Pear (Pyrus L.) is an important commercial fruit in the world. The fruit size is one of the important characters in fruit quality. The previous research reported that the fruit size of pear was mainly caused by the number of cell in about 40 days after blossom (DAB) in nature. However, studies about the mechanisms underlying cell division in young fruit development are very limited in pear. Two pear accessions codenamed ‘GH59B’ with big fruit and ‘GH81S’ with small fruit in three stages were sampled and the RNA-seq high-throughput sequencing was used to evaluate changes of gene transcription levels in the early stage of fruit development. The difference of cell size among two samples was little in 40 DAB, implying that the difference of the fruit size was caused by the number of the cell. More than 274,517,982 high quality reads from six libraries of fruit development were sequenced. A total of 797 differentially expressed genes (DEGs) were identified. Three cytokinin dehydrogenase genes and two gibberellin 2-beta-dioxygenase gene were identified in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways related to zeatin and gibberellin. Their expression was upregulated at 20 DAB in ‘GH81S’ and at 30 DAB in ‘GH59B’, suggesting that the small fruit size might be related to the early degradation of cytokinin and gibberellin inducing a short period of cell division. A total of 38 DEGs of transcription factors were found and 23 DEGs including NAC, ERF and bHLH transcription factors were highly related with cytokinin dehydrogenase and gibberellin dioxygenase genes. Altogether, the results of the present study provide information from a comprehensive gene expression analysis and insight into the molecular mechanism underlying the difference of fruit size in Pyrus pyrifolia.

Sugar content is an important trait of fruits. The genetic background of fruits can affect their sugar content in different cultivars. The quantitative trait loci and genes related to sugar content during fruit ripening remain unclear. In this study, we performed quantitative trait locus (QTL) mapping of sugar content. Two QTLs (qSugar-LG6-Chr7 and qSugar-LG12-Chr3) were identified based on their total sugar contents. A total of 577 and 519 genes were annotated around these two QTL loci. The contents of fructose, sorbitol, glucose, and sucrose were measured at six time points in four cultivars before fruit maturation, including two sweet cultivars (“Zaoshengxinshui” and “ZQ65”) and two general cultivars (“Qiushui” and “ZQ82”). In sweet cultivars, sucrose and fructose accumulate substantially, and sorbitol content decreases significantly during fruit ripening. A transcriptome analysis identified 125 upregulated and 222 downregulated differentially expressed genes (DEGs) in sweet cultivars. Two sucrose transport genes (PpSUT, LOC103964096, and LOC103940043) were negatively correlated with sugar content. A weighted gene co-expression network analysis showed that two genes, sorbitol dehydrogenase (PpSDH, LOC103960512 and LOC103960513), around the locus of qSugar-LG6-Chr7 were negatively co-expressed with the total sugar content, which was downregulated in the sweet cultivars. PpSDH and PpSUT may play important roles in regulating sugar content during pear ripening. Transcriptome analysis also revealed that some DEGs were related to sugars (PpS6PDH and ATP-PpPFK), hormones (PpARG7), and transcription factors (PpEMB1444, PpCYP734A1, and PpWRKY50). In conclusion, this study provides new insights into the molecular mechanisms associated with sugar content in the fruit ripening of Pyrus pyrifolia.

A large proportion of the genome of ‘Suli’ pear (Pyrus pyrifolia) contains long terminal repeat retrotransposons (LTR-RTs), which suggests that LTR-RTs have played important roles in the evolution of Pyrus. Further analysis of retrotransposons, particularly of high-copy-number LTR-RTs in different species, will provide new insights into the evolutionary history of Pyrus. A total of 4912 putative LTR-RTs classified into 198 subfamilies were identified in the ‘Suli’ pear genome. Six Asian pear accessions, including cultivars and wild species, were resequenced. The comparison of copy number for each LTR-RT subfamily was evaluated in Pyrus accessions, and data showed up to four-fold differences for some subfamilies. This contrast suggests different fates for retrotransposon families in the evolution of Pyrus. Fourteen high-copy-number subfamilies were identified in Asian pears, and more than 50% of the LTR-RTs in the genomes of all Pyrus accessions were from these 14 identified LTR-RT subfamilies. Their average insertion time was 3.42 million years ago, which suggests that these subfamilies were recently inserted into the genome. Many homologous and specific retrotransposon insertion sites were identified in oriental and occidental pears, suggesting that the duplication of retrotransposons has occurred throughout almost the entire origin and evolution of Pyrus species. The LTR-RTs show high heterogeneity, and their copy numbers vary in different Pyrus species. Thus, our findings suggest that LTR-RTs are an important source of genetic variation among Pyrus species.