**3. Discussion**

Although recent studies have revealed that *V. dahliae* modulates the xylem microbiome in olive plants by increasing the diversity of bacterial communities when the pathogen is present in the soil [27], the influence that olive cultivar, biological treatments, and their interaction could have on the endosphere against the viability of the infectious structures of *V. dahliae* has not yet been reported. Thus, this work represents a first approach to elucidate the role of cultivar and biological treatments in modifying the effect of the endosphere contents on the pathogen in olive plants.

In this study, 6-month-old potted olive plants were used because, based on our extensive experience working with the olive tree and *V. dahliae* pathosystem, it is not only the age limit to reproduce the symptoms of the disease by artificial inoculation of plants in small pots [3,14], but also considering the limitations that the size and physiological structure of plant tissues could have in the extraction of contents from the endosphere by Cadahía's method [28]. Regarding this last aspect, we have to consider that Cadahía calls 'sap' to the total liquid extract from the endosphere (called OSE in this study) that comes from both the xylem and phloem of the plant. The ideal would probably be to obtain raw sap (xylem) and elaborated sap (phloem) separately to evaluate the effect of the raw sap alone, but despite this limitation, we chose this method instead of the sap-extraction method using a Scholander chamber [29] for the following reasons: (i) because it has been demonstrated that the extract of pure sap from the xylem of olive plants contains a wide diversity of bacterial communities [27], which could mask the effect of OSE on the germination of conidia in *in vitro* sensitivity tests; (ii) consequently to this first reason, the extraction of OSE using Cadahía's method guaranties the absence of living microorganisms, as well as cellular debris in the extract, since the plant material is subjected to consecutive immersion and freezing treatments in diethyl ether during the extraction process; and (iii) because it has been recognized as reliable method to determine the nutritional levels of plants, since results using 'sap-like' extracts (OSE) contrast well with those of the yield and quality of the harvest of several fruit crops, including the olive tree, in more than 40 years of experience [28].

Only the effect of OSE on the germination of conidia was evaluated in this study because conidia are the infectious structures of *V. dahliae* directly affected by the sap, considering the life cycle of the pathogen, i.e., the pathogen infects plants through the root by the germinated MS and then systemically colonizes the infected plants by producing conidia in the xylem vessels [1,6]. In addition, the treatment combinations (treatments and/or mode of application) evaluated in this study were selected because they all resulted in high effectiveness in inhibiting the viability of the infectious structures of *V. dahliae*, as well as reducing the progress of the disease in previous studies by López-Moral et al. [3].

Our results revealed that OSE from 'Frantoio' (tolerant) showed higher RGI values than that from 'Arbequina' (moderately susceptible) and 'Picual' (highly susceptible), whereas the inhibition did not differ markedly between the last two cultivars (*Experiment I*). In 'Picual', the influence of OSE from the treated plants on the inhibition of conidia germination varied significantly between the evaluated treatments. In this cultivar, the BCA *B. amyloliquefaciens* PAB-024, and a phosphonate salt [copper phosphite (Phoscuprico®)] were the most effective in inhibiting conidia germination by the OSE (*Experiment II*). In addition, when the four selected treatments were applied to the three olive cultivars, their influence on the effect of OSE on the inhibition of conidia germination was not significant between treatments for 'Frantoio', whereas moderate and markedly significant differences between treatments were observed for 'Arbequina' and 'Picual', respectively (*Experiment III*). Finally, although some differences can be observed between both positive and negative controls in RGI as the OSE concentration increases (Figures 2 and 3), the EC50 data did not show significant differences between either control in any case. Therefore, these data suggest that the biotic stress caused by the infection of the pathogen in the plant does not influence the effect of OSE on the conidial germination of *V. dahliae*.

As a first conclusion, our results indicate that the influence of biological treatments against the pathogen depends on the genotype, since the greater the resistance of the cultivar, the lower the influence of the treatments on the ability of OSE to inhibit conidia germination. Thus, the results suggest that the high tolerance to *V. dahliae* conferred by the 'Frantoio' genotype prevails over the treatments, even those that were more effective against the pathogen in susceptible cultivars. These results are in agreement with those obtained recently by López-Moral et al. [7], who determined the influence of cultivars and biological treatments on the effect of root exudates from olive plants on the viability of MS and conidia of *V. dahliae*. These authors demonstrated that root exudates induced germination of conidia and MS of *V. dahliae* and that the genotype significantly affected this ability; the root exudates from 'Frantoio' did not show a significant effect on the induction of MS and conidia germination compared to the control, whereas those from 'Arbequina' and 'Picual' showed a moderate and marked effect, inducing the viability of both MS and conidia of *V. dahliae*, respectively [7].

Regarding genetic resistance, our study reveals new knowledge about the relationship that the sap and the olive cultivar could have, favouring or interfering with the colonization of the xylem by the pathogen. However, in order to elucidate how the cultivar factor could influence the effect of its sap on the colonization of the pathogen, further interaction studies with the xylem anatomy of each cultivar should be conducted in the future. In fact, previous studies that evaluated the anatomy of the xylem of healthy olive trees of cvs. Frantoio and Picual showed significant differences not only in the parameters related to water transport but also in the density of vessels associated with a larger or smaller conduction area in the xylem tissue, both parameters being significantly higher in 'Frantoio' than in 'Picual' [30].

Regarding the influence of the treatments on the effect of OSE on the viability of the conidia, our results are also in agreement with those obtained by López-Moral et al. [7]. These authors revealed that the root exudates from plants of the three cultivars (Frantoio, Arbequina, and Picual) treated with the same four treatments evaluated in *Experiment III* of the present study showed significant differences in their effect on MS and conidia germination, and the genotype also significantly affected this ability. In this case, the treatment with *A. pullulans* AP08 was most effective, showing a significant effect inhibiting conidia germination in 'Arbequina' and MS germination in 'Arbequina' and 'Picual', but non-significant effect in these two parameters was observed in 'Frantoio'. Regarding the rest of the evaluated treatments, the root exudates from plants treated with copper phosphite, potassium phosphite, and *B. amyloliquefaciens* PAB-24 gave rise to a significant inhibition in the germination of conidia or MS but only in cv. Arbequina [7]. In addition, previous studies conducted by López-Moral et al. [19] to evaluate these treatments as potential inducers of host resistance against VWO showed that the two BCAs *A. pullulans* AP08 and *B. amyloliquefaciens* PAB-024, as well as the phosphonate salt Phoscuprico®, had

the ability to accumulate jasmonic acid (JA) and JA-isoleucine in leaves, stem, or roots of treated olive plants of cv. Picual. These results also suggest an implication of JA in the host resistance induced by these treatments. This last aspect could be directly related to our results, since it is well known that xylem infections by vascular pathogens cause drastic metabolic changes in the cells of the xylem parenchyma adjacent to the infected vessels. These metabolic changes lead to the accumulation of different proteins and secondary metabolites in xylem sap during pathogen colonization, including pathogenesis-related proteins (PR proteins), enzymes (e.g., peroxidases, proteases, xyloglucan-endotransglycosylase, xyloglucan-specific endoglucanase protein inhibitor), phenols, phytoalexins, and lignin, which help to enhance the natural defence mechanisms of the plant [31–35]. On the other hand, due to the fact that the OSE did not have a live microbiome as a consequence of the extraction method used, studies to determine whether the treatments applied in this work could influence the modification of the xylem microbiome of olive plants must be conducted. In this way, recent studies performed by Anguita-Maeso et al. [27] determined that the xylem microbiome of olive plants inoculated with *V. dahliae* increases the diversity of bacterial communities compared to non-inoculated plants. However, how the xylem microbiome could be modified by biological treatments favouring *V. dahliae* inhibition is still uncertain.

In summary, the method used in this study to obtain endosphere contents of olive plants, called OSE or 'sap-like', for further analysis in the laboratory against *V. dahliae* can be considered valid and useful, since all our results agree with those obtained in previous studies [3,7,19]. The knowledge generated here represents a first approach in the study of how genotype and/or biological treatments can influence the extracts of olive plants by inhibiting the germination of conidia or of MS of *V. dahliae*. This knowledge could be useful in the future to prevent infections or mitigate the progression of disease within the framework of the current 'from-farm-to-fork' strategy to obtain safe and healthy fruits.
