Scientific Publications - Work Done by Microbiology Reader Bioscreen C

Plant Science Volume 139, Issue 1 , 11 December 1998, Pages 83-91

Activity of  different antibacterial peptides  on Erwinia amylovora growth,  and evaluation of the phytotoxicity  and stability of cecropins

Fabienne Mourgues^a, Marie-Noëlle Brisset^{a, b} and Elisabeth Chevreau<sup>a

a</sup> INRA, Station d'Amélioration des Espèces Fruitières et Ornementales, BP 57, 49071, Beaucouzé Cedex, France
^b INRA, Station de Pathologie Végétale, BP 57, 49071, Beaucouzé Cedex, France

Received 18 June 1998; revised 24 August 1998; accepted 14 September 1998. Available online 23 November 1998.


 

ABSTRACT

Antibacterial peptides of plant or non-plant origin are attractive candidates for pear (Pyrus communis L.) genetic engineering to increase resistance to fire blight caused by Erwinia amylovora. The antibacterial activity of several peptides isolated from cereals (purothionins and puroindolines), insects (cecropins), animals (hen egg white (HEW) lysozyme) and microorganisms (T₄ lysozyme) was tested against E. amylovora. Cecropins had the higher bactericidal efficiency, followed by T₄ lysozyme. A synergistic effect between Cecropin B and HEW lysozyme was also demonstrated. No toxicity of cecropins towards pear cell suspension cultures was observed at the bactericidal concentrations. However, the incubation of cecropins with pear leaf extracellular fluids (ECF) caused a loss of their antibacterial activity and a degradation of ECF-treated cecropins was observed. We conclude that the expression of a cecropin gene in transgenic pear will be efficient only if the peptide breakdown in pear tissue is counterbalanced by its continuous secretion into the intercellular spaces to maintain a bactericidal concentration at the point of infection.

Author Keywords: Pear; Cecropins; Thionins; Lysozymes; Pyrus communis; Erwinia amylovora

Abbreviations: cfu, colony-forming unit; ECF, extracellular fluid; FDA, fluorescein diacetate; G-6-PD, glucose-6-phosphate dehydrogenase; HEW, hen egg white; KB, King's medium B; MBC, minimal bactericidal concentration; MES, morpholino ethane sulfonic acid; MIC, minimal inhibitory concentration; NADP+, nicotinamide adenine dinucleotide phosphate; PCV, packed cell volume; PVP-10, polyvinylpyrrolidone wt. 10000

Index Terms: plant growth; peptide analysis; phytotoxicity; drug stability; peptide hormone; cecropin; antibiotic agent

1. INTRODUCTION

Bacterial diseases cause great losses in several important crops such as rice, vegetables and fruits. Various projects are attempting to introgress genes that encode antibacterial peptides into plants to control plant pathogens [1]. The success of such strategies depends on the efficient antibacterial activity of the peptides and on their stability in plant tissue. Additionally, it is a prerequisite that the product of the transferred gene is non-toxic towards plant cells. In the case of edible crops, absence of toxicity towards mammalian cells is also requested.

The antibacterial activity of several peptides such as purothionins [2], cecropins [3, 4, 5, 6 and 7] and lysozyme [8] has already been established on plant or animal pathogenic bacteria. However, these studies have been accomplished following different procedures on various pathogens that make it difficult to compare these peptides on a given pathogen. Fire blight, caused by Erwinia amylovora (Burill) Winslow et al., is the most important bacterial disease of Maloideae. Pear varieties, which are particularly susceptible to fire blight, are now amenable to genetic engineering [9]. The present study compares the in vitro antibacterial activity of cecropins, purothionins, puroindolines and lysozymes on E. amylovora. Efficacy of antibacterial peptides was measured on E. amylovora growth, in pure culture or in presence of pear cells. In addition, the stability of antibacterial peptides in pear tissue and their non-toxicity towards plant cells were studied.

2. MATERIAL AND METHODS

2.1. Antibacterial peptides

All antibacterial peptides tested are presented in Table 1. These peptides were solubilized in sterile distilled water, except lysozyme T₄ which was dissolved in HEPES (N-(2-hydroxyethyl) piperazine-N′-2-ethane sulfonic acid) buffer, pH 7.4. Peptides were stored at −20°C.

 

Table 1. Antibacterial substances tested against E. amylovora^a

 

2.2. Bacterial strains

E. amylovora strain CFBP1430 and its transposon mutant PMV6023 were used [10]. PMV6023, mutated in hrp region, is unable to induce either the disease on host plants, or the hypersensitive reaction on non-host plants. Bacteria were cultivated on King's medium B (KB) [11], supplemented with 20 mg l⁻¹ chloramphenicol for the transposon mutant.

2.3. Pear cell suspension

Callus formation was obtained from in vitro internode explants of Pyrus communis cv. Doyenné du Comice, plated on modified Murashige and Skoog basal medium with 1 mg l⁻¹ kinetin and 1 mg l⁻¹ 2-4-dichlorophenoxyacetic acid. After 1 month, a suspension cell culture was established from friable calli in the same liquid medium. Cells were grown at 25°C in a 250 ml Erlenmeyer flask, on a rotary shaker (120 rpm) under 16 h photoperiod. Cells were subcultured every 10 days, by adding 7-8 g fresh weight of cells to 120 ml of fresh medium. Viability of plant cells was determined by staining with fluorescein diacetate (FDA) [12].

2.4. Extracellular fluid extraction

Leaf extracellular fluid (ECF) was extracted from the two youngest leaves, taken from 1-month-old seedlings of cv. Kirchensaller, grown in the greenhouse. Leaves were vacuum infiltrated for 5 min at 30 hPa with 5 mM MES buffer pH 6.0 supplemented with 10 mM ascorbic acid, 10 mM reduced glutathion, 55 mM mannitol and 2 mM PVP-10. Whole leaves were then blotted dry and centrifuged (10000×g, 15 min) in tubes with plastic mesh set about 0.5 cm above the bottom. The ECF collected from the bottom of the tube was used immediately or stored at −20°C until used. The collected volume was between 10 and 100 l per leaf. For each experiment, ECFs from 100 leaves were pooled. Their total soluble protein content assayed according to Bradford [13] was on average 50 g ml⁻¹. The glucose-6-phosphate dehydrogenase (G-6-PD) activity (EC 1.1.1.49) was used as a marker to control the absence of cytoplasmic contaminants in ECF extracted from pear leaves. The G-6-PD activity of extracted ECF was measured by spectrophotometry [14] using the following reaction mixture, 60 mM Tricine buffer, pH 8.0, 2.5 mM glucose-6-phosphate, 0.3 mM NADP⁺and 150 l extract in a final volume of 400 l. Total soluble protein extract, obtained from leaves grounded in the same buffer used in ECF extraction, was used as the positive control. The increase of optical density at 340 nm, due to the reduction of NADP⁺, was monitored at 30°C.

2.5. In vitro antibacterial assay

Minimal inhibitory concentration (MIC) was determined in liquid KB. Bacteria CFBP1430 were harvested from solid KB after growing for 24 h at 25°C, resuspended in liquid KB to yield approximately 10⁷ cfu ml⁻¹, and dispensed in microwell plates (300 l). Antibacterial peptides were added at various concentrations (see below) and incubation was performed at 26°C with rapid continuous shaking for 72 h. The bacterial population growth was estimated automatically every 2 h by optical density measurement at 600 nm (Microbiology Analyser Bioscreen C, Labsystems). The MIC was taken as the lowest concentration of peptide completely inhibiting growth. The minimal bactericidal concentration (MBC) was determined through subculture of inhibited broth on solid KB, incubated for 48 h. The MBC was defined as the lowest concentration of applied peptides after which no regrowth was observed after 48 h. Concentration of each peptide was tested in triplicate. Each experiment was repeated one to five times depending on availability of the peptide. Tested concentrations were 0, 50, 100 and 200 M for the lysozymes, purothionins and puroindolines, 0, 5, 10, 15, 20 and 50 M for Shiva-1, 0, 5, 10, 15, and 20 M for SB-37, and 0, 0.5, 1, 2.5, 5, 10, 15 and 20 M for Cecropin B. Synergistic antibacterial effect of hen egg white (HEW) lysozyme (150 M) and Cecropin B (1 M) was also measured, in two different experiments, following the procedure mentioned above.

2.6. Activity of cecropins on the E. amylovora/pear cell interaction

Antibacterial activity of Cecropin B and SB-37 was tested on E. amylovora in the presence of pear cells, using a procedure adapted from Brisset and Paulin [15]. Bacteria were washed in liquid medium M5 (5 mM MES, 3% mannitol, pH 6.0) and pelleted by centrifugation (12000×g, 10 min at 4°C). The pellet was then resuspended in the same medium. Pear cell suspensions (4-5 days-old) were washed twice in M5 medium then pelleted by centrifugation (150×g, 3 min) and resuspended in M5 medium to reach 10% packed cell volume (PCV). This PCV was previously determined by centrifugation of 10 ml of cell suspension at 600×g for 10 min in graduated tubes. Washed pear cells were pre-incubated for 1.5 h in Falcon tubes in a rotary shaker (120 rpm), then aliquoted in four well-plates, resulting in 500 l of cell suspension per well. Bacteria and Cecropin B or SB-37 were added in order to obtain a final concentration of 10³-10⁷ cfu ml⁻¹ for bacteria, 5 M for Cecropin B, and 15 M for SB-37. Plates were then incubated under cool white fluorescent tubes, 20-30 mol m⁻² s⁻¹, with continuous shaking (120 rpm) at 24°C for 48 h. Pear cell viability was measured 48 h after co-culture (three samples per well, three wells per treatment) following the FDA procedure [12].

2.7. Stability of cecropins in ECF

Cecropin B (5 M) was incubated for 1 h with ECF or alternatively with ECF boiled at 100°C for 20 min. Antibacterial activity of treated Cecropin B on E. amylovora growth was measured by optical density following the procedure described above. The putative Cecropin B degradation in ECF was checked by electrophoresis. ECF samples (2.5 ml) were mixed with 2.5 ml of Cecropin B (100 M) and incubated for various times (0, 1, 2.5, 5 or 24 h) at room temperature. Activity was stopped by freezing the samples at −20°C. Before loading on the gel, samples were mixed with 5 l of 2x sample buffer [16]. According to the Tricine-SDS-Page method of Schägger and Von Jagow [16], the gel was composed of a 5.5 cm resolving gel (16.5% T, 6% C), overlaid with a 1.5 cm spacer gel (10% T, 3% C) and a 1 cm stacking gel (4% T, 3% C). After electrophoresis at constant voltage 100 V, proteins were linked to the gel matrix with fixative solution (18% ethanol, 3.5% formaldehyde, 0.8 g l⁻¹ Coomassie Blue) then stained for 1 h in staining solution (25% ethanol, 0.35% formaldehyde, 1.2 g l⁻¹ Coomassie Blue) and destained overnight with 25% ethanol +0.35% formaldehyde.

3. RESULTS

3.1. Effect of tested peptides on Erwinia amylovora growth in vitro

Results of in vitro antibacterial assays are summarized in Table 2. Compounds isolated from plants did not show antibacterial activity against the bacteria in the range of tested concentrations (0-200 M). Among the substances from non-plant origin, only three peptides were bactericidal on CFBP1430: Cecropin B, SB-37 and T₄ lysozyme. Their minimal lethal concentrations were 5, 15, and 150 M, respectively. Whatever the substances tested, the observed MIC and MBC were nearly identical between the three different wells of each treatment in the same experiment and between the different experiments. Detailed results of Cecropin B activity are given in Fig. 1. Total absence of growth was observed only at 5 M, but sub-lethal concentrations (0.5, 1 and 2.5 M) caused serious growth retardation of the bacteria.

 

Table 2. Antibacterial activity of different peptides on E. amylovora

 

Fig. 1. CFBP1430 E. amylovora strain growth in the absence () and in the presence of Cecropin B at 0, 0.5 (),1 (), 2.5 (), 5 (♦) and 10 M (•). Data points represent the mean of nine replicates (three wells in three repeated experiments); vertical bars represent standard errors of the means.

 

3.2. Synergistic antibacterial effect of HEW lysozyme and Cecropin B

The growth of strain CFBP1430 was measured in the presence of both HEW lysozyme (150 M) and Cecropin B (1 M) (Fig. 2). At these concentrations, when applied separately, HEW lysozyme did not affect the growth rate of E. amylovora, while cecropin caused growth retardation in comparison with the control. However, when applied simultaneously, complete growth inhibition was observed at the given concentrations. Moreover, after plating on solid KB, the absence of colonies confirmed the bactericidal effect of Cecropin B (1 M) and HEW lysozyme (150 M) applied together.

 

Fig. 2. CFBP1430 E. amylovora strain growth in the absence () and in the presence of Cecropin B at 1 M (), HEW lysozyme at 150 M (•) or Cecropin B (1 M)+ HEW lysozyme (150 M) () respectively. Data points represent the mean of six replicates (three wells in two repeated experiments); vertical bars represent standard errors of the means.

 

3.3. Activity of Cecropin B and SB-37 on the E. amylovora/pear cell interaction

The viability of pear cells at the beginning of the experiments was 100%. After 48 h, controls showed no loss of viability (Fig. 3). In absence of antibacterial peptides, after 48 h of co-culture strain CFBP1430 provoked a loss of viability of pear cells which increased with the bacterial concentration. At 10³ cfu ml⁻¹, the viability was not significantly different from the control while at 10⁷ cfu ml⁻¹, mortality was nearly complete. The high viability of pear cells after co-cultivation with the mutant strain PMV6023 (10⁷ cfu ml⁻¹) indicated that the loss of viability observed with the strain CFBP1430 at the same concentration was evidently due to the pathogenic factors of this strain.

 
(17K)

Fig. 3. Viability of pear cells, after 48 h co-culture, in the absence and presence of Cecropin B (5 M) or SB-37 (15 M) and E. amylovora strains CFBP1430 and PMV6023. The viability was measured on samples of about 100 cells. For each treatment, histogram bars represent the mean of 18 measurements (three samples per well, three wells per treatment in two repeated experiments). Vertical bars represents standard errors of the means.

 

In presence of Cecropin B (5 M) or SB-37 (15 M), the viability of pear cells after 48 h was not affected. At concentrations lethal to bacteria, these peptides were non-toxic to pear cells. Cecropin B at 20 M was not toxic to pear cells but a concentration of 50 M caused 98% mortality (data not shown).

The addition of Cecropin B (5 M) and SB-37 (15 M) to the pear cells limited the mortality caused by strain CFBP1430 (Fig. 3). But, in the presence of pear cells, the lethal concentration of Cecropin B or SB-37 (5, and 15 M, respectively) was not sufficient to kill all bacterial cells. Indeed, the viability of cell suspension co-cultivated with bacteria at 10⁶ and 10⁷ cfu ml⁻¹ in presence of Cecropin B was close to the viability of cells co-cultivated with the same strain at a concentration 10 times lower (10⁵, and 10⁶ cfu ml⁻¹, respectively). This effect was less marked for SB-37.

3.4. Loss of activity of ECF-treated Cecropin B and SB-37 on E. amylovora growth

To test the stability of Cecropin B in pear tissue, these peptides were incubated with extracellular fluids extracted from pear leaves, then tested as described above for their activity against E. amylovora. The extracellular origin of ECF was previously confirmed by the absence of enzymatic activity of the G-6-PD cytoplasmic marker (results not shown). In the presence of ECF-treated Cecropin B (5 M), the growth of CFBP1430 was only slightly different from the control (Fig. 4) whereas in absence of ECF, 5 M of cecropin was lethal to bacteria. This decreased activity was also observed when Cecropin B was incubated with ECF which had been boiled beforehand at 100°C for 20 min, but to a lesser extent. In this case, the growth of CFBP1430 was similar to the bacterial growth in the presence of 1 M Cecropin B. This experiment, repeated three times, indicates a probable instability of Cecropin B in pear extracellular spaces. The same results were obtained with SB-37 peptide (data not shown).

 

Fig. 4. CFBP1430 E. amylovora strain growth in the absence () and in the presence of ECF (x), Cecropin B at 1 () or 5 M (♦), Cecropin B (5M) incubated with ECF (•), Cecropin B (5M) incubated with boiled ECF (). Data points represent the mean of six replicates (three wells in two repeated experiments); vertical bars represent standard errors of the means.

 

3.5. Electrophoretic analysis of Cecropin B degradation.

Cecropin B (1 g) was mixed (v/v) with ECF and incubated at room temperature during 0, 1, 2.5, 5 or 24 h. Electrophoretic analysis showed the progressive loss of the polypeptide band corresponding to Cecropin B (MW 3832) (Fig. 5). After 24 h (lane 5), in presence of ECF, no Cecropin B band could be detected by Coomassie Blue staining. When ECF was treated for 20 min at 100°C, no degradation of 1 g Cecropin B was observed even after 24 h.

 

Fig. 5. Electrophoresis gel of Cecropin B incubated with ECF. Lanes L: molecular weight ladder (Sigma ref. MW-SDS-17S), lanes C: Cecropin B (1 g), lanes 1-5: Cecropin B (1 g) incubated with pear leaf ECF, lanes 6-10: Cecropin B (1 g) incubated with boiled ECF. Periods of incubation were 0 h (lanes 1 and 6), 1 h (lanes 2 and 7), 2.5 h (lanes 3 and 8), 5 h (lanes 4 and 9) and 24 h (lanes 5 and 10). Only the separating gel layer is shown.

 

4. DISCUSSION

These results constitute the first comparative study of antibacterial efficiency realized on various peptides towards the same pathogen, following a unique procedure. The in vitro antibacterial activity test is a fast and simple method to measure the bactericidal activity of different peptides on a bacterial population. These tests have been achieved on a bacterial population growing under optimal conditions (high concentration, optimum culture medium and temperature). They provide information about bacteriostatic or bactericidal activity of peptides under specific conditions, and we can suppose that an antibacterial activity observed under optimal conditions for the bacteria will also be expressed in plant tissue environment where conditions are less favorable to the pathogen.

The results obtained in this study are consistent with previously published work. Purothionins were reported to inhibit both Gram-positive and Gram-negative phytopathogenic bacteria, but had a very low efficiency on E. amylovora with a MIC of 540 g ml⁻¹ [2]. This result is confirmed by our tests where no bactericidal activity has been found up to 1 mg ml⁻¹. In the case of puroindolines, which were recently isolated [17], our tests also confirmed preliminary studies showing a strong antifungal but a low antibacterial effect of theses peptides (pers. com. D. Marion). The possible mode of action of purothionins and puroindolines is a modification of the cell membrane permeability causing lysis, and this activity seems to vary between species and between strains of bacteria [18].

The action of lysozymes is a lytic action due to degradation of the bacterial cell wall peptidoglycan. Therefore, these peptides are mostly active against Gram-positive bacteria [8], which can explain the inefficiency of HEW lysozyme on E. amylovora, a Gram-negative bacterium. The higher lytic action of T₄ lysozyme, compared to HEW lysozyme or silk moth lysozyme, has already been reported [19]. Applied together with Cecropin B, HEW lysozyme (150 M) lowered the MBC of Cecropin B against E. amylovora from 5 M to about 1 M. These results confirmed the synergistic interaction of these two types of antibacterial peptides already observed in insect immunity studies [8]: cecropins cause membrane lysis while lysozymes degrade the cell wall.

Cecropins are small peptides for which high antibacterial activity on Gram-positive and Gram-negative bacteria, has been known for several years [3, 4, 6 and 7]. In addition, toxicity of cecropins towards mammalian cells has not been observed [8]. Among the different cecropins tested, Cecropin B and SB-37 showed a very efficient bactericidal activity while Shiva-1 had no activity on E. amylovora up to 50 M. The low homology (46%) between Shiva-1 and Cecropin B sequences [6] could explain this difference of activity. However, in another study, Shiva-1 had been reported to be more active than Cecropin B or SB-37 against Ralstonia solanacearum [20].

The decreased activity of Cecropin B and SB-37 after contact with pear cells or ECF has been clearly demonstrated in our study. Cecropin breakdown has already been observed in tobacco [21] and peach [22] and attributed to the action of plant serine proteases [21]. In our study even ECF pre-treated for 20 min at 100°C was able to partially decrease Cecropin B activity, whereas the same treatment of peach or tobacco ECF inactivated plant peptidases. Thus, the cecropin degradation mechanism taking place in pear is probably different from the one described in tobacco and peach.

The breakdown of cecropins by pear ECF challenges the use of these peptides for pear genetic engineering. The steady state level of a secreted protein results from the balance between its secretion and its degradation. Thus, the introduction of a cecropin gene with a constitutive promoter could drive a continuous synthesis of cecropin sufficient to maintain a bactericidal concentration in pear tissue despite its breakdown. An alternative could be the use of a pathogen inducible promoter leading to a high level of synthesis shortly after infection. The co-integration of a cecropin gene and a T₄ lysozyme gene could also constitute an interesting field of investigation. Because of their synergistic effect, a bactericidal activity could be obtained with low concentrations of both peptides in pear tissue. Finally, another approach could be the use of a more stable cecropin analog. Interestingly, a recent study demonstrated that the modification of a single amino acid in Cecropin B modified the protease sensitive site of this protein, and diminished its degradation by leaf ECF [23]. If this modified Cecropin B maintains its strong antibacterial activity against E. amylovora without increasing its toxicity against plant or animal cells, its integration in pear tissue could be an efficient strategy.
 

ACKNOWLEDGEMENTS
 

We thank Dr H.S. Aldwinckle for his careful reading of the manuscript.
 

REFERENCES

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005