Introduction

 

Wolfberry (Lycium barbarum L., also known as ‘goji’) is a kind of traditional Chinese medicine, belonging to the genus Lycium in Solanaceae. Wolfberry contains lots of bioactive compounds and trace elements, and has many medicinal functions (Zhou et al. 2018). In traditional Chinese medicine, wolfberry was used as a mild tonic (Potterat 2009) and reported to reduce blood sugar and lipids and hypertension, and exhibit other bioactivities including anti-aging, immune regulation, anti-cancer, anti-fatigue, protection of liver and kidneys, and promotion of male fertility (Luo et al. 2004; 2006; Li et al. 2007; Amagase and Farnsworth 2011; Li et al. 2011). And it also has shown that wolfberry can increase retinal ganglion cell survival after rat partial optic nerve transection (Li et al. 2019). As a result, wolfberry has become one of the most popular fruits in the global marketplace, and is also processed into different products, such as wolfberry juice and wine (Benzie et al. 2006; Ren et al. 2018). At the same time, some researches have combined wolfberry with other food ingredients. Can made of chrysanthemum flower (Chrysanthemum morifolium cv. Hangju) and wolfberry can resist oxidation and inflammatory (Zhang et al. 2019a).

Wolfberry is mainly produced in the northwest of China, mainly including Ningxia, Gansu, Inner Mongolia, Xinjiang and Qinghai Provinces. Of these, the quality of the wolfberry in Zhongning County, Ningxia Province is the best, and Zhongning County is known as the birthplace of Lycium barbarum L. in the world, with more than 600 years history of wolfberry cultivation (Zhang et al. 2017). The Ministry of Agriculture, China formally approved the registration and protection of geographical indications for ‘Zhongning wolfberry’ as one of the agricultural products on January 10, 2017. In recent years, a number of new varieties have been produced through screening of natural dominant mutants and one of the main varieties widely cultivated in the northwest of China is ‘Ningqi No. 1’. However, due to the different geographical sources, cultivation methods, light intensities, temperatures, precipitations, soils and other environmental factors, wolfberries from different habitats show great differences in medical functions (Meng et al. 2019). At present, the researches on wolfberry from different habitats mainly focused on the determination of the contents of active compounds (Yin and Dang 2008; Wang et al. 2010), fingerprint evaluation (Peng et al. 2004; Lu et al. 2014; Liu et al. 2015), cultivation technology (Zhang et al. 2013), etc.

Proteomics had been applied in many fields. The effect of different concentrations of nitrogen on different plants was studied by proteomics (Tan et al. 2017; Xiong et al. 2019; Tang et al. 2019), and proteomics was combined with other omics to analyze the effects of different culture conditions on crops (Guzman-Albores et al. 2019), as well as the tolerance and molecular defense mechanisms of plants under various stresses (Wei et al. 2019a; Jia et al. 2019; Almutrairi 2019). Label-free proteomic and untargeted metabolomic analysis were used to characterize and differentiate ginger samples from China and Ghana, and a total of 180 significantly different proteins were found, which could be the underlying cause of the intraspecific differences in ginger samples (Yin et al. 2017). The protein expression level of different varieties of kiwifruit (Actinidia Lindl.) were compared by proteomics, which was helpful to understand the metabolic pathway and biological process of different kiwifruit flesh color changes (Lin et al. 2017). The molecular mechanism of tobacco response to different climatic environments was preliminarily clarified by analyzing the protein expression level of the same variety from different habitats (Cai et al. 2013). However, it has not been found that the geographical differences of wolfberry from the perspective of proteins. It is essential to study wolfberry by proteomics in order to further determine the differences of wolfberry from different habitats in China.

In this study, the quantitative proteomics of wolfberry from five habitats was studied by iTRAQ and 2D LC-MS/MS, and the differences of wolfberry proteome expression levels between Zhongning, Ningxia and other four habitats (Jinghe, Xinjiang, Urat Front Banner, Inner Mongolia, Guazhou, Gansu, and Delingha, Qinghai) were analyzed respectively. Through the bioinformatics analysis, we obtained the protein information of wolfberry from different habitats. The purpose of this study was to determine the main differences of wolfberry proteins in different habitats, to provide a reference for the study of synthesis and accumulation of active compounds in wolfberry, and to help clarify the quality formation mechanism of wolfberry.

 

Materials and Methods

 

Sample sources

 

Wolfberry (variety ‘Ningqi No. 1’) fruits were collected from Ningxia, Xinjiang, Inner Mongolia, Gansu and Qinghai, China from July 20 to August 10, 2017. According to the harvest time of different habitats, we harvested fresh fruits which could represent the characteristics of the local wolfberry. For the sake of minimizing the influence of the field management on the quality of wolfberry, the factors such as tree age, density and pruning conditions were fully taken into account in sampling. ‘Ningqi No. 1’ wolfberries with 6–10 years old were randomly selected from each habitat. In the garden, we selected three normal plants grown continuously in the middle of the garden, and the mixed fresh fruits of the wolfberry were collected from all parts of the trees. After harvesting, the fruits were frozen in liquid nitrogen immediately, and stored at -80oC for further analysis.

 

Protein extraction

 

Trichloroacetic acid (TCA)/acetone and SDT lysis method were used to extract the total protein of wolfberry (Zhu et al. 2014). 5% (m/v) TCA/acetone (1:9) was added into the finely powder wolfberry and mixed it with vortex. The mixture was precipitated at -20oC for 4 h. After centrifuged (6,000 × g, 4oC 40 min), the precipitate was air dried. 30% (m/v) of SDT buffer was added into 20 µg powder, then mixed and boiled for 5 min. The homogenate was sonicated for 80 W, 10 s, intermittent for 15 s, 10 times and then boiled for 15 min. After centrifuged (14,000 × g, 40 min), the supernatants were filtered by 0.22 µm filters. Then the BCA Protein Assay Kit (Bio-Rad, USA) was used to determine the protein content. And the samples were stored at -80ºC.

 

SDS-polyacrylamide gel electrophoresis (PAGE)

 

The experimental steps were same as previously mentioned (Li et al. 2017). 20 μg of proteins for each sample were added to 5 × loading buffer (10% SDS, 0.5% Bromophenol Blue, 50% glycerol, 500 mM DTT, 250 mM Tris-HCL, pH 6.8), and then boiled in water for 5 min. After that, 5% stacking gels and 12.5% resolving gels (Seebio, Shanghai, China) were used for SDS-PAGE analysis of the samples respectively (14 mA, 90 min), and the protein bands were visualized by Coomassie Blue R-250 (Invitrogen) staining.

 

Protein digestion

 

The filter-aided sample preparation (FASP) procedure was used to digest the protein (Wisniewski et al. 2009). 200 μg of wolfberry protein sample was dissolved in 30 μL SDT buffer (4% SDS, 100 mM DTT and 150 mM Tris-HCl, pH 8.0), incubated in boiling water for 5 min and cooled them down to the room temperature. Then the experimental steps were same as previously described (Chen et al. 2017). The sample was added with 200 μL UA buffer (8 M urea, 150 mM Tris-HCl, pH 8.0), placed in a 10 kD filter (Sartorius, German), then centrifuged (14,000 × g, 15 min). This step was repeated once. The sample was added with 100 μL IAA buffer (100 mM IAA in UA), incubated for 30 min in darkness, then centrifuged (14,000 × g, 15 min). Then filters were washed three times with 100 μL UA buffer, and washed twice with 100 μL Dissolution buffer (DS buffer). 40 μL of trypsin (Promega, Madison, WI, USA) buffer (4 μg trypsin in 40 μL DS buffer) was added, then the samples were digested overnight at 37℃. After centrifuged (14,000 × g, 15 min), the BCA Protein Assay Kit was used to determine the protein content.

iTRAQ labeling

 

According to the manufacturer’s instructions, 100 μg peptide mixture of each sample was labeled using the iTRAQ reagent (Applied Biosystems, Framingham, MA, U.S.A.). And the protein samples were labeled with 113 (Zhongning, Ningxia), 114 (Jinghe, Xinjiang), 115 (Urat Front Banner, Inner Mongolia), 116 (Guazhou, Gansu) and 117 (Delingha, Qinghai) respectively.

 

Strong cation exchange (SCX) chromatography

 

iTRAQ labeled peptides were fractionated using the AKTA Purifier system (GE Healthcare, Litter Chalfont, United Kingdom) by SCX chromatography. Recombined the dried peptide mixtures, acidified it with buffer A (10 mM KH2PO4 in 25% of ACN, pH 3.0) and loaded it into a Poly SULFOETHYL 4.6 × 100 mm column (5 µm, 200 Å, Poly LC Inc., MD, USA). The peptide was eluted with buffer B (500 mM KCl, 10 mM KH2PO4 in 25% of ACN, pH 3.0) at a flow rate gradient of 1 mL/min. The linear gradient of buffer B absorbance value and follow-up steps could be found in the previous study (0–8% for 22 min, 8–52% during 22–47 min, 52–100% during 47–50 min, 100% during 50–58 min, and the buffer B was reset to 0% after 58 min) (Lin et al. 2017). The elution was monitored at 214 nm, the components were collected every minute and a total of 30 fractions were collected. Then each group of samples was divided into 3 portions, the collected fractions were desalted on C18 Cartridges and concentrated by vacuum centrifugation.

 

LC-MS/MS analysis

 

Every sample was separated using a HPLC liquid phase system Easy nLC (Thermo Fisher Scientific, Odense, Denmark). 0.1% formic acid was Buffer A, 0.1% formic acid (84% acetonitrile) was Buffer B, and the column was equilibrated with 95% Buffer A. The sample was loaded into a loading column (Thermo Scientific Acclaim PepMap100, 100 μm × 2 cm, nano Viper C18) from an autosampler and separated by an analytical column (Thermo Scientific Easy Column, 10 cm, ID 75 μm inner diameter, 3 μm resin, C18-A2) at a flow rate of 300 nL/min by IntelliFlow technology. LC-MS/MS analysis was performed on a Q Exactive mass spectrometer (Thermo Scientific, San Jose, CA, USA), and it was coupled to Easy nLC. And the situation was same as previously described (Wang et al. 2013).

 

Database search and protein identification and quantification

 

The identification and quantification of mass spectrometry were performed by MASCOT engine (Matrix Science, London, U.K., version 2.2) embedded into Proteome Discoverer 1.4. The MS/MS data were searched against the database for protein identification and quantification, and the criteria were as followed: uniprot_Solanoideae_139867_20170221 protein database (downloaded February 2017, 818,444 sequences). The MASCOT parameters were set same as previously mentioned (Wang et al. 2013).

The reported data were based on at least one unique peptide with 99% confidence for protein identification as determined by false discovery rate (FDR)≤1% (Yuan et al. 2018). There was at least one unique peptide in each successfully identified protein. All peptide ratios were standardized with the median protein ratio, which should be 1. Up- or down-regulated proteins were determined with a 1.2-fold cutoff, and a P < 0.05 (Chu et al. 2015; Miao et al. 2015). The protein ratio was further analyzed by Student's t-test, and the statistical package was Perseus1.3.0.4. And the ‘Ningqi No. 1’ wolfberry in Zhongning, Ningxia was considered as a relative quantitative reference.

 

Bioinformatic and statistical analysis

 

We performed GO annotation (http://www.geneontology.org/) and the KEGG pathway (http://www.genome.jp/kegg/) annotation on DEPs using Blast2GO (Gotz et al. 2008) program and KAAS (KEGG Automatic Annotation Server) software (Moriya et al. 2007) respectively. And then Fisher's Exact Test was used to perform an enrichment analysis of GO annotations and KEGG pathway annotations for the DEPs.

 

Results

 

Analysis of SDS-PAGE in protein samples of wolfberry

 

As shown in Fig. 1, each strip was distributed between 14.4 and 116 kD, and the molecular weights of most proteins were between 18.4 and 25 kD. The distribution patterns of strips among different samples were similar. The results preliminarily showed that the proteins extracted from wolfberry met the requirements of iTRAQ technology, and thus could be further analyzed by protein labeling and chromatography.

 

Protein identification and quantification

 

Three repeated experiments were carried out to compare the wolfberry proteome by LC-MS/MS technology. As shown in Table 1, total of 818,444 LC MS/MS spectra were matched to known spectra; among them, the number of matched spectra was 60,795, and the utilization rate of the spectrum was 7.43%; a total of 16,384 and 12,375 unique peptides were obtained. These peptides could identify 4,852 proteins of the wolfberry.

Table 1: Statistics of the protein identification results

 

No.

Total spectra

Spectra (PSM)

Peptides

Unique peptides

Protein groups

1

276883

21380

11360

8877

4093

2

268829

19591

10657

8350

3903

3

272732

19824

10740

8410

3903

1,2,3, Combination and results

818444

60795

16384

12375

4852

 

 

Fig. 1: SDS-PAGE analysis. Markers represent the protein standards. A/B/C/D/E represent the proteins in wolfberry from Zhongning, Ningxia, Jinghe, Xinjiang, Urat Front Banner, Inner Mongolia, Guazhou, Gansu and Delingha, Qinghai respectively, and 1/2/3 represents three repeated experiments

 

Fig. 2: Characteristics of the peptides. (a) Distribution of the identified proteins with different molecular weights (kD). (b) Distribution of the identified proteins with different isoelectric points. (c) Distribution of the identified proteins with different peptide length. (d) Distribution of the identified proteins with protein sequence coverage (%)

Mascot software was used to visualize the identified proteins. Most of the identified proteins (72.88%) had molecular weights ranging from 10–20 kD (661 proteins), 20–30 kD (823 proteins), 30–40 kD (806 proteins), 40–50 kD (662 proteins), or 50–60 kD (584 proteins) (Fig. 2a). As shown in Fig. 2b, the isoelectric points of most of the identified proteins (90.56%) were 6–10 (4,394 proteins). In the identified proteins, the number of amino acids was mainly 5–19, of which 7–11 were the most (Fig. 2c). As shown in Fig. 2d, the identified proteins had high peptide coverage.

In comparison group with the wolfberry samples in Ningxia, a total of 1,437 DEPs were separated under the three biological replications. As shown in Fig. 3a, of the 1,437 DEPs, 160 (286), 55 (111), 438 (528) and 498 (517) proteins were up-regulated (down-) in Xinjiang/Ningxia, Inner Mongolia/Ningxia, Gansu/Ningxia and Qinghai/Ningxia respectively.

We could find that the number of DEPs was different between Ningxia and other habitats: Ningxia/Qinghai had the largest number of DEPs, followed by Gansu/Ningxia and Xinjiang/Ningxia and the least DEPs were found in Ningxia/ Inner Mongolia. In addition, we could find that the number of up-regulated DEPs was always less than the down-. Meanwhile, we used Venn diagrams to study the overlap of DEPs among the four comparison groups. As shown in Fig. 3b, we found 380 DEPs in one group, 452 DEPs were found in any two groups, 258 DEPs were found in any three groups and 56 DEPs were found in all three groups. It was found that the number of DEPs was different between Ningxia and other habitats: Ningxia and Qinghai had the largest number of DEPs, followed by Gansu and Xinjiang, and the least DEPs were found in Ningxia and Inner Mongolia.

Bioinformatics analysis

 

Fig. 3: Differential expression of proteins. (a) The number of up-regulated, down-regulated and total DEPs in each group. (b) The overlap in total DEPs among the four groups. A/B/C/D/E notes were shown in Fig. 1

 

GO functional classification: The functional classification of all DEPs was determined by analyzing their biological process (BP), molecular function (MF) and cellular component (CC). The comparison between Ningxia and Xinjiang is shown in Fig. 4a, DEPs were classified into 47 functional groups, of which BP accounted for 20 GO terms, MF accounted for 11 Go terms, and CC accounted for 16 GO terms. The GO terms included ‘developmental process’ (P = 0.01147), ‘growth’ (P = 0.03839), ‘reproduction’ (P = 0.00693), ‘multicellular organismal process’ (P = 0.01608), ‘reproductive process’ (P = 0.00617) and ‘response to stimulus’ (P = 0.01123).

As depicted in Fig. 4b, in the comparison group of Ningxia and Inner Mongolia, DEPs were classified into 41 functional groups, of which BP accounted for 19 GO terms, MF accounted for 9 GO terms, and CC accounted for 13 GO terms. The GO terms included ‘detoxification’ (P = 0.00810), ‘antioxidant activity’ (P = 0.00689) and ‘membrane’ (P = 0.02226).

In the comparison group of Ningxia and Gansu (Fig. 4c), DEPs were classified into 49 functional groups, of which BP accounted for 21 GO terms, MF accounted for 12 GO terms, and CC accounted for 16 GO terms. The GO terms included ‘metabolic process’ (P = 0.04212) and ‘structural molecule activity’ (P = 0.00337).

In the comparison group of Ningxia and Qinghai (Fig. 4d), DEPs were classified into 48 functional groups, of which BP accounted for 21 GO terms, MF accounted for 11 GO terms, and CC accounted for 16 GO terms. The GO terms included ‘detoxification’ (P = 0.03390), ‘antioxidant activity’ (P = 0.02454) and ‘structural molecule activity’ (P = 0.00637).

In these 4 groups, ‘metabolic process’, ‘cellular process’ and ‘single−organism process’ were the most important terms among biological processes, ‘catalytic activity and ‘binding’ were the most important terms among molecular functions, and ‘cell’, ‘cell part’ and ‘organelle’ were the most important terms among cellular components.

KEGG pathway analysis: Different proteins usually work together to perform their biological functions and we can use a pathway-based analysis to learn more about the biological functions of proteins. The comparison between Ningxia and Xinjiang is shown in Fig. 5a, the most aplenty DEPs in KEGG pathway were bound up with ‘carbon metabolism’ (10 up-regulated, 17 down-regulated), and the main pathways in the KEGG enrichment analysis were ‘peroxisome’ (P = 0.01815).

As shown in Fig. 5b, in the comparison group of Ningxia and Inner Mongolia, DEPs with KEGG pathway mostly affected protein processing in ‘endoplasmic reticulum’ (1 up-regulated, 10 down-regulated), and the main pathways in the KEGG enrichment analysis were ‘protein processing in endoplasmic reticulum’ (P = 0.00761), ‘purine metabolism’ (P = 0.04420) and ‘phenylpropanoid biosynthesis’ (P = 0.04708).

In the comparison group of Ningxia and Gansu (Fig. 5c), the most aplenty DEPs in KEGG pathway were bound up with ‘carbon metabolism’ (27 up-regulated, 31 down-regulated), and the main pathways in the KEGG enrichment analysis were ‘ribosome’ (P = 0.00018), ‘carbon metabolism’ (P = 0.04189), ‘biosynthesis of amino acids’ (P = 0.02000), ‘glycolysis / Gluconeogenesis’ (P = 0.02571), ‘carbon fixation in photosynthetic organisms’ (P = 0.00164), ‘glyoxylate and dicarboxylate metabolism’ (P = 0.0204), ‘2-Oxocarboxylic acid metabolism’ (P = 0.00724), ‘peroxisome’ (P = 0.009062) and ‘fructose and mannose metabolism’ (P = 0.02895).

In the comparison group of Ningxia and Qinghai (Fig. 5d) the most aplenty DEPs in KEGG pathway were bound up with ‘carbon metabolism’ (23 up-regulated, 32 down-regulated), and the main pathways in the KEGG enrichment analysis were ‘2-Oxocarboxylic acid metabolism’ (P = 0.01217), ‘protein processing in endoplasmic reticulum’ (P = 0.01607), ‘phenylpropanoid biosynthesis’ (P = 0.04519), ‘longevity regulating pathway-multiple species’ (P = 0.04141) and ‘ascorbate and aldarate metabolism’ (P = 0.04350).

 

Fig. 4: GO classification of DEPs from different comparison groups. (a) Xinjiang/Ningxia. (b) Inner Mongolia/Ningxia. (c) Gansu/Ningxia. (d) Qinghai/Ningxia

 

 

Fig. 5: Kyoto encyclopedia of genes and genomes (KEGG) classification of DEPs from different comparison groups (Top 20). (a) Xinjiang/Ningxia. (b) Inner Mongolia/Ningxia. (c) Gansu/Ningxia. (d) Qinghai/Ningxia

 

Fig. 6: Distribution of sampling spatial points of wolfberry. A/B/C/D/E represent the proteins in wolfberry from Zhongning, Ningxia, Jinghe, Xinjiang, Urat Front Banner, Inner Mongolia, Guazhou, Gansu and Delingha, Qinghai respectively

 

Discussion

 

In this study, proteomics was used to study the protein differences of wolfberry between Zhongning, Ningxia and other different habitats in China. A research has shown that there were no shared DEPs between goat and bovine comparison groups, probably because they are two different species (Wei et al. 2019b). But it was found there were some shared DEPs between the four comparison groups in our paper, this could be because they belong to the same wolfberry species ‘Ningqi No. 1’. It was found that the results of GO functional annotation analysis of peach fruits produced at different varieties of kiwifruit (Lin et al. 2017; Huan et al. 2019) were basically consistent with this study, suggesting that our results were credible. Central carbon metabolism was one of the most basic cellular metabolic pathways of all living organisms (Bar-Even et al. 2012). Through the KEGG pathway analysis we found that ‘carbon metabolism’ was the most represented pathway. Among the four groups, one of the most significant differences was ‘carbon metabolism’ in photosynthetic organisms. Photosynthetic is the decisive factor of sugar synthesis and transportation and connects the environmental and biological factors that regulate fruit development (Chen et al. 2002; Hu et al. 2019).

And some researches have shown that environment factors are the main causes of wolfberry fruit morphological changes of ‘Ningqi No. 1’ (Su et al. 2015) and the global distribution of wolfberry has obvious regional characteristics, the high temperature, low precipitation and high altitude are the main factors limiting the growth of wolfberry (Amagase and Farnsworth 2011). According to the distribution of wolfberry sampling spatial points (Fig. 6), the altitude of Guazhou, Gansu (2452 meters) and Delingha, Qinghai (2982 meters) are obviously higher than Zhongning, Ningxia (1225 meters), Urat Front Banner, Inner Mongolia (1022 meters) and Jinghe, Xinjiang (332 meters). Precipitation and temperature are usually lower at high altitudes, and the soil factors will be different, too (Haag et al. 2019; Zhang et al. 2019b). We speculated that the different climate and environmental factors, especially the altitude, may lead to different expression levels of wolfberry proteins.

Although the genome of an organism is usually stable and unchangeable, the expression and composition of proteome has been changing in the process of growth and various physiological processes (Hua et al. 2015). Ecological factors in different habitats such as altitude, light intensities, temperatures, precipitations and soil may induce the change of expression products, leading to differential expression of proteins (Cai et al. 2013; Yin et al. 2017), and accumulation of different secondary metabolites was further induced (Feng et al. 2018; Fu et al. 2019). Wolfberry fruit is the storage organ of the main medicinal parts and effective medicinal components, and the variety, content and proportion of sugar in the fruit are factors to determine the variety and its commercial value (Yao et al. 2011). When the environmental conditions are different, different soil and meteorological factors will have a certain impact on the active ingredients of wolfberry (Su et al. 2015; Abd El-Wahab et al. 2018). Therefore, the study of proteomics is helpful to study the differences of wolfberry in different habitats.

 

Conclusion

 

The proteins expression levels of the wolfberry were different between Ningxia and the other four habitats. The number of DEPs in Qinghai/Ningxia and Gansu/Ningxia was obviously more than that in Inner Mongolia/Ningxia and Xinjiang/Ningxia. And in the main functional items, the number of DEPs in Qinghai/Ningxia and Gansu/Ningxia was significantly higher than that in Inner Mongolia/Ningxia and Xinjiang/Ningxia. According to the above conclusions, it can be inferred that the wolfberry in Ningxia is significantly different from Qinghai and Gansu, and less different from Inner Mongolia and Xinjiang. This study may provide a reference for analyzing wolfberry quality in different habitats. However, the exact biological function and interaction among differentially expressed proteins need to be further studied.

 

Acknowledgement

 

We would like to thank the financial supports of National Natural Science Foundation (31660438) of China, Postgraduate Innovation Project of Ningxia University (2019), Innovation Platform Funds of Ningxia Key Laboratory for Food Microbial-Applications Technology and Safety Control (2018DPC05026), and Project of Support Local Colleges and University Reform and Development Funds of China (2018).

 

References

 

Abd El-Wahab RH, AR Al-Rashed, A Al-Dousari (2018). Influences of physiographic factors, vegetation patterns and human impacts on aeolian landforms in arid environment. Arid-Land Ecosyst 8:97‒110

Almutrairi ZM (2019). Plant molecular defense mechanisms promoted by nanoparticles against environmental stresses. Intl J Agric Biol 21:259‒270

Amagase H, NR Farnsworth (2011). A review of botanical characteristics, phytochemistry, clinical relevance in efficacy and safety of Lycium barbarum fruit (goji). Food Res Intl 44:1702‒1717

Bar-Even A, A Flamholz, E Noor, R Milo (2012). Rethinking glycolysis: on the biochemical logic of metabolic pathways. Nat Chem Biol 8:509‒517

Benzie IFF, WY Chung, J Wang, M Richelle, P Bucheli (2006). Enhanced bioavailability of zeaxanthin in a milk-based formulation of wolfberry (Gou Qi Zi; Fructus barbarum L.). Braz J Nutr 96:154‒160

Cai YZ, PX Zhou, FL Li, CL Zhao, C Lin, HW Yang, ZC Mao (2013). Proteomic analysis of tobacco rosette stage leaves under different climatic conditions. Sci Agric Sin 46:859‒870

Chen J, S Zhang, L Zhang, Z Zhao, J Xu (2002). Fruit photosynthesis and assimilate translocation and partitioning: their characteristics and role in sugar accumulation in developing Citrus unshiu fruit. Acta Bot Sin 44:158‒163

Chen YY, XM Fu, X Zhou, SH Cheng, LT Zeng, F Dong, ZY Yang (2017). Proteolysis of chloroplast proteins is responsible for accumulation of free amino acids in dark-treated tea (Camellia sinensis) leaves. J Proteom 157:10‒17

Chu P, GX Yan, Q Yang, LN Zhai, C Zhang, FQ Zhang, RZ Guan (2015). iTRAQ-based quantitative proteomics analysis of Brassica napus leaves reveals pathways associated with chlorophyll deficiency. J Proteom 113:244‒259

Feng YC, TX Fu, X Li, CY Wang, DJ Zhang (2018). Study on metabonomics of rice from various producing areas based on GC-MS technology. Chin J Biol 31:902‒909

Fu T, Y Feng, L Zhang, X Li, C Wang (2019). Metabonomics study on rice from different geographical areas based on gas chromatography-mass spectrometry. Food Sci 40:176‒181

Gotz S, JM Garcia-Gomez, J Terol, TD Williams, SH Nagaraj, MJ Nueda, M Robles, M Talon, J Dopazo, A Conesa (2008). High-throughput functional annotation and data mining with the Blast2GO suite. Nucl Acids Res 36:3420‒3435

Guzman-Albores JM, ML Ramirez-Merchant, EC Interiano-Santos, LA Manzano-Gomez, JH Castanon-Gonzalez, R Winkler, M Abud-Archila, JA Montes-Molina, FA Gutierrez-Miceli, VM Ruiz-Valdiviezo (2019). Metabolomic and proteomic analysis of Moringa oleifera cultivated with vermicompost and phosphate rock under water stress conditions. Intl J Agric Biol 21:786‒794

Haag I, PD Jones, C Samimi (2019). Central Asia's changing climate: How temperature and precipitation have changed across time, space, and altitude. Climate 7:1‒10

Hu JW, X Dai, T Song, GY Sun (2019). Effects of different light qualities on growth and photosynthetic characteristics of mulberry seedlings. Zhiwu Yanjiu 39:481‒489

Hua YJ, SN Wang, LS Zou, XH Liu, JY Xu, YY Luo, JX Liu (2015). iTRAQ-based quantitative proteomics of pseudostellariae radix from different habitats. J Chin Mass Spectron Soc 37:236‒246

Huan C, Y Xu, XJ An, ML Yu, RJ Ma, XL Zheng, ZF Yu (2019). iTRAQ-based protein profiling of peach fruit during ripening and senescence under different temperatures. Postharv Biol Technol 151:88‒97

Jia XM, YF Zhu, Y Hu, R Zhang, L Cheng, ZL Zhu, T Zhao, X Zhang, YX Wang (2019). Integrated physiologic, proteomic, and metabolomic analyses of Malus halliana adaptation to saline-alkali stress. Hortic Res 6:1‒19

Li HY, M Huang, QY Luo, X Hong, S Ramakrishna, KF So (2019). Lycium barbarum (wolfberry) increases retinal ganglion cell survival and affects both microglia/macrophage polarization and autophagy after rat partial optic nerve transection. Cell Transplant 28:606‒618

Li L, Y Tian, JK Yu, X Song, RY Jia, QK Cui, WZ Tong, YF Zou, LX Li, LZ Yin, XX Liang, CL He, GZ Yue, G Ye, L Zhao, F Shi, C Lv, SJ Cao, ZQ Yin (2017). iTRAQ-based quantitative proteomic analysis reveals multiple effects of Emodin to Haemophilus parasuis. J Proteom 166:39‒47

Li SY, D Yang, CM Yeung, WY Yu, RCC Chang, KF So, D Wong, ACY Lo (2011). Lycium barbarum polysaccharides reduce neuronal damage, blood-retinal barrier disruption and oxidative stress in retinal ischemia/reperfusion injury. PLoS One 6:1–13

Li XM, YL Ma, XJ Liu (2007). Effect of the Lycium barbarum polysaccharides on age-related oxidative stress in aged mice. J Ethnopharmacol 111:504‒511

Lin MM, JB Fang, XJ Qi, YK Li, JY Chen, LM Sun, YP Zhong (2017). iTRAQ-based quantitative proteomic analysis reveals alterations in the metabolism of Actinidia arguta. Sci Rep 7:1‒11

Liu W, JN Xu, R Zhu, YQ Zhu, Y Zhao, P Chen, C Pan, WB Yao, XD Gao (2015). Fingerprinting profile of polysaccharides from Lycium barbarum using multiplex approaches and chemometrics. Intl J Biol Macromol 78:230‒237

Lu WY, QQ Jiang, HM Shi, YG Niu, BY Gao, LL Yu (2014). Partial least-squares-discriminant analysis differentiating Chinese wolfberries by UPLC-MS and flow injection mass spectrometric (FIMS) fingerprints. J Agric Food Chem 62:9073‒9080

Luo Q, Z Li, X Huang, J Yan, S Zhang, YZ Cai (2006). Lycium barbarum polysaccharides: Protective effects against heat-induced damage of rat testes and H2O2-induced DNA damage in mouse testicular cells and beneficial effect on sexual behavior and reproductive function of hemicastrated rats. Life Sci 79:613‒621

Luo Q, YZ Cai, J Yan, M Sun, H Corke (2004). Hypoglycemic and hypolipidemic effects and antioxidant activity of fruit extracts from Lycium barbarum. Life Sci 76:137‒149

Meng J, Z Liu, CL Gou, KM Rogers, WJ Yu, SS Zhang, YW Yuan, L Zhang (2019). Geographical origin of Chinese wolfberry (goji) determined by carbon isotope analysis of specific volatile compounds. J Chromatogr B Anal Technol Biomed Life Sci 1105:104‒112

Miao JY, FL Chen, S Duan, XY Gao, G Liu, YJ Chen, W Dixon, H Xiao, Y Cao (2015). iTRAQ-based quantitative proteomic analysis of the antimicrobial mechanism of peptide F1 against Escherichia coli. J Agric Food Chem 63:7190‒7197

Moriya Y, M Itoh, S Okuda, AC Yoshizawa, M Kanehisa (2007). KAAS: an automatic genome annotation and pathway reconstruction server. Nucl Acids Res 35:182‒185

Peng Y, SQ Sun, ZZ Zhao, HW Leung (2004). A rapid method for identification of genus Lycium by FTIR spectroscopy. Guangpuxue Guangpu Fenxi 24:679‒681

Potterat O (2009). Goji (Lycium barbarum and L. chinense): phytochemistry, pharmacology and safety in the perspective of traditional uses and recent popularity. Plant Med 76:7‒19

Ren J, SY Wang, Y Ning, MZ Wang, LY Wang, BL Zhang, BQ Zhu (2018). The impact of over-maturation on the sensory and nutritional quality of Gouqi (Chinese wolfberry) wine. J Inst Brew 124:57‒67

Su X, G Qi, G Zheng, J Yang, J Liu, H Bao, J Wang (2015). Effect of meteorological factors of different regions on sugar accumulation in Lycium barbarum L. fruit. Acta Bot Bor-Occident Sin 35:1634‒1641

Tan J, Q Li, JH Zhou, LJ Chen, YY Zhang, Y Zhang, J Bin, RZ Wang (2017). Physiological response and comparative proteomic analysis of tobacco seedling roots to NH4+. Intl J Agric Biol 19:1270‒1278

Tang JC, ZG Sun, QH Chen, RN Damaris, BL Lu, ZR Hu (2019). Nitrogen fertilizer induced alterations in the root proteome of two rice cultivars. Intl J Mol Sci 20:1‒20

Wang CC, SC Chang, BS Inbaraj, BH Chen (2010). Isolation of carotenoids, flavonoids and polysaccharides from Lycium barbarum L. and evaluation of antioxidant activity. Food Chem 120:184‒192

Wang LX, WY Liang, JH Xing, FL Tan, YY Chen, L Huang, CL Cheng, W Chen (2013). Dynamics of Chloroplast Proteome in Salt-Stressed Mangrove Kandelia candel (L.) Druce. J Proteom Res 12:5124‒5136

Wei XD, DW Shi, X Li, XW Fang, JN Wu (2019a). Enhanced tolerance of transgenic rice plants over-expressing maize C-4 phosphoenolpyruvate carboxylase gene to low nitrogen conditions. Intl J Agric Biol 22:727‒736

Wei YC, X Li, DQ Zhang, YF Liu (2019b). Comparison of protein differences between high- and low-quality goat and bovine parts based on iTRAQ technology. Food Chem 289:240‒249

Wisniewski JR, A Zougman, N Nagaraj, M Mann (2009). Universal sample preparation method for proteome analysis. Nat Meth 6:359‒362

Xiong QQ, L Zhong, TH Shen, CH Cao, HH He, XR Chen (2019). iTRAQ-based quantitative proteomic and physiological analysis of the response to N deficiency and the compensation effect in rice. BMC Genom 20:1‒16

Yao X, LJ Xu, W Xiao, Y Peng, PG Xiao (2011). Analysis of Lycium barbarum polysaccharide from different lycii fructus. Herald Med 30:426‒428

Yin GH, YL Dang (2008). Optimization of extraction technology of the Lycium barbarum polysaccharides by Box-Behnken statistical design. Carbohydr Polym 74:603‒610

Yin XJ, SL Wang, RN Alolga, E Mais, P Li, PF Yang, S Komatsu, LW Qi (2017). Label-free proteomic analysis to characterize ginger from China and Ghana. Food Chem 249:1‒7

Yuan W, CL Fang, SJ Liu, WQ Yang, LS Wei, XJ Lei, F Hu, HY Huang, W Li, W Chen, LM Li, YS Long (2018). Long-term moderate exercise enhances specific proteins that constitute neurotrophin signaling pathway: A TMT-based quantitative proteomic analysis of rat plasma. J Proteom 185:39‒50

Zhang N, ZJ He, SY He, P Jing (2019a). Insights into the importance of dietary chrysanthemum flower (Chrysanthemum morifolium cv. Hangju) wolfberry (Lycium barbarum fruit) combination in antioxidant and anti-inflammatory properties. Food Res Intl 116:810‒818

Zhang SS, YM Wei, S Wei, HY Liu, BL Guo (2017). Authentication of Zhongning wolfberry with geographical indication by mineral profile. Intl J Food Sci Technol 52:457‒463

Zhang TB, YH Kang, SQ Wan (2013). Shallow sand-filled niches beneath drip emitters made reclamation of an impermeable saline-sodic soil possible while cropping with Lycium barbarum L. Agric Manage Water Qual 119:54‒64

Zhang YX, Y Li, GR Zhu (2019b). The effects of altitude on temperature, precipitation and climatic zone in the Qinghai-Tibet Plateau. J Glaciol Geocry 41:505‒515

Zhou LS, LL Huang, H Yue, K Ding (2018). Structure analysis of a heteropolysaccharide from fruits of Lycium barbarum L. and anti-angiogenic activity of its sulfated derivative. Intl J Biol Macromol 108:47‒55

Zhu Y, H Xu, H Chen, JJ Xie, MM Shi, BY Shen, XX Deng, C Liu, X Zhan, CH Peng (2014). Proteomic analysis of solid pseudopapillary tumor of the pancreas reveals dysfunction of the endoplasmic reticulum protein processing pathway. Mol Cell Proteom 13:2593‒603