Introduction

Cassava (Manihot esculenta Crantz), belonging to the Euphorbiaceae family, is native to the southern Amazon basin and is widely cultivated in tropical and subtropical regions, ranking as the world’s sixth-largest crop [1]. Despite the starch-rich storage roots, frequent consumption can lead to hidden hunger due to low levels of other nutrients [2]. In recent years, leaves have gained recognition as a substitute source of protein and micronutrients in sub-Saharan Africa and parts of Asia [3]. Research indicates that cassava tuberous roots with different flesh colors exhibit varying nutritional profiles, with darker-colored cassava potentially harboring elevated levels of secondary metabolites and nutrients, particularly carotenoids and flavonoids [

Fig. 4
figure 4

RNA-seq analysis of differentially expressed transcripts in WT and OE leave. (A) Volcano plot of the number of DEGs in leaves. (B) Bar graph of the number of DEGs in leaves. (C) Heatmap of the DEGs in leaves. WT, wild-type cassava. OE, MeFtsZ2-1 overexpression cassava

GO classification analysis was performed on the DEGs of OEL/WTL (Fig. 5). In the OEL/WTL comparison, a majority of DEGs were categorized under defense response clusters in relation to biological processes. DEGs associated with cellular components were linked to cell nucleus, plasma membrane, cytoplasm, membrane components, and chloroplasts. The functionality of DEGs was primarily related to protein binding, ATP binding, and DNA binding transcription factor activity.

Fig. 5
figure 5

Gene ontology (GO) enrichment classification of differentially expressed genes in cassava leaves. All annotated unigenes were divided into three functional GO categories: biological process, cellular component and molecular function

KEGG pathway analysis showed that the main pathways of OEL/WTL included glycolysis, gluconeogenesis, citric acid metabolism, fatty acid degradation, glycolysis/gluconeogenesis, phenylpropionic acid biosynthesis, anthocyanin biosynthesis and carotenoid metabolism (Supplementary Figure S3A).

Identification of genes involved in chlorophylls, carotenoids and anthocyanins biosynthesis pathways

A comparison between OEL/WTL revealed a total of 26 differentially expressed genes (DEGs) within the chlorophylls, carotenoids and anthocyanins synthesis pathway (Supplement Table S3). We identified 6 DEGs associated with chlorophylls biosynthesis and metabolism pathways. Among these, transcription factor GLABRA 3 (GL3), Protein STAY-GREEN homolog (SGR), ABC transporter C family member 3 (ABCC3), transcription factor PIF1 (PIF1), and putative U-box domain-containing protein 42 (PUB42) exhibited higher expression levels in OEL. We identified six DEGs related to carotenoids biosynthesis in OEL, in which the expression of indole-3-acetaldehyde oxidase (AAO1) was up-regulated, 9-cis-epoxycarotenoid dioxygenase NCED3 (NCED3) was down-regulated (Fig. 6). The expression levels of genes related to the anthocyanins synthesis pathway, such as 4-coumarate–CoA ligase (4CL), Naringenin,2-oxoglutarate 3-dioxygenase (F3H), and UDP-glycosyltransferases (UGT), were significantly increased in OEL (Fig. 6).

Fig. 6
figure 6

Heatmaps of DEGs related to anthocyanidin and β-carotenoid biosynthesis in the WT and OE cassava. PAL, phenylalanine ammonia lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4coumaroyl CoA ligase; CHS, chalcone synthase; CHI, chalcone isomerase; F3H, flavanone 3-hydroxylase; DFR, dihydroflavonol-4-reductase; LDOX, leucoanthocyanidin dioxygenase; ANR, Anthocyanidin reductase; LAR, leucocyanidin reductase; UGT, UDP glycosyltransferase; PSY, phytoene synthase; PDS, phytoene desaturase; Z-ISO, ζ-carotene isomerase; ZDS, ζ-carotene desaturase; LYCE, lycopene epsilon-cyclase; BCH, β-carotene hydrolase; ZEP, zeaxanthin epoxidase; VDE, violaxanthin de-epoxidase; NCED, 9-cis-epoxycarotenoid dioxygenase; ABA2, xanthoxin dehydrogenase; Significant changes in gene expression levels are shown by the intensities of colors expressed in log2 values. WT, wild-type cassava. OE, MeFtsZ2-1 overexpression cassava

Identification of transcription factors (TFs)

In the transcriptomic databases of OEL and WTL, 49 differentially expressed genes (DEGs) encoding transcription factors (TFs) were identified and classified into 11 different TF families (Supplementary Figure S3B). The WRKY family had the highest number of DEGs (10 genes), followed by MYB (9 genes), bHLH (8 genes), AP2-ERF (5 genes), and TFIID (2 genes). These results indicate that the MeFtsZ2-1 gene can have multiple regulatory mechanisms in leaves.

DEGs associated with hormone signaling pathways

In OE, a total of 41 differentially expressed genes (DEGs) related to plant hormones were identified in leaves, respectively (Supplement Table S4 and Figure S4). Compared to the WT, within the auxin signaling pathway, auxin-induced proteins (AUX), auxin-responsive protein genes (SAUR/IAA), and auxin transporter protein (LAX) exhibited higher expression levels in the OEL. However, indole-3-acetic acid-amido synthetases (GH3.1/GH3.9) and auxin response factor (ARF) showed decreased expression in OEL. In the salicylic acid signaling pathway, the expression of transcription factor TGA (TGA1) and pathogenesis-related protein 1 (PRB1) was upregulated in OE. Within the gibberellin signaling pathway, chitin-inducible gibberellin-responsive protein 1 (CIGR1) was significantly downregulated in OEL, while scarecrow-like transcription factor PAT1 (PAT1) and transcription factor PIF1 (PIF1) were markedly upregulated. In the ABA signaling pathway, abscisic acid receptors (PYL) and serine/threonine-protein kinase (SRK2A) were upregulated, whereas probable protein phosphatase 2 C (PP2C51), transcription factor HHO (HHO2), and histidine kinase 5 (AHK5) were downregulated.

Quantitative real-time PCR (qRT-PCR) analysis

Ten randomly selected DEGs from both leaves and tuberous roots were subjected to qRT-PCR analysis to validate the reliability of the transcriptome sequencing data (Supplement Table S5). The qRT-PCR results were consistent with the transcriptomic data, indicating the high reliability of the transcriptome data (Fig. 7).

Fig. 7
figure 7

Ten candidate genes were randomly selected from leaves for qRT-PCR verification. All measurements were replicated three times. Vertical bars represent S.E. of the mean (n = 3). Data are mean ± SE of three individual experiments, each performed in triplicate (*: p < 0.05). WT, wild-type cassava. OE, MeFtsZ2-1 overexpression cassava