Abstract

   An increasing attention is being given to treat fruits with ultraviolet
   C (UV-C) irradiation to extend shelf-life, senescence, and protection
   from different diseases during storage. However, the detailed
   understanding of the pathways and key changes in gene expression and
   metabolite accumulation related to UV-C treatments are yet to be
   explored. This study is a first attempt to understand such changes in
   banana peel irradiated with UV-C. We treated Musa nana Laur. with
   0.02 KJ/m^2 UV-C irradiation for 0, 4, 8, 12, 15, and 18 days and
   studied the physiological and quality indicators. We found that UV-C
   treatment reduces weight loss and decay rate, while increased the
   accumulation of total phenols and flavonoids. Similarly, our results
   demonstrated that UV-C treatment increases the activity of defense and
   antioxidant system related enzymes. We observed that UV-C treatment for
   8 days is beneficial for M. nana peels. The peels of M. nana treated
   with UV-C for 8 days were then subjected to combined transcriptome and
   metabolome analysis. In total, there were 425 and 38 differentially
   expressed genes and accumulated metabolites, respectively. We found
   that UV-C treatment increased the expression of genes in secondary
   metabolite biosynthesis related pathways. Concomitant changes in the
   metabolite accumulation were observed. Key pathways that were
   responsive to UV-C irradiation include flavonoid biosynthesis,
   phenylpropanoid bios6ynthesis, plant-pathogen interaction, MAPK
   signaling (plant), and plant hormone signal transduction pathway. We
   concluded that UV-C treatment imparts beneficial effects on banana
   peels by triggering defense responses against disease, inducing
   expression of flavonoid and alkaloid biosynthesis genes, and activating
   phytohormone and MAPK signaling pathways.

   Keywords: antioxidant activity, banana storage, flavonoid biosynthesis,
   hypersensitive response, postharvest treatment, UV-C treatment,
   phenylpropanoid biosynthesis

Introduction

   China is a major consumer as well as producer of banana. In 2019, China
   produced 11,655.65 thousand metric tons of bananas; Guangdong being the
   major banana production area (4,648.31 thousand metric tons)
   ([32]https://www.statista.com/statistics/242954/banana-production-in-ch
   ina-by-region/). Furthermore, China imported 1.94 million tons of
   banana in 2019 alone ([33]www.producereport.com). The produced and
   imported banana needs to be transported to the consumers fresh and
   disease free. In order to keep bananas fresh, there are several storage
   methods; one of which is getting popularity nowadays i.e., treatment
   with ultraviolet C (UV-C) irradiation. UV-C suppresses several
   postharvest diseases. It is a nonionizing and nonthermal way which
   improves the storability of fruits and reduces deterioration of fruits
   and vegetables by triggering defense responses ([34]Wituszyńska and
   Karpiński, 2013; [35]Ding and Ling, 2014; [36]Xu and Liu, 2017).
   However, the dose, application period, and the mechanism of UV-C action
   on fruits is still at the exploratory stage.

   UV-C has been proven to be much safer than the longer wavelength UVs
   i.e., UV-A and UV-B. Also, UV-C is considered more effective in
   controlling bacteria than the longer wavelengths ([37]Bintsis et al.,
   2000). In particular reference to banana, UV-C treatment (0.03 KJ/m^2)
   has been reported to improve abiotic stress tolerance e.g., reduced
   chilling injury in peels. This increased tolerance was attributed to
   reduced damage to membranes, lower browning and damage to chlorophyll
   ([38]Pongprasert et al., 2011). Another study tested the effect of
   0.01–0.30 KJ/m^2 UV-C irradiation on the control of crown rot disease,
   quality of postharvest fruits, and changes in antioxidant capacity of
   banana fruits (Musa AAA “Berangan”). The study found that UV-C
   irradiation was able to not only control the crown rot disease severity
   but also resulted in increase of total phenolics, ferric reducing
   antioxidant power (FRAP), and 1,1-diphenyl-2-picrylhydrazyl (DPPH)
   values ([39]Mohamed et al., 2017). These studies indicate that the
   effect of UV-C treatment on banana (and other fruits) is dose dependent
   and can induce changes in antioxidant capacity, secondary metabolites,
   and disease resistances. However, the knowledge on the molecular
   mechanism of UV-C irradiation effects in these processes is still not
   explored in detail.

   It has been demonstrated that UV-C treatments eliminate/control the
   postharvest diseases and improve shelf-life of fruits. At the same
   time, it does not impart negative physiological impacts on the stored
   tissues of the fruits and vegetables ([40]Urban et al., 2016). In this
   regard, it is known that UV-C treatment may delay senescence and induce
   the biosynthesis of secondary metabolites, which have health benefits
   ([41]Poiroux-Gonord et al., 2010). Though a complete understanding of
   the molecular control of such effects is unknown, studies reported that
   UV-C changes the expression of several genes (succinic dehydrogenases
   and cytochrome c oxidases) which control the important steps in
   tricarboxylic acid cycle ([42]Yang et al., 2014). Similarly, how does
   UV-C treatment affect reactive oxygen species (ROS) generation and
   delay oxidative damage is not completely understood. However, it is
   known that when fruits (e.g., freshly harvested mangoes, strawberries,
   and pineapples) are treated with UV-C irradiation, the activity of
   enzymes related to antioxidant systems as well as defense mechanisms is
   enhanced ([43]González Aguilar et al., 2007; [44]Erkan et al., 2008;
   [45]Alothman et al., 2009a). Regardless of such a diverse knowledge,
   there are limited studies which explored the gene expression of
   pathways that are associated with antioxidant systems, defense
   responses, and similar other pathways that could be associated with
   signaling and growth e.g., hormone signaling and MAPK signaling
   pathways ([46]Urban et al., 2016). Since, the postharvest handling of
   different fruits differs, the mechanism underlying UV-C induced changes
   in the stored tissues could be different. Similarly, the tissue
   structure, biochemical nature, and genetic make-up of the fruits may
   differ, therefore, the UV-C dose, storage conditions, and shelf-life
   will need separate studies to explore related mechanisms.

   Recent developments in transcriptome and metabolome profiling have
   enabled researchers to explore the key molecules that are expressed in
   response to specific stimuli. For example, metabolome profiling of
   mangosteen subjected to different ripening treatments enabled
   researchers to understand the accumulation of certain metabolites in
   response to the applied treatments ([47]Parijadi et al., 2019). Another
   study used multi-omics (transcriptome and metabolome) approach to
   delineate the effects of blue LED on flavonoid and lipid metabolism in
   tea ([48]Wang et al., 2020a). In a similar attempt, we irradiated Musa
   nana Laur. with 0.02 KJ/m^2 UV-C for different days after harvesting
   (0, 4, 8, 12, 15, and 18 days) under similar storage conditions. Based
   on different physiological indicators, we selected the bananas
   irradiated for 8 days and studied the transcriptome and metabolome
   profiles of the peels. Through this approach, we report the key
   transcriptomic and metabolomic changes regarding flavonoid
   biosynthesis, phenylpropanoid biosynthesis, phytohormone signaling, and
   defense responses in M. nana peel under the influence of UV-C
   treatment.

Results

Physiological Analyses

Effect of UV-C Treatments on Banana Peel Appearance and Quality

   In this study, M. nana peel was irradiated with a dose of 0.02 KJ/m^2
   for 0–18 days under controlled temperature and humidity. We observed
   that, when banana was stored under the UV-C dose for a longer time
   i.e., 18 days, a few black spots appeared and the peel was shiny, while
   the CK peel had more black spots and its color was light yellow
   ([49]Figure 1). There was no decay from 0 to 8 days, while from day
   12–18, the decay rate of the UV-C group showed upward trend, while that
   of CK group was highly significantly low (p < 0.01) ([50]Figure 2A).
   The weight loss rate of the CK was higher (significant at p < 0.01)
   than UV-C group at all time points ([51]Figure 2B). These two
   observations suggest that UV-C treatment up to 8 days controls the
   banana peel decay rate, while the weight loss rate remains lower for
   UV-C treated bananas. Another indicator of fruit peel quality is the
   activity of cellulase activity, where it is known that its activity
   affects the hardness and the quality of fruits and vegetables
   ([52]Kannahi and Elangeswari, 2015). We observed an increased cellulase
   activity with an increase in time. The CK cellulase activity was higher
   after 4 days, while that of UV-C treated bananas did not differ
   significantly till 8 days. Interestingly, from the eighth day till the
   end of the experiment (18th day), the cellulase activity of CK and UV-C
   at the same sampling time was significantly different; the activity was
   significantly higher in CK as compared to UV-C group ([53]Figure 2C).
   These observations suggest that after 8 days of UV-C treatment, the
   cellulase activity is reduced as compared to CK and it possibly
   maintains the hardness and the quality of the banana peels.

FIGURE 1.

   [54]FIGURE 1
   [55]Open in a new tab

   M. nana peel treated with UV-C (0.02 kJ/m^2) or without UV-C (CK) for
   0, 4, 8, 12, 15, and 18 days.

FIGURE 2.

   [56]FIGURE 2
   [57]Open in a new tab

   Physiological index analyses of banana peel treated with UV-C
   (0.02 kJ/m^2) or without UV-C (CK) for 0, 4, 8, 12, 15, and 18 days.
   (A) peel decay rate, (B) weight loss rate, (C) cellulase activity, (D)
   total phenols, (E) total flavonoids, (F) phenylalanine ammonia lyase
   (PAL) activity, (G) superoxidase dismutase (SOD) activity, (H)
   peroxidase (POD) activity, (I) 2, 20
   -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), (J)
   1,1-diphenyl-2-picrylhydrazyl radical scavenging capacity (DPPH), and
   (K) ferric reducing antioxidant power (FRAP). The error bars represent
   standard deviation. The ns, *, and ** show that the differences between
   UV-C and CK treatments are non-significant, significant, and highly
   significant, respectively. The blue and orange bars represent CK and
   UV-C treated banana samples.

   It is known that the total phenol content of the banana increases with
   maturity. In our experiment, we observed that the total phenol contents
   of the UV-C treated peels were higher than CK. The total phenol
   contents showed an upward accumulation content in the UV-C treated
   banana as well as CK. Overall, we observed that the phenolics contents
   didn’t differ significantly between UV-C and CK except on eighth day.
   These observations suggest that total phenols content increases with
   maturity and that the UV-C treatment has minor effect on the total
   phenols content and such an effect can be seen from day 8 ([58]Figure
   2D). On the contrary, a sharp increase in the total flavonoids content
   was observed with the increase in storage time ([59]Figure 2E). UV-C
   treatment significantly affected (increased) the total flavonoids
   content in M. nana peels except for day 12 of storage.

Effect of UV-C Treatment on the Activity of Defense and Anti-Oxidant Related
Enzymes

   Phenylalanine ammonia lyase (PAL) plays important roles in resistance
   to stresses (low temperature, drought, UV irradiation) in plants. The
   activity of PAL increased with the prolongation of the storage time.
   Considering the effect of UV-C treatment, we noticed that PAL activity
   of the treatment group was significantly higher than control group at
   all time points particularly at day 8 ([60]Figure 2F). It could be
   stated that UV-C treatment causes an increase in PAL activity and
   therefore protects banana from abiotic stresses. On the other hand, we
   also studied the effect of UV-C on the activity of superoxide dismutase
   (SOD), and peroxidase (POD), which are indispensable enzymes for
   anti-oxidation and anti-aging systems. The SOD activity of the two
   groups of banana fruits decreased briefly in the early stage of
   storage, and then increased significantly. During the storage period,
   the SOD activity of the treatment group was always higher than that of
   the control group, and the difference was extremely significant (p <
   0.01). The difference was the largest on the eighth day. The SOD
   activity of the treatment group was 4.75 times that of the control
   group. ([61]Figure 2G). The results show that UV-C treatment can
   effectively induce the increase of SOD activity of banana fruits, which
   is beneficial to the elimination of oxygen free radicals generated by
   the ripening and senescence during storage. The activity of the key
   enzyme in the antioxidant system of plants i.e., POD was increased with
   UV-C treatment however, the differences were not significant until day
   8; significant differences were observed from day 12 and later
   ([62]Figure 2H). These observations suggest that the UV-C treatment
   enhances the activity of POD and hence activates the antioxidant system
   in banana peels.

Effect of UV-C Treatment on ABTS, DPPH, and FRAP of M. nana Peels

   In this experiment, three indicators i.e., 2, 20
   -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) value, DPPH,
   and FRAP were used to evaluate the antioxidant capacity of banana peels
   before and after UV-C treatment. The ABTS clearance rate of the two
   groups of bananas continued to increase with the days of storage. The
   ABTS clearance rate of the UV-C treated peels was significantly higher
   than that of the CK at 4, 8, and 18 days ([63]Figure 2I). The DPPH
   clearance rate of the two groups of bananas continued to increase
   during the storage period, though on the day 0, the differences between
   the treated and CK group were non-significant. These observations
   suggest that DPPH as like ABTS clearance rate is affected by storage
   time as well as the UV-C treatment ([64]Figure 2J). The FRAP values
   only differed significantly between UV-C and CK groups after 8 and 18
   days, however, the FRAP value of the two groups of bananas increased
   continuously during the storage period, and that the FRAP value of the
   treatment group was always higher than that of the control group
   ([65]Figure 2K). These observations indicate a positive impact of UV-C
   treatment on the antioxidant capacity of the tested bananas.

Transcriptome Comparison of CK and UV-C

Transcriptome Sequencing

   Based on the effect of UV-C treatment on banana peel appearance,
   quality, and other physiological parameters, it was concluded that
   storage under the experimental conditions for 8 days were suitable for
   maintenance of banana quality and hardness. Therefore, we further
   processed the banana peels treated with UV-C for 8 days ([66]Figures 1,
   [67]2).

   The transcriptome sequencing of six banana peel samples resulted in
   48.14 Gb clean data with Q30 base % >90%. The clean reads ranged from
   50,539,726 to 58,994,130 reads per sample (average 53,803,030 reads)
   and the average GC content per sample was 51.61% ([68]Supplementary
   Table S1). Quantification of gene expression results for the UV-C
   treated and CK samples is shown in [69]Figure 3A; the Fragments Per
   Kilobase of Transcript per Million fragments mapped (FPKM) of UV-C
   treated banana peels was slightly higher than CK ([70]Figure 3A). The
   gene expression quantification results were highly reliable evident
   from higher Pearson Correlation Coefficient (PCC) between the
   expression of treatments and replicates ([71]Figure 3B).

FIGURE 3.

   [72]FIGURE 3
   [73]Open in a new tab

   (A) Overall Fragments Per kilobase of Transcript per Million fragments
   mapped for UV-C treated M. nana (T) and control (CK). (B) Pearson
   Correlation Coefficient between the UV-C and CK replicates. (C) Heatmap
   and hierarchical clustering of the expression of differentially
   expressed genes (DEGs) between UV-C and CK bananas. Differential gene
   expression between UV-C and CK.

Differential Gene Expression Between UV-C and CK

   The screening conditions for selection of the differentially expressed
   genes (DEGs) was log2 Fold Change ≥1 and false discovery rate (FDR) <
   0.05. The screening resulted in the identification of 425 DEGs; 94 and
   331 genes were downregulated and upregulated, respectively, in UV-C
   treated banana as compared to CK. Top-5 highly downregulated genes in
   UV-C treated banana were Protein of unknown function (DUF1191;
   Ma11_g18320), MYB44-like transcription factor (TF) Ma07_g11110,
   ethylene-responsive transcription factor ERF057 (Ma04_g24880), auxin
   efflux carrier component 3a-like (Ma07_g18160), and LOB
   domain-containing protein 18 (Ma04_g39080) ([74]Supplementary Table
   S2). The changes in the expression of these genes indicate that UV-C
   treatment affects ethylene responses, auxin transport, responses to
   abiotic stresses ([75]Jaradat et al., 2013), and possibly regulates
   cell wall loosening ([76]Lee and Kim, 2013). Top-5 genes that showed
   higher log two fold change values (higher expression in UV-C treated
   banana) were glutathione S-transferase (Ma05_g07440), isoleucyl-tRNA
   synthetase (Ma09_g10880), Wound-induced protein (Ma11_g17160),
   alcohol-forming fatty acyl-CoA reductase (Ma09_g30440), and histone H3
   (Ma05_g00960) ([77]Supplementary Table S2). These increased expressions
   indicate that UV-C treated banana has a stronger wounding response
   ([78]Yen et al., 2001), increased detoxification of toxic substances by
   their conjugation with glutathione ([79]Yen et al., 2001), higher
   protein synthesis ([80]Saga et al., 2020), higher fatty alcohol
   synthesis ([81]Metz et al., 2000), and cell cycle and development
   ([82]Otero et al., 2014).

   KEGG pathway enrichment analysis indicated that the DEGs were
   significantly enriched in taurine and hypotaurine metabolism, cutin,
   suberine and wax biosynthesis, aminoacyl-tRNA biosynthesis,
   phenylpropanoid biosynthesis, flavonoid biosynthesis, and stilbenoid,
   diarylheptanoid and gingerol biosynthesis pathways. We also found that
   the DEGs were enriched in Flavone and flavonol biosynthesis, starch and
   sucrose metabolism, plant-hormone signal transduction, MAPK signaling
   pathway-plants, plant-pathogen interaction, and carotenoid
   biosynthesis. Detailed changes in each pathway are presented below
   ([83]Supplementary Figure S1).
     * a. Differential regulation of signaling and defense related
       pathways

   Our results showed that the DEGs were enriched in two signaling related
   pathways i.e., plant-hormone signaling (17 DEGs) and MAPK signaling
   pathway (13 DEGs). Four DEGs i.e., auxin response factor (ARF), two
   jasmonate ZIM domain-containing protein (JAZ), and a protein
   brassinosteroid insensitive 1 (BRI1) had lower expression in UV-C
   treated bananas. The expression of three JAZs; two downregulated and
   one upregulated in UV-C treated bananas, indicate that Jasmonic acid
   (JA) signaling is affected by UV-C irradiation. Other genes in JA
   signaling i.e., coronatine-insensitive protein 1 (COI1), JA-amino
   synthetase (JAR1), and two MYC2 genes were also upregulated in UV-C
   treated banana peels. The UV-C irradiation on banana resulted in the
   increased expression of gibberellin (gibberellin receptor GID1 and
   DELLA protein), abscisic acid (ABA, PYR/PYL family protein and protein
   phosphatase 2C, PP2C), ethylene (ethylene insensitive protein 3, EIN3
   and ethylene-responsive transcription factor 1, ERF1), and salicylic
   acid (SA, pathogenesis-related protein 1, PR1) signaling related genes.
   Therefore, it could be suggested that the UV-C irradiation activates
   the phytohormone signaling mechanisms in banana peels. These changes
   affect fruit ripening and senescence possibly due to changes in the
   expression of EIN3 and ERF1 ([84]Lado et al., 2015). Additionally, the
   UV-C treatment impart changes in anthocyanin accumulation in fruits in
   response to differential expression of PYR/PYL family protein ([85]Gao
   et al., 2018; [86]Yang et al., 2019). The delay in softening is
   possibly due to changes in the expression of gibberellin signaling
   related genes ([87]Wongs-Aree et al., 2014). Finally, UV-C treatment
   may also help in maintaining fruit quality by activating SA related
   genes ([88]Ennab et al., 2020) ([89]Supplementary Table S3).

   The MYC2 TFs, EIN3, ERF1, PP2C, PYR/PYL, and PR1 genes were also
   significantly enriched in the MAPK signaling. Among other genes that
   showed increased expression in response to UV-C irradiation, we found
   two calmodulin (CaM4, Ma00_g02460 and Ma00_g06660), a WRKY22
   (Ma04_g08390), respiratory burst oxidase (RbohD, Ma05_g29360), LRR
   receptor-like serine/threonine-protein kinase ERECTA (Ma10_g00450), and
   transmembrane protein 222 (Ma04_g08810). Increased expression of WRKY22
   suggests that reactive oxygen species (ROS) are induced in response to
   UV-C treatment in banana peel as compared to CK ([90]Adachi et al.,
   2015). While the increased expressions of CaM4s and RbohD maintains the
   homeostasis of ROS in UV-C treated peels ([91]Taj et al., 2010;
   [92]Pathak et al., 2013) ([93]Supplementary Table S3). Thus, UV-C
   treatment protects bananas from decay.

   Nineteen DEGs were enriched in plant-pathogen interaction pathway; only
   two DEGs (3-ketoacyl-CoA synthases, Ma04_g08880 and Ma03_g01910) showed
   reduced expression after UV-C treatment as compared to control.
   However, two other genes annotated as 3-ketoacyl-CoA synthase
   (Ma02_g23450 and Ma03_g02070) were upregulated. Genes that were
   enriched in the MAPK signaling pathway were also enriched in the
   plant-pathogen interaction pathway i.e., RbohD, WRKY22, CaM4, and PR1.
   Most importantly, we noticed the UV-C treatment increased the
   expression of pattern recognition receptors (PRRs i.e., EIX44198 and
   CERK1 (chitin elicitor receptor kinase 1)), genes related to
   hypersensitive response (PBS1, serine/threonine-protein kinase PBS1,
   and HSP90, molecular chaperone HtpG), and pathogenesis-related genes
   transcriptional activator PTI5 ([94]Supplementary Table S3). These
   expression changes suggest that UV-C treatment induced a hypersensitive
   response in banana peels as compared to CK and thus protects banana
   during storage.

   Other enzymes that are parts of antioxidant and defense responses i.e.,
   PAL (Ma01_g04420 and Ma05_g03720) and POD (Ma10_g05180, Ma07_g23680,
   and Ma11_g04630) also showed increased expressions after UV-C
   treatment. These observations also confirm the activity of PAL and POD
   as presented in [95]Figure 2. Thus, UV-C treatment activates a defense
   response in banana peels.
     * b. Effect of UV-C treatment on secondary metabolites related
       pathways

   It has been reported that secondary metabolite biosynthesis increases
   when tomato is treated with UV-C irradiation ([96]Liu et al., 2018). To
   this regard, our findings that DEGs were enriched in many secondary
   metabolite related pathways is quite relevant. Seven DEGs annotated as
   cysteamine dioxygenases (enriched in taurine and hypertaurine
   biosynthesis metabolism) showed increased expression in UV-C treated
   banana peels as compared to CK. These genes are known for their role in
   enhanced plant survival under abiotic stresses ([97]Licausi et al.,
   2011) by increasing the biosynthesis of taurine ([98]Kotzamanis et al.,
   2020). Hence, it is possible that UV-C treatment affected their
   expression in order to respond to changes in O[2] levels in peels,
   since cysteamine dioxygenases have also been shown to be sensitive to
   O[2] levels ([99]White et al., 2018). Their function is directly linked
   with ERFs ([100]White et al., 2020), thus it could be suggested that
   together, ERFs and cysteamine dioxygenases are activated under the
   influence of UV-C in banana peel and protect it during storage. Among
   other pathways, we found that 13 DEGs were enriched in the
   phenylpropanoid biosynthesis pathway. Only one gene i.e., shikimate
   O-hydroxycinnamoyltransferase (Ma03_g26620) was downregulated in
   response to UV-C treatment, while all other genes showed higher
   expression in the UV-C group; three PODs, three trans-cinnamate
   4-monooxygenases, two cinnamoy-CoA reductases, and two PALs
   ([101]Figure 4). The higher expression of all these genes leads towards
   increased biosynthesis of lignins (p-hydroxyphenyl lignin, guaiacyl
   lignin, coniferin, and syringyl lignin). The activity of callose
   ([102]Figure 2C) is also in line with these observations. The three
   trans-cinnamate 4-monooxygenases were also enriched in flavonoid
   biosynthesis pathway and stilbenoid, diarylheptanoid and gingerol
   biosynthesis ([103]Figure 5; [104]Supplementary Table S3). Furthermore,
   the UV-C treatment resulted in increased expression of two chalcone
   synthases, a shikimate O-hydroxycinnamoyltransferase (also enriched in
   stilbenoid, diarylheptanoid and gingerol biosynthesis), and a flavonoid
   3′,5′-hydroxylase. All these genes are located in the upstream of
   anthocyanin biosynthesis, dihydrotricetin and dihydromyricetin
   biosynthesis. These expression changes suggest that UV-C treatment
   increases the production of these compounds which are substrates for
   production of unstable leucopelargonidin, leucocyanidin, and
   leucodelphinidin. However, we did not detect the differential
   expression of any of the genes in anthocyanin biosynthesis pathway
   ([105]Hua et al., 2013). Thus, specifically, overproduction of
   dihydrotricetin and dihydromyricetin in banana peels under UV-C
   treatment might be for different protective activities. However, we
   detected the increased expression of zeaxanthin epoxidase (Ma07_g03980)
   and (+)-abscisic acid 8′-hydroxylase (Ma07_g07210). Zeaxanthin
   epoxidase converts zeaxanthin into antheraxanthin and then to
   violaxanthin and protects plants against damage by excess light
   ([106]Hieber et al., 2000).
     * c. Effect of UV-C treatment on other important pathways

FIGURE 4.

   [107]FIGURE 4
   [108]Open in a new tab

   Differential regulation of phenylpropanoid biosynthesis pathway in M.
   nana peel under the influence of UV-C irradiation as compared to
   control. Number in the boxes represent E.C. numbers of the enzymes. Red
   and green color indicates upregulation and downregulation of genes in
   response to UV-C treatment.

FIGURE 5.

   [109]FIGURE 5
   [110]Open in a new tab

   Differential regulation of flavonoid biosynthesis pathway in banana
   peels treated with or without UV-C irradiation. Number in the boxes
   represent E.C. numbers of the enzymes. Red and green color indicates
   upregulation and downregulation of genes in response to UV-C treatment.

   The KEGG enrichment analysis also revealed that the DEGs were
   significantly enriched in cutin, suberine and wax biosynthesis and
   aminoacyl-tRNA biosynthesis. Seven DEGs were enriched in cutin,
   suberine and wax biosynthesis. A gene annotated as fatty acid
   omega-hydroxylase (Ma03_g26150) was upregulated in UV-C treated peels.
   It converts 9,10-epoxy-18-hydroxystearate into
   9,10,18-Trihydroxystrearate (an unsaturated fatty acid, UFA).
   Previously, it was shown that the proportion of UFAs in peels is
   closely related with membrane fluidity and tolerance against abiotic
   stresses. A higher proportion of UFAs is found in plants that show
   tolerance to cold ([111]Liang et al., 2020). Thus, it could be proposed
   that UV-C irradiation increases the UFA content in banana peel.
   Further, we found four DEGs (three alcohol-forming fatty acyl-CoA
   reductases and an aldehyde decarbonylase) that were upregulated in
   response to UV-C irradiation. Both genes control essential steps in the
   biosynthesis of wax ([112]Cheesbrough and Kolattukudy, 1984; [113]Metz
   et al., 2000). Higher wax production in fruit offers a better
   shelf-life; cuticle is known as a modulator of postharvest quality of
   fruits ([114]Lara et al., 2014; [115]Lara et al., 2019).

   Apart from wax and UFA biosynthesis, we found that 10 DEGs were
   enriched in aminoacyl-tRNA biosynthesis pathway ([116]Supplementary
   Table S3). Interestingly, all the genes were annotated as
   isoleucyl-tRNA synthetases. These genes are known to play essential
   roles in plants e.g., (together with dirigent proteins) maintenance of
   plant cell wall integrity ([117]Paniagua et al., 2017). Thus, it is
   possible that UV-C treatment increases the expression of isoleucyl-tRNA
   synthetases, which in turn helps banana to maintain cell wall to avoid
   being flaccid.

qRT-PCR Analysis of Selected DEGs

   The qRT-PCR analysis of 17 M. nana genes showed that the expression
   profiles were consistent with the transcriptome sequencing results in
   M. nana peels treated with UV-C as compared to CK ([118]Figure 6).
   These results validated the RNA sequencing data and confirmed its
   reliability.

FIGURE 6.

   [119]FIGURE 6
   [120]Open in a new tab

   qRT-PCR analysis of selected M. nana genes that were differentially
   expressed in UV-C treated peels (UV-C) as compared to control (CK). The
   error bars on columns represent the standard deviation.

Metabolome Analysis of UV-C Treated M. nana vs CK

   The metabolome analysis resulted in the identification of 38
   differentially accumulated metabolites (DAMs) ([121]Table 1). The DAMs
   belonged to a range of metabolite classes i.e., alkaloids, amino acids
   and derivatives, anthocyanins, diterpenoids, flavanols, flavonoids,
   flavonols, free fatty acids, glycerol ester, phenolic acids, tannin,
   and xanthone. Eight of the 38 metabolites were accumulated in lower
   quantities in UV-C treated bananas as compared to CK. These DAMs
   included two amino acids i.e., L-tryptophan and
   L-valyl-L-Phenylalanine, three free fatty acids, a glycerol ester, a
   phenolic acid (disooctyl phthalate), and a tannin (cinnamtannin A2).
   Interestingly, all alkaloids, flavonoids, and terpenoids showed
   increased accumulation in banana peels treated with UV-C irradiation
   ([122]Table 1). These observations are in accordance with the results
   of total phenols and flavonoids contents ([123]Figures 2D,E). Top five
   DAMs that were accumulated in response to UV-C treated are
   4-p-Cumaroyl-rhamnosyl-(1→6)-D-glucose, 5-Aminovaleric acid,
   6,8-dihydroxy-1,2,3-trimethoxyxanthone,
   quercetin-3-O-(2'''-p-coumaroyl) sophoroside-7-O-glucoside, and
   Hydroxyanigorufone ([124]Table 1).

TABLE 1.

   List of metabolites that were differentially accumulated in Musa nana
   peel before and after UV-C treatment.
   Compounds Index CK (ion intensity) UV-C (ion intensity) VIP (variable
   importance in projection) Log2FC
   Alkaloids
    3-Carbamyl-1-methylpyridinium (1-Methylnicotinamide) pme1738 18,232
   37,087 1.04 1.02
   Amino acids and derivatives
    5-Aminovaleric acid pme0120 9 24,860 1.67 11.43
    L-Tryptophan mws0282 2,257,100 1,128,307 1.58 -1.00
    L-Valyl-L-Phenylalanine Lmhp002001 29,604 11,240 1.47 -1.40
   Anthocyanins
    Cyanidin-3-O-(6″-O-p-Coumaroyl) glucoside Lmpp003789 8,470,067
   19,123,333 1.63 1.17
    Delphinidin-3-O-(6″-O-p-coumaroyl)glucoside Lmpp003662 12,015,000
   27,235,000 1.62 1.18
   Diterpenoids
    Cafestol pme3459 25,221 61,500 1.62 1.29
   Flavonols
    7-O-Galloyltricetiflavan Lmhn004960 20,380 57,734 1.33 1.50
    Catechin gallate mws0355 28,317 78,391 1.17 1.47
    Epigallocatechin-3-gallate mws0034 62,222 1,72,137 1.25 1.47
   Flavonoid
    Luteolin-7-O-neohesperidoside (Lonicerin) pmp001079 26,332 64,707 1.65
   1.30
    Hesperetin-8-C-glucoside-3′-O-glucoside* pmb0618 62,916 1,38,263 1.60
   1.14
   Flavonols
    6-Hydroxykaempferol-7-O-glucoside pmp001309 3,419,267 6,891,400 1.30
   1.01
    Quercetin-3-O-(6″-acetyl)galactoside Hmln002199 1979 7,946 1.48 2.01
    Quercetin-7-O-(6″-malonyl)glucoside pmp000589 15,604 36,131 1.56 1.21
    Kaempferol-3-O-glucoside-7-O-rhamnoside Lmsp004670 7,849,067
   19,585,000 1.58 1.32
    Kaempferol-3-O-neohesperidoside* Lmjp002867 7,249,867 18,705,333 1.61
   1.37
    Quercetin-3-O-apiosyl (1→2)galactoside Lmtp004044 3,513 15,564 1.01
   2.15
    Quercetin-3-O-glucoside-7-O-rhamnoside Lmsp004166 12,898,333
   26,437,667 1.62 1.04
    Quercetin-7-O-rutinoside pmb0711 10,995,400 24,707,000 1.55 1.17
    Quercetin-3-O-rutinoside (Rutin) mws0059 4,823,300 1,1,302,000 1.55
   1.23
    6-Hydroxykaempferol-3,6-O-Diglucoside pmp001310 32,312 72,397 1.56
   1.16
    Kaempferol-3-O-neohesperidoside-7-O-glucoside pmp001105 1,49,923
   3,41,013 1.63 1.19
    Quercetin-3-O-rutinoside-7-O-glucoside Lmmp002334 1,90,217 4,25,890
   1.59 1.16
    Quercetin-3-O-(2‴-p-coumaroyl)sophoroside-7-O-glucoside Lmwp004293
   8,003 62,906 1.62 2.97
    Quercetin-3-O-(6″-feruloyl)glucoside-7-O-rutinoside Lmdp004461 5,339
   12,901 1.62 1.27
   Free fatty acids
    Palmitoleic Acid mws0361 6,847 2,752 1.27 -1.32
    9-Hydroperoxy-10E,12,15Z-octadecatrienoic acid pmb2791 1,26,204 15,634
   1.33 -3.01
    13S-Hydroperoxy-6Z,9Z,11E-octadecatrienoic acid pmb2789 1,80,035
   24,968 1.30 -2.85
   Glycerol ester
    PI(18:2/0:0) Lmqn008369 53,432 25,019 1.65 -1.09
   Others
    Hydroxyanigorufone HJAP129 3,28,233 2,456,067 1.59 2.90
    3,5,7,4′-Tetrahydroxy-Coumaronochromone Hmyp002315 61,487 1,90,140
   1.64 1.63
   Phenolic acids
    Diisooctyl Phthalate Lmmp010562 4,090,667 8,67,850 1.22 -2.24
    4-p-Cumaroyl-rhamnosyl-(1→6)-D-glucose Zmhn002508 9 32,638 1.67 11.82
    Di-O-Glucosylquinic acid Zmhn000785 46,582 1,14,807 1.66 1.30
    Syringic acid-4-O-(6″-feruloyl) glucoside pmb0824 6,162 12,872 1.60
   1.06
   Tannin
    Cinnamtannin A2 pmn001649 11,418 5,410 1.55 -1.08
   Xanthone
    6,8-Dihydroxy-1,2,3-trimethoxyxanthone pmp001011 2,687 34,100 1.54
   3.67
   [125]Open in a new tab

   We found that the DAMs were significantly enriched in eighteen
   different pathways ([126]Supplementary Figure S2). Significant
   enrichment of DAMs was observed in anthocyanin biosynthesis
   (cyanidin-3-O-(6″-O-p-Coumaroyl) glucoside and
   delphinidin-3-O-(6″-O-p-coumaroyl) glucoside) and flavone and flavonol
   biosynthesis (quercetin-3-O-rutinoside (Rutin) and
   quercetin-3-O-(6″-feruloyl) glucoside-7-O-rutinoside) pathways. The
   changes in the accumulation pattern of the anthocyanin biosynthesis
   pathway related metabolites are consistent with the increased
   expression of the genes that were present upstream in this pathway
   ([127]Figure 5). Thus, the changes in the expression of phenylpropanoid
   pathway genes and accumulation of these metabolites might be a reason
   for the color differences between the UV-C and CK bananas. Similarly,
   the higher accumulation of the two metabolites in flavone and flavonol
   biosynthesis pathway is consistent with the observed higher flavonoid
   contents in UV-C treated bananas ([128]Figure 2E). The enrichment of
   DAMs in other pathways is consistent with that of DEGs’ enrichment in
   KEGG pathways ([129]Supplementary Figures S1-2; [130]Figure 7A).

FIGURE 7.

   [131]FIGURE 7
   [132]Open in a new tab

   (A) Co-joint analyses of DEGs and DAMs between UV-C treated M. nana and
   CK. (A) KEGG pathway enrichement, (B) nine-quadrant graph, and (C)
   network diagrams of DAMs and DEGs based on PCC (≥0.8).

Co-Joint Analysis of DEGs and DAMs

   The nine-quadrant graph based on the PCC between DEGs and DAMs (PCC
   ≥0.8) was generated. Within the nine-quadrant graph, we looked at the
   third and seventh quadrant since the genes and metabolites in these
   quadrant have consistent regulatory trend in response to UV-C treatment
   ([133]Figure 7B). The DEGs and DAMs in these two quadrant have positive
   correlation and hence it is possible that the DAMs are regulated by the
   DEGs within these quadrants. These genes and metabolites are presented
   in [134]Supplementary Table S4. The newtwork analysis based on PCC
   indicated that a metabolite mws0282 (L-Tryptophan) had negative
   correlation with isoleucyl-tRNA synthetases ([135]Figure 7C first
   panel). Similarly, luteolin-7-O-neohesperidoside (Lonicerin) had
   positive correlation (0.92) with flavonoid 3′,5′-hydroxylase
   (Ma11_g14250) and quercetin-3-O-rutinoside (Rutin) had positive
   correlation (0.974) with isoflavone
   7-O-glucoside-6″-O-malonyltransferase (Ma04_g31740) ([136]Figure 7C).
   These combined analyses indicate that UV-C irradiation had impacted the
   flavonoid as well as phenylpropanoid biosynthesis pathways, which is
   supported by the individual omics analyses.

Discussion

UV-C Treatment Affects M. nana Fruit Quality and Antioxidant Capacity

   Increasing attention is being paid to the effects of UV irradiation on
   the postharvest preservation of fruits and vegetables. Studies have
   shown that the UV light improves the quality and shelf life of fruits
   e.g., strawberry, grapes, mango, and pineapple [ ([137]Zhang and Jiang,
   2019) and references therein]. Our observations that the decay rate and