Two clones of Vitis vinifera L. cv. Tempranillo plants (VN40 and RJ43) grafted on Richter-110 rootstock were provided by two commercial nurseries (Vitis Navarra and Viveros Villanueva, Navarra, Spain, respectively). The plants (n = 30 per clone) were grown in a containment level 2 greenhouse (NCB2), in 5 L pots containing an enriched nutrient substrate of organic matter (90 %), sphagnum peat (160 g/L), calcium carbonate (7 g/L), NPK fertiliser (1.5 g/L) and trace elements (PotgrondH, Klasmann-Deilmann GmbH, Germany). The plants were grown for 2 months, with a 16 h day/8 h night photoperiod at an average room temperature of 18 ºC and were irrigated to field capacity when necessary.

When the plants had 10 to 12 fully expanded leaves, they were treated with either R. okamurae extract (RO, 6 g/L), water (W) or copper (BB, cuprous oxide 4.2 g/L, Nordox 75 WG, NordoxIndustrier, Oslo, Norway) spraying 3 times on leaves once every 48 hours (Figure 1). The soil in the pots was protected from the rinse with filter paper during each application. Ten plants were used per treatment type and clone type. To all the treatments (W, BB, RO) 0,1 % Retenol (Daymsa, Zaragoza, Spain) was added as a coadjutant.

dPB= days post biostimulation, dPI= days post inoculation.

Three crucial time-points were studied: S ₀ sampling, the initial sampling prior to the application of the treatments (Day 0); S ₁ sampling, 24 h after foliar application of the last treatment, referred to as the biostimulation time-point (1 day post-biostimulation, 1dPB); S ₂ sampling, 24 h after P. viticola inoculation, referred to as the post-inoculation time-point (1dPI) (Figure 1).

To assess the initial physiological state of the plants, the photochemical parameters of the leaves and plant height were determined in S ₀ (n = 60 plants). At the biostimulation time-point (S ₁, 1dPB), the 5th leaf (starting from the apex) was sampled, and a leaf from two plants was pooled to create five replicate samples per treatment type/clone. Leaves were milled in liquid nitrogen and were then used for gene expression and secondary metabolites analysis (including enzyme activity and polyphenols). The 7th leaf was used to study photochemical parameters in n=10 plants/treatment. In this case, while the same sampling approach was used 1 day after inoculation (S ₂, 1dPI), the 4th and 6th leaf of two plants were pooled together for gene, polyphenol and enzyme analysis (n = 5 biological replicates per treatment/clone). In addition, disease development was evaluated by visual observation in all the treated plants (W, BB and RO, n = 60) at three different monitoring times: 3, 10 and 17 days after observing the first symptoms. Incidence and severity rates were measured as a percentage and then transformed into a 0–7 scale similar to the EPPO evaluations for the efficacy evaluation of fungicides: a sporulation of 0 % was given a “0” value, a sporulation lower than 5 % a “1” value, a sporulation lower than 10 % a “2” value, a sporulation lower than 25 % a “3” value, a sporulation lower than 50 % a “4” value, a sporulation lower than 75 % a “5” value, and a sporulation higher than 75 % a “6” value. Those values were transformed into a disease index (DI) using the Townsend-Heuberger formula (1943) to obtain the Incidence Disease Index and the Severity Disease Index, respectively.

Differential expression analysis was carried out on the targeted 89 genes (Table S3) following three applications of R. okamurae (RO) and Water treatment (W). The results revealed that the extract significantly enhanced the expression of three genes belonging to the PR protein group (PR3, PR4bis and PIN) and four other genes, each involved in a different functional group: CHI2 (only in RJ43), HSR-203J, GST1 and CAD2 genes related to the flavonoid pathway, lipoxygenase pathway, redox status regulation and cell wall reinforcement, respectively (Figure 2A). In addition, the applied extended gene expression analysis using NeoViGen microarray strategy (compared to the 15 targeted genes in Zarraonaindia et al. (2023)) allowed down-regulated genes to be identified after seaweed applications, among which PR8, PR12, CHORS2, CAGT, EDS1b and SAPB2a showed the highest inhibition values, with log values (RO versus Water plants) close to -2 (Figure 2A).

Figure 2. Transcript levels of defense-related genes (log transformed) in grapevine leaves induced by applications of Rugulopteryx okamurae (S1, 1dPB) in A) and by Plasmopara viticola (S2, 1dPI) in B) relative to the water treated samples. Only genes with a log-transformed-fold change greater than 1.5 or less than -1.5 relative to water for any clone/sampling point are shown. Black bars refer to the VN40 clone and grey bars refer to the RJ43 clone. The values represent 5 biological replicates per treatment/clone. The error bars indicate the variation range of the Log2 FC values calculated as follows: ((Log2 FC (RO vs Water plants) x (RO ∆Cq variation coefficient (SD/∣mean∣) + Water plants ∆Cq variation coefficient))/ 100.

The GST1 gene, commonly associated with ascorbate peroxidase (APX) levels ( Gullner et al., 2018), exhibited the highest induction values following RO treatment, with a 4.62-fold change in VN40 and a 4.83-fold change in RJ43, consistent with previous findings in Zarraonaindia et al. (2023), which showed that RO extract preferentially induced jasmonic acid (JA) related genes, while inhibiting salicylic acid (SA) responsive genes. In particular, PR3, PR4bis and PIN (gene function of PR6), whose expressions are known to be strongly dependent on JA signalling ( Ali et al., 2018; Van Loon et al., 2006), were significantly higher in RO-treated leaves. Moreover, the expression of HSR-203J, known to be enhanced by JA signalling ( Dufour et al., 2016), was also increased by RO treatments. On the other hand, RO was found to down-regulate PR-8, PR-12, PR-14, EDS1b, SAMPT1 and SAPB2a genes, which are known to be related to SA signalling ( Adrian et al., 2012; Bodin et al., 2020; Dufour et al., 2016); this thus corroborates the antagonistic regulation of both JA and SA pathways suggested by other authors ( Paolacci et al., 2017).

Both clones showed similar upregulated and downregulated gene profiles after R. okamurae treatments. However, overall, VN40 plants showed a more consistent downregulation of five genes, three of which were related to SA pathway ones (PR8, PR12 and SAPB2a), a nitrate transporter (NrT2) and a glycosyltransferase (CAGT). By contrast, in five cases (PR3, PIN, CHI2, GST1 and CAD2) upregulation was more pronounced in the RJ43 clone (Figure 2A). CAD2 expression, involved in the biosynthesis of monolignols (main component of lignin), was significantly upregulated by RO, especially in RJ43 plants. Lignans play an important role in plant defense and, indeed, one of the characteristic responses of plants to injury and disease is the reinforcement of cell walls ( Ninkuu et al., 2022). Therefore, RO treated plants, and particularly the RJ43 clone, might have higher protection responses than untreated plants in the event of a pathogen infection.

The concentrations of polyphenols in the leaves of Tempranillo VN40 plants increased after treatment with R. okamurae (Figure 3A). The concentrations of caftaric acid and coutaric acids were significantly higher in RO- and BB-treated plants than in the control (water). Consistent with this, at this sampling point, significant increases in stilbenes (total piceid and E-resveratrol) were also observed in RJ43 plants relative to the control ones (Figure 3B). Viniferins, on the other hand, were detected under the detection limits. The same stilbenes as those found in Zarraonaindia et al. (2023) were identified, albeit in higher concentrations, which is probably due to the additional third RO treatment applied in the present experiment.

Figure 3. Polyphenol content in grapevine leaves after treatment (S ₁, 1dpB) of A) VN40 clone plants, and B) RJ43 clone plants. Different letters above the bars indicate significant differences between treatments for each compound (ANOVA, p < 0.05).

In general, polyphenols reached higher concentrations in VN40 leaves than in RJ43 without any significant differences. However, stilbenes, key compounds for responding to biotic stresses, reached similar concentrations in RO leaves for both clones, which suggests that biostimulation with regards to these compounds is similar in these clones. Flavonols (quercetin 3-glucoside + glucuronide and myricetin 3-glucoside) were not affected by the treatments, which corraborates our previous results in Zarraonaindia et al. (2023). An increase in flavonoids has previously been observed after seaweed treatment in the berries of Tempranillo blanco plants ( Cataldo et al., 2022), but, to our knowledge, such an increase has not been associated with leaves. In general, the results regarding phenolic compounds indicate that both clones are equally prepared for a possible infection.

SOD, APX and GR enzymes have been described as being related to defense mechanisms in plants. Generally, biostimulants derived from algae are known to induce the antioxidant system ( García-Sánchez et al., 2022). However, in VN40 plants, SOD and GR activities did not show any significant differences between the three different treatments (W, BB and RO). By contrast, APX activity significantly increased for RO- and BB-treated plants (1.6 and 2.9 times higher than that of water respectively; Figure 4A), indicating that both treatments were able to reduce H ₂O ₂. This could be related to a H ₂O ₂ signaling mediated by the biostimulation, preparing VN40 for any potential stress. At the end of signalling, H ₂O ₂ must be metabolised, for which APX is needed. This is an interesting strategy, since H ₂O ₂ is a reactive oxygen species which can damage plant cells ( Hasanuzzaman & Fujita, 2022). Importantly, the already high intrinsic SOD values found in the VN40 plants indicate that this clone relies heavily on H ₂O ₂ signalling and regulation in its basal state (under no biotic or abiotic stress). According to Pérez-López et al. (2015), high SOD activity could be considered an advantage as it helps plants protect themselves against potential oxidative damage. Therefore, overall, this high basal SOD could be the reason why the antioxidant system response to biostimulation was less pronounced in VN40 plants. Unfortunately, no reference was found in the literature to the effect of biostimulants on Vitis vinifera antioxidant enzymes. However, by comparing our results with those obtained under non-stress conditions, it can be noted that the SOD values observed in our study are higher than those reported for SOD in non-stressed Tempranillo (Salazar-Parra et al., 2012) and Öküzgözü ( Özden et al., 2009).

Figure 4. Antioxidant enzymes in treated leaves at different sampling times: A) and B) after 24 hours of treatment applications (S1, 1dPB) for VN40 and RJ43 respectively, C) and D) after 24 hours of Plasmopara viticola inoculation (S2, 1dPI) for VN40 and RJ43 respectively. Different letters above the bars denote significant differences between treatments (ANOVA, p < 0.05) within each compound.

By contrast, all the enzymes studied in RJ43 plants were affected by the RO treatment. SOD activity decreased 2.2- and 1.8-fold for RO and BB treatments relative to the control plants, respectively. Meanwhile, for the RO and BB treatments, there was an approximate 1.9-fold increase in APX and 1.7- and 1.3-fold increase in GR (Figure 4B). The increase in GR in RJ43 plants allows them to regenerate the important antioxidant molecule glutathione and increase the ability of plants to prepare for an oncoming stress ( Noctor & Foyer, 1998). It is important to note that RJ43 seems to rely on APX rather than SOD when reducing the accumulation of H ₂O ₂.

Thus, our results indicate that each clone modulates antioxidant enzymes differently after RO biostimulation, with the RJ43 clone modulation benefiting more, since GR is activated to a greater extent than in the VN40 clone. However, it should be noted that elicitation may have been lower in the VN40 plants due to their already high basal SOD activity.

While the composition of photosynthetic pigments in RJ43 clone plants were not affected by treatments, VN40 clone plants decreased both chlorophyll a and b, as well as carotenes and xanthophylls, which is in line with the SPAD results (Table 1). Conversely, maximum photochemical efficiency ( φ _PSII ) decreased in the RJ43 plants, while the VN40 clone maintained it at levels similar to those of the control samples (W), despite significantly decreasing the greenness (SPAD) (Table 1). Previous studies have shown that the application of seaweeds in general, and the brown macroalgae Ascophyllum nodosum in particular, results in increased chlorophyll content in the leaves of different plants ( e.g., tomato, dwarf French bean, wheat, barley and maize); this is likely due to the inhibition of chlorophyll degradation, which is caused partly by compounds, such as betaines, present in the seaweed extracts ( Sangha et al., 2014). The R. okamurae extract used here does not contain betaine lipids, as was revealed by biochemical characterisation conducted by Córdoba-Granados et al. (2024), and as would be expected for an aqueous extract, which might justify the differences found with the aforementioned publication.

In general, our results revealed no effect, or just slight negative effects (depending on the clone) of the RO treatment on photosynthetic pigments and the photochemical parameters associated with them. However, it is worth mentioning that the VN40 plants were at a more advanced phenological stage than the RJ43 plants, which may also have contributed to the reduction in photosynthetic pigments observed in VN40. Nonetheless, the reduction was slight, and the values were still within the established range representing good photosynthesis activity in grapevine leaves ( Casanova-Gascon et al., 2018).

The expression levels of most RO-modulated genes either remained stable or decreased at S ₂ (1dPI) compared to S ₁ (1dPB), a pattern that could be due to Tempranillo's inherent sensitivity to P. viticola. Tempranillo’s response to infection can usually be expected to be moderate, although P. viticola is known to be able to modulate JA pathway-related genes in certain grapevine varieties ( Figueiredo et al., 2015; Guerreiro et al., 2016). In this study, in the RJ43 clone, six out of the seven RO-induced genes, particularly the PR4bis gene in the JA pathway, showed lower expression values after inoculation (Figure 2B; Table 2). Interestingly, while the expression levels also decreased in RO-treated VN40 plants, CHI2 expression markedly increased (332 %-fold change from S ₂ to S ₁), suggesting that this gene was primed by the seaweed application.

Remarkably, while the inhibition of SA-related genes was not maintained after infection in the VN40 plants, their downregulation was strengthened in the RJ43 clone (Figure 2B). Gene expression downregulation was particularly marked in PR12 and NrT2, as well as in PR8 and CALS, in RJ43 after P. viticola inoculation; this suggests that the extract was able to prime the plants, and therefore the inhibition of the expression of these genes became stronger when the grapevines were further subjected to biotic stress. Several stimuli have been shown to prime plants for an improved response to future stress, thereby triggering earlier, faster and/or stronger responses ( Hilker et al., 2016; Mauch-Mani et al., 2017), among which seaweed extracts can be highlighted ( Islam et al., 2021). It should be noted that elicitors may also be potentially used in defense priming strategies by downregulating certain genes ( Landi et al., 2017), as is the case of susceptibility genes (S-genes) ( Liu et al., 2022). However, to our knowledge, no previous evidence of downregulation associated with the priming of plant defense response genes has been described in the literature to date; therefore, the downregulation observed for PR12, NrT2, PR8 and CALS in the previously biostimulated plants requires further validation.

Total polyphenol concentration increased in all plants after infection; however, no clear trend was observed between treatments or clones (data not shown). Only the RJ43 clone plants exhibited an increase in total piceid, this enrichment being more acute in the plants previously treated with seaweed (1.15-fold for W, 1.31-fold for BB and 1.84-fold for RO). However, piceids, in their glucoside form, are known to be inactive against P. viticola ( Gabaston et al., 2017). Other stilbenes have been directly correlated with resistance to fungal disease, and differences in the content and accumulation of stilbenes between susceptible and resistant genotypes is known to be key to overcoming infections ( Burdziej et al., 2021; Paolocci et al., 2014; Vrhovsek et al., 2012). The biosynthesis of complex stilbenes in grapevine leaves is typically observed during infection around six days after inoculation ( Vrhovsek et al., 2012; Burdziej et al., 2021), which may explain the low concentrations and differences observed in our early sampling; i.e., just 24 hours after infection. Despite this early sampling, ε-viniferin and ω-viniferin were detected in both clone plants in all the treatments (W, BB, RO), ε-viniferin concentration being significantly higher in the RO samples of RJ43 plants (Table S4). Susceptible genotypes are thought to accumulate resveratrol and piceid during infection, while other stilbenes, such as ε- and δ-viniferins, have been found to be mostly produced in resistant genotypes ( Boso et al., 2012; Paolocci et al., 2014). Therefore, the increase in viniferins observed in the present study suggests that, despite Tempranillo being considered a susceptible variety, the clones studied here were exhibiting a certain level of resistance to infection. Further samplings at later time-points would provide valuable insights into the extent of this resistance.

In both clones, SOD activity decreased with respect to the S ₁ sampling point, especially in plants that had not been previously biostimulated, indicating that the oomycete was likely downregulating SOD activity in order to inhibit the plants’ defense mechanisms ( Nascimento et al., 2019). Interestingly, in the VN40 clone, plants previously elicited by RO showed an increase of 138 %, 93 % and 73 % for SOD, APX and GR respectively in their leaves, relative to the water-treated plants. In the RJ43 clone, however, RO-biostimulated plants did not alter SOD in response to P. viticola inoculation, but APX and GR activities increased by 112 % and 57 % relative to the water samples. The significant induction of APX correlated with the pronounced overexpression of the GST1 gene (Figure 2A) in the RJ43 clone, suggesting that the ROS scavenging and detoxification mechanisms triggered by RO application in S ₁ were further enhanced following P. viticola infection, thus helping to protect the cells from the combined signal induced by both the seaweed and the pathogen.

Overall, the results indicate that RO biostimulated the plant´s antioxidant metabolism, which was activated after pathogen inoculation, maintaining most of the enzymes at high values in both clones. This indicates that both clones potentially reduce the development capacity of the oomycete. The high SOD values in VN40 could promote some resistance, as high H ₂O ₂ levels are toxic for fungi and prevent the development of the pathogen. In fact, reactive oxygen species play an important role in pathogen resistance by directly strengthening host cell walls via the cross-linking of glycoproteins, lipid peroxidation and the activation of ROS-signalling networks, leading to the establishment of a resistance response ( Figueiredo et al., 2017). However, as mentioned previously, H ₂0 ₂ could also be toxic for plants ( Hasanuzzaman & Fujita, 2022; Mittler et al., 2004); therefore, we hypothesise that to counteract it, VN40 regulated the high H ₂0 ₂ levels with a higher activation of APX. These strategies seem to allow the delicate equilibrium between ROS production and scavenging to be maintained ( Mittler et al., 2004).

Thus, we can conclude that R. okamurae seaweed treatment was able to prime both clones of the Vitis vinifera plants, allowing the plants to respond more efficiently when the infection occurred than the plants that had not been previously biostimulated by R. okamurae.

Few significant differences were found in terms of pigments and photochemical parameters between treatments after infection (S ₂). The VN40 clone plants treated with RO did not show any significant differences when compared to the water treated plants (Table S5). With regards to the RJ43 clone plants, no differences in terms of pigments, SPAD or photosynthetic efficiency were found between treatments; this demonstrates that any negative effect observed at the biostimulation sampling point (S ₁) (Table 3) was transitory and quickly reverted.

P. viticola has been shown to affect the photosynthesis of susceptible grapevine cultivars by reducing the diffusion of CO ₂ in leaf mesophyll ( Jermini et al., 2020) and by inhibiting the enzymes involved in chlorophyll synthesis ( Nogueira Júnior et al., 2020). However, in the present study, the sampling was performed one day after pathogen inoculation, and therefore the above effects had not yet been observed.