Trichomes are the special structures derived from the epidermal cell of plant aerial part (Werker, 2000). Nearly 30% of all vascular plants have multicellular glandular trichomes (Glas et al., 2012). Because of the different secondary metabolism abilities, trichomes are segmented into non-glandular and glandular trichomes (Serna and Martin, 2006; Huchelman et al., 2017). Two types of trichomes are covered on both the adaxial and abaxial leaf surfaces of Artemisia annua, T-shaped trichomes (TSTs) and GSTs, and GST is identified as the main spot for artemisinin synthesis and accumulation (Ascensao and Pais, 1987; Duke and Paul, 1993). Besides, characteristic essential oil and various of potential secondary metabolites including flavonoids, phenolics, and terpenoids/terpenes may be accumulated in GSTs (Bhakuni et al., 2001; Ahmad Malik et al., 2009; Ferreira et al., 2010). Accordingly, GSTs are considered to be intriguing natural products troves with substantial economic potential for medicines, essential oils, and natural insecticide, where numerous known or undetected metabolites and metabolic pathways can be discovered (Glas et al., 2012; Sallets et al., 2014; Huchelman et al., 2017).

It is feasible to directly harvest trichomes from plants only adhering single type of trichome by brushing the plant surfaces, shaking the leaves or buds in aqueous solution, or brushing the frozen leaves (McCaskill et al., 1992; Wang et al., 2001; Feng et al., 2021). However, in most cases, multiple types of trichome appear simultaneously in plant. Thus, these approaches are not suitable for them. In some species, like A. annua, the stalk cells of GSTs are firmly stuck on the epidermis, shaking buds in aqueous solution might be the only feasible way to obtain enough GSTs, while they are still mixed with TSTs and limited by the floral bud stage (Wang et al., 2009). In tomato, it is able to isolate trichomes by brushing the leaves (Balcke et al., 2014), but subsequent selection of each type is time-consuming, and laser capture microdissection (LCM) is required.

Laser capture microdissection is an extraordinary technology using microscope and laser to harvest specific populations of tissue cells and even organelles (Day et al., 2005; Espina et al., 2006). The collected samples need proper fixation which is one of the most critical steps during sample preparation for LAM and then they can be used to extract DNA, RNA, or protein for genomic, transcriptomic, proteomic studies of specific cell types (Kerk et al., 2003; Nakazono et al., 2003; Day et al., 2005; Fang and Schneider, 2014). In plants, it has been broadly used to explore the gene expression in different tissues (Hacquard et al., 2010). To date, however, LCM used for metabolic analysis has been rarely reported because of the limitation by the use of various fixatives (Gautam and Sarkar, 2015). The fixation is not helpful for tissue dedicated to the analysis of small molecules which can be soluble in the solvent used (Hölscher and Schneider, 2008). In addition, the fixative-free procedure of cryosectioning gave the only chance to exclude the total extraction of secondary metabolites (Hölscher and Schneider, 2008). Nevertheless, the large vacuoles may cause the loss of integrity of plant cells (Nelson et al., 2006). Thus, in order to isolate high-quality single type of trichomes from live plants without fixation or cryosections for deep studies, an improved method is needed.

In this study, we developed an improved strategy for LCM without tissue fixation or cryosections to capture unitary and purer trichomes. 150 high-quality GSTs of A. annua were captured efficiently in 30 min and used for artemisinin measurement using UPLC/IM-QTOF-HDMS. The artemisinin and arteannuin B were detected clearly, and 15 potential compounds in GSTs were identified. This improvement may be used to exploit the studies on plant metabolic profiles, as well as protein analyses.

Researches have shaped our understanding of the multitudinous secondary metabolites in glandular trichomes over the past few decades. In Mentha spicata L., two types of glandular trichomes, capitate and peltate were observed (Champagne and Boutry, 2013). Volatile essential oils as secondary metabolites are mostly produced in peltate glandular trichomes (PGTs), including some monoterpenes, limonene and carvone (Alonso et al., 1992; Champagne and Boutry, 2013; Markus Lange and Turner, 2013; Wang et al., 2016). There are five types of trichomes on Cannabis sativa L., three of them are glandular trichomes, namely capitate-stalked, capitate-sessile, and bulbous trichomes (Dayanandan and Kaufman, 1976; Happyana et al., 2013). The glandular trichomes are as the main production and storage site to terpenes, and cannabinoids which is famous for their psychoactive and therapeutic effects (Kim and Mahlberg, 1991; Mahlberg and Kim, 2004). Two types of trichomes are described on the leaves of tobacco, the short type may secret hydrophilic nicotine, while the long type produces a resin containing diterpene cembratrienediol (Meyberg et al., 1991; Amme et al., 2005). There are multiple trichomes classified as types I–VII in tomato (S. lycopersicum). Types I, IV, VI and VII are glandular trichomes, whereas types II, III and V are non-glandular trichomes (Jin-Ho et al., 2010). Type I and type IV trichomes are involved in acyl sugar biosynthesis, while type VI trichomes are considered as the foremost spot for the secretion of terpenoid, flavonoid and methyl ketone (Bleeker et al., 2011; Feng et al., 2021). Therefore, glandular trichomes of plants are considered as a great site for secreting and storing abundant specialized secondary metabolites including terpenoids/terpenes, phenylpropanoids, flavonoids, methyl ketone, polyphenols, and acyl sugars, and these compounds can be broadly used as natural medications, insect repellents or essential oils (Tattini et al., 2000; Dixon, 2001; Fridman et al., 2005; Gershenzon and Dudareva, 2007; Schilmiller et al., 2008; Tissier, 2012; Huchelman et al., 2017). So that understanding the high-efficiency metabolic network of glandular trichomes and the transport and storage of metabolites could be the next challenge (Schuurink and Tissier, 2020).

Laser capture microdissection (LCM) is an advanced technology that allows the identification, selection, and isolation of pure cell populations. Generally speaking, LCM requires histological processing and fixation to stabilize proteins and RNA, nevertheless, this procedure using various fixatives, coagulative fixatives such as acetone and alcohol or cross-linking fixatives such as formaldehyde, may wash out or damage the metabolites (Gillespie et al., 2002; Inada and Wildermuth, 2005; Schad et al., 2005; Schneider and Hlscher, 2007; Li et al., 2012; Fang and Schneider, 2014; Goldsworthy et al., 2015). Moreover, using cryosections can extract RNA, proteins, and metabolites without histological fixation (Asano et al., 2002; Nakazono et al., 2003; Schad et al., 2005, 2010; Li et al., 2012). However, unlike animal cells, plant cells will lose their integrity in cryosections as a result of the large vacuoles and cell wall of variable rigidity (Nelson et al., 2006; Gautam and Sarkar, 2015). Besides, in RNA analysis, adequate RNA samples for studies can be achieved by linear amplification (Kerk et al., 2003; Nelson et al., 2006). But this amplification is not suitable for proteins and metabolites, which are limited by the abundance of cellular targets (Fang and Schneider, 2014). So that, it is necessary to develop a method dispense with tissue fixation or cryosections.

Comparing to the traditional way using brushes, LCM is a more direct way to isolate pure trichomes. Accordingly, LCM of plant trichomes is applied in some cases. In A. annua, LCM was used to isolate different whole cells of A. annua GSTs from slices of leaves and flower buds fixing with 4% para-formaldehyde. While trichomes perpendicular to the laser beam are necessary, which increase the randomness and uncontrollability (Olsson et al., 2009). In C. sativa, capitate-sessile and capitate-stalked trichomes were isolated over the flowering period using LCM with neither fixation nor freeze (Happyana et al., 2013). In Colquhounia coccinea, peltate glandular trichomes was collected directly from unfixed leaves using LCM (Li et al., 2013). However, the obtained trichomes were not trichome-exclusive and always adhered with redundant mesophyll. On the other hand, a special angle of trichomes is needed. So far, capturing unfixed trichomes or other plant cells is still difficult and costly.

Strikingly, the method we described in this study provides a very instructive solution to shape the metabolic network of plant trichomes. In our study, the extract of 150 GSTs provided a very clear peak and strong signal of artemisinin, and offered a number of potential compounds using database available. In addition, the expression level of AaADS, AaCYP71AV1, AaDBR2 and AaALDH1 (Figure 5), which are the key structural genes involved in the artemisinin biosynthesis pathway, verified the reliability of our method. Accordingly, collecting more numerous GSTs may get stronger signals and more information. That is to say, it can give us insights into the known or undiscover molecules with bioactive potential in plant GSTs.

In conclusion, we have provided an alternative and promising strategy, which is suitable for many other plant species to efficiently capture trichomes, to isolate GSTs without fixation by LCM. 150 GSTs of A. annua were captured in 30 min and used for metabolites analysis using UPLC/IM-QTOF-HDMS. The artemisinin and arteannuin B were detected, respectively, and 15 potential compounds in GSTs were identified. This improvement may offer the possibility of trichome-specific proteomics and metabolomics, or other single cell analysis.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

WQ and KT designed the research. WQ, YL, BP, HL, TC, XY, YZ, CW, and XGY performed most of the experiments. WQ and YL drafted the manuscript. XF, LL, and KT revised the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by National Key R&D Program of China (2018YFA0900600), the Bill & Melinda Gates Foundation (OPP1199872), the National Natural Science Foundation of China (31770327), SJTU Trans-med Awards Research (20190104) and SJTU Global Strategic Partnership Fund (2020 SJTU-CORNELL).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.