Vaccine 33 (2015) 2118–2124

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

Bacterial phospholipases C as vaccine candidate antigens against cystic fibrosis respiratory pathogens: The Mycobacterium abscessus model Vincent Le Moigne a , Martin Rottman a , Céline Goulard a , Benoît Barteau b , Isabelle Poncin c , Nathalie Soismier a , Stéphane Canaan c , Bruno Pitard d,e , Jean-Louis Gaillard a , Jean-Louis Herrmann a,∗ a

INSERM U1173, UFR Simone Veil, Versailles-Saint-Quentin University, 78180 Saint-Quentin en Yvelines, France IN-CELL-ART, Nantes, France c CNRS—Aix-Marseille Université—Enzymologie Interfaciale et Physiologie de la Lipolyse UMR 7282, 31 chemin Joseph Aiguier, 13402 Marseille Cedex 20, France d INSERM UMR 1087, CNRS UMR 6291, Nantes, France e Université de Nantes, l’institut du thorax, Nantes, France b

a r t i c l e

i n f o

Article history: Received 28 October 2014 Received in revised form 5 March 2015 Accepted 11 March 2015 Available online 21 March 2015 Keywords: Mycobacteria PLC DNA vaccine Pulmonary diseases Cystic fibrosis F508

a b s t r a c t Background: Vaccine strategies represent one of the fighting answers against multiresistant bacteria in a number of clinical settings like cystic fibrosis (CF). Mycobacterium abscessus, an emerging CF pathogen, raises difficult therapeutic problems due to its intrinsic antibiotic multiresistance. Methods: By reverse vaccinology, we identified M. abscessus phospholipase C (MA-PLC) as a potential vaccine target. We deciphered here the protective response generated by vaccination with plasmid DNA encoding the MA-PLC formulated with a tetra functional block copolymer 704, in CF (F508) mice. Protection was tested against aerosolized smooth and rough (hypervirulent) variants of M. abscessus. Results: MA-PLC DNA vaccination (days 0, 21, 42) elicited a strong antibody response. A significant protective effect was obtained against aerosolized M. abscessus (S variant) in F508 mice, but not in wild-type FVB littermates; similar results were observed when: (i) challenging mice with the “hypervirulent” R variant, and; (ii) immunizing mice with purified MA-PLC protein. High IgG titers against MA-PLC protein were measured in CF patients with M. abscessus infection; interestingly, significant titers were also detected in CF patients positive for Pseudomonas aeruginosa versus P. aeruginosa-negative controls. Conclusions: MA-PLC DNA- and PLC protein-vaccinated mice cleared more rapidly M. abscessus than ␤galactosidase DNA- or PBS- vaccinated mice in the context of CF. PLCs could constitute interesting vaccine targets against common PLC-producing CF pathogens like P. aeruginosa. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Fighting against multiresistant infectious diseases needs to find new strategies, involving more the host response than focusing on developing new antimicrobials. Chronic obstructive pulmonary disease (COPD) and cystic fibrosis (CF), a Mendelian deficit due to a series of mutation in the cftr gene encoding the Cl− channel [1], represent critical examples whereby multiresistant bacteria raised

∗ Corresponding author at: UMR1173, UFR des Sciences de la Santé, 2 avenue de la Source de la Bièvre, 78180 Montigny le Bretonneux, France. Tel.: +33 1 47 10 79 50. E-mail address: [email protected] (J.-L. Herrmann). http://dx.doi.org/10.1016/j.vaccine.2015.03.030 0264-410X/© 2015 Elsevier Ltd. All rights reserved.

therapeutic problems with severe even fatal infections [2,3]. Lung infections in CF patients represent the most frequent but also the more serious manifestations since they are responsible for more than 90% of CF patient deaths [4]. Mycobacterium abscessus complex [5] is a rapidly growing mycobacterium that can be responsible for a broad spectrum of diseases in humans [6–14]. Like Pseudomonas aeruginosa and other non-fermenting Gram-negative bacteria, M. abscessus complex strains demonstrated of a peculiar link with CF patients [6–9]. Recent reports of human-to-human transmission in the context of CF care have been described [15,16]. The most typical form of M. abscessus infection in CF is a pseudotuberculous lung disease with granuloma and caseous necrosis [17]. M. abscessus pathology is mostly related to a morphotype transition (smooth “S” to rough “R”

V. Le Moigne et al. / Vaccine 33 (2015) 2118–2124

form) for which the molecular mechanisms [18] and host responses were recently unraveled [19,20]. M. abscessus raises very challenging therapeutic issues because of its natural resistance to most available antibiotics [21,22]. Severe, even fatal, infections in CF patients have been described due to therapeutic deadlock [23]. M. abscessus infection is a contraindication for lung transplantation in several countries [24], leaving CF patients without therapeutic options. Finally, we recently demonstrated the significant link between previous intravenous antibiotic courses and the isolation of M. abscessus in CF patient lungs [25] further underlining the role of broad-spectrum antimicrobial therapy in the emergence of M. abscessus disease. Vaccine strategy represents one of the new approaches to fight multidrug resistant bacteria. In the recent years, the search for vaccine targets have benefited from reverse vaccinology, whereby genomes sequence comparison between pathogenic versus nonpathogenic bacterial genomes allowed identification of key factors of bacterial pathogenicity as vaccine candidates [26]. Deciphering the M. abscessus type strain whole genome sequence obtained by Sanger sequencing [21] evidenced “non-mycobacterial” virulence genes acquired by horizontal gene transfer (HGT) from major CF pathogens such as Pseudomonas aeruginosa or Burkholderia cenocepacia [21]. One of these genes is MAB 0555, encoding a phospholipase C absent from the genomes of M. smegmatis or M. chelonae. The M. abscessus PLC (MA-PLC) has 45% amino acid (aa) identity (length of comparison: 503/692 aa) with the PLC of P. aeruginosa (versus only 38 to 42% identity with the four PLC of Mycobacterium tuberculosis) [27]. The documented role of phospholipases as virulence factors involved in the intracellular lifestyle of several pathogens including CF pathogens [28] made MA-PLC a likely virulence factor and a promising vaccine target candidate. DNA vaccine technology has been used successfully in the context of cancer and infectious diseases [26]. Over the last 15 years, DNA vaccines have proven effective in animal models including against HIV, malaria and influenza [29]. DNA vaccines have been extensively evaluated in humans with a recent review identifying 72 Phase I, 20 Phase II and two Phase III human trials [30]. Mycobacteria, particularly M. tuberculosis, have received a particular attention in this respect [31,32]. Synthetic delivery system (SDS) consisting of tetra-functional block copolymers (TFBC) demonstrated of a dramatic enhancement of target protein expression in preclinical models [33–35], maximizing access of the plasmid DNA to the cytosol, with activation of DNA sensors that trigger an innate immune response. Benefiting of this expertise, we set up a DNA vaccine model using the TFBC 704, which is under regulatory development for the treatment by vaccination of the human hepatocellular-carcinoma, to document the protective role of MA-PLC immunization in CF hosts using F508 (the most frequent CF mutation) mutated mice [36].

2119

Saint Isle, France). F508 FVB mice, supplemented with movicol (Norgine, The Netherlands), and their wild type FVB littermates [36] were from INRA (Jouy en Josas, France).

2.2. Vaccination experiments Animal experiments were performed according to institutional and national ethical guidelines (Agreement n◦ 92-033-01). Before injection, mice were anesthetized by isoflurane inhalation (Abbvie, Rungis, France). Mice were immunized by the intramuscular (im) injection into each quadriceps muscle of 50 ␮l of the MA-PLC encoding plasmid DNA or the control ␤-galactosidase encoding plasmid DNA, formulated with the copolymer 704 (In Cell Art, Nantes, France); following a prime/boost scheme (days 0, 21 and 42) (supplementary Figure 1 C). For protein immunization, mice were subcutaneously (sc) injected with recombinant purified MAPLC (20 ␮g per mouse in a total volume of 200 ␮l) in PBS buffer with aluminum hydroxide following an identical scheme. Control mice received 200 ␮l PBS with aluminum hydroxide. Blood samples were collected from retro-orbital sinus; samples were warmed at 37 ◦ C for 1 h and centrifuged (10,000 g, 10 min). Collected sera were stored at −40 ◦ C. 2.3. Antibody response analysis 2.3.1. ELISA assays Assay plates (Maxisorp, Nunc, Roskilde, Denmark) were coated overnight with 100 ng of purified MA-PLC [37] or ␤-galactosidase in 100 ␮l of carbonate-bicarbonate buffer (0.1 M, pH 9.6) at 4 ◦ C. Plates were washed (phosphate-buffered saline-Tween 20 (PBST); 0.05% v/v) and blocked for 1 h at 37 ◦ C with PBS-T + 0.5% BSA. Serial dilutions of the assayed mouse sera were incubated at 37 ◦ C (90 min) and washed four times with PBS-T. Phosphatase-alkalineconjugated goat anti-mouse IgG was used (1/5000; 90 mn, 37 ◦ C) and washed four times. Liquid phosphatase-alkaline substrate was added and the plates were read at 405 nm. Serum samples from CF patients enrolled in the French study cohort “OMA” (Vaincre La Mucoviscidose project N◦ RF20110600446/1/3/130) [9] were diluted 1:400 for MA-PLC antibody response evaluation.

2.3.2. Western blot (WB) Transfected Hela cells were lyzed after 24 h by the use of Ripa buffer and centrifuged. Protein concentration was determined in the supernatant. 40 ␮g of proteins were separated on SDS-PAGE, transferred onto nitrocellulose membrane, incubated with the mouse sera, washed and peroxidase-conjugated goat anti-mouse (IgG; 1/5000) was added, washed and solid peroxidase substrate added. Purified MA-PLC was used as a control.

2. Materials and methods

2.4. Aerosol M. abscessus infection

2.1. Plasmids, bacterial strains and mice

Mice were aerosolized with M. abscessus on day 56, i.e. 14 days after the last boost (supplementary Figure 1 C). Briefly, aerosol was generated by a Micro Mist® small volume nebulizer (Hudson RCI-Teleflex medical, Research Triangle Park, NC, USA) containing 6 ml of 1–5 × 108 CFU/ml bacterial suspension in isotonic injection grade saline. Mice were anesthetized with isoflurane (AbbVie, Rungis, France) and 200 ␮g of etomidate (Hypnomidate, JanssenCilag, Issy-les-Moulineaux, France) and placed for 15 min in a filter-confined 50 ml container allowing the selective exposure of their snout to the nebulized suspension. Five to seven mice were aerosolized with the same inoculum during each experimental run, with 3 consecutive runs performed per data point.

Plasmid pVAX1-MAB0555 (4371 bp) contains kanamycin resistance gene and cDNA-MAB 0555 insert of 1440 bp under the control of a CMV promoter (supplementary Figure 1A). MA-PLC gene sequence was optimized for expression in eukaryotic cells by rare codons substitution. Non-relevant plasmid pCMV ␤-galactosidase contains an ampicillin resistance gene and a cDNA-E. coli ␤galactosidase gene (3141 bp) under the control of a CMV promoter (supplementary Figure 1B). S and R variants of M. abscessus CIP 104536T were used for infectious challenges and prepared as previously described [19,37]. BALB/c mice were from Janvier (Le Genest

V. Le Moigne et al. / Vaccine 33 (2015) 2118–2124

2.5. Organ sampling – CFU counts Mice were sacrificed on days 1, 7, 14 and 21 (DNA vaccination) or on days 1, 12 and 21 post-challenge (protein vaccination). Lungs, spleen and liver were removed aseptically, and CFU counts were determined as described [21,38]. Vaccinated and non-vaccinated mice cleared the infection at day 28.

A Antibody titer

2120

3000 2000 1000

2.6. Statistical analysis

0

Fisher’s exact test and Student t test were used. A p value