Evolution and functional divergence of glycosyltransferase genes shaped the quality and cold tolerance of tea plants

Abstract

Plant uridine diphosphate-dependent glycosyltransferases (UGTs) play a key role in plant growth and metabolism. Here, we examined the evolutionary landscape among UGTs in 28 fully sequenced species from early algae to angiosperms. Our findings revealed a distinctive expansion and contraction of UGTs in the G and H groups in tea (Camellia sinensis), respectively. Whole-genome duplication and tandem duplication events jointly drove the massive expansion of UGTs, and the interplay of natural and artificial selection has resulted in marked functional divergence within the G group of the sinensis-type tea population. In Cluster II of group G, differences in substrate selection (e.g. abscisic acid) of the enzymes encoded by UGT genes led to their functional diversification, and these genes influence tolerance to abiotic stresses such as low temperature and drought via different modes of positive and negative regulation, respectively. UGTs in Cluster III of the G group have diverse aroma substrate preferences, which contribute a diverse aroma spectrum of the sinensis-type tea population. All Cluster III genes respond to low-temperature stress, whereas UGTs within Cluster III-1, shaped by artificial selection, are unresponsive to drought. This suggests that artificial selection of tea plants focused on improving quality and cold tolerance as primary targets.

Figure 1.

Identification and classification of UGTs in different cultivars of tea plants. A) Total UGT identification in different tea plant cultivars. B) The synonymous mutation rate (Ks) of UGTs' gene pairs in DASZ, YK10, and SCZ, respectively. C) Subgroup identification and classification of UGT in different tea plants. The line graph shows the change of UGT group in each breed relative to the total numbers. D) Collinearity analysis of G group in Arabidopsis, cacao, grape, and tea (SCZ). E) Collinearity analysis of H group in Arabidopsis, cacao, grape, and tea (SCZ).

Figure 2.

H group retains significant disease resistance. A to D) Expression level of UGTH1 in low temperature, dehydration, and fungal infection and E. oblique feeding on tea plants, respectively. CK, control; 4℃_6 h, 4 ℃ for 6 h; 4℃_7d, 4 ℃ for 7 days; 0℃_7d, 0 ℃ for 7 days; 24h, 48 h, 72 h, tea samples were treated with PEG-6000 for 24, 48, and 72 h, respectively; 1d, 4d, 7d, fungi infected tea samples for 1, 4, and 7 days, respectively. T1 to T3, biological replication of the E. oblique feeding on tea leaves. E) Disease symptoms after fungal infection for 4 days were observed under a stereomicroscope after suppressing UGTH1. F) The relative expression of control (sODN) and UGTH1-suppressed (AsODN) in tea plants. G) The lesion area of control (sODN) and UGTH1-suppressed (AsODN) in tea plants after fungal infection for 4 days. H to J) The relative contents of hormones of control (sODN) and UGTH1-suppressed (AsODN) in tea plants before fungal infection. FW, fresh weight. K to L) Relative expression levels of NPR1 and PR1 in tea plants of control (sODN) and UGTH1-suppressed (AsODN). M) Working models of UGTH1 from H group in response to disease resistance in tea plants. Data were expressed as the mean ± Sd from at least 5 biological replicates. All statistical analysis was performed by Student's t-tests.

Figure 3.

Evolutionary analysis of the G group in tea plants. A) Collinearity analysis of G group UGTs in DASZ, YK10, and SCZ. B) Phylogenetic relationships of the UGT genes for G group in the SCZ. CSS0040815 from the OG group in the homologous classification was selected as an outgroup. C) Hypothetical evolutionary histories of the UGTs genes in G group (left). The letters T and W in the schematic diagram showing the hypothetical origins of UGT genes indicate tandem duplication and WGD, respectively. The heat map (right) represents the transcript levels of G group genes in different tissues of tea plant, the heat map color is calculated using log₂(TPM + 1). AB, apical bud; FL, flower; FR, fruit; YL, young leaf; ML, mature leaf; OL, old leaf; RT, root; ST, stem. D) Phylogenetic relationships of G group (UGT85) genes in plants. Phylogenetic relationships were reconstructed using IQ-Tree (Ultrafast bootstrap = 5000), and sequences were aligned using MAFFT. AtUGT85A1 (AAF18537), Arabidopsis thaliana; PdUGT85A19 (ABV68925), Prunus dulcis; SbUGT85B1 (AAF17077), Sorghum bicolor; MsUGT85K4 (AEO45781), MsUGT85K5 (AEO45782), Manihot esculenta; CoUGT85N1 (AEB61489), Consolida orientalis; VvGT14 (XP_002285770.1), VvGT16 (XP_002263158.1), Vitis vinifera; CsUGT85V1 (KF446241), CsUGT85U1 (KF446243), CsUGT85U2 (KF446242), Crocus sativus; PlZOG1 (AAD04166), PlZOX1 (AAD51778), Phaseolus lunatus. UGT85A1 was selected as an outgroup. E) Expression patterns of Cluster II (left) and Cluster III (right) UGTs from G group under abiotic stresses. F to G) Phylogenetic relationships of G group genes in DASZ, YK10, and SCZ. Phylogenetic tree was reconstructed using IQ-Tree (Ultrafast bootstrap = 5000), and sequences were aligned using MAFFT. CSS0040815, CSS0032998, and CSS0033747 from the OG group in the homologous classification was selected as an outgroup. H) Selective sweep regions around Cluster III-1 genes' locus were evaluated by different summary statistics. The blue bar denotes the location of the genes within Cluster III-1. The arrow denotes the genes subjected to selection. The statistics were calculated separately for populations from wild, cultivator, landrace, and elite accessions.

Figure 4.

Functional validation of UGT genes in Cluster II from G group. A) LC–MS/MS analyses of enzymatic products formed by UGT85A53. The mass spectra of the enzymatic reaction products produced by recombinant UGT85A53 are listed. B) LC–MS/MS analyses of enzymatic products formed by UGT1. The mass spectra of the enzymatic reaction products produced by recombinant UGT1 are listed. C) Chlorophyll fluorescence images of control (sODN-1) and UGT85A53-suppressed (AsODN-1) tea plants under 4 ℃ for 48 h. The color-coded values indicate the degree of damage, with smaller values indicating more severe damage. D) Chlorophyll fluorescence images of control (sODN-2) and UGT1-suppressed (AsODN-2) tea plants under 4 ℃ for 48 h. E, G) Relative expression of control (sODN-1/sODN-2) and UGT85A53-suppressed (AsODN-1)/UGT1-suppressed (AsODN-2) in tea plants, respectively. F, H) Statistical analysis of maximum photochemical efficiency of photosystem II (F_v/F_m) in E) and G). I, J) Relative content of hormones in control (sODN-1/sODN-2) and UGT85A53-suppressed (AsODN-1)/UGT1-suppressed (AsODN-2) in tea plants under 4 ℃, respectively. K, L) The tea plant phenotypes of control (sODN-3/sODN-4) and UGT85A53-suppressed (AsODN-3)/UGT1-suppressed (AsODN-3) after drought stress for 24 h, respectively. M, P) Relative expression of control (sODN-3/sODN-4) and UGT85A53-suppressed (AsODN-3)/UGT1-suppressed (AsODN-4) in tea plants, respectively. N, Q) Leaf inclination of tea plants in control (sODN-3/sODN-4) and UGT85A53-suppressed (AsODN-3)/UGT1-suppressed (AsODN-4) after drought stress for 24 h, respectively. O, R) Relative moisture content of tea plants in control (sODN-3/sODN-4) and UGT85A53-suppressed (AsODN-3)/UGT1-suppressed (AsODN-4) after drought stress for 24 h, respectively. S, T) Relative content of hormones in control (sODN-3/sODN-4) and UGT85A53-suppressed (AsODN-3)/UGT1-suppressed (AsODN-4) in tea plants after drought stress for 24 h, respectively. U) Working models of UGT85A53 and UGT1 from Cluster II in response to cold and drought stress response in tea plants. Images in C), D), K), and L) were digitally extracted for comparison. Data were expressed as the mean ± Sd from at least 3 biological replicates. All statistical analysis was performed by Student's t-tests.

Figure 5.

Functional validation of UGT genes in Cluster III from G group. A) LC–MS/MS analyses of enzymatic products formed by UGT2. The mass spectra of the enzymatic reaction products produced by recombinant UGT2 are listed. B) LC–MS/MS analyses of enzymatic products formed by UGT3. The mass spectra of the enzymatic reaction products produced by recombinant UGT3 are listed. C) Chlorophyll fluorescence images of control (sODN-5) and UGT2-suppressed (AsODN-5) in tea plants under 4 ℃ for 48 h. The color-coded values indicate the degree of damage, with smaller values indicating more severe damage. D) Chlorophyll fluorescence images of control (sODN-6) and UGT3-suppressed (AsODN-6) in tea plants under 4 ℃ for 48 h. E, G) Relative expression of control (sODN-5/sODN-6) and UGT2-suppressed (AsODN-5)/UGT3-suppressed (AsODN-6) in tea plants, respectively. F, H) Statistical analysis of maximum photochemical efficiency of photosystem II (F_v/F_m) in C), D). I, and J) The relative content of HMF-glucoside/perillyl alcohol-glucoside of control (sODN-5) and UGT2-suppressed (AsODN-5) in tea plants, respectively. K) The relative content of phytol-glucoside of control (sODN-6) and UGT3-suppressed (AsODN-6) in tea plants. L, M) Relative content of hormones in control (sODN-5/sODN-6) and UGT2-suppressed (AsODN-5)/UGT3-suppressed (AsODN-6) in tea plants under 4 ℃, respectively. N, O) The tea plant phenotypes of control (sODN-7/sODN-8) and UGT2-suppressed (AsODN-7)/UGT3-suppressed (AsODN-8) after drought stress for 24 h, respectively. P, U) Relative expression of control (sODN-7/sODN-8) and UGT2-suppressed (AsODN-7)/UGT3-suppressed (AsODN-8) in tea plants, respectively. Q, V) Leaf inclination of tea plants in control (sODN-7/sODN-8) and UGT2-suppressed (AsODN-7)/UGT3-suppressed (AsODN-8) after drought stress for 24 h, respectively. R, W) Relative moisture content of tea plants in control (sODN-7/sODN-8) and UGT2-suppressed (AsODN-7)/UGT3-suppressed (AsODN-8) after drought stress for 24 h, respectively. S, T) The relative content of HMF-glucoside/perillyl alcohol-glucoside of control (sODN-7) and UGT2-suppressed (AsODN-7) in tea plants, respectively. X, Y) Relative content of hormones in control (sODN-7/sODN-8) and UGT2-suppressed (AsODN-7)/UGT3-suppressed (AsODN-8) in tea plants after drought stress for 24 h, respectively. Z) Working models of UGT2 and UGT3 from Cluster III in response to cold and drought stress response in tea plants. Images in C), D), N), and O) were digitally extracted for comparison. Data were expressed as the mean ± Sd from at least 3 biological replicates. All statistical analysis was performed by Student's t-tests.