Litchi possesses unique flower morphology and adaptive reproduction strategies. Although previous attention has been intensively devoted to the mechanisms underlying its floral induction, the molecular basis of flower sex determination remains largely unknown. MADS-box genes are promising candidates for this due to their significant roles in various aspects of inflorescence and flower organogenesis. Here, we present a detailed overview of phylogeny and expression profiles of 101 MADS-box genes that were identified in litchi. These LcMADSs are unevenly located across the 15 chromosomes and can be divided into type I and type II genes. Fifty type I MADS-box genes are subdivided into Mα, Mβ and Mγ subgroups, while fifty-one type II LcMADSs consist of 37 MIKC^C -type and 14 MIKC *-type genes. Promoters of both types of LcMADS genes contain mainly ABA and MeJA response elements. Tissue-specific and development-related expression analysis reveal that LcMADS51 could be positively involved in litchi carpel formation, while six MADS-box genes, including LcMADS42/46/47/75/93/100, play a possible role in stamen development. GA is positively involved in the sex determination of litchi flowers by regulating the expression of LcMADS51 (LcSTK). However, JA down-regulates the expression of floral organ identity genes, suggesting a negative role in litchi flower development.

Litchi florescence has large flower spikes and volume; reasonable control of the ratio of male to female litchi flowers is the key operational aspect of litchi orchards for preserving quality and increasing production. To achieve the rapid detection of male and female litchi flowers, reduce manual statistical errors, and meet the demand for accurate fertilizer regulation, an intelligent detection method for male and female litchi flowers suitable for deployment to low-power embedded platforms is proposed. The method uses multi-teacher pre-activation feature distillation (MPFD) and chooses the relatively complex YOLOv4 and YOLOv5-l as the teacher models and the relatively simple YOLOv4-Tiny as the student model. By dynamically learning the intermediate feature knowledge of the different teacher models, the student model can improve its detection performance by meeting the embedded platform application requirements such as low power consumption and real-time performance. The main objectives of this study are as follows: optimize the distillation position before the activation function (pre-activation) to reduce the feature distillation loss; use the LogCosh-Squared function as the distillation distance loss function to improve distillation performance; adopt the margin-activation method to improve the features of the teacher model passed to the student model; and propose to adopt the Convolution and Group Normalization (Conv-GN) structure for the feature transformation of the student model to prevent effective information loss. Moreover, the distilled student model is quantified and ported for deployment to a field-programmable gate array (FPGA)-embedded platform to design and implement a fast, intelligent detection system for male and female litchi flowers. The experimental results show that compared with an undistilled student model, the mAP of the student model obtained after MPFD feature distillation is improved by 4.42 to 94.21%; the size of the detection model ported and deployed to the FPGA-embedded platform is 5.91 MB, and the power consumption is only 10 W, which is 73.85% and 94.54% lower than that of the detection models on the server and PC platforms, respectively, and it can better meet the application requirements of rapid detection and accurate statistics of male and female litchi flowers.

Litchi (Litchi chinensis Sonn.) is an economically important evergreen fruit tree widely cultivated in subtropical areas. Low temperature is absolutely required for floral induction of litchi, but its molecular mechanism is not fully understood. Leaves of litchi played a key role during floral induction and could be the site of low temperature perception. Therefore, leaves were treated under different temperature (15 °C/25 °C), and high-throughput RNA sequencing (RNA-Seq) performed with leaf samples for the de novo assembly and digital gene expression (DGE) profiling analyses to investigate low temperature-induced gene expression changes. 83,107 RNA-Seq unigenes were de novo assembled with a mean length of 1221 bp and approximately 61% of these unigenes (50,345) were annotated against public protein databases. Differentially-expressed genes (DEGs) under low temperature treatment in comparison with the control group were the main focus of our study. Hierarchical clustering analysis arranged 2755 DEGs into eight groups with three significant expression clusters (p-value ≤ 0.05) during floral induction. With the increasing contents of sugars and starch, the expression of genes involved in metabolism of sugars increased dramatically after low temperature induction. One FT gene (Unigene0025396, LcFT1) which produces a protein called ‘florigen’ was also detected among DEGs of litchi. LcFT1 exhibited an apparent specific tissue and its expression was highly increased after low temperature induction, GUS staining results also showed GUS activity driven by LcFT1 gene promoter can be induced by low temperature, which indicated LcFT1 probably played a pivotal role in the floral induction of litchi under low temperature. Our study provides a global survey of transcriptomes to better understand the molecular mechanisms underlying changes of leaves in response to low temperature induction in litchi. The analyses of transcriptome profiles and physiological indicators will help us study the complicated metabolism of floral induction in the subtropic evergreen plants.

Titanium dioxide nanoparticles (nTiO₂) are widely used as fertilizers in agricultural production because they promote photosynthesis and strong adhesion. Low pollination and fertilization due to rainy weather during the litchi plant’s flowering phase result in poor fruit quality and output. nTiO₂ would affect litchi during the flowering and fruiting stages. This study considers how nTiO₂ affects litchi’s fruit quality and pollen viability during the flowering stage. The effects of nTiO₂ treatment on pollen vigor, yield, and fruit quality were investigated. nTiO₂ effectively improved the pollen germination rate and pollen tube length of litchi male flowers. The germination rate reached 22.31 ± 1.70%, and the pollen tube reached 237.66 μm in the 450 mg/L reagent-treated group. Spraying with 150 mg/L of nTiO₂ increased the germination rate of pollen by 2.67% and 3.67% for two types of male flowers (M1 and M2) of anthesis, respectively. After nTiO₂ spraying, the fruit set rates of ‘Guiwei’ and ‘Nomici’ were 46.68% and 30.33%, respectively, higher than those of the boric acid treatment group and the control group. The edibility rate, titration calculation, and vitamin C of nTiO₂ treatment were significantly higher than those of the control. The nTiO₂-treated litchi fruit was more vividly colored. Meanwhile, the adhesion of nTiO₂ to leaves was effectively optimized by using ATP and BCS to form nTiO₂ carriers and configuring nTiO₂ complex reagents. These results set the foundation for future applications of titanium dioxide nanoparticles as fertilizers for agriculture and guide their application to flowers and fruits.

Prolonged exposure to cold temperatures often results in a relatively low flowering rate in litchi (Litchi chinensis Sonn.) trees with younger leaves. This study aimed to verify the impact of stem girdling on litchi flowering by identifying and characterizing the induced metabolic changes. After a 60 day exposure to cold treatment at 15 °C/10 °C (12 h/12 h), the flowering rate of the girdled trees was 100%, while that of the non-girdled trees was 20%, indicating that girdling improved litchi flowering at its turning stage. The metabolic profiles of litchi leaves with and without stem girdling during floral induction were compared and 505 metabolites potentially associated with litchi flowering were detected. Most metabolites were involved in the metabolism of starch and sucrose, fatty acid, and phenylpyruvic acid. The metabolic pathways concerned with the biosynthesis of epinephrine, sucrose, and d-maltose were induced in leaves after girdling treatment. The level of galactitol, phenylpyruvic acid, acetyl-CoA, linoleic acid, alpha-linolenic acid, and 13-HPOT biosynthesis remained stable in the leaves from girdled trees but changed drastically in the leaves from non-girdled trees. In addition, 379 metabolites concerning flowering rate were characterized. Metabolism pathways of starch and sucrose, galactose, and linoleic acid are of great significance to the flowering of litchi. Linoleic acid exhibited the most significant variations between girdled trees and non-girdled trees with fold changes of up to 13.62. These results contribute to understanding the biological mechanism of litchi floral induction and the metabolic changes after stem girdling.

The relative content of chlorophyll, assessed through the soil and plant analyzer development (SPAD), serves as a reliable indicator reflecting crop photosynthesis and the nutritional status during crop growth and development. In this study, we employed machine learning methods utilizing unmanned aerial vehicle (UAV) multi-spectrum remote sensing to predict the SPAD value of litchi fruit. Input features consisted of various vegetation indices and texture features during distinct growth periods, and to streamline the feature set, the full subset regression algorithm was applied for dimensionality reduction. Our findings revealed the superiority of stacking models over individual models. During the litchi fruit development period, the stacking model, incorporating vegetation indices and texture features, demonstrated a validation set coefficient of determination (R²) of 0.94, a root mean square error (RMSE) of 2.4, and a relative percent deviation (RPD) of 3.0. Similarly, in the combined litchi growing period and autumn shoot period, the optimal model for estimating litchi SPAD was the stacking model based on vegetation indices and texture features, yielding a validation set R², RMSE, and RPD of 0.84, 3.9, and 1.9, respectively. This study furnishes data support for the precise estimation of litchi SPAD across different periods through varied combinations of independent variables.