ABSTRACT: Purpose: Ubiquitin-mediated proteolysis by the 26S proteasome is a critical regulatory mechanism controlling many biological processes. The multi-subunit Skp1-Cullin1-F-box (SCF) complexes are designed to select substrates for ubiquitylation in all eukaryotes. Within these complexes, the Skp1 protein is defined as the bridge between the F-box and the Cullin1 proteins. In Arabidopsis thaliana, there are 19 Arabidopsis Skp1-Like (ASK) genes that encode Skp1 proteins. Our previous genomics and evolutionary studies suggest that ASK1 plays a predominant role in the ASK family. To demonstrate this hypothesis, we genetically introduced the same Ds insertion allele of ask1-1 in Ler into the Arabidopsis Col-0 reference genome. The ask1-1 allele was initially obtained from an ask1-1(+/-)ask2-1(+/-)/LerWs double heterozygous mutant (Liu et al., 2004). To identify the breadth and depth of SCF functions, we applied RNA-Seq sequencing to systematically analyze the perturbation of transcriptomes in both floral bud and open flower tissues in this novel ask1 mutant. Method: Two replicates of total RNA were extracted from 10-50 mg of floral buds or open flowers from Col-0, ask1, and ASK1 (WT allele) primary inflorescences five hours after dawn grown under a long-day photoperiod condition. The NucleoSpin® RNA kit (Macherey-Nagel) was used for RNA extraction following the manufacturer’s instructions. 1 μg each from the 12 RNA samples was used for pair end 150-mer deep sequencing using a NovaSeq 6000 platform (Novogene). Adapter and low-quality sequences in the raw FASTQ sequencing data were removed to produce valid sequences using Trimmomatic (phred=33, minimum length=36; (Bolger et al., 2014)), which were further aligned to the A. thaliana Col-0 reference genome (TAIR10; www.arabidopsis.org) via TOPHAT2 (Kim et al., 2013a) to identify accepted hits of each locus. HTSeq was then used to calculate an absolute expression level (counts) of each locus based on the accepted hits (Anders et al., 2015). The resulting counts per locus were normalized to counts per million reads (CPM) among all 12 libraries and used to identify groups of SDEGs in bud and open flower tissues in ask1 and ASK1 through comparing to the WT bud and open flower tissues, respectively, by edgeR (Log2FC ≥ 1; FDR < 0.01%; (Robinson et al., 2010)). Results: In total, 52 and 54 SDEGs were identified to be up-regulated and down-regulated, respectively, in ASK1 BC10F4 bud tissues; and 13 and 3 SDEGs were identified to be up-regulated and down-regulated, respectively, in ASK1 BC10F4 open flower tissues (2-fold change (FC) with a false discovery rate (FDR) < 0.01). However, 306 and 941 SDEGs (in total 1,247) in ask1 bud tissues and 392 and 1,137 SDEGs (in total 1,529) in ask1 open flower tissues were identified to be up-regulated and down-regulated, respectively (2-FC and FDR < 0.01). Conclusions: Compared to Col-0, 527 and 640 SDEGs were identified in Ler and Ws, respectively, which were greatly higher than the number of SDEGs identified in ASK1 floral bud (Fisher’s exact test, p = 6.0e-82 and 7.5e-112, respectively) and open flower tissues (Fisher’s exact test, p = 7.4e-150 and 1.2e-186, respectively). Therefore, we concluded that the genetic background of ASK1 and ask1 are isogenic to Col-0. The large array of SDEGs identified in both floral buds and open flower tissues in ask1 demonstrated a wide range of roles of ASK1-containing SCF complexes.
