RNA-Sequencing Analysis of Abscission-related Transcriptome in Citrus Calyx Abscission Zone Reveals Distinguishing Profile of Phytohormone Expression Associated with Abscission of HLB-affected Sweet Orange

Citrus greening or huanglongbing (HLB) disease is caused by Candidatus liberibacter asiaticus (CLas) and associated with an increase in pre-harvest fruit drop, for which the molecular mechanisms remain unknown. In order to understand the molecular basis of the HLB-associated fruit abscission, by means of RNA-Sequencing analysis (RNA-Seq), transcriptomes in citrus calyx abscission zones were analyzed and compared among fruit dropped (D) or retained (R) from healthy (h) or HLB-diseased (d) trees upon shaking the trees. Cluster analysis based on the transcript reads indicates that dropped fruit from HLB-diseased trees (Dd) have largest distances from all other groups. Differentially expressed genes (DEGs) were identified between Dd and Rd, Dh and Rh, Rd and Rh, Dd and Dh. Wilcoxon test of the whole dataset of DEGs revealed that consistently up-regulated genes in Dd versus Rd and Dd versus Dh are in the functional categories of “secondary metabolism, lipid metabolism and hormone-ethylene and –jasmonate”; while those down-regulated genes didn’t show clear pattern of regulation. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses indicated that the significant biological processes or pathways involved in HLB-related fruit abscission were those related to defense response, secondary metabolism and hormone signaling. Among them, “response to chitin” was the most significant (p= 9.95E-13) biological process, and “jasmonic acid biosynthesis” was the most significant (p= 4.6E-5) pathway. Genes related to synthesis and signaling of ethylene (ET) and jasmonate (JA) were consistently up-regulated, while abscisic acid, auxin, brassinosteroid, cytokinin, and gibberellin were generally down-regulated in Dd; but not in Dh. Consistent with the transcriptomic data, fruit ethylene production was detected in two third of the Dd fruit, but none of Rd, Dh or Rh fruit. And, in agreement with the hormone expression profiles, substantial numbers of downstream JA/ET-responsive defense related genes were up-regulated in Dd, but not in Dh. Thirty representative DEGs covering categories of hormone, secondary metabolism, and JA/ET responsive defense responses were verified by qRT-PCR. The results indicate that HLB-associated pre-harvest fruit abscission is mediated by JA/ET signaling, which has been known to be triggered by infection of necrotrophic pathogens. Overall design: Six-year old ‘Hamlin’ orange trees (Citrus sinensis (L.) Osbeck), about 2.5-3.0 m tall, on 'Swingle' citrumelo (C. paradisi Macf. × Poncirus trifoliata (L) Raf.) rootstock, in a commercial grove located in Southern Florida. Thirty six trees were selected for the experiment, of which eighteen were CLas negative (healthy) and the other eighteen were CLas positive (HLB), as tested by qPCR using the method of Li et al. (2006). The selected trees were similar in size, and all were grown under similar agro-climatic conditions and received common cultural practices and the grower’s standard pest and disease management. Fruit were harvested on 1 Dec. 2014 (during commercial harvest season). The ground under the trees was cleaned just before shaking the trees, and trees were shaken manually. For HLB-diseased trees (d), many of the fruit dropped (Dd) upon shaking the trees. For healthy trees (h), the forces of shaking were strong enough to get enough number of dropped fruit (Dh) from the trees. The dropped fruit (Dd or Dh) were collected, and the retained fruit on HLB-diseased (Rd) or healthy trees (Rh) were hand-picked off the trees (Fig. 1A). The fruit from nine trees were pooled together as a group. There were two groups (biological replicates) for each of the Dd, Rd, Dh and Rh. Thirty fruit were randomly picked from each of the Dd, Rd, Dh and Rh groups, and the AZ-C were excised (Fig.1B) and immediately frozen in liquid nitrogen and stored at –80 ºC for RNA isolation. Another thirty fruit from each of the groups were transported to the laboratory where fruit ethylene production was measured.