The effect of Galleria mellonella hemolymph polypeptides on Legionella gormanii

Among Legionella species, which are recognized to be pathogenic for humans, L. gormanii is the second prevalent causative agent of community-acquired pneumonia after L. pneumophila. Anti-L. gormanii activity of Galleria mellonella hemolymph extract and apolipophorin III (apoLp-III) was examined. The extract and apoLp-III at the concentration 0.025 mg/ml caused 75% and 10% decrease of the bacteria survival rate, respectively. The apoLp-III-induced changes of the bacteria cell surface were analyzed for the first time by atomic force microscopy. Our studies demonstrated the powerful anti-Legionella effects of the insect defence polypeptides, which could be exploited in drugs design against these pathogens.


INTRODUCTION
Currently, members of the family Legionellaceae comprise 58 described species that are highly successful in colonizing natural aquatic environments (Euzeby, 2013).A particular hallmark of these bacteria is their dual host system allowing intracellular growth in protozoa (from the genera Acanthamoeba, Naegleria, Hartmanella) and in human alveolar macrophages during infection (Fields et al., 2002).The ability to survive and replicate inside unicellular organisms contributed to the acquisition of virulence factors that enable Legionella to overcome the antimicrobial activities of human macrophages.The clinical manifestations of Legionella infections are primarily related to the respiratory tract.The most common presentation is acute pneumonia, which varies in severity from mild illness to fatal multilobar pneumonia.Legionella is widely recognised as an important source of community-and hospital-acquired pneumonia.The second form of a respiratory illness is called Pontiac fever, which is a flu-like infection (Palusinska-Szysz & Cendrowska-Pinkosz, 2009).The first described Legionella species, L. pneumophila, is unceasingly the dominating species among clinical isolates, whereas L. gormanii is the second prevalent causative agent of community-acquired pneumonia (Lode, 1987).Legionella infections are difficult to treat because of their intracellular localization in phagocytic cells.Moreover, bacteria released from A. castellanii are more resistant to chemical disinfectants and antibiotics used to treat pneumonia in comparison with those residing out-side amoebae in the environment or laboratory cultured on Legionella artificial medium.A combination of fluoroquinolones with macrolides is an effective method of Legionnaires' disease treatment.However, the mortality rate among patients with hospital-acquired pneumonia is 14%, and with community-acquired pneumonia ranges from 5 to 10% (Benin et al., 2002;Palusinska-Szysz, 2011).Therefore, the development of new antibacterial agents for Legionella infections is urgently needed.
A rich source of natural defence peptides (antimicrobial peptides; AMPs) with different biochemical properties and antimicrobial activity is insect's hemolymph (Bulet et al., 2004;Bulet & Stöcklin, 2005).Due to amino acid composition, amphipathicity, cationic charge, and molecular size, AMPs can interact with microbial cell membranes forming toroidal or barrel-stave pores, or micelles composed of peptides and the membrane phospholipid molecules.Disturbing a proper structure of a cell membrane leads to depolarization, increased permeabilization, and even membrane fragmentation, which results in death of the invading microbes (Bulet et al., 2004).These properties and selective toxicity towards some pathogens make them the patterns for designing drugs alternative to antibiotics.
In order to test antimicrobial activity of G. mellonella defence compounds against other species of Legionella, the effects of hemolymph methanolic extracts and the purified apoLp-III on L. gormanii cells were investigated.In addition, the apoLp-III-induced changes of the bacteria cell surface topography and properties were imaged and analyzed by atomic force microscopy (AFM).Our study is an attempt at assessment of the potential of the new agents in elimination of L. gormanii.
Insects immunization and preparation of hemolymph methanolic extracts.The larvae of G. mellonella (Lepidoptera: Pyralidae) were reared on honeybee nest debris (a natural diet) at 30 o C in the dark.The immune challenge was performed by puncturing of the last instar larvae with a needle dipped into a pellet containing live E. coli D31 and M. luteus cells.The immune hemolymph was collected 24 h after the challenge.The methanolic extracts containing antimicrobial peptides and proteins below 30 kDa were prepared from the hemocyte-free hemolymph and deprived of lipids as described earlier (Cytryńska et al., 2007).The protein concentration was determined by a Bradford method (Bradford, 1976) using bovine serum albumin as a standard.
Purification of apolipophorin III.G. mellonella apoLp-III was purified from the immune hemolymph extract as described in our previous study (Zdybicka-Barabas et al., 2013).Briefly, the freeze-dried hemolymph extract dissolved in 0.1% trifluoroacetic acid (TFA) was subjected to the HPLC chromatography using a Discovery Bio Wide Pore C18 4.6 mm × 250 mm column (Sigma-Aldrich, USA) and two buffer sets, A: 0.1% TFA (v/v), B: 0.07% TFA, 80% acetonitrile (v/v).A linear gradient from 30 to 70% of buffer B over 35 min and 1 ml/min flow rate was applied.The homogenous fraction containing apoLp-III was freeze-dried, subjected to weighting, redissolved in sterile deionized water, and stored at -80 o C until use.The homogeneity and identity of apoLp-III was confirmed by SDS-PAGE electrophoresis (Schägger & von Jagow, 1987) and by N-terminal sequencing on an automatic protein sequencer (Procise 491, Applied Biosystems).
Antimicrobial assays.The activity of immune hemolymph extract and apoLp-III against L. gormanii was carried out using a colony counting assay as described previously (Palusinska-Szysz et al., 2012).Briefly, 10 µl of a bacterial suspension (obtained by 2×10 -4 dilution of a suspension with OD 620 = 0.1) was incubated without (control) or with the extract (final protein concentrations 0.025-0.8mg/ml) or apoLp-III (final protein concentrations 0.025-0.2mg/ml) at 37°C for 1 h.Then, the incubation mixtures were spread onto BCYE medium plates, incubated for 4 days at 37°C, and the number of the colony-forming units (CFU) was determined.The minimal inhibitory concentration (MIC) was defined at the concentration which yielded 95% inhibition of bacterial growth.The data were calculated from three independent experiments, each performed in triplicate.
Atomic force microscopy imaging of L. gormanii.Forty μl of a water suspension (OD 620 =0.2) containing the L. gormanii cells grown on BCYE medium were incubated without (control) and in the presence of purified G. mellonella apoLp-III (final concentration 0.2 mg/ml) at 37 o C for 1 h.After centrifugation (8 000 × g, 10 min., 4 o C) the bacteria were suspended in 5 μl of pyrogen-free water, applied on the surface of mica discs and allowed to dry overnight at room temperature.
L. gormanii cell surface imaging was carried out using NanoScope V AFM (Veeco, USA) in "PeakForce QNM" operation mode with a NSG 30 silicon tip (spring constant of 20N/m; NT-MDT, Russia) (Analytical Laboratory, Faculty of Chemistry, UMCS, Lublin, Poland).The data were analyzed with Nanoscope Analysis ver.1.40 software (Veeco, USA).Two fields on each mica disc were imaged.The roughness values were measured over the entire bacterial cell surface on 3 µm×3 µm areas.The average surface root-mean-square (RMS) roughness was calculated from forty fields (265 nm×265 nm).
Statistics.Statistical analysis was performed using the Wilcoxon's paired test.The data were presented as ± standard deviation (SD) from three independent experiments.

RESULTS AND DISCUSSION
The antibacterial activity of G. mellonella immune hemolymph extract against L. gormanii was evaluated.The results showed dose-dependent killing of L. gormanii cells by the extract (Table 1).The incubation of L. gormanii in the presence of the extract at the concentration 0.025 mg/ml for 1 h caused more than 75% decrease of the bacteria survival rate compared to the control, L. gormanii was incubated with the G. mellonella immune hemolymph extract at the concentrations of 0-0.8 mg/ml at 37°C for 1 h.The bacteria were then seeded on BCYE agar plates and the colonies were counted after four days incubation at 37°C in a 5% CO 2 atmosphere.Data were expressed as means ± S.D. of three independent experiments.
i.e. non-treated bacteria.The MIC value defined as the concentration yielding at least 95% inhibition of bacterial growth for the extract was determined to be 0.8 mg/ml.However, the extract used at the concentration 0.4 mg/ ml reduced the bacteria survival rate to almost the same extent, i.e. by 94%.Recently, we have reported that antimicrobial proteins and peptides of the G. mellonella immune hemolymph extract inhibited L. dumoffii growth.
When the extract was used at the concentration 0.4 mg/ ml, ca.50% decrease of L. dumoffii survival rate was observed (Palusinska-Szysz et al., 2012).
A main protein component of G. mellonella hemolymph extract is apoLp-III (Cytryńska et al., 2007).Hence, the effect of apoLp-III on L. gormanii survival rate was studied.A colony counting assay using apoLp-III at the concentrations 0.025-0.2mg/ml revealed that L. gormanii cells were sensitive to apoLp-III.The protein at the concentration 0.2 mg/ml caused ca.50% decrease of the bacteria survival rate (Fig. 1).Interestingly, when apoLp-III was used at the concentration 0.025 mg/ml, the survival rate of L. gormanii decreased only by 10%, whereas the hemolymph extract at the same concentration caused 75% reduction of bacterial survivability (Fig. 1).The results indicated that although L. gormanii was susceptible to apoLp-III action, other proteinaceous (e.g.defence proteins and peptides) and non-proteinaceous compounds present in G. mellonella hemolymph extract were probably also involved in killing of the bacteria.Increasing of the hemolymph extract and apoLp-III concentrations above 0.8 mg/ml and 0.2 mg/ml, respectively, did not reduce further L. gormanii survival rate (data not shown).However, the explanation of this fact needs further investigations.
In our previous papers, usefulness of an atomic force microscopy for analysis of an influence of G. mellonella defence factors on bacterial and fungal cell surface has been demonstrated (Zdybicka-Barabas et al., 2011;2012a;2012b;2013).In this study, AFM was used for examining of the G. mellonella apoLp-III effect on L. gormanii cell surface.AFM imaging revealed that the surface of L. gormanii control cells was covered with small uniform granules (Figs. 2, 3).On the surface of some control cells few shallow depressions were also visible.The bacteria were incubated with the G. mellonella apoLp-III at the concentrations of 0-0.2 mg/ml at 37°C for 1 h.Cells were then plated on BCYE agar and the number of colonies was counted after four days incubation at 37°C in a 5% CO 2 atmosphere.Experimental results were mean ± S.D. of three independent experiments.The Wilcoxon's paired test was used for comparisons within groups.p values *p≤0.001 were considered.In addition, the cells with uneven surface covered with furrows, some of them 10-20 nm deep and even 200 nm wide, were detected in the control samples.In contrast, the most of the apoLp-III-exposed cells were decorated with numerous rounded bubble-like features of different size, 15-20 nm in height and 100-200 nm in diameter (Figs. 2, 3).The alterations of the L. gormanii cell surface caused by apoLp-III were also reflected by 1.44-fold increase in cell surface roughness, the parameter used to describe the structural heterogeneity of the bacterial cell surface.The RMS roughness values for the control and apoLp-III-treated cells were calculated as 2.935 nm (±1.016) and 4.2305 nm (±1.003; p=0.00103), respectively.One of the reasons of the increase of the surface roughness could be rupturing of bacterial cell following the apoLp-III binding to LPS and phospholipids, which was reported by Pratt & Weers (2004).
Although both L. dumoffii and L. gormanii were sensitive to G. mellonella apoLp-III, the 8-fold lower concentration of apoLp-III was sufficient to exert the same bactericidal effect on L. gormanii in comparison to L. dumoffii (Palusinska-Szysz et al., 2012).The high sensitivity of L. gormanii to apoLp-III in comparison with L. dumoffii indicates that different Legionella species could exhibit diverse susceptibility to the insect-derived antimicrobial factors, possibly reflecting differences in the cell surface properties.
Searching for antimicrobials effective against L. gormanii is especially important since this species was isolated from peadiatric cases.Moreover, antimicrobial ther-apy commonly used in empirical pneumonia treatment in young patients is not effective against Legionella spp.(Ephros et al., 1989;Greenberg et al., 2006).

Figure 1 .
Figure 1.The effect of G. mellonella apoLp-III on L. gormanii survival rate.The bacteria were incubated with the G. mellonella apoLp-III at the concentrations of 0-0.2 mg/ml at 37°C for 1 h.Cells were then plated on BCYE agar and the number of colonies was counted after four days incubation at 37°C in a 5% CO 2 atmosphere.Experimental results were mean ± S.D. of three independent experiments.The Wilcoxon's paired test was used for comparisons within groups.p values *p≤0.001 were considered.

Figure 2 .
Figure 2. The effect of G. mellonella apoLp-III on L. gormanii cell surface topography.The cells were incubated without (control) or in the presence of apoLp-III (0.2 mg/ml) at 37 o C for 1 h and then imaged by AFM.The height and "peak force error" images are presented.The white and black arrowheads indicate granules and furrows, respectively, on the control cells surface.The white arrows mark bubble-like features appearing on apoLp-III-treated cells.

Figure 3 .
Figure 3. Profile section analysis of L. gormanii cell surface.The cells were incubated without (control) or in the presence of apoLp-III (0.2 mg/ml) at 37 o C for 1 h and then imaged by AFM.The upper panels present the height images of the cell surface.The bottom panels demonstrate the section profiles corresponding to the lines (a, b) shown in the height images.