**1. Introduction**

Blueberries (*Vaccinium* spp.) contain bioactive compounds which offer potential health benefits and have witnessed a large increase in production over the last two decades [1,2]. Blueberries are native to North America and some common cultivated species include lowbush (*Vaccinium angustifolium* Ait.), northern highbush (*Vaccinium corymbosum* L.), rabbiteye (*V. virgatum* Ait.), and southern highbush (hybrids of *V. corymbosum, V. virgatum,* and *V. darrowii* Camp.) [3–5]. During fruit development, blueberry fruit on a branch mature at different rates, resulting in a non-uniform ripening period extending over 2 to 3 weeks [6]. As a result, blueberries intended for the fresh fruit market are hand harvested three to five times depending on the variety. This makes harvesting a labor intensive and expensive component of blueberry production, requiring up to 520 h of labor/acre and costing up

to \$0.70 per pound of harvested fruit [7–9]. Concentrating the period of ripening could help reduce the required number of harvests and reduce costs associated with production. Ripening is regulated by multiple plant hormones such as ethylene, abscisic acid, auxins, and jasmonates [10]. External applications of plant growth regulators (PGRs) that influence the levels or activity of these plant hormones may alter the progression of ripening and thereby help in concentrating the period of fruit ripening for efficient harvesting. Therefore, understanding the progression of ripening and developing tools such as PGR applications can help in improving the efficiency of blueberry harvesting.

Fruit ripening is a coordinated process involving changes in fruit texture, color, flavor, and susceptibility to biotic and abiotic factors [11,12]. Although all fruit display these changes during ripening, fruits can be generally classified into one of two types depending on physiological and biochemical changes accompanying the initiation and progression of ripening: climacteric and non-climacteric. In climacteric fruits such as tomato (*Solanum lycopersicum*), banana (*Musa* spp.), and apple (*Malus* × *domestica*), ripening is accompanied by a peak in respiration and ethylene production [11–14]. In such fruits, once ethylene production is triggered at ripening, it is autocatalytic and is one of the key factors that regulate changes associated with ripening. Non-climacteric fruits, such as strawberry (*Fragaria* × *ananassa*) and grape (*Vitis vinifera*), do not exhibit an increase in respiration and ethylene in association with ripening. In these fruits, the role of ethylene and other signals in regulating ripening are not completely understood [12,14–16]. The roles of climacteric respiration and ethylene in the progression of fruit ripening in blueberry are unclear. Some previous studies observed an increase in respiration and ethylene during blueberry ripening, suggesting a potential climacteric nature to the ripening process [6,17,18]. Also, external application of the ethylene-releasing compound ethephon accelerated the progression of ripening and reduced the harvest time in blueberry [19–21]. However, several other studies have classified blueberry as a non-climacteric fruit that does not display a substantial climacteric rise in respiration or ethylene evolution [22]. Hence, further studies are required to better understand the contribution of ethylene in blueberry ripening, and to determine if manipulation of this plant hormone offers a viable option for controlling ripening.

Abscisic acid (ABA), another plant hormone, plays an important role in many developmental processes such as adaptation to stress and seed dormancy. In addition, recent work has suggested a role for ABA during ripening in climacteric as well as non-climacteric fruit. Abscisic acid concentration increases during fruit ripening in apple [23], orange (*Citrus sinensis*) [24], cherry (*Prunus avium*) [25], strawberry [26,27], and grape [28]. In strawberry, decreased expression of *9-cis-epoxycarotenoid dioxygenase* (*FaNCED1*), a gene coding for an enzyme involved in ABA biosynthesis, lowered ABA levels and prevented fruit from ripening normally [27]. In grape, ABA applications improved red color and helped achieve early harvest, underlining its potential for accelerating ripening [29–31]. Further, in tomato, ABA may function upstream of ethylene and induce the expression of ethylene biosynthesis genes to regulate ripening [32,33]. Similarly, in banana, ABA applications may enhance ethylene sensitivity and coordinate ethylene-regulated ripening [34]. In bilberry (*V. myrtillus* L.), which is closely related to blueberry, ABA has been implicated in the regulation of ripening [35]. In highbush blueberry (*V. corymbosum*), ABA levels increase at the onset of ripening and may be involved in regulating the production of flavonoids [36]. However, external applications of ABA delayed ripening and increased fruit firmness in southern highbush blueberry (*V. corymbosum* interspecific hybrids) [37]. Although these studies sugges<sup>t</sup> a potential role for ABA in regulating blueberry ripening, it requires further investigation, especially to determine if external ABA applications can be used to reliably manipulate the progression of this process across different blueberry species.

Jasmonates are another group of phytohormones with well-characterized roles in defense responses and developmental processes such as senescence [38]. Jasmonates have been implicated recently in the regulation of fruit ripening [14,39]. In tomato and apple, jasmonates promoted ethylene biosynthesis by inducing the expression of genes involved in its biosynthesis [40]. In apple, methyl jasmonate (MeJA) applications influenced the production of aromatic volatiles, an integral component of fruit flavor, in a cultivar-dependent manner [41]. In peach (*Prunus persica*), jasmonates

delayed ripening [42]. Although MeJA had a negative effect on ethylene biosynthesis during ripening in peach, it still promoted anthocyanin biosynthesis [43]. In non-climacteric fruits such as cultivated strawberry and Chilean wild strawberry (*Fragaria chiloensis*), application of MeJA increased ethylene evolution and respiration, and promoted color development thereby accelerating ripening [44,45]. In raspberry (*Rubus idaeus*), MeJA application increased flavonoid content, total soluble solids (TSS) content, and total sugars, and lowered titratable acidity (TA), thus influencing multiple ripening characteristics [46]. Together, these emerging data sugges<sup>t</sup> that the effect of jasmonates on fruit ripening may be species-specific, requiring further evaluation in the species of interest. Further, preharvest and postharvest applications of MeJA may not only improve fruit quality but also offer a protective role by limiting pathogen growth as seen in strawberry and peach [39]. MeJA applications on highbush blueberry resulted in changes in total sugar content, total anthocyanin content, and expression of anthocyanin biosynthesis genes in a cultivar-dependent manner [47]. However, the specific role of MeJA in blueberry ripening and its effect on postharvest fruit quality attributes is not clear and has not been investigated previously.

While the effects of multiple PGRs on fruit ripening have been evaluated in various fruit crops, these have not ye<sup>t</sup> been tested extensively in blueberry. Blueberry production could greatly benefit from the use of PGRs that help manipulate the time of ripening. Hence, the main goal of this research was to evaluate three PGRs, ethephon, ABA, and MeJA, for their ability to alter the progression of ripening in two rabbiteye blueberry cultivars. These three PGRs were selected for further study due to previous research suggesting their potential as described above. Furthermore, as preharvest applications of these PGRs can influence postharvest fruit quality and storage characteristics including disease symptom development, the effects of their application on postharvest fruit quality and disease incidence were also evaluated.

#### **2. Materials and Methods**

#### *2.1. Plant Material and PGRs*

Two rabbiteye blueberry cultivars, Premier and Powderblue (both at 5 years since planting), grown at the Durham Horticulture Farm in Watkinsville, GA were used for this study. All applications were performed when around 30–40% of fruit on the plant were ripe. Whole plants were sprayed using a hand-held sprayer until run-off. For the early-maturing Premier, the treatments consisted of: control (water), 250 mg L−<sup>1</sup> ethephon, 600 mg L−<sup>1</sup> ABA, and 0.5 mM MeJA. All treatments were applied on 20 June 2016 along with an adjuvant (0.15% Latron B-1956; Simplot, Lathrop, CA, USA). The doses were determined based on preliminary studies. Applications on Premier were made in the evening close to sunset to minimize photo-destruction of ABA. For the later-maturing Powderblue, the same treatments were applied on 9 July 2016 except that the concentration of ABA and MeJA were increased to 1000 mg L−<sup>1</sup> and 1 mM, respectively. Due to potential rainfall in the late afternoon, all applications on Powderblue were made early in the morning. For each treatment, four replicates consisting of four individual plants were used in both cultivars.

#### *2.2. Rate of Ripening*

Prior to PGR application, three 50 to 100-cm-long shoots, each consisting of a total of approximately 50–100 fruit, were tagged per replicate. Very small immature as well as ripe fruit were removed from the tagged branches. The number of green, pink, and ripe fruit was counted prior to and after PGR applications at regular intervals (2–4 days) up to 11 days and 13 days for Premier and Powderblue, respectively. Fruit counted as pink ranged from having around 25% pink color (75% green) to around 75% pink (25% blue) on the fruit surface. Fruit was considered ripe when the color of the entire fruit was blue. The percentage of green, pink, and ripe fruit was calculated from these data for each assessment date.

#### *2.3. Postharvest Fruit Quality and Disease Incidence*

Two additional shoots containing around 300 fruit (total) were tagged on each replicate to study the effect of PGR applications on postharvest fruit quality and disease incidence. Very small immature and ripe fruit at the time of application were removed. Ripe fruit were hand-harvested approximately 10 days after application of PGRs and split into three groups for postharvest fruit quality analyses. These groups were randomly assigned to one of the following treatment periods for postharvest evaluation: PH + 1 (postharvest + 1 day); PH + 15 (postharvest + 15 days); and PH + 29 (postharvest + 29 days). For postharvest storage, fruit were placed in a walk-in cooler set to 4 ◦C and a relative humidity of 90–95%. For each storage period and replicate, around 40 fruit were used for fruit quality evaluation and around 60 fruit were used for disease incidence evaluation. For fruit quality analysis, visual evaluation of quality was conducted and weight, texture, pH, TA, TSS content, and berry color were measured. For visual evaluation of fruit quality, 30 fruit per replicate were scored for symptoms of bruising such as tears, dents, leakiness, and appearance of mold to determine the percent defective fruit. Fruit weight was measured on 20 fruit, using a balance (Quintix® Precision Balance, Sartorius, Bohemia, NY, USA). Fruit texture measurements were made using a fruit texture analyzer (GS-15, Güss Manufacturing, Strand, South Africa). Two tests, compression and skin puncture, were performed on 12 fruit per replicate for determining fruit texture by orienting the fruit on the equatorial plane. For compression analyses, a probe with a 15-mm diameter end plate was used with parameters set at a measure speed of 5 mm s<sup>−</sup><sup>1</sup> and measure distance of 1 mm. To measure skin puncture force, a 1.5-mm probe was used with parameters set at a measure speed of 5 mm s<sup>−</sup><sup>1</sup> and measure distance of 3 mm.

For measuring pH, TA, and TSS, juice from around 30 g of fruit was extracted using a blender followed by centrifugation for 10 min at 3901× *g* on a benchtop centrifuge (Allegra X-22, Beckman Coulter Life Sciences, Indianapolis, IN, USA). The supernatant was filtered through cheesecloth. Around 0.3 mL of supernatant was used to determine TSS using a digital handheld refractometer (Atago USA, Belleveue, WA, USA). To determine pH and TA, the supernatant was titrated using an automatic mini titrator (Hanna Instruments, Woonsocket, RI, USA) and alkaline titrant. The titrator has a pH electrode which provided an initial pH value of the supernatant before titration is initiated. For TA, the data are expressed as percent citric acid (CA) equivalents. Fruit color was determined on 20 fruit using a handheld colorimeter (3nh Technology Co., Shenzhen, China).

To determine natural postharvest disease incidence, fruits were maintained at 23–25 ◦C for 4 days after removing them from cold storage at the three postharvest intervals described above. Fruit displaying symptoms of disease and/or signs of plant pathogens were recorded using around 60 fruit per replicate. The associated pathogens were identified microscopically as described in Mehra et al. [48].

Statistical analysis (one-way analysis of variance for a randomized complete block design) was performed separately for every time-point after harvest using JMP Pro 12 (SAS Institute, Cary, NC, USA). Means were separated using Tukey's honestly significant difference (HSD) test ( α = 0.05).
