Fig 1.
In vitro and in vivo activities of the monoclonal antibody (mAb) VSX targeting P. aeruginosa core LPS.
(A) In vitro killing of P. aeruginosa PAO1 in the presence of PMN and complement (OPKA) by VSX. The lot#-472 is representative of mAbs able to bind P. aeruginosa but without any detectable OPKA activity. C = complement. HIC = heat inactivated complement. PMN = polymorphonuclear leukocytes. Abs = antibodies. (B) Acute lung infection model. Challenge dose: 2x106 CFU. Inoculation: intranasal, 106 CFUs in each nostril. Mice = 10 per group (two experiments with 5 animals per group each time). mAbs were injected IP 4 hours post-infection. Dose of the mAbs: VSX = 15 mg/kg. Control mAb (against Clostridioides difficile) = 15 mg/kg. P-value = 0.04, measured in a log-rank test.
Fig 2.
(A) The workflow for identification of an AMP to deploy in the construction of an ADC. (B) The AMP P297 showed rapid bactericidal activity in a time-kill assay using P. aeruginosa ATCC 27853. A growth control was compared with P297 at 0.5, 1, 2, 4, 8, or 16x MIC of peptide or with ciprofloxacin at 8x MIC. Note that the 4x, 8x, 16x, and ciprofloxacin 8x MICs all overlap. (C) Calcein leakage. The mechanism of action for P297 likely involves membrane disruption, as assessed by measuring calcein leakage from DOPE/DOPG liposomes. Various concentrations of peptide (from 0.05–100 μg/ml) were incubated with liposomes for defined lengths of time (5–45 min). Release of calcein was assessed by measuring the increase in fluorescence at 530 nm, compared with a non-peptide reference. Peptide concentrations above 1 μg/ml resulted in a measurable increase in fluorescence. (D) Assessment of resistance rates for P297 vs. colistin, using two different procedures. (E) Comparison of wild-type P. aeruginosa ATCC strain 27853 with two P297-resistant mutants indicates differences in drug sensitivity and phenotype. The mutant strains were less sensitive to P297.
Table 1.
In vitro Activity and Toxicity of Representative Peptide Variants.
Fig 3.
Synthesis, characterization and in vitro evaluation of VSX conjugates.
(A) Synthesis of antibody-AMP conjugates, using sortase ligation. VSX was expressed in Expi293 cells with a (GS)15 flexible linker and a sortase acceptor tag (LPETGGSG) present at the C-terminus of both the light chain and the heavy chain. Next, a (GGG)-modified AMP (P297, for example) was added covalently via incubation with recombinantly produced sortase for a target DAR of 4. This linker/sortase addition strategy was employed for all constructs analyzed in the present study. (B) Size exclusion chromatography (SEC-HPLC) of VSX (red line) and VSX conjugate (blue line), indicating an earlier shift in elution time for the modified construct vs. the starting antibody. The constructs did not differ significantly in the peak width and height, indicating a relative degree of homogeneity in the sortase modification. (C) In vitro killing activity of VSX conjugates. P. aeruginosa ATCC strains 27853 or 39324 were treated with VSX conjugates with a DAR of ~4 containing AMP peptides 271, 293, 294, 295, or 297. Conjugate IgG concentrations that resulted in 50% killing are recorded. (D) VSX-1 (VSX with peptide P297 and a DAR of ~4) was characterized in several assays, including in vitro bactericidal activity, opsonic activity, hemolytic activity, and cytotoxic activity.
Fig 4.
Synergy of P297 with various antibiotic classes.
MICs for both meropenem and colistin were determined alone and then in the presence of varying concentrations of P297, in order to measure the effect on the MIC for P. aeruginosa ATCC 27853. (A) Examination of the MIC for meropenem in the presence of various concentrations of P297. (B) As (A), except that colistin was used as the antibiotic. Not all antibiotics demonstrated synergy; for example, the aminoglycoside tobramycin did not exhibit an enhanced MIC in the presence of P297.
Fig 5.
Evaluation of conjugates in in vivo models of P. aeruginosa lung infection.
(A) Neutropenic animals were co-administered VSX-1 and bacteria (ATCC 27853). The CFU burden in the lung was measured at eight hours. Co-administration of 10 μg of VSX-1 resulted in a multi-log reduction in bacterial burden, with reduction to the limit of detection upon administering 200 μg of ADC. (B) Neutropenic animals were infected with P. aeruginosa (ATCC 27853) and were treated 1 hour post-infection with either vehicle or VSX-1 (200 μg). The CFU burden in the lungs was measured just before treatment (pre-treatment) and eight hours post-injection. (C) The acute lung infection model with P. aeruginosa PA14 (2x106 CFU/animal, 1x106 in each nostril), C57/Bl6, 10 animals per group (two experiments with 5 animal per group each time), intranasal inoculation, IP dosing (15 mg/kg) 4 hours post-infection.
Fig 6.
(A) In vivo imaging of VSX-1’s biodistribution. Labeled VSX (the antibody alone) or VSX-1 (the ADC) was administered to animals IP and monitored by imaging up until 48 hours post-injection (h.p.i.). Both agents distributed throughout the animal (i.e. to all perfused organs) but VSX-1 had a shorter elimination time, with little remaining agent at 48 h.p.i. (B) Analysis of serum levels (using an ELISA) of VSX-1 (DAR = 4: grey bars) at 1, 24, and 72 hours post-injection, and DAR = 2 constructs (black bars). (C) The same imaging procedure as in (A); the ADC with a DAR = 2 demonstrates better biodistribution and a longer half-life.
Fig 7.
In vitro and in vivo assessments of VSX-2.
(A) Activity of VSX-2 compared with VSX-1 with regard to bacterial killing, red blood cell hemolysis (mean lytic concentration (MLC)) and toxicity against mammalian cells (CC50). The bactericidal activity of VSX-2 is similar to (but slightly lower than) that of VSX-1, with a lower DAR, and with a similar inability to lyse RBCs or kill mammalian cells. (B) VSX-2 has similar activity in vivo in the acute lung infection model with P. aeruginosa PA14 (2 x 106 CFUs/animal, 1 x 106 CFUs in each nostril), C57/Bl6, 25 animals per group (five experiments with five animals per group each time), intranasal inoculation, and IP dosing (15 mg/kg) 4 hours post-infection.
Fig 8.
Prevention and treatment of P. aeruginosa PA14 biofilm formation in a dynamic model labelled with Syto9 (green) and PI (red) fluorochromes.
Fig 8: Effect of preventive and curative treatments of P. aeruginosa PA14 biofilm in a dynamic model. (A-B). The live/dead bacteria within the biofilm were assessed with Syto9 (green = live) and propidium iodide (red = dead) fluorochromes. Fluorescence was measured at 47 h (top), 30 h (middle) and 24 h (bottom) as readouts to 3D-reconstruct the biofilm and determine its thickness in the control condition (A) and in the treatment condition with 6 μg/mL of VSX-2 (DAR2) for 1 h (B). (C) For the prevention of biofilm formation, fluorescence was measured after the injection of 6 μg/mL of DAR2 (VSX-2) during bacterial inoculation at time 0. (D) Fluorescence volume of the biofilm biomass, with the distribution of Syto9 and PI staining; scale bar = 100 μm. Fluorescent microscopy images of two adjacent fields from one sample, and calculation using Imaris software. The experiments were repeated three times.