All the studied strains showed high efficacy in protecting pear fruitlet slices against fire blight (Table 1). Although no significant differences in the severity of the disease were found after their preventive treatment with individual strains (degree of infection 0.0-0.6 compared to untreated 4.0), the 59M strain deserves attention, as it protected the slices at 100% (Fig. 1). The A506 and C9-1 strains used for comparison also proved to be very effective (efficacy 96.5% and 95.0%, respectively). The comparison of the efficacy of the studied strains on pear fruitlet slices with that on apple flowers and shoots (Mikiciński et al. 2020) shows some differences, which may be related to the different biochemical compositions of the protected organs and the possibility of their compounds being used by bacteria. The 43M strain, the efficacy of which in the protection of pear fruitlet slices in the current study was 91.3%, did not show the ability to inhibit the growth of E. amylovora on five agar media (NAS, King B, LB, R2A and NAG), while 3M and 59M formed inhibition zones on all these media. However, L16 highly inhibited pathogen growth only on NAS, King B, and NAG (Mikiciński et al. 2020). The demonstration that some strains limited the growth of the pathogen only on certain media indicates that their composition is essential for such activity. For example, pear tissue was found to have a low concentration of histidine and a high concentration of alanine (Lewis et al. 1964). Studies by Wright et al. (1998) on the potential of strain Eh318 Pantoea agglomerans to biosynthesize the antibiotics pantocin A and pantocin B, which are involved in the biocontrol of fire blight, showed that the presence of histidine abolished the activity of pantocine A, whereas the presence of arginine abolished the activity of pantocine B. Ishimaru et al. (1988) also proved that antibiotic herbicolin O produced by the C9-1 strain (previously E. herbicola, currently Pantoea vagans) was inhibited by 1-histidine. Treatment of immature pear fruit slices with whole cells of strain C9-1 or with partially purified herbicolins promoted a significant reduction in the severity of fire blight. Research by Wodziński et al. (1987) showed that the ability of the strains Eh252 and Eh318 to produce antibiotics on a glucose-aspartic medium was inhibited by the addition of casamino acid. As may be expected, depending on the conditions, the production of toxic metabolites by bacteria may undergo various modifications in the environment, which translates into their antagonistic activity against E. amylovora. Gerami et al. (2013) showed that the E. amylovora antagonists Pseudomonas putida and Serratia marcescens were effective in planta but not in vitro. With these considerations in mind, the research of Duffy and Défago (1999) documented liquid culture assays with different Pseudomonas fluorescens strains. Their experiments broadened the information on the biosynthetic regulation of antimicrobial compounds, including antibiotics and siderophores. Several minerals and carbon sources were proven to have a differential influence on the production of 2,4-diacetylphloroglucinol, pyoluteorin, pyrrolnitrin and pyochelin by the model strain CHA0. For example, siderophore pyochelin production was increased by Co2+, fructose, mannitol, and glucose.
Table 1
Severity of fire blight on pear fruitlet slices cv. Conference after protective treatment with bacterial suspensions of tested strains
Strain
|
Distribution of fire blight incidence on fruitlets slices among susceptibility classes
|
Severity of fire blight
|
Efficacy
(%)
|
0
|
1
|
2
|
3
|
4
|
43M
59M
3M
L16
A506
C9-1
Control (water)
|
70.0
100.0
79.0
86.0
92.5
87.5
0.0
|
25.0
0.0
5.0
0.0
2.5
5.0
0.0
|
5.0
0.0
5.0
0.0
5.0
7.5
0.0
|
0.0
0.0
7.0
0.0
0.0
0.0
0.0
|
0.0
0.0
4.0
14.0
0.0
0.0
100.0
|
0.35 ± 0.1a
0.00 ± 0.0a
0.48 ± 0.3a
0.60 ± 0.7a
0.13 ± 0.1a
0.20 ± 0.3a
4.00 ± 0.0b
|
91,3
100
88.0
85.0
96.8
95.0
-
|
Assessment was done 7 days after inoculation with E. amylovora; rating scale of severity: 0–4 (0 – no necrosis, 4 – total necrosis of slide); each treatment consisted of 40 slices (10 × 4 repetitions); means within column followed by the same letter are not significantly different at P < 0.05 according to Newman-Keuls test (means ± SE) |
Our study showed that in strain 59M, the phlD, pltB, pltC and gacA genes were detected, while in strain 3M, only the genes prnD and gacA were detected. However, none of the sought genes were found in L16 (Table 2) or in the earlier studied strain 49M Pseudomonas graminis (Mikiciński et al. 2016b). These results indicate that different mechanisms may play a role in the antagonism of these strains against E. amylovora, including the presence and expression of genes determining antibiosis other than those that were detected. In this context, it is interesting that Cabrefiga et al. (2007) showed that the strain EPS62e P. fluorescens, which was selected for its high efficiency in controlling E. amylovora infection, does not synthesize antibiotics (PCA, Phl and Prn) and does not have the appropriate biosynthetic genes. The authors surmise that competition for certain available nutrients is another mechanism that may be involved in the biocontrol of E. amylovora by this strain. EPS62e was also found to exhibit a more versatile spectrum of nutrient sources since it used 51 of 95 carbon sources compared to the 27 used by E. amylovora (Cabrefiga et al. 2007).
Table 2
Potential mechanisms of bacterial species first time recognized for their biocontrol activity against fire blight (Erwinia amylovora)
Strain
|
Production of
|
Nicoti-
nic
acid
degra-
dation
|
Moti-
lity
|
Presence of genes*
|
Side-
ropho-
res
|
SA
|
IAA
|
AHL
|
HCN
|
Biosur-
factant
|
phlD
|
phzA, phzF,
phzC, phzD
|
pltB,
pltC
|
prnD
|
gacA
|
L16
Pseudomonas
vancouverensis
|
+*
|
+*
|
++*
|
‒*
|
+*
|
+*
|
++*
|
+*
|
‒
|
‒ ‒ ‒ ‒
|
‒ ‒
|
‒
|
‒
|
3M
Pseudomonas
chlororaphis
subsp. aureofaciens
|
+
|
‒
|
‒
|
+
|
+
|
‒
|
++
|
++
|
‒
|
‒ ‒ ‒ ‒
|
‒ ‒
|
+
|
+
|
59M
Pseudomonas
protegens
43M
Enterobacter ludwigii
|
+
‒
|
‒
‒
|
+
+
|
‒
‒
|
+
+
|
‒
‒
|
++
++
|
++
+++
|
+
nd
|
‒ ‒ ‒ ‒
nd
|
+ +
nd
|
‒
nd
|
+
nd
|
A506
Pseudomonas fluorescens
|
+
|
‒
|
+
|
‒
|
+
|
‒
|
++
|
++
|
‒
|
‒ ‒ ‒ ‒
|
‒ ‒
|
‒
|
‒
|
Pf5
Pseudomonas
protegens
C9-1
Pantoea vagans
|
nd
+
|
nd
‒
|
nd
+++
|
nd
++
|
nd
+
|
nd
+++
|
nd
+++
|
+
++
|
+
nd
|
‒ ‒ ‒ ‒
nd
|
+ +
nd
|
+
nd
|
+
nd
|
*Ability of siderophores, salicylic acid (SA) and AHL production: (+) positive, (‒) negative; IAA production – supernatant pink discoloration: weak (+), moderate (++), strong (+++), negative (-); Nicotinic acid degradation - diameter of bacterial colony: (+) > 2 mm, (++) 2–4 mm, (+++) < 4 mm; HCN production - change in the colour of the filter paper from yellow to light brown, brown or reddish-brown: weak (+), moderate (++), strong (+++); Motility - diameter of the bacterial colonies: (‒) < = 5 mm, (+) 6–15 mm, (++) 16–30 mm, (+++) 31–50 mm, (++++) > = 51 mm; Biosurfactant production - diameter of the circle created by oil: (+) 1–6 mm, (++) 7–15 mm, (+++) 16–30 mm, (++++) > 30 mm, (‒) negative; presence of genes: (+), (‒) negative; nd- not done. |
We have demonstrated that strains L16, 3M, 59M, A506 and C9-1 produced siderophores (Table 2) similar to the previously described strain 49M P. graminis (Mikiciński et al. 2016b). Siderophores are a major group of iron-chelating agents that play a key role in bacterial iron homeostasis (Schalk et al. 2020). The production of these compounds can be assumed to be one of the factors responsible for the antagonism and protective abilities of our studied strains. The production of siderophores was also confirmed in the EPS62e strain (Cabrefiga et al. 2007). Mondal et al. (2000) found that P. fluorescens, through siderophore-mediated competition for Fe+ 3, reduces the development of Xanthomonas axonopodis pv. malvacearum, causing economic losses in cotton cultivation. The strains of species Pseudomonas fulva and P. putida have also been proven to reduce decay caused by Dickeya sp. on potato tubers, showing the ability to compete for iron ions by the production of siderophores but not inhibiting pathogens on tryptic soy agar (Czajkowski et al. 2012). Regarding our research, the efforts of Temple et al. (2004) to elucidate the bioavailability of iron to P. fluorescens strain A506 on flowers of pear and apple deserve attention. The flowers of the mentioned plants were found to represent an iron-limited environment for this strain. A506 produces a pyoverdine siderophore on iron-depleted media but not in media amended with 0.1 mM ferric chloride. The authors concluded that production of the iron-induced antibiotic by A506 is unlikely to occur in the field and that competition for nutrients and sites may be the main mechanism of its activity (Temple et al. 2004). The possibility of producing siderophores in nature, proven in vitro as a possible mechanism of antagonism, is also worth considering. Santos Kron et al. (2020) demonstrated that strain F9 P. orientalis harbours biosynthesis genes for the siderophore pyoverdine as well as for the antibiotics safracin and phenazine. They believe that mechanisms related to antibiosis and competition are related to a number of factors or causes and not solely dependent on antibiotic and/or siderophore production. Youard et al. (2007) proved that some species of the genus Pseudomonas (e.g., P. aeruginosa, P. fluorescens) produce pyochelin and enantio-pyochelin, which are siderophores formed via the condensation of salicylic acid and two molecules of cysteine.
Of our strains tested, only L16 produced SA (Table 2), which is considered an important plant hormone, mediating host responses to biotic stress. The treatment of apple and pear trees with the commercial preparation Bion, of which the active compound is S-methylbenzo-1,2,3-thiadiazole-7-carbothiate (BTH), a functional analogue of SA, is well-documented to result in a significant increase in resistance to fire blight (Norelli et al. 2003; van der Zwet et al. 2012). Skłodowska et al. (2010) showed that BTH applied to apple trees caused changes in the prooxidant-oxidant balance in leaves but in different ways in the enzymatic and nonenzymatic antioxidants. Wang and Liu (2012) found that exogenously applied SA on navel orange (Citrus sinensis) plants significantly increased the endogenous free and bound SA. At the same time, pretreatment with SA plants and inoculation with Xanthomonas axonopodis pv. citri was proven to lower citrus canker incidence rate, which was accompanied by repression of bacterial growth at the lesion sites. Additionally, treatment with pakchoi (Brassica campestris ssp. chinensis Makino) with SA resulted in a reduction in clubroot (Plasmodiophora brassicae brassicae) in the susceptible cultivar Xinxiaqing (Xi et al. 2021). According to Zheng et al. (2020), exogenous SA stimulated the resistance response and initiated a similar defence pathway compared to Phytophthora infestans infection in the late blight-resistant potato genotype. Studies by Anand et al. (2008) showed that Nicotiana benthamiana plants treated with SA showed decreased susceptibility to Agrobacterium infection related to a reduction in pathogen growth, a decrease in its attachment to plant cells and inhibition of virulence gene induction.
Among all the tested strains, C9-1 and L16 showed the highest activity in the production of indole acetic acid (IAA), and only strain 3M did not produce this compound (Table 2). Although IAA is well known as a plant growth and development regulator, it can also plays a role in interactions between plants and microorganisms influencing plant resistance to infection by pathogens (Kunkel and Harper 2018). By mutagenesis analysis Yang et al. (2007) found that IAA may directly impact virulence gene expression in Erwinia chrysanthemi (Dickeya didantii) and, when applied exogenously, stimulated the production of pectate lyase, an enzyme involved in plant cell wall maceration and causing soft rot. Overall, there is no information in the literature about the direct effect of IAA produced by antagonistic bacteria on pathogens causing diseases on various plant organs.
Another issue is the degradation of nicotinic acid (NiAc), which is a specific growth factor for E. amylovora in vitro. The activity of all tested strains in this area was found to be moderate and only high for the C9-1 strain (Table 2). Both NiAC and NiNh2 (nicotinamide) are present on the hypanthia of apple and pear blossoms (Paternoster et al. 2009) and are sites of massive tree infections by this pathogen. The degradation of these compounds by the bacteria used to control fire blight is expected to reduce the disease in the orchard as well, as has been demonstrated in apple blossoms under greenhouse conditions (Paternoster et al. 2010). Additionally, the total content of NicAc and NicNH₂ was determined approximately 6 to 23 times higher in pear and approximately 1.2 to 3.5 times higher in apple than the amounts of NicAc or NicNH₂ necessary to support maximum E. amylovora growth in vitro (Paternoster et al. 2011).
We also tested the ability of our strains to produce N-acyl homoserine lactones (AHLs), which are considered bacterial mediators and autoinducers. Thanks to these compounds, bacteria can interact with each other in a process called quorum sensing (QS), regulating some life processes (Venturi 2006). Such an ability was observed only in the 3M and C9-1 strains, but the latter was more active (Table 2). Fray et al. (1999) established that AHLs play a critical role in plant/microbe interactions. The N-(3-oxohexanoyl)-L-homoserine lactone (OHHL) induces exoenzymes that degrade the plant cell wall by the pathogenic bacterium Erwinia carotovora. Conversely, the antifungal activity of the biocontrol bacterium Pseudomonas aureofaciens 30–84 is related (at least in part) to phenazine antibiotics with synthesis regulated by N-hexanoylhomoserine lactone (HHL) (Fray et al. 1999). Jakovljevic et al. (2008) demonstrated the presence of AHLs related to the mechanism of QS in bacterial isolates of the species Erwinia billingiae showing high antagonistic activity towards E. amylovora on pear fruitlet slices. This bacterium may be involved in controlling fire blight. Other researchers dealing with this issue have proven that AHL particles regulate the production of secondary metabolites such as pyrrolnitrin or phenazine but also the biofilm formation of P. chlororaphis strain PA23 (Selin et al 2012). A research of Shah et al. (2020) indicated that QS governs the expression of approximately 13% of the PA23 genome, affecting different functions in secondary metabolite production and general metabolism.
All tested strains demonstrated the ability to produce hydrogen cyanide (HCN) (Table 2), a gaseous compound, toxic towards a number of plant pathogens. Anand et al. (2020) demonstrated that HCN can act as an intracellular and extracellular volatile signalling molecule contributing to growth, motility, and biofilm formation. It is well involved in the production of other secondary metabolites, such as siderophores and phenazines. These authors suggest that bacteria use endogenous HCN not only to control their own cellular functions but are also able to remotely influence the behaviour of other bacteria sharing the same environment. Extensive literature data indicate the importance of HCN biosynthesis, however, mostly to rhizosphere bacteria (Haas and Defago 2005; Zdor 2014; Sehrawat et al., 2022. Nagarajkumar et al. (2004), by testing 14 strains of P. fluorescens, showed that all of these strains produced HCN, but two (PfMDU2 and PfMDU3) revealed the highest activity. The authors suppose that the increased production of HCN by the PfMDU2 strain might be related to the highest inhibition of mycelial growth of Rhizoctonia solani in vitro. Voisard et al. (1989) believe that the ability of bacteria to produce HCN plays an important role in the biological protection of plants against black root rot of tobacco (Thielaviopsis basicola). As an example, they suppressed this disease control by P. fluorescens, proving it by mutagenesis of the CHA0 strain obtained by insertional inactivation of HCN production. Strain LBUM300 Pseudomonas sp. was also found to significantly repress Clavibacter michiganensis subsp. michiganensis growth in vitro, whereas its respective HCN nonproducing mutant (hcnC−) showed less or none of such activity. LBUM300 was also capable to significantly reduce bacterial canker development on tomato plants and the rhizospheric population of its causal agent (Lanteigne rt al. 2012). These studies suggest that cyanogenesis is an important trait of biological control agents.
In the context of the discussion of our results, the conclusion from the study by Anand et al. (2020) is relevant, suggesting that while HCN is a major contributor to the restriction of Phytophthora infestans mycelial growth, it does not play a major role in the inhibition of other disease-related features, such as zoospore germination or infection of plant tissues. There are reports indicating that many organisms have developed different strategies to avoid cyanide intoxication, such as chemical conversion of HCN to thiocyanate by the rhodanese enzyme or the use of cyanide-insensitive oxidases (Frangipani et al. 2014; Cipollone et al. 2017).
Our strains possessed some other mechanisms that potentially enhance their colonization, adaptation and survival in E. amylovora host plants. Motility is noteworthy. All strains revealed this ability, with the highest degree in 43M and the lowest in L16 and Pf5 (Table 2). According to Czajkowski et al. (2012), motility enhances the colonization of surfaces and may be an advantage in competing with motile Dickeya spp. both in plants and in soil. Strains L16 and C9-1 produced biosurfactant, but the latter was shown to be more active (Table 2).
The studied strains representing bacterial species recognized for their biocontrol activity against fire blight (Erwinia amylovora) showed good efficacy in the protection of apple flowers and leaves (Mikiciński et al. 2020) and pear fruitlet slices. Among the studied mechanisms that could be related to this activity, the L16 strain was characterized by the highest antagonistic activity, showing the ability to produce siderophores, biosurfactant, HCN, SA and IAA. The L16 strain also degraded nicotinic acid. The 43M strain showed the lowest activity, producing only IAA and degrading nicotinic acid. A study on the detection of genes encoding antibiotics characteristic of pseudomonads showed the presence of prnD and gacA in the 3M strain and phlD, pltB, pltC and gacA in 59M. However, none of the sought genes were detected in the L16 strain. When analysing the presence of such genes, it is also important to consider their expression on the protected organ of the plant.
In conclusion, we would like to emphasize that using in vitro tests, it is not possible to clearly determine how a given strain will behave on or in the treatment of various plant organs. Laboratory tests only indicate the existence of potential mechanisms that may or may not appear on a protected plant. An important issue is also the possibility of interaction between the various mechanisms of antagonism, especially in the context of synergism between those mechanisms to obtain the greatest possible efficiency. However, the results of in vitro studies are useful and may be helpful in pointing out the inactivity of some mechanism, which may also be of great practical importance. The success of the application of beneficial bacteria depends largely on various factors in a specific environment, including the capacity of the adaptation of these bacteria and their utilization of nutrients.