Optimization of a virus-induced gene silencing system and functional elucidation of MaBAM9b in postharvest banana fruits

Background: The genetic basis of metabolic pathways that operate during fruit ripening needs to be understood before the nutritional value of the banana can be improved. The banana is a typical starch conversion fruit, and β-amylase is a key enzyme that may play an important role in starch degradation during the ripening process. Musa acuminata β-amylase 9b (MaBAM9b) is closely related to starch degradation. However, its exact function in starch degradation has not been demonstrated in banana. Stable genetic transformation to identify gene function is time- and energy-consuming. Thus, an ecient and rapid method is needed for functional identication. Virus-induced gene silencing (VIGS) is a reverse-genetics method based on RNA-mediated antiviral plant defense that has been used to rapidly identify gene functions in plants. The aim of this study is to optimize a VIGS system and functional elucidation of MaBAM9b in postharvest banana fruits. Results: Using 0.5% iodine-potassium-iodide (I 2 -KI) staining for 150 s, we found that 1:3 TRV1:TRV2-MaBAM9b cultivated at 30 mmHg for 30 s to an optical density (OD) of 0.8 at 600 nm, and kept on Murashige & Skoog (MS) media for 5 days produced the best silencing results. Under these conditions, the total starch content was greatly increased, whereas the β-amylase activity, soluble sugar content, and expression of endogenous MaBAM9b greatly decreased. Conclusions: The system described here is particularly useful for studying genes and networks involved in starch conversion in fruit, which alone would not produce a visual phenotype. This system will provide a platform for functional genomics and fruit quality improvement in the banana.


Background
Banana is an important and pleasantly avored tropical fruit that is popular with consumers worldwide [1]. As a typical starch conversion fruit, the starch content of banana fruit upon harvesting can reach 12-35% fresh weight (FW) [2]. Starch is degraded and converted to sucrose rapidly during the post-harvest banana fruit ripening process [3]. The soluble sugars may reach up to 20% FW of the pulp in the ripe fruit, with sucrose accounting for approximately 80%, whereas glucose and fructose make up almost all the remaining 20% of the soluble sugars in equal proportions [2]. During banana fruit ripening process, the sucrose content varied from 12.2 to 411.8 mg g -1 FW, while the glucose and fructose contents varied from 2.1 to 100.5 mg g -1 FW and 1.5 to 103.8 mg g -1 FW, respectively (not published data). The conversion e ciency directly decides the fruit sweetness, rmness, avor, nutritional quality, and shelf life. Therefore, banana is a model fruit for investigating fruit starch metabolism, β-amylase is a key enzyme that converts starch to glucose [4]. Musa acuminata β-amylase 9b (MaBAM9b, Ma05_t07800.1) is highly expressed during banana fruit ripening and is closely related to starch degradation [5][6]. However, its exact function in starch degradation has not been demonstrated in banana. Elucidating the function of MaBAM9b in banana using genetic transformation could take in excess of three years due to various banana-speci c challenges such as slow growth process, low transgenic e ciency and high dependence on genotype [7]. Therefore, developing an e cient way to rapidly and exactly identify gene function is important for banana functional genomics.
In this report, a VIGS-mediated banana fruit slice system was developed to rapidly elucidate the role of MaBAM9b in banana fruit starch metabolism. The effects of different ratios of pTRV2-MaBAM9b with pTRV1, the OD600 values of the bacterial uid, and the co-cultivation times on the silencing results were carefully investigated. The results showed that the co-cultivation of 2-4-mm-thick fruit slices at a 1:3 ratio of pTRV1: pTRV2-MaBAM9b at OD600 0.8 at 30 mmHg for 30 sec and kept on Murashige & Skoog (MS) media for 5 d could obtain the most e cient silencing results based on an analysis of the contents of total starch, glucose, β-amylase activity, and MaBAM9b expression. These results will provide a platform for banana functional genomics on starch metabolism and should contribute towards developing a technique for controlling starch degradation.

Results
The effects of different I 2 -KI solutions and times on the staining results As shown in Fig. 1, as the treatment duration and concentration increased, the blue stain gradually became darker. The 0.1% (a) and 0.25% (b) I 2 -KI solutions did not complete their staining in any of the tested treatment times ( Fig. 1A-F). The 0.75% (d) and 1.0% (e) I 2 -KI solutions resulted in an uneven dark or light color under long or short treatment times, respectively. The banana fruit slices stained with 0.5% I 2 -KI solution for 150 s obtained a more uniform staining effect that allowed for easier identi cation.
The effects of different co-cultivation times on the silencing results As shown in Fig. 4A, as the co-cultivation time progressed, the blue stain of I 2 -KI became gradually darker.
When the co-cultivation time reached 5 d, the color was the darkest. However, when the co-cultivation time increased to 7 d, the staining became weaker. This result demonstrated that co-cultivation for 5 d was best for suppressing starch degradation. Further investigation found that the total starch contents at 3 d, 5 d, and 7 d were 198.4, 315.7, and 301.4, increasing by 12.5%, 79.1%, and 71.0% compared to 1 d, respectively (Fig. 4B). Conversely, the β-amylase activities at co-cultivation times of 3 d, 5 d, and 7 d were signi cantly reduced compared to 1 d, decreasing by 10.7%, 29.8%, and 27.4%, respectively (Fig. 4C). The contents of glucose, fructose, and sucrose also decreased greatly by 60.5%, 53.8%, and 64.1% at an cocultivation time of 5 d, respectively (Fig. 4D). The expression of endogenous MaBAM9b was greatly suppressed by 26.0%, 58.0%, and 42.0%, respectively. These results suggested that 5 d of co-cultivation obtained the best silencing e ciency.

Discussion
Banana is an important global tropical climacteric fruit with high nutritional value. Fruit ripening involves a quality formation process, during which many irreversible physiological and biochemical changes occur. Starch is the main component and the building block of banana fruit. Starch conversion to sugar is a primary metabolic process that directly in uences the fruit quality, such as the texture, avor, and shelf life [3]. Understanding the molecular mechanism of fruit starch degradation metabolism is necessary for improving fruit quality. MaBAM9b is a key enzyme gene that might play an important role in starch degradation during the banana fruit ripening process [5]. However, its function has not been explicitly identi ed in banana. Identifying gene functions using stable genetic transformation is time-and energyconsuming. It is thus necessary that an e cient means for rapid functional identi cation is developed.
VIGS is a reverse-genetics method based on RNA-mediated antiviral plant defense that has been used to rapidly identify gene function in plants [25]. The e ciency of VIGS in inducing endogenous gene silencing is related to multiple factors, such as the correct viral vectors, the characteristics of the target gene, the infection buffer, the activity of Agrobacterium in the bacterial suspension, and the plant growth environment suitable for silencing [26]. Using MaBAM9b as an example, we optimized the VIGS system to facilitate gene function analysis.  has been widely applied in rice [27], Zea mays [28], and Arabidopsis [29]. The result of I 2 -KI staining is in uenced by factors such as staining solution and time. However, these factors have not been optimized in banana. Here, we investigated the effects of the concentration of the staining solution and the staining duration on the staining e ciency. The results indicated that 2-4 mm-thick banana fruit slices stained with 0.5% I 2 -KI solution for 150 s obtained the best staining effect. This result provides a reference for the rapid identi cation of starch contents, especially for high-starch fruits.
Successful gene silencing depends upon a dynamic interplay between virus spread and plant growth, both of which can be in uenced by various conditions [26]. Different fruits have different optimized conditions for VIGS. For Lycium plants, the optimized conditions included an OD600 of 1.0, with the cells incubated at room temperature for 3-4 h. The pTRV1 and pTRV2 solutions were mixed at a ratio of 1:1 for injection [30]. In tomato, the OD600 was adjusted to 0.5 or 1.0 for all of the vectors, and the resuspended cells with pTRV1 and pTRV2 were then combined at a ratio of 1:1 and kept for more than 10 d in a plastic tray [8,31]. In loquat, the OD600 was adjusted to 1.2, and the resuspended cells with pTRV1 and pTRV2 were combined at a ratio of 1:1 and co-cultivated for a week [18]. In the present study, the effects of different ratios of pTRV2-MaBAM9b and pTRV1, OD600 values, and co-cultivation durations on the silencing results were systematically investigated. The results showed that a 1:3 ratio of pTRV1: pTRV2-MaBAM9b, an OD600 of 0.8, and 5 d of cultivation obtained the most e cient silencing results.
This system was something different from other fruits and re ected the distinguishing qualities of banana fruit.

Conclusion
A TRV-mediated VIGS system was successfully constructed and used to e ciently silence MaBAM9b, which resulted in increased total starch contents and decreased β-amylase activity, soluble sugar content, and endogenous MaBAM9b expression. These ndings support that MaBAM9b plays a vital role in banana fruit starch degradation metabolism, thus providing a platform for banana functional genomics and for the breeding of new banana varieties.

Plant materials
Banana (Musa AAA group cv Brazilian) fruits were harvested from the banana plantation of the Institute of Tropical Bioscience and Biotechnology (Chengmai, Hainan, 20N, 110E) and transported back to the laboratory. The healthy middle fruit comb was selected and separated into a single fruit nger. After surface-sterilization for 10 min with 0.1% sodium hypochlorite, the fruit ngers were air-dried and stored for 8 d at 22°C and 80% relative humidity.
A sharp blade that has been surface-sterilized with 75% ethanol was used to uniformly section the fruit ngers into 2-4-mm-thick slices, with the fruit ends discarded.

Quanti cation of physiological index
The total starch content was detected following the method of Miao et al. (2014) [35]. The β-amylase activity was measured according to the method of Hou et al. (2017) [36]. The soluble sugar (fructose, glucose and sucrose) contents were analyzed using High-Performance Liquid Chromatography (HPLC; Waters, Milford, CT, USA). The HPLC method was developed and validated using an AMINEX 87H column, 10 mmol L −1 sulfuric acid as eluent and refraction index detection [37].

Statistical analysis
At least three biological replicates were assessed, and all of the data were analyzed using One-way analysis of variance (ANOVA) and Student's t-tests for determination of signi cant differences.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests. Figure 1 The effects of different I2-KI solution concentrations and treatment times on the staining results. A-F, staining for 30 s, 60 s, 90 s, 120 s, 150 s, and 180 s, respectively; a-e, I2-KI solutions of 0.1%, 0.25%, 0.5%, 0.75%, and 1.0%, respectively Figure 2 The effects of different ratios of pTRV1: pTRV2-MaBAM9b on the silencing results. A, the effects of I2-KI staining; B, soluble sugar contents; C, β-amylase activity; D, total contents of starch; E, expression of MaBAM9b. a-c, the ratios of 1:1, 1:2, and 1:3, respectively. d, empty TRV2 plasmid as a negative control.