Pseudomonas chlororaphis PCL1606 is a rhizobacterum isolated from avocado roots characterized by 2-hexyl,5-propyl resorcinol (HPR) production, the main compound involved in biocontrol (Cazorla et al. 2006), but also in other relevant phenotypes, such as biofilm formation and colonization (Calderón et al. 2019; Arrebola et al. 2019). A previous analysis of additional antifungal antibiotic compounds produced by PCL1606 revealed the production and presence of coding genes for hydrogen cyanide (HCN) and pyrrolnitrin (PRN); however, phenazine production and the presence of coding genes were not found in PCL1606 (Calderón et al. 2015). Additionally, the presence of genes homologous to the fitABCDEFGH cluster, encoding a putative cytotoxin similar to Fit toxin, was detected in the genome of P. chlororaphis PCL1606 (Calderón et al. 2015). According to Ruffner et al. (2015), the Fit cytotoxin is restricted to a particular group of rhizobacteria comprised of P. protegens and P. chlororaphis, and it is strongly correlated with high insect toxicity.
PCL1606 displayed insecticidal activity at the same level as the control strains P. chlororaphis subsp aurantiaca (former P. aurantiaca) BL915 and P. protegens Pf-5, where HPR and Fit production were first described, respectively (Nowak-Thompson et al. 2003; Péchy-Tarr et al. 2008). On the other hand, the nonantagonistic rhizobacterium P. alcaligenes AVO110 did not show insecticidal capacity, did not have the fit or any antifungal antibiotic genes in its genome (Pintado et al. 2021), and did not produce antifungal secondary metabolites (Pliego et al. 2007; Pintado et al. 2021). The typical symptoms produced by the wild-type strain PCL1606 correspond to mortality accompanied by intense melanization of the G. mellonella larvae from 24 h postinoculation. Melanization is a typical insecticidal symptom corresponding to the synthesis and deposition of melanin to encapsulate pathogens at the wound site followed by haemolymph coagulation and opsonisation (Tsai et al. 2016).
Single and double mutants impaired the production of antifungal hydrogen cyanide (HCN) or pyrrolnitrin (PRN) compounds in PCL1606, resulting in a mortality reduction of G. mellonella larvae, but with no differences among them. This suggests that both compounds could only contribute to the insecticidal background and were not the main insecticidal compounds produced, as previously described for other insecticidal strains (Flury et al. 2017). It has been previously described that the insecticidal virulence of P. protegens CHA0 and P. chlororaphis PCL1391 mutants lacking one or several antibiotics, such as 2,4-diacetylphloroglucinol, phenazine, pyrrolnitrin or pyoluterin, was also not reduced (Flury et al. 2017).
The role of HCN and PRN on insecticidal and nematicidal activity by antagonistic bacteria has been illustrated by other authors (Nandi et al. 2015; Kang et al. 2019), and our results indicate their involvement in the insecticidal phenotype displayed by the model strain PCL1606. In addition, this study included the PCL1606 derivative mutant defective in gacS, since gacS is a global regulator that can interfere with activities related to the production of secondary compounds, such as antifungals and insecticidals (Saraf et al. 2011; Nandi et al. 2015). Previous studies have reported that mutants deficient in the global regulator gacS showed no insecticide activity (Kang et al. 2019). These results do not agree with our observation where the impaired mutants in the gacS gene of PCL1606 still retain insecticidal activity, suggesting a gacS-regulated independent pathway responsible for the insecticidal phenotype.
For this, the effect of HPR and Fit products was studied. Single mutants of these genes scarcely lowered G. mellonella larval mortality, but the absence of both genes completely impaired the insecticidal activity of the derivative bacteria, even in the presence of hcnB and prnC genes.
The insecticidal and nematicidal effectiveness of the Fit cytotoxin in P. protegens and P. chlororaphis has been documented (Kupferschmied et al. 2013; Péchy-Tarr et al. 2008). Heterologous expression of the Fit toxin in Escherichia coli resulted in the capacity of the bacterium to kill the insect host upon injection (Kupferschmied et al. 2013). In addition, the deletionally defective fitD mutants of P. protegens Pf-5 and CHA0 show a decrease in insecticidal activity, demonstrating that FitD makes an important contribution to insect virulence but may not be essential for the capacity for these bacteria to multiply in the infected host. The fact that the fitD mutants still retained a certain level of toxicity indicated to the authors that additional factors may contribute to anti-insect activity (Péchy-Tarr et al. 2008), in agreement with our observations on PCL1606. However, the PCL1606 derivative mutant with the fit toxin gene disrupted showed the capacity to kill G. mellanella at almost 80% 30 h postinfection. This result is not only due to HCN and PRN, as reported by the double derivative mutant in HPR and Fit toxin, with no virulence at this time postinfection. On the other hand, the double mutant HCN and PRN still showed that 60% of G. mellonella larvae died. Therefore, we can conclude that HCN and PRN could contribute to the insecticidal capacity of PCL1606, but they could be considered accompanying compounds. Thus, HPR and the Fit toxin could be considered the main toxins responsible for insecticidal activity. Single derivative mutants in HPR and Fit toxin displayed the same levels of G. mellonella mortality, suggesting that the mutation of one of them can complement the mortality index by the other gene and vice versa. However, the mortality shown by the derivative mutant PCL1606::darBfitD was innocuous after approximately one day and a half, when the first sign of mortality started to appear.
HPR has been described as the main antibiotic against fungal pathogens such as Rosellinia necatrix or Fusarium oxysporum (Cazorla et al. 2006). However, HPR has been revealed as a versatile compound involved in bacterial adhesion, colony morphology and typical air-liquid interphase pellicles produced by PCL1606 (Calderón et al. 2019). Due to the chemical nature of HPR, it is possibly not an integral component of the biofilm matrix, together with the evidence of a regulatory role in Photorhabdus asymbiotica (Hapeshi and Waterfield 2017), which extends its putative role in PCL1606.
The combined antimicrobial and regulatory roles of HPR in PCL1606 could explain several results obtained in the present study. The toxic nature of HPR against different organisms (Kanda et al. 1975) could directly affect G. mellonella cells, as happens with antifungal phenazines (Wand et al. 2013), helping the insecticidal characteristics of P. chlororaphis PCL1606. This insecticidal activity would rival FitD in toxicity, thus justifying the single mutants PCL1606:darB and PCL1606:fitD results in comparison with double mutant PCL1606::darBfitD, also revealing that because of the absence of HPR and Fit toxin, there was no mortality due to HCN and PRN. On the other hand, HPR could have additional regulatory roles on secondary metabolites since alkylresorcinols can be involved as signalling molecules in a novel quorum sensing two-component regulatory system (Brameyer et al. 2015). Belonging to these sets of secondary metabolites could be chitinase and phospholipase C, putative genes that have been found in the PCL1606 genome (PCL1606_RS13585 as chitinase and PCL1606_RS14060 as phosphatase C). Both enzymes have also been reported to be important in insect pathogenicity (Flury et al. 2016). This also helps to explain the gacS mutant results, whose killing ability was not affected, suggesting a regulation of all the compounds at a higher hierarchy, but this regulatory role of HPR is still a hypothesis under study.
In summary, PCL1606 could produce the Fit toxin, a described compound with insecticidal capacity, and HPR, which has been shown to have insecticidal potential, to which its fungicidal character and its possible regulatory role must be added. This confirms HPR as one the main compounds produced by P. chlororaphis PCL1606 involved in the beneficial phenotypes displayed by this model bacterium.