Abstract

   Sonodynamic therapy (SDT) has good feasibility to deeply seated
   infections, but SDT alone is insufficient being highly effective
   against multidrug‐resistant (MDR) bacteria. SDT combined with
   triphenylphosphanium bromide (P^+Ph[3]Br^−) is expected to solve this
   problem. This work develops a pseudo‐conjugated polymer P^FCPS‐P
   containing cationic P^+Ph[3]Br^−‐modified sonosensitizer FCPS (FCPS‐P)
   and ROS‐sensitive thioketal bonds. P^FCPS‐P is assembled with
   DSPE‐mPEG[2000] to generate nanoparticle NP^FCPS‐P. FCPS has SDT effect
   and generates ROS under ultrasound (US) stimulation. ROS triggers the
   degradation of NP^FCPS‐P and release of FCPS‐P, endowing highly favored
   biosafety. FCPS‐P targets to bacterial surface through electrostatic
   interaction and achieves bacterial killing under a synergistic action
   of SDT and P^+Ph[3]Br^−. In vitro, NP^FCPS‐P+US gives >90% inhibition
   rates against MDR ESKAPE pathogens, moreover, it causes bacterial
   metabolic disorders including inhibited nucleic acid synthesis,
   disordered energy metabolism, excessive oxidative stress, and
   suppressed biofilm formation and virulence. In mice, NP^FCPS‐P+US
   exhibits a 99.3% bactericidal rate in Pseudomonas aeruginosa‐induced
   sublethal pneumonia and renders a 90% animal survival rate in lethal
   pneumonia, and additionally immunological staining and transcriptomics
   analyses reveal that NP^FCPS‐P+US induces inhibited inflammatory
   response and accelerated lung injury repair. Taken together,
   NP^FCPS‐P+US is a promising antibiotics‐alternative strategy for
   treating deeply seated bacterial infections.

   Keywords: bacterial pneumonia, cationic triphenylphosphanium bromide,
   ESKAPE pathogens, pseudo‐conjugated polymer, sonodynamic therapy
     __________________________________________________________________

   Bacterial pneumonia is the sixth leading cause of death. In this study,
   an inhalable polymeric nanoparticle NP^FCPS‐P is developed, which under
   US stimulation showed high efficacy in the treatment of acute bacterial
   pneumonia. This nanoplatform would be much less prone to select
   bacterial resistance and represented a promising measure for
   antibiotics‐alternative therapy of deeply seated infections.

   graphic file with name ADVS-12-2417469-g008.jpg

1. Introduction

   Bacterial pneumonia is the sixth leading cause of death and the only
   infectious disease in the top ten causes of death.^[ [36]^1 ^] ESKAPE
   pathogens contain the following six pathogenic bacteria Enterococcus
   faecium (Ec. faecium), Staphylococcus aureus (S. aureus), Klebsiella
   pneumoniae (K. pneumoniae), Acinetobacter baumannii (A. baumannii),
   Pseudomonas aeruginosa (P. aeruginasa), and Enterobacteriaceae species
   such as Eb. Hormaechei,^[ [37]^2 ^] and they account for about
   two‐thirds of the total nosocomial infections worldwide
   ([38]https://microbenotes.com/eskape‐pathogens‐antimicrobial‐resistance
   ). S. aureus, P. aeruginasa, K. pneumoniae, A. baumannii represent the
   most frequent pathogens of hospital‐acquired bacterial pneumonia), ^[
   [39]^3 ^] which is among the most common nosocomial infections.^[
   [40]^4 ^] Most ESKAPE isolates are multidrug resistant (MDR) and
   capable of escaping bactericidal actions of traditional antibiotics,
   which poses a risk of the ending of Antibiotic Era and urgently
   prioritizes in need of developing antibiotics‐alternative therapeutic
   strategies^[ [41]^5 ^] such as sonodynamic therapy (SDT).^[ [42]^6 ^]

   As a prerequisite for SDT, sonosensitizer is activated by ultrasound
   (US) to produce reactive oxygen species (ROS) through cavitation effect
   and sonoluminescence effect.^[ [43]^6 ^] By at least causing protein
   denaturation and DNA damage, the generated ROS is the main reason for
   SDT‐based bacterial killing; meanwhile, ultrasonic cavitation endows
   sonomechanical damage and increased surrounding temperatures acting as
   additional bactericidal mechanisms.^[ [44]^6 ^] SDT generally uses
   low‐intensity focused US (0.5–3.0 W cm^−2, 1.0–2.0 MHz), which has
   minor tissue damage and higher biosafety compared to high‐intensity
   focused US.^[ [45]^7 ^] US takes advantage of superior tissue
   penetrability, enabling SDT to have great feasibility to deeply seated
   infections including pneumonia.^[ [46]^6 ^]

   Organic sonosensitizers exhibit good biodegradability and biosafety,
   but they often suffer from poor water solubility and short blood
   circulation.^[ [47]^9 ^] Liposome or polymer nanocarriers encapsulated
   with organic sonosensitizers would enhance their controlled release
   property and pharmacokinetic profile in blood circulation.^[ [48]^9 ^]
   In addition, sonosensitizers are needed to enhance their targeting
   ability to bacterial cells over mammalian cells.^[ [49]^10 ^] Through
   electrostatic interaction, cationically charged organic sonosensitizers
   would increase their selectively targeting affinity to bacteria, which
   are typically more anionic than mammalian cells.^[ [50]^10 ^]

   Inspired by the above principles, we designed a biodegradable
   pseudo‐conjugated polymer P^FCPS‐P composed of ROS‐sensitive thioketal
   bonds and FCPS‐P, for which a polymer sonosensitizer FCPS was
   chemically bonded with cationic triphenylphosphanium bromide
   (P^+Ph[3]Br^−) at its side chain (Scheme [51]1 ). P^FCPS‐P was
   assembled with commercially available DSPE‐mPEG[2000] to generate the
   nanoparticle NP^FCPS‐P, which could be delivered into the lungs via
   aerosolized intratracheal inoculation. NP^FCPS‐P under US stimulation
   (NP^FCPS‐P+US) produced high levels of ROS owing to FCPS‐mediated SDT
   activity. The action of the generated ROS on ROS‐sensitive thioketal
   bonds in the polymers resulted in the degradation of NP^FCPS‐P and the
   release of FCPS‐P. This ROS‐responsive biodegradability granted the
   highly favorable biosafety of NP^FCPS‐P at both cellular and animal
   levels. The cationic charge of FCPS‐P endowed the targeting
   accumulation of FCPS‐P on the negatively charged cell surface of
   bacteria, and moreover, the FCPS‐induced SDT worked in synergy with
   P^+Ph[3]Br^− (known as a classical antimicrobial agent to disrupt
   bacterial cell membrane^[ [52]^11 ^]) to afford the enhanced
   antibacterial efficacy. NP^FCPS‐P+US gave >90% inhibition rates against
   ESKAPE pathogens in vitro. NP^FCPS‐P+US exhibited a 99.3% bactericidal
   rate in a mouse model of P. aeruginosa‐induced sublethal pneumonia and
   moreover rendered a 90% mouse survival rate in the corresponding lethal
   pneumonia model. NP^FCPS‐P+US represented a promising
   antibiotics‐alternative strategy for the treatment of deeply seated
   infections caused by MDR bacteria.

Scheme 1.

   Scheme 1
   [53]Open in a new tab

   Schematic illustration of synthesis and bactericidal action of
   NP^FCPS‐P. a) The polymer P^FCPS was synthesized from the four monomers
   M1, M2, M3, and M4, among which the first three made up the polymer
   sonosensitizer FCPS while the last one harbored the ROS‐sensitive
   thioketal bond. Furthermore, P^FCPS‐P was obtained by conjugating
   P^+Ph[3]Br^− to M2 in FCPS. Here, FCPS‐P manifested as
   P^+Ph[3]Br^−‐modified sonosensitizer FCPS. b) P^FCPS‐P and
   DSPE‐mPEG[2000] were self‐assembled together into the nanoparticle
   NP^FCPS‐P. c) NP^FCPS‐P was delivered in the lung of mouse via
   aerosolized intratracheal inoculation. NP^FCPS‐P under US stimulation
   (NP^FCPS‐P+US) produced high levels of ROS owing to FCPS‐mediated SDT
   activity. The generated ROS led to the breakage of ROS‐sensitive
   thioketal bonds in the polymers, resulting in the degradation of
   NP^FCPS‐P and the release of FCPS‐P. This ROS‐responsive degradability
   endowed the highly favorable biosafety of NP^FCPS‐P. The electrostatic
   interaction between cationic FCPS‐P and negatively charged bacteria
   surfaces would enhance the targeted attack on bacteria. SDT along with
   P^+Ph[3]Br^− exhibited a synergistic effect for anti‐infection.
   NP^FCPS‐P+US represented a highly biocompatible nanoplatform for
   antibiotics‐alternative treatment of acute bacterial pneumonia.

2. Result and Discussion

2.1. Synthesis and Characterizations of NP^FCPS‐P

   There were totally three polymers (P^FCPS, P^FCPS‐P, and commercially
   available DSPE‐mPEG[2000]) and two nanoparticles (NP^FCPS and
   NP^FCPS‐P) as described in Scheme [54]1 and Figure [55]S1 (Supporting
   Information). P^FCPS was synthesized from four monomers M1, M2, M3, and
   M4 via a so‐called Stille reaction (Scheme [56]1a). M1, M2, and M3
   constituted the polymer sonosensitizer FCPS with a molecular weight of
   14 KDa. M1 acted to connect M2 and M3. M2 could expand the conjugation
   system, reduce the electronic cloud density and the orbital energy
   level, and facilitate the electronic transition of P^FCPS, thereby
   enhancing the sonodynamic effect of P^FCPS. M3, as a classic structural
   monomer of a sonosensitizer, inherently possesses significant
   sonodynamic efficacy. M4 contained the ROS‐sensitive thioketal bond,
   rendering P^FCPS biodegradable. P^+Ph[3]Br^− was bonded to the side
   chain of FCPS to form FCPS‐P with a molecular weight of 15 KDa, and
   thus FCPS‐P contained cationic triphenylphosphanium bromide known as a
   classical antimicrobial agent^[ [57]^11 ^] and was sill recognized as a
   polymer sonosensitizer.

   The successful synthesis of P^FCPS and P^FCPS‐P was confirmed by ^1H
   NMR (Figures [58]S1, S2, Supporting Information). In the ^1H NMR
   spectrum of P^FCPS, the peaks observed at ≈0.8 and 1.3 ppm can be
   assigned to the alkane chains of M1 and M2, respectively. The peak near
   4.04 ppm corresponds to the methyl group of M3, while the peak ≈1.5 ppm
   is attributed to the methyl group of M4. Additionally, the peaks in the
   range of 7.0 to 8.0 ppm are associated with the protons of the benzene
   ring in M3 and the protons of the thiophene ring in M1. In contrast to
   P^FCPS, the ^1H NMR spectrum of P^FCPS‐P shows an additional peak at
   7.3 ppm, which is attributed to triphenylphosphine. This result
   indicated that triphenylphosphine has been successfully grafted onto
   the side chain of P^FCPS.

   The hydrophobic polymer P^FCPS‐P or P^FCPS in the absence of
   DSPE‐mPEG[2000] would aggregate in water and could not be assembled
   into nanoparticles. In order to improve the water solubility of
   P^FCPS‐P or P^FCPS and meanwhile to reduce the side effect of
   P^+Ph[3]Br^−, the amphiphilic DSPE‐mPEG[2000] was used to encapsulate
   P^FCPS‐P or P^FCPS to obtain NP^FCPS‐P (Scheme [59]1b) or NP^FCPS
   (Scheme [60]S1, Supporting Information) as the micelle, respectively.
   P^FCPS‐P or P^FCPS was located at the core of NP^FCPS‐P or NP^FCPS,
   respectively. As observed by transmission electron microscopy (TEM) and
   scanning electron microscopy (SEM), NP^FCPS‐P (Figure [61] 1a,b and
   Figure [62]S3, Supporting Information) or NP^FCPS (Figure [63]S4,
   Supporting Information) presented uniform and complete spherical shape
   with an average particle size of 118 or 117 nm, respectively. Energy
   dispersion‐X‐ray spectroscopic scanning (EDXS) confirmed the presence
   of elements C, N, O, S, and P in NP^FCPS‐P (Figure [64]1c) and P^FCPS
   (Figure [65]S5, Supporting Information). As shown by UV–vis absorption
   spectrum, NP^FCPS‐P and NP^FCPS gave almost the same absorption curve
   with the maximum absorption peak at ≈660 nm (Figure [66]1d). These
   demonstrated the successful assembly of NP^FCPS‐P and NP^FCPS.

Figure 1.

   Figure 1
   [67]Open in a new tab

   Characterization of NP^FCPS‐P. a) TEM images of NP^FCPS‐P. Scale bar =
   100 nm. b) SEM images of NP^FCPS‐P. Scale bar = 100 nm. c) STEM and
   EDXS images of NP^FCPS‐P. Scale bar = 50 nm. d) UV–vis‐NIR absorption
   spectra of NP^FCPS‐P and NP^FCPS. e) Absorption curves of DPBF as a
   selective trapping agent to detect ROS in the reaction mixture of
   NP^FCPS‐P under US stimulation (1.0 MHz, 1.5 W cm^−2, 50% duty cycle).
   f) Trend of UV absorption at 410 nm for the above DPBF‐based detection
   results. g) ESR spectra demonstrating ^1O[2] generation of NP^FCPS‐P or
   NP^FCPS with US stimulation (1.5 W cm^−2, 1.0 MHz, 50% duty cycle,
   1 min). h) Absorption curves of DPA as a selective trapping agent to
   detect ^1O[2] in the reaction mixture of NP^FCPS‐P under US stimulation
   (1.0 MHz, 1.5 W cm^−2, 50% duty cycle). i) Trend of UV absorption at
   378 nm for the above DPA‐based detection results. DLS detection of j)
   size distributions of NP^FCPS‐P with or without US stimulation
   (1.0 MHz, 1.5 W cm^−2, 50% duty cycle, 6 min).

   P^+Ph[3]Br^− alone could not be encapsulated with DSPE‐mPEG[2000] into
   a nanoparticle. Each of M1, M2 and M3 could be packaged with
   DSPE‐mPEG[2000] to obtain NP^M1, NP^M2, and NP^M3, respectively, and
   their SDT‐mediated ROS production ability was determined as below; as
   expected, the absorbance of NP^M3 was decreased by 24%, which was
   higher than the 11%, 14%, and 11% decrease values of NP^M1, NP^M2,
   1,3‐diphenylisobenzofuran (DPBF) blank control, respectively (Figure
   [68]S6, Supporting Information). These results confirmed M3 as a core
   determinant of sonodynamic effect.

   To further validate the sonodynamic effect of NP^FCPS‐P or NP^FCPS,
   2 mm DPBF as the ROS probe was mixed with 20 µg mL^−1 NP^FCPS‐P or
   NP^FCPS followed by US stimulation for 0 to 9 min. As determined by a
   multimode reader, the absorption peak of NP^FCPS‐P at 410 nm was
   decreased rapidly under US and the peak intensities was gradually
   decreased as the prolonging of US stimulation times (Figure [69]1e);
   after 9 min of US stimulation, the absorbance of NP^FCPS‐P decreased by
   69%, which was higher than the 34% decrease of DPBF blank control
   (Figure [70]1f). NP^FCPS showed similar results (Figure [71]S7,
   Supporting Information). These proved that NP^FCPS‐P and NP^FCPS had
   almost the same SDT‐mediated ROS production ability.

   The ability of NP^FCPS‐P or NP^FCPS to produce ^1O[2] and •OH under US
   stimulation was systematically characterized by a series of electron
   spin resonance (ESR) experiments. First, 3 µg mL^−1
   2,2,6,6‐tetramethylpiperidine (TEMP) and 10 µg mL^−1
   5,5‐dimethyl‐1‐pyrroline‐N‐oxide (DMPO), as the capture agents for
   ^1O[2] and •OH, respectively, were mixed with 100 µg mL^−1 of NP^FCPS‐P
   or NP^FCPS followed by US stimulation for 1 min. NP^FCPS‐P+US produced
   a strong ^1O[2] signal peak (Figure [72]1g), while the signal peak of
   •OH was not detected (Figure [73]S8, Supporting Information).
   NP^FCPS+US showed similar results (Figure [74]1g and Figure [75]S8,
   Supporting Information). These disclosed that NP^FCPS‐P and NP^FCPS
   produced ^1O[2] but not •OH upon US stimulation. Second, 1 mg mL^−1
   9,10‐dibenzoanthraquinone (DPA) as the ^1O[2]‐specific probe was mixed
   with 20 µg mL^−1 NP^FCPS‐P or NP^FCPS, and the peak intensity at 378 nm
   was detected after US stimulation for 0 to 5 min to verify the effect
   of ^1O[2] production. The absorption peak of NP^FCPS‐P (Figure [76]1h)
   was decreased rapidly under US and the peak intensities at 378 nm were
   gradually decreased with the prolonging of US stimulation times; after
   5 min of US stimulation, the absorbance of NP^FCPS‐P decreased by 59%,
   which was higher than the 11% decrease of DPA blank control
   (Figure [77]1i). NP^FCPS+US showed similar results (Figure [78]S9,
   Supporting Information). These demonstrated that the production of
   ^1O[2] by NP^FCPS‐P+US and NP^FCPS+US was time‐dependent. Third, 10 µg
   mL^−1 methylene blue (MB) as the •OH‐specific capture agent was mixed
   with 20 µg mL^−1 NP^FCPS‐P or NP^FCPS, and the effect of •OH production
   was characterized by detecting its peak intensity at 664 nm after US
   stimulation for 0 to 5 min. For both NP^FCPS‐P and NP^FCPS, the peak
   intensity at 664 nm showed no obvious change with the prolonging of US
   stimulation times (Figure [79]S10, Supporting Information), confirming
   that there was no •OH production as detected by ESR above.
   Collectively, NP^FCPS‐P and NP^FCPS under US stimulation could produce
   ^1O[2] in a time‐dependent manner but they did not generate •OH. This
   is attributed to the ability of the sonosensitizer to convert O[2] to
   ROS after absorbing the energy of ultrasound.

   The sensitivity of NP^FCPS‐P or NP^FCPS to ROS was examined in vitro.
   First, as shown by gel permeation chromatography (GPC) curves, the
   molecular weight of P^FCPS‐P (3 mg mL^−1) was gradually decreased from
   14.7 into 1.5 KDa after co‐incubation with 10 mm H[2]O[2] from 0 to
   8 h, indicating that the ROS‐sensitive group of P^FCPS‐P was broken
   under the action of H[2]O[2] (Figure [80]S11a, Supporting Information).
   P^FCPS‐P showed similar results (Figure [81]S11b, Supporting
   Information). Second, as observed by TEM and SEM, NP^FCPS‐P (Figure
   [82]S12a,b, Supporting Information) and NP^FCPS (Figure [83]S12c,d,
   Supporting Information) after US stimulation for 6 min became smaller
   in size but still had uniform spherical shapes. Third, as measured by
   dynamic light scattering (DLS), the average diameter of NP^FCPS‐P after
   US stimulation for 6 min was changed from 115.9 to 53.65 nm
   (Figure [84]1j), while its polydispersity index (PDI) had no obvious
   change (Figure [85]S13a, Supporting Information), and its average Zeta
   potential was changed from ‐24.1 to ‐33.2 mV (Figure [86]S13b,
   Supporting Information). NP^FCPS gave similar DLS results (Figure
   [87]S14, Supporting Information). Collectively, US treatment of
   NP^FCPS‐P or NP^FCPS would lead to the degradation of NP^FCPS‐P or
   NP^FCPS, which would include the breakage of thioketal bonds, the
   release of FCPS‐P or FCPS, respectively, and the dispersing of
   DSPE‐mPEG[2000] in the system (Figures [88]S13, [89]S14, Supporting
   Information). However, after the termination of US stimulation, the
   dispersed DSPE‐mPEG[2000] would re‐encapsulate the released FCPS‐P or
   FCPS into new nanoparticles with smaller sizes and lower negative
   charges (Figures [90]S13, [91]S14, Supporting Information). To verify
   this re‐encapsulation, P^FCPS‐P or P^FCPS was pre‐treated by 10 mm
   H[2]O[2] for 24 h to respectively obtain the released FCPS‐P or FCPS in
   the system; subsequently, DSPE‐mPEG[2000] was used to repackage each
   H[2]O[2]‐treated system, forming NP^FCPS‐P+H2O2 or NP^FCPS+H2O2,
   respectively; as expected, the characterization results (Figure
   [92]S15, Supporting Information) showed that the particle size and
   potential of this repackaged NP^FCPS‐P+H2O2 or NP^FCPS+H2O2 were
   comparable to the above US‐treated NP^FCPS‐P or NP^FCPS, respectively.
   Taken together, NP^FCPS‐P possessed excellent sonodynamic activity to
   generate abundant ^1O[2] upon US stimulation, and the generated ROS
   promoted the degradation of NP^FCPS‐P and the release of FCPS‐P.

2.2. Antibacterial Efficacy of NP^FCPS‐P Against MDR ESKAPE Pathogens in
vitro

   The clinical MDR ESKAPE isolates, including Ec. faecium HJP554,^[
   [93]^12 ^] S. aureus USA300‐FPR3757,^[ [94]^12 ^] K. pneumoniae ATCC
   BAA‐2146,^[ [95]^13 ^] A. baumannii LAC‐4,^[ [96]^14 ^] P. aeruginosa
   F291007,^[ [97]^12 ^] and Eb. hormaechei ATCC BAA‐2082
   ([98]https://www.atcc.org/products/baa‐2082) were subjected to
   antibacterial experiments in vitro by setting a total of 6 treatment
   groups namely Mock, US, NP^FCPS, NP^FCPS+US, NP^FCPS‐P, and
   NP^FCPS‐P+US (Figure [99]2 ). As determined by plate counting method
   (Figure [100]2a to [101]2f), the three control groups Mock, NP^FCPS,
   and NP^FCPS‐P had no obvious inhibitory activity against all ESKAPE
   bacteria, denoting that the nanoparticles alone had no bactericidal
   effect. The NP^FCPS+US group (SDT alone) showed higher bactericidal
   activity than the US group. Compared to all the other groups, the
   NP^FCPS‐P+US group (SDT plus P^+Ph[3]Br^−) gave the highest inhibition
   rates (96.07%, 91.79%, 90.05%, 98.76%, 90.93% and 96.85%, respectively)
   against E. faecium, S. aureus, K. pneumoniae, A. baumannii, P.
   aeruginosa and Eb. hormaechei. It should be noted that higher
   irradiances (W cm^−2) and longer stimulation times were needed to
   achieve >90% inhibition rates against Gram‐positive bacteria compared
   to Gram‐negative ones, most likely because Gram‐positive bacteria had
   thicker cell walls and thus were more tolerant to US. P. aeruginosa and
   S. aureus were chosen as the representatives of Gram‐negative and
   Gram‐positive ESKAPE bacteria, respectively, to explore the synergistic
   antibacterial effect of SDT and P^+Ph[3]Br^−. Based on the inhibition
   rates, the synergy factor Q values (calculation formula and evaluation
   criteria shown in Supporting Information) were calculated at 1.34 and
   1.16 for these two bacteria, respectively. These two calculated Q
   values were >1.15, a cutoff indicator of “strong synergy”.^[ [102]^15
   ^] Furthermore, live/dead dual fluorescent dye staining, for which the
   green dye SYTO‐9 and the red dye propidium iodide (PI) differentially
   stained live and dead bacteria, respectively, were performed for P.
   aeruginosa and S. aureus post‐treatment (Figure [103]2g). A large
   number of dead bacterial cells and a small number of survived cells
   were observed in the NP^FCPS+US group, while almost all bacteria died
   in the NP^FCPS‐P+US group, reconfirming the synergy. In conclusion,
   under the synergistic action of SDT and P^+Ph[3]Br^−, NP^FCPS‐P+US in
   vitro gave the best and highly effective bactericidal action against
   MDR ESKAPE pathogens especially Gram‐negative bacteria.

Figure 2.

   Figure 2
   [104]Open in a new tab

   Antibacterial efficacy against ESKAPE pathogens in vitro. US
   stimulation: 2.0 W cm^−2, 1.0 MHz, 50% duty cycle, 15 min for
   Gram‐positive bacteria; 1.5 W cm^−2, 1.0 MHz, 50% duty cycle, 6 min for
   Gram‐negative bacteria. a–f) Representative photographs of plate count
   agars for ESKAPE pathogens post‐treatment, and corresponding
   statistical analysis of bacterial counts. Data are presented as mean ±
   standard deviation (SD) (n = 3). ns (not significant): P ≥ 0.05, ^*: P
   < 0.05, ^**: P < 0.01, ^***: P < 0.001, one‐way analysis of variance
   (ANOVA) with Tukey's multiple comparisons test. g) Representative
   images (n = 3) of live/dead bacteria post‐treatment as viewed by
   confocal laser scanning microscopy (CLSM). Red and green fluorescence
   signals correspond to dead and live bacterial cells, respectively.
   Scale bar = 20 µm.

2.3. Mechanisms of Bactericidal Action of NP^FCPS‐P in vitro

   Cell membrane damages and destructive biochemical changes of P.
   aeruginosa and S. aureus post‐treatment with the above 6 groups were
   explored (Figure [105]3 ). First, bacterial morphological shrinkage and
   concavity were observed by SEM (Figure [106]3a) and TEM (Figure
   [107]S16, Supporting Information). Second, leakage of bacterial
   cytoplasmic contents, which were mainly attributed to cell membrane
   damages,^[ [108]^15 ^] was examined by detecting DNA, protein, and K^+
   contents in the supernatants of bacterial cultures (Figure [109]3b‐d).
   GSH levels in bacterial cultures were detected by the Reduced GSH
   Content Assay Kit (Figure [110]3e), for which the detected decrease in
   GSH corresponded to weaken ROS defense along with increased oxidative
   stress in bacterial cells.^[ [111]^15 ^] Third, DNAs of bacterial cell
   nucleus were stained with the blue fluorescence dye
   4′,6‐diamidino‐2‐phenylindole (DAPI) (Figure [112]3f), which was used
   as the negative indicator of bacterial DNA damage.^[ [113]^15 ^]
   Fourth, ROS levels were detected by using the ROS‐specific green
   fluorescent probe 2′,7′‐dichlorofluorescein diacetate (DCFH‐DA)
   (Figure [114]3g and Figure [115]S17, Supporting Information), which was
   used as the positive indicator of SDT‐based ROS generation levels in
   bacterial cultures.^[ [116]^15 ^] As shown by the above 4 different
   assays, a total of 5 detection indexes (namely bacterial morphological
   shrinkage/concavity, leakage of bacterial cytoplasmic contents,
   bacterial DNA damage, elevated ROS level in bacterial culture, and
   decreased GSH level in bacterial culture) were positively detected for
   the 3 treatment groups NP^FCPS‐P+US, NP^FCPS+US, and US, and moreover
   all these 5 indexes gave the same statistically different trend:
   NP^FCPS‐P+US > NP^FCPS +US > US; by contrast, no obvious positive
   results were detected for all the other treatment groups.

Figure 3.

   Figure 3
   [117]Open in a new tab

   Bacterial morphological and biochemical view of antibacterial
   mechanisms in vitro. US stimulation: 2.0 W cm^−2, 1.0 MHz, 50% duty
   cycle, 15 min for Gram‐positive bacteria; 1.5 W cm^−2, 1.0 MHz, 50%
   duty cycle, 6 min for Gram‐negative bacteria. a) SEM images of P.
   aeruginosa and S. aureus post‐treatment. Red arrows represent typical
   disruptions of bacterial morphology. Scale bar = 1 µm. b) DNA, c)
   Protein, d) K^+ and e) GSH contents in supernatants of P. aeruginosa
   and S. aureus cultures post‐treatment. f) CLSM images of DAPI‐stained
   bacterial DNAs post‐treatment. Scale bar = 1 µm. g) Fluorescence images
   of P. aeruginosa and S. aureus post‐treatment indicating the generation
   of ROS. The scale bar is 100 µm. Data are presented as mean ± SD (n =
   3). ns: P ≥ 0.05, ^*: P < 0.05, ^**: P < 0.01, ^***: P < 0.001, one‐way
   ANOVA with Tukey's multiple comparisons test.

   The metabolomes of P. aeruginosa were compared between the NP^FCPS‐P+US
   group and the Mock group (Figure [118]4 ). First, as disclosed by
   principal component analysis (PCA), the samples in each treatment group
   were densely distributed, indicating the existence of limited variation
   within each group; by contrast, the two groups of data fell into two
   clusters far apart from each, denoting a high level of difference
   between the two groups (Figure [119]4a). Second, as shown in the
   Volcano plot (Figure [120]4b) and histogram (Figure [121]S18,
   Supporting Information), a total of 2257 differential metabolites
   (DRMs) were identified for NP^FCPS‐P+US relative to Mock, of which 931
   were down‐regulated and 1326 were up‐regulated. Third, KEGG (Kyoto
   Encyclopedia of Genes and Genomes) analysis (Figure [122]4c) enriched
   11 down‐regulated and 3 up‐regulated pathways, overall illustrating how
   NP^FCPS‐P+US induced the obstruction of bacterial virulence, growth and
   proliferation. Fourth, 10 up‐regulated and 13 down‐regulated
   metabolites were arbitrarily selected as the representative DRMs for
   clustering into a heat map (Figure [123]4d), and moreover, these 23
   selected DRMs could be assigned into a dysregulated metabolic pathway
   map (Figure [124]4e). cAMP was down‐regulated, which would affect the
   activity of Vfr as a master positive regulator of biofilm formation and
   virulence factors in P. aeruginosa.^[ [125]^16 ^] Up‐regulation of
   L‐glutamate, glutathione, and NADP^+ indicated the enhancement of GSH
   biosynthesis, which would be a compensatory increase of GSH for
   resistance to SDT‐generated ROS.^[ [126]^17 ^] Up‐regulation of
   D‐gluconate‐1,5‐lactone, D‐gluconate, and pyruvate denoted the
   elevation of the pentose phosphate pathway (PPP), which would promote
   bacterial tolerance to oxidative stress by providing NADPH.^[ [127]^18
   ^] Down‐regulation of glycerate‐3P would act as a metabolic switch to
   inhibit central carbon pathways such as glycolysis and the TCA cycle,
   redirecting the carbohydrate flux to PPP to produce more NADPH and GSH
   to counteract oxidative stress.^[ [128]^19 ^] In addition,
   up‐regulation of succinate in the TCA cycle was generally recognized as
   a biomarker of oxidative stress.^[ [129]^20 ^] Down‐regulation of
   pyrimidine metabolism‐related 2′,3′‐cyclic UMP, uridine, cytidine,
   dCTP, and UTP would interfere with DNA and mRNA biosynthesis,^[
   [130]^21 ^] while down‐regulation of L‐asparagine, L‐aspartate,
   threonine, and isoleucine would affect multiple tRNA biosynthesis
   pathways.^[ [131]^22 ^] Sixth, through ROC (Receiver Operating
   Characteristic) curve analysis (Figure [132]S19, Supporting
   Information), an AUC (area under ROC curve) value of 0.96 was detected
   for putrescine. This was larger than the cutoff value 0.9.^[ [133]^23
   ^] The up‐regulated putrescine could be used as a biomarker for
   predicting the pathological state of P. aeruginosa treated with SDT
   plus P^+Ph[3]Br^−, which was consistent with our previous observation
   for P. aeruginosa treated with PTT (photothermal therapy) plus
   CDT(chemodynamic therapy).^[ [134]^24 ^] In conclusion, NP^FCPS‐P+US
   leads to suppressed biofilm formation and virulence, excessive ROS
   production and enhanced oxidative stress, inhibited glycolysis and TCA
   cycle and inhibited DNA, mRNA, and tRNA biosynthesis in bacteria.

Figure 4.

   Figure 4
   [135]Open in a new tab

   Metabolomics analysis of P. aeruginosa after treatment. a) PCA plot of
   clustered datasets between the two treatment groups NP^FCPS‐P+US and
   Mock. US represents US stimulation (1.5 W cm^−2, 1.0 MHz, 50% duty
   cycle, 8 min). b) Volcano plot of DRMs identified for NP^FCPS‐P+US
   compared to Mock; Up: Up‐regulated metabolites. Down: Down‐regulated
   metabolites. c) KEGG pathway enrichment analysis of DRMs. d) Heat maps
   of selected DRMs, and e) diagrams of related dysregulated metabolic
   pathways. Red represents up‐regulation and blue stands for
   down‐regulation. Experiments were performed with 5 independent
   biological replicates.

2.4. Biosafety and Therapeutic Feasibility of NP^FCPS‐P

   The biological targeting, biosafety, and therapeutic feasibility of
   NP^FCPS‐P were validated prior to in vivo anti‐infective therapy
   experiments. First, to evaluate the targeting affinity of NP^FCPS‐P to
   bacteria in vitro (Figure [136]S20, Supporting Information), the
   mixture of P. aeruginosa and BEAS‐2B cells was treated by 20 µg mL^−1
   NP^FCPS‐P or NP^FCPS with or without US stimulation (1.5 W cm^−2,
   1.0 MHz, 50% duty cycle, 2 min). In NP^FCPS‐P+US group, FCPS‐P was
   significantly aggregated around P. aeruginosa rather than BEAS‐2B
   cells. After US‐triggered degradation, NP^FCPS‐P released positively
   charged FCPS‐P that would accumulate on the bacterial cell surface
   through positive and negative electric absorption, achieving the
   selectively targeting affinity to bacteria. Second, the cytotoxicity of
   NP^FCPS‐P on mouse embryonic normal fibroblast cell line NIH‐3T3 and
   human ovarian normal epithelial cell line IOSE‐80 was detected by Cell
   Counting Kit‐8 (CCK‐8) assay (Figure [137]S21, Supporting Information);
   when cells were incubated with 100 µg mL^−1 NP^FCPS‐P for 24 h (none of
   US stimulation was added), the cell viability of each cell line reached
   >80%.^[ [138]^12 , [139]^15 , [140]^25 ^] In addition, the cytotoxicity
   of NP^FCPS‐P on NIH‐3T3 or IOSE‐80 cells in the presence of 100 µg
   mL^−1 NP^FCPS‐P together with US stimulation (1.5 W cm^−2, 1.0 MHz, 50%
   duty cycle, 6 min) was also tested (Figure [141]S22, Supporting
   Information), and the results showed that the cell viability of each
   cell line still reached >80%. Third, through hemolysis test, mouse red
   blood cells (RBCs) after incubation with 100 µg mL^−1 NP^FCPS‐P for 4 h
   (none of US stimulation was added) gave the hemolysis rate was <5%^[
   [142]^12 , [143]^15 , [144]^25 ^] (Figure [145]S23, Supporting
   Information). In addition, the hemolysis rate of RBCs in the presence
   of 100 µg mL^−1 NP^FCPS‐P together with US stimulation (1.5 W cm^−2,
   1.0 MHz, 50% duty cycle, 6 min) was also measured (Figure [146]S24,
   Supporting Information), and the results showed that the hemolysis rate
   was still <5%. Fourth, IVIS spectral in vivo imaging system (Figure
   [147]S25, Supporting Information) was used to observe the fluorescence
   distributions in infection‐free mice, which were treated with the 6
   groups Mock, US, NP^FCPS, NP^FCPS+US, NP^FCPS‐P, and NP^FCPS‐P+US (for
   which NP^FCPS or NP^FCPS‐P was encapsulated with the red fluorescent
   dye Cy7.5) using the same route and dose as anti‐infective therapy
   experiments. On day 0 post‐treatment, much more uniform fluorescence
   distribution in the lungs was observed for NP^FCPS+US (compared to
   NP^FCPS) or NP^FCPS‐P+US (relative to NP^FCPS‐P), indicating that
   nanoparticles could be accurately delivered into the lungs and
   US‐promoted the dispersion of nanoparticle in the lungs. The
   fluorescence was decreased as time went on, indicating that the
   nanoparticle was gradually metabolized in the body. Fifth,
   infection‐free mice were treated with the above 6 groups, each of which
   was co‐delivered with DAPI as a positive control and DCFH‐DA for the
   detection of ROS (Figure [148]S26, Supporting Information). Compared to
   the other groups, the NP^FCPS+US and NP^FCPS‐P+US groups showed no
   obvious difference in DCFH‐DA green fluorescence but gave much stronger
   green fluorescence compared to the other groups, indicating that ROS
   could be efficiently produced by NP^FCPS or NP^FCPS‐P upon the US in
   the lungs. Sixth, infection‐free mice were treated with the above 6
   groups, along with the lipopolysaccharide (1 mg kg^−1) group as the
   positive control; hematoxylin and eosin (H&E) staining of the tissues
   from heart, liver, spleen, lung, and kidney on day 6 post‐treatment
   showed that there was no inflammation and pathological lesions (Figure
   [149]S27, Supporting Information). For the lipopolysaccharide group,
   significant inflammatory cell infiltration was observed in lung tissues
   but not in any of the other four organs. Seventh, as determined by
   blood biochemical tests on day 6 post‐treatment, the two primary liver
   function indexes alanine aminotransferase (ALT) and aspartate
   aminotransferase (AST), and the two primary renal function indexes
   blood urea nitrogen (BUN) and creatinine (CREA) showed no abnormal
   changes (Figure [150]S28, Supporting Information). In summary,
   NP^FCPS‐P showed the selectively targeting affinity to bacteria in
   vitro, and the highly favored biosafety at both cellular and animal
   levels; US promoted the diffusion of delivered NP^FCPS‐P and its
   CDT‐based ROS generation ability in the lungs.

2.5. Therapeutic Efficacy Against Acute Bacterial Pneumonia in Mice

   P. aeruginosa F291007, belonging to the high‐risk P. aeruginosa ST235
   clone characteristic of MDR and hypervirulence,^[ [151]^26 ^] was used
   as the representative to infect mice. Therapeutic efficacy against
   acute lung infections in mice was evaluated by the programmed
   procedures including i) co‐delivery of P. aeruginosa and nanoparticle
   via aerosolized intratracheal inoculation, ii) establishment of
   bacterial infection, and iii) applying of US stimulation to lunch
   one‐time anti‐infective therapy with the above 6 groups (Figure
   [152]5a), which was followed by iv) sampling at different days
   post‐therapy for subsequent microbiological, immunological, and
   transcriptomic experiments (Figure [153]5b). This co‐delivery strategy
   was employed in our previous study,^[ [154]^1 ^] and its reasonability
   was based on the following three aspects. First, the in vitro
   bactericidal experiment (Figure [155]2) proved that NP^FCPS‐P or
   NP^FCPS in the absence of US stimulation has no killing effect on
   bacteria. US stimulation was applied at 24 h post co‐delivery to lunch
   anti‐infective therapy (Figure [156]5a), and therefore this co‐delivery
   strategy would not affect the establishment of bacterial infection.
   Second, this co‐delivery strategy could make nanoparticles and bacteria
   to be better co‐dispersed at the sites of infection, achieving a stable
   and reliable therapeutic effect upon US stimulation. Third, mice need
   to be anesthetized to achieve lung delivery, and this co‐delivery
   strategy will reduce the number of times of anesthetization, meeting
   the 3R principle of animal research. All animal experiment protocols
   were reviewed and approved by the Institutional Animal Care and Use
   Committee of the Academy of Military Medical Sciences (Permit number:
   IACUC‐DWZX‐2023‐P025).

Figure 5.

   Figure 5
   [157]Open in a new tab

   Therapeutic efficacy against P. aeruginosa‐induced lung infections in
   mice. US stimulation: 1.0 MHz, 1.5 W cm^−2, 50% duty cycle, 2 min. a)
   Schematic illustration of pulmonary delivery, infection establishment,
   and anti‐infective therapy. b) Timing diagram for various tests
   post‐therapy. c) Representative photographs of plate count agars for
   the lungs once infective therapy was applied, and d) corresponding
   statistical analysis of bacterial load counts. Data are presented as
   mean ± SD (n = 6). e) Pathological scores of lung tissues (n = 3) on
   days 0, 2, and 4 post‐therapy. ns: P ≥ 0.05, ^*: P < 0.05, ^**: P
   < 0.01, ^***: P < 0.001, one‐way ANOVA with Tukey's multiple
   comparisons test. f) Representative photographs of H&E staining of the
   lungs. Scale bar = 100 µm. Blue arrows denote inflammatory cells such
   as neutrophils, macrophages, and lymphocytes. Red arrows stand for
   local bleeding. g) Survival curves and h) body weights of infected mice
   post‐therapy. Data are presented as mean ± SD (n = 10). ns: P ≥ 0.05,
   ^*: P < 0.05, ^**: P < 0.01, ^***: P < 0.001, log‐rank Mantel–Cox test.

   A sublethal pneumonia model was established by challenging mice with 2
   × 10^5 colony‐forming units (CFUs), which was the predetermined maximum
   P. aeruginosa dose leading to no death (LD[0]). US stimulation was
   applied on 24 h post bacterial inoculation to launch anti‐infective
   therapy. First, bacterial loads in the lung, liver, and spleen were
   determined by plate counting method once infective therapy was applied
   (Figure [158]5c,d). Abundant bacterial loads were detected in the lungs
   for the three control groups Mock, NP^FCPS, and NP^FCPS‐P, indicating
   the successful establishment of lung infections. By contrast, the
   NP^FCPS‐P+US, NP^FCPS+US, and US groups gave the statistically
   different inhibition rates at 99.3%, 86.8%, and 48.2%, respectively,
   indicating their very quick bactericidal actions. Meanwhile, a very
   small number of bacteria was detected in the liver and spleen,
   indicating that only a small portion of bacteria could enter into the
   bloodstream from the lung; there was statistically significant
   difference in inhibition rates for all those 6 groups, which might be
   due to the presence of quite too few bacteria (Figure [159]S29,
   Supporting Information). Second, H&E staining was performed to observe
   the histopathologic changes in the 5 major organs (lung, heart, liver,
   spleen, and kidney) on days 0, 2, and 4 post‐treatment. On day 0, high
   levels of neutrophil infiltration and local hemorrhage with a
   pathological score of Grade 4 were found in the lung for all the 6
   groups; on day 2 or 4, the NP^FCPS‐P+US group gave the lowest levels of
   inflammatory response and pathological score in the lungs
   (Figure [160]5e,f). On days 0, 2, and 4 post‐therapy, H&E staining of
   the other four organs showed no obvious pathologic lesions (Figure
   [161]S30, Supporting Information), confirming that the infection was
   restricted in the lung and few bacteria could spill into the
   bloodstream but cleared by mouse immune system.

   An additional lethal pneumonia model was established by challenging
   mice with 7 × 10^5 CFUs, which was the predetermined P. aeruginosa
   absolute lethal dose (LD[100]). US‐triggered anti‐infective therapy was
   launched on 2 h post bacterial inoculation. All mice died on day 3
   post‐therapy for the three control groups Mock, NP^FCPS, and NP^FCPS‐P.
   By contrast, the NP^FCPS‐P+US, NP^FCPS+US, and US groups gave
   statistically different survival rates at 90%, 50%, and 20%,
   respectively (Figure [162]5g); meanwhile, the body weight curves gave
   no statistically significant difference between these 3 groups
   (Figure [163]5h).

   Taken together, NP^FCPS‐P+U had the best and most highly effective
   anti‐infective action against P. aeruginosa‐induced sublethal or lethal
   pneumonia in mice, and the therapeutic efficacies gave the trend as
   expected: NP^FCPS‐P+US > NP^FCPS+US > US.

2.6. Mechanisms of Anti‐Infective Action Viewed from Host Response

   Mechanisms of in vivo anti‐infective action were systematically
   explored from the view of immunological and transcriptomic response in
   the lungs post‐therapy of sublethal P. aeruginosa‐induced infection.
   First, to characterize the expression of three major proinflammatory
   cytokines TNF‐α, IL‐6, and IFN‐β, the bronchoalveolar lavage fluids on
   days 0, 1, 2, and 4 post‐therapy were collected for enzyme‐linked
   immunosorbent assay (ELISA) (Figure [164]6a). With the days went on,
   the expression of these cytokines was progressively decreased in the
   NP^FCPS‐P+US group. As for the comparison of different treatment
   groups, day 2 post‐therapy represented the ideal time point, at which
   the NP^FCPS‐P+US group gave the lowest expression of each cytokine and
   moreover the expression levels of all these cytokines displayed the
   statistically different trend as expected: NP^FCPS‐P+US < NP^FCPS+US <
   US. Second, the lung tissues on day 2 post‐therapy were subjected to
   immunohistochemical staining (IHS) for detecting these 3 cytokines
   (Figure [165]6b,c), confirming the ELISA results. Third, the lung
   tissues on day 2 post‐therapy were also used for immunofluorescence
   staining (IFS) to determine the 3 major lung injury repair proteins,
   namely SP‐C (generation of alveolar type II epithelial cells^[ [166]^27
   ^]), α‐SMA (generation of myofibroblasts and deposition of
   extracellular matrix^[ [167]^28 ^]), and VEGF‐A (angiogenesis^[
   [168]^29 ^]) (Figure [169]6d and Figure [170]S31, Supporting
   Information). The NP^FCPS‐P+US group gave the highest expression of
   each protein, with the statistically different trend for the three
   treatment groups: NP^FCPS‐P+US > NP^FCPS+US > US. In conclusion,
   NP^FCPS‐P+US had the best and highly effective action to inhibit the
   inflammatory response and meanwhile promote lung injury repair.

Figure 6.

   Figure 6
   [171]Open in a new tab

   ELISA, IHS, and IFS assays of proinflammatory cytokines and lung injury
   repair proteins. US stimulation: 1.0 MHz, 1.5 W cm^−2, 50% duty cycle,
   2 min. a) Concentrations of proinflammatory cytokines TNF‐α, IL‐6, and
   IFN‐β in bronchoalveolar lavage fluids on days 0, 1, 2, and 4
   post‐therapy. b) Representative IHS photographs and c) corresponding
   statistical analyses for expression levels of TNF‐α, IL‐6, and IFN‐β in
   the lungs on day 2 post‐therapy. Yellow‐brown represents positive
   detection. Scale bar = 100 µm. d) Representative IFS photographs and e)
   corresponding statistical analyses for expression levels of lung injury
   repair proteins SP‐C, α‐SMA, and VEGF‐A on day 2 post‐therapy. Red
   represents positive detection. Data are expressed as mean ± SD (n = 3).
   ns: P ≥ 0.05, ^*: P < 0.05, ^**: P < 0.01, ^***: P < 0.001, one‐way
   ANOVA with Tukey's multiple comparisons test.

   In addition, the RNA‐seq transcriptomes of infected lung tissues on day
   2 post‐therapy were compared between the two treatment groups
   NP^FCPS‐P+US and Mock (Figure [172]7 ). First, the PCA plot
   (Figure [173]7a) showed a high level of data reproducibility within
   each group and a high degree of data difference between the two groups.
   Second, a total of 857 down‐regulated genes (including the above 3
   proinflammatory cytokines) and 307 up‐regulated genes (including the
   above 3 lung injury repair proteins) were identified for NP^FCPS‐P+US
   in comparison to Mock (Figure [174]7b). Third, Gene Ontology (GO)
   pathway enrichment analysis of these DEGs identified not only the 14
   down‐regulated pathways (Figure [175]7c; neutrophil migration,
   neutrophil chemotaxis, phagocytosis, cytokine‐mediated signaling
   pathway, IL‐1 production, TNF production, IL‐6 production, regulation
   of innate immune response, cellular response to lipopolysaccharide,
   defense response to bacterium, myeloid cell differentiation, antigen
   processing and presentation, Toll‐like receptor signaling pathway, and
   canonical NF‐κB signal transduction), but also the 14 up‐regulated
   pathways (Figure [176]7d; regulation of epithelial cell proliferation,
   VEGF signaling pathway, endothelial cell proliferation, regulation of
   angiogenesis, multivesicular body, smooth muscle cell proliferation,
   positive regulation of cell activation, endothelial cell
   differentiation, muscle cell proliferation, blood vessel endothelial
   cell migration, epithelial cell migration, collagen‐containing
   extracellular matrix, regulation of neurogenesis, and FGFR signaling
   pathway). Notably, all these enriched down‐regulated pathways
   contributed to the resolution of hyperinflammation, whereas all these
   enriched up‐regulated pathways were responsible for the promotion of
   lung injury repair, providing a global perspective on clearance of
   bacterial infection along with restoration of damaged lung. Fourth, as
   displayed by the heat map (Figure [177]7e), an array of 20 key DEGs
   were arbitrarily selected from the above‐enriched pathways, and they
   would represent the principal host‐responsive contributors to endow the
   treatment outcome of NP^FCPS‐P+US. In conclusion, as viewed from the
   host immunological and transcriptomic response, the excellent
   therapeutic effect of NP^FCPS‐P+US depended largely on the suppression
   of inflammatory response and simultaneously the acceleration of lung
   injury repair.

Figure 7.

   Figure 7
   [178]Open in a new tab

   RNA‐seq assay of infected lung tissues on day 2 post‐therapy. a) PCA
   plot of clustered datasets between the two treatment groups
   NP^FCPS‐P+US and Mock. US stimulation: 1.0 MHz, 1.5 W cm^−2, 50% duty
   cycle, 2 min. b) Volcano plot of DGEs for NP^FCPS‐P+US compared to
   Mock. Up: Up‐regulated genes. Down: Down‐regulated genes. Non‐SIG:
   non‐significantly differentially regulated genes. GO pathway enrichment
   analysis was performed for c) Down‐regulated and d) up‐regulated genes.
   e) Heat map of selected DRGs involved in inhibited inflammatory
   response and accelerated lung injury repair. Boxes indicate the 6 DRGs
   revealed simultaneously by IHS and RNA‐seq. Experiments were performed
   with 4 independent biological replicates.

3. Conclusion

   In summary, we developed an inhalable polymeric nanoparticle NP^FCPS‐P
   for the treatment of acute bacterial pneumonia. P^FCPS‐P was a
   pseudo‐conjugated polymer contained FCPS‐P (manifested as
   P^+Ph[3]Br^−‐modified sonosensitizer FCPS) as well as ROS‐sensitive
   thioketal bonds, and it was wrapped by DSPE‐mPEG[2000] in an aggregated
   fashion to locate at a central position of NP^FCPS‐P. NP^FCPS‐P
   displayed an excellent sonodynamic effect, generating ^1O[2] upon US
   stimulation in a time‐dependent manner. The generated ^1O[2] triggered
   the degradation of NP^FCPS‐P and the release of FCPS‐P. The released
   cationic FCPS‐P targeted to bacterial surface through electrostatic
   interaction and achieved bacterial killing under a synergistic action
   of SDT and P^+Ph[3]Br^−. NP^FCPS‐P+US gave >90% in vitro inhibition
   rates against MDR ESKAPE pathogens, and it could cause extensive
   morphological and biochemical changes in bacteria, especially including
   morphological shrinkage and concavity, leakage of cytoplasmic contents,
   elevated DNA damage, and decreased cellular GSH levels. A metabolomics
   assay of P. aeruginosa revealed that NP^FCPS‐P+US could lead to a
   comprehensive disorder in bacterial metabolism, particularly including
   suppressed biofilm formation and virulence, excessive ROS production
   and enhanced oxidative stress, inhibited glycolysis and TCA cycle, and
   inhibited DNA, mRNA, and tRNA biosynthesis. The ROS‐sensitive property
   of NP^FCPS‐P endowed its highly favored biosafety at cellular and
   animal levels. When NP^FCPS‐P was delivered in the lungs of mice, US
   stimulation could promote the diffusion of delivered NP^FCPS‐P and its
   SDT‐based ROS generation ability in the lungs. In mice, NP^FCPS‐P+US
   exhibited a 99.3% bactericidal rate in P. aeruginosa‐induced sublethal
   pneumonia and rendered a 90% animal survival rate in lethal pneumonia.
   As disclosed by immunological staining and transcriptomics analyses,
   the excellent therapeutic effect of NP^FCPS‐P+US depended largely on
   the suppression of pro‐inflammatory response and simultaneously
   acceleration of lung injury repair in mice. NP^FCPS‐P+US represented a
   promising antibiotics‐alternative strategy for treating deeply seated
   infections caused by MDR bacteria.

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [179]ADVS-12-2417469-s001.docx^ (35.6MB, docx)

Acknowledgements