Abstract

   Blueberry belongs to the genus Vaccinium L. in the Ericaceae and is an
   economically important shrub that produces small berries that are rich
   in nutrients. There were differences in the appearance of blueberry
   leaves under different shade treatments. To explore the differences in
   metabolites in blueberry leaves under different shading treatments,
   nontargeted liquid chromatography–mass spectrometry (LC–MS) metabonomic
   analysis was performed. Different shade intensities resulted in
   significant differences in the contents of metabolites. A total of 6879
   known metabolites were detected, including 750 significantly
   differentially expressed metabolites, including mainly lipids and
   lipid-like molecules and phenylpropanoid and polyketide superclass
   members. Based on a Kyoto Encyclopedia of Genes and Genomes (KEGG)
   pathway analysis, the flavone and flavonol biosynthesis pathways were
   the most significantly enriched. The results of this study provide a
   reference and scientific basis for the establishment of a high-quality
   and high-yield shaded blueberry cultivation system.

   Keywords: blueberry, shading, leaf metabolite, multivariate statistical
   analysis, metabolism pathway

1. Introduction

   Light, as an important ecological factor, affects the growth and
   development of plants. Light often acts on plants in the form of
   environmental signals, causing plants to undergo different
   physiological, morphogenetic and metabolic changes to adapt to
   different light environments [[36]1,[37]2,[38]3]. Therefore, light
   conditions are often considered to be one of the environmental factors
   most closely related to plant development and secondary metabolism
   [[39]4]. Plants of the same kind have different light environment
   requirements at different growth stages, and people often adopt a
   variety of technical measures to adjust the light intensity to meet the
   needs of plant growth [[40]5]. Jeltsch et al. [[41]6] found that the
   distribution and increase in carbon with plants were significantly
   affected under the inhibition of strong light. Light intensity plays
   contrasting roles in regulating metabolite compositions in Brassica
   rapa var. parachinensis [[42]3]. The plant growth, biomass
   accumulation, pigment content and phenolic content of Cyclocarya
   paliurus were significantly affected by different light quality
   treatments [[43]2]. Shading is a commonly used technique to regulate
   light intensity and can reduce the photosynthetically active radiation
   of plants, thus affecting photosynthesis and photomorphogenesis
   [[44]7,[45]8,[46]9]. Previous studies have pointed out that plants can
   adapt to changes in the light environment by increasing their specific
   leaf area, reducing leaf thickness, increasing photosynthetic pigment
   contents and increasing the absorption of mineral elements [[47]10]. In
   addition, previous studies have shown a relationship between plant
   secondary metabolism and light intensity. The carbon/nutrient (C/N)
   balance hypothesis holds that under the condition of sufficient
   nutrients, shading can inhibit photosynthesis and decrease C/N in
   plants, resulting in a decrease in C-based phenols (including
   flavonoids) and terpenes [[48]11]. Spatiotemporal shading can regulate
   anthocyanin, proanthocyanidin and sucrose accumulation in black soybean
   seeds [[49]7]. Due to the important role of plant secondary metabolites
   in improving plant stress resistance and coordinating the relationship
   with the environment, their synthesis and accumulation are more easily
   affected by environmental conditions than those of primary metabolites
   [[50]12].

   Blueberry is an important small berry fruit tree in economic forests
   [[51]13,[52]14]. Blueberry fruit is rich in nutrients, such as phenols,
   anthocyanins, mineral elements and vitamins. Moreover, compared with
   traditional fruits, such as grapes, apples and tomatoes, blueberry
   fruits contain more antioxidants, among which anthocyanins are the most
   effective [[53]15,[54]16]. Eating these antioxidants greatly benefits
   human health and can effectively prevent a variety of diseases. At
   present, it has been confirmed that blueberries contain antioxidants,
   antihypertensives, cardiovascular protectants and other substances that
   prevent chronic diseases and perform other functions [[55]17,[56]18].
   Therefore, blueberries are listed as one of the top five health foods
   by the Food and Agriculture Organization of the United Nations (FAO)
   and are known as “golden berries” [[57]19]. With blueberry
   industrialization, the growth and accumulation of secondary metabolites
   in blueberries are the top two main concerns. While growth is the basis
   of yield formation, there are few studies on the effects of
   environmental factors, especially shade treatment, on the growth and
   secondary metabolites of blueberry.

   Metabonomics technology can be used to study the types and contents of
   metabolites in the same plant at different developmental stages or in
   different tissues and organs [[58]20,[59]21]. For example, Wu et al.
   [[60]21] analysed the kernels of Styrax tonkinensis at different
   developmental stages by metabonomics and fully elucidated the metabolic
   substances and enriched pathways related to oil synthesis during seed
   development. Bai et al. [[61]22] comprehensively investigated the
   phytochemical constituents and multiple biological activities of
   different parts of tree peony fruits harvested from Paeonia ostii and
   Paeonia rockii. Guo et al. [[62]23] used metabonomics and
   transcriptomics to analyse the similarities and differences of
   phenylpropanoid and flavonoid synthesis-related genes and metabolites
   in young and mature leaves of Ginkgo biloba and provided useful
   information for the practical application of ginkgo leaves in the
   pharmaceutical industry. The diversity of metabolites in plants makes
   them an ideal target for studying the regulatory mechanisms of
   biosynthesis [[63]24]. With the progress of science and technology,
   metabonomics has become a widely used tool in systems biology and
   molecular breeding, especially for high-throughput analysis,
   chromatography and the analysis of metabolites; the availability of
   bioinformatics tools and databases has also improved
   [[64]25,[65]26,[66]27,[67]28].

   In this study, different shading treatments were used in the
   cultivation of potted blueberry seedlings. The goal was to deeply
   understand the effects of different shading intensities on the
   metabolites of blueberry leaves to determine which appropriate shading
   intensity is conducive to the growth of blueberries and the
   accumulation of metabolites. To achieve this goal, (i) we identified
   all metabolites in blueberry leaves; (ii) the metabolic substances were
   classified, and the change trends of the main metabolic substances
   under different shading intensities were determined; and (iii) the
   differentially expressed metabolites (DEMs) were identified across
   different shading treatments, and the main DEMs were enriched by a
   Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.
   Overall, the aim of this study was to determine the optimum conditions
   for blueberry leaf growth and metabolite accumulation under different
   light intensities to provide a reference and scientific basis for the
   establishment of a high-quality and high-yield shaded blueberry
   cultivation system. In addition, this research provides strong evidence
   for optimizing the cultivation conditions and intensive management of
   blueberry in the summer.

2. Results

2.1. Phenotypic Observation and Leaf Colour Analysis

   Shading treatment was carried out using a one-layer black shading net
   (T1, 50% shading), two-layer black shading net (T2, 80% shading) and no
   shading (CK, total light). After shading for 2 months, there were some
   differences in the appearance of blueberry leaves under different
   treatments ([68]Figure 1). Compared with the leaves of the CK group,
   the leaves of the T1 group were greener ([69]Figure 1a,b), and the
   leaves of the T2 group were more yellow–green ([70]Figure 2c).
   Furthermore, to explore the difference in blueberry leaf colour among
   the CK, T1 and T2 treatment groups, the values of the colour parameters
   L* (colour brightness), a* (+red/−green degree) and b* (+yellow/−blue
   degree) were measured and analysed ([71]Figure 1d–f). The results
   showed that the L* and b* values of blueberry leaves in the three
   groups were greater than 0, indicating that their colour was brighter.
   The L* and b* values of the blueberry leaves were significantly
   different between the CK and T2 groups. The order of the L* values were
   as follows: T2 > T1 > CK ([72]Figure 1d); that of the b* values was as
   follows: CK > T1 > T2 ([73]Figure 1f). The a* values of blueberry
   leaves in the CK, T1 and T2 groups were less than 0, indicating that
   their colour was green, and there were very significant differences
   among the different groups ([74]Figure 1e).

Figure 1.

   [75]Figure 1
   [76]Open in a new tab

   Phenotypic observations (a–c) and colour parameters (d–f) of blueberry
   leaves. *, *** and **** represents a significant correlation at the
   0.05, 0.001 and 0.0001 level, respectively. The ns represents no
   significant correlation.

Figure 2.

   [77]Figure 2
   [78]Open in a new tab

   Multivariate statistical analysis of blueberry samples in positive and
   negative ion modes. Principal component analysis in positive (a) and
   negative ion modes (d); partial least-squares discriminant analysis
   score chart in positive (b) and negative ion modes (e); and loading
   plot in positive (c) and negative ion modes (f).

2.2. Multivariate Statistical Analysis

   The base peak chromatogram (BPC) is a map obtained by continuously
   depicting the intensity of the strongest ion in the mass spectrum at
   each time point. [79]Figures S1 and S2 are BPC diagrams of 9 samples in
   positive and negative ion modes, respectively. According to the
   qualitative metabolite results, there were 24,529 substance peaks in
   blueberry leaves from the CK, T1 and T2 groups (15,218 in positive ion
   mode and 9311 in negative ion mode), while the total number of known
   metabolites was 6879, including 4727 in positive ion mode and 2152 in
   negative ion mode. We used unsupervised principal component analysis
   (PCA) to observe the overall distribution among samples and the
   stability of the whole analysis process. The elliptical region
   represents the 95% confidence interval. By observing the dispersion of
   samples, it was concluded that the stability and reliability of the
   instrumental analysis were good, and the samples between different
   groups were relatively far away on the abscissa and ordinate score
   diagrams. The degree of variation was large, and there may have been
   significant differences. The variability within three biological repeat
   samples in the group was small, which showed that biological
   repeatability in the same group was good ([80]Figure 2a positive ion
   mode, and 2d negative ion mode). Supervised partial least-squares
   discriminant analysis (PLS-DA) with the addition of grouping variables
   can compensate for the deficiencies of the PCA. In positive ion mode,
   the parameter R^2X (cum) = 0.90, the interpretation rate R^2Y (cum) and
   the prediction rate Q^2 (cum) were close to 1, indicating that the
   PLS-DA model can better explain and predict the differences in samples
   among the different groups ([81]Figure 2b). In negative ion mode, the
   parameters R^2X (cum) = 0.95 and Q^2 (cum) = 0.99 indicated that the
   model established by this method was stable and reliable ([82]Figure
   2e). In addition, loading plots can be used to determine the intensity
   of the impact of metabolites in different comparisons. A load farther
   away from 0 indicates that the variable has a greater impact on the
   component, and a load closer to 0 indicates that the variable has a
   lesser impact on the component ([83]Figure 2c positive ion mode, and 2f
   negative ion mode).

2.3. DEMs Analysis

   We used the combination of multidimensional analysis and
   one-dimensional analysis to screen the DEMs between groups and further
   verified whether the DEMs between groups were significant by T test. To
   more intuitively show the relationships and the expression differences
   of metabolites between different samples, we performed hierarchical
   clustering based on the expression of all significant DEMs (750)
   ([84]Figure 3; [85]Table S1). The colours from blue to red indicate the
   expression of metabolites from low to high; that is, the redder colour
   indicates that the expression of DEMs is higher. In addition, based on
   the specific numbers of DEMs in each comparison group, the highest
   number of DEMs was found between the T2 and CK groups, at 735, while
   there were 636 DEMs between the T1 and CK groups, and 690 DEMs between
   the T2 and T1 groups ([86]Figure 4a). Interestingly, 229 DEMs were
   common to all comparison groups ([87]Figure 4b). We classified all DEMs
   according to their chemical properties and obtained 150 unclassified
   DEMs, 314 DEMs belonging to lipids and lipid-like molecules, and 122
   DEMs belonging to phenylpropanoids and polyketides ([88]Figure 4c). The
   remaining DEMs were distributed among 11 different superclasses, and
   the number of DEMs in each superclass was no more than 50. To better
   understand the classification of DEMs between different groups, we
   performed statistical analysis on the classification of DEMs in each
   comparison group. There was only one DEM in the three superclasses of
   organic nitrogen compounds, organohalogen compounds, and organosulfur
   compounds in the T1 and CK comparison ([89]Figure 4d). In the T2 and CK
   comparison, the number of DEMs in the six superclasses of alkaloids and
   derivatives, hydrocarbons, organic nitrogen compounds, organohalogen
   compounds, organosulfur compounds, lignans, neolignans and related
   compounds was less than 5 ([90]Figure 4e). Between the T2 and T1
   groups, the most abundant DEMs were lipids and lipid-like molecules
   (281), followed by unclassified molecules (133) ([91]Figure 4f). Since
   the greatest difference in the expression abundance of DEMs was found
   between the T2 and CK groups, the correlation analysis of the top 50
   DEMs in this group showed that there was a positive correlation between
   six common flavonoids (quercitrin, rutin, quercetin
   3-(6″-malonyl-glucoside), quercetin-3-galactoside, isoquercitrin, and
   kaempferol 3-(2″,4″-di-(Z)-p-coumaroylrhamnoside)), and the correlation
   coefficients were greater than 0.97 ([92]Figure 4g).

Figure 3.

   [93]Figure 3
   [94]Open in a new tab

   Vertical and horizontal cluster heatmap of significantly differentially
   expressed metabolites (DEMs) in blueberry leaves. The abscissa
   represents the sample name, and the ordinate represents the DEMs.

Figure 4.

   [95]Figure 4
   [96]Open in a new tab

   Statistical and analytical charts of differentially expressed
   metabolites (DEMs). (a) DEMs in the different comparisons. (b) The
   number of DEMs in the different comparisons. (c) Classification of all
   DEMs classification. (d–f) DEMs in different superclasses between the
   T1 and CK groups, T2 and CK groups, and T2 and T1 groups, respectively.
   (g) Correlation analysis of the top 50 DEMs.

2.4. KEGG Pathway Enrichment Analysis of DEMs

   The significant DEMs were annotated by the KEGG database, and the main
   enriched metabolic pathways of the DEMs were analysed. The results
   showed that the DEMs were enriched in 62 KEGG metabolic pathways, of
   which 41 pathways were found across all comparisons, and only 3
   pathways existed in only one comparison (ko00232 caffeine metabolism,
   ko00592 alpha linolenic acid metabolism and ko00942 anthocyanin
   biosynthesis) ([97]Figure 5a). We analysed the top 10 pathways with
   significant KEGG enrichment in the different comparisons and found that
   the flavone and flavonol biosynthesis pathways were the most
   significantly enriched ([98]Figure 5b–d). In addition, the enrichment
   of six pathways, namely, flavonoid biosynthesis, ABC transporters,
   arginine biosynthesis, aminoacyl-tRNA biosynthesis, nitrogen
   metabolism, and alanine, aspartate and glutamate metabolism, ranked in
   the top 10 in each comparison.

Figure 5.

   [99]Figure 5
   [100]Open in a new tab

   KEGG pathway enrichment analysis based on differentially expressed
   metabolites (DEMs). (a) The main enriched metabolic pathways of the
   DEMs. (b–d) The top 10 enriched KEGG pathways. The horizontal
   coordinate is the name of the pathway, the vertical coordinate is the
   −log10 (p value) and the two horizontal dashed lines represent the
   positions of p < 0.01 and p < 0.05.

2.5. Analysis of the Metabolites in Six Pathways

   There were six metabolites (L-glutamic acid, L-threonine, biliverdin,
   coproporphyrinogen III, mesobilirubinogen and O-phospho-L-threonine) in
   the porphyrin and chlorophyll metabolism pathway (ko00860), of which
   the content of L-glutamic acid was the highest, especially in the T1
   group ([101]Figure 6a). In the carotenoid biosynthesis pathway
   (ko00906), 11 metabolites were annotated. The contents of canthaxanthin
   and presqualene diphosphate were relatively high, and the contents of
   (S)-(+)-abscisic acid and lutein B/calthaxanthin/3′-epilutein were
   relatively low ([102]Figure 6b). The chlorogenic acid content was
   relatively high in the phenylpropanoid biosynthesis (ko00940) and
   flavonoid biosynthesis (ko00941) pathways ([103]Figure 6c,d). Moreover,
   a total of 6 metabolites were annotated into the anthocyanin
   biosynthesis pathway (ko00942) ([104]Figure 6e). The trends in the
   contents of petanin, malvidin-3-glucoside, peonidin-3-glucoside and
   delphinidin-3-glucoside in each group were CK > T1 > T2. In the flavone
   and flavonol biosynthesis pathway (ko00944), 13 metabolites were found,
   with isoquercitrin and rutin having the highest content; the trend for
   the content of these two metabolites in each group was CK > T1 > T2
   ([105]Figure 6f).

Figure 6.

   [106]Figure 6
   [107]Open in a new tab

   Analysis of the metabolites enriched in six important pathways. (a)
   Porphyrin and chlorophyll metabolism pathway; (b) carotenoid
   biosynthesis pathway; (c) phenylpropanoid biosynthesis; (d) flavonoid
   biosynthesis; (e) anthocyanin biosynthesis pathway; and (f) flavone and
   flavonol biosynthesis pathway.

3. Discussion

   With the expansion of blueberry planting areas, large-scale breeding of
   blueberry seedlings is important to the healthy development of the
   blueberry industry. To cultivate blueberry seedlings with strong
   growth, high survival rates and good quality, the effects of different
   shade treatments on blueberry growth and physiological metabolite
   indexes were studied to determine the shading degree suitable for
   blueberry growth in the summer (T1: 50% shading) and to provide a
   technical reference for the scientific cultivation, management and
   large-scale breeding of blueberry.

   As one of the most important environmental factors, light is
   indispensable for plant compound biosynthesis. Key factors related to
   light include the intensity (quantity), photoperiod (duration) and
   light quality (frequency or wavelength) [[108]29]. The results showed
   that shading had important effects on leaf gas exchange, leaf pigments
   and secondary metabolites in Polygonum minus Huds., an aromatic
   medicinal herb [[109]30]. Similarly, in a study on the effect of light
   intensity on the metabolite content of Centella plants, it was also
   found that a 50% shading treatment would reduce the contents of
   triterpenoids, flavonoids and chlorogenic acid in the leaves [[110]31],
   and in blueberries [[111]32] and Berberis microphylla G. Forst,
   consistent results were also found for anthocyanin synthesis [[112]33].
   However, some studies reported that moderate shading could induce the
   accumulation of plant secondary metabolites [[113]34] and increase the
   anthocyanin content [[114]35]. Light plays an important role in plant
   growth and development, and plants have different stress responses to
   light [[115]36,[116]37,[117]38]. This study found that compared with
   the control, the blueberry leaves under the T1 shading treatment were
   significantly darker and greener than those under other treatments
   ([118]Figure 1a,b). This result may have been because appropriate
   shading can increase the chlorophyll contents in the leaves [[119]39],
   which can cause the leaves to capture more photons and enhance
   photosynthesis under weak light conditions. However, the colour of
   blueberry leaves treated with excessive shading was significantly
   lighter ([120]Figure 1c), which was similar to the results of Chen et
   al.’s study [[121]40]. It is inferred that moderate shading can
   increase the chlorophyll content of blueberry leaves, but excessive
   shading will lead to the decomposition of chlorophyll. Some studies
   have also shown that shading may balance the source–sink gradient,
   thereby reducing feedback inhibition, leading to photoinhibition and
   oxidative stress [[122]41]. Moderate shading alleviates sink limitation
   by preventing excessive starch accumulation and increasing leaf sucrose
   levels [[123]41]. The relationship between plant structure and the
   environment plays an important role in exploring the stress resistance
   of plants. Most related studies have been performed on leaves. Usually,
   plant leaves are the most common organ in which to study photosynthesis
   and changes in the external environment. Under different shading
   environments, the changes in plant growth and morphological
   characteristics were the most obvious. These adaptive changes are
   conducive to improving plant survival, fitness and competitiveness.

   Metabolomics is a comprehensive analysis of the whole metabolome under
   specific experimental conditions. It ultimately defines the chemical
   phenotype of organisms at a certain time point [[124]42]. This chemical
   phenotype, the metabolome, is a collection of all metabolites of an
   organism and is the result of genomic functioning in a specific
   environment [[125]43]. Therefore, both genomic and environmental
   changes will eventually appear in the metabolome and finally result in
   gene and protein expression. Plant metabolomics aims to comprehensively
   and in an unbiased manner determine the response of metabolite
   concentrations in organisms to interference by biotic or abiotic
   factors [[126]44,[127]45]. Wu et al. [[128]21] analysed the metabolites
   of kernels at different developmental stages of Styrax tonkinensis
   through metabolomics, obtaining 187 and 1556 metabolites through GC–MS
   and LC–MS, respectively. Zou et al. [[129]46] conducted extensive
   targeted metabolomics analysis based on LC–MS/MS for loquat, and a
   total of 536 metabolites were identified. Among them, 193 metabolites
   (including 7 carbohydrates, 12 organic acids and 8 amino acids) were
   different between the different cultivars. Through enrichment analysis,
   it was also found that there were significant differences in the
   phenolic pathways between different cultivars [[130]46]. In this study,
   nontargeted LC–MS analysis was used to obtain a total of 6879 known
   metabolites, including 2152 in negative ion mode and 4727 in positive
   ion mode. There were significant differences in the number of
   metabolites between different species/tissues.

   The synthesis and accumulation of secondary metabolites in plants are
   induced by certain environmental conditions [[131]47]. As a result, the
   secondary metabolite synthesis pathway is distributed in different
   organs, tissues and cells of plants and is regulated by developmental
   processes. Sometimes the environmental conditions that result in the
   highest flavone content may not be the most suitable growth conditions
   for plants. From the perspective of plant cultivation, selecting a
   sub-suitable ecological environment with certain environmental
   pressures is sometimes used to balance the cultivation and growth of
   some medicinal plants and the accumulation of secondary products
   [[132]48]. Correlation analysis can help to measure the correlation
   between significant DEMs and further reveal the relationships between
   metabolites in the process of biological state change. Herein, we found
   that there was a positive correlation between six common flavonoids,
   and the correlation coefficients were greater than 0.97 and p value
   were less than 0.05, indicating that when one metabolite increased, the
   other metabolites with a very significant correlation also increased.
   Several studies have found that after shading, the polyphenol content
   of tea will decrease and the amino acid content will increase to reduce
   the phenol:ammonia ratio and improve the quality of tea [[133]49].
   However, the results of this study are not completely consistent. The
   changes in the contents of various polyphenols in different groups were
   not exactly the same, indicating that there may be competition for
   common substrates among various phenolic metabolites. Light energy
   beyond the acceptable range of plant light response centres can lead to
   light stress and photoinhibition [[134]50]. In addition to possibly
   affecting blueberry growth and fruit flavour, flavonol glycosides are
   effective attenuators of sunlight, so they are considered to play a key
   role in plant photoprotection [[135]51,[136]52]. In this study, after
   shading treatment, the content of most flavonoid (alcohol) glycosides
   decreased further with an increase in the degree of shading
   ([137]Figure 6e,f), which is similar to the results of Wang et al.’s
   [[138]53] study. KEGG pathway enrichment analysis showed that the
   flavone and flavonol biosynthesis pathways were the most significantly
   enriched in this study among the different comparisons. It is suggested
   that changes in this product from this pathway may be important in the
   response of blueberry leaves to different shading stresses. Several
   studies have indicated that flavonoids play a certain role in
   protecting plants from light [[139]54]. However, Liu et al.’s [[140]55]
   study showed that shading treatment reduced the leaf flavonoid content
   of Cyclocarya paliurus. Hence, different plants have different
   accumulation patterns of secondary metabolites, including flavonoids,
   under different light intensities or light qualities. The contents of
   most secondary metabolites in the anthocyanin biosynthesis pathway
   (ko00942) and flavone and flavonol biosynthesis pathway (ko00944)
   decreased with the increase in the shading degree in blueberry, similar
   to the results from Wang et al.’s [[141]53] study. This study provides
   a basis for creating shading gradients and examining the metabolite
   synthesis and accumulation of blueberry in the summer.

4. Materials and Methods

4.1. Plant Materials

   Two-year-old potted ‘Biloxi’ blueberry seedlings with uniform growth
   and no diseases or pests were selected as the experimental materials.
   The matrix ratio was 3:3:3:1 peat, coconut bran, pine bark and perlite,
   and the pH was approximately 5.5. There were three replicates per
   treatment and 3–4 seedlings per replicate. The seedlings grew in the
   test base of Nanjing Lishui White Horse Industrial Park, Institute of
   Botany, Jiangsu Province, Chinese Academy of Sciences (119°03′ E,
   31°35′ N). Shade treatments were carried out from the end of July to
   the end of September using a one-layer black shading net (Taizhou
   Jinnong Screen Factory, Zhejiang, China) (T1, 50% shading), two-layer
   black shading net (T2, 80% shading) and no shading (CK, total light).
   The light intensity changes constantly on any given day in the summer.
   Under total light, the minimum light intensity was 35,400 lx, and the
   maximum high light intensity was 88,400 lx. In most cases, the CK light
   intensity was about 66,100 lx.

4.2. Determination of Blueberry Leaf Colour

   For blueberry leaf colour measurement, photos were taken under uniform
   conditions, and the colour parameters were recorded using Adobe
   Photoshop CS6 software, in which L*, a* and b* values were directly
   measured by a colour detector. In brief, the middle, upper, lower,
   left, and right parts of each sample leaf were measured, and four
   biological replicates were performed [[142]56].

4.3. Sample Preparation

   Sample preparation was carried out according to previously published
   research in our laboratory [[143]56]. A total of nine samples were
   analysed. In brief, 80 mg per blueberry leaf was placed into 20 μL of
   internal standard (l-2-chlorophenylalanine, 0.3 mg/mL; methanol
   configuration) and 1 mL of methanol water (v/v = 7:3). Then, we added
   two small steel balls, precooled samples at −20 °C for 2 min, and
   ground them in a grinder at 60 Hz for 2 min. Ultrasonic extraction was
   performed in an ice water bath for 30 min followed by incubation
   overnight at −20 °C. Then, the samples were centrifuged for 10 min
   (13,000 rpm, 4 °C), and 150 μL of the mixture was aspirated with a
   syringe. Finally, the mixture was filtered through a 0.22 μm organic
   phase pinhole filter, transferred to an LC injection vial and stored at
   −80 °C until LC–MS analysis. All extraction reagents were precooled at
   −20 °C before use.

4.4. Sample Processing

   A Dionex Ultimate 3000 RS UHPLC system fitted with a Q-Exactive
   quadrupole-Orbitrap mass spectrometer equipped with a heated
   electrospray ionization (ESI) source (Thermo Fisher Scientific,
   Waltham, MA, USA) was used to analyse the metabolic profiles in both
   ESI-positive and ESI-negative ion modes. An ACQUITY UPLC HSS T3 column
   (1.8 μm, 2.1 × 100 mm) was employed. The binary gradient separation
   system consisted of (a) water (containing 0.1% formic acid, v/v) and
   (b) acetonitrile (containing 0.1% formic acid, v/v). The following
   gradients were used for separation: 0 min, 5% B; 2 min, 5% B; 4 min,
   30% B; 8 min, 50% B; 18 min, 80% B; 14 min, 100% B; 15 min, 100% B;
   15.1 min, 5% B; and 16 min, 5% B. The flow rate was 0.35 mL/min, and
   the column temperature was 45 °C. For other sample processing-specific
   details and parameters, refer to He et al. [[144]57] and Wang et al.
   [[145]58].

4.5. Data Preprocessing

   Data preprocessing was conducted according to Wang et al. [[146]58]. In
   brief, the raw LC–MS data were identified and analysed by Progenesis QI
   v2.3 software (Nonlinear Dynamics, Newcastle, UK). The precursor
   tolerance was set at 5 ppm, fragment tolerance was set at 10 ppm,
   product ion threshold was set at 5% and retention time (RT) tolerance
   was set at 0.02 min. For statistical analysis of the specific data,
   please refer to Liu et al. [[147]59]. An Excel file was used for
   three-dimensional data sets, including m/z, peak RT and peak
   intensities, and RT–m/z pairs were used as the identifier for each ion.

4.6. Statistical Analysis

   The PCA and (orthogonal)PLS-DA were carried out to visualize the
   metabolic alterations among experimental groups after mean centring and
   Pareto variance scaling, respectively. Variable importance in the
   projection (VIP) ranks the overall contribution of each variable to the
   OPLS-DA model, and those variables with VIP > 1 are considered relevant
   for group discrimination. The DEMs were selected on the basis of the
   combination of a statistically significant threshold of VIP values
   obtained from the OPLS-DA model and p values from a two-tailed
   Student’s t test on the normalized peak areas, where metabolites with
   VIP values larger than 1 and p values less than 0.05 were considered
   DEMs. Moreover, based on the KEGG database, metabolic pathway
   enrichment analysis of DEMs was performed. A hypergeometric test was
   used to find the pathway entries significantly enriched in the
   significantly DEMs compared with the overall background [[148]60].

5. Conclusions

   In conclusion, moderate shading is conducive to the growth and
   metabolism of blueberry, but transitional shading will strongly inhibit
   the accumulation of metabolites in blueberry leaves. Among the top 10
   pathways with significant KEGG enrichment across different comparisons,
   it was found that enrichment of the flavone and flavonol biosynthesis
   pathway was the most significant. This study showed that blueberry
   growth and metabolite synthesis and accumulation were optimal under a
   shading net that allowed, 50% of the summer light. The results provide
   a reference for the cultivation conditions to obtain high-quality and
   high-yield blueberry in the summer.

Supplementary Materials

   The following are available online at
   [149]https://www.mdpi.com/article/10.3390/metabo12010079/s1, Figure S1:
   The base peak chromatogram diagrams of 9 samples in positive ion mode,
   Figure S2: The base peak chromatogram diagrams of 9 samples in negative
   mode, Table S1: The significantly differentially expressed metabolites
   of blueberry leaves.
   [150]Click here for additional data file.^ (2.3MB, zip)

Author Contributions

   Y.W. and C.Z. designed the study and performed the experiments; Y.W.,
   L.L. performed formal analysis; H.Y. and Z.H. analysed the data; Y.W.
   wrote the manuscript; W.W. provided the plants; W.L. and W.W. revised
   the manuscript. All authors have read and agreed to the published
   version of the manuscript.

Funding

   The research was supported by the Science and Technology Planning
   Project of Jiangsu Province (BE2019399, BE2020344), Jiangsu Agriculture
   Science and Technology Innovation Fund (JASTIF)(CX(21)3172), earmarked
   fund for Jiangsu Agricultural Industry Technology System
   (JATS[2021]511), Central Finance Forestry Technology Promotion and
   Demonstration Project (SU[2021]TG08), and Jiangsu Institute of Botany
   Talent Fund (JIBTF202105).

Institutional Review Board Statement

   Not applicable.

Informed Consent Statement

   Not applicable.

Data Availability Statement

   The data and materials supporting the conclusions of this study are
   included within the article.

Conflicts of Interest

   The authors declare no conflict of interest.

Footnotes

   Publisher’s Note: MDPI stays neutral with regard to jurisdictional
   claims in published maps and institutional affiliations.

References