Tissue-Specific High-Quality Protoplast Isolation for Single-Cell Analyses in Arabidopsis

Single-cell RNA sequencing (scRNAseq) has revolutionized plant biology by enabling the analysis of gene expression at an unprecedented resolution. scRNAseq provides transcriptome information on cellular heterogeneity, developmental trajectories, and responses to environmental stimuli at cell-type-specific resolution^1,2,3.

Unlike in animal systems, the rigid plant cell wall and the structural complexity of plant tissues present significant barriers to the efficient isolation of single cells for transcriptomic analysis. Among the available methods, enzymatic removal of the cell wall to produce protoplasts has emerged as a widely used method, which was developed more than 60 years ago^4,5,6. Despite its widespread use, protoplast isolation remains a technically demanding procedure that requires tissue-specific optimization to balance cell yield, viability, and a stable transcriptome. Alternative methods, such as single-nucleus RNA sequencing (snRNA-seq), avoid the need for cell wall digestion by isolating nuclei⁷. However, this comes at the cost of losing information about mature transcripts present in the cytoplasm. Other physical dissociation methods, such as mechanical and laser-assisted microdissection, avoid enzymatic cell wall digestion but are often susceptible to cross-contamination and can result in lower cell viability and limited throughput^8,9,10.

Many plant cell atlases have been developed in recent years, particularly for Arabidopsis, providing important insights into the transcriptomic landscapes of various tissues^{2,11,12,13,14,15,16,17,18}. However, these atlases are primarily based on plants in the juvenile or seedling stages, mainly focusing on either root or leaf tissues. The protocols used in these studies require optimization processes for older Arabidopsis tissues, such as mature leaves, flowers, and stems. Additionally, until the use of cell-type-specific sequencing studies such as scRNAseq or Fluorescence-Activated Cell Sorting (FACS) combined with RNA sequencing, the transcriptome changes that the protoplast isolation was causing were not fully accounted for¹⁹.

The method presented here builds on previous efforts while introducing several critical improvements: it accounts for the physiological and anatomical variability across plant tissues by optimizing buffer concentrations for both root and above-ground tissues (leaves, stems, flowers) and incorporates gentle purification steps to minimize cellular damage and improve the purity of the protoplasts. Compared to nuclei-based approaches, this method has the potential to retain the whole-cell transcriptome, enabling higher resolution information of mature mRNA molecules present in the cytosol. Importantly, the procedure yields high protoplast viability, heterogeneity, and reproducible concentrations across tissues, ranging from approximately 9.8 × 10³ cells/mg in flowers to 2.2 × 10² cells/mg in leaves.

The overall goal of the presented method is to present a robust and reproducible protoplast isolation protocol tailored for multiple tissues of Arabidopsis thaliana, addressing a critical need for high-quality input that covers diverse cell types for plant single-cell studies.

The reagents and the equipment used for this study are listed in the Table of Materials.

1. Buffer preparation

1. Prepare the respective protoplast isolation buffer^17,18according to Table 1 in a 15 mL centrifuge tube; Root protoplast isolation buffer for root tissue or flower/stem/leaf protoplast isolation buffer for above-ground tissues.
   NOTE: Approximate amount of protoplast isolation buffer required for five harvested Arabidopsis thaliana flowering plants (stage 11-12 with closed flower buds): Root (~ 375 mg): 7 mL, Leaf (~930 mg): 5 mL, Stem (~215 mg): 3 mL, Flower (~10 mg): 2 mL.
2. Add 2% (w/v) Cellulase RS and 1% (w/v) Macerozyme R10 fresh before using the root/leaf protoplast isolation buffer. Mix well by putting the protoplast isolation buffer on a rotator until the powder is completely dissolved (around 10 min at 25 °C). Avoid creating air bubbles and make sure that there are no powder clumps left inside the buffer.

2. Tissue digestion

1. Filter the respective protoplast isolation buffer through a 0.2 µm filter before using. Use a standard-sized Petri dish (100 mm x 15 mm) for leaves and roots, and a small-sized Petri dish (30 mm x 15 mm) for small amounts of tissues such as flowers. Add a few drops of protoplast isolation buffer into the petri dish before transferring the tissues.
2. Transfer the respective tissue into the petri dish and chop it into smaller pieces by vertical strokes using a razor blade as finely and fast as possible (Figure 1A).
   NOTE: Always use different blades and tweezers for each sample to avoid cross-contamination. Wear protective gloves while handling the razor blade.
3. Add the remaining protoplast isolation buffer into the respective Petri dish to immerse the chopped tissue and incubate at 25 °C for 1 h. Keep the above-ground tissues on the bench (no shaking) and transfer the root tissues to an orbital shaker platform (60 rpm).

3. Filtration

1. Root sample after the incubation:
   1. Using a Pasteur pipette, transfer the protoplasts from the petri dish to a 40 µm Cell Strainer placed on top of a pre-cooled open 50 mL centrifuge tube.
   2. To harvest the remaining protoplasts, add the exact same volume of W5 solution (2 mM MES pH 5.7, 154 mM NaCl, 125 mM CaCl₂, 5 mM KCl) into the same Petri dish to collect the protoplasts adhered to the base of the Petri dish. Transfer the protoplasts resuspended in the W5 solution to the same 40 µm Cell Strainer placed on top of a pre-cooled open 50 mL centrifuge tube.
   3. Pass it a second time through a new 40 µm Cell Strainer into a new pre-cooled 50 mL centrifuge tube. From now on, keep the samples on ice and be very gentle when handling the protoplasts.
2. Above-ground samples after the incubation:
   1. Using a Pasteur pipette, transfer the protoplasts from the Petri dish to a 70 µm Cell Strainer placed on top of a pre-cooled open 50 mL centrifuge tube.
   2. To harvest the remaining protoplasts, add the exact same volume of W5 solution (2 mM MES pH 5.7, 154 mM NaCl, 125 mM CaCl₂, 5 mM KCl) into the same Petri dish to collect the protoplasts adhered to the base of the petri dish. Transfer the protoplasts resuspended in the W5 solution to the same 70 µm Cell Strainer placed on top of a pre-cooled open 50 mL centrifuge tube. From now on, keep the samples on ice and be very gentle with the protoplasts.

4. Washing and Ficoll purification

1. Pre-cool the centrifuge to 4 °C and collect (pellet) the protoplasts for 5 min at 500 × g with zero brake setting of the centrifuge.
2. Root sample after centrifugation:
   1. Discard the supernatant without disturbing the protoplast pellet.
   2. Resuspend the protoplasts very gently in pre-cooled (4 °C) 600-700 µL root resuspension buffer (equal parts root protoplast isolation buffer (without enzymes) and W5 solution) using a Pasteur pipette or 1 mL Gilson pipette with a cut tip to avoid shearing.
   3. Transfer the suspension to 1.5 mL pre-cooled (4 °C) tubes.
3. Above-ground samples after centrifugation:
   1. Discard the supernatant without disturbing the pellet.
   2. Resuspend the pellet very gently in pre-cooled (4 °C) 600-700 µL flower/stem/leaf resuspension buffer (equal parts flower/stem/leaf protoplast isolation buffer (without enzymes) and W5 solution) using a Pasteur pipette or 1 mL Gilson pipette with a cut tip to avoid shearing.
   3. Transfer the suspension to 1.5 mL pre-cooled (4 °C) tubes.
   4. Prepare 300 µL of 13.3 % (w/v) Ficoll PM 400 solution in a 1.5 mL tube. Mix and keep on ice.
   5. Add the resuspended protoplast sample (600-700 µL) slowly on top of the prepared Ficoll solution without touching/mixing the Ficoll layer.
   6. Centrifuge at 4 °C for 5 min at 700 × g with zero break setting of the centrifuge.
   7. Collect the protoplasts that are present in the middle layer (interface) together with the upper phase (first layer) (Figure 1B). Be careful not to collect the lower third layer, which contains Ficoll and the separated debris. Store at 4 °C till further use.

5. Counting and quality check

1. Use 10 µL of the protoplast solution to count the number of isolated protoplasts with a Neubauer hemocytometer.
   OPTIONAL: 1% (w/v) Evans Blue dye can be added for evaluating the viability of the protoplasts (Figure 1C).
2. Dilute the protoplast solution to the required concentration of protoplasts/mL using pre-cooled (4 °C) resuspension buffer (explained in step 4.2.2 for root and step 4.3.2 for above-ground tissues) of the respective tissue.
3. Proceed with the intended experiment as fast as possible, e.g., single-cell RNA sequencing and/or FACS^15,19.

The implemented protoplast isolation protocols for above-ground and root tissues successfully enable efficient digestion of the cell wall and the release of intact single cells from all Arabidopsis thaliana tissues. Across the tested tissue types, which included flowers, stems, leaves, and roots, the enzymatic digestion process produced well-separated protoplasts with minimal numbers of undigested cell groups (Figure 2A-D). Evans Blue dye assays confirmed that the protoplasts that displayed spherical morphology with smooth plasma membrane boundaries are viable cells (Figure 1C). For each tissue type, spherical protoplasts were counted. Flowers contained on average 97,875 ± 7,531 (9,788 cells/mg, n = 3), stems 119,025 ± 22,508 (550 cells/mg, n = 4), leaves 210,483 ± 15,572 (225 cells/mg, n = 3), and roots 225,363 ± 52,561 (599 cells/mg, n = 4).

The protoplast suspensions showed a broad distribution of cell sizes and the presence of green or pale plastids, reflecting the heterogeneous cellular composition of each harvested tissue (Figure 3A-D). To confirm this, a protoplast isolation with a transgenic companion cell-specific YFP reporter line was performed, and the suspension contained viable small protoplasts with a YFP signal representing companion cells (Figure 3E). This underscores that the used protoplast isolation protocol is suitable for difficult-to-isolate vascular cells with rigid cell walls.

With the presented optimized method, one can isolate a high number of viable protoplasts with a minimal amount of debris from various tissues of adult plants. To improve the purification of above-ground samples, we included a Ficoll cushion cleaning step to remove released organelles and other cell debris from the digested samples while preserving the transcriptome quality (Figure 4).

In the scRNAseq cDNA libraries generated from the protoplasts, highly expressed housekeeping transcripts such as ACTIN2 and rare/low expressed transcripts such as the homeodomain transcription factor KNAT1, a marker specific to cambial and meristematic cells, as well as the companion cell markers SUC2 and PARCL1, were detected by specific cDNA PCR assays15 (Figure 5). This indicates a broad cell-type representation and heterogeneity of the obtained protoplast population, which was confirmed further by analyzing the scRNAseq data produced by this method15.

Figure 1: Workflow of protoplast isolation steps. Six main steps are shown: (1) buffer preparation, (2) tissue digestion, (3) filtration, (4) wash and Ficoll purification, (5) counting and quality control, and (6) continuing with downstream experiments. Please click here to view a larger version of this figure.

Figure 2: Images of protoplast preparation steps. (A) Representative picture of the chopping stage of the leaf sample (Step 2.1). The upper image shows the harvested tissue with a few drops of buffer in a standard-sized Petri dish and a razor blade for chopping. The image below, on the left, represents the tissue after a few vertical strokes with the razor blade (see arrow). The image below, on the right, represents the chopped sample, ready for the next step. (B) Image of a Ficoll cushion after centrifugation in a 1.5 mL tube (Step 4.3.7). The white arrow indicates the middle layer, which contains protoplasts. (C) Representative microscope image of leaf protoplasts incubated in Evans Blue dye. Scale bar: 250 µm. Please click here to view a larger version of this figure.

Figure 3: Microscope images of generated protoplasts. Protoplast images from each tissue type are displayed. (A) Flower, (B) Stem, (C) Leaf, (D) Root. Scale bar: 250 µm. (E) Confocal microscope images of a companion cell-specific YFP reporter line. The first image (left) represents the expression of the YFP companion cell reporter in the leaf vasculature (green color). The second image (right) represents protoplasts from cells expressing the YFP reporter (green signal). Purple/blue color represents chloroplast auto-fluorescence. Scale bar 10 µm. Please click here to view a larger version of this figure.

Figure 4: Microscope images of leaf protoplasts before (A) and after (B) Ficoll cushion cleaning. Scale bars: 250 µm. Please click here to view a larger version of this figure.

Figure 5: Quality control PCRs of scRNAseq libraries. Representative gel images of quality control PCRs of root and above-ground tissue single-cell RNAseq libraries. Tested genesSUC2, PARCL, KNAT1, ACT2 (A to D, respectively), and for each tissue: flower, stem, leaf, root from left to right, respectively. cDNA (+) is used as a positive control, and water is used as a negative control. Please click here to view a larger version of this figure.

Table 1: Protoplast isolation buffer compositions. Protoplast isolation buffer recipes for root and above-ground (flower, stem, leaf) tissues are provided.

This optimized protocol for isolating protoplasts from various Arabidopsis tissues was tailored for scRNAseq applications, analyzing the whole transcriptome present in the cytoplasm and nuclei. Tissue-specific optimization of the buffers and additional steps for digestion and purification resulted in robust and reproducible yields across different tissues. Specifically, flower protoplasts averaged ~98,000 cells per sample (~9,788 cells per mg tissue), stem protoplasts ~119,000 cells (~550 cells per mg), leaf protoplasts ~210,500 cells (~225 cells per mg), and root protoplasts ~225,000 cells (~599 cells per mg), reflecting the structural and physiological diversity of these tissues.

A key refinement in this protocol was the use of a Ficoll-based cushion instead of the more commonly used sucrose-based gradients/cushions for purification. Unlike sucrose, which is taken up by cells and has a major effect on the transcriptome^13,20,21, Ficoll is osmotically inert and membrane-impermeable, thereby preserving RNA integrity during isolation²². This step also reduced the number of free organelles, such as chloroplasts, from broken cells and cell debris in the suspension. The presence of such debris can interfere with droplet formation by clogging the channels of microfluidic devices or resulting in the presence of more than one cell in formed droplets, which contributes to ambient RNA contamination. All of these factors can lead to incorrect conclusions or biases in the analysis of scRNA-seq data.

It was shown that protoplast isolation can alter the transcriptome of plant cells^2,15,16,19. A key strategy in this protocol to minimize this artifact is maintaining protoplasts at 4 °C (on ice) after enzymatic digestion, which slows down metabolic and transcription activity and, thus, reduces transcriptional changes known to be associated with protoplast isolation. In addition, it is crucial to keep tissue harvesting time consistent across replicates and align it with the study objective, as variations in timing have been shown to influence both protoplast-induced genes and circadian-regulated transcripts detected in scRNA-seq datasets¹⁵.

In this method, besides producing single cells with minimal debris, another very important point was keeping the heterogeneity of the tissues to potentially represent all cell types. The diversity in the protoplast suspensions was confirmed by the detection of different morphologies of protoplasts, including diverse sizes and the presence of signals from cell-type-specific reporter lines within the obtained population (Figure 2). Additionally, the protoplast population was evaluated by detecting rare and cell-type-specific as well as housekeeping transcripts in scRNA-seq cDNA libraries using PCR (Figure 4). These included rare marker genes of the shoot apical meristem, and markers of companion cells¹⁵. The successful capture of these transcripts confirms that the presented method allows for the isolation of diverse cell types with high integrity, making it a valuable tool for advancing single-cell research in plants. In addition to scRNAseq, this method can be used for flow cytometry, transient gene expression analysis, and plant regeneration assays as well.