Edited by:
Reviewed by:
*Correspondence:
†Present address: Anna Kärkönen, Natural Resources Institute Finland (Luke), Green Technology, Latokartanonkaari 7, Helsinki, Finland
‡These authors have contributed equally to this work.
This article was submitted to Technical Advances in Plant Science, a section of the journal Frontiers in Plant Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Laser capture microdissection (LCM) enables precise dissection and collection of individual cell types from complex tissues. When applied to plant cells, and especially to woody tissues, LCM requires extensive optimization to overcome such factors as rigid cell walls, large central vacuoles, intercellular spaces, and technical issues with thickness and flatness of the sections. Here we present an optimized protocol for the laser-assisted microdissection of developing xylem from mature trees: a gymnosperm (Norway spruce,
Laser capture microdissection (LCM) was originally developed to assist semi-automatic collection of individual cells and tissues from complex biological specimens for subsequent genomic and transcriptomic analyses (
Development of new techniques for high-throughput data acquisition, concurrent development of bioinformatic analyses and increasing availability of sequenced genomes have made LCM a method of choice for elucidating the cell type-specific molecular features. Downstream applications can cover virtually every aspect of cell biology. Above all, gene expression analyses are, nowadays, often performed through genome-wide expression profiling based on microarray or next generation sequencing (NGS) technologies (whole genome sequencing, targeted genome sequencing, RNA sequencing and ribosome profiling). The combination of LCM with NGS techniques is a powerful approach to study cell- and tissue-specific processes (
Microdissection of plant tissues and individual cells poses a challenge due to the presence of a cell wall, a large central vacuole in most fully differentiated cells and, in some tissues, large intercellular spaces (
Non-paraffin embedding procedures have been introduced recently to LCM of plant tissues. An alternative strategy of sample preparation which guaranteed RNA quality and morphological integrity, replaces paraffin with Steedman’s wax. This polyester wax has a lower melting point than paraffin (38-40°C
LCM has also been used to study tissue- and cell-specific processes of non-xylem tissues in woody plants. Abbott and co-workers successfully isolated resin ducts and cambial tissues from tangential cryosections of 2-year-old white spruce (
Xylem tissue with its highly lignified cell walls has proven challenging for LCM applications.
Here we developed a protocol for LCM of developing xylem tissues from 40-year-old Norway spruce (
We developed protocols for the LCM of wood sections for extraction of cell type-specific total RNA from developing xylem of Norway spruce and aspen. Disks of wood stems of 40-year-old spruce and aspen trees were collected during the active secondary growth and stored at -80°C. For an easier handling of the samples during preparation of the cryosections, the stem disks were further sawn into smaller blocks. The blocks were then used to prepare 30/40 μm thick tangential (spruce) or cross and radial (aspen) sections using a cryomicrotome. The cryosections, mounted on microscope glass slides, were fixed with ethanol. Spruce and aspen wood required different procedures for dehydration and flattening of the sections. Spruce tangential sections were flattened with a pipette tip during the last dehydration step in 100% ethanol. Aspen cross and radial sections were fixed in 75% ethanol and then flattened using a microscope slide sandwich during dehydration in a freeze-dryer chamber. Intrinsic differences in wood anatomy between gymnosperms and angiosperms determined the sectioning plane and, subsequently, the flattening method for the cryosections. For example, the presence of large vessels in aspen prevented the use of tangential sections, because of the high probability of section breakage along the vessel. Individual cell types (ray parenchymal cells, tracheids, and fibers) were isolated by LCM using a PALM Micro-Beam microscope at room temperature, and collected into adhesive caps of collecting tubes. Dissected material was snap-frozen in liquid nitrogen and stored at -80°C until RNA isolation. Specific cell types collected over several LCM sessions were pooled together for RNA extraction.
A comprehensive LCM protocol for the developing xylem of both angio- and gymnosperms, applicable to tangential, radial and cross sections.
Less laborious protocol, avoiding long embedding procedures common in several published protocols.
Replaces membrane-coated slides with glass microscope slides, and introduces the easy assembly of a “glass sandwich” to keep anatomical integrity and flatness of the xylem sections.
Enables isolation of treachery elements, fibers and ray cells for RNA extraction.
Enables isolation of cell-type specific, good quality total RNA for high-throughput analyses.
The method is divided into five stages from A to E: (A) tissue sampling, handling and preparation of the frozen specimens; (B) cryosectioning; (C) fixation and dehydration; (D) LCM; (E) RNA extraction and quality assessment (
A 40-year-old spruce tree was felled during active secondary growth. Disks of the trunk were sawn and immediately frozen in dry ice, and then stored in plastic bags at -80°C.
In order to avoid thawing and to facilitate the subsequent preparation of cryosections, small wood blocks were cut from the frozen stem disks with a jigsaw. For convenience, the stem disks were first sawn in halves. Further sawing was conducted in two parts: first, a long strip of wood, around 10.0 cm × 1.2 cm × 1.2 cm (H × W × D), was sawn tangentially from the half stem disk (
Developing xylem could be identified on the tangential side of the woody cube as a white, almost translucent layer, as opposed to the yellowish, previous year’s growth ring. This facilitated the orientation of the cube during cryosectioning. Alternatively, the side of the mature xylem in the cube was marked red by a quick dipping in 0.5% Safranin O (in 50% ethanol), before storage at -80°C.
Aspen wood specimens were collected from a 40-year-old tree according to the scheme illustrated in
For collection of wood samples, different strategies of cutting can be applied (
During the preparation of the small wood blocks, we advise to define a reference to easily identify the developing xylem zone, for example, by staining as in the case of spruce, or by keeping the bark in place as we did for aspen.
Jigsaw; plastic bags; aluminum foil; 50 ml tubes; dry ice; liquid nitrogen; 0.5% Safranin O in 50% ethanol).
The orientation of the block determines the direction of sectioning (
Choice of the direction of sectioning.
Section orientation | Advantages | Disadvantages |
---|---|---|
Tangential | Easy isolation of pure ray cell lines. Easy to prepare cryosections, as fibers/tracheids go along the cutting plane. | Difficult to identify the boundaries between the annual growth rings during sectioning. A preliminary estimation of the thickness of developing xylem is required. In hardwoods, the lignified cell walls of fibers/vessels cause longitudinal cracks in the section, and hamper the adequate flatness of the sections on a glass slide. |
Radial | Easy to identify fibers and ray cells, and isolate them in large amounts. If the section is perfectly radial (not oblique), the orientation of fibers/tracheids/vessels helps to keep the section flat. | If the section is not perfectly radial, the direction of fibers/tracheids/vessels makes it difficult to fix the sample on the slide and keep it flat. Contamination from the surrounding fiber/tracheid cells is not excluded. |
Cross | Easy to distinguish the developing xylem from the mature xylem, and to localize different types of cells. The porous appearance of the section facilitates the laser capture microdissection cut. | When isolating ray cells, contamination from the surrounding fiber/tracheid cells cannot be excluded. |
Glass slides were acid-washed in 1 M HCl for 16 h at c. 55°C. After several washing steps in dH2O, the slides were soaked in 100% ethanol, air-dried on a rack and sterilized at 180°C for 4 h. Alternatively, the slides were first washed with RNase decontamination solution or 0.1% DEPC (diethyl pyrocarbonate)-treated water, then with 100% ethanol and finally autoclaved. As a third alternative treatment to remove RNase contamination, the glass slides can also be incubated in a sterile Petri dish under UV radiation for 30 min.
Prior to cryosectioning all the tools, including tweezers and razor blades, were cleaned with the RNase decontamination solution. We do not recommend the use of this solution to decontaminate the cryomicrotome, which can be cleaned with 100% ethanol instead. The temperature of the cryomicrotome was set to -20°C, at least 1 h before starting sectioning. During the first hour of pre-cooling, mounting disks and glass slides were kept in sterile Petri dishes, and together with tweezers, razor blades, and ethanol solutions cooled inside the cryomicrotome chamber.
Membrane-coated slides used in other published protocols (
The thickness of developing xylem in spruce was determined using an ocular micrometer from 20 μm thick cross sections stained with Safranin
The block of developing xylem was mounted with tweezers on a cooled mounting disk with the tissue freezing medium. We first performed a trimming step, cutting a series of 20 μm thick sections, to remove about 100 μm of the outermost layer of developing xylem. Then, 40 μm thick sections were cut and placed on pre-cooled, RNase-free glass slides. Each slide carried 3–4 sections. According to the previous estimation of the thickness of the developing xylem, the cutting depth did not exceed 600 μm. To facilitate the attachment of the sections to the slide, the slide was slightly warmed from the back with a gloved finger at the site of the sections for ca. 10 s. This allowed collection of the freshly cut sections directly from the cryomicrotome stage, and avoided any further handling or damaging of the sections with tweezers.
The frozen wood block was mounted as described above for spruce. Bark and the oldest layers of secondary phloem were removed by cutting serial, 20 μm thick tangential sections. The youngest layers of the secondary phloem, 500–700 μm from the cambium, were left intact to facilitate the cross- and radial-sectioning, and to provide a reference for the orientation of the sections (
Three or four 40 μm thick cross or radial sections (around 1 cm × 1 cm/1 cm × 1.5 cm) were cut and delicately positioned using sterile tweezers in 50 μl of 75% ice-cold ethanol on a pre-cooled, RNase-free microscope slide. The addition of ethanol helped the sections to adhere to the slide. The glass slide was kept uncovered and transferred into a 50 ml tube, which was previously cooled in liquid nitrogen. The tube carrying the slide was closed tightly, snap-frozen in liquid nitrogen and kept on dry ice until the freeze-drying step.
This stage of the protocol implies the longest handling time for the samples before the fixation. For this reason, appropriate measures should be taken to limit the risk of RNA degradation. Tools and equipment need to be sterile or treated with a RNase decontamination solution or, if not possible, with ethanol. Moreover, the whole procedure has to be done quickly and at low temperature. We recommend to keep all the tools, e.g., razor blades, tweezers, mounting slides, in the cryomicrotome chamber.
Cryomicrotome (Leica CM3050 S); UV Crosslinker (Spectrolinker XL-1000); mounting disks; freezing mounting medium (Leica Tissue Freezing Medium, Ref. 14020108926); microscope glass slides; glass Petri dishes; 50 ml tubes; tweezers; razor blades; liquid nitrogen; RNaseZAP® decontamination solution (AMBION AM9780); 100% ethanol; 1 M HCl; 0.1% DEPC (diethyl pyrocarbonate; Sigma–Aldrich®, 1609-47-8)-treated water; 75% ethanol solution in RNase-free water (AMBION AM9930); solution (1:1, v/v) of 0.5% Safranin
Two minutes after the last section was placed onto the slide, ice-cold 70% ethanol was pipetted in excess on top of the sections to fully cover the whole area. The slide was then transferred in a sterile glass Petri dish into a laminar flow cabinet, where 70% ethanol was replaced with ice-cold 100% ethanol. The subsequent fixation step was performed at room temperature in the laminar flow cabinet. During the 2-min incubation in 100% ethanol, the sections were gently stretched and flattened with a pipette tip (see critical notes). After the removal of ethanol by pipetting, sections were air-dried for 15 min. Slides with dry sections were collected into sterile 50 ml tubes. The tubes were tightly closed and stored at -80°C for no longer than 7 days.
The freeze-dryer chamber was first cleaned with the RNase decontamination solution and pre-cooled to -50°C. The cryosections were dried as follows: (i) in a closed 50 ml tube for 20 min. The tube cap was kept loose but not totally unscrewed; (ii) in the totally opened tube for additional 5–10 min; (iii) in a closed glass sandwich kept in the 50 ml tube, overnight. The glass sandwich was prepared by covering the glass microscope slide carrying the sections with a second one and closing them on their short sides with a pair of clips (
This stage is particularly important to obtain flat and dry sections required for successful LCM and good-quality RNA extraction (see critical notes). The glass sandwich keeps the specimens flat during and after drying. However, we observed that the sections were optimal for LCM if they were first kept 30 min uncovered in the freeze-dryer. During this incubation, sections curl and become partially detached from the glass slide, but are still attached all along the phloem and the cambium. After this initial drying step the sections can be flattened without cracks by closing the glass sandwich for the following overnight dehydration.
Spruce: ice-cold 70% ethanol in RNase-free water (QIAGEN); 100% ethanol; glass slides; pipette tips; 50 ml tubes; laminar flow cabinet; Aspen: freeze-dryer (Scanvac CoolSafeTM), RNaseZAP® decontamination solution (AMBION AM9780); 50 ml tubes; microscope glass slides; office clips.
Laser capture microdissection step was performed similarly for spruce and aspen sections. The only difference was that the aspen samples required an additional preparatory step to immobilize the sections on glass slides. At the end of the overnight freeze-drying, the glass sandwich was removed from the freeze-dryer and delicately opened in the laminar flow cabinet. The dried sections were stably attached to the glass with strips of tape on their longer sides (
The spruce samples were transported to the LCM apparatus in a 50 ml tube kept in liquid nitrogen. Prior to LCM, the tube containing the spruce cryosections was kept at room temperature for 20 min to prevent condensation onto the slide.
The specimens were dissected using the PALM® MicroBeam System (Carl Zeiss AB). Optimal laser dissection was achieved with both types of wood according to the following settings: cut speed 20–30; cut energy 60–65; laser power catapulting energy 25–30. The best focus was achieved in the range of 65–70 with delta focus at –2. The optimal magnification for outlining and cutting was 20× (spruce) and 10× (aspen). This is similar to the most suitable magnification for catapulting (20× for spruce and 10× for aspen) (
The dissected pieces were collected by catapulting into the adhesive cap of a 500 μl collecting tube. Each tube and each slide were used during a single LMC session for no longer than 1.5 h. In this time, we were able to catapult around 15–25 and 200–300 microdissected pieces for aspen and Norway spruce, respectively, which generally corresponded to a total area of 0.5–0.75 mm2 and 0.2–0.4 mm2 of aspen and spruce wood tissue, respectively, collected in a single adhesive cap. Specimens, dissected during several LCM sessions, were combined to obtain a detectable amount of extracted RNA. For example, the total area of spruce LCM-isolated ray parenchymal cells was of ca. 12 mm2. At the end of every session (usually 1.5 h long), the collecting tube was snap-frozen in liquid nitrogen. All the dissected samples were stored at -80°C until RNA isolation.
We suggest to always prepare more than one slide with 3–4 sections for one LCM session in order to have enough sections in case one appears unsuitable for LCM. The use of RNase-free tools (tweezers and needles) during addition of tape to the freeze-dried sections is recommended although the sections are now fixed, and thus, more protected against RNA degradation. In a troubleshooting
Troubleshooting list for LCM and practical recommendations.
Problem | Possible reason | Way out |
---|---|---|
No or very low amount of RNA detected. | RNA degradation during the section preparation or during the LCM session. Not enough dissected cells. | Improve the procedures for RNase decontamination. Shorten the time at room temperature before or during the LCM session. Increase the number of dissected cells for RNA extraction. |
Section sticks to the slide, catapulting of dissected cells is impossible. | Section is not dry enough. | Prolong air-drying or freeze-drying step. |
Burning of the section during cutting. | Laser power is too high. Section is too thick. | Reduce laser power. Decrease the thickness. Do not exceed 60 μm in case of hardwood sections. |
Loss of power cut during cutting. | The section area is out-of-focus or became out-of-focus during the cut. Oblique sectioning during preparation of cryosections. | Repeat the cut along the outline. Adjust the focus slightly until the laser beam turns thinner and effective again. Improve the orientation of the wood block during sectioning. |
Wavy, not perfectly flat section. | Oblique sectioning during preparation of the cryosections. Section not perfectly dry. Section not stably attached to the mounting slide. | Improve the orientation of the wood block during sectioning. Prolong the dehydration step. Avoid condensation onto the slide. Keep the slides in a desiccator chamber. After freeze-drying, add tape to the section more tightly. |
Cracked section after the freeze-drying step | Glass sandwich was assembled too early. | Wait longer before assembling the glass sandwich. Slightly increase the section thickness. |
Laser microdissection microscope (PALM MicroBeam, Zeiss); adhesive cap tubes 500 μl opaque (Carl Zeiss AB, 415190-9201-000); liquid nitrogen; 50 ml tubes; additionally for aspen: tweezers; needles; office transparent tape (MagicTM invisible 3M, Scotch®); scissors; desiccator.
RNA was extracted from the whole cryosections and dissected cells pooled from several LCM sessions. Total RNA quality was assessed using the RNA Pico Assay for 2100 Bioanalyzer.
Aspen sections required a longer preparation time than those of spruce, and were consequently more susceptible to RNA integrity impairment, especially during the last pre-LCM stage at room temperature. Total RNA quality was therefore measured at every step of the protocol, i.e., in a single intact cryosection, in a single intact cryosection fixed and left at room temperature in the desiccator for two days, in microdissected cells, and in residual sections used for LCM (
In spruce, the RNA quality was assessed first in whole cryosections. Twenty sections, each 20 μm thick, were pooled and processed in the same way as microdissected samples, i.e., fixed and dehydrated in ethanol series, and then incubated for 2 h at room temperature, equal to the time of the LCM session. As a control, an identical set of whole cryosections were kept at -80°C for no longer than 7 days and analyzed. No significant impairment of the RNA integrity was observed in the sections stored at room temperature. The quality of total RNA isolated from the separately dissected tracheids and ray parenchymal cells was of adequate quality for RNA sequencing (RIN 6–9,
The RNA yield we could obtain with our protocol was for spruce ray parenchymal cells and xylem tracheids c. 60 ng, and for the whole sections from 90 to 180 ng. In the case of aspen, the total RNA yield measured on LCM dissected samples was between 5 and 20 ng on average for xylem fibers and rays. The RNA yield per volume of the tissue was 0.5 ng/mm3 (spruce whole sections), 3.36 ng/mm3 (spruce tracheids), 9.6 ng/mm3 (spruce ray cells), and 0.7–2.1 ng/mm3 (aspen ray cells). To test the suitability of aspen samples for transcript analysis, two LCM samples containing approximately 4 and 6 ng RNA was amplified and subjected to reverse transcriptase-PCR (RT-PCR) using actin and ubiquitin primers. RT-PCR products were detected on an agarose gel, a cDNA library from total developing wood was used as a positive control (
RNA extraction kit for LCM material/small samples (RNAqueous® Micro Kit, AMBION) (aspen); RNeasy® Plant Mini Kit (QIAGEN, 74904) (spruce); RNase free water (AMBION AM9930 (aspen); QIAGEN (spruce)); RNA assay kit (AGILENT RNA 6000 PICO KIT 5067-1513); Agilent 2100 G2938A Bioanalyzer for RNA quantitation and quality assessment. MessageAmpTM II aRNA kit (ThermoFisher (aspen)); RT-PCR kit (Thermofischer Maxima First Strand cDNA Synthesis Kit for RT-qPCR, with dsDNase); actin and ubiquitin primers (aspen): ACT-F: TGAGGATAT TCAGCCCCTTG, ACT-R: CCATGCTCAATGGGGTATTT, UBI-F: AGATGTGCTGTTCATGTTGTCC, UBI-R: ACAGCCACTCCAAACAGTACC.
The presence of the cell wall in plant tissues poses a challenge in laser microdissection of plant samples. Lignification of secondary cell walls adds another challenge in LCM of woody samples. Softer plant tissues, such as developing phloem, can be cut by LCM from 100 μm thick sections (
In LCM the target cells are visualized
The strategies of sectioning and fixation selected in our protocol helped to optimize the flatness of the sections and the overall quality of the plant samples. In contrast to the majority of the protocols on LCM of plant tissues (
During the fixation step, xylem sections do not stably adhere on the membrane-coated slides. PEN membrane slides are widely used in LCM of animal and some plant tissues (
The goal of our protocol was to extract good quality total RNA from specific cell types of developing xylem of spruce and aspen. The protocol was developed to limit RNase contamination and RNA degradation during sampling and further processing of the samples. We prevented RNase contamination by working with sterile instruments, by cleaning all the tools and equipment with RNase decontamination solution or with ethanol, and using UV radiation to pretreat glass slides. RNA degradation was minimized by limiting the handling of the wood blocks and the sections at room temperature prior to the fixation step. Similarly, we minimized the exposure of spruce sections at room temperature during fixation, applying a fast procedure based on two sequential ethanol incubations, the first of which was performed in the cryomicrotome chamber at -24°C. Fixed spruce sections were stored at -80°C for no longer than 7 days, and good quality RNA was extracted (RIN higher than 7) from the sections (
The protocol described here allows isolation of ray parenchymal cells, tracheary elements, and fibers from developing xylem of both angiosperm and gymnosperm tree species by combining optimized techniques of wood section preparation and LCM technology. The resulting RNA is of high quality demonstrating that the protocol can be applied to transcriptome analyses aimed at deciphering cell-specific molecular events.
OB and CV contributed to the concept and design of the work, performed the experiments and wrote the manuscript; KS performed experiments and wrote the manuscript; LZ performed experiments and was involved in drafting the manuscript; AK, TN, and KF substantially contributed to the concept and design of the experiments and took part in writing of the manuscript.
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.
We thank Dr. Mikael Niku, the head of the laser microdissection core facility at the Department of Veterinary Biosciences, University of Helsinki, for valuable discussions, instructions and practical help with the Micro Beam instrument. We deeply appreciate the help of Annika Korvenpää, Department of Agricultural Sciences, University of Helsinki, for technical help with microdissection, and Eija Rinne, Department of Biosciences, University of Helsinki, for guidance in cryosectioning.