Characterization and discrimination of two varieties of eggplants using multi-element and metabolomics profiles with chemometrics

doi:10.21203/rs.3.rs-1624357/v1

Download PDF

Research Article

Characterization and discrimination of two varieties of eggplants using multi-element and metabolomics profiles with chemometrics

https://doi.org/10.21203/rs.3.rs-1624357/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The main purpose of the present study was to clarify the differences present in the multi-element and metabolite profiles between two varieties of eggplants. A total of 54 elements were identified by inductively coupled plasma mass spectrometry, 16 elements were significantly different between peel (n = 3) and flesh (n = 3). Besides, untargeted metabolomics combined with chemometric approach based on UHPLC-Q-TOF/ MS was used to identify the peel (n = 4) and flesh (n = 4).178 metabolites were screened with criteria of p < 0.05, fold change > 1.5 or < 0.67 and variable importance > 1 for the PLS-DA model prediction score. Amino acid compounds, maltitol and Kaempferol 3-O-rutinoside are the most important identification components of the two varieties of eggplant. Results showed that the two varieties of eggplants could be distinguished based on their multi-element and metabolite profiles, which may provide new directions for eggplant function research.

Eggplants

Discrimination

Multi-element

Metabolomics

Chemometrics

Eggplants (Solanum melongena L.) belong to the Solanaceae family. This is a vegetable crop that grows in tropical and subtropical regions, such as in Asia, Africa, Europe and the Near East (Nergiz et al., 2018). And Asia is the largest producer. Half of the global production comes from Asia, most of which is distributed in two countries, China and India (Agregán et al., 2021; Niño-Medina et al., 2017). The scarlet eggplant(S. aethiopicum L.) and the gboma eggplants (S. macrocarpon L.) are mainly grown in Africa (Taher et al., 2017; Capello et al., 2019).

Eggplant has gained attention and is widely grown as a high-yielding and inexpensive crop (Karimi et al., 2021; Arivalagan et al., 2012). There are many varieties of eggplants, they have different forms, sizes and colors (Al-Gburi et al., 2021; Wu et al., 2013). Eggplant contains polyphenols, such as flavonoids. It also contains minerals, as well as small amounts of vitamins, nitrogen and sulphur compounds (Braga et al., 2016; Dong et al., 2020). Previous studies have shown that the content of phenolic acids in eggplant pulp is relatively high, and appropriate techniques should be used to extract phenolic acids, anthocyanins and other substances from eggplant (Niño-Medina et al., 2017; Agregán et al., 2021). Anthocyanins are mainly extracted from the peel of eggplant, and there have been studies on the extraction of pectin from the peel of eggplant (Karimi et al., 2021; Braga et al., 2016; Kazemi et al., 2019; Ferarsa et al., 2018). In addition, studies on the content of vitamin P in different varieties of eggplants have been reported. The results showed that the content of vitamin P in fresh fruit flesh and peel was different (Dong et al., 2020). We can find that recent studies have detailed some beneficial phytochemicals in eggplant, such as phenolic compounds, carotenoids and so on (Nergiz et al., 2018; Luthria et al., 2010). The differences between different varieties of eggplant may be due to factors such as living environment and soil, but there are not many detailed researches on composition differences. This piqued our interest.

At present, metabolomics is a hot topic of research, but there are few studies on the comparison of composition differences among eggplant varieties by using high resolution non-targeted UHPLC-Q-TOF/MS metabolomics method (Mihailova et al., 2021). The present study was conducted to establish chemical element and metabolite profiles and to identify potential chemical element and metabolite markers for the isolation of two varieties of eggplants. Multi-element profiles were obtained by simultaneous quantification of 54 chemical elements in a single inductively coupled plasma mass spectrometry (ICP-MS) run, while a complete metabolomic snapshot of the eggplant was obtained by using a UHPLC-Q-TOF/MS-based metabolomics approach.

2.1. Sample collection and preparation

long and slender in shape eggplant (S. melongena var. serpentinum. MVS) and egg-shaped eggplant (S. melongena var. esculentum. MVE) (Supplementary Fig. 1) were purchased directly from the supermarket (Baoding, China). The samples were stored at -20°C in the freezer until processed.

Prior to multi-element analysis, the peel and flesh of eggplant are dried at 60°C. Then, the samples were grinded into powder, passed through a No. 200 mesh sieve, packed in sealed bags and stored at room temperature for further use.

Prior to metabolomics analysis, separate treatment of MVS peel (MVSP), MVS flesh (MVSF), MVE peel (MVEP) and MVE flesh (MVEF). In other words, the samples were slowly thawed at 4℃, and added an appropriate amount of sample to the pre-chilled solution. The solution consists of methanol, acetonitrile and water in a 2:2:1 volume ratio. And then they are mixed by vortex. The mixture is ultrasonic at low temperature for 30 minutes, allowed to stand at -20℃ for 10 minutes. The mixture was centrifuged for 15 min (14000g, 4°C). Then we took the supernatant for vacuum drying. For analysis, added 100 µL acetonitrile/water (1:1, v/v) were added for re-dissolution, vortex and centrifuged for 15 min (14000g, 4°C), analyzed supernatant.

“QC” (quality control) samples are made by mixing equal amounts of the samples to be tested, and the QC samples are processed in the same way as the samples to be tested.

2.2. Multi-element analysis by inductively coupled plasma mass spectrometry

Two kinds of eggplant peel and pulp powder was accurate weighted. Each sample requires 250 mg. Microwave digestion and multiple calibration standard were performed according to the reported method (Mi et al., 2020). Quantification of chemical elements using inductively coupled plasma mass spectrometry (ICP-MS 7700X, Agilent Technologies Inc. Palo Alto, California, USA). ICP-MS operation with forward power set to 1280 w, Its spray chamber temperature and sampling depth are 2°C and 8 mm respectively; coolant gas flow rate, 1.47 L/min; carrier gas flow rate, 1 L/min; makeup gas flow rate, 1 L/min. The precision of the ICP-MS method was evaluated by using internal standard elements (72Ge, 115In and 209bi) with relative standard deviations generally within 10%. This ensures that it is the variation in elemental concentrations between the analyzed samples that causes the standard deviation of the experimental mean, and not the analytical method. To monitor the measurement reproducibility of the entire ICP-MS analysis, regent blanks (5% HNO₃) and QC samples need to be injected every 3 samples.

2.3. Metabolomics analysis by UHPLC-Q-TOF/MS

Metabolic profiling was performed with reference to the reports already available (Benton et al., 2014; Becerra-Martinez et al., 2017). Analyses were performed using an UHPLC (1290 Infinity LC, Agilent Technologies) coupled to a quadrupole time-of-flight (AB Sciex TripleTOF 6600). For HILIC separation, samples were analyzed using a 2.1 mm × 100 mm ACQUIY UPLC BEH 1.7 µm column (waters, Ireland) (Wu et al., 2013; Mi et al., 2020). Its column temperature is set at 25°C; flow rate 0.5 mL/min; injection volume 2 µL; The mobile phase A was composed of water, 25 mM ammonium acetate and 25 mM ammonia;B: acetonitrile. The gradient was 95% B for 0.5 min and was linearly reduced to 65% in 6.5 min, and then was reduced to 40% in 1 min, and kept 1 min, and then increased to 95% in 0.1 min, with a 2.9 min re-equilibration period employed. The sample was placed in a 4°C autosampler throughout the analysis. To avoid the effect of fluctuations in the instrument detection signal, a random order was used for continuous analysis of the samples. QC samples were inserted in the sample queue to monitor and evaluate the stability of the system and the reliability of the experimental data.

2.4. Data processing and statistics

Raw data in Wiff format was converted to the.mzXML format by ProteoWizard and then XCMS software was used for peak alignment, retention time correction and extraction of peak areas.

The data was analyzed by Addinsoft XLSTAT-Premium v2019.4.1. Data are statistically significant at p ≤ 0.05. Data mining using principal component discriminant analysis and partial least squares discriminant analysis to classify samples. Principal component analysis (PCA) was used to explore data. Partial least square discriminant analysis (PLS-DA) was used to classify samples.

3.1. Chemical elemental analysis of peel and flesh of two varieties of eggplants

A total of 54 chemical elements of eggplant peel and flesh were analyzed. Fifty-one and 50 chemical elements were detected in the peel and flesh of two varieties of eggplants, respectively. The most abundant elements in MVS were K, Ca, Mg, Na, B, Rb, Sr, and K, Ca, Mg, Na, B is rich in MVE (Table 1). However, we found that the mineral concentrations obtained this time were different from those previously reported (Nergiz et al., 2018; Ahmad et al., 2021). This may be related to eggplant varieties and growing environment (Arivalagan et al., 2012). In MVSP, the concentration of Cu, Fe, Zn were 4.30, 31.51, 33.35 µg/g, and in MVSF, their contents were 3.07, 21.16, 19.07 µg/g. In MVEP, the concentration of Cu, Fe, Zn were 3.82, 64.82, 37.87 µg/g, and in MVEF, their contents were 1.85, 10.90, 16.26 µg/g. This is partially consistent with the previous report that eggplant contains not only abundant phenols, but also a certain amount of trace elements (Arivalagan et al., 2012; Rosa-Martínez et al., 2021; Al-Gburi et al., 2021; Raigón et al., 2008). The results of this study show that the content of these minerals is mainly concentrated in eggplant peel. Among them, Fe was mainly present in MVEP, while Cu and Zn were not much different in the peel of both kinds of eggplant.

At the same time, the potential toxic elements As, Cr, Cd and Pb were detected in two kinds of eggplant. In MVSP and MVEP, the contents of arsenic were 0.067, 0.091 µg/g, the chromium contents were 0.58, 0.69 µg/g, the cadmium contents were 0.016, 0.014 µg/g, and the lead contents were 0.086, 0.22 µg/g. By comparison, except for arsenic and cadmium, the concentrations of the other two were above the permissible limits of 0.5, 0.5, 0.1 and 0.1 mg/kg, respectively, set by Standardization Adminis-tration of the People's Republic of China (Mi et al., 2020; Al-Gburi et al., 2021). The contents of these four elements in the pulp are all lower than the specified value.

3.2. Multi-element comparison of two varieties of eggplants

Univariate and multivariate data analyses were used to visualize differences in the chemical elemental composition between peel (n = 3) and flesh (n = 3) of MVS and MVE. Statistical analysis (p ≤ 0.05) was performed for 16 mineral elements of peel and flesh of eggplants (Table 1). Then, unsupervised principal component analysis (PCA) was performed on the data containing 51 chemical elements. Figure 1a shows the PCA score plot with a clear separation between the peel and flesh of the two eggplants. The first two main constituents (PC) accounted for 76.49% of the total variance, with PC1 at 58.02% and PC2 at 18.47%. The data containing 51 chemical elements were subsequently analyzed using PLS-DA, and Fig. 1b depicts the PLS-DA score plot for the multi-element data set. The peel and flesh of the two eggplant species were clearly divided into four groups, with R²X = 0.94, R²Y = 0.99, Q² = 0.96. R² represents the degree of fit of the model to the data, while Q² indicates the predictive power of the model for the new data set (Mi et al., 2020).

The most important distinguishing factor between groups was selected according to the variable importance criterion of projection (VIP) score > 1 and p ≤ 0.05 (Arivalagan et al., 2012). The results showed (Table 1) that 9 of the 16 elements had significantly higher concentrations in MVSP than in MVEP. These elements included B, Ca, Cu, Rb, Sr, Pd, Cd, Ba, Tl. A close correlation between these elements and MVSP was confirmed by correlation (Fig. 1c). Five of these elements are closely distributed around the MVEP. These elements are Cs, K, As, Mg, Na(Table 1 and Fig. 1c).

Seven elements were significantly more concentrated in MVSP than in MVSF. These elements were B, K, Cu, Rb, Pd, Cs, Ba. There were 13 elements with significantly higher concentrations in MVEP than in MVEF, including B, Na, Mg, K, Ca, Cu, As, Rb, Sr, Cd, Cs, Ba, Tl. There were 13 elements with higher concentrations in MVSF than in MVEF, namely B, Mg, Ca, Cu, As, Rb, Sr, Pd, Cd, Sn, Cs, Ba, Tl. As can be seen from Fig. 1c, these significant differences chemical elements are mainly distributed in the eggplant peel.

To sum up, there are certain differences in the element spectrum between the two kinds of eggplant, which may be caused by planting conditions and varieties. Although we found differences in chemical elements between the two kinds of eggplant in this study, there may be some limitations due to the small number of samples, so the sample size can be expanded in subsequent studies to continue verification.

3.3. Metabolomic comparison of two varieties of eggplant

3.3.1. Comparison of metabolite composition of two varieties of eggplant peels

Principal component analysis was used to analyze dataset of metabolites detected in the peel of two varieties of eggplant in ESI + and ESI- modes. The PCA score plots shows an obvious separation between MVSP (n = 4) and MVEP (n = 4) in both ESI+ (Fig. 2a) and ESI- (Fig. 2b) modes, which indicates that there is a large difference in the peel of eggplant between the two varieties. The first two principal components (PC) explain 57.4% and 57.1% of the total variance for ESI + and ESI- modes, respectively.

To further determine the metabolites that distinguish between the two varieties of eggplant peels, we performed a supervised PLS-DA on a dataset of these metabolites. Figure 3a and b show the PLS-DA scores of eggplant peel in ESI + and ESI- modes. It can be seen from the graph that the MVSP (n = 4) and the MVEP (n = 4) have a better separation on the graph. We evaluate the overall goodness of fit and predictive power of the PLS-DA models by R²X and Q² values. In the present study, the R²X and Q² of two varieties of eggplant peels in the ESI + mode were 0.55 and 0.93, respectively. R²X and Q² in ESI- mode are 0.56 and 0.92, respectively. These results indicate that the PLS-DA models developed have good predictive power (Q² > 0.50) (Mi et al., 2020; Mihailova et al., 2021) to identify new data set into groups. Then the VIP scores of metabolites were calculated according to the PLS-DA model, and significant metabolites were selected according to the criteria of VIP score > 1 (Mi et al., 2020; Becerra-Martinez et al., 2017; Wu et al., 2017).

Binary comparison of the overall metabolome of two varieties of eggplant peels using volcano plots to represent the differences in individual metabolites between the groups (Fig. 4a for ESI + and Fig. 4b for ESI-). This led to further screening of metabolites mainly used to distinguish two varieties of eggplant peels with the criteria of VIP > 1, FC > 1.5 or < 0.67 and p value < 0.05. A total of 62 significantly differential compounds were identified (Table 2), suggesting that there are large metabolomic differences between two varieties of eggplant peels. In the comparison of MVSP and MVEP, the number of up- regulated and down-regulated metabolites was found to be 52 and 10, respectively (Table 2).

The top 10 most up-regulated metabolites in MVSP were maltitol, L-rhamnose, 11(Z),14(Z)-eicosadienoic acid, chlorpromazine, (+)-abscisic acid, argininosuccinic acid, phenylacetic acid, trans-cinnamate, N-acetyl-D-lactosamine, L-tryptophan. Among them, the levels of maltitol and l-rhamnose in MVSP were 36.58 and 35.77 times higher than those in MVEP (Table 2). Grembecka et al. (2014) used high pressure liquid chromatography coupled to Corona charged aerosol detector (CAD) to determine maltitol in food. Maltitol is a ditangitol that has a solubility similar to sucrose (Bensouissi et al., 2010; Sun et al., 2015). So it is often used as a substitute for sucrose (Ding et al., 2019). This is probably why eggplant is a little bit sweeter when eaten. Studies have reported maltitol as a sugar substitute, for example, as a sweetener in sugar-free products, with the aim of providing sweetness and swelling (Joshi et al., 2016; Ding et al., 2019). Most of the research that has been done on maltitol has focused on its own function, that is, adding maltitol to some foods as an additive and then measuring its function (Joshi et al., 2016; Son et al., 2018).

3.3.2. Comparison of metabolite composition of MVSP and MVSF

Principal component analysis was used to analyze dataset of metabolites detected in MVSP and MVSF in ESI + and ESI- modes. The PCA score plots shows an obvious separation between MVSP (n = 4) and MVSF (n = 4) in both ESI+ (Fig. 2c) and ESI- (Fig. 2d) modes, which indicates that there is a large difference in MVSP and MVSF. The first two principal components (PC) explain 61.5% and 67.2% of the total variance for ESI + and ESI- modes, respectively.

To further determine the metabolites that distinguish between MVSP and MVSF, we performed a supervised PLS-DA on a dataset of these metabolites. Figure 3c and d show the PLS-DA scores of eggplant peel and flesh in ESI + and ESI- modes. It can be seen from the graph that the MVSP (n = 4) and MVSF (n = 4) have a better separation on the graph. We evaluate the overall goodness of fit and predictive power of the PLS-DA models by R²X and Q² values. In the present study, the R²X and Q² of MVSP and MVSF in the ESI + mode were 0.59 and 0.96, respectively. R²X and Q² in ESI- mode are 0.57 and 0.97, respectively. These results indicate that the PLS-DA models developed have good predictive power (Q² > 0.50) to identify new data set into groups. Then the VIP scores of metabolites were calculated according to the PLS-DA model, and significant metabolites were selected according to the criteria of VIP score > 1 (Wu et al., 2017).

Binary comparison of the overall metabolome of peel and flesh of eggplant using volcano plots to represent the differences in individual metabolites between the groups (Fig. 4c for ESI + and Fig. 4d for ESI-). This led to further screening of metabolites mainly used to distinguish peel and flesh with the criteria of VIP > 1, FC > 1.5 or < 0.67 and p value < 0.05. A total of 47 significantly differential compounds were identified (Table 3), suggesting that there are large metabolomic differences between peel and flesh. In the comparison of peel and flesh, the number of up- regulated and down-regulated metabolites was found to be 39 and 8, respectively (Table 3).

The top 10 most up-regulated metabolites in MVSP were maltitol, kaempferol 3-O-rutinoside, trans-3-Coumaric acid, L-rhamnose, tyramine, quinic acid, 11(Z),14(Z)-eicosadienoic acid, chlorpromazine, 2-methylbutyroylcarnitine, phenylacetic acid. When comparing the two, the levels of maltitol and kaempferol 3-O-rutinoside in MVSP were 131.6 and 116.70 times higher than those in MVSF (Table 3). Except maltitol and kaempferol 3-O-rutinoside, others are organic acids, tyramine and so on. Maltitol has been mentioned in the previous comparison of the two kinds of eggplant peels and will not be explained too much here. Kaempferol 3-O-rutinoside is a flavonoid that is often found in vegetables and fruits (Hu et al., 2020; Joshi et al., 2016). Previous studies have shown that kaempferol 3-O-rutinoside has protective effects on cerebral ischemia and dementia with multiple infarcts. In addition, kaempferol has protective effects on the liver, so it may be a natural source of protection from oxidative damage to the liver (Wang et al., 2015). Studies using animal experiments have proved that kaempferol 3-O-rutinoside has a certain anti-inflammatory effect (Hu et al., 2020; Li et al., 2019). Kaempferol 3-O-rutinoside is mainly found in the peel of eggplant, so the peel of eggplant also has some uses and can be recycled.

3.3.3. Comparison of metabolite composition of MVEP and MVEF

Principal component analysis was used to analyze dataset of metabolites detected in MVEP and MVEF in ESI + and ESI- modes. The PCA score plots shows an obvious separation between MVEP (n = 4) and MVEF (n = 4) in both ESI+ (Fig. 2e) and ESI- (Fig. 2f) modes, which indicates that there is a large difference MVEP and MVEF. The first two principal components (PC) explain 56.8% and 65.1% of the total variance for ESI + and ESI- modes, respectively.

To further determine the metabolites that distinguish between MVEP and MVEF, we performed a supervised PLS-DA on a dataset of these metabolites. Figure 3e and f show the PLS-DA scores of eggplant peel and flesh in ESI + and ESI- modes. It can be seen from the graph that the MVEP (n = 4) and MVEF (n = 4) have a better separation on the graph. We evaluate the overall goodness of fit and predictive power of the PLS-DA models by R²X and Q² values. In the present study, the R²X and Q² of MVEP and MVEF in the ESI + mode were 0.56 and 0.95, respectively. R²X and Q² in ESI- mode are 0.60 and 0.94, respectively. These results indicate that the PLS-DA models developed have good predictive power (Q² > 0.50) to identify new data set into groups. Then the VIP scores of metabolites were calculated according to the PLS-DA model, and significant metabolites were selected according to the criteria of VIP score > 1 (Wu et al., 2017).

Binary comparison of the overall metabolome of MVEP and MVEF using volcano plots to represent the differences in individual metabolites between the groups (Fig. 4e for ESI + and Fig. 4f for ESI-). This led to further screening of metabolites mainly used to distinguish peel and flesh with the criteria of VIP > 1, FC > 1.5 or < 0.67 and p value < 0.05. A total of 41 significantly differential compounds were identified (Table 4), suggesting that there are large metabolomic differences between peel and flesh. In the comparison of peel and flesh, the number of up- regulated and down-regulated metabolites was found to be 30 and 11, respectively (Table 4).

The top 10 most up-regulated metabolites in MVEP were kaempferol 3-O-rutinoside, tyramine, quinic acid, 5-Methylcytosine, 3,4-Dihydroxymandelic acid, cyanidin 3-glucoside cation, deoxycytidine, isoscopoletin, thymidine, thymine. In MVE, the levels of kaempferol 3-O-rutinoside and tyramine were higher in the peel than in the flesh, with kaempferol 3-O-rutinoside being the highest (Table 4).

3.3.4. Comparison of metabolite composition of two varieties of eggplant flesh

Principal component analysis was used to analyze dataset of metabolites detected in the two varieties of eggplant flesh in ESI + and ESI- modes. The PCA score plots shows an obvious separation between MVSF (n = 4) and MVEF (n = 4) in both ESI+ (Fig. 2g) and ESI- (Fig. 2h) modes, which indicates that there is a large difference in the flesh of two varieties of eggplant. The first two principal components (PC) explain 58.2% and 63.1% of the total variance for ESI + and ESI- modes, respectively.

To further determine the metabolites that distinguish between MVSF and MVEF, we performed a supervised PLS-DA on a dataset of these metabolites. Figure 3g and h show the PLS-DA scores of two varieties of flesh in ESI + and ESI- modes. It can be seen from the graph that the MVSF (n = 4) and MVEF (n = 4) have a better separation on the graph. We evaluate the overall goodness of fit and predictive power of the PLS-DA models by R²X and Q² values. In the present study, the R²X and Q² of two varieties of eggplant flesh in the ESI + mode were 0.57 and 0.93, respectively. R²X and Q² in ESI- mode are 0.69 and 0.96 respectively. These results indicate that the PLS-DA models developed have good predictive power (Q² > 0.50) to identify new data set into groups. Then the VIP scores of metabolites were calculated according to the PLS-DA model, and significant metabolites were selected according to the criteria of VIP score > 1 (Wu et al., 2017).

Binary comparison of the overall metabolome of two varieties of eggplant flesh using volcano plots to represent the differences in individual metabolites between the groups (Fig. 4g for ESI + and Fig. 4h for ESI-). This led to further screening of metabolites mainly used to distinguish two varieties of eggplant flesh with the criteria of VIP > 1, FC > 1.5 or < 0.67 and p value < 0.05. A total of 28 significantly differential compounds were identified (Table 5), suggesting that there are large metabolomic differences between two varieties of eggplant flesh. In the comparison of MVSF and MVEF, the number of up- regulated and down-regulated metabolites was found to be 6 and 22, respectively(Table 5).There are only six metabolites that are regulated on the MVSF, including D-proline, L-tryptophan, N-acetyl-D-lactosamine, N-acetylputrescine, D-aspartic acid, D-ribose.

To the best of our knowledge, previous studies on eggplant metabolites have been published (Nergiz et al., 2018). But the primary metabolites such as amino acids are not described. Amino acids are widely used as nutritional supplements in commercial products (Chen et al., 2021). It also plays an important role in the immune response (Ikeda et al., 2021). This not only supplemented the research on amino acids in eggplant, but also provided directions for the follow-up nutrition research.

In this study, an histological approach was used in studying the multi-element and metabolite composition of two eggplant varieties. Multivariate analysis showed a clear separation of the two eggplant species. Specifically, the maltitol content of MVSP was significantly higher when compared to MVEP, as well as when compared to MVSF. In contrast, when comparing MVSF with MVEF, the D-proline content was significantly higher in MVSF, while the nicotinamide and sucrose content were significantly lower. And in this study, these significantly altered metabolites provide some basis for an in-depth study of the bioactive components of eggplant. The results indicate that multi-elemental analysis and metabolomic approaches combined with chemometric techniques are a very effective tool for the identification of candidate chemical markers of MVSP and MVEF.

Data Availability

Data sets generated and/or analyzed in the current study can be obtained from corresponding authors upon reasonable request.

Conflict of interest

All authors of this work have no relevant conflicts of interest to disclose.

Acknowledgments

Thanks to Hebei Agricultural University for the help of the experiment.

Funding

This study was financially supported by the Industrial Innovation Team Program of Vegetables of Modern Agricultural Industry Technology System of Hebei Province [grant number HBCT2018030208]. Food Processing Discipline Group of Hebei Agricultural University [grant number 2021-09].

Author's contribution statement

Xuemei Niu: Writing – original draft, Writing – review & editing. Si Mi: Writing – review & editing. Qiuxia Jin: Writing – original draft. Yaxin Sang: Investigation, Supervision. Xianghong Wang: Supervision, Project administration, Funding acquisition.

Author information

Xuemei Niu: [email protected]

Si Mi: [email protected]

Qiuxia Jin: [email protected]

Yaxin Sang: [email protected]

*Xianghong Wang: [email protected]

Agreg n, R., Munekata, E. S., Xi, F., Astray, G., Gull n, B., & Lorenzo, M. (2021). Recent advances in the extraction of polyphenols from eggplant and their application in foods. LWT, 146, 111381.
Al-Gburi, B. (2021). Effect of different control applications on cuscuta campestris, and biochemical content of eggplant. Journal of the Saudi Society of Agricultural Sciences, 20, 209-216.
Arivalagan, M., Gangopadhyay, K. K., Kumar, G., Bhardwaj, R., Prasad, T. V., Sarkar, S. K., & Roy, A. (2012). Variability in mineral composition of indian eggplant (solanum melongena l.) genotypes. Journal of Food Composition & Analysis, 26, 173-176.
Ahmad, I., Rawoof, A., Dubey, M., & Ramchiary, N. (2021). Icp-ms based analysis of mineral elements composition during fruit development in capsicum germplasm. Journal of Food Composition and Analysis, 101, 103977.
Becerra-Martinez, E., Florentino-Ramos, E., Perez-Hernandez, N., Gerardo Zepeda-Vallejo, L., Villa-Ruano, N., & Velazquez-Ponce, M., et al. (2017). ¹H NMR-based metabolomic fingerprinting to determine metabolite levels in serrano peppers ( capsicum annum l.) grown in two different regions. Food Research International, 102, 163.
Braga, P. C., Scalzo, R. L., Dal Sasso, M. D., Lattuada, D., Greco, V., & Fibiani, M., (2016). Characterization and antioxidant activity of semi-purified extracts and pure delphinidin-glycosides from eggplant peel (solanum melongena l.). Journal of Functional Foods, 20, 411-421.
Benton, H. P., Ivanisevic, J., Mahieu, N. G., Kurczy, M. E., Johnson, C. H., Franco, L., Rinehart, D., Valentine, E., Gowda, H., Ubhi, B. K., Tautenhahn, R., Gieschen, A., Fields, M. W., Patti, G. J., & Siuzdak, G. (2014). Autonomous metabolomics for rapidmetabolite identification in global profiling. Analytical Chemistry, 87, 884–891.
Bensouissi, A., Roge, B., & Mathlouthi, M. (2010). Effect of conformation and water interactions of sucrose, maltitol, mannitol and xylitol on their metastable zone width and ease of nucleation. Food Chemistry, 122, 443-446.
Chen Q. C., Xue , C. Y., Chen, J., He, Z. Y., Wang, Z. J., Qin, F., Wang, Q., Cao, Q. Q., Fatih, O., Emel, O., Chen, J., & Zeng, M. M. (2021). Simultaneous determination of the phip-proline adduct and related precursors by uplc-ms/ms for confirmation of direct elimination of phip by proline. Food Chemistry, 365, 130484.
Dong, R., Yu, B., Yan, S. S., Qiu, Z. K., Lei, J. J., Chen, C. M., Li, Y., & Cao, B. (2020). Analysis of vitamin p content and inheritance models in eggplant. Horticultural Plant Journal, 6, 240-246.
Ding, S.Y., Peng, B., Li, Y., & Yang, J. (2019). Evaluation of specific volume, texture, thermal features, water mobility, and inhibitory effect of staling in wheat bread affected by maltitol. Food Chemistry, 283, 123-130.
Ferarsa, S., Zhang, W. X., Moulai-Mostefa, N., Ding, L. H., Jaffrin, M. Y., & Grimi, N. (2018). Recovery of anthocyanins and other phenolic compounds from purple eggplant peels and pulps using ultrasonic-assisted extraction. Food & Bioproducts Processing, 109, 19-28.
Hu, W. H., Dai, D. K., Zheng, Z. Y., Duan, R., Chan, K. L., Dong, T. X., Qin, Q. W., & Tsim, W. K . The binding of kaempferol-3-o-rutinoside to vascular endothelial growth factor potentiates anti-inflammatory efficiencies in lipopolysaccharide-treated mouse macrophage raw264.7 cells. Phytomedicine : international journal of phytotherapy and phytopharmacology, 80, 153400.
Ikeda, H. (2021). Plasma amino acid levels in individuals with bacterial pneumonia and healthy controls. Clinical Nutrition ESPEN, 44, 204-210.
Joshi, K., Kumari, A., Arora, S., & Singh,A. K. (2016). Development of an analytical protocol for the estimation of maltitol from yoghurt, burfi and flavoured milk. LWT- Food Science & Technology, 70, 41-45.
Karimi, A., Kazemi, M., Samani, S. A., & Simal-Gandara, J. (2021). Bioactive compounds from by-products of eggplant: functional properties, potential applications and advances in valorization methods. Trends in Food Science & Technology. 112, 518-531.
Kazemi, M., Khodaiyan, F., & Hosseini, S. S. (2019). Eggplant peel as a high potential source of high methylated pectin: ultrasonic extraction optimization and characterization. Lebensmittel Wissenschaft Und Technologie, 105, 182-189.
Luthria, D., Singh, A. P., Wilson, T., Vorsa, N., Banuelos, G. S., & Vinyard, B. T. (2010). Influence of conventional and organic agricultural practices on the phenolic content in eggplant pulp: plant-to-plant variation. Food Chemistry, 121, 406-411.
Li, j., Wang, H., Lu, Y., Mao, T. F., Xiong, J., He, S. L., & Liu, H. (2019). Inhibitory effect of tartary buckwheat seedling extracts and associated flavonoid compounds on the polyphenol oxidase activity in potatoes(solanum tuberosum l.). Journal of Integrative Agriculture, 18, 2173-2182.
Mi, S., Yu, W. L., Li, J., Liu, M. X., & Wang, X. H. (2020). Characterization and discrimination of chilli peppers based on multi-element and non-targeted metabolomics analysis. LWT- Food Science and Technology, 131, 109742.
Mihailova, A., Kelly, S. D., Chevallier, O. P., Elliott, C. T., Maestroni, B. M., & Cannavan, A. (2021). High-resolution mass spectrometry-based metabolomics for the discrimination between organic and conventional crops: a review. Trends in Food Science & Technology, 110, 142-154.
Mi, S., Shang, K., Li, X., Zhang, C. H., Liu, J. Q., & Huang, D. Q. (2019). Characterization and discrimination of selected China's domestic pork using an LC-MS-based lipi-domics approach. Food Control, 100, 305–314.
Nergiz, G., Selman, U., Anne, F., Amy, F., & Sami, D. (2018). Health benefits and bioactive compounds of eggplant. Food Chemistry, 268, 602-610.
Niño-Medina, G., Urías-Orona, V., Muy-Rangel, M. D., & Heredia, J. B. (2017). Structure and content of phenolics in eggplant (solanum melongena) - a review - sciencedirect. South African Journal of Botany, 111, 161-169.
Rosa-Martínez, E., García-Martínez, M. D., Adalid-Martínez, A. M., Pereira-Dias, L., Casanova, C., Soler, E., Fig s, M. R., Raig n, M.D., Plazas, M., Soler, S., & Prohens, J. (2021). Fruit composition profile of pepper, tomato and eggplant varieties grown under uniform conditions. Food Research International, 147, 110531.
Raigón, M. D., Prohens, J., Mu oz-Falc n, J. E., & Nuez, F. (2008). Comparison of eggplant landraces and commercial varieties for fruit content of phenolics, minerals, dry matter and protein. Journal of Food Composition & Analysis, 21, 370-376.
Sun, Q. J., Chong, N., Lei, D., & Liu, X. (2015). Effect of heat-moisture treatment with maltitol on physicochemical properties of wheat starch. LWT- Food Science & Technology, 62, 319-324.
Son, Y. J., Choi, S. Y., Yoo, K. M., Lee, K. W., Lee, S. M., Hwang, I. K., & Kim, S. (2018). Anti-blooming effect of maltitol and tagatose as sugar substitutes for chocolate making. LWT- Food Science and Technology, 88, 87-94.
Taher, D., Solberg, S., Prohens, J., Chou, Y. Y., Rakha , M., & Wu, T. H. (2017). World vegetable center eggplant collection: origin, composition, seed dissemination and utilization in breeding. Frontiers in Plant Science, 8.
Wu, S. B., Meyer, R. S., Whitaker, B. D., et al. (2013). A new liquid chromatography-mass spectrometry-based strategy to integrate chemistry, morphology, and evolution of eggplant (Solanum) species. Journal of Chromatography A, 1314, 154-172.
Wang, Y., Tang, C., & Zhang, H. (2015). Hepatoprotective effects of kaempferol 3-o-rutinoside and kaempferol 3-o-glucoside from carthamus tinctorius l. on CCl4-induced oxidative liver injury in mice. Journal of Food and Drug Analysis, 23, 310-317.
Wu, J. L., Zhou, C. X., Wu, P. J., Jin, X., Guo, Y. Q., & Fei, X., et al. (2017). Brain metabolomic profiling of eastern honey bee (apis cerana) infested with the mite varroa destructor. Plos One, 12, Article e0175573.

Table 1 List of significantly different chemical elements between peel and flesh of two varieties of eggplants.

No.	Elements	Concentration			p value	VIP score
No.	Elements	EP1	EP2	EF1	EF2
1	B	15.58±0.38a	14.19±0.49b	13.83±1.02b	11.25±0.23c	0.000	1.115
2	Na	152.01±1.22d	342.54±1.62a	158.70±1.54c	204.88±3.20b	<0.0001	1.195
3	Mg	1214.86±1.64c	1597.71±20.70a	1297.90±13.14b	1280.49±20.72b	<0.0001	1.369
4	K	24046.95±130.01b	26535.13±52.43a	18508.51±169.77d	19539.47±314.52c	<0.0001	1.024
5	⁴³Ca	1934.64±16.24b	1683.81±17.29c	2487.97±51.00a	939.04±27.64d	<0.0001	1.639
6	⁴⁴Ca	1877.57±10.32b	1646.52±6.91c	2383.62±24.04a	927.39±27.99d	<0.0001	1.632
7	Cu	4299.15±13.29a	3817.04±201.97b	3068.47±37.70c	1853.35±720.50d	<0.0001	1.077
8	As	66.95±1.89c	91.04±3.53a	71.51±1.39b	65.86±1.47c	<0.0001	1.290
9	Rb	7267.12±50.52a	4606.04±23.15c	5297.48±98.91b	3217.86±22.15d	<0.0001	1.436
10	Sr	6801.70±16.66b	3880.76±47.96c	7051.52±58.94a	2646.42±41.22d	<0.0001	1.366
11	Pd	4.34±0.74a	0.57±0.99b	3.63±1.25a	1.81±1.57ab	0.017	1.161
12	Cd	16.47±0.60b	13.58±0.26c	17.21±0.26a	8.64±0.23d	<0.0001	1.383
13	Sn	1445.97±81.07b	1502.80±134.54b	1897.01±40.11a	1530.56±249.15b	0.022	1.687
14	Cs	8.05±0.44a	9.02±0.62a	2.96±0.62b	2.95±0.33b	<0.0001	1.062
15	Ba	1524.80±23.22a	1224.44±49.52c	1407.53±25.60b	603.72±13.98d	<0.0001	1.243
16	Tl	1.59±0.23a	1.56±0.11a	1.62±0.10a	0.86±0.18b	0.001	1.205

Notes: The values in the table indicate the average and standard deviation (n = 3). Different letters in the same row indicate significant differences at p ≤ 0.05. The levels of B, Na, Mg, K and Ca are expressed in microgram per gram of skin and fresh of eggplants on fresh weight basis, and the other elements are given in microgram per kilogram of skin and fresh of eggplants on fresh weight basis. EP1, MVS peel; EP2, MVE peel; EF1, MVS flesh; EF2, MVE flesh. VIP, the different importance on the projection.

Table 2 The top 10 most significantly up- or down-regulated metabolites between peels (n = 4) of two varieties of eggplants.

No.	Compounds	Adduct ions	FC (EP1/EP2)	p value	VIP
1	Maltitol	[M-H2O-H]^-	36.58	7.36E-06	3.67
2	L-Rhamnose	[M-H]^-	35.77	3.10E-05	1.85
3	11(Z),14(Z)-Eicosadienoic Acid	[M-H]^-	10.64	2.65E-02	2.66
4	Chlorpromazine	[M+H]⁺	9.16	6.72E-06	1.23
5	(+)-Abscisic acid	[M-H]^-	5.95	4.86E-03	2.02
6	Argininosuccinic acid	[M+H]⁺	5.43	7.03E-08	2.20
7	Phenylacetic acid	[M+H]⁺	5.43	3.06E-03	1.31
8	trans-cinnamate	[M+H]⁺	5.13	1.60E-03	1.15
9	N-Acetyl-D-lactosamine	M^-	5.11	6.40E-04	2.37
10	L-Tryptophan	[M+H]⁺	4.56	1.10E-03	2.32
11	Thymidine	[M-H]^-	0.49	4.97E-02	1.77
12	Adenine	[M+H]⁺	0.41	2.53E-02	10.04
13	Val-Val	[M+K]⁺	0.41	3.23E-02	1.77
14	L-Pipecolic acid	[M+H]⁺	0.35	2.07E-03	1.75
15	3-Hydroxydodecanoic acid	[M+Na-2H]^-	0.34	4.60E-02	1.51
16	2'-O-methyladenosine	[M+H]⁺	0.33	4.67E-02	5.40
17	sn-Glycerol 3-phosphoethanolamine	[M-H]^-	0.33	5.80E-03	2.53
18	Deoxycytidine	[M+H]⁺	0.32	4.34E-04	1.23
19	Eicosapentaenoic Acid ethyl ester	M⁺	0.32	1.90E-02	10.39
20	Glycerophosphocholine	M⁺	0.26	1.95E-02	16.18

Notes: m/z, mass to charge ratio; FC, fold change; EP1, MVS peel; EP2, MVE peel; VIP, variable importance in projection.

Table 3 The top 10 most significantly up- regulated or top 10 down-regulated metabolites between peels (n = 4) and flesh (n = 4) of long and slender in shape eggplant.

No.	Compounds	Adduct ions	FC (EP1/EF1)	p value	VIP
1	Maltitol	[M-H2O-H]^-	131.56	6.52E-06	3.11
2	Kaempferol 3-O-rutinoside	[M+H]⁺	116.70	1.24E-02	2.28
3	trans-3-Coumaric acid	[M+H-H2O]⁺	87.07	1.28E-03	2.52
4	L-Rhamnose	[M-H]^-	76.71	2.83E-05	1.56
5	Tyramine	[M+H]⁺	21.37	1.61E-03	2.57
6	Quinic acid	[2M-H]^-	15.08	2.79E-06	15.21
7	11(Z),14(Z)-Eicosadienoic Acid	[M-H]^-	9.35	2.86E-02	2.30
8	Chlorpromazine	[M+H]⁺	7.34	9.39E-06	1.10
9	2-Methylbutyroylcarnitine	[M-2H+3K]⁺	7.09	9.70E-04	1.15
10	Phenylacetic acid	[M+H]⁺	7.09	1.55E-03	1.22
11	D-Fructose	(2M-H)-	0.70	3.22E-02	3.57
12	Dihydroxyacetone	(M-H)-	0.70	4.08E-02	3.99
13	1-Stearoyl-sn-glycerol 3-phosphocholine	(M+H)+	0.62	4.04E-02	3.66
14	Pantothenol	(M+H)+	0.59	1.05E-02	1.19
15	D-Glucono-1,5-lactone	(M-H)-	0.56	1.52E-02	5.31
16	D-Arabinono-1,4-lactone	(M-H)-	0.48	2.74E-02	8.31
17	Galactonic acid	(M-H2O-H)-	0.37	2.64E-03	4.90
18	Nobiletin	(M+H)+	0.16	2.33E-02	7.75
19	Tangeritin	(M+H)+	0.16	3.15E-02	6.19
20	1-Stearoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine	(M+H)+	0.09	2.66E-02	3.65

Notes: m/z, mass to charge ratio; FC, fold change; EP1, MVS peel; EF1, MVS flesh; VIP, variable importance in projection.

Table 4 The top 10 most significantly up- or down-regulated metabolites between peels (n = 4) and flesh (n = 4) of egg-shaped eggplant.

No.	Compounds	Adduct ions	FC (EP2/EF2)		p value		VIP
1	Kaempferol 3-O-rutinoside	[M+H]⁺	59.15	5.33E-03		1.55
2	Tyramine	[M+H]⁺	25.53	3.97E-05		1.85
3	Quinic acid	[2M-H]^-	17.39	1.23E-02		11.82
4	5-Methylcytosine	[M+H]⁺	15.05	4.58E-05		1.23
5	3,4-Dihydroxymandelic acid	[M-H2O-H]^-	9.38	4.22E-02		1.25
6	Cyanidin 3-glucoside cation	M⁺	8.54	2.98E-02		1.01
7	Deoxycytidine	[M+H]⁺	8.43	1.49E-06		1.41
8	Isoscopoletin	[M-H]^-	8.34	1.69E-02		1.00
9	Thymidine	[M-H]^-	7.09	5.77E-04		2.13
10	Thymine	[M+H]⁺	6.02	4.93E-04		1.76
11	Dihydroxyacetone	(M-H)-	0.51	2.48E-03		3.82
12	Galactonic acid	(M-H2O-H)-	0.46	2.23E-04		4.18
13	.alpha.-D-(+)-Talose	(M+H-H2O)+	0.45	6.00E-03		4.18
14	Alpha-D-Glucose	(M-H)-	0.45	1.19E-02		1.37
15	D-Lyxose	(M-H)-	0.41	6.28E-04		2.92
16	D-Fructose	(2M-H)-	0.35	5.68E-04		5.12
17	D-Allose	(2M-H)-	0.34	1.77E-04		4.96
18	Sucrose	(M-H)-	0.33	1.56E+00		12.36
19	Galactinol	(M+CH3COO)-	0.28	2.10E-03		1.34
20	Serotonin	(M+H)+	0.09	2.45E-02		1.65

Notes: m/z, mass to charge ratio; FC, fold change; EP2, MVE peel; EF2, MVE flesh; VIP, variable importance in projection.

Table 5 The top 8 most significantly up- or top 12 down-regulated metabolites between flesh (n = 4) of two varieties of eggplants.

No.	Compounds	Adduct ions	FC (EF1/EF2)	p value	VIP
1	D-Proline	[M+H]⁺	2.28	1.48E-02	2.16
2	L-Tryptophan	[M+H]⁺	2.26	4.37E-02	2.35
3	N-Acetyl-D-lactosamine	M^-	2.22	3.06E-02	2.15
4	N-Acetylputrescine	[M+H]⁺	2.09	3.15E-02	2.52
5	D-Aspartic acid	[M-H]^-	1.70	2.01E-02	1.46
6	D-Ribose	[M+CH3COO]^-	1.68	2.01E-02	1.17
7	Allantoin	(M-H)-	1.41	4.79E-02	2.70
8	Diacetyl	(M+H)+	1.33	5.00E-02	10.73
9	9-OxoODE	(M-H)-	0.39	4.86E-02	3.36
10	Adenine	(M+H)+	0.37	3.06E-02	9.53
11	L-Pipecolic acid	(M+H)+	0.34	1.97E-02	1.78
12	Eicosapentaenoic Acid ethyl ester	M+	0.31	5.48E-03	7.16
13	Cytidine	(M+H)+	0.31	3.58E-02	2.94
14	Ascorbic acid 6-palmitate	(M+H)+	0.28	2.353-03	1.22
15	alpha-Linolenic acid	(M+H)+	0.28	3.51E-02	1.80
16	Guanosine	(M+H)+	0.25	4.12E-02	5.07
17	Ribothymidine	(M-H)-	0.23	3.22E-02	1.07
18	Serotonin	(M+H)+	0.15	3.23E-02	1.53
19	9(S)-HODE	(M-H)-	0.15	9.04E-04	10.93
20	Stearidonic Acid	(M-H)-	0.10	1.82E-02	2.08

Notes: m/z, mass to charge ratio; FC, fold change; EF1, MVS flesh; EF2, MVE flesh; VIP, variable importance in projection.

Supplementary Figure 1 is not available with this version.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Characterization and discrimination of two varieties of eggplants using multi-element and metabolomics profiles with chemometrics

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1. Sample collection and preparation

2.2. Multi-element analysis by inductively coupled plasma mass spectrometry

2.3. Metabolomics analysis by UHPLC-Q-TOF/MS

2.4. Data processing and statistics

3. Results And Discussion

3.1. Chemical elemental analysis of peel and flesh of two varieties of eggplants

3.2. Multi-element comparison of two varieties of eggplants

3.3. Metabolomic comparison of two varieties of eggplant

3.3.1. Comparison of metabolite composition of two varieties of eggplant peels

3.3.2. Comparison of metabolite composition of MVSP and MVSF

3.3.3. Comparison of metabolite composition of MVEP and MVEF

3.3.4. Comparison of metabolite composition of two varieties of eggplant flesh

4. Conclusions

Declarations

References

Tables

Supplementary Figures

Additional Declarations

Status:

Version 1