We generated triploid hybrids of I. trifida and I. tabascana, providing insights into sweetpotato evolution and bridge germplasm for genetic improvement. Sweetpotato [Ipomoea batatas (L.) Lam.] (2n = 6x = 90) is an important root crop whose genomic polyploidisation process remains incompletely understood and highly complex. Its wild relatives are favoured for the study of genome formation, polyploidy processes and genetic improvement in sweetpotato. In this study, we obtained two triploid hybrids via reciprocal crosses between the closest diploid (I. trifida, 2n = 2x = 30) and the closer tetraploid (I. tabascana, 2n = 4x = 60) wild relatives of sweetpotato. To verify the authenticity of these two hybrids, molecular marker analysis, flow cytometry analysis, chromosome counting, morphological characterisation, pollen viability testing and meiotic observation were performed. Through InDel marker analysis, nine primer pairs consistently amplified parent-specific bands in the hybrids. Ploidy characterisation showed that both hybrids were triploid with 45 chromosomes. Morphologically, the two hybrids inherited many traits from both parents whilst producing some variation, such as both round and pentagonal corollas and purple stigmas with two or three lobes. Some pollen grains from the hybrids were fertile. The hybrids showed many abnormalities during meiosis, such as lagging chromosomes and uneven chromosome distribution. The triploid hybrids produced in this study can be used to infer early triploid genomic events during the origin and evolution of sweetpotato and to develop bridge germplasm for its genetic improvement.

Sweetpotato (Ipomoea batatas (L.) Lam.) is an important multifunctional crop with great value in food supply, industrial processing and bioenergy utilization. Crude protein content (CPC) is a core target trait for sweetpotato quality breeding. To dissect the genetic basis of CPC and identify key candidate genes, we used an F1 population of 212 individuals. CPC was measured by near-infrared reflectance spectroscopy (NIRS) in 2020 and 2021, and QTL mapping was performed using a high-density SNP genetic linkage map. Candidate genes were explored via a genome-wide association study (GWAS), multiple-database functional annotation, and quantitative real-time PCR (qPCR) validation. The results showed that: (1) CPC in the population exhibited a continuous normal distribution with high inter-year stability, and phenotypic variation was mainly controlled by genetic factors; (2) one stable minor-effect QTL for CPC, qCPC09-1, was mapped to Chr09: 7906895-8614924 bp, explaining 5.7% of phenotypic variation; (3) GWAS detected no significant SNP loci, suggesting that CPC is regulated by multiple minor-effect genes; (4) genes within the qCPC09-1 interval were significantly enriched in three protein synthesis-related KEGG pathways: ribosome, nitrogen metabolism and ubiquinone and other terpenoid-quinone biosynthesis; (5) qPCR verified that itf09g13420 and itf09g13230 were upregulated in the low-CPC parent Yushu 10 and negatively correlated with CPC, while itf09g13550 was upregulated in the high-CPC parent Xin 24 and positively correlated with CPC. These three genes exhibited expression patterns highly consistent with phenotypic differences. This study provides a theoretical basis and technical support for molecular marker-assisted breeding and elite germplasm innovation in sweetpotato.

Starch, a key biological macromolecule accounting for 50-80% of dry weight in sweetpotato (Ipomoea batatas [L.] Lam.) storage roots, underpins food and industrial applications. However, sweetpotato starch characterization is limited by local-sectioning approaches that fail to capture the whole-root granule dynamics. Here, we established a new morphological observation system covering three key root regions based on two representative cultivars: Okinawa 100 (V100), and Yanshu25 (Y25). It was effective and convenient for in situ starch observation and analysis in sweetpotato roots. The whole-root in situ microscopy, starch physicochemical profiling, and transcriptomic correlation were integrated to resolve starch dynamics in Y25 and V100. We identified widespread simple starch granules (SSGs)-compound starch granule (CSG) coexistence across the whole root tissues, with Y25 exhibiting programmed CSG fragmentation driven by ARCs/FtsZ-mediated amyloplast envelope destabilization and concomitant AMY/BMY upregulation. Y25 had a higher amylose content and a higher proportion of medium/long chains, but the average degree of polymerization was slightly lower. Transcriptomic analyses revealed that the differentially expressed genes were annotated in pathways of carbohydrate metabolism, and the differentially expressed genes in the starch metabolism pathway were analyzed. Weighted gene co-expression network analysis further identified the hub genes from different modules and analyzed the co-expression networks. This work will not only advance the understanding of starch granule assembly and remodeling in sweetpotato, but also provide a robust methodological and transcriptome-guided framework for starch-focused germplasm screening and quality improvement.

Abstract Background The development of sustainable agriculture requires the creation of efficient biostimulants that enhance crop yield and nutritional quality while minimizing environmental impact. Biomass-derived carbon dots, with their unique physicochemical properties and low toxicity, have emerged as promising candidates. Results This study synthesized ligustrum lucidum carbon dots (LL-CDs) and evaluated their biological effects on vegetable sweetpotato over short-term periods (7 and 14 days). Foliar application of LL-CDs significantly promoted plant growth, biomass accumulation, photosynthetic efficiency, nutrient uptake, and antioxidant capacity. Transcriptomic and metabolomic analyses indicated that LL-CDs treatment synergistically upregulated photosynthesis-related pathways, including light capture, electron transport, and carbon fixation, and increased the accumulation of bioactive compounds. Additionally, LL-CDs improved soil nutrient availability and positively influenced the rhizosphere microbial community structure. Conclusions These preliminary results suggest that LL-CDs have the potential to be developed as agricultural nanomaterials, providing initial theoretical support and technical reference for the development of novel biostimulants.

To uncover the variation patterns of the nutritional components in sweetpotato storage roots during long-term storage comprehensively, the general nutrients, phytochemicals, and starch properties of nine sweetpotato varieties with different flesh colors were quantified and analyzed by chemical and physical techniques. During the storage, the starch content decreased firstly and then increased, with sugar content the opposite. The crude protein content and the total dietary fiber content both increased continuously. The β-carotene content decreased or kept constant, while the anthocyanin content showed different variation patterns in the three purple-fleshed varieties. The four types of polyphenols and two types of flavonoids showed no obvious content changes during the storage. The amylose contents of all varieties showed various patterns, while the crystallinity was C-type. The proportion of small-sized starch granules reduced, and the combined proportion of medium-sized and large-sized granules increased. New correlations among the nutritional parameters for each variety were revealed for the first time. Principal component analysis indicated that the orange-fleshed varieties were distinguished from other varieties. Finally, the most storage-resistant variety ZZ3 and the suitable variety for each quality trait was selected. This study provides not only theoretical basis for comprehensive understanding of the nutrient’s variations in sweetpotato storage roots during long-term storage, but also guidelines for evaluation of nutritional quality of sweetpotato roots during storage and improvement of storage methods.

The National Sweetpotato Genebank in Vitro (Xuzhou, China) recently collected 1183 sweetpotato accessions, including potential duplicates. To assess genetic diversity and identify redundant accessions, we conducted comprehensive analyses using seven SSR (Simple Sequence Repeat) marker pairs coupled with agronomic trait evaluation. These markers demonstrated high polymorphism, with an average PIC (Polymorphic Information Content) value of 0.7821. Genetic distance-based clustering revealed the genetic similarity among accessions originating from the same geographic regions. Molecular analysis identified 543 duplicates (each sharing zero genetic distance with at least one counterpart) among the 1183 accessions, while 640 accessions were confirmed as genetically unique. Further classification of the 543 duplicates resulted in 130 distinct genetic groups based on zero genetic distance. Subsequent phenotypic characterization, including evaluations of skin color, flesh color, and aboveground morphological traits, revealed 190 additional unique accessions within these groups. Combining these with the initial 640 unique accessions yielded a final collection of 830 genetically distinct germplasm resources. This study significantly enhances germplasm identification efficiency for conservation purposes, provides valuable insights into the genetic diversity and geographical distribution patterns of sweetpotato germplasm, and underscores the importance of targeted collection efforts in regions exhibiting high genetic diversity.

Leaf color mutants are valuable resources for studying photosynthesis, pigment metabolism, and gene regulatory networks in plants. In this study, a naturally occurring sweetpotato albino mutant exhibiting a stable white-leaf phenotype across developmental stages was identified and compared with its green-leaf wild type to elucidate the molecular mechanisms underlying albinism. The mutant showed a dramatic 98.8% reduction in total chlorophyll content and a markedly decreased Fv/Fm value (0.59), indicating severe impairment of PSII efficiency. Integrated transcriptomic analysis identified 3520 differentially expressed genes (DEGs), while metabolomic profiling revealed 270 differentially accumulated metabolites (DAMs). Genes involved in chlorophyll and carotenoid biosynthesis, chloroplast development, and photosynthetic electron transport were strongly repressed, including key regulators such as GLK1, PORA, and PORB. Metabolomic alterations were mainly enriched in flavonoids, phenylpropanoids, and amino acid-derived pathways, reflecting broad reprogramming of both primary and secondary metabolism. These changes were accompanied by severely disrupted chloroplast ultrastructure, suggesting a primary defect in plastid development. Collectively, the integrated multi-omics evidence provides a comprehensive understanding of the coordinated transcriptional and metabolic alterations driving the albino phenotype in sweetpotato and establishes this mutant as a potential model for studying the interplay between chloroplast biogenesis, photosynthesis, and secondary metabolism.

[Objective]The purpose of this study was to analyze the quantitative trait loci(QTLs)related to stem nematode resistance in sweetpotato(Ipomoea batatas(L.)Lam.),lay a foundation for the fine mapping,cloning,and functional analysis of stem nematode resistance genes in sweetpotato.It also aimed to provide support for the study of the genetic mechanisms of stem nematode resistance,as well as the breeding of resistant varieties in sweetpotato.[Method]An F1 population of 212 progenies derived from a cross between the highly resistant cultivar Yushu 10 and the susceptible line Xin 24 was used.In this study,field-based natural infection assays were conducted to evaluate stem nematode resistance.QTL mapping was performed using composite interval mapping(CIM),and candidate genes within QTL confidence intervals were predicted.Additionally,genome-wide association studies(GWAS)were carried out using the rMVP software(a memory-efficient,visualization-enhanced,and parallel-accelerated R package).For further validation,the resistant cultivar Zhenghong 22 was artificially inoculated with D.destructor,and samples were collected at different post-inoculation time points.The expression patterns of five candidate genes(itf02g19880,itf02g20080,itf02g20100,itf13g18480 and itf13g18550)were analyzed via qRT-PCR.[Result]Three QTLs(qSNR02-1,qSNR02-2,and qSNR13-1)were identified,distributed on chromosomes 2 and 13.The phenotypic contribution rate of the individual QTL related to stem nematode resistance ranged from 9.6％to 11.7％.GWAS revealed one significantly associated locus with stem nematode resistance on chromosome 6.Based on the genomic annotation information,36 candidate genes related to stem nematode resistance were predicted within the QTL confidence intervals,including members of the ABC transporter family,multidrug and toxic compound extrusion(MATE)efflux proteins,E3 ubiquitin ligases,and glutathione S-transferases,which are involved in defense mechanisms,post-translational modification,and stress response.qRT-PCR results showed that the expression patterns of the five candidate genes were significantly different.The expression level of itf02g20100 reached a peak at 3 days after inoculation,which was 6.2 times that of the control;the expression level of itf02g19880 increased sharply and reached the highest level at 0.5 days after inoculation,which was 43.2 times that of the control;the expression patterns of itf13g18480 and itf13g18550 were similar,and both reached a peak at 7 days after inoculation.This indicates that different candidate genes may play different regulatory roles in the defense response after Ditylenchus destructor infection.[Conclusion]Three QTLs related to stem nematode resistance in sweetpotato were identified,and 36 related candidate genes were screened out,which can be used for the subsequent cloning and functional study of genes related to stem nematode resistance in sweetpotato.

The sweetpotato (Ipomoea batatas [L.] Lam) is a globally significant crop, valued for its nutritional and economic importance. The tuberous roots of the sweetpotato are rich in carotenoids, which contribute to their vibrant colors and health benefits. This study focuses on three elite fresh-consumption sweetpotato cultivars: “Kokei No. 14,” “Xinxiang,” and “Zheshu81” with distinct flesh colors. To elucidate the metabolic pathways and genetic mechanisms underlying carotenoid biosynthesis in the sweetpotato, 20 types of carotenoids were quantified using targeted metabolomic analyses, and the key genes involved in carotenoid synthesis were identified with transcriptomic analyses. The results revealed significant differences in carotenoid content and composition among the cultivars, with “Zheshu81” exhibiting the highest carotenoid levels. Weighted gene co-expression network analysis further highlighted key regulatory genes and transcription factors influencing carotenoid accumulation. This study identifies key transcriptional regulators associated with carotenoid accumulation, sheds light on sweetpotato carotenoid biosynthesis mechanisms, and lays a foundation for breeding to improve its nutritional quality and flesh color.

The lack of genomic and genetic research on wild sweetpotato species has hindered the advancement of sweetpotato variety development through modern crop improvement techniques. To facilitate the use of genomic and genetic approaches in sweetpotato variety development, we conducted a comprehensive assessment of the genome size and ploidy of six closely related wild sweetpotato species using flow cytometry and chromosome counting. Additionally, we acquired insights into their genomic characteristics through high-throughput sequencing. Based on the 17-mer frequency distribution, the genome sizes of these species ranged from 518.47 Mb to 1,505.04 Mb. Notably, most diploid species exhibited genome sizes of approximately 500 Mb, with the diploid wild species I. purpurea standing out as having a significantly larger genome size compared to other diploid species. A substantial proportion of repeats (ranging from 57.47 to 81.07

Sweet potato (Ipomoea batatas L.) is a crop native to the tropics that is weakly resistant to cold. Low temperatures are important factors limiting its storage. By analyzing the phenotypic and tuberous root color differences between Xushu 32 and Yanshu 25 under low-temperature treatment, we found that Xushu 32 showed less low-temperature damage. Transcriptome analysis was performed on the tuberous roots of the two sweet potatoes at different storage times to further investigate the molecular mechanism. There were 7039 differentially co-expressed genes in Xushu 32 and 11033 differentially co-expressed genes in Yanshu 25 after 4 and 8 days of low-temperature treatment. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses showed that the percentage of differentially expressed genes in the glutathione metabolism pathway in Xushu 32 was significantly higher than that in Yanshu 25. Weighted correlation network analysis (WGCNA) analysis indicated that the expression modules of MElightcyan1 and MEmegenta3 were up-regulated under low-temperature stress and that glutathione S-transferases (GSTs) were hub genes with high connectivity. The candidate genes IbGST2 and IbGST4 were highly similar to a rice glutathione S-transferase (GST) (Os10g0528900) by heatmap analysis and evolutionary tree analysis and may play a role in low-temperature stress. Subsequently, gene-silenced plants of the homologous genes SlBI-GST and SlGST-T4 in tomatoes were constructed using virus-induced gene silencing (VIGS) technology, and the contents of reactive oxygen species (ROS) metabolites and expression of BI-GST, GST-T4, and the C-repeat binding transcription factors (CBFs) were determined. This suggested that BI- GST and GST- T4 genes could scavenge ROS and improve the cold tolerance of tomatoes. These results provide important insights into the roles of IbGST2 and IbGST4 in the storage characteristics of sweet potatoes under low-temperature stress.

Sweetpotato, Ipomoea batatas (L.) Lam., is an important worldwide crop used as feed, food, and fuel. However, its polyploidy, high heterozygosity and self-incompatibility makes it difficult to study its genetics and genomics. Longest vine length (LVL), yield per plant (YPP), dry matter content (DMC), starch content (SC), soluble sugar content (SSC), and carotenoid content (CC) are some of the major agronomic traits being used to evaluate sweetpotato. However limited research has actually examined how these traits are inherited. Therefore, after selecting 212 F1 from a Xin24 × Yushu10 crossing as the mapping population, this study applied specific-locus amplified fragment sequencing (SLAF-seq), at an average sequencing depth of 26.73× (parents) and 52.25× (progeny), to detect single nucleotide polymorphisms (SNPs). This approach generated an integrated genetic map of length 2441.56 cM and a mean distance of 0.51 cM between adjacent markers, encompassing 15 linkage groups (LGs). Based on the linkage map, 26 quantitative trait loci (QTLs), comprising six QTLs for LVL, six QTLs for YPP, ten QTLs for DMC, one QTL for SC, one QTL for SSC, and two QTLs for CC, were identified. Each of these QTLs explained 6.3 to 10% of the phenotypic variation. It is expected that the findings will be of benefit for marker-assisted breeding and gene cloning of sweetpotato.

BACKGROUND:The utilization of sweetpotato starch in the food industry is significantly influenced by the granule size of the starch. To isolate sweetpotato starch fractions with different sizes, an efficient isolation method is in demand. The differences in thermal properties of starch fractions with different sizes from various sweetpotato varieties were revealed insufficiently. RESULTS:In this study, we devised a time-saving isolation technique to effectively isolate sweetpotato starch fractions based on granule sizes. The new technique was proved applicable for sweetpotato varieties with different flesh colors. The amylose contents of the isolated starch fractions were in the range 16.49-23.27%. A positive association was observed between amylose content, relative crystallinity of starch fractions and their granule size. Conversely, both the swelling power and water solubility at 95 °C displayed a consistent decline from more than 30 g g-1 to lower than 20 g g-1 as the granule size increased. Tp, To and Tc decreased gradually with an increase of starch granule size, while the medium- or small-sized starch fractions showed higher ΔH. In the first stage of thermogravimetric analysis curves, the weight of the small-sized starch fractions decreased the slowest, but no definite pattern was detected in the second or third stage. CONCLUSION:Therefore, the newly established technique and the results of this study will help better understand the properties of sweetpotato starch fractions with different sizes and certainly provide guidelines for the utilization of sweetpotato starch in food processing and product development. © 2024 Society of Chemical Industry.

The domain of unknown function 668 (DUF668) is a gene family that plays a vital role in responses to adversity coercion stresses in plant. However, the function of the DUF668 gene family is not fully understood in sweet potato. In this study, bioinformatics methods were used to analyze the number, physicochemical properties, evolution, structure, and promoter cis-acting elements of the IbDUF668 family genes, and RNA-seq and qRT–PCR were performed to detect gene expression and their regulation under hormonal and abiotic stress. A total of 14 IbDUF668 proteins were identified in sweet potato, distributed on nine chromosomes. By phylogenetic analysis, IbDUF668 proteins can be divided into two subfamilies. Transcriptome expression profiling revealed that many genes from DUF668 in sweet potato showed specificity and differential expression under cold, heat, drought, salt and hormones (ABA, GA3 and IAA). Four genes (IbDUF668-6, 7, 11 and 13) of sweet potato were significantly upregulated by qRT-PCR under ABA, drought and NaCl stress. Results suggest that the DUF668 gene family is involved in drought and salt tolerance in sweet potato, and it will further provide the basic information of DUF668 gene mechanisms in plants.

To evaluate scientifically the yield, adaptability, stability and testing sites discrimination and representativeness of varieties（lines） in leaf-vegetable sweetpotato joint identification trials in China, the AMMI model and GGE biplot were used to analyze the yield data of ten varieties（lines） in nine testing sites during 2018-2019. The productive and stable varieties suitable for large scale promotion as well as the testing sites with good discrimination and representativeness were screened. The results showed that the yield of leaf-vegetable sweetpotato reached highly significant in genotype effect, environment effect and interaction effect. Environment effect accounted for the largest proportion, followed by interaction effect, and genotype effect was the smallest. Xiangcaishu No.3 had higher productivity and stability, and it was the closest to the ideal variety, which was suitable for large area promotion. Qiancaishu No.2, Shulü No.2, EC15, Haida 7798 and Fucaishu 25 had better productivity, but had mediocre stability, so they were suitable for promotion in specific areas. The most suitable variety（lines） in Changsha, Hunan province was Shulü No.2, followed by Haida 7798 and Fucaishu 25. For Wuhan of Hubei, Chongqing, Qionghai of Hainan, Fuzhou of Fujian and Guangzhou of Guangdong, the most suitable variety（lines） was Qiancaishu No.2, followed by EC15. Qionghai of Hainan and Fuzhou of Fujian were the more discriminating and representative testing sites and more suitable as identification test sites for vegetable sweetpotato. In this study, AMMI and GGE analytical models were used simultaneously in vegetable sweetpotato yield analysis for the first time, and the application effects were demonstrated. And the varieties（lines） suitable for large-scale promotion and planting in specific areas were identified, and the best identification test points were determined, which provided a theoretical reference for the rational promotion of new leaf-vegetable sweetpotato varieties and scientific regional test.

Sweetpotato is an important crop that exhibits hexaploidy and high heterozygosity, which limits gene mining for important agronomic traits. Here, 314 sweetpotato germplasm resources were deeply resequenced, and 4 599 509 SNPs and 846 654 InDels were generated, among which 196 124 SNPs were nonsynonymous and 9690 InDels were frameshifted. Based on the Indels, genome-wide marker primers were designed, and 3219 of 40 366 primer pairs were selected to construct the core InDel marker set. The molecular ID of 104 sweetpotato samples verified the availability of these primers. The sweetpotato population structures were then assessed through multiple approaches using SNPs, and diverse approaches demonstrated that population stratification was not obvious for most Chinese germplasm resources. As many as 20 important agronomic traits were evaluated, and a genome-wide association study was conducted on these traits. A total of 19 high-confidence loci were detected in both models. These loci included several candidate genes, such as IbMYB1, IbZEP1, and IbYABBY1, which might be involved in anthocyanin metabolism, carotenoid metabolism, and leaf morphogenesis, respectively. Among them, IbZEP1 and IbYABBY1 were first reported in sweetpotato. The variants in the promoter and the expression levels of IbZEP1 were significantly correlated with flesh color (orange or not orange) in sweetpotato. The expression levels of IbYABBY1 were also correlated with leaf shape. These results will assist in genetic and breeding studies in sweetpotato.

甘薯茎叶富含蛋白质、维生素、膳食纤维、矿物质等营养成分,具有抗氧化、调节血脂等多个保健功能,可作为叶菜、饲料来源.另外甘薯茎叶具有丰富多彩的叶色和变化多样的叶型,具有观赏和园林绿化等商业价值.甘薯茎叶研究基础较为薄弱,主要集中于营养、生理活性成分等研究,而分子遗传、加工利用等相关研究起步较晚.系统梳理了甘薯茎叶的基础研究进程,并从叶菜、观赏、食品加工、饲用等多个方面分析了甘薯茎叶的利用进展.针对研究与利用现状及存在的不足,提出了针对性的建议与展望,为后续甘薯茎叶相关研究及产业发展提供参考.

Background Sweetpotato is an important vegetable and food crop that is bred through sexual crosses and systematic selection. The use of in vitro germination of sweetpotato pollen to test its viability has important theoretical and practical implications for improving the efficiency of sweetpotato crossbreeding by controlling pollination and conducting research on sweetpotato pollen biology. Results In this study, we observed the morphological structure of sweetpotato pollen under a scanning electron microscope (SEM), developed an effective method for the in vitro germination of sweetpotato pollen, and examined the viability of sweetpotato pollen after treating plants at different temperatures before blossoming. Sweetpotato pollen grains are spherical, with an average diameter of 87.07 ± 3.27 μm (excluding spines), with multiple germination pores and reticulate pollen surface sculpture. We applied numerous media to sweetpotato pollen germination in vitro to screen the initial medium and optimised the medium components through single-factor design. The most effective liquid medium for in vitro sweetpotato pollen germination contained 50 g/L Sucrose, 50 g/L Polyethylene glycol 4000 (PEG4000), 100 mg/L Boric acid and 300 mg/L Calcium nitrate, with a pH = 6.0. The optimum growth temperature for pollen development in sweetpotato was from 25 to 30 °C. Neither staining nor in situ germination could accurately determine the viability of sweetpotato pollen. Conclusions In vitro germination can be used to effectively determine sweetpotato pollen viability. The best liquid medium for in vitro germination of sweetpotato pollen contained 50 g/L Sucrose, 50 g/L Polyethylene glycol 4000 (PEG4000), 100 mg/L Boric acid and 300 mg/L Calcium nitrate, with the pH adjusted to 6.0. This study provides a reliable medium for the detection of sweetpotato pollen viability, which can provide a theoretical reference for sweetpotato genetics and breeding.

Drought is a common environmental stress with great negative impacts on plant growth, development and geographical distribution as well as agriculture and food production. Sweet potato is characterized by starchy, fresh and pigmented tuber, and is regarded as the seventh most important food crop. However, there has been no comprehensive study of the drought tolerance mechanism of different sweet potato cultivars to date. Here, we studied the mechanism for drought response of seven sweet potato drought-tolerant cultivars using the drought coefficients, physiological indicators and transcriptome sequencing. The seven sweet potato cultivars were classified into four groups of drought tolerance performance. A large number of new genes and transcripts were identified, with an average of about 8000 new genes per sample. Alternative splicing events in sweet potato, which were dominated by first exon and last exon alternative splicing, were not conserved among different cultivars and not significantly affected by drought stress. Furthermore, different drought-tolerance mechanisms were revealed through differentially expressed gene analysis and functional annotation. Two drought-sensitive cultivars, Shangshu-9 and Xushu-22, mainly resisted drought stress by up-regulating plant signal transduction. The other drought-sensitive cultivar Jishu-26 responded to drought stress by down-regulating isoquinoline alkaloid biosynthesis and nitrogen/carbohydrate metabolism. In addition, the drought-tolerant cultivar Chaoshu-1 and drought-preferred cultivar Z15-1 only shared 9% of differentially expressed genes, as well as many opposite metabolic pathways in response to drought. They mainly regulated flavonoid and carbohydrate biosynthesis/metabolism in response to drought, while Z15-1 increased photosynthesis and carbon fixation capacity. The other drought-tolerant cultivar Xushu-18 responded to drought stress by regulating the isoquinoline alkaloid biosynthesis and nitrogen/carbohydrate metabolism. The extremely drought-tolerant cultivar Xuzi-8 was almost unaffected by drought stress and responded to drought environment only by regulating the cell wall. These findings provide important information for the selection of sweet potatoes for specific purposes.