- Email: [email protected]
Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Bladder Cancer Editorial by XXX on pp. x–y of this issue
Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo Estela Noguera-Ortega a,y, Silvia Secanella-Fandos a,y, Hasier Eran˜a a,z, Jofre Gasio´n a, Rosa M. Rabanal b, Marina Luquin a, Eduard Torrents c, Esther Julia´n a,* a
Departament de Gene`tica i de Microbiologia, Facultat de Biocie`ncies, Universitat Auto`noma de Barcelona, Bellaterra, Spain; b Unitat de Patologia Murina i Comparada,, Departament de Medicina Animal i Cirurgia, Facultat de Veterina`ria, Universitat Auto`noma de Barcelona, Bellaterra, Spain; c Bacterial Infections and Antimicrobial Therapy Group, Institute for Bioengineering of Catalonia, Barcelona, Spain
Article info
Abstract
Article history: Accepted March 27, 2015
Background: Bacillus Calmette-Gue´rin (BCG) prevents tumour recurrence and progression in non–muscle-invasive bladder cancer (BC). However, common adverse events occur, including BCG infections. Objective: To find a mycobacterium with similar or superior antitumour activity to BCG but with greater safety. Design: In vitro, ex vivo, and in vivo comparisons of the antitumour efficacy of nonpathogenic mycobacteria and BCG. Intervention: The in vitro antitumour activity of a broad set of mycobacteria was studied in seven different BC cell lines. The most efficacious was selected and its ex vivo capacity to activate immune cells and its in vivo antitumour activity in an orthotopic murine model of BC were investigated. Outcome measurements and statistical analysis: Growth inhibition of BC cells was the primary outcome measurement. Parametric and nonparametric tests were use to analyse the in vitro results, and a Kaplan-Meier test was applied to measure survival in mycobacteria-treated tumour-bearing mice. Results and limitations: Mycobacterium brumae is superior to BCG in inhibiting lowgrade BC cell growth, and has similar effects to BCG against high-grade cells. M. brumae triggers an indirect antitumour response by activating macrophages and the cytotoxic activity of peripheral blood cells against BC cells. Although no significant differences were observed between BCG and M. brumae treatments in mice, M. brumae treatment prolonged survival in comparison to BCG treatment in tumour-bearing mice. In contrast to BCG, M. brumae does not persist intracellularly or in tumour-bearing mice, so the risk of infection is lower. Conclusions: Our preclinical data suggest that M. brumae represents a safe and efficacious candidate as a therapeutic agent for non–muscle-invasive BC. Patient summary: We investigated the antitumour activity of nonpathogenic mycobacteria in in vitro and in vivo models of non–muscle-invasive bladder cancer. We found that Mycobacterium brumae effectively inhibits bladder cancer growth and helps the host immune system to eradicate cancer cells, and is a promising agent for antitumour immunotherapy. # 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Associate Editor: Shahrokh Shariat Keywords: Animal models Bacillus Calmette-Gue´rin Cytokines Immunomodulation Immunotherapy Mycobacteria Urothelial cell line
y
These authors contributed equally to this work. Current address: CIC bioGUNE, Derio, Vizcaya, Spain. * Corresponding author. Departament de Gene`tica i de Microbiologia, Facultat de Biocie`ncies, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain. Tel. +34 93 5813096; Fax: +34 93 5812387. E-mail address: [email protected] (E. Julia´n). z
http://dx.doi.org/10.1016/j.euf.2015.03.003 2405-4569/# 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 2
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
1.
Introduction
Despite the well-known benefits of bacillus CalmetteGue´rin (BCG) therapy for non–muscle-invasive bladder cancer (NMIBC) [1], it is not a perfect treatment. BCG, an attenuated mycobacterium obtained from the pathogenic Mycobacterium bovis, exerts both direct and indirect antitumour activities involving inhibition of cellular proliferation and stimulation of a robust immune response, respectively [1,2]. However, nonsevere toxicity is frequently observed [1], and severe adverse events, including BCG infections [3–5], limit the clinical applicability of this immunotherapy. Many studies have attempted to find agents as efficacious as BCG but safer. Killed BCG and BCG cell extracts are less efficacious than live BCG [2]. Moreover,
for the first instillations in a treatment regimen, live bacteria are needed to obtain a favourable immune response [6,7]. Other weaknesses regarding BCG functionality also have to be considered. First, different BCG strains can differ in effectiveness [8,9]. Second, BCG supply shortages worldwide have highlighted its costly production, mainly because of its slow growth. An attractive approach to overcome adverse BCG effects is the use of nonpathogenic mycobacteria. Most mycobacteria species (Fig. 1) are nonpathogenic and potentially share immunomodulatory antigens with BCG. Mycobacterium smegmatis and Mycobacterium phlei were initially proposed as alternatives to BCG for NMIBC treatment [10]. Later, researchers investigated Mycobacterium vaccae and Mycobacterium indicus pranii,
Fig. 1 – Genealogical tree of mycobacterial species selected for the study. Maximum likelihood phylogenetic tree of type strains from the Mycobacterium genus based on 16S rRNA gene sequences, indicating slow and rapid Mycobacterium growers. Arrows indicate the species selected for the study.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
initially used as immunotherapeutic agents for tuberculosis and leprosy, for treatment of other types of cancer [11]. All of these studies used cell extracts or heat-killed mycobacteria, probably because M. phlei, M. smegmatis, and M. vaccae infections have been described [12–14]. Therefore, the same limitations observed for heat-killed BCG also affect these other mycobacteria. Only recombinant M. smegmatis expressing human tumour necrosis factor (TNF)-a has been investigated as a live mycobacterium, and it is vastly superior to BCG, as demonstrated in in vitro assays and in vivo in mouse models [15]. However, genetically modified M. smegmatis and BCG have not been studied in humans. This previous research indicates that nonpathogenic mycobacterial species could have antitumour activity that is similar or superior to that of BCG. In this study we focused on nonpathogenic, rapidly growing species because of two advantages over BCG: a lower risk of infection and lower production costs. After selecting the mycobacterial species with the highest antitumour activity in vitro, we investigated its ex vivo capacity to activate immune cells and its in vivo antitumour activity.
2.5.
3
Cytotoxic activity
Experiments on inhibition of cell growth were performed using human peripheral blood mononuclear cells (PBMCs) as previously described [17]. The optimal MOI and time for cytokine production were first determined on a small set of PBMC samples. Sample collection was approved by the Ethics Committee of Universitat Auto`noma de Barcelona (UAB).
2.6.
Cytokine analysis
To quantify cytokines, enzyme-linked immunosorbent assays were performed on cell culture supernatants using commercially available kits (Supplementary Table 2). Nitric oxide (NO) production was assessed by measuring nitrite concentrations in the Griess reaction (Sigma– Aldrich, Madrid, Spain).
2.7.
Mycobacterial viability
To determine the intracellular viability of mycobacteria, T24 cells and J774 macrophages were infected and collected at different time points after infection as previously described [9].
2.8.
Orthotopic model of BC and intravesical treatment
Animal experiments were performed according to procedures approved
2.
Materials and methods
2.1.
Sequence alignment and phylogenetic inferences
by the Animal Care Committee of UAB. The murine model was developed as previously described [6,8,19]. In brief, C57Bl/6 female mice (7–9 wk old; Charles River Laboratories, Barcelona, Spain) were anaesthetised with isoflurane, and chemical lesions were induced on the bladder urothelium
16S rRNA sequences (1200 nucleotides) were obtained from the Ribosomal Database Project (release 11; http://rdp.cme.msu.edu/index. jsp) [16]. A set of 143 good-quality 16S rRNA sequences (only Mycobacterium genus-type strains) was downloaded and used for sequence alignment (Slow accurate alignment algorithm) and phylogenetic analysis (Neighbour-joining algorithm) using CLC Main Workbench v. 6.9.1 (http://www.clcbio.com/products/clc-main-workbench/). Tree topology was re-examined using the bootstrap (1000) method of resampling.
by instilling 50 ml of L-poly-lysine intravesically through a 24-gauge catheter (dwell time 10 min). Subsequently, 105 MB49 bladder tumour cells were instilled to induce tumours (dwell time 60 min). At 1 d after tumour implantation, the mice were randomly divided into three groups (7–14 animals/group) and administered mycobacterial cells (dwell time 60 min) or PBS. The animals were treated weekly for 4 wk (Fig. 5B) and evaluated on a daily basis in terms of health status and the presence of gross haematuria. After 60 d, surviving animals were sacrificed and bladders were fixed in buffered formalin (10%) and embedded in paraffin. Haematoxylin-eosin staining was performed on 4-mm sections for
2.2.
Bacterial strains and cell lines
histopathologic examination. Spleens were collected, disrupted, serially diluted, and plated to determine colony-forming units.
All mycobacterial strains, cell lines, and growth conditions are described in the supplementary material.
2.9.
2.3.
Student’s t-test (Prism software, GraphPad, San Diego, CA, USA) was used
Direct growth inhibition experiments
Statistical analysis
to assess the statistical significance of differences among groups for Tumour cells were infected with mycobacteria at different multiplicity
direct inhibition of BC cell growth, cytokine levels produced by BC cells,
of infection (MOI) levels, and cell proliferation was analysed as
and mycobacterial viability in infected cultures. The Wilcoxon-Mann-
previously described [9,17].
Whitney rank test was used to compare cytokine levels and percentage
2.4.
the statistical significance of Kaplan-Meier survival curves.
cytotoxicity (Prism software). Log-rank tests were applied to determine
Macrophage infection and surface receptor analysis
Murine J774 macrophages were seeded into six-well cell culture plates.
3.
Results
3.1.
M. brumae has a better antitumour effect than BCG against
After 24 h, lipopolysaccharide or mycobacterial suspension was added. After 3 h, the cells were washed and complete medium was added. At 24 h after infection, cell culture supernatants were harvested, and the cells were collected and centrifuged [18]. The pellets were resuspended in phosphate-buffered saline (PBS) with antimouse CD32/CD16 (Mouse FcBlock, Becton-Dickinson, Pharmingen, San Jose, CA, USA) and incubated on ice for 15 min. Then phycoerythrin-conjugated anti-CD80, anti-CD86, and anti-CD40 (BD Pharmingen) antibodies were added. Isotype control antibodies were also included. The cells were then analysed on a FacsCalibur flow cytometer using CellQuest software (both from BD Biosciences, San Jose, CA, USA).
low-grade tumour cells
We assessed the ability of eight different mycobacteria to inhibit tumour cell growth. The antiproliferative activity of mycobacteria depended on both time and MOI (Supplementary Fig. 1). Comparison of tumour growth inhibition revealed that not all mycobacteria had antitumour effects (Fig. 2). BCG, M. brumae, M. phlei, and M. vaccae smooth
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 4
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 2 – Mycobacterium brumae had the greatest inhibitory effect on the growth of low-grade tumour cells among all the mycobacteria studied. (A) Growth inhibition and cytokine production for high-grade (T24 and J82) and low-grade (RT4) bladder cancer cells infected with mycobacteria (multiplicity of infection [MOI] 12.5:1) at 72 h after infection for T24 and RT4 cells, and 120 h for J82 cells. Data denote percentage cell survival relative to untreated cells (control). Values represent the mean W standard deviation (SD) for triplicate preparations. Data are representative of one of at least three independent experiments. (B) Growth inhibition of T24 and low-grade (SW780, 5637, RT112, and MG-HU-3) bladder tumour cells infected with mycobacteria (MOI 12.5:1) at 72 h after infection. Data represent percentage cell survival relative to untreated cells. Each column represents the average W SD for triplicate cultures from at least three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05 versus BCG-infected cells, ** p < 0.05 for infected versus noninfected cells, Student’s t-test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 3 – Induction of an immune response and antitumor activity by macrophages and peripheral blood mononuclear cells (PBMCs) infected with Mycobacterium brumae. (A) Induction of activation markers on mycobacteria-infected macrophages. J774 murine macrophages were not stimulated (NS), treated with lipopolysaccharide (LPS; 10 mg/ml), or infected with mycobacteria (multiplicity of infection [MOI] 1:1 for BCG and 10:1 for the other mycobacteria) for 24 h. The cells were then stained with anti-CD40, anti-CD80, or anti-CD86 antibodies. Results are expressed as the mean fluorescence intensity (Geo MFI from 10 000 events for each histogram) W standard deviation (SD) for three independent experiments. (B) Production of cytokines (interleukin [IL]-6, IL-12) and nitric oxide (NO) in uninfected and infected macrophages was measured 48 h after infection. (C) Cytokine production by mycobacteria-infected PBMCs. PBMC were infected with BCG (MOI 0.1:1) or M. brumae (MOI 1:1) for 7 days, and cytokine production was measured. ND = not determined (data were below the minimum detection value of the test). (d). Cytotoxicity of BCG-activated (blue column) and M. brumae-activated (red column) PBMCs, or soluble factors (cell culture supernatants) from BCG-activated (blue) and M. brumae-activated (red) PBMCs against T24 cells after 48 h. PBMCs alone or the corresponding cell culture supernatant against T24 cells (black columns) and T24 alone (white columns) were used as controls. In (B–D), values represent the mean W SD for triplicate cell cultures from three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05 versus BCG treatment, ** p < 0.05 versus nontreated cells, Mann-Whitney test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
5
EUF-20; No. of Pages 10 6
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
variant (S) exhibited similar antitumour activity in highgrade T24 and J82 cell lines (Fig. 2). In low-grade RT4 cells, M. brumae and M. phlei inhibited cell proliferation to a greater extent than BCG and M. vaccae (S). Infection with BCG and M. brumae induced the highest production of cytokines, apart from in RT4 cells (Fig. 2A). Given that M. brumae and M. phlei were more efficacious than BCG in inhibiting cell growth in low-grade RT4 cells, we evaluated the antitumour activity of a number of other mycobacterial species in four additional low-grade cell lines and the high-grade T24 cell line. In T24 cells, all mycobacteria showed lower antitumour activity than M. brumae, BCG, or M. phlei (Fig. 2B). In low-grade cell lines, M. brumae significantly inhibited cell growth to a greater extent than the other mycobacteria (Fig. 2B). As Mycobacterium gastri had the lowest inhibition and cytokine induction rates, it was used as a negative control in subsequent experiments. We did not observe growth inhibition of the murine normal urothelium G/G cell line after infection with BCG or M. brumae (data not shown). 3.2.
(Fig. 5A), we performed in vivo experiments (Fig. 5B). Approximately 9 2 d after tumour induction, all animals exhibited haematuria, a hallmark of established tumours. While 100% of nontreated animals died within 40 d of tumour implantation, mycobacteria treatment significantly prolonged survival in tumour-bearing mice. As for BCG [20,21], the response to M. brumae in the animal model is dose-dependent (Supplementary Fig. 2). At 60 d after tumour implantation, no significant differences were observed between the BCG and M. brumae treatment groups, although M. brumae treatment prolonged survival compared to BCG treatment (Fig. 5C).
M. brumae activates immune cells and triggers cytotoxic
activity against tumour cells
To evaluate the immunomodulatory potential of M. brumae, we evaluated its ability to activate macrophages and PBMCs. M. brumae triggered the expression of activation markers and the production of cytokines and NO in macrophages (Fig. 3A,B). Notably, levels of CD40 expression significantly increased after infection with M. brumae or M. vaccae in comparison to BCG and M. phlei (Fig. 3A). In PBMCs, M. brumae triggered the production of interleukin (IL)-8, IL-10, IL-12, and TNF-a to levels similar to those induced by BCG infection (Fig. 3C). It was not possible to measure the response to M. phlei because this mycobacterium formed a biofilm on the surface of PBMC cultures 48 h after infection. We then analysed the ex vivo antitumour activity of mycobacteria-activated PBMCs. Both BCG- and M. brumae-infected PBMCs triggered direct cytotoxic activity against T24 cells (Fig. 3D). In addition, soluble factors derived from both mycobacteria-infected cultures showed cytotoxic effects against T24 cells. Coculture with mycobacteria-activated PBMCs or nonactivated PBMCs significantly increased constitutive IL-8 and IL-6 production in T24 cells (Fig. 3D). 3.3.
M. brumae does not persist in T24 tumour cells or
J774 macrophages
We corroborated the nonpathogenicity of M. brumae by investigating its survival in nonprofessional and professional phagocytic cells. Unlike BCG, M. brumae and M. vaccae S were not viable in tumour cells or macrophages at 24 and 96 h after infection (Fig. 4A,B). 3.4.
Intravesical treatment with M. brumae enhances the
survival of tumour-bearing mice
After we demonstrated that M. brumae inhibits the growth of murine MB49 tumour cells in vitro, similar to BCG
Fig. 4 – Mycobacterial survival in infected cell cultures. (A) T24 tumour cells and (B) J774 macrophages were infected at a multiplicity of infection of 10:1 for 3 h. After washing, samples were plated on Middlebrook agar and viable intracellular mycobacteria were counted. Results are expressed as mean of colony-forming units (CFU)/well W standard deviation for triplicate wells from three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05, ** p < 0.01 versus BCG infection, Student’s t-test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 5 – Mycobacterium brumae treatment prolongs survival in tumour-bearing mice. (A) Growth inhibition and interleukin (IL)-6 and keratinocyte chemoattractant (KC) production for high-grade murine MB49 cells infected with mycobacteria (multiplicity of infection 10:1). Data represent percentage cell survival relative to untreated cells. Values represent the mean W standard deviation for triplicate preparations from two independent experiments. ** p < 0.05 versus control group, * p < 0.05 versus BCG group, Student’s t-test. (B) Schematic schedule of the in vivo experiments. C57BL/6 mice (n = 7–14) received four weekly intravesical instillations of phosphate-buffered saline (PBS), BCG, or M. brumae (indicated by arrows) after tumour implantation. (C) Kaplan-Meier analysis of survival in tumour-bearing mice after treatment with PBS (black line), 2–4 T 106 BCG (blue), or 2 T 107 M. brumae (red). ** p < 0.01 versus PBS group, log-rank (Mantel-Cox) test. (D) Colony-forming units (CFU) for splenocyte cultures from mice treated with M. brumae (red) or BCG (blue). The number of viable bacteria from each spleen is represented as a dot; the line identifies the median for the group. (E) Representative histologic images (hematoxylin-eosin staining) of bladder sections from nontreated (PBS), M. brumae-treated, and BCGtreated tumour-bearing mice. Scale bar, 1 mm (upper images) and 50 or 100 mm (lower images).
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
7
EUF-20; No. of Pages 10 8
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Examination of splenocyte cultures revealed that BCG-treated mice had viable bacteria counts, while no mycobacteria were detected in surviving M. brumae-treated mice (Fig. 5D). Histopathologic examination of a nontreated bladder showed a solid mass growing within the bladder lumen and infiltrating the mucosa. Tumours were homogeneous in pattern, with several haemorrhagic and necrotic foci. In addition, mitotic cells were observed at higher magnification (Fig. 5E). Mycobacteria-treated mice had chronic inflammation, with moderate to intensive lymphocytic infiltration into the lamina propia and activation of lymphoid follicles. No granulomatous inflammation was observed (Fig. 5E). 4.
Discussion
Among all the mycobacteria studied, M. brumae had the strongest inhibitory activity on growth. M. brumae inhibited high-grade tumour growth to a similar extent to BCG and M. phlei, but was superior to the other mycobacteria in inhibiting low-grade tumour cells. The antiproliferative properties are not related to species origin because the most efficacious bacteria, M. brumae, BCG, M. phlei, and M. vaccae, are not phylogenetically related (Fig. 1), whereas species close to M. brumae (M. fallax and M. chitae) were not as efficacious as M. brumae. Our results indicate that all mycobacteria enter T24 cells, as previously shown for BCG and M. smegmatis [22], but the outcome after internalisation differs among mycobacterial species: some remain alive, while others are destroyed (Fig. 4). We corroborated these results for BCG and M. brumae using microscopy techniques (data not shown). Therefore, the antigens exposed and their role could differ among different mycobacteria. Further studies are needed to unravel the M. brumae antigen(s) related to BC inhibitory activity. After our initial encouraging results for M. brumae, we studied its capacity to activate an immune response. In vitro and in vivo studies have demonstrated that an effective immunotherapeutic agent should activate antigen-presenting cells and strongly induce Th1 cells [2,23–25]. M. brumae elevated the expression of activation surface markers. Interestingly, M. brumae triggered higher expression of CD40 than BCG did. Activation of CD40 upregulates CD80 and CD86, which improve the pattern of macrophages with the M1 phenotype, which is necessary for activating effector T cells and effective antitumour therapy [23]. M1 macrophages participate in the in vitro growth inhibition of T24 cells, suggesting that the same behaviour could occur in vivo after mycobacterial treatment [26]. In addition to activating macrophages, M. brumae induces cytokine secretion and activates the antitumour capacity of infected PBMCs, although to a lesser extent than BCG. In vitro and in vivo studies demonstrated the influence of cytokines and the host immune response on the outcome of BCG immunotherapy in BC [1,2]. Several BCG subcomponents are potent inducers of IL-12 and interferon (IFN)-g production by PBMCs, promoting Th1 cell differentiation and enhancing cytotoxicity against BC cells [27,28]. IL-12 and IFN-g are critical for effective immunotherapy in BC mouse models [19,24]. Thus, our results suggest that
M. brumae shares some components with BCG that are critical for its immunomodulatory activity. Finally, we demonstrated that M. brumae treatment significantly prolongs survival in tumour-bearing mice. Importantly, BCG was recovered from splenocytes of BCGtreated mice, whereas M. brumae was not recovered from the spleen of surviving M. brumae-treated animals. This is in accordance with our in vitro results, which demonstrated that M. brumae is cleared within a few days after infection of BC cells and macrophages, whereas BCG persists (Fig. 4). Although the limitations of the mouse model must be taken into account, this result is indicative of the BCG infections observed in NMIBC patients after treatment. Accidents after traumatic instillations in humans can lead to systemic BCG invasion and possible infection. Numerous reports have documented the onset of BCG infection. Bowyer et al [29] detected acid-fast bacilli in a bladder 16 mo after the completion of intravesical instillations. Systemic infections have even been observed 3 y [4] and 6 y [3] after completion of BCG treatment due to reactivation of the bacilli when a patient’s immunocompetence is decreased. The fact that viable M. brumae was not recovered in an animal model implies that use of M. brumae for BC treatment could avoid this adverse event associated with BCG. M. brumae is a saprophyte mycobacterium associated with water and soil. Although a case of catheterrelated infection was attributed to M. brumae, it was subsequently confirmed that the isolate did not correspond to M. brumae [30]. Therefore, infections due to M. brumae have not been described. Our study has some limitations. The molecular mechanisms by which M. brumae inhibits tumour cell growth and triggers an immune response have not been addressed. Although this was not the primary aim of our study, we are currently performing in vitro and in vivo experiments to elucidate the mechanisms. Moreover, although the animal model we used is currently the most appropriate and widely accepted for comparing potential agents with BCG treatment [6,19], other animal models (eg, using rats and/ or pigs) could be used to corroborate our results before clinical assays of the safety and tolerability of M. brumae in humans. 5.
Conclusions
In summary, our data indicate that M. brumae effectively inhibits BC growth and helps the host immune system to eradicate cancer cells. Moreover, M. brumae grows three times faster than BCG and is therefore faster and cheaper to produce than BCG. Although further research is needed, our preclinical results suggest that M. brumae represents a safe and efficacious agent for antitumour immunotherapy and may be a promising option for NMIBC treatment. Author contributions: Esther Julia´n had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Julia´n.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
˜ a, Gasio´n, Acquisition of data: Noguera-Ortega, Secanella-Fandos, Eran Rabanal, Torrents, Julia´n. Analysis and interpretation of data: Noguera-Ortega, Secanella-Fandos, ˜ a, Gasio´n, Rabanal. Luquin, Torrents, Julia´n. Eran Drafting of the manuscript: Noguera-Ortega, Torrents, Julia´n. Critical revision of the manuscript for important intellectual content: Noguera-Ortega, Rabanal, Luquin, Torrents, Julia´n. Statistical analysis: Noguera-Ortega, Secanella-Fandos, Julia´n. Obtaining funding: Julia´n. Administrative, technical, or material support: Luquin, Torrents, Julia´n. Supervision: Julia´n. Other: None. Financial disclosures: Esther Julia´n certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Silvia Secanella-Fandos, Marina Luquin, and Esther Julia´n are the inventors of patent PCT/ ES2013/070547. The remaining authors have nothing to disclose.
9
bacillus Calmette-Guerin determine the succeeding Th1 response. J Urol 2003;170:2004–8. [8] Rentsch CA, Birkha¨user FD, Biot C, et al. Bacillus Calmette-Gue´rin strain differences have an impact on clinical outcome in bladder cancer immunotherapy. Eur Urol 2014;66:677–88. [9] Secanella-Fandos S, Luquin M, Julia´n E. Connaught and Russian strains showed the highest direct antitumor effects of different bacillus Calmette-Gue´rin substrains. J Urol 2013;189:711–8. [10] Yarkoni E, Rapp HJ. Immunotherapy of experimental cancer by intralesional injection of emulsified nonliving mycobacteria: comparison of Mycobacterium bovis (BCG), Mycobacterium phlei, and Mycobacterium smegmatis. Infect Immun 1980;28:887–92. [11] Grange JM, Bottasso O, Stanford CA, Stanford JL. The use of mycobacterial adjuvant-based agents for immunotherapy of cancer. Vaccine 2008;26:4984–90. [12] Paul E, Devarajan P. Mycobacterium phlei peritonitis: a rare complication of chronic peritoneal dialysis. Pediatr Nephrol 1998;12: 67–8. [13] Hachem R, Raad I, Rolston KV, et al. Cutaneous and pulmonary infections caused by Mycobacterium vaccae. Clin Infect Dis 1996; 23:173–5.
Funding/support and role of the sponsor: This work was funded by the
[14] Khatter S, Singh UB, Arora J, Rana T, Seth P. Mycobacterial infections
Spanish Science and Education Ministry (SAF2006-05868), the Spanish
in human immuno-deficiency virus seropositive patients: role of
Ministry of Science and Innovation (Instituto de Salud Carlos III – PI10/
non-tuberculous mycobacteria. Indian J Tuberc 2008;55:28–33.
01438), the European Regional Development Fund (FEDER), and the
[15] Young SL, Murphy M, Zhu XW, et al. Cytokine-modified Mycobacte-
Generalitat of Catalunya (2009SGR-108). Silvia Secanella-Fandos was supported by a Ph.D. fellowship from Universitat Auto`noma de
rium smegmatis as a novel anticancer immunotherapy. Int J Cancer
Barcelona. These agencies had no role in study design, data collection
[16] Cole JR, Wang Q, Fish JA, et al. Ribosomal database project: data and
and analysis, decision to publish, or preparation of the manuscript.
2004;112:653–60. tools for high throughput rRNA analysis. Nucleic Acids Res 2014; 42:D633–42.
Acknowledgments: We thank all the members of the Animal Facility and
[17] Secanella-Fandos S, Noguera-Ortega E, Olivares F, Luquin M, Julia´n
Manuela Costa from the Cytometry Unit at UAB for their support and advice. We also thank Dr. Francesc Padro´s, Dra. Mercedes Ma´rquez, and
E. Killed but metabolically active Mycobacterium bovis bacillus Calmette-Gue´rin retains the antitumor ability of live bacillus Calmette-Gue´rin. J Urol 2014;191:1422–8.
Ester Blanco for histopathologic examination of bladders.
[18] Kan-Sutton C, Jagannath C, Hunter RL. Trehalose 6,60 -dimycolate on
Appendix A. Supplementary data
the surface of Mycobacterium tuberculosis modulates surface marker expression for antigen presentation and costimulation in
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. euf.2015.03.003.
murine macrophages. Microbes Infect 2009;11:40–8. [19] Zaharoff D, Hoffman BS, Hooper HB, et al. Intravesical immunotherapy of superficial bladder cancer with chitosan/interleukin-12. Cancer Res 2009;69:6192–9. [20] Matsushima M, Horinaga M, Fukuyama R, et al. Enhanced antitu-
References [1] Gontero P, Bohle A, Malmstrom P-U, et al. The role of bacillus Calmette-Gue´rin in the treatment of non-muscle-invasive bladder cancer. Eur Urol 2010;57:410–29. [2] Alexandroff AB, Nicholson S, Patel PM, Jackson AM. Recent advances in bacillus Calmette-Guerin immunotherapy in bladder cancer.
mor effect of combination intravesical mitomycin C and bacillus Calmette-Guerin therapy in an orthotopic bladder cancer model. Oncol Lett 2011;2:13–9. [21] Xiao Z, Hanel E, Mak A, Moore RB. Antitumor efficacy of intravesical BCG, gemcitabine, interferon-a and interleukin-2 as mono- or combination-therapy for bladder Cancer in an orthotopic tumor model. Clin Med Insights Oncol 2011;5:315–23.
Immunotherapy 2010;2:551–60. [3] Alvarez-Mu´gica M, Ferna´ndez J, Bulnes V, et al. Fatal outcome due to
[22] Redelman-Sidi G, Iyer G, Solit DB, Glickman MS. Oncogenic activa-
sepsis by Mycobacterium bovis six years after BCG intravesical
BCG entry into bladder cancer cells. Cancer Res 2013;73:1156–67.
instillations. J Med Cases 2010;1:81–3. [4] Gonzalez OY, Musher DM, Brar I, et al. Spectrum of bacille CalmetteGue´rin (BCG) infection after intravesical BCG immunotherapy. Clin Infect Dis 2003;36:140–8.
tion of Pak1-dependent pathway of macropinocytosis determines [23] Kimura T, Ohashi T, Kikuchi T, Kiyota H, Eto Y, Ohishi Y. Antitumor immunity against bladder cancer induced by ex vivo expression of CD40 ligand gene using retrovirus vector. Cancer Gene Ther 2003;10: 833–9.
[5] Nadasy KA, Patel RS, Emmett M, et al. Four cases of disseminated
[24] O’Donnell MA, Luo Y, Chen X, Szilvasi A, Hunter SE, Clinton SK.
Mycobacterium bovis infection following intravesical BCG instilla-
Role of IL-12 in the induction and potentiation of IFN-gamma in response to bacillus Calmette-Gue´rin. J Immunol 1999;163:
tion for treatment of bladder carcinoma. South Med J 2008;101:91–5. [6] Biot C, Rentsch C, Gsponer JR, et al. Preexisting BCG-specific T cells improve intravesical immunotherapy for bladder cancer. Sci Transl Med 2012;4:137ra72. [7] De Boer EC, Rooijakkers SJ, Schamhart DH, Kurth K-H. Cytokine gene expression in a mouse model: the first instillations with viable
4246–52. [25] Luo Y, Henning J, O’Donnell MA. Th1 cytokine-secreting recombinant Mycobacterium bovis bacillus Calmette-Gue´rin and prospective use in immunotherapy of bladder cancer. Clin Dev Immunol 2011: 728930.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 10
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
[26] Dufresne M, Dumas G, Asselin E´, Carrier C, Pouliot M, Reyes-Moreno
[28] Ito T, Hasegawa A, Hosokawa H, et al. Human Th1 differentiation
C. Pro-inflammatory type-1 and anti-inflammatory type-2 macro-
induced by lipoarabinomannan/lipomannan from Mycobacterium
phages differentially modulate cell survival and invasion of human bladder carcinoma T24 cells. Mol Immunol 2011;48: 1556–67. [27] Zlotta AR, Van Vooren JP, Denis O, et al. What are the immunologically active components of bacille Calmette-Guerin in therapy of superficial bladder cancer? Int J Cancer 2000;87:844–52.
bovis BCG Tokyo-172. Int Immunol 2008;20:849–60. [29] Bowyer L, Hall RR, Reading J, Marsh MM. The persistence of bacille Calmette-Guerin in the bladder after intravesical treatment for bladder cancer. Br J Urol 1995;75:188–92. [30] Jime´nez MS, Julia´n E, Luquin M. Misdiagnosis of Mycobacterium brumae infection. J Clin Microbiol 2011;49:1190–1.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Bladder Cancer Editorial by XXX on pp. x–y of this issue
Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo Estela Noguera-Ortega a,y, Silvia Secanella-Fandos a,y, Hasier Eran˜a a,z, Jofre Gasio´n a, Rosa M. Rabanal b, Marina Luquin a, Eduard Torrents c, Esther Julia´n a,* a
Departament de Gene`tica i de Microbiologia, Facultat de Biocie`ncies, Universitat Auto`noma de Barcelona, Bellaterra, Spain; b Unitat de Patologia Murina i Comparada,, Departament de Medicina Animal i Cirurgia, Facultat de Veterina`ria, Universitat Auto`noma de Barcelona, Bellaterra, Spain; c Bacterial Infections and Antimicrobial Therapy Group, Institute for Bioengineering of Catalonia, Barcelona, Spain
Article info
Abstract
Article history: Accepted March 27, 2015
Background: Bacillus Calmette-Gue´rin (BCG) prevents tumour recurrence and progression in non–muscle-invasive bladder cancer (BC). However, common adverse events occur, including BCG infections. Objective: To find a mycobacterium with similar or superior antitumour activity to BCG but with greater safety. Design: In vitro, ex vivo, and in vivo comparisons of the antitumour efficacy of nonpathogenic mycobacteria and BCG. Intervention: The in vitro antitumour activity of a broad set of mycobacteria was studied in seven different BC cell lines. The most efficacious was selected and its ex vivo capacity to activate immune cells and its in vivo antitumour activity in an orthotopic murine model of BC were investigated. Outcome measurements and statistical analysis: Growth inhibition of BC cells was the primary outcome measurement. Parametric and nonparametric tests were use to analyse the in vitro results, and a Kaplan-Meier test was applied to measure survival in mycobacteria-treated tumour-bearing mice. Results and limitations: Mycobacterium brumae is superior to BCG in inhibiting lowgrade BC cell growth, and has similar effects to BCG against high-grade cells. M. brumae triggers an indirect antitumour response by activating macrophages and the cytotoxic activity of peripheral blood cells against BC cells. Although no significant differences were observed between BCG and M. brumae treatments in mice, M. brumae treatment prolonged survival in comparison to BCG treatment in tumour-bearing mice. In contrast to BCG, M. brumae does not persist intracellularly or in tumour-bearing mice, so the risk of infection is lower. Conclusions: Our preclinical data suggest that M. brumae represents a safe and efficacious candidate as a therapeutic agent for non–muscle-invasive BC. Patient summary: We investigated the antitumour activity of nonpathogenic mycobacteria in in vitro and in vivo models of non–muscle-invasive bladder cancer. We found that Mycobacterium brumae effectively inhibits bladder cancer growth and helps the host immune system to eradicate cancer cells, and is a promising agent for antitumour immunotherapy. # 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Associate Editor: Shahrokh Shariat Keywords: Animal models Bacillus Calmette-Gue´rin Cytokines Immunomodulation Immunotherapy Mycobacteria Urothelial cell line
y
These authors contributed equally to this work. Current address: CIC bioGUNE, Derio, Vizcaya, Spain. * Corresponding author. Departament de Gene`tica i de Microbiologia, Facultat de Biocie`ncies, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain. Tel. +34 93 5813096; Fax: +34 93 5812387. E-mail address: [email protected] (E. Julia´n). z
http://dx.doi.org/10.1016/j.euf.2015.03.003 2405-4569/# 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 2
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
1.
Introduction
Despite the well-known benefits of bacillus CalmetteGue´rin (BCG) therapy for non–muscle-invasive bladder cancer (NMIBC) [1], it is not a perfect treatment. BCG, an attenuated mycobacterium obtained from the pathogenic Mycobacterium bovis, exerts both direct and indirect antitumour activities involving inhibition of cellular proliferation and stimulation of a robust immune response, respectively [1,2]. However, nonsevere toxicity is frequently observed [1], and severe adverse events, including BCG infections [3–5], limit the clinical applicability of this immunotherapy. Many studies have attempted to find agents as efficacious as BCG but safer. Killed BCG and BCG cell extracts are less efficacious than live BCG [2]. Moreover,
for the first instillations in a treatment regimen, live bacteria are needed to obtain a favourable immune response [6,7]. Other weaknesses regarding BCG functionality also have to be considered. First, different BCG strains can differ in effectiveness [8,9]. Second, BCG supply shortages worldwide have highlighted its costly production, mainly because of its slow growth. An attractive approach to overcome adverse BCG effects is the use of nonpathogenic mycobacteria. Most mycobacteria species (Fig. 1) are nonpathogenic and potentially share immunomodulatory antigens with BCG. Mycobacterium smegmatis and Mycobacterium phlei were initially proposed as alternatives to BCG for NMIBC treatment [10]. Later, researchers investigated Mycobacterium vaccae and Mycobacterium indicus pranii,
Fig. 1 – Genealogical tree of mycobacterial species selected for the study. Maximum likelihood phylogenetic tree of type strains from the Mycobacterium genus based on 16S rRNA gene sequences, indicating slow and rapid Mycobacterium growers. Arrows indicate the species selected for the study.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
initially used as immunotherapeutic agents for tuberculosis and leprosy, for treatment of other types of cancer [11]. All of these studies used cell extracts or heat-killed mycobacteria, probably because M. phlei, M. smegmatis, and M. vaccae infections have been described [12–14]. Therefore, the same limitations observed for heat-killed BCG also affect these other mycobacteria. Only recombinant M. smegmatis expressing human tumour necrosis factor (TNF)-a has been investigated as a live mycobacterium, and it is vastly superior to BCG, as demonstrated in in vitro assays and in vivo in mouse models [15]. However, genetically modified M. smegmatis and BCG have not been studied in humans. This previous research indicates that nonpathogenic mycobacterial species could have antitumour activity that is similar or superior to that of BCG. In this study we focused on nonpathogenic, rapidly growing species because of two advantages over BCG: a lower risk of infection and lower production costs. After selecting the mycobacterial species with the highest antitumour activity in vitro, we investigated its ex vivo capacity to activate immune cells and its in vivo antitumour activity.
2.5.
3
Cytotoxic activity
Experiments on inhibition of cell growth were performed using human peripheral blood mononuclear cells (PBMCs) as previously described [17]. The optimal MOI and time for cytokine production were first determined on a small set of PBMC samples. Sample collection was approved by the Ethics Committee of Universitat Auto`noma de Barcelona (UAB).
2.6.
Cytokine analysis
To quantify cytokines, enzyme-linked immunosorbent assays were performed on cell culture supernatants using commercially available kits (Supplementary Table 2). Nitric oxide (NO) production was assessed by measuring nitrite concentrations in the Griess reaction (Sigma– Aldrich, Madrid, Spain).
2.7.
Mycobacterial viability
To determine the intracellular viability of mycobacteria, T24 cells and J774 macrophages were infected and collected at different time points after infection as previously described [9].
2.8.
Orthotopic model of BC and intravesical treatment
Animal experiments were performed according to procedures approved
2.
Materials and methods
2.1.
Sequence alignment and phylogenetic inferences
by the Animal Care Committee of UAB. The murine model was developed as previously described [6,8,19]. In brief, C57Bl/6 female mice (7–9 wk old; Charles River Laboratories, Barcelona, Spain) were anaesthetised with isoflurane, and chemical lesions were induced on the bladder urothelium
16S rRNA sequences (1200 nucleotides) were obtained from the Ribosomal Database Project (release 11; http://rdp.cme.msu.edu/index. jsp) [16]. A set of 143 good-quality 16S rRNA sequences (only Mycobacterium genus-type strains) was downloaded and used for sequence alignment (Slow accurate alignment algorithm) and phylogenetic analysis (Neighbour-joining algorithm) using CLC Main Workbench v. 6.9.1 (http://www.clcbio.com/products/clc-main-workbench/). Tree topology was re-examined using the bootstrap (1000) method of resampling.
by instilling 50 ml of L-poly-lysine intravesically through a 24-gauge catheter (dwell time 10 min). Subsequently, 105 MB49 bladder tumour cells were instilled to induce tumours (dwell time 60 min). At 1 d after tumour implantation, the mice were randomly divided into three groups (7–14 animals/group) and administered mycobacterial cells (dwell time 60 min) or PBS. The animals were treated weekly for 4 wk (Fig. 5B) and evaluated on a daily basis in terms of health status and the presence of gross haematuria. After 60 d, surviving animals were sacrificed and bladders were fixed in buffered formalin (10%) and embedded in paraffin. Haematoxylin-eosin staining was performed on 4-mm sections for
2.2.
Bacterial strains and cell lines
histopathologic examination. Spleens were collected, disrupted, serially diluted, and plated to determine colony-forming units.
All mycobacterial strains, cell lines, and growth conditions are described in the supplementary material.
2.9.
2.3.
Student’s t-test (Prism software, GraphPad, San Diego, CA, USA) was used
Direct growth inhibition experiments
Statistical analysis
to assess the statistical significance of differences among groups for Tumour cells were infected with mycobacteria at different multiplicity
direct inhibition of BC cell growth, cytokine levels produced by BC cells,
of infection (MOI) levels, and cell proliferation was analysed as
and mycobacterial viability in infected cultures. The Wilcoxon-Mann-
previously described [9,17].
Whitney rank test was used to compare cytokine levels and percentage
2.4.
the statistical significance of Kaplan-Meier survival curves.
cytotoxicity (Prism software). Log-rank tests were applied to determine
Macrophage infection and surface receptor analysis
Murine J774 macrophages were seeded into six-well cell culture plates.
3.
Results
3.1.
M. brumae has a better antitumour effect than BCG against
After 24 h, lipopolysaccharide or mycobacterial suspension was added. After 3 h, the cells were washed and complete medium was added. At 24 h after infection, cell culture supernatants were harvested, and the cells were collected and centrifuged [18]. The pellets were resuspended in phosphate-buffered saline (PBS) with antimouse CD32/CD16 (Mouse FcBlock, Becton-Dickinson, Pharmingen, San Jose, CA, USA) and incubated on ice for 15 min. Then phycoerythrin-conjugated anti-CD80, anti-CD86, and anti-CD40 (BD Pharmingen) antibodies were added. Isotype control antibodies were also included. The cells were then analysed on a FacsCalibur flow cytometer using CellQuest software (both from BD Biosciences, San Jose, CA, USA).
low-grade tumour cells
We assessed the ability of eight different mycobacteria to inhibit tumour cell growth. The antiproliferative activity of mycobacteria depended on both time and MOI (Supplementary Fig. 1). Comparison of tumour growth inhibition revealed that not all mycobacteria had antitumour effects (Fig. 2). BCG, M. brumae, M. phlei, and M. vaccae smooth
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 4
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 2 – Mycobacterium brumae had the greatest inhibitory effect on the growth of low-grade tumour cells among all the mycobacteria studied. (A) Growth inhibition and cytokine production for high-grade (T24 and J82) and low-grade (RT4) bladder cancer cells infected with mycobacteria (multiplicity of infection [MOI] 12.5:1) at 72 h after infection for T24 and RT4 cells, and 120 h for J82 cells. Data denote percentage cell survival relative to untreated cells (control). Values represent the mean W standard deviation (SD) for triplicate preparations. Data are representative of one of at least three independent experiments. (B) Growth inhibition of T24 and low-grade (SW780, 5637, RT112, and MG-HU-3) bladder tumour cells infected with mycobacteria (MOI 12.5:1) at 72 h after infection. Data represent percentage cell survival relative to untreated cells. Each column represents the average W SD for triplicate cultures from at least three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05 versus BCG-infected cells, ** p < 0.05 for infected versus noninfected cells, Student’s t-test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 3 – Induction of an immune response and antitumor activity by macrophages and peripheral blood mononuclear cells (PBMCs) infected with Mycobacterium brumae. (A) Induction of activation markers on mycobacteria-infected macrophages. J774 murine macrophages were not stimulated (NS), treated with lipopolysaccharide (LPS; 10 mg/ml), or infected with mycobacteria (multiplicity of infection [MOI] 1:1 for BCG and 10:1 for the other mycobacteria) for 24 h. The cells were then stained with anti-CD40, anti-CD80, or anti-CD86 antibodies. Results are expressed as the mean fluorescence intensity (Geo MFI from 10 000 events for each histogram) W standard deviation (SD) for three independent experiments. (B) Production of cytokines (interleukin [IL]-6, IL-12) and nitric oxide (NO) in uninfected and infected macrophages was measured 48 h after infection. (C) Cytokine production by mycobacteria-infected PBMCs. PBMC were infected with BCG (MOI 0.1:1) or M. brumae (MOI 1:1) for 7 days, and cytokine production was measured. ND = not determined (data were below the minimum detection value of the test). (d). Cytotoxicity of BCG-activated (blue column) and M. brumae-activated (red column) PBMCs, or soluble factors (cell culture supernatants) from BCG-activated (blue) and M. brumae-activated (red) PBMCs against T24 cells after 48 h. PBMCs alone or the corresponding cell culture supernatant against T24 cells (black columns) and T24 alone (white columns) were used as controls. In (B–D), values represent the mean W SD for triplicate cell cultures from three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05 versus BCG treatment, ** p < 0.05 versus nontreated cells, Mann-Whitney test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
5
EUF-20; No. of Pages 10 6
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
variant (S) exhibited similar antitumour activity in highgrade T24 and J82 cell lines (Fig. 2). In low-grade RT4 cells, M. brumae and M. phlei inhibited cell proliferation to a greater extent than BCG and M. vaccae (S). Infection with BCG and M. brumae induced the highest production of cytokines, apart from in RT4 cells (Fig. 2A). Given that M. brumae and M. phlei were more efficacious than BCG in inhibiting cell growth in low-grade RT4 cells, we evaluated the antitumour activity of a number of other mycobacterial species in four additional low-grade cell lines and the high-grade T24 cell line. In T24 cells, all mycobacteria showed lower antitumour activity than M. brumae, BCG, or M. phlei (Fig. 2B). In low-grade cell lines, M. brumae significantly inhibited cell growth to a greater extent than the other mycobacteria (Fig. 2B). As Mycobacterium gastri had the lowest inhibition and cytokine induction rates, it was used as a negative control in subsequent experiments. We did not observe growth inhibition of the murine normal urothelium G/G cell line after infection with BCG or M. brumae (data not shown). 3.2.
(Fig. 5A), we performed in vivo experiments (Fig. 5B). Approximately 9 2 d after tumour induction, all animals exhibited haematuria, a hallmark of established tumours. While 100% of nontreated animals died within 40 d of tumour implantation, mycobacteria treatment significantly prolonged survival in tumour-bearing mice. As for BCG [20,21], the response to M. brumae in the animal model is dose-dependent (Supplementary Fig. 2). At 60 d after tumour implantation, no significant differences were observed between the BCG and M. brumae treatment groups, although M. brumae treatment prolonged survival compared to BCG treatment (Fig. 5C).
M. brumae activates immune cells and triggers cytotoxic
activity against tumour cells
To evaluate the immunomodulatory potential of M. brumae, we evaluated its ability to activate macrophages and PBMCs. M. brumae triggered the expression of activation markers and the production of cytokines and NO in macrophages (Fig. 3A,B). Notably, levels of CD40 expression significantly increased after infection with M. brumae or M. vaccae in comparison to BCG and M. phlei (Fig. 3A). In PBMCs, M. brumae triggered the production of interleukin (IL)-8, IL-10, IL-12, and TNF-a to levels similar to those induced by BCG infection (Fig. 3C). It was not possible to measure the response to M. phlei because this mycobacterium formed a biofilm on the surface of PBMC cultures 48 h after infection. We then analysed the ex vivo antitumour activity of mycobacteria-activated PBMCs. Both BCG- and M. brumae-infected PBMCs triggered direct cytotoxic activity against T24 cells (Fig. 3D). In addition, soluble factors derived from both mycobacteria-infected cultures showed cytotoxic effects against T24 cells. Coculture with mycobacteria-activated PBMCs or nonactivated PBMCs significantly increased constitutive IL-8 and IL-6 production in T24 cells (Fig. 3D). 3.3.
M. brumae does not persist in T24 tumour cells or
J774 macrophages
We corroborated the nonpathogenicity of M. brumae by investigating its survival in nonprofessional and professional phagocytic cells. Unlike BCG, M. brumae and M. vaccae S were not viable in tumour cells or macrophages at 24 and 96 h after infection (Fig. 4A,B). 3.4.
Intravesical treatment with M. brumae enhances the
survival of tumour-bearing mice
After we demonstrated that M. brumae inhibits the growth of murine MB49 tumour cells in vitro, similar to BCG
Fig. 4 – Mycobacterial survival in infected cell cultures. (A) T24 tumour cells and (B) J774 macrophages were infected at a multiplicity of infection of 10:1 for 3 h. After washing, samples were plated on Middlebrook agar and viable intracellular mycobacteria were counted. Results are expressed as mean of colony-forming units (CFU)/well W standard deviation for triplicate wells from three independent experiments. M. vaccae S = smooth variant; M. vaccae R = rough variant. * p < 0.05, ** p < 0.01 versus BCG infection, Student’s t-test.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Fig. 5 – Mycobacterium brumae treatment prolongs survival in tumour-bearing mice. (A) Growth inhibition and interleukin (IL)-6 and keratinocyte chemoattractant (KC) production for high-grade murine MB49 cells infected with mycobacteria (multiplicity of infection 10:1). Data represent percentage cell survival relative to untreated cells. Values represent the mean W standard deviation for triplicate preparations from two independent experiments. ** p < 0.05 versus control group, * p < 0.05 versus BCG group, Student’s t-test. (B) Schematic schedule of the in vivo experiments. C57BL/6 mice (n = 7–14) received four weekly intravesical instillations of phosphate-buffered saline (PBS), BCG, or M. brumae (indicated by arrows) after tumour implantation. (C) Kaplan-Meier analysis of survival in tumour-bearing mice after treatment with PBS (black line), 2–4 T 106 BCG (blue), or 2 T 107 M. brumae (red). ** p < 0.01 versus PBS group, log-rank (Mantel-Cox) test. (D) Colony-forming units (CFU) for splenocyte cultures from mice treated with M. brumae (red) or BCG (blue). The number of viable bacteria from each spleen is represented as a dot; the line identifies the median for the group. (E) Representative histologic images (hematoxylin-eosin staining) of bladder sections from nontreated (PBS), M. brumae-treated, and BCGtreated tumour-bearing mice. Scale bar, 1 mm (upper images) and 50 or 100 mm (lower images).
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
7
EUF-20; No. of Pages 10 8
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
Examination of splenocyte cultures revealed that BCG-treated mice had viable bacteria counts, while no mycobacteria were detected in surviving M. brumae-treated mice (Fig. 5D). Histopathologic examination of a nontreated bladder showed a solid mass growing within the bladder lumen and infiltrating the mucosa. Tumours were homogeneous in pattern, with several haemorrhagic and necrotic foci. In addition, mitotic cells were observed at higher magnification (Fig. 5E). Mycobacteria-treated mice had chronic inflammation, with moderate to intensive lymphocytic infiltration into the lamina propia and activation of lymphoid follicles. No granulomatous inflammation was observed (Fig. 5E). 4.
Discussion
Among all the mycobacteria studied, M. brumae had the strongest inhibitory activity on growth. M. brumae inhibited high-grade tumour growth to a similar extent to BCG and M. phlei, but was superior to the other mycobacteria in inhibiting low-grade tumour cells. The antiproliferative properties are not related to species origin because the most efficacious bacteria, M. brumae, BCG, M. phlei, and M. vaccae, are not phylogenetically related (Fig. 1), whereas species close to M. brumae (M. fallax and M. chitae) were not as efficacious as M. brumae. Our results indicate that all mycobacteria enter T24 cells, as previously shown for BCG and M. smegmatis [22], but the outcome after internalisation differs among mycobacterial species: some remain alive, while others are destroyed (Fig. 4). We corroborated these results for BCG and M. brumae using microscopy techniques (data not shown). Therefore, the antigens exposed and their role could differ among different mycobacteria. Further studies are needed to unravel the M. brumae antigen(s) related to BC inhibitory activity. After our initial encouraging results for M. brumae, we studied its capacity to activate an immune response. In vitro and in vivo studies have demonstrated that an effective immunotherapeutic agent should activate antigen-presenting cells and strongly induce Th1 cells [2,23–25]. M. brumae elevated the expression of activation surface markers. Interestingly, M. brumae triggered higher expression of CD40 than BCG did. Activation of CD40 upregulates CD80 and CD86, which improve the pattern of macrophages with the M1 phenotype, which is necessary for activating effector T cells and effective antitumour therapy [23]. M1 macrophages participate in the in vitro growth inhibition of T24 cells, suggesting that the same behaviour could occur in vivo after mycobacterial treatment [26]. In addition to activating macrophages, M. brumae induces cytokine secretion and activates the antitumour capacity of infected PBMCs, although to a lesser extent than BCG. In vitro and in vivo studies demonstrated the influence of cytokines and the host immune response on the outcome of BCG immunotherapy in BC [1,2]. Several BCG subcomponents are potent inducers of IL-12 and interferon (IFN)-g production by PBMCs, promoting Th1 cell differentiation and enhancing cytotoxicity against BC cells [27,28]. IL-12 and IFN-g are critical for effective immunotherapy in BC mouse models [19,24]. Thus, our results suggest that
M. brumae shares some components with BCG that are critical for its immunomodulatory activity. Finally, we demonstrated that M. brumae treatment significantly prolongs survival in tumour-bearing mice. Importantly, BCG was recovered from splenocytes of BCGtreated mice, whereas M. brumae was not recovered from the spleen of surviving M. brumae-treated animals. This is in accordance with our in vitro results, which demonstrated that M. brumae is cleared within a few days after infection of BC cells and macrophages, whereas BCG persists (Fig. 4). Although the limitations of the mouse model must be taken into account, this result is indicative of the BCG infections observed in NMIBC patients after treatment. Accidents after traumatic instillations in humans can lead to systemic BCG invasion and possible infection. Numerous reports have documented the onset of BCG infection. Bowyer et al [29] detected acid-fast bacilli in a bladder 16 mo after the completion of intravesical instillations. Systemic infections have even been observed 3 y [4] and 6 y [3] after completion of BCG treatment due to reactivation of the bacilli when a patient’s immunocompetence is decreased. The fact that viable M. brumae was not recovered in an animal model implies that use of M. brumae for BC treatment could avoid this adverse event associated with BCG. M. brumae is a saprophyte mycobacterium associated with water and soil. Although a case of catheterrelated infection was attributed to M. brumae, it was subsequently confirmed that the isolate did not correspond to M. brumae [30]. Therefore, infections due to M. brumae have not been described. Our study has some limitations. The molecular mechanisms by which M. brumae inhibits tumour cell growth and triggers an immune response have not been addressed. Although this was not the primary aim of our study, we are currently performing in vitro and in vivo experiments to elucidate the mechanisms. Moreover, although the animal model we used is currently the most appropriate and widely accepted for comparing potential agents with BCG treatment [6,19], other animal models (eg, using rats and/ or pigs) could be used to corroborate our results before clinical assays of the safety and tolerability of M. brumae in humans. 5.
Conclusions
In summary, our data indicate that M. brumae effectively inhibits BC growth and helps the host immune system to eradicate cancer cells. Moreover, M. brumae grows three times faster than BCG and is therefore faster and cheaper to produce than BCG. Although further research is needed, our preclinical results suggest that M. brumae represents a safe and efficacious agent for antitumour immunotherapy and may be a promising option for NMIBC treatment. Author contributions: Esther Julia´n had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Julia´n.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
˜ a, Gasio´n, Acquisition of data: Noguera-Ortega, Secanella-Fandos, Eran Rabanal, Torrents, Julia´n. Analysis and interpretation of data: Noguera-Ortega, Secanella-Fandos, ˜ a, Gasio´n, Rabanal. Luquin, Torrents, Julia´n. Eran Drafting of the manuscript: Noguera-Ortega, Torrents, Julia´n. Critical revision of the manuscript for important intellectual content: Noguera-Ortega, Rabanal, Luquin, Torrents, Julia´n. Statistical analysis: Noguera-Ortega, Secanella-Fandos, Julia´n. Obtaining funding: Julia´n. Administrative, technical, or material support: Luquin, Torrents, Julia´n. Supervision: Julia´n. Other: None. Financial disclosures: Esther Julia´n certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Silvia Secanella-Fandos, Marina Luquin, and Esther Julia´n are the inventors of patent PCT/ ES2013/070547. The remaining authors have nothing to disclose.
9
bacillus Calmette-Guerin determine the succeeding Th1 response. J Urol 2003;170:2004–8. [8] Rentsch CA, Birkha¨user FD, Biot C, et al. Bacillus Calmette-Gue´rin strain differences have an impact on clinical outcome in bladder cancer immunotherapy. Eur Urol 2014;66:677–88. [9] Secanella-Fandos S, Luquin M, Julia´n E. Connaught and Russian strains showed the highest direct antitumor effects of different bacillus Calmette-Gue´rin substrains. J Urol 2013;189:711–8. [10] Yarkoni E, Rapp HJ. Immunotherapy of experimental cancer by intralesional injection of emulsified nonliving mycobacteria: comparison of Mycobacterium bovis (BCG), Mycobacterium phlei, and Mycobacterium smegmatis. Infect Immun 1980;28:887–92. [11] Grange JM, Bottasso O, Stanford CA, Stanford JL. The use of mycobacterial adjuvant-based agents for immunotherapy of cancer. Vaccine 2008;26:4984–90. [12] Paul E, Devarajan P. Mycobacterium phlei peritonitis: a rare complication of chronic peritoneal dialysis. Pediatr Nephrol 1998;12: 67–8. [13] Hachem R, Raad I, Rolston KV, et al. Cutaneous and pulmonary infections caused by Mycobacterium vaccae. Clin Infect Dis 1996; 23:173–5.
Funding/support and role of the sponsor: This work was funded by the
[14] Khatter S, Singh UB, Arora J, Rana T, Seth P. Mycobacterial infections
Spanish Science and Education Ministry (SAF2006-05868), the Spanish
in human immuno-deficiency virus seropositive patients: role of
Ministry of Science and Innovation (Instituto de Salud Carlos III – PI10/
non-tuberculous mycobacteria. Indian J Tuberc 2008;55:28–33.
01438), the European Regional Development Fund (FEDER), and the
[15] Young SL, Murphy M, Zhu XW, et al. Cytokine-modified Mycobacte-
Generalitat of Catalunya (2009SGR-108). Silvia Secanella-Fandos was supported by a Ph.D. fellowship from Universitat Auto`noma de
rium smegmatis as a novel anticancer immunotherapy. Int J Cancer
Barcelona. These agencies had no role in study design, data collection
[16] Cole JR, Wang Q, Fish JA, et al. Ribosomal database project: data and
and analysis, decision to publish, or preparation of the manuscript.
2004;112:653–60. tools for high throughput rRNA analysis. Nucleic Acids Res 2014; 42:D633–42.
Acknowledgments: We thank all the members of the Animal Facility and
[17] Secanella-Fandos S, Noguera-Ortega E, Olivares F, Luquin M, Julia´n
Manuela Costa from the Cytometry Unit at UAB for their support and advice. We also thank Dr. Francesc Padro´s, Dra. Mercedes Ma´rquez, and
E. Killed but metabolically active Mycobacterium bovis bacillus Calmette-Gue´rin retains the antitumor ability of live bacillus Calmette-Gue´rin. J Urol 2014;191:1422–8.
Ester Blanco for histopathologic examination of bladders.
[18] Kan-Sutton C, Jagannath C, Hunter RL. Trehalose 6,60 -dimycolate on
Appendix A. Supplementary data
the surface of Mycobacterium tuberculosis modulates surface marker expression for antigen presentation and costimulation in
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. euf.2015.03.003.
murine macrophages. Microbes Infect 2009;11:40–8. [19] Zaharoff D, Hoffman BS, Hooper HB, et al. Intravesical immunotherapy of superficial bladder cancer with chitosan/interleukin-12. Cancer Res 2009;69:6192–9. [20] Matsushima M, Horinaga M, Fukuyama R, et al. Enhanced antitu-
References [1] Gontero P, Bohle A, Malmstrom P-U, et al. The role of bacillus Calmette-Gue´rin in the treatment of non-muscle-invasive bladder cancer. Eur Urol 2010;57:410–29. [2] Alexandroff AB, Nicholson S, Patel PM, Jackson AM. Recent advances in bacillus Calmette-Guerin immunotherapy in bladder cancer.
mor effect of combination intravesical mitomycin C and bacillus Calmette-Guerin therapy in an orthotopic bladder cancer model. Oncol Lett 2011;2:13–9. [21] Xiao Z, Hanel E, Mak A, Moore RB. Antitumor efficacy of intravesical BCG, gemcitabine, interferon-a and interleukin-2 as mono- or combination-therapy for bladder Cancer in an orthotopic tumor model. Clin Med Insights Oncol 2011;5:315–23.
Immunotherapy 2010;2:551–60. [3] Alvarez-Mu´gica M, Ferna´ndez J, Bulnes V, et al. Fatal outcome due to
[22] Redelman-Sidi G, Iyer G, Solit DB, Glickman MS. Oncogenic activa-
sepsis by Mycobacterium bovis six years after BCG intravesical
BCG entry into bladder cancer cells. Cancer Res 2013;73:1156–67.
instillations. J Med Cases 2010;1:81–3. [4] Gonzalez OY, Musher DM, Brar I, et al. Spectrum of bacille CalmetteGue´rin (BCG) infection after intravesical BCG immunotherapy. Clin Infect Dis 2003;36:140–8.
tion of Pak1-dependent pathway of macropinocytosis determines [23] Kimura T, Ohashi T, Kikuchi T, Kiyota H, Eto Y, Ohishi Y. Antitumor immunity against bladder cancer induced by ex vivo expression of CD40 ligand gene using retrovirus vector. Cancer Gene Ther 2003;10: 833–9.
[5] Nadasy KA, Patel RS, Emmett M, et al. Four cases of disseminated
[24] O’Donnell MA, Luo Y, Chen X, Szilvasi A, Hunter SE, Clinton SK.
Mycobacterium bovis infection following intravesical BCG instilla-
Role of IL-12 in the induction and potentiation of IFN-gamma in response to bacillus Calmette-Gue´rin. J Immunol 1999;163:
tion for treatment of bladder carcinoma. South Med J 2008;101:91–5. [6] Biot C, Rentsch C, Gsponer JR, et al. Preexisting BCG-specific T cells improve intravesical immunotherapy for bladder cancer. Sci Transl Med 2012;4:137ra72. [7] De Boer EC, Rooijakkers SJ, Schamhart DH, Kurth K-H. Cytokine gene expression in a mouse model: the first instillations with viable
4246–52. [25] Luo Y, Henning J, O’Donnell MA. Th1 cytokine-secreting recombinant Mycobacterium bovis bacillus Calmette-Gue´rin and prospective use in immunotherapy of bladder cancer. Clin Dev Immunol 2011: 728930.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003
EUF-20; No. of Pages 10 10
EUROPEAN UROLOGY FOCUS XXX (2015) XXX–XXX
[26] Dufresne M, Dumas G, Asselin E´, Carrier C, Pouliot M, Reyes-Moreno
[28] Ito T, Hasegawa A, Hosokawa H, et al. Human Th1 differentiation
C. Pro-inflammatory type-1 and anti-inflammatory type-2 macro-
induced by lipoarabinomannan/lipomannan from Mycobacterium
phages differentially modulate cell survival and invasion of human bladder carcinoma T24 cells. Mol Immunol 2011;48: 1556–67. [27] Zlotta AR, Van Vooren JP, Denis O, et al. What are the immunologically active components of bacille Calmette-Guerin in therapy of superficial bladder cancer? Int J Cancer 2000;87:844–52.
bovis BCG Tokyo-172. Int Immunol 2008;20:849–60. [29] Bowyer L, Hall RR, Reading J, Marsh MM. The persistence of bacille Calmette-Guerin in the bladder after intravesical treatment for bladder cancer. Br J Urol 1995;75:188–92. [30] Jime´nez MS, Julia´n E, Luquin M. Misdiagnosis of Mycobacterium brumae infection. J Clin Microbiol 2011;49:1190–1.
Please cite this article in press as: Noguera-Ortega E, et al. Nonpathogenic Mycobacterium brumae Inhibits Bladder Cancer Growth In Vitro, Ex Vivo, and In Vivo. Eur Urol Focus (2015), http://dx.doi.org/10.1016/j.euf.2015.03.003