, USA), with 10 % OADC supplement (BD Diagnostics), 0.05 % Tween 80 (Sigma) and 2 mg/l mycobactin J (Synbiotics Corporation, Lyon, France). MAH (MAC 101, ATCC 700898, an isolate from an AIDS patient) was grown in Middlebrook 7H9 with 10 % ADC supplement (BD Diagnostics) and 0.05 % Tween 80. Cultures were incubated for 2 weeks at 37 °C with shaking. Inocula were prepared by centrifugation at 13,000 rpm for 15 min, followed by 2 washes in saline with 0.05 % Tween 80 and a final resuspension in the same solution at a concentration corresponding to an OD (540 nm) of 1 (approximately 1 × 108 bacteria/ml). Inocula were aliquoted and stored frozen at −70 °C. For clump removal prior to infection, both inocula were briefly sonicated (GM-2200, Bandelin Sonorex, Berlin, Germany) and MAP inoculum was further treated by 10 passages through a 24G needle. For studies involving MAP uptake and LAMP-1 colocalization, MAP was stained before infection with Syto BC green fluorescent nucleic acid stain (Molecular probes, Life Technologies). Briefly, an aliquot of MAP inoculum (0.2 ml) was centrifuged at 10,000×g for 5 min, resuspended in 50 nM Syto BC and incubated for 15 min at room temperature. After incubation, bacteria were washed in PBS and resuspended in 0.2 ml of infection RPMI. MAP staining was confirmed by fluorescence microscopy. Macrophage cultures prepared as above were infected with MAP or MAH at a MOI (multiplicity of infection) of 10 bacteria: 1 cell. We used different staining methods for uptake assessment for MAH and MAP, since MAP cell wall loss inside human macrophages may occur [33– 36], making it unsuitable to use in ZN-dependent assays. For MAH uptake evaluation, 3 h post infection MAHinfected macrophages were stained by a conventional Ziehl–Neelsen method (ZN). Briefly, the glass coverslips, facing upwards, were covered with carbol-fuchsin solution (Sigma) and heated until vapors started rising. After 15 min, the coverslips were washed with distilled water and destained with 3 % hydrochloric acid in isopropyl alcohol for 2–3 min. Coverslips were again washed in distilled

13



water, counterstained for 1 min with methylene blue, rinsed in distilled water, air-dried and examined. For evaluation of MAP uptake and colocalization with LAMP-1, macrophages in chambered slides were infected with Syto BClabeled MAP. After 3 h (MAP uptake) or 24 h (MAP colocalization with LAMP-1), monolayers were washed, fixed with 4 % paraformaldehyde and processed for immunofluorescence analysis as described below. Mycobacterial uptake was determined by counting the number of ZN-positive (MAH) or DNA-stained (MAP) bacteria in a minimum of 100 cells, using a Nikon Eclipse 50i microscope equipped with a digital camera (Nikon) at 1000× magnification under oil immersion (for MAH enumeration) or a Leica Microsystems TCS SP2 AOBS confocal microscope (for MAP enumeration). Mycobacterial growth determined by CFU counts The number of viable mycobacterial cells was determined at 3 h (T3h) and 7 days (T7d) after infection, by culture on solid media with subsequent counting of colony-forming units (CFU). Briefly, the cell monolayers were lysed with 10 % saponin, and supernatants were plated on Middlebrook 7H9 with 10 % OADC supplement, 0.05 % Tween 80, 2 mg/l mycobactin J and 1.5 % agar (MAP) or Middlebrook 7H10 with 10 % ADC supplement and 0.05 % Tween 80 (MAH). Colony counts were determined after incubation of plates at 37 °C until colonies appeared, usually after 2 weeks (MAH) or 5 weeks (MAP). Viability assay using pre16SrRNA DNA and pre16SrRNA RNA The existence of non-culturable phenotypes of MAP makes culturability not enough to measure viability. As such, MAP viability was additionally determined using a previously tested ribosomal turnover assay [27]. This assay is based on the determination of RNA/DNA copy ratio from a single genome copy gene pre-16SRNA region, which is indicative of loss (ratio