Background

The Tandem Tag (TT) assay is a widespread approach for quantifying autophagic flux in yeast and mammalian cells (Zhou et al., 2012; Klionsky et al., 2016; Yoshii and Mizushima, 2017). It has been previously described to be applicable for plant cells in a study using tobacco BY-2 cell suspension cultures (Hanamata et al., 2013; Klionsky et al., 2016). Here we provide a detailed protocol for in planta quantification of autophagic flux in epidermal root cells of Arabidopsis thaliana seedlings (Figure 1). The TT assay employs ratiometric quantification of red and green fluorescence and allows to quantify relative induction or inhibition of autophagy. Another advantage of this method is that it provides valuable information on subcellular localization of ATG8, i.e., translocation of ATG8 from the nuclei to cytoplasm (Zhou et al., 2012), incorporation into puncta of different morphologies and motility, and its accumulation in the vacuole.

For this protocol we used stable transgenic Arabidopsis lines expressing the optimized TT (Huang et al., 2015) fused to AtATG8a and driven by a double 35S promoter. The TT-AtATG8a fusion was introduced into wild-type or atg5atg7 double knockout backgrounds, to produce reporter and control lines, respectively (Dauphinee et al., 2019). Under normal growth conditions TT-AtATG8a is localized in the cytoplasm and in the nuclei of the root epidermal cells, but upon induction of autophagy it is gradually translocated to the cytoplasm, then incorporated into autophagosome membranes and delivered together with the cargo to the lytic vacuole. While TagRFP shows relatively high tolerance to the pH of the lytic compartments (Huang et al., 2015), fluorescence of mWasabi is significantly reduced under the same conditions. Thus, ratiometric measurement of the TagRFP and mWasabi fluorescence allows to estimate the delivery rate of the fusion protein to the lytic compartment and eliminates potential bias coming from differences in fusion protein expression. Furthermore, since the assay relies on confocal microscopy image acquisition and analysis, it is possible to obtain time-resolved and dose-dependent data about the changes in autophagic activity. Although we describe only a protocol for detection of the TT-AtATG8a delivery to the vacuole, it is also possible to use the same imaging data to visualize and quantify dynamics of autophagosome formation.

We observed that response to known modulators of autophagic activity, such as AZD8055, which induces autophagy by inhibiting TORC1 activity, and concanamycin A (ConA) (Dauphinee et al., 2019), which inhibits autophagy by inactivating vacuolar vATPase and thus changing the pH of the lytic compartment, in Arabidopsis roots significantly varied depending on the area of the root zone scanned. By comparing data obtained from different root zones, we established that the most reproducible measurements could be obtained by analyzing images of root epidermal cells located in the beginning of the differentiation zone. Nevertheless, we still observed quite significant variation between responses in tricho- and atrichoblasts. Hence, a relatively large number of images was required for the data analysis to obtain a reliable mean value representative of average autophagic activity in root cells.

This assay provides a significant improvement of autophagic activity measurement in planta. It is based on high-throughput image analysis, thus improving reproducibility and robustness of the results. It relies on the use of designated macros written in ImageJ Macro Language (IJM) and R scripts. We ensured that it can be universally used for imaging data obtained using confocal laser scanning microscopes (CLSMs) of various manufacturers. Furthermore, to enable applicability of the protocol for images obtained on different CLSM systems that will naturally vary in efficacy of detection, we added an extra step of threshold values adjustments that will maximize the capacity of the vacuole area selection tool.

Importantly, the TT assay described here allows quantification of autophagy-dependent delivery of ATG8 to the lytic vacuole. Although it can be used as a very good indication for estimating autophagic activity, it is still preferable to combine this assay with other techniques verifying degradation of autophagic cargo e.g., the GFP-ATG8 cleavage assay or long-lived proteins assay (Dauphinee et al., 2019; Klionsky et al., 2016).


Figure 1. A workflow scheme of the Tandem Tag assay. The protocol for the assay comprises six major steps. The time required for each step was estimated based on experienced user progress. It is recommended to use a seed plating guide in the first step to establish equal distance between seedlings and reproducible growth conditions. In the second step, seedlings are grown on the vertically positioned plates to ensure that roots remain on the surface of the medium. We suggest to perform drug treatments in liquid medium for faster, more even and more reproducible delivery of the drugs into the root cells. The required treatment time will be compound-specific. The time required for confocal microscopy might vary depending on microscope configuration and software.