**1. Introduction**

Anthracnose is a key leaf disease in the tea plant (*Camellia sinensis* (L.) O. Kuntze) induced by *Colletotrichums* [1,2], which has caused a great loss in tea yield and quality owing to its strong infectivity and widespread distribution [3]. Although there have been hundreds of *Colletotrichum* isolates detected in tea plant [4,5], the dominant species or major pathogen on the tea plant are considered to be

*Colletotrichum gloeosporioidesis*, *C. camelliae* and *C. fructicola* [3–5], which grow and spread quickly in humid and hot summer and autumn seasons, leading to gray sunken or shrunk necrotic lesions on tea leaves and twigs [6,7]. The loss of tea yield induced by anthracnose was estimated to range from 5% to 20% or even more, depending on the cultivar planted [8,9]. Although plant activators [10], antagonistic bacteria [11] and agronomic techniques such as leaf trimming and rational fertilization have been used to control the anthracnose disease, spraying fungicides is still a major control method in tea fields, resulting in a serious fungicides residue problem. Breeding tea cultivars resistant to anthracnose is considered to be the most effective measure to control anthracnose because there is a great difference in resistance to this disease between various tea cultivars [6,8,9].

Revealing molecular defense responses of tea plant to anthracnose infection and developing a molecular assisted selection (MAS) method are badly needed in the tea plant breeding field. However, little is known about the molecular mechanisms regulating the defense response in tea plants [6] although attempts were made to probe this. Transcriptional analysis and histochemistry revealed that the hypersensitive response (HR) and H2O2 play critical roles in tea plant defense response to *C. fructicola* [12], and chemical changes of caffeine is considered to be associated with tea–fungi interaction [13]. Nonpathogenic species of *Colletotrichum* was more vulnerable to catechins and caffeine, and differentiation in secondary metabolites might be an important factor leading to the difference in pathogenicity between cultivars [14]. Different communities of *Colletotrichum* with little variability within internal transcribed spacer (ITS) and glyceraldehyde 3-phosphate dehydrogenase (*GAPDH*) regions in the DNA sequence have different virulence [15].

Generally, it was reported that salicylic acid (SA), a secondary metabolite belonging to phenolic compounds, acts as a plant signal or hormone with a regulatory role in a variety of physiological processes under abiotic and biotic stresses, especially as a module in activating disease resistance [16–18]. The endogenous SA concentration is related to the conditions of stresses. For instance, SA accumulation, which is responsible for the phenylpropanoid synthesis pathway, was observed in wheat leaves after heavy metal cadmium treatment [19]. In tobacco (*Nicotiana tabacum* L. cv. *Xanthinc*), a transient sharp increase in SA induced by ultraviolet UV light, ozone and mosaic virus parallelly led to pathogenesis-related proteins accumulation and enhanced disease resistance [20]. The *Arabidopsis thaliana* mutants of *sid* and *pad4*, which are involved in synthesis of SA and camalexin (a kind of phytoalexin), were defective in SA synthesis, resulting in them being more susceptible to pathogens [21,22]. The defense compounds including pathogenesis-related proteins (PR-1, PR-2, PR-5), camalexin and H2O2 could be induced by endogenous SA [22–24]. Transgenic plants with salicylate hydroxylase, which converts SA to other compounds, accumulated almost no SA and lost the resistance to virus [25]. When blocked by an inhibitor of phenylpropanoid (precursor of SA) synthesis, the exogenous SA recovered the ability to plant resistance [23]. All these show that SA indeed plays the key role and an important signaling effect in plant resistance, especially in systemic acquired resistance [25,26]. Due to the fact that there has been no study revealing the relationship of SA to the response of tea plant to diseases, it will be interesting to reveal the responses of SA biosynthesis genes and molecular function of SA in the defense response of tea plant to anthracnose infection.

RNA-seq can be focused on assessing the degree of RNA processing and the types of RNA modification, both of which may play an important role in the disease infection process [27,28]. Information of RNA-seq can be used to profile gene expression levels and to reveal the genes involved in plant defense responses [29,30]. RNA-seq analysis needs no reference genome [31] even though tea plant genome data has been available [32]. Developing a transcriptome assembly dataset generating from anthracnose infected tea leaves will be helpful to reveal the molecular defense responses of tea plant to the anthracnose and also to mine molecular markers for MAS used in resistance breeding.

In the present study, transcriptome sequencing by Trinity was performed on the platform of Illumina HiSeq to construct a de novo transcriptome assembly database generated from anthracnose infected tea leaves of two susceptible tea cultivars "Longjing 43" (LJ43) and "Zhenong 139" (ZN139), in which the unigenes were generated and annotated, and important functional genes and metabolic

pathways were also revealed. The data sets will provide references for further study on gene expression profiles, biochemical processes and regulation networks associated with tea plant immunity against anthracnose.
