(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



The role of naturally acquired intracellular Pseudomonas aeruginosa in the development of Acanthamoeba keratitis in an animal model [1]

['Binod Rayamajhee', 'School Of Optometry', 'Vision Science', 'Faculty Of Medicine', 'Health', 'Unsw', 'Sydney', 'Mark Willcox', 'Fiona L. Henriquez', 'School Of Health']

Date: 2024-02

Sequences of the Rns genes from the two Acanthamoeba strains AK1 and AK11 used this study were aligned using the MUSCLE algorithm and compared to the NCBI reference strains to confirm genotypes. GenBank accession numbers of AK1 and AK11 are OR263302 and OR263297, respectively. A neighbour–joining phylogenetic tree with 1,000 bootstraps was constructed using Kimura parameter and reference sequences from genotypes T1, T2, T3, T4 (A–G), T5, T6, T12, T13, and T23. Both isolates assessed in this study belong to genotype T4F, indicating they are very closely related allelic forms of the Rns with shared features and minor distinctions between them ( S1 and S3 Figs).

The 16S rRNA PCR showed the presence of intracellular bacteria in the Acanthamoeba strain AK1 which had been isolated from the domestic tap water of an AK patient ( S1 and S2 Figs). Amplicon sequencing was performed to identify the bacteria, and the blast n analysis confirmed the bacteria as P. aeruginosa (GenBank accession number OR297627). Additionally, MALDI–TOF MS analysis confirmed the strain as P. aeruginosa (score = 1.99; S5 Table ). Phylogenetically, the P. aeruginosa AK1–PA was closely related to previous isolates of P. aeruginosa, mostly obtained from environmental samples such as guano, soil, water, cloaca of Bothrops insularis, and intestinal tract of termites ( S4 Fig ).

(A) Overview of a trophozoite with intracellular P. aeruginosa. (B) Higher magnification showing the rod–shaped intracellular bacteria enclosed within phagolysosome like structure. A bacterial cell is in the process of binary fission. (C–D) Rod shaped bacteria like structure was seen inside empty cyst. Symbols: M: Mitochondria; N: Nucleus; Arrowhead: Binary fission; Black arrow: Rod–shaped bacteria; Yellow arrow: Multi–layered membrane–bound compartment; Asterisk (*): Digested bacterial cell, Endo: Endo–cyst wall; Ecto: Ecto–cyst wall. Scale bars, A: 1μm, B: 500nm, C and D: 2μm.

Under transmission electron microscopy, AK1–PA bacteria exhibited a rod–shaped morphology, enclosed within multi–layered phagolysosome like structure ( Fig 5 ). A distinct phagolysosomal membrane was evident, encapsulating the engulfed bacteria. No intranuclear stage was identified but a few cells were observed in close proximity to the nuclear membrane. Within the phagolysosome, it was intriguing to observe transverse bacterial cell division through binary fission ( Fig 5B ). Both undigested and digested bacterial cells were found within the same phagocytic vacuole as intact and disintegrated with granules, respectively. In the cystic stage of the host AK1, a bacteria–like structure was detected close to endo–cyst wall ( Fig 5C–5D ).

Probes EUK516 conjugated with Cy5 (red), targeting Acanthamoeba, and pB–383 conjugated with FITC (shown in green) targeting P. aeruginosa 16S rRNA were used in double FISH assay. DAPI was used in mounting medium when visualized by a fluorescence microscope. White arrow indicates rod–shaped bacterial cells. Scale bar in each image represents 10 μm.

Positive hybridization was observed with a fluorescent dye (FITC)–conjugated DNA probe specific to P. aeruginosa in Acanthamoeba AK1 isolated from domestic tap water of an AK patient ( Fig 4 ). The bacterial cells were present throughout the cytoplasm of the trophozoites and were observed in all amoebal cells in the population. Confocal Z–stack images also confirmed the presence of P. aeruginosa cells inside Acanthamoeba trophozoites instead of sitting on the host surface ( S5 Fig ). This observation showed an average of 4±2.2 bacteria (mean±SD) per trophozoite.

AlkDala labelling was tested on AK1 trophozoites (first panel) and the green fluorescence indicates metabolically viable intracellular bacteria. A. castellanii (ATCC 30868) was used as a control (second panel) and DAPI was used to stain host nucleus. Scale bars, 10μm (first panel) and 15μm (second panel).

To confirm the viability of intracellular P. aeruginosa (AK1–PA) cells within trophozoites, we used (R)–α–Propargylglycine as a D–alanine analogue in the assay to incorporate into the bacterial cell’s peptidoglycan during cell wall biosynthesis. Subsequently, the cells were labelled with a fluorescent Azide probe using click chemistry. Prior to use with Acanthamoeba cells, alkDala–labelling was tested for specificity and efficacy with PAO1 and S. aureus (SA32, clinical isolate) using both viable and heat killed bacteria. As expected, heat–killed bacteria or bacteria treated only with D–alanine (without alkyne group) did not label. A. castellanii ATCC 30868, devoid of any intracellular bacteria was used as a control. The confocal microscopy showed alkDala–labelled bacterial cells within the amoebal host indicating the presence of metabolically active bacteria harboured by Acanthamoeba strain AK1 ( Fig 6 ). These results are consistent with the electron microscopy observation where bacterial cells undergoing binary fission were seen indicating the presence of live P. aeruginosa residing within the host cell.

Effect of Acanthamoeba adaptation on the motility of intracellular P. aeruginosa (AK1–PA) compared with PA01; ( i ) swimming, ( ii ), swarming, and ( iii ) twitching. Data are mean ± SEM. The mean motility data were analysed using an unpaired t test (all p–values >0.05).

Acanthamoeba adapted P. aeruginosa AK1–PA showed slightly greater swimming motility compared to the non–adapted wild–type strain PA01, but the difference was not significant (p>0.05, Fig 7i ). The mean swarming motility of AK1–PA strain was about double that of the PA01, but the difference was not significant (p>0.05, Fig 7ii ). Similarly, the twitching distance (radius) exhibited by AK1–PA was 1.3mm higher compared to the PA01, but again this was not significant (p>0.05, Fig 7iii ).

The intracellular P. aeruginosa AK1–PA possessed exoU but not exoS ( S6 Fig ). Two open reading frames (ORFs) were identified in the 1176–bp sequence of exoU. The amino acid sequence coded by ORF1 (322→1071, 249 aa) showed a high similarity with the T3SS effector putative cytotoxic exoU in blastp research. A few single–nucleotide polymorphisms [SNPs; V1056C (GGC→GCG), V1084G (CAA→CGA), V1090G (CAA→CGA), V1101T (GAA→GTA), V1113G (CCA→CGA), and V1121T (CAA→CTA)] and deletion mutations [Δ1 bp (12), Δ1 bp (1028), Δ1 bp (1092), Δ1 bp (1103), Δ1 bp (1126), Δ1 bp (1157), and Δ1 bp (1168)] were identified when the sequence of exoU of AK1–PA (1176 bp sequence) was compared with the exoU genes of other P. aeruginosa strains deposited in NCBI. However, high–throughput whole genome sequencing is required to confirm the SNPs and deletions.

Representative confocal microscopy images of hybridization assay (i) of hMDMs cells after three hours post–infection; A . mock–infected cells, B . infected with wild–type (PA01), and C . infected with P. aeruginosa (AK1–PA). Compared to uninfected cells, the structure of macrophage cells infected with AK1–PA exhibited a slight disorganization. Bacterial numbers in hMDMs cells were enumerated by confocal microscopy ( ii ). Student’s t test was used to compare the number of AK1–PA/hMDM versus wild–type PA01 at 3, 6, and 12 hrs p.i. White arrow shows bacteria within macrophage cells. Scale bar represents 20μm.

EUB338 probe labelling also revealed a higher number of P. aeruginosa AK1–PA in macrophage cells compared to the wild–type strain 3 hrs p.i. ( Fig 9i ), indicating enhanced intracellular proliferation of Acanthamoeba–adapted P. aeruginosa in human immune cells. Uninfected and PA01–infected macrophage cells showed normal morphology, while those infected with AK1–PA were slightly disorganized ( Fig 9i , C ). These results suggest that amoeba–adapted bacteria were more toxic to the macrophages than the non–adapted strain. To confirm the intracellular presence of bacteria at the single–cell level, we enumerated bacterial load per hMDM cell at 3–12 hrs post–infection. At 3 hrs p.i., we observed approximately 3–fold more bacteria per hMDM infected by AK1–PA compared to PA01, but this difference was not significant (p>0.05) ( Fig 9ii ) . However, at 6 and 12 hrs post–infection, the bacterial load in macrophage cells harbouring AK1–PA strain was significantly higher than that of the wild–type strain (p<0.05).

The data represents the mean CFUs ± SEM from three independent experiments (n = 3). Student’s t test was used to compare the intracellular numbers of AK1–PA and PA01 at different time points p.i. (*, p<0.05; **, p<0.01; ***, p<0.001).

Fig 8. To assess the intra–vacuolar replication of P. aeruginosa strains (AK1–PA and PA01) within hMDMs monolayers, infected hMDMs monolayers were lysed at 3, 6, 12, and 24 h p.i. and serial dilutions were plated on agar plates to quantify the colony–forming units (CFUs).

To assess whether the intracellular survival of Acanthamoeba–adapted bacteria extends to other higher eukaryotic phagocytic cell, the survival abilities of adapted AK1–PA and non–adapted P. aeruginosa PA01 were compared using primary human macrophages (hMDMs) ( S7 Fig ). At 3 hrs post–infection, the AK1–PA strain infected macrophages had more than 4–fold greater numbers of intracellular bacteria compared to the wild–type strain ( Fig 8 , p<0.05). The difference was approximately 3–fold at 24 hrs p.i. (p<0.05).

3.7. The Acanthamoeba strain with intracellular bacteria induced acute keratitis

The study investigated whether naturally acquired intracellular bacteria play a role in inducing severe Acanthamoeba keratitis in rats’ eyes. In the first phase, axenically cultured Acanthamoeba cells with (AK1) and without (AK11) intracellular bacteria were trans–corneally inoculated into the rat’s cornea and the progression of infection was observed microscopically (Fig 3). Hybridization assay has shown approximately four bacteria in each trophozoite of the AK1 strain, so 4×104 P. aeruginosa bacteria were inoculated when 104 trophozoites were delivered into each eye of group B rats, while group A received only 104 trophozoites (AK11). The clinical features of AK were assessed, recorded, and graded using slit lamp examinations from day 1 to 5 post–infection. Group A rats cornea inoculated with Acanthamoeba devoid of any intracellular bacteria, showed a few focal infiltrates on day 4 of infection. The corneas remained transparent during the infection period with very mild keratitis showing no signs of inflammation (Fig 9i).

In contrast, group B rats infected with an Acanthamoeba strain containing viable intracellular P. aeruginosa exhibited severe infection with a large ring infiltrate in the centre of the cornea within 48 hrs p.i. All rats of group B rapidly developed keratitis with anterior chamber inflammation, severe conjunctivitis, diffuse infiltrates, and mild corneal edema. By day 4, the ocular lesions had progressed to corneal epithelial ulceration, accompanied by extensive stromal inflammation. Fluorescein staining of the cornea was not performed for group B rats on days 4 and 5 due to acute infection (Fig 10i). Blood vessels and random superficial lesions were developed with an extensive zone of corneal opacity and necrosis at the centre on day 4 in group B rats. Based on slit lamp examination, infection was at its peak on day 3 in group B rats, but it remained mild even on day 5 in group A rats. The mean clinical scores of group A and group B rats revealed a significant difference on each day of post–inoculation (Fig 10ii). Five days after infection, rats were euthanised due to acute keratitis, which was the clinical end point. The weights of both groups rats remained normal compared to starting weight (S8 Fig).

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 10. Representative slit lamp photographs of experimentally induced Acanthamoeba keratitis in rats’ right eyes caused by Acanthamoeba alone (group A) or Acanthamoeba with viable P. aeruginosa (AK1) (group B) from day 1 to 5 p.i. The first panel of each group represents bright field, and the second shows a fluorescein–stained micrographs (i). The mean clinical score of group A and B rats was compared using an unpaired t–test (ii) (*, p≤0.05, ***, p≤0.001, ****, p≤0.0001, n = 4 in each group). https://doi.org/10.1371/journal.pntd.0011878.g010

The HE staining of corneal microsections from the control left eyes had a normal appearance showing a well–defined stromal structure with regularly arranged stromal fibres and the absence of any changes in the corneal epithelium (Fig 11A). However, the corneal epithelium and endothelium displayed areas of necrosis in the stromal region of group A rats (Fig 11B). In contrast, histological analysis showed the inflammatory infiltrate at centre of stroma, and the epithelium and endothelium of cornea were completely collapsed in group B rats, with haemorrhagic necrosis and desquamated cells throughout the cornea (Fig 11C). Furthermore, the observation showed substantial quantities of cellular debris and accumulated fibrin, neovascularisation surrounded by infiltrate leukocytes and profusion of granulation tissue.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 11. Histological observation of AK induced in Wister rats by intrastromal injection of amoebal trophozoites. Five μm thick corneal section of the uninfected left eye stained with HE exhibited well–defined epithelium and stroma without any infiltration of inflammatory cells (A), but a portion of endothelium was fell off during micro–sectioning in the cryostat. Corneas infected by Acanthamoeba alone (group A) showed a few areas of necrosis and blood cells with the epithelium and endothelium slightly disorganized (B). In group B, the stromal region infected by Acanthamoeba with intracellular bacteria exhibited haemorrhagic necrosis with fibrin deposition, desquamated cells, inflammatory infiltrate, and a collapse of the corneal epithelium and stromal structure (C). Indicators: White arrow, corneal epithelium, and endothelium; Yellow arrow: RBCs; Asterisk (*): Necrosis; Hash (#): Haemorrhagic necrosis; and Arrowhead: Inflammatory infiltrate. Scale bar represents 30μm for A, 20μm for B and C. https://doi.org/10.1371/journal.pntd.0011878.g011

Cyst–like structures were observed in the corneal sections of group B rats when examined under a light microscope. Therefore, calcofluor white stain (CFW), a chemo–fluorescent dye that binds with cellulose in the cyst cell wall was used to stain suspected Acanthamoeba cysts. Interestingly, those suspected cyst–like structures in the corneal sections of group B rats exhibited a bluish white colouration under a fluorescence microscope (Fig 12). The corneal sections of group A rats showed no CFW staining.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 12. Thin (10μm) corneal sections from group A (A) and B (B) rats were stained with CFW. Corneal sections from group B exhibited blue fluorescence with cyst–like structures resembling amoebic cysts, while no staining was observed in group A. Arrowhead indicates Acanthamoeba cysts like structures under light and fluorescence microscopes. Scale bar represents 20μm. https://doi.org/10.1371/journal.pntd.0011878.g012

To confirm cells stained with CFW were Acanthamoeba cysts, suspected cells (~150 cells) were precisely collected from heterogeneous cell population from corneal sections of group B rats. Corneal sections were transferred onto a PEN (polyethylene naphthalate) membrane slide (Carl Zeiss Microscopy, GmbH, Germany), and cyst–like cells were selected, excised, and collected using non–contact laser pressure catapult (LPC) procedure. The catapulted cells were collected in a collection tube and gDNA was extracted followed by PCR using the Acanthamoeba specific JDP primer pair. PCR confirmed that the cyst–like cells labelled with CFW in the corneas of group B rats were Acanthamoeba cysts (S9 Fig).

Corneal homogenates were cultured from the infected right eyes of both group rats to re–isolate intrastromally inoculated Acanthamoeba from group A and Acanthamoeba along with intracellular P. aeruginosa from group B. Non–nutrient agar (NNA) and TSA were used to culture Acanthamoeba and intracellular P. aeruginosa from the homogenates, respectively. No growth was observed from any of the group A homogenates while Acanthamoeba trophozoites and P. aeruginosa colonies were grown from group B. NNA plates were cultured for three weeks at 32°C and trophozoites number were counted using an inverted microscope (IX71, Olympus America, NY, USA). Among the four corneal tissues of group B, the trophozoites count was not significantly different in the cornea six days after infection (p>0.05). Trophozoites recovered from corneal tissues were approximately 18 to 25–fold lower compared to the original inoculum (104/cornea) but the mean count across four corneal tissues was not significantly different (p>0.05) (Fig 13A). Similarly, P. aeruginosa counts ranged from 9.2x104 to 1.8x104/cornea, and the counts were not significantly different (p>0.05) except between rat B.2 and B.4 (p = 0.04) (Fig 13B). Acanthamoeba trophozoites (AK10) recovered from group B rats’ corneal tissue appeared to have expelled all their intracellular P. aeruginosa during infection in rat’s eye as no intracellular bacteria seen in hybridization assay (Fig 13C). Acanthamoeba (AK10) and P. aeruginosa re–isolated from the corneal tissues of group B rats were utilised in the second phase of experimental AK study to investigate whether the severe keratitis observed in group B was attributed to the co–presence of both Acanthamoeba and intracellular bacteria.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 13. The mean trophozoites and P. aeruginosa counts in each corneal homogenate of group B rats’ eyes (A, B). Acanthamoeba trophozoites recovered from the corneal homogenate of group B rats (AK10) were used for FISH assay to assess presence of originally acquired intracellular P. aeruginosa using the pB–383 probe. Representative FISH images demonstrate the absence of intracellular bacteria in any of the amoebal cells within the population (C). Scale bar represents 10μm. https://doi.org/10.1371/journal.pntd.0011878.g013

In the second phase, Acanthamoeba (AK10) was inoculated into group C (n = 3) rats, while P. aeruginosa recovered from corneal tissues was used to induce keratitis in group D (n = 3) rats, following the same procedure as that used in phase I. Acanthamoeba (AK1) contained an average of four bacterial cells per trophozoite. Therefore, a total of 4x104 P. aeruginosa cells were inoculated into the group D rats to achieve a bacterial load similar to that of group B rats. In group C rats, minor focal and diffuse infiltrates along with faint linear epithelial corneal opacities were noted 72 hrs p.i. In contrast, group D rats exhibited moderate conjunctivitis, diffuse infiltrates, slight corneal opacity, and diffuse central edema. Retained fluorescein stain was also clearly visible 48 hrs p.i. (Fig 14i). Clinical corneal lesions were similar between all rats in group D with the infection reaching its peak on day 3, which was significantly higher compared to group C (Fig 14ii). Between p.i. days 4 and 5, corneal opacities and infection slowly decreased in group B, but the infection remained almost similar in group C characterised by focal and diffuse infiltrates at the centre of cornea. Consistent to slit lamp observation, HE staining of corneal microsections from group C rats showed a discrete accumulation of red blood cells (RBCs) accompanied by slight epithelial disorganization. No evidence of inflammatory infiltrates or severe necrosis was detected. Necrosis–like structures with RBCs, desquamated cells, mild inflammatory infiltrate, and notable disorganization of both the corneal epithelium and stromal structure were seen in corneal tissues of group D rats (S10 Fig). Similar to phase I, the weights of rats in both groups did not change significantly compared to their starting weights (S11 Fig.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 14. Clinical photographs of rats’ right eyes with experimentally induced Acanthamoeba keratitis. Development of keratitis in rats’ corneas inoculated only with Acanthamoeba (AK10) (group C) or P. aeruginosa (group D) from day 1 to 5 p.i. Representative slit lamp photographs (i), before (first panel of each group) and after (second panel of each group) application of fluorescein stain. The mean clinical score of group C and D rats was compared using an unpaired t–test (ii) (*, p≤0.05, ***, p≤0.001, ****, p≤0.0001, n = 3 in each group). https://doi.org/10.1371/journal.pntd.0011878.g014

We also compared the mean clinical scores of group A rats with those of group C rats, which were infected by different strains of Acanthamoeba devoid of any intracellular bacteria. The average clinical score of group C rats was significantly higher (p<0.05) than that of group A rats at 24, 72, 96, and 120 hrs p.i. (Fig 15i). This indicates that Acanthamoeba may have become more virulent after adapting to the rat cornea, as group C rats were infected with Acanthamoeba re–isolated from corneal tissue. Similarly, infection severity was significantly higher (p<0.05) in presence of Acanthamoeba with naturally acquired intracellular P. aeruginosa (group B) compared to cases where only intracellular P. aeruginosa was inoculated (group D) at all time points from day 1 to 5 p.i. (Fig 15ii).

[END]
---
[1] Url: https://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0011878

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.