Abstract
BACKGROUND:
New sweet cherries rootstocks and cultivars can be valorized on the market if they express improved plant yield efficiency and fruit quality, and also including highest content of antioxidant compounds (mainly anthocyanins but also significant amounts of phenolic acids and flavonols).
OBJECTIVE:
The goal of this study was to compare the effect of 10 new rootstock genotypes, now available for cherry, on plant growth and fruit quality of the cultivar “Sweetheart”, a very late auto-fertile sweet cherry cultivar.
METHODS:
Plants vegetative and productive parameters, as well as fruit sensorial quality, were measured during a 5-year cycle (2011–2015), while fruit nutritional quality was investigated for the last 3 years (2013–2015).
RESULTS:
For the agronomical traits, the vigorous rootstock Adara/Major and intermediate Gisela 6 stands out for the plant yield. From the qualitative point of view, the intermediate MaxMa 14 and Gisela 6, and the dwarf Gisela 5, differed for reduced fruit size but with optimal sugar/acid ratio. The higher fruit nutritional value were detected for “Sweetheart” trees grafted on Adara/Major (vigorous), the best MaxMa 14 (weaker) and Gisela 5 (dwarfing) rootstocks.
CONCLUSIONS:
There is not a univocal combination between the scion and rootstock equally valid for all the considered parameters. The rootstock capable of ensuring good yields, sensorial and nutritional quality is Adara/Major. From the productive and qualitative point of view, Gisela 6 is well adapted to these conditions, while MaxMa 14 rootstock stands out for nutritional quality, although induces a high reduction of vigor, poor vegetative renewal and low average fruit weight.
Introduction
Fruit trees cultivation is widespread in different pedo-climatic conditions and is adapted to different cultivation systems; thanks to the utilization of new selected rootstocks with a different capacities of controlling tree growth [1], total yield and yield efficiency, [1, 2] and floral and foliar nutrition [3]. Furthermore, the qualitative and nutritional attributes of the fruit can also be modified by the use of different rootstocks [1, 5].
The rootstock plays a major role in plant development and vigour [6], mainly for the anatomical differences of the vascular system in comparison with the scion. Additionally, it also controls other traits associated with molecular messages induced with the transfer of micromolecules, and in particular of interfering RNA (RNAi); which is now widely studied in transferring virus resistance from the rootstocks to the scion [7]. Preliminary results on the application of this technology have been achieved in sweet cherry for inducing resistance to the Prunus necrotic ringspot virus (PNRSV) [8]. Although, all these mechanisms of rootstock/scion interaction are not fully understood, different studies have demonstrated how the type of rootstock used affects the vegetative, productive, sensorial (including fruit shape, soluble solid concentration, acidity) and nutritional (including fruit antioxidant, polyphenols and anthocyanin contents) characteristics of different Prunus fruit species [9–13].
In recent years, a large increase of sweet cherry cultivation in different fruit producing areas of the world occurred (http://faostat3.fao.org/download/Q/QC/E). This increase has happened mostly because of breeding programs that release new improved sweet cherry cultivars; in particular for the plant auto-fertile habit, and rootstocks for reduced vigor or dwarfed stature. The release of new weak or dwarf rootstocks opened the possibility of introducing new more intensive cultivation systems with the use of the new compact trailing system named “Spanish bush”. This combination produced an easier control of plant development with reduced costs for pruning and harvesting [14, 15]. However, these systems are adapted only to specific soil conditions and require high water availability; otherwise they are not recommended if the conditions are different as this would risk reduced plant development and plant yield thus, requiring intermediate or even vigorous rootstocks.
Currently, GiSelA and PHL are the two most popular dwarfing rootstocks used for the new intensive cherry cultivation [16], while MaxMa 14 is a cherry rootstock selected for its good agronomic adaptation to different soils [17] and cultivation systems requiring intermediate levels of vigor [2].
Sweet cherry is mainly consumed as a fresh fruit and in many areas comes as a first fresh fruit of the season, but it is not always easy to have high fruit quality as expected by the consumer. Several new cultivars are being widely cultivated mainly because of specific traits such as yield, ripening season, and auto-fertility but they don’t always necessarily yield fruits of high quality. Nevertheless, sweet cherry fruit is popular for its sweetness, color, and nutritional compounds content [18]. Therefore, to increase the production efficiency and fruit quality of these new cultivars, it is important to extend the knowledge on the effect of rootstock type not only on plant adaptability and vigor, but also on fruit sensorial and nutritional quality.
Increasing number of studies demonstrate that the use of functional foods as source of bioactive phytochemicals or nutraceutical compounds in the diet represents a valuable strategy for counteracting oxidative stress damages [19–22]. In the past years, epidemiological evidences have highlighted a strong correlation between plant metabolites having bioactive properties and improvement of human health [23–25]. Edible plants, that counteract free radical overproduction during oxidative stress, are considered an effective tool to prevent several diseases, and increase human well-being. Phenolic compounds have attracted special attention due to their health-promoting characteristics [26, 27]. These natural products, with considerable diversity in their structure, are important components of plant foods to which they contribute to flavor, color, and sensory properties such as bitterness and astringency [28].
Sweet cherries are known to possess various antioxidant compounds [29–31]. The major class of phenolic antioxidants in sweet cherries are anthocyanins, but they also showed significant amounts of phenolic acids and flavonols [18, 32–34].
The bioactive compounds content of fruit is mostly influenced by the cultivar, but also the effect of rootstock on polyphenol and anthocyanin content of “Stella” and “Lapins” sweet cherry fruits has been reported [35]. This rootstock-related variation could be due to the influence of the type of rootstock on mineral uptake and transfer to shoots and leaves [5].
Very few studies are available on the effect of new dwarf rootstocks on the fruit production and nutritional quality of sweet cherry trees trailed with the “Spanish bush” trailing system, and in particular on “Sweetheart” cultivar.
“Sweetheart” sweet cherry (Prunus avium L.) was developed at the Agriculture and Agri-Food Canada, Research Centre, Summerland, British Columbia as part of a breeding program aimed at improving the quality and productivity of sweet cherries [36]. This is an auto-fertile late cultivar (its fruits mature approximately 9 days later than “Lapins”), with medium vigor, and fruits characterized by large size (they could be smaller in trials with low tree vigor combined with heavy fruit set), intense bright red skin (it lies between Royal Horticultural Society color chart number 46A and l87B [37]), firm flesh, good handling characteristics and excellent resistance to storage [36]. This cultivar is preferred in cold areas [38] and has been studied during storage under modified atmosphere for the evaluation of fruit quality [39]. However, this widespread cultivar has not been studied for its adaptability to different rootstock types.
Therefore, the primary purpose of this work is to compare the influence that rootstocks of 10 different genotypes with different vigor capacities have on plant vegetative and productive parameters, and fruit sensorial and nutritional quality of the “Sweetheart” cultivar trailed with “Spanish bush” system.
Materials and methods
Plant material
In the spring of 2007, one-year-old plants of “Sweetheart” cultivar grafted on rootstocks of 10 different genotypes (Adara/Major, Adara/Marianna, Argot, P. avium, Gisela 5, Gisela 6, MaxMa14, MaxMa60, PHLA, PHLC) (Table 1) were planted in the experimental farm of the Agenzia per i Servizi nel Settore Agroalimentare della Regione Marche (ASSAM), situated in the province of Fermo (43°03′09.6′N - 13°41′11.6′E), Italy. The trees were spaced 4.5 m within and between the rows and trailed with the “Spanish bush” trailing system. The experimental field included 16 plants, divided in 4 randomized plots (4 plants each), for each of the 10 rootstock genotypes.
The cherry plants were grown under open field conditions, using sour mulch and a standard integrated pest management cultivation system. The orchard was managed according to the integrated production disciplinary of Marche Region, with mineral fertilization restitution of 70 kg/ha of nitrogen, 40 kg/ha of phosphorus, 40 kg/ha of potassium, and rescue irrigation with an over-crown sprinkling system. The soil analyses were performed by ASSAM - Regional Agrochemical Center, according to the procedures regulated in “D.M. 13/09/99 GU SO n.248 of 21/10/1999”. The soil structure was tendentially clayey, with high active limestone (85 g/kg), high organic matter (25.2 g/kg) and slightly alkaline pH (8.0). The high amount of assimilable phosphorus (20.6 mg/kg) and exchangeable potassium (410 mg/kg) justified the low doses of mineral phosphorus and potassium fertilization.
Main characteristics of rootstocks utilized in this study (adapted from [60])
Main characteristics of rootstocks utilized in this study (adapted from [60])
At harvest time (June and July), all fruits at full ripening stage, of uniform size and color (according to “Sweetheart” colorimetric cards), were harvested. The cumulative yield per plant was determined for 5 production cycles (2011 to 2015), allowing to better identify the behavior of all rootstocks from juvenile to productive stage and to get a rather complete overall picture of the adaptation of plants to the environmental and pedo-climatic conditions of the specific area of the study.
During all the five production cycles (2011 to 2015) after harvesting, summer pruning was performed, and the total weight of pruned fresh shoots was collected. We also obtained a single data set derived from the sum of summer prunings for the 5 years.
During all the five production cycles (2011 to 2015), monitoring of plant growth was carried out after the vegetative period. This was done by measuring the trunk cross sectional area (TCSA) 10 cm above the grafting union [40]. Only results obtained in the fifth year (2015) are presented, as we were interested in evaluating the different vigor capacities induced by rootstocks at the end of the five-year trial.
Finally, plant yield efficiency [9] was calculated, corresponding to the ratio between the cumulative yield of the plant from 2011 to 2015 and the trunk area measured in the year 2015.
Fruit sensorial analyses
Fruits from the first, second and third main pickings were used to prepare bulk of samples for each genotype during each of the five seasons (2011–2015). From these bulks, 30 fresh fruits were used to measure the average fruit weight (AFW) and then they were frozen at –20°C until the Soluble Solids (SS) and Titratable Acidity (TA) analyses were performed. The SS were measured using a hand-held refractometer (Atago, Italy). The refractive index was recorded as °Brix, with the refractometer prism cleaned with distilled water after each sample. The TA was determined from 10 ml of juice diluted with distilled water (1:2, v/v) and titrated with 0.1 N NaOH solution to pH 8.2. The titratable acidity was expressed as meq NaOH/100 g fresh weight (FW). The pH of the cherry juice samples was measured using a digital pH meter (Model 420A, Orion bench-top pH meter, Allometrics Inc., Seabrook, USA).
Fruit nutritional analyses
A fruit sample (300 g) of each genotype obtained from 2013–2015 production cycles was stored in a freezer at –20°C until the extraction and analysis of fruit nutritional quality.
Chemicals
Methanol (99%, ACS-ISO) from Carlo Erba Reagents (Milan, Italy). Folin-Ciocalteu reagent, sodium carbonate (anhydrous), potassium chloride, sodium acetate, hydrochloric acid, glacial acetic acid, dihydrogen potassium phosphate, dipotassium hydrogen phosphate, 2,2’-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt (ABTS•+), 2,4,6-tris(2-pyridyl)-striazine (TPTZ, 99%), 6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (Trolox), potassium persulphate, 3,4,5–trihydroxybenzoic acid (gallic acid), and sodium hydroxide, all these reagents were obtained from Sigma-Aldrich (Sigma-Aldrich s.r.l., Milan, Italy).
Fruit extraction
The fruit nutritional analyses were performed on 10 g of the randomly selected fruits harvested in three seasons (2013–2015), stored at –20 °C and placed into test-tubes for the extraction procedure with 100 mL methanolic solution (20:80 water:methanol and 1% of acetic acid). After this step, samples were homogenized using an Ultraturrax T25 homogeniser (Janke and Kunkel, IKA Labortechnik, Staufen, Germany). The homogenized suspensions were placed at 4°C in the dark. After 48 hours, the suspensions were centrifuged at 2000 rpm for 20 min (Rotofix32 centrifuge, Hettich Zentrifugen, Tuttlingen, Germany) and the recovered supernatants were collected and stored in individual amber vials at –20°C [41, 42].
Total Antioxidant Capacity
The fruit Total Antioxidant Capacity (TAC) was evaluated using the TEAC (Trolox Equivalent Antioxidant Capacity) assay, according to a previously validated procedure [42, 43]. Briefly, a test tube was filled with 1.9 ml ABTS. Afterwards 0.1 ml of the diluted sample (1:10) was added and stored in the dark for 6 min. The absorbance of the sample was measured after exactly 6 min at 734 nm.
The TAC activity is expressed as mg Trolox eq per kg fresh weight (mg TE/kg FW). The calibration was calculated by linear regression from the dose-response of the Trolox standards.
Total Phenol Content
The fruit Total Phenol Content (TPH) was evaluated using the Folin-Ciocalteu reagent method [44], with gallic acid as the standard for the calibration curve. Briefly, glass test-tubes were filled with 7.0 ml water, and 1 ml diluted samples (1:3) was added. This step was followed by the addition of 500μl of Folin-Ciocalteu reagent and vortexing of the samples. After 3 min, 1.5 ml of sodium carbonate (0.53 mol/l) was added and the test-tubes were mixed again and then stored in the dark for 60 min. The absorbance of the samples was measured at 760 nm. Data are expressed as mg gallic acid equivalents per kg fresh weight (mg GA eq/kg FW).
Total Anthocyanin Content
The fruit Total Anthocyanin Content (ACY) was measured using the pH differential shift method [45]. This assay is based on the characteristic change in intensity of the hue of the anthocyanins according to the pH shift. Briefly, the samples were diluted (1:3) with potassium chloride (pH 1.00) and, separately, with sodium acetate (pH 4.50). Then, the corresponding maximum absorbance for both solutions was measured at the wavelengths of 510 nm and 700 nm. The data are expressed as mg pelargonidin 3-glucoside per kg fresh weight (mg Pel-3-Glu/kg FW).
Experimental trial and data analysis
Titratable acidity and soluble solids were analyzed in triplicate for each sample, while TEAC, TPH and ACY were all analyzed six times for each sample.
All the vegetative, productive, qualitative and nutritional parameters were analyzed using one-way analysis of variance (ANOVA), with each genotype as an independent variable. Significant differences within samples were calculated according to Student’s Newman-Keuls (SNK) tests, and differences at p≤0.05 were considered as significant. Furthermore, the correlations among vegetative, productive and qualitative parameters (years 2011–2015) and among nutritional parameters (years 2013–2015) were calculated through the Pearson’s correlation matrix (p≤0.05).
3. Results and discussion
Plant vigour
Summer pruning data indicateed the capacity of vigorous rootstocks to produce more vegetation and therefore higher number of pruned shoots (Fig. 1). These results confirmed the effect of the different rootstocks in determining plant development in none intensive trailing system and Mediterranean pedoclimatic conditions.
In the 5 years of study, from Adara/Marianna trees the highest amount of shoots was pruned, with a total 57.61 kg/plant (p < 0.05). This rootstock was followed by trees grafted on P. avium (48.74 kg/plant), Argot (42.07 kg/plant) and Adara/Major (39.14 kg/plant), statistically similar to each other. The behavior observed on plant vigor induced by the Adara/Marianna, Adara/Major and P. avium rootstocks was in line with what was expected for vigorous rootstocks.
Trees grafted on the dwarf rootstock PHLC produced an intermediate number of pruned shoots (27.95 kg/plant), while PHLA, MaxMa 60, and Gisela 6 trees showed lower and statistically similar values (19.84, 16.96, and 15.60 kg/plant, respectively). Among these rootstocks, MaxMa60 is generally considered vigorous, however, from the lower weight of shoots pruned in the summer it is comparable to other weaker and dwarf rootstocks.

Summer pruning (sum of years 2011–2015 ± standard errors) and Trunk Cross Sectional Area (TCSA) (mean of years 2011–2015±standard errors) divided according to the rootstock vigor. Histograms with different letters are statistically different (SNK test, p≤0.05). Uppercase letters belong to Summer pruning, lowercase letters belong to TCSA.
The weaker rootstock MaxMa 14 showed low adaptability to these conditions by producing a very low amount of pruned shoots, corresponding to only 3.08 kg/plant of total pruned shoots in 5 years. The cultivation conditions adopted for this trial were probably too limiting for this rootstock in that, its vigor was further reduce which led to production of a plant even smaller than Gisela 5 grafted plants (9.59 kg/plant), which until now was considered the weakest rootstock. In fact, in a study done by Jakobek et al. [46], in different pedoclimatic conditions and with a different scion (“Lapins” cv.), the less vigorous plants resulted from the Gisela 5 rootstock, wheras MaxMa 14 plants produced more vigorous plants.
The differences observed among rootstocks in the total weight of summer pruning were also observed in the TCSA. The highest vigor of trees grafted on vigorous rootstocks was confirmed by larger trunk areas (Fig. 1), with Adara/Marianna (287.04 cm2) giving the highest measurements, followed by Adara/Major and P. avium, with 207.27 and 184.32 cm2 respectively. Adara showed the highest TCSA as observed in another trial performed with the cultivar “Van” [47], and one of the highest for the cultivar “Stark Hardy Giant”, confirming its vigorous behavior with different cultivars and other pedoclimatic condition of the Mediterranean area.
Among weaker rootstocks, PHLA (133.81 cm2) and Gisela 6 (100.31 cm2) induced similar values of trunk area above grafting union, statistically higher compared to values obtained from the dwarf rootstock Gisela 5 (66.85 cm2) (p < 0.05). These outcome is in accordance with Robinson et al. [48] which revealed a higher TCSA in cv. “Lapins”, “Regina” and “Hudson” grafted on Gisela 6 than on Gisela 5. Robinson and Hoying [49] confirmed the higher TCSA in Gisela 6 than Gisela 5 for “Lapins” and “Hedelfingen”, while for “Sweetheart” the results were contrary.
As for the summer pruning, trees grafted with MaxMa 60 showed lower values than expected also for TCSA (113.98 cm2), the other parameter taken in consideration for determining plant vigor.
The low adaptability of MaxMa 14 rootstock to these cultivation conditions was also evidenced by the very low values of TCSA (25.34 cm2). This outcome is in contrast with other studies showing the capacity of MaxMa 14 to induce larger TCSA values of “Van” and “Stark Hardy Giant” cultivars [47]. The same was observed by Long et al. [50] testing MaxMa 14 in three different cherry cultivars and training systems, reporting even slightly higher values than Gisela 6 rootstock. Furthermore, Domozetova and Radomirska [51] described higher vigor and trunk area of “Kordia” and “Hedelfinger” trees grafted with MaxMa 14 than with Gisela 5.
Cumulative yield
The cumulative yield per tree from 2011 to 2015 was analyzed with the aim characterizing the different production capacity of the cv. “Sweetheart” when grafted on these different rootstocks. The cumulative yield for 5 years of study ranged from 9.15 to 27.16 kg/plant. Those results confirmed the range of cherry plant yield cultivated in Italy, even if in different areas and with different rootstock/cultivar interactions. Godini et al. reported a cumulative yield (for 7 years) range for cv “Lapins” of 2.5 to 45.5 kg/plant [52]; for the same cultivar, a very wide range (9.7 – 165.7 kg/plant) of cumulative yield for 9 years was indicated by de Salvador et al. [53], but with big variations according to the zone of Italy considered; finally, an interesting study on plant cumulative yield from year 3 to year 6 of cultivation on “Lapins” and “Regina” cvs showed very similar results with our study, with a range of 8.2 – 37.6 kg/plant (according to rootstock/cultivar interaction) [54].
The total yield determined by the rootstock/cultivar interaction in our study was related to the different vigor of the plant. The cherry trees grafted with vigorous rootstock Adara/Major showed the highest total yield in the 5 years trial (27.16 kg/plant) (Fig. 2). The high yield capacity of Adara rootstock was observed also in other studies [47, 55]. Also Gisela 6 has induced a high cumulative fruit production of “Sweetheart” cultivar (25.75 kg/plant), resulting in the best rootstock for the weaker category (p < 0.05).

Cumulative yield of years 2011–2015 divided according to the rootstock vigor. Histograms with different letters are statistically different (SNK test, p≤0.05).
Trees grafted with the dwarf rootstocks Gisela 5 and PHLC showed a very good amount of total yield (20.13 kg/plant, 19.51 kg/plant, respectively). These data demonstrated that Gisela 5 performed well in our study, but data on Gisela 5 in literature are conflicting: e.g., it was reported that Gisela 5 registered the lowest yield value for both “Van” and “Stark Hardy Giant” cultivars [47]. On the other hand, Gisela 5 trees had a higher yield than MaxMa 14 in Domozetova and Radomirska [51], and it showed higher total yield than Gisela 6 in “Lapins”, “Regina”, “Hudson” [48], and “Hedelfingen” [49]. In this last study, Gisela 5 gave a lower cumulative yield than Gisela 6 for “Lapins”, and registered almost the same cumulative yield than Gisela 6 for “Sweetheart”.
The lower cumulative yields were harvested from trees grafted on P. avium (11.16 kg/plant) and in particular on MaxMa 14, that in 5 years’ production gave only a total of 9.15 kg/plant. MaxMa 14 registered the lowest value as also happened for plant vigor. In a study done by Long et al. [50], MaxMa14 showed a lower total yield than Gisela 6 in “Sweetheart” cultivar. However, in Font i Forcada et al. [47], considering the cv “Stark Hardy Giant”, MaxMa 14 showed the highest yield among 8 different rootstocks.
The highest plant yield efficiency, calculated as ratio between the cumulative plant production from 2011 to 2015, and the trunk area in 2015 (Fig. 3), MaxMa 14 had 0.352 for trees. However, the interesting value observed for MaxMa 14 combination was mostly due to very low vegetative development of the tree thus, a very low production. A similar behavior was observed in trees grafted on the dwarf rootstock Gisela 5 with a ratio of 0.334, confirming observation made on GiSelA rootstocks by Robinson et al. [48], where Gisela 5 showed the highest plant yield efficiency for three different cultivars, followed by Gisela 6. The same author also showed a similar cumulative yield efficiency over 11 years for Gisela 5 and 6 in “Sweetheart”, with Gisela 6 slightly better than Gisela 5 [49].

Plant yield efficiency (mean±standard errors) divided according to the rootstock vigor. Histograms with different letters are statistically different (SNK test, p≤0.05).
The vigorous rootstocks showed the worst plant yield efficiency. Adara/Marianna, which presented a good total yield, showed also the highest TCSA and the ratio of these two parameters gave a very low plant yield efficiency (0.071). Differently, in Font i Forcada [47], Adara showed better plant yield efficiency than MaxMa 14 for two cultivars, but this result was due to the different behavior of the two rootstocks with cultivars other than “Sweetheart” and in different pedoclimatic conditions.
Also P. avium rootstock showed a very low plant yield efficiency, while all other combinations resulted with more equilibrated tree vigor and yield.
Average Fruit Weight (AFW)
“Sweetheart” fruit weight was highly influenced by the type of rootstock (Fig. 4). The largest fruits were harvested from trees grafted on PHLA rootstock (9.29 g/fruit). Large fruit but slightly smaller than PHLA were harvested from trees grafted on P. avium (9.10 g/fruit), Argot (9.02 g/fruit) (two vigorous rootstocks) and PHLC (8.86 g/fruit). Trees of MaxMa 60 (8.72 g/fruit) and Adara/Major (8.66 g/fruit) yielded fruit with an intermediate AFW. The smallest fruits were harvested from MaxMa 14 (8.17 g/fruit).

Average fruit weight (AFW) (mean of years 2011–2015±standard errors) divided according to the rootstock vigor. Histograms with different letters are statistically different (SNK test, p≤0.05).
This result could be explained by the low vegetative development induced by MaxMa 14 rootstock, which corresponded to the differentiation of a high number of productive branches with very small fruit, until the limit case of year 2015, when fruits of some of these trees were declared unmarketable for their very small dimension. AFW values induced by MaxMa 14 are highly variable in literature, demonstrating a great variability of this parameter according to the scion, the pedoclimatic conditions and the training form. For example, in Jakobek et al. [46], MaxMa 14 showed the lowest average fruit weight (lower than Gisela 5), and in Long et al. [50] it showed lower fruit size of “Sweetheart” cv. than Gisela 6 in two training system of three. Contrarily, in Font i Forcada [47] and Domozetova and Radomirska [51], MaxMa 14 induced optimal fruit size in different cultivars, even better than the one induced by Adara and Gisela 5.
Similarly to MaxMa 14, Gisela 5 and 6 rootstocks sustained a limited fruit development (8.33 and 8.42 g/fruit respectively), confirming that conditions limiting plant vegetative development can have an effect in increasing crop load but with fruits of low commercial value. Also in Robinson and Hoying [49], Gisela 5 and 6 showed very similar values of AFW, but even smaller than in our study (7.76 and 7.68 g/fruit, respectively).
The type of rootstock also influenced the quality of “Sweetheart” fruits, in particular accumulation of soluble solids (SS), with a difference of about 2.5°Brix between the highest fruit content, induced by MaxMa 14 (19.5°Brix), and the lowest fruit content, induced by P. avium and Adara/Marianna (17.0°Brix) (Fig. 5). The data obtained from this study indicates that vigorous rootstocks reduce the quality of the fruit by decreasing their sugar content. This phenomenon is generally justified by the plant developmental physiology, which involves competition among fruit and shoot development. Shoots generally are more competitive, thus reducing carbohydrate availability to the fruits and the increasing plant density corresponding with a decrease of fruit sweetness [56, 57]. Another factor that could influence the low amount of SS in the fruits of vigorous rootstock plants is the low light exposition of these fruits, due to the thickness of the canopy. On the contrary weaker rootstocks induce a more equilibrate vegeto-reproductive development with positive effects also on fruit quality.

Soluble Solids content (SS) and Titratable Acidity (TA) (mean of years 2011–2015±standard errors) divided according to the rootstock vigor. Histograms with different letters are statistically different (SNK test, p≤0.05). Uppercase letters belong to SS, lowercase letters belong to TA.
Gisela 6 and 5 rootstocks belong to weaker and dwarf classes respectively, but both induced medium-high levels of SS fruit content. This result can be related to their capacity to reduce canopy development and foliage so to ensure an optimal light exposition of fruits, promoting also a slight anticipation of fruit ripening. The capacity of Gisela 5 and 6 to increase fruits sweetness was observed also in “Lapins”, “Regina” and “Hudson” cultivars [48] and “Stark Hardy Giant” cultivar [47], differing in particular to fruit sweetness achieved from trees grafted with Adara and MaxMa 14. In our study, the other dwarf rootstock (PHLC) exhibited a completely different trend. In fact, its behavior was more similar to what was observed in vigorous rootstocks, producing fruits with 1.3°Brix lower than Gisela 5.
Regarding fruit titratable acidity, differences were substantial among fruits from trees grafted with different rootstocks (Fig. 5). The highest value was registered for fruit harvested from P. avium rootstock (15 meq NaOH/100 g), while the lowest from Gisela 6 (12.2 meq NaOH/100 g). It is interesting to note that, among all the analyzed fruits, those deriving from P. avium rootstock showed the worst sensorial quality, giving the excessive acidity and the insufficient sweetness of the fruits.
Nutritional parameters
Total Antioxidant Capacity (TAC)
Many studies investigated the variability of health promoting compounds in different sweet cherry cultivars [57], but only few studies underlined the importance of rootstock/scion interaction in determining the content of healthy compounds in sweet cherry fruits [46, 59].
In our study, fruit TAC showed an interesting range of variability, trees grafted on MaxMa 14 yielded fruit with the highest value (10.25 mM Trolox eq/kg FW) while those grafted on P. avium rootstock the lowest (6.98 mM Trolox eq/kg FW). Similar intermediate TAC values were detected on fruit harvested from trees grafted on PHLA (9.36 mM Trolox eq/kg FW), Gisela 5 (9.28 mM Trolox eq/kg FW), Adara/Major (9.21 mM Trolox eq/kg FW), and Adara/Marianna (9.18 mM Trolox eq/kg FW), while the two most vigorous rootstocks yielded fruit with the lowest values: MaxMa 60 (8.83 mM Trolox eq/kg FW) and Argot (8.22 mM Trolox eq/kg FW) (Table 2).
Total Phenol Content (TPH)
Fruit TPH contents showed a similar trend as the one observed in TAC (Table 2), but in this case trees grafted on Adara/Major yielded fruit with the highest content of polyphenols (1078.31 mg GA eq/kg FW), followed by MaxMa 14 (989.76 mg GA eq/kg FW).
Total Antioxidant Capacity (TAC), Total Phenol Content (TPH) and Total Anthocyanin Content (ACY) of the different rootstocks. Data are expressed as mean of years 2013–2015±standard errors. Different superscript letters for each Nutritional parameter analyzed indicate significant differences (SNK test, p≤0.05)
Total Antioxidant Capacity (TAC), Total Phenol Content (TPH) and Total Anthocyanin Content (ACY) of the different rootstocks. Data are expressed as mean of years 2013–2015±standard errors. Different superscript letters for each Nutritional parameter analyzed indicate significant differences (SNK test, p≤0.05)
Differently from what would be expected from the high TAC value, the PHLA induced a reduced polyphenolic concentration (772.28 mg GA eq/kg FW), which suggests the presence of other components (e.g. vitamin C or other compounds) able to induce antioxidant activity.
Among the dwarfing subjects, there seems to be a direct correlation between TAC and polyphenolic concentration: in particular, as it can be estimated by the TAC analysis, Gisela 5 which achieved interesting results for TPH (886.53 mg GA eq/kg FW).
Briefly, regarding fruit ACY content, each rootstock/scion habitus is characterized by an accentuated variability and there is not a group that could be indicated as responsible for ACY-rich fruits. MaxMa 60 induced high anthocyanin pigments concentration in the fruits (66.04 mg Pel-3-Glu/kg FW) (Table 2), statistically similar to that realized by the weaker MaxMa14 (64.95 mg Pel-3-Glu/kg FW) and dwarfing Gisela 5 rootstocks (62.62 mg Pel-3-Glu/kg). These three rootstocks achieved excellent results both for TAC and ACY, suggesting that the higher antioxidant capacity could be attributable largely to the increased accumulation of anthocyanins. Adara/Major (55.28 mg Pel-3-Glu/kg FW) and Gisela 6 (54.70 mg Pel-3-Glu/kg FW) showed medium-high levels of anthocyanins. The worst result was registered for fruits from P. avium rootstock (42.93 mg Pel-3-Glu/kg FW).
Correlation matrix
The performing of Pearson’s correlation on the parameters analyzed in the years 2011–2015 showed some interesting correlations among the productive parameters (Table 3). First of all, the TCSA and the summer pruning showed a very high positive correlation (0.88): this is not surprising, because both these parameters indicate the vigor of a plant, and a vigorous plant which shows a high TCSA will also have a massive vegetative system, which produces high amounts of shoots to be pruned during summer (Adara/Marianna is the best example). The TCSA is positively correlated also with the total yield (0.33). A high vigor of the plant tends to diminish the plant yield efficiency, given that the TCSA is the denominator in the efficiency ratio: for this reason, the plant yield efficiency is negatively correlated both to the TCSA (–0.73) and to the summer pruning (–0.68). The plant yield efficiency is also negatively correlated to AFW (–0.36).
Regarding the sensorial parameters (Table 3), soluble solids content and total acidity are negatively correlated to each other (–0.61). In particular, SS are negatively correlated also to TCSA (–0.49) and AFW (–0.19), but positively correlated to the plant yield efficiency (0.52). On the contrary, TA is positively correlated to TCSA (0.30) and AFW (0.38), but negatively correlated to the plant yield efficiency (–0.44). This means that fruits obtained from vigorous plants (with a high TCSA) tend to be less sweet and more acid. Also fruits with higher AFW seemed to be more acidic and less sweet than smaller fruits.
Pearson’s correlation matrix of the parameters analyzed in the period 2011–2015. Red numbers with an asterisk indicate a significant correlation with p≤0.05. ns = not significantly correlated
Pearson’s correlation matrix of the parameters analyzed in the period 2011–2015. Red numbers with an asterisk indicate a significant correlation with p≤0.05. ns = not significantly correlated
Finally, regarding the nutritional parameters evaluated in the years 2013–2015 (Table 4), TAC, TPH and ACY have been directly correlated to each other: in fact, fruits with high amount of total phenols and anthocyanins, both high antioxidants, show also a high antioxidant potential through the TAC value.
Pearson’s correlation matrix of the parameters analyzed in the period 2013–2015. Red numbers with an asterisk indicate a significant correlation with p≤0.05. ns = not significantly correlated
The commercial rootstocks tested in this paper showed a different capacity in promoting fruit yield, sensorial and nutritional quality characteristics of “Sweetheart”, the late and self-fertile sweet cherry cultivar. The use of a single reference variety has allowed to enhance the features, induced by 10 rootstocks, fixing them on a unique base thus enabling an easier comparison. This is a positive aspect but it imposes some limitations such as, the analyses are considered valid on a limited cultivation area and related to specific climatic conditions. The results obtained only with one cultivar can give an indication of the behavior of the tested rootstock genotypes in the cultivation conditions, that can be helpful to identify the best rootstock to be used; but keeping in mind the possible different responses induced by different cultivars.
From the analyses carried out, there is not a univocal combination between the scion and rootstock equally valid for all the considered parameters. For the agronomical traits, taking into account the total productions, the vigorous rootstock Adara/Major and the weaker Gisela 6 stands out for cherry yield. From the sensorial point of view, which includes parameters such as the AFW, SS and TA, rootstocks belonging to the class of weaker, including the MaxMa 14 and the Gisela 6, and the dwarf Gisela 5 stand out, producing fruits with low AFW but optimal sugar/acid ratio. The valid combinations for nutritional aspects are the Adara/Major for vigorous, the best ever MaxMa 14 for weaker and Gisela 5 for dwarfing rootstocks.
The results of this study reveal that, for non-intensive trial systems in Mediterranean pedoclimatic conditions, the rootstock that can ensure good yields, sensorial and nutritional quality of “Sweetheart” cultivar is the Adara/Major. From the productive and qualitative point of view, Gisela 6 is well adapted to these conditions, while the less productive Gisela 5 is recommended for the fruit nutritional quality but it requires more appropriate cultivation practices for improving the productive parameters. The MaxMa 14 rootstock, despite excelling for nutritional quality of its fruits, induces a high reduction of vigor associated with poor vegetative renewal and insufficient AFW, hence unsuitable for this testing. The individuals grafted on Argot and P. avium are less recommended for this area: these plants presented fruits with big dimensions, but very acid, with low sugar content and antioxidant compounds.
Funding
The authors report no funding.
Conflict of interest
The authors have no conflict of interest to report.
Footnotes
Acknowledgments
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Francesca Balducci and Luca Mazzoni were supported by the H2020-SFS-2014-2015 GoodBerry project No. 679303. The authors would like to thank to Dr. Cecilia Limera for extensive editing of the manuscript.
