Analysis of ethylene biosynthesis gene expression profile during titanium dioxide (TiO2) treatment to develop a new banana postharvest technology

Banana is an important crop that demands proper methods in postharvest handling. As a climacteric fruit, the banana fruit ripening process is affected by ethylene. Several methods have been developed to extend the shelf life of a banana, such as using ethylene scrubbers. In this study, titanium dioxide (TiO2), a photocatalyst, was used as an alternative method to delay the fruit ripening process. The effect of TiO2 on the ripening‐related gene MaACS1 was investigated. Banana fruits were placed in a TiO2‐coated glass chamber and observed for ten days. Fruit ripening in the treated chamber was delayed for eight days compared to the control. Total RNA was extracted from control and TiO2‐treated fruit pulp and synthesized into cDNA. Reverse transcription PCR was performed to investigate the gene expression, which showed that MaACS1 expression was relatively lower than treated control. The finding of these studies suggested that the TiO2 chamber has the potential to extend the shelf life of banana by delaying its ripening process and decreasing the expression of MaACS1. To the best of our knowledge, no previous study has investigated the effect of TiO2 on the expression of genes related to banana fruit ripening.


Introduction
Banana is one of the most consumed fruits and catego rized as an important commodity in various countries. The highly nutritious contents in banana such as vitamins A, B1, B2, and C make the fruit popular around the world (Cano et al. 1997). Indonesia is one of the top 10 countries in banana production (FAO 2019). However, Indonesia also faces problems in the postharvest handling of banana production. The majority of postharvest handling of ba nana is performed traditionally, which consequently leads to lowquality banana production. Some of the banana agroindustries around the world have attempted using a modified atmosphere and a controlled atmosphere tech nology to extend the shelf life of the fruit. However, such techniques may not be effectively applicable, especially for small Indonesian farmers in rural areas. Therefore, there is a need for other alternatives such as low energy technologies to extend the shelf life of banana by delaying its ripening process.
Ripening is a natural process that occurs in fruits and is highly influenced by a gaseous phytohormone, ethy lene (C 2 H 4 ) (Guo and Ecker 2004). Ethylene is syn thesized from Sadenosyl methinonine (AdoMet), which is then converted into 1aminocyclopropane1 carboxylic acid (ACC) with the help of the enzyme ACC synthase (Yang and Hoffman 1984). ACC is the intermediate pre cursor to generate ethylene. ACC is then converted into ethylene with the help of an ACC oxydase enzyme. Stud ies have demonstrated that MaACS1, a member of the ACC synthase (ACS) gene family, is responsible for the ripening process of the fruit (Liu et al. 1999; Karmawan et al. 2009; Dwivany et al. 2016. Titanium dioxide (TiO 2 ) has been used as a photocata lyst to degrade ethylene that is emitted by fruits at low tem peratures (Hussain et al. 2010). Another recent study con ducted using TiO 2 nanofiber has demonstrated a delay in banana softening and color change (Zhu et al. 2019). It has been reported that ethylene degradation by TiO 2 produces carbondioxide and water and triggers a low ratio of O 2 to CO 2 in the atmosphere (Charoenshap et al. 2012). These conditions result in the reduction of ethylene biosynthe sis since oxygen is needed to convert ACC to ethylene by ACC oxidase (Kanellis et al. 2009). As mentioned above, ACC synthase is also an important enzyme which its activity also triggered by ethylene autocatalytic reac tion. Thus, lower ethylene production will influence this reaction (Inaba et al. 2007). In the present study, we in vestigated the effect of TiO 2 applied as a thin coating on the process of fruit ripening and the expression of the ba nana ripeningrelated ACC synthase gene, MaACS1. The results of this study demonstrated that TiO 2 has the poten tial to extend the shelf life of banana fruit.

Fruit Storage Chamber Preparation
Fruit storage chambers made from glass (1.5 L) were used in the experiment. The chambers were coated with a thin layer of TiO 2 that was prepared by mixing crystal TiO 2 (J25) nanoparticles in ethanol as a solvent. The anatase crystal had a pH range of 3.5-4.5. The TiO 2 nanoparticles were coated on the chambers using an airbrush sprayer and then heated using a shot gun. All preparations were per formed at the Chemistry Department, Institut Teknologi Bandung. Charcoal pouches were placed in all chambers to absorb water produced from respiration.
Best quality Cavendish bananas were assorted, placed inside the chambers, and stored until ripening for approx imately ten days. The observation time points were cho sen from day 0 to day 8 (Dday, D1, D3, D6, and D 8). Control group bananas were placed in a closed cham ber without light exposure, whereas treated group bananas were placed in a closed chamber with UV radiation (300-400 nm) for 24 h. Both treatments were performed at room temperature (26°C-27°C). The observation was per formed at Bioscience and Biotechnology Research Center, Institut Teknologi Bandung.

Physical and Physiological Analysis
Physical and physiological characteristics of the bananas were evaluated based on peel color changes, starch con tent measurements using the iodine test, pulptopeel ra tio measurements, and sugar content values measured us ing total soluble solids (TSSs) (Dadzie and Orchard 1997). Peel discoloration and starch content measurements based on the iodine test during ripening were documented using a Canon IXUS 230 HS digital camera. Meanwhile, TSS measurements were performed using a hand refractometer (Atago™).

Gene Expression Profile Analysis
Molecular analysis was conducted to investigate the pat tern of MaACS1 gene expression in control and treated groups using genespecific primers (Karmawan et al. 2009; Handayani andDwivany 2014), wherein MaGA PDH was used as a reference gene to compare the expres sion profile. Total RNA was extracted as described by Cordeiro et al. (2008) with some modifications. The to tal RNA was then synthesized into complementary DNA (cDNA) using iScript™ cDNA Synthesis (Biorad cata log number: 1708890). Gene expression profile was analyzed by reverse transcription PCR (RTPCR) using the GreenTaq Master Mix kit (Promega, catalog num ber: M7122). All the PCR results were confirmed by electrophoresis on 1% (w/v) agarose. Semiquantitative analysis of gene expression was performed using ImageJ quantification (http://www.imagej.net) as well as by Ttest analysis.

Physical and Physiological Analysis
Before conducting molecular analysis, the physical and physiological characteristics of all bananas from the con trol and treatment groups were evaluated (Dadzie and Or chard 1997). Based on the ripening stage, the fruit can be divided into two categories, climacteric, and nonclimac teric (Friend and Rhodes 1981). Banana is a climacteric fruit and has the unique feature of ethylene production during its ripening stage. During the entire ripening pro cess, the cells undergo several physical and physiological as well as molecular changes. Banana peel color changes can be easily monitored by physical analysis before con ducting physiological and molecular analyses.
The assessments in this study included observation of color changes in the banana peel, starch content analysis, pulptopeel ratio measurement, and TSS analysis. Color changes in the peel in both the control and treatment groups were observed during the ripening stage ( Figure 1). Data were collected at the observation points Dday, D1, D3, D6, and D8. Results showed that treat ment using the TiO 2 chamber could delay the ripening pro cess as indicated by the color changes from mature green to yellow and brown. Banana peels in the control group were generally brown in color since day 6, whereas the treated banana peels were green to yellow in color.
Regarding starch content analysis, Dadzie and Or chard (1997) had stated that this method is simple, rapid, and inexpensive to detect starch conversion into sugar dur ing the ripening process visually. In this study, the treated bananas showed differences compared with the control group bananas, and after six days of observation, the pulp still had the black coloration (Figure 1). This indicated that the treated group had abundant starch content com pared with the control group. This result was also con firmed by the TSS values ( Figure 2). Bananas from both the control and treatment groups exhibited an increment in the o Brix during ripening. The faster ripening process in the control group was also confirmed by its o Brix value that was higher from Dday to D6 than that in the treated group. However, on D8, the sugar content in the con trol group was found to be lower than that in the treatment group. Regarding the color changes in the banana peel from green to yellow and brown after being overripened, studies have reported that these changes are caused by  the degradation of chlorophyll by the chlorophyllase en zyme (Matile et al. 1996; Duan et al. 2007). Furthermore, the yellow coloration on the peel is due to the increasing amount of carotenoid pigments in the fruit (Subagio et al. 1996). The brown spot or senescent spotting that appeared on the ripe banana could be due to cell necrosis as a result of chlorophyll degradation (Mosera et al. 2009). The color and size of the spot could increase rapidly during the ripen ing process (Karmawan et al. 2009). In addition, it has been reported that starch contributes to 20% of the major component in the fruit and gets converted into a carbon source during ripening (Bierhals et al. 2004). During the process of ripening, starch conversion results in a sweeter fruit. In this study, the starch content was assessed qualita tively by the iodine test, in which iodine reacts with starch and results in black color around the fruit flesh. On the other hand, the sugar content analysis done by measuring the TSS in the pulp of the fruit indicated that TSS values increased as the ripening process progressed (Dadzie and Orchard 1997).
These results also corresponded to the pulptopeel ra tio measurement. Based on the data obtained, both the control and treatment groups exhibited an increase in the pulptopeel ratio during the ripening process ( Figure 3). However, the pulptopeel ratio has been considered to be constant and better in assessing the ripening index (Dadzie and Orchard 1997). A change in the pulptopeel ratio is one of the physical indicators of ripening -sugar con centration increases in the pulp, which causes differences in the osmotic pressure in the tissue. Meanwhile, the peel loses its water content due to transpiration. Hence, the pulptopeel ratio shows an increasing trend during the ripening process. This measurement can also be per formed using the peeltopulp ratio, but this would show a concomitantly decreasing trend with the ripening pro cess. In this study, bananas in the treated group exhibited a lower ratio than the control group bananas, indicating a correlation with a slower ripening process in the treated group.

Gene Expression Profile Analysis
The MaACS1 gene expression profile analysis was con ducted as described by Handayani and Dwivany (2014) us ing MaGAPDH as the reference gene ( Figure 4). Results showed that the expression level of MaACS1 in the con trol group was statistically significantly higher than that in the treated group (p<0.05). Lower MaACS1 gene expres sion may result in lower ethylene synthesis, as the gene has been reported to be a member of the ACC synthase gene family that converts AdoMet into ACC with the help of the enzyme ACC synthase. As mentioned earlier, ACC is the intermediate precursor of ethylene and influenced by ethylene autocatalytic reaction. The change of O 2 to CO 2 ratio in the atmosphere by TiO 2 treatment may result in a reduction of autocatalytic reaction and ethylene biosyn thesis. The MaACS1 has been reported as a marker for two different treatments to prolong banana fruit ripening, such as low temperature and fungicide storage as well as chitosan coating treatment (Dwivany et al. 2016; Lustriane et al. 2018).
On the basis of previous studies, the expression level of the MaACS1 gene increases significantly during early ripening and then decreases (Inaba et al. 2007; Karmawan et al. 2009). In this study, the expression of MaACS1 was increased until day six and then decreased on day 8.

Conclusions
This study has provided further information about the ef fects of using the TiO 2 chamber on the physical, physio logical, and gene expression changes during banana ripen ing. Earlier studies have also used TiO 2 application to delay papaya and tomato ripening (Maneerat andHayata 2006; Lourenço et al. 2017). However, to the best of our knowledge, this study is the first investigation on banana ripening using the TiO 2 chamber treatment, wherein the results provided new insights into important gene expres sion changes occurring during ripening and further sug gested that these changes can be used as an important biomarker for evaluating banana ripening. The findings of this study can provide a novel strategy to increase the postharvest quality of banana by prolonging its shelf life. Finally, these results can also provide a better understand ing of the process of banana ripening and aid the develop ment of postharvest technology.