3.3. Influence of RLBV-infection, Harvest Year and Locality on Polyphenolic Profile of Raspberry Fruits
Three-factorial ANOVA and Tukey’s multiple comparison tests were used to analyze the influence of certain factors on the polyphenolic profile of raspberry fruit samples. Factors of variation (viral status, locality, harvest year) and their interactions are presented in
Table 4. It is noteworthy that the influence of harvest year and locality might be actually understood as the influence of weather conditions and soil composition, respectively. Viral infection showed the least effect on the contents of CA, Q3-RHA, Q, PHENOL, and ANTHO (no statistical differences), while such influence was the most dominant on the contents of
pCOU, EA, RUT, ISO-Q, and CY3-SOP (
p < 0.001). The contents of all detected compounds are highly influenced by harvest year (
p < 0.001 for all compounds, except
pCOU and Q3-RHA, which are
p < 0.01) and locality (
p < 0.001) (
Table 4 & Appendix).
Comparing all infected and all healthy samples, regardless of the locality and harvest year,
Figure 1 represents the influence of the viral infection on the polyphenolic profile of raspberry fruit samples. It is obvious that in most of the cases, viral infection prompted either an increase in the content of certain polyphenolics (
pCOU, EA, RUT, ISO-Q, KAE, CY3-GLU, CY3-SOP) or the concentrations remained intact (CA, Q, PHENOL, ANTHO). Only the content of FA decreased after infection. The same behavior was observed when analyzing the dual interaction between
viral status ×
harvest year (A × B) and
viral status ×
locality (A × C).
To present it in a more illustrative manner, PCA was applied to determine how infected and healthy raspberry samples of the ‘Willamette’ in different orchards were grouped based on concentrations of 11 individual compounds and 2 classes of compounds (
Figure 2). The first two principal components explain 60.02% (39.03% and 20.99%, respectively) of the total variance. PCA analysis revealed no segregation whatsoever regarding the viral status of the plant. Furthermore, it is evident that infected and uninfected pairs (rounded in
Figure 2) stand in close proximity.
The influence of harvest year can be more clearly observed in
viral status ×
harvest year (A × B) and
harvest year ×
locality (B × C) interactions. Analyzing the B × C interaction revealed that within the same locality, the content of EA was higher in 2020 compared to 2019, while content of FA was decreased in 2020 (Appendix). Anthocyanins were highly influenced by weather conditions, so the contents of both detected anthocyanins (CY3-GLU and CY3-SOP) were higher in 2019 than in 2020, in all three experimental orchards. For instance, contents of CY3-GLU in raspberry samples harvested in Bedina Varoš, Devići and Cerova were 101.4, 56.9 and 106.7 mg/100 g fw in 2019, and 35.8, 21.4 and 35.7 mg/100 g fw in 2020, respectively. Flavonols did not show any conclusive trend. Interaction A × B showed very similar trend within the same viral status, meaning that weather conditions highly affect the formation of certain polyphenolics. The influence of the weather conditions observed through the harvest years, can be more clearly seen in PCA. All raspberry samples harvested during 2019 are grouped on the negative side of PC1, while 2020-harvested samples formed the other group on the positive side of the same axis (
Figure 2). This separation might be attributed to the different weather conditions in 2019 and 2020 in all orchards, which is clearly seen in
Table 2. While the average monthly air temperatures were similar in both examined years, the rainfall amounts significantly differed in almost all examined months (April−October). Cheng et al. [
24] proved that anthocyanins were more readily accumulated in the skin of grapes (
Vitis vinifera L.) if the soil contained less water. On the contrary, Li et al. [
25] showed that the contents of almost all anthocyanins were increased in the rain-shelter cultivated wine grapes.
The effect of the locality on the polyphenolic profile cannot be clearly seen through dual interactions:
viral status ×
locality (A × C) and
harvest year ×
locality (B × C). It turned out that PCA showed good separation among raspberry samples (
Figure 2). Namely, the different behavior of individual polyphenolics in Bedina Varoš locality compared to Devići and Cerova orchards can be clearly observed by PCA. Raspberry samples harvested in Bedina Varoš locality formed the first group, situated on the positive side of PC2. A second group, situated on the negative side of PC2, was formed by the fruit samples harvested in the remaining two orchards, Devići and Cerova. Such a trend can be attributed to the different soil compositions of three orchards (
Table 5). Namely, it is clear that contents of phosphorus and potassium in the soil of Bedina Varoš orchard are significantly higher compared to Devići and Cerova orchards, which certainly influence the polyphenolic profile of grown fruits [
26]. The content of phosphorus is 27.7-fold and 5.6-fold higher in Bedina Varoš orchard than in Devići and Cerova orchards, respectively. Furthermore, the weather conditions in 2019 and 2020 in Bedina Varoš were quite different compared to the remaining two orchards, especially regarding the rainfall amounts (
Table 2). The polyphenolic profile of raspberry fruits appears to be more influenced by weather and soil characteristics than by RLBV infection. Given the wide range in soil composition, particularly in terms of phosphorus and potassium content, and rainfall amounts (for example, the amount of rainfall in the Devići locality in August 2020 was 5.7 times and 1.7 times higher than that in Bedina Varoš and Cerova, respectively), it is not surprising. Delgado et al. [
27] proved that anthocyanin content in ‘Tempranillo’ grapes at veraison increased in proportion to the potassium dose applied in fertilization. Also, increased accumulation of polyphenols in grape fruits with no nitrogen fertilization was observed, while this trend diminished as the potassium fertilization dose was increased. In contrary, purple-blue potatoes showed the highest anthocyanins content with the highest nitrogen fertilization and a lack of phosphorus and potassium fertilization [
28]. It seems that every fruit and vegetables species and cultivar require certain level and ratio of nitrogen, phosphorus, and potassium contents in the soil to maximize polyphenolic content.
Polyphenolic compounds belong to the group of secondary metabolites, responsible for defense mechanisms in plants, primarily towards the microbiological infections, including viral infections. The contents of these compounds in fruits depends on various factors, such as fruit species, variety, maturity stage, soil substrate, plant health status, agricultural practices, etc. [
29,
30,
31]. In this research, the influence of the cultivar was excluded since only ‘Willamette’ was utilized for analyses. On the other hand, three influences were included: viral status, harvest year (weather conditions), and locality (soil composition). It is very difficult to isolate only the effect of infection on the content of polyphenolics and to draw any clear conclusion or trend regarding the influence of the virus presence. Statistical analyses showed an evident influence of locality and harvest year since PCA clearly differentiated all samples harvested in the first harvest year from the samples harvested the following year, as well as all samples harvested in Bedina Varoš locality from all samples harvested in the remaining two orchards (
Figure 2).
It is well known that stress conditions, such as the presence of pathogens, might cause alteration in the contents of flavonoids, anthocyanins, and hydroxycinnamic acids, as a sign of the plant’s response to infection [
32,
33,
34]. Such influence is most obvious in the Bedina Varoš locality, where all four hydroxycinnamic acids, as well as RUT and Q3-GLU, were additionally deposited in fruits as a consequence of infection during both harvest years (
Table 3). Some authors claimed that an increased amount of hydroxycinnamic acids is a sign of a plant’s response to pathogens [
31]. On the other hand, Usenik et al. [
32] found a lower content of hydroxycinnamic acids in PPV-infected plums compared to the healthy samples. Such an effect was explained by stress-induced alteration of the biosynthetic pathway of flavonoids, and thus an increased amount of flavonoids was synthesized on account of reduced synthesis of hydroxycinnamic acids in infected plum samples. In our group’s research on PPV influence on phenolic compounds, it was noticed that none of the two suggested options occurred, giving rise to the conclusion that the examined plum cultivar could be sorted as highly tolerant towards the virus [
18]. In this research, it cannot be concluded that cv ‘Willamette’ is tolerant towards RLBV, but it is also unclear the effect that such a virus caused. The effect of RLBV was indisputable, but this viral effect on polyphenolic compounds cannot be isolated from other effects in our designed experimental setup. Jevremović et al. [
12] concluded that locality played an important role in raspberry polyphenolic profiles when infected by RLBV.
Viral infection had no impact on CA, Q3-RHA, Q, PHENOL and ANTHO (non-significant), then a medium impact on FA (
p < 0.05), KAE, and CY3-GLU (
p < 0.01), while a high influence on
pCOU, EA, ISO-Q, and CY3-SOP was detected (
p < 0.001) (
Table 4,
Figure 1). Taking this into consideration, it could be concluded that virus presence definitely showed a certain influence on polyphenolic compounds. Nevertheless, such influence was certainly limited and prevailed due to the effects of locality and weather conditions. Since these two parameters played an important role, the different behavior of some phenolic compounds in two successive harvest years might be explained by various weather conditions (for instance, the content of CY3-SOP in raspberry samples harvested in 2019 in Bedina Varoš increased due to the infection from 304.0 to 594.4 mg/100 g fw, while in 2020 it decreased from 186.1 to 95.2 mg/100 g fw. On the other hand, since contents of some compounds showed different trends in various orchards during the same harvest year, such behavior could be due to the different localities and also different weather conditions (for instance, content of CY3-GLU in raspberry samples harvested in Bedina Varoš in 2019 increased as a result of infection from 69.9 to 132.8 mg/100 g fw, while during the same harvest year content of this anthocyanin remained intact after infection in the other two localities). Apparently, the influence of viral status was overpowered by the influence of weather and locality on the phenolic profile of raspberry ‘Willamette’. On the other hand, Jevremović et al. [
12] concluded that RLBV significantly decreased the dimensions and weight of infected raspberry ‘Willamette’ fruits in the same three orchard’s localities during 2019. The decreases in fruit length, width, and height were 4.01 ‒ 9.8%, 4.06 ‒ 9.47% and 5.88 ‒ 14.9%, respectively, while the decreases in fruit weight due to RLBV infection ranged from 9.15 ‒ 27.49%. Regardless of the orchard locality, RLBV presence, without exception, causes a decrease in the fruit’s dimensions and weight. It is clear that RLBV infection overpowers the effects of environmental conditions on ‘Willamette’ productivity. As for the polyphenolic profile, viral infection was suppressed by environmental conditions, which is clearly demonstrated by PCA. This information could be of great interest to ‘Willamette’ producers.