Dupouy Et Al - Technical Paper On Microfluidic Microchip
Dupouy Et Al - Technical Paper On Microfluidic Microchip
1
Institut de Biologie Moléculaire des Plantes (IBMP), CNRS, Université de Strasbourg,
Strasbourg, France
2
Université Aix Marseille, CEA, CNRS, BIAM, UMR7265, 13108, Saint-Paul-lez-
Durance, France
3
Institute of Microstructure Technology (IMT), Karlsruhe Institute of Technology (KIT),
Hermann-von-Helmholtz-Platz 1, Eggenstein-Leopoldshafen, Germany
4
Université de Strasbourg, CNRS, Architecture et Réactivité de l'ARN, UPR 9002,
Strasbourg, 67000 France
*: Corresponding authors:
marie-edith.chaboute@ibmp-cnrs.unistra.fr
alexandre.berr@ibmp-cnrs.unistra.fr
1
Abstract
nuclear dynamics relies notably on the connection of the nucleus with the cytoskeleton
and environmental cues. This is particularly true in the plant root system, which is
constantly exposed to a wide range of internal and external stimuli. Currently, studying
imaging for quantitative analyses under controlled conditions. Microfluidic systems for
plant cell studies are valuable analytical tools that provide a precise control of culture
treatments. As exemplified here, our microfluidic platform can be conveniently used for
thaliana roots.
2
1. Introduction
The nucleus is a complex and sophisticated organelle where major cellular processes,
including DNA protection, compaction, repair, replication and transcription, take place.
The nuclear content is isolated from other cell compartments by the nuclear envelope,
which acts as a key nexus between the cytoskeleton and the nuclear interior. As a
central hub, the nuclear envelope serves for signal sensing, transduction, molecule
domains (for review see [1][2]). Underneath the nuclear envelope, the nucleoskeleton
appears as a dynamic nuclear network, bridged to the cytoskeleton via the linker of
nucleoskeleton and cytoskeleton (LINC) complex, that ensures the overall structural
integrity of the nucleus and provides support for nuclear functions. The nucleus is
well as higher-order chromatin foldings [3]. In addition, the nucleus is also highly
dynamic. This is illustrated during each cell cycle with the assembly/disassembly of the
nuclear envelope, nuclear movements during the day/night cycles, changes in nuclear
shape and its mechanical properties, chromosome folding and positioning in the
Arabidopsis thaliana was widely used as a model species to track nuclear and
3
envelope, and telomeres often associated with the nucleolus [7][8]. Interestingly, the
response to a wide range of external and internal signals, which are used to regulate
various developmental and stress responses. Indeed, dynamic changes in the higher-
order nuclear organization were for example reported depending on light availability
[9], mechanical stress [10], heat stress [11], or even osmotic and salt stresses [12][13].
From these few examples, the nucleus was proposed to act as a rheostat to integrate
stress signals from outside to inside and modulate transcriptional control and
Yet, these data were mainly gained from fixed material at single time-point and should
therefore not depict a complete and comprehensive view of the nuclear dynamics from
its early to late response towards external stimuli, as well as its recovery upon return
to normal conditions. In this context, one of the challenges was to establish systems
for live-cell imaging allowing rapid and reversible treatments while ensuring the long-
term integrity of the living material. Towards this end, microfluidic devices were
quantitatively analyze images and finally establish biological principles [24]. Here we
report a method to setup a microfluidic system that enables dynamic live-cell imaging
system allows to easily and rapidly apply any treatments that can be added in solution
(biotic/abiotic stresses) via passive pumping and proper washout conducted in parallel
with long-term (i.e., up to 36h) high-resolution in vivo imaging. For example, drugs
affecting the integrity of key cytoskeleton components (for an exhaustive list see [25])
4
can be easily applied as exemplified in this protocol. Drugs known as “epidrugs” able
to target chromatin-related enzymes (for a review see [26]) can also be used to
network. Finally, the microfluidic system is also suitable to apply chemical agents such
as polyethylene glycol (PEG), mannitol, and salt (NaCl) to simulate osmotic stress. An
interesting outcome from this system is that the very first instants of a treatment can
be imaged. Another advantage of the system is that it is suitable to grow and image
seedlings at high temporal and spatial resolutions for up to 36-48 h. Moreover, the
system can be later recovered by simply replacing the feeding medium again with the
original one. Finally, as a demonstration, we show how our microfluidic system can be
2. Materials
4. Micro-milling machine equipped with milling tools (e.g., circular shaft mills with
6. Binocular microscope
7. Photomask
8. Silicon wafer
9. Acetone
5
10. Isopropyl alcohol
12. Compressed N2
assembly
3. Vacuum pump
4. Beaker with three dripless pouring spouts (e.g., Tripour Beaker, Haslab, or
similar)
6. Protective glasses
7. Adhesive tape
8. Nitrile gloves
12. Compressed N2
6
13. Plasma etcher
16. Oven
diameter)
20. Isopropanol
21. Acetone
4. Beakers
7
12. Cleaning wipes
15. pH meter
19. Liquid ½ MS medium: 2.2 g/l Murashige and Skoog basal medium, 0.5%
Sucrose, pH 5.7
20. Solid ½ MS medium: 2.2 g/l Murashige and Skoog basal medium 0.5%
3. Vacuum chamber
4. Lab marker
diameter)
8
11. Paper towel
13. Syringes 50 mL
14. Electric syringe pump system that can receive up to two syringes
15. Liquid ½ MS medium: 2.2 g/l Murashige and Skoog basal medium, 0.5%
Sucrose, pH 5.7
23. pH meter
2.5 Loading the microfluidic device with seedlings grown inside tips
7. Cleaning wipes
9
8. Nitrile gloves
9. Oven (65°C)
2. Liquid ½ MS medium: 2.2 g/l Murashige and Skoog basal medium, 0.5%
Sucrose, pH 5.7
3 Methods
This section describes two distinct procedures classically used to prepare molds that
will precisely mirror the microfluidic device. Preparing molds by micro-milling involves
the use of a high-speed micrometer milling tool to precisely cut patterns into a mold
material, in our case a synthetic polymer named Polymethylmethacrylate (PMMA) or
acrylic glass (see Note 1). Preparing molds by photolithography is based on the use
of a photoresist material, in our case SU-8 (see Note 2) that is patterned by exposure
to UV light through a mask. Both methods can produce high-quality molds with
extremely high resolution, high repeatability and relatively easy customization of molds
to fit specific design requirements. Unfortunately, both methods are complex and
require specialized/expensive equipment such as a clean room, as well as skilled
operators to perform effectively. Regarding technical aspects, photolithography can
produce molds with higher resolution than micro-milling with feature sizes as small as
a few hundred nanometers (compared to a few micrometers for micro-milling). Finally,
micro-milling is a highly versatile method since it can be applied to a wide range of
materials, including metals, plastics, and ceramics, while photolithography is limited to
photoresist materials. In the example provided here, the mold produced by micro-
milling contains three cultivation channels, each with a quadratic cross-section of 200
10
µm x 200 µm (Fig. 1a and c). In addition, two 0.8 mm diameter pillars are added per
channel at the bottom of the mold for both nutritional fluid in- and outlets. The seedling
inlet is milled in the chip after PDMS solidification. A single mold can be reused multiple
times and multiple different molds can be designed, for instance with a zig-zag
structure designed directly ahead of the cultivation channels to ensure the proper
mixing of several liquid media injected simultaneously (Fig. 1b). Exact parameters and
conditions for both micro-milling and photolithography may vary depending on the
specific materials and equipment used, as well as the desired pattern and resolution.
The entire procedure for both methods require from one to several hours depending
on the complexity of the mold design.
1. Design the 3D mold pattern using a Computer-Aided Design (CAD) software. The
pattern should include all features required for the microfluidic device, such as
channels or reservoirs at the microscale level. This step is common to both methods.
2. Mold preparation
2a. For micro-milling, generate a list of G-code commands (i.e., the programming
language specific for the milling machine) corresponding to the design using a
Computer-Aided Manufacturing (CAM) software compatible with the CAD software.
The G-code should specify the tool path, cutting speed, and cutting depth.
3a. Mount the mold material, in our case PMMA, onto the work table of the milling
machine using clamps. Load the milling tool with a diameter of 50 µm into the spindle
of the milling machine and set cutting parameters such as cutting speed, depth and
feed rate (see Note 3).
4a. Apply cutting fluid to the cutting area to lubricate and cool both the mold and the
tool during the milling process. Then, start the milling process and monitor the progress
to ensure that the mold features are being accurately machined.
11
5a. Once the milling process is complete, remove the mold from the machine and clean
it with a soft brush and compressed air to remove any debris or residue. Inspect the
mold under a binocular microscope to ensure that all features are properly machined
and that there are no defects or inaccuracies.
2b. The procedure described below provides a general guide for performing
photolithography using a single photomask (see Note 4). Prior photolithography,
photomasks printing should be performed (e.g., in our case one mask for the three
parallel main channels was sufficient) at high resolution (50,800 dpi) in negative mirror
and usually by a specialized company (e.g., Selba S.A.).
4b. Activate the wafer in a plasma cleaner (40 kHz, 100 Watts) for 30 seconds with 0.5
mbar of oxygen at 25% power and pre-bake it for 3 minutes at 95°C on a hot plate to
evaporate any remaining solvent.
5b. Dispense SU-8 2000.5 photoresist on the center of the activated wafer for a 0.5
µm layer (i.e., measure circa. 10g, see Note 5) and evenly distribute the photoresist
on the wafer surface using a spin coater following manufacturer instructions, first at
500 rpm for 5 seconds, then at 3,000 rpm for 30 seconds (see Note 6).
6b. Expose the wafer to UV light for 20 seconds (i.e., corresponding to an energy
exposure of 80 mJ/cm2) using a mask aligner and post-bake it for 3 min at 95°C to
further harden the SU-8 photoresist.
7b. Generate a second layer of 100 µm on the first 0.5 µm one by adding two 50 µm
layers of SU-8 2025. Pre-bake the wafer for 15 minutes at 95°C on a hot plate. Spread
the first 50 µm layer of the photoresist using a spin coater, first at 500 rpm for 5
12
seconds, then at 1,700 rpm for 30 seconds and expose it to UV light for 95 seconds
(i.e., corresponding to an energy exposure of 380 mJ/cm2) using the mask aligner with
the photomask on the top of the resist. Finally, post-bake the wafer for 10 min at 95°C.
8b. Remove unbound photoresist by placing the wafer on the spin coater, covering it
for 3 min with SU-8 Developer and spinning it 25 seconds at 2,500 rpm. Repeat the
development step 3 times and finally wash the wafer with isopropyl alcohol followed
by 25 seconds of spinning at 2,500 rpm still in the spin coater.
9b. Repeat the same procedure to add the second 50 µm layer with a pre-bake of 10
min and finally remove cracks of the photoresist by a hard bake for 10 minutes at
200°C. At this step, the design of the channels should be clearly visible (see Note 7).
assembly
devices (Fig. 1d and e). A mold precisely mirroring the microfluidic device design is
The PDMS casting process requires about two and a half hours, including the PDMS
polymerization time.
1. Thoroughly mix the PDMS elastomer base with the curing agent (i.e., the
catalytic agent which when added to resin causes polymerization) in a 1/10 ratio
properties of PDMS, the use of disposable fork and beaker for mixing is
preferable.
2. Pour the PDMS mixture into a mold beforehand placed into a Petri dish until
13
pour a 5 mm layer of PDMS in a second Petri dish that will serve to later
3. Degas the PDMS mixture by placing the mold under vacuum for 20 minutes.
The removal of air bubbles trapped in the PDMS mixture will prevent the
4. Cure the PDMS mixture in the mold in an oven at 65 °C for one hour.
5. Let the cast PDMS cool down to room temperature until solidification. Then,
very carefully demold PDMS microfluidic device using a clean scalpel blade
while avoiding to damage both the mold and the microfluidic device. Small
pieces of adhesive tape can be used to remove PDMS debris and dust. Wear
protective glasses during the demolding procedure to prevent any injuries due
to eventual flying sparkles and wear nitrile gloves to avoid leaving fingerprints
6. Drill straight through the PDMS microfluidic device at locations coincident with
the above-mentioned pillars using a drill equipped with a 1 mm diameter drill bit
to open the nutritional fluid in- and outlets. Do the same for the seedling inlets
by applying a 45° angle. Such angle will later facilitate the root seedling entering
7. Before final assembling, clean the PDMS microfluidic device from any drilling-
water bath for 5 minutes each. Do not use solvents such as acetone as PDMS
swells in it.
minutes each.
14
9. Carefully dry all parts using compressed N2.
10. Place the PDMS microfluidic device and the glass cover with their bonding
surfaces facing up inside a plasma etcher and activate bonding surfaces with a
11. Open the etching chamber and immediately assemble the glass cover on the
12. Place the microfluidic device in an oven at 65 °C for 15 minutes to finalize the
assembly.
13. Test the microfluidic device for blockages and leakages by connecting it to a
syringe electric pump and pumping isopropanol at a flow rate of 500 µL / min
14. Finally, dry the functional microfluidic devices with compressed N2 and package
ethanol for sterilization. The durability of the bond between the PDMS and the
This section describes the procedure for germination and early cultivation of plantlets
inside ½ MS agar medium filled pipette tips, a preliminary step necessary before
loading the microfluidic device (Fig. 2a-d). The entire procedure requires about an
hour and should be carried out under sterile conditions in a laminar flow hood. Be
aware that the microfluidic device described here was configured to allow the growth
of a total of three plantlets in parallel per device. The model of electric syringe pump
15
listed in the “Material” section can receive up to two syringes, for example to supply
1. For each fluorescent Arabidopsis tagged line (see Note 11), sterilize
dish (i.e., one Petri dish per fluorescent Arabidopsis tagged line).
4. Wait 20 seconds for the ½ MS agar medium to solidify inside tips and finally
eject filled pipette tips aside in a second sterile and empty Petri dish. Prepare a
line analyzed.
5. When the required number of pipette tips is prepared, cut each tip just below
the top level of the solidified ½ MS agar medium using a sterile scalpel to obtain
6. Plant cut tips inside the ½ MS agar medium filled Petri dish (see step 3 in section
3.3) using a fine straight tweezer in order to maintain them in a vertical position.
The use of a tall Petri dish is recommended here to avoid any contact between
7. At this stage, sterilized seeds must be dried. Drop one seed per cut tip on the
top of the agar surface using a tweezer or a sterilized toothpick. When all cut
tips are loaded, close the Petri dish and seal it with micropore surgical tape.
16
8. For stratification, place Petri dishes containing cut tips loaded with seeds at 4°C
9. After stratification, transfer Petri dishes to a long day (16 h light, 8 h night, 21°C)
in vitro growth chamber for 4-5 days until root tips reach the edge of cut tips
(Fig. 2d).
This section describes the assembly procedure allowing to connect the microfluidic
device to the electric syringe pump (Fig. 2e-j). The entire procedure should be carried
out under sterile conditions in a laminar flow hood and requires about one and a half
hours.
and by placing the device in a vacuum chamber for at least 5 minutes. Besides
sterilizing the device, this 70% ethanol bath will also later prevent the formation
2. While the sterilized microfluidic device dries, continue by preparing the Petri
dish that will serve as a container for the microfluidic device to keep sterile and
humid conditions during the entire experimental process. Label with a marker
on the lid of a Petri dish the positions of the in- and outlet tubing of the
hypodermic needle while ensuring that the diameter of the holes is about the
outer diameter of the PTFE microtubing (i.e., slightly higher than 1.07 mm; Fig.
2e).
3. Continue by preparing the bubble trap system. Slice a PDMS cylinder from the
Petri dish prepared at step 3.1.2 using a 7 mm biopsy punch (i.e., corresponding
17
to the internal diameter of a 0.65 mL microcentrifuge tube) and drill 4 holes in it
(i.e., in the microfluidic device presented here, 3 x 20 cm long for inlet tubing
5. Using the same sterile surgical blade, cut an additional microtubing that will
serve as a primary tubing connected at one of its ends to the syringe. Be aware
that the length of this tube may vary depending on the space available at your
acquisition tool (e.g., a 20 cm length primary tube for a LSM700 Laser Scanning
Confocal Microscope).
6. Using a tweezer, insert the three 20 cm inlet tubing and the primary tubing into
the holes of the PDMS cylinder prepared at step 3 section in 3.4 and let tubing
protruding from the PDMS at one side of the cylinder for about 0.5 cm long (Fig.
2f).
7. Insert the resulting PDMS cylinder in a 0.65 mL microcentrifuge tube like a cork
with the 0.5 cm long excess tubing at the inside of the microcentrifuge tube (Fig.
2f).
8. Using a tweezer, pass the other end of the in- and outlet tubing through the
holes drilled in the Petri dish lid prepared at step 2 in section 3.4 in order to
leave around 1 cm of microtubing length on the interior side of the Petri dish.
Still from the interior side of the Petri dish, tubing can be cut with a 45° angle to
help inserting them inside the PDMS-based microfluidic device (Fig. 2g).
18
9. Assemble all parts by carefully pushing the in- and outlets tubing from the Petri
dish lid inside their respective in- and outlets holes on the PDMS microfluidic
10. Dispose a piece of autoclaved paper towel at the periphery of the Petri dish
while leaving enough space at its center to later dispose of the microfluidic
device.
11. Add sterile water to the paper towel contained in the Petri dish to keep humidity
in the system and place the PDMS microfluidic device inside (Fig. 2i).
12. Fill a 50 mL syringe with ½ MS liquid medium, place a needle at its tip and
13. Using a tweezer, carefully push the other end of the primary tubing onto the
needle. Be careful not to damage the tubing with the needle tip.
14. Setup the liquid flow rate at 2,000 µL / minute and fill the entire microfluidic
15. Carefully control whether all tubing and channels are filled completely and all
air bubbles are removed. If necessary close the Petri dish with micropore
16. Stop the electric syringe pump and continue with the final assembling step of
3.5 Loading the microfluidic device with seedlings grown inside tips
This section describes the assembly of seedlings grown inside tips on the microfluidics
system. This final assembly requires 5-10 minutes and should be carried out under
sterile conditions in a laminar flow hood. Be careful not to let root tips grow beyond the
tip cone edge to prevent any damage or additional stress during this final assembly. A
19
regular control of seedling growth using a binocular microscope or a magnifying glass
is highly recommended. Usually, around 20% of the prepared cut tips loaded with a
seed can be used for following steps due to proper seedling growth.
1. Gently remove a single cut tip holding a properly grown seedling from the agar
plate (i.e., for better grip, avoid extra thin tweezers) and carefully insert it,
exerting a gentle pressure along a spiral path, inside one of the seedling inlet
holes. Be careful not to insert it too deep as this could break the underneath
2. Repeat this step as many times as necessary to fill each seedling inlet hole
available on the microfluidic device (i.e., three in the example shown in this
protocol).
3. Close the Petri dish containing the fully assembled microfluidic device and wrap
4. Transfer the system in an in vitro growth chamber and set the flow rate to 350
µL / min for 30 secondes while flicking the outlet tubing to remove any air bubble
5. Position the syringe electric pump at the same height compared to the Petri dish
containing the fully assembled microfluidic in order to maintain the inlet tubing
as vertically as possible and tilt the Petri dish at an angle of approximately 45°
to facilitate the entrance of the root inside the observation channel (fig. 2j).
6. Finally, reduce the flow rate to 8 µL / minute and let seedlings grow (see Note
13).
20
This section describes the real-time imaging procedure of root-grown inside the
microfluidic device. Be aware that the microfluidic system presented here was
optimized for an inverted microscopy platform and that the settings have to be adjusted
channels (i.e., usually after 3-4 days of growth), the system can be used for
imaging.
2. Be careful to keep the microfluidic system plugged to its fluid supply during the
3. Unwrap the Petri dish and open it in order to place the microfluidic device on
the microscope platter with the glass cover facing downward (see Note 14).
syringe containing ½ MS medium with the drug (at the concentration of interest)
on the second slot of the electric syringe pump (see Note 15).
6. Move the needle connected to the primary tubing from the syringe containing
the control ½ MS medium to the other one containing the drug and remove the
old syringe.
minute for a 1 min period in order to entirely replace the control medium by
minute.
21
9a. using a confocal microscope LSM 700. Perform a Z-stack every minute at high
speed (e.g., circa 20 slices with a step size of 1 µm every minute for 20 minutes
medium-of-interest for the study) followed by 90 minutes with one Z-stack every
three minutes (recovery from the treatment). Set the excitation parameters and
Airy unit.
9b. using a confocal spinning disk. The scan speed can be much faster in
This allows higher temporal resolution with images (circa 20 slices, 1 µm z-step)
stable over 5-10 minutes imaging time, for GFP and RFP reporter proteins. This
increasing temporal resolution increases the size of the image file. This is
10. Proceed to image analysis using ImageJ or Fiji and quantify parameters of
22
imaged in 3D [29][30]. Note that quantitative analyses require careful
and density) can be performed directly using ImageJ [31] or through plugins
The system described here enables us to dynamically follow the impact of any
envelope markers was followed during treatment of Arabidopsis roots with low
SUN2-RFP localized to the nuclear envelope, while the tagged centromeric specific
Using these markers, we can follow changes in the nuclear shape as well as changes
parts, including the root cap, elongation zone, mature zone and also root hairs. Finally,
4. Notes
23
1. As a versatile material, PMMA has a number of properties that make it an attractive
material. It has an excellent optical clarity as well as a high resistance to impact, UV
radiation and many chemicals including acids and alkalis.
2. SU-8 is a commonly used negative-tone epoxy-based photoresist material used in
microfabrication processes that has a thickness range of 5-200 μm and a high chemical
resistance and mechanical strength. Exposure to ambient light should be minimized
since SU-8 is sensitive to the UV contained in it, which can cause unintended exposure
and reduce the contrast of the pattern. Rather use a yellow or red safe light in the
lithography area and cover the photoresist-coated substrate when transferring it
between equipment.
3. Cutting parameters should be carefully selected to achieve the desired results since
they can affect the quality of the cut, the accuracy of the mold features, and the lifespan
of the cutting tool.
4. Typically, a photomask can be used in the range of 50 to 100 cycles. It is important
to verify the quality of the photomask by inspection or metrology before re-using it.
After photolithography and prior reuse, rinse the photomask with isopropyl alcohol to
ensure complete removal of any residual photoresist or contaminants. Finally, dry the
photomask with nitrogen gas.
5. The SU-8 resin being highly viscous, the easiest way to pour it onto the wafer is
directly from the bottle. Dispense a volume equivalent to 10g for uniform coating. A too
small volume can result to a highly uneven surface after spinning.
6. There are typically two spin coating steps during the photolithography process to
ensure a uniform and controlled thickness of the photoresist layer on the wafer surface
and minimize defects such as bubbles and edge bead. The first spin coating step is
referred to as the "prime" or "adhesion" layer while the second is referred to as the
"main" or "photoresist" layer. Spin coating steps are following manufacturer
instructions.
7. To facilitate the subsequent removal of cast PDMS, few droplets of
tridecafluoro1,1,2,2-tetrahydrooctyl-1-trichlorosilane can be added on the surface of
the mold.
24
8. Plasma etching is a dry-etching technique that combines the principles of physical
etching due to the impact of non-reactive ions on the material’s surface and the
principles of chemical etching due to chemical reactions between the etching gas and
the material surface. If the etch time is kept short, it can be used for activating surfaces
as well. Such a surface activation is done for bonding PDMS to glass (i.e., both having
similar surface-chemistry). Through activation by an O2-plasma etching process PDMS
forms Silanol-groups (Si-OH) to which glass often binds itself. When the activated
glass and PMDS are brought together e.g., by pressing them together, the Si-OH-
groups of glass and PDMS condense and form strong intermolecular bonds through
newly formed Si-O-Si-groups.
9. PDMS not sticking to the glass cover could result from glass cover and/or PDMS
surfaces are not clean enough. To avoid this, make sure that the coverslip has been
thoroughly cleaned with water and that both glass cover and PDMS surfaces have
autoclaving as it can provoke swelling that may cause leakage rendering the chip
unusable.
11. Practically, all transgenic plants with translational fusion of a protein-of-interest with
a fluorescent tag can be used, provided that the fusion protein is expressed at least in
root tissues. Below is a list of possible Arabidopsis transgenic plants expressing fusion
Nucleoporins [37].
25
3. Nucleoplasm makers: such as tagged histones for the H3 isoforms [41]; the
12. Agar can dry up too quickly in the tip after sowing if the agar surface in the tip is
not flat when cutting the tip after filling with ½ MS medium. Therefore, make sure to
13. Contamination may occur during microfluidic device plugging or syringe exchange.
To avoid high contamination rate, make sure to cut tubing with a scalpel beforehand
soaked in 70% ethanol and flame sterilized. When changing the syringe make sure to
keep the pump surroundings sterile. Finally, make sure that the coverslip is intact
14. Images can appear not steady during acquisition under a microscope. PDMS being
a squishy material the positioning of the microfluidic device on the microscopic plate
can be tricky to keep it steady. Make sure that the PDMS block is shorter than the
15. When applying precious drugs or purified bioactive compounds with limited
availability, prefer using smaller syringes (< 50 mL) with an adapted pump system.
Acknowledgments
The work needed to establish this protocol was funded by the Centre National de la
Recherche Scientifique (CNRS), the Human Frontier Science Program (HFSP, grant
26
project REWIRE and ANR-20-CE13-0025 project MechaNUC). In addition, this work
was also partly funded by the European Fund for Regional Development (Interreg
Upper Rhine, project DialogProTec, coordinator: Prof. Peter Nick, Botanical Institute,
References
27
Arabidopsis. Plant Cell 22:3118–3129. https://doi.org/10.1105/tpc.110.078493
12. Goswami R, Asnacios A, Milani P, Graindorge S, Houlné G, Mutterer J, Hamant
O, Chabouté M-E (2020) Mechanical Shielding in Plant Nuclei. Current Biology
30:2013-2025.e3. https://doi.org/10.1016/j.cub.2020.03.059
13. Wang Z, Casas-Mollano JA, Xu J, Riethoven J-JM, Zhang C, Cerutti H (2015)
Osmotic stress induces phosphorylation of histone H3 at threonine 3 in pericentromeric
regions of Arabidopsis thaliana. Proc Natl Acad Sci U S A 112:8487–8492.
https://doi.org/10.1073/pnas.1423325112
14. Fal K, Asnacios A, Chabouté M-E, Hamant O (2017) Nuclear envelope: a new
frontier in plant mechanosensing? Biophys Rev 9:389–403.
https://doi.org/10.1007/s12551-017-0302-6
15. Goswami R, Asnacios A, Hamant O, Chabouté M-E (2020) Is the plant nucleus
a mechanical rheostat? Curr Opin Plant Biol 57:155–163.
https://doi.org/10.1016/j.pbi.2020.09.001
16. Busch W, Moore BT, Martsberger B, Mace DL, Twigg RW, Jung J, Pruteanu-
Malinici I, Kennedy SJ, Fricke GK, Clark RL, Ohler U, Benfey PN (2012) A microfluidic
device and computational platform for high-throughput live imaging of gene
expression. Nat Methods 9:1101–1106. https://doi.org/10.1038/nmeth.2185
17. Fendrych M, Akhmanova M, Merrin J, Glanc M, Hagihara S, Takahashi K,
Uchida N, Torii KU, Friml J (2018) Rapid and reversible root growth inhibition by TIR1
auxin signalling. Nat Plants 4:453–459. https://doi.org/10.1038/s41477-018-0190-1
18. Grossmann G, Guo W-J, Ehrhardt DW, Frommer WB, Sit RV, Quake SR, Meier
M (2011) The RootChip: an integrated microfluidic chip for plant science. Plant Cell
23:4234–4240. https://doi.org/10.1105/tpc.111.092577
19. Massalha H, Korenblum E, Shapiro OH, Aharoni A (2019) Tracking Root
Interactions System (TRIS) Experiment and Quality Control. Bio Protoc 9:e3211.
https://doi.org/10.21769/BioProtoc.3211
20. Pereira D, Alline T, Singh G, Chabouté M-E, Asnacios A (2023) A Microfluidic-
Like System (MLS) to Grow, Image, and Quantitatively Characterize Rigidity Sensing
by Plant’s Roots and Root Hair Cells. In: Zaidel-Bar R (ed) Mechanobiology: Methods
and Protocols. Springer US, New York, NY, pp 121–131
21. Singh G, Pereira D, Baudrey S, Hoffmann E, Ryckelynck M, Asnacios A,
Chabouté M-E (2021) Real-time tracking of root hair nucleus morphodynamics using
a microfluidic approach. The Plant Journal 108:303–313.
https://doi.org/10.1111/tpj.15511
22. Sun L, Liu L, Lin X, Xia Z, Cao J, Xu S, Gu H, Yang H, Bao N (2021) Microfluidic
Devices for Monitoring the Root Morphology of Arabidopsis Thaliana in situ. ANAL SCI
37:605–611. https://doi.org/10.2116/analsci.20P281
23. Kaiser C-F, Perilli A, Grossmann G, Meroz Y (2023) Studying root-environment
interactions in structured microdevices. J Exp Bot erad122.
https://doi.org/10.1093/jxb/erad122
24. Colin L, Martin-Arevalillo R, Bovio S, Bauer A, Vernoux T, Caillaud M-C,
Landrein B, Jaillais Y (2022) Imaging the living plant cell: From probes to quantification.
The Plant Cell 34:247–272. https://doi.org/10.1093/plcell/koab237
25. Caillaud M-C (2022) Tools for studying the cytoskeleton during plant cell
division. Trends in Plant Science 27:1049–1062.
https://doi.org/10.1016/j.tplants.2022.05.006
26. Fal K, Tomkova D, Vachon G, Chabouté M-E, Berr A, Carles CC (2021)
Chromatin Manipulation and Editing: Challenges, New Technologies and Their Use in
Plants. International Journal of Molecular Sciences 22:512.
28
https://doi.org/10.3390/ijms22020512
27. Dubos T, Poulet A, Gonthier-Gueret C, Mougeot G, Vanrobays E, Li Y, Tutois
S, Pery E, Chausse F, Probst AV, Tatout C, Desset S (2020) Automated 3D bio-
imaging analysis of nuclear organization by NucleusJ 2.0. Nucleus 11:315–329.
https://doi.org/10.1080/19491034.2020.1845012
28. Dubos T, Poulet A, Thomson G, Péry E, Chausse F, Tatout C, Desset S, van
Wolfswinkel JC, Jacob Y (2022) NODeJ: an ImageJ plugin for 3D segmentation of
nuclear objects. BMC Bioinformatics 23:216. https://doi.org/10.1186/s12859-022-
04743-6
29. Dumur T, Duncan S, Graumann K, Desset S, Randall RS, Scheid OM,
Prodanov D, Tatout C, Baroux C (2019) Probing the 3D architecture of the plant
nucleus with microscopy approaches: challenges and solutions. Nucleus 10:181–212.
https://doi.org/10.1080/19491034.2019.1644592
30. Randall RS, Jourdain C, Nowicka A, Kaduchová K, Kubová M, Ayoub MA,
Schubert V, Tatout C, Colas I, Kalyanikrishna, Desset S, Mermet S, Boulaflous-
Stevens A, Kubalová I, Mandáková T, Heckmann S, Lysak MA, Panatta M, Santoro R,
Schubert D, Pecinka A, Routh D, Baroux C (2022) Image analysis workflows to reveal
the spatial organization of cell nuclei and chromosomes. Nucleus 13:279–301.
https://doi.org/10.1080/19491034.2022.2144013
31. Higaki T (2017) Quantitative evaluation of cytoskeletal organizations by
microscopic image analysis. Plant Morphology 29:15–21.
https://doi.org/10.5685/plmorphol.29.15
32. Boudaoud A, Burian A, Borowska-Wykręt D, Uyttewaal M, Wrzalik R,
Kwiatkowska D, Hamant O (2014) FibrilTool, an ImageJ plug-in to quantify fibrillar
structures in raw microscopy images. Nat Protoc 9:457–463.
https://doi.org/10.1038/nprot.2014.024
33. Nakamura M, Naoi K, Shoji T, Hashimoto T (2004) Low Concentrations of
Propyzamide and Oryzalin Alter Microtubule Dynamics in Arabidopsis Epidermal Cells.
Plant and Cell Physiology 45:1330–1334. https://doi.org/10.1093/pcp/pch300
34. Graumann K, Runions J, Evans DE (2010) Characterization of SUN-domain
proteins at the higher plant nuclear envelope. Plant J 61:134–144.
https://doi.org/10.1111/j.1365-313X.2009.04038.x
35. Tamura K, Iwabuchi K, Fukao Y, Kondo M, Okamoto K, Ueda H, Nishimura M,
Hara-Nishimura I (2013) Myosin XI-i Links the Nuclear Membrane to the Cytoskeleton
to Control Nuclear Movement and Shape in Arabidopsis. Current Biology 23:1776–
1781. https://doi.org/10.1016/j.cub.2013.07.035
36. Zhou X, Graumann K, Wirthmueller L, Jones JDG, Meier I (2014) Identification
of unique SUN-interacting nuclear envelope proteins with diverse functions in plants.
Journal of Cell Biology 205:677–692. https://doi.org/10.1083/jcb.201401138
37. Tamura K, Fukao Y, Iwamoto M, Haraguchi T, Hara-Nishimura I (2010)
Identification and Characterization of Nuclear Pore Complex Components in
Arabidopsis thaliana. The Plant Cell 22:4084–4097.
https://doi.org/10.1105/tpc.110.079947
38. Goto C, Tamura K, Fukao Y, Shimada T, Hara-Nishimura I (2014) The Novel
Nuclear Envelope Protein KAKU4 Modulates Nuclear Morphology in Arabidopsis. The
Plant Cell 26:2143–2155. https://doi.org/10.1105/tpc.113.122168
39. Sakamoto Y, Sato M, Sato Y, Harada A, Suzuki T, Goto C, Tamura K, Toyooka
K, Kimura H, Ohkawa Y, Hara-Nishimura I, Takagi S, Matsunaga S (2020) Subnuclear
gene positioning through lamina association affects copper tolerance. Nat Commun
11:5914. https://doi.org/10.1038/s41467-020-19621-z
29
40. Batzenschlager M, Lermontova I, Schubert V, Fuchs J, Berr A, Koini MA,
Houlné G, Herzog E, Rutten T, Alioua A, Fransz P, Schmit A-C, Chabouté M-E (2015)
Arabidopsis MZT1 homologs GIP1 and GIP2 are essential for centromere architecture.
Proceedings of the National Academy of Sciences 112:8656–8660.
https://doi.org/10.1073/pnas.1506351112
41. Otero S, Desvoyes B, Peiró R, Gutierrez C (2016) Histone H3 Dynamics Reveal
Domains with Distinct Proliferation Potential in the Arabidopsis Root. Plant Cell
28:1361–1371. https://doi.org/10.1105/tpc.15.01003
42. Rutowicz K, Puzio M, Halibart-Puzio J, Lirski M, Kroteń MA, Kotliński M,
Kniżewski Ł, Lange B, Muszewska A, Śniegowska-Świerk K, Kościelniak J, Iwanicka-
Nowicka R, Żmuda K, Buza K, Janowiak F, Jõesaar I, Laskowska-Kaszub K, Fogtman
A, Zielenkiewicz P, Tiuryn J, Kollist H, Siedlecki P, Ginalski K, Świeżewski S,
Koblowska M, Archacki R, Wilczyński B, Rapacz M, Jerzmanowski A (2015) A
specialized histone H1 variant is required for adaptive responses to complex abiotic
stress and related DNA methylation in Arabidopsis. Plant Physiol pp.00493.2015.
https://doi.org/10.1104/pp.15.00493
43. Lermontova I, Schubert V, Fuchs J, Klatte S, Macas J, Schubert I (2006)
Loading of Arabidopsis Centromeric Histone CENH3 Occurs Mainly during G2 and
Requires the Presence of the Histone Fold Domain. The Plant Cell 18:2443–2451.
https://doi.org/10.1105/tpc.106.043174
44. Yelagandula R, Stroud H, Holec S, Zhou K, Feng S, Zhong X, Muthurajan UM,
Nie X, Kawashima T, Groth M, Luger K, Jacobsen SE, Berger F (2014) The Histone
Variant H2A.W Defines Heterochromatin and Promotes Chromatin Condensation in
Arabidopsis. Cell 158:98–109. https://doi.org/10.1016/j.cell.2014.06.006
45. Staiger CJ, Sheahan MB, Khurana P, Wang X, McCurdy DW, Blanchoin L
(2009) Actin filament dynamics are dominated by rapid growth and severing activity in
the Arabidopsis cortical array. Journal of Cell Biology 184:269–280.
https://doi.org/10.1083/jcb.200806185
46. Smertenko AP, Deeks MJ, Hussey PJ (2010) Strategies of actin reorganisation
in plant cells. Journal of Cell Science 123:3019–3028.
https://doi.org/10.1242/jcs.071126
47. Marc J, Granger C, Brincat J, Fisher D, Kao T, McCubbin A, Cyr R (1998) A
GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living
epidermal cells. Plant Cell 10:1927–1940
48. Shaw SL, Kamyar R, Ehrhardt DW (2003) Sustained Microtubule Treadmilling
in Arabidopsis Cortical Arrays. Science 300:1715–1718.
https://doi.org/10.1126/science.1083529
30
Figure legends
Figure 1: Molding of the PDMS microfluidic device. (a) Mold used for casting a replica
of the microfluidic device adapted to the simultaneous growth of three seedlings. The
zoom-ring shows one of the 0.8 mm pillars that will serve as a nutritional fluid outlet.
(b) Example of an alternative mold design in which three inlet-channels for three
different substances converging into a single mixing channel with a “zig-zag” structure
designed for an optimal substance mixing. The zoom-ring shows the 0.8 mm pillars
that will serve as a seedling root inlet upstream of the cultivation/observation channel.
(c) Scheme of the PDMS microfluidic device obtained from the mold presented in (a).
This device contains three cultivation/observation channels in which seedling roots are
grown. Each cultivation/observation channel has a nutritional fluid inlet on one side
and a nutritional fluid outlet on the other side for connecting nutritional fluid inlet and
opening for inserting a tip in which a seedling is grown. Each opening forms a 45°
angle with its cultivation/observation channel. (d) Top (left) and side (right) views of the
PDMS-based microfluidic device bonded via plasma bonding to a cover glass on the
observation side.
Figure 2: Assembly of a microfluidic device for root imaging. 10 µL tips are filled with
melt ½ MS agar medium (a), cut just above the 2 µL mark (b) and plant vertically on a
solidified ½ MS agar plate before loading sterilized seeds on it. (d) Tips containing 4-
5-day-old seedlings just before transfer on the microfluidic device. (e) Preparation of
the Petri dish that will serve as a container for the microfluidic device. (f) Assembly of
the bubble trap system from a PDMS cylinder (top left), that will be inserted in a 0.65
31
mL microcentrifuge tube (middle left) after inserting the three inlet tubings and the
primary tubing (bottom left). (g) Follow-up of the tubing mounting, including the three
outlet tubings, through the Petri dish lid prepared in (e). (h) Connection of the PDMS
microfluidic device to the different in- and outlet tubing. (i) Microfluidic device
periphery of the Petri dish to keep the system humid. (j) Microfluidic device loaded with
a seedling grown inside a tip, connected to an electric syringe pump and placed in an
in vitro growth chamber before imaging. (k) Microfluidic real-time imaging on a confocal
LSM 700 (Zeiss) microscope. (l) Scheme of the microfluidic system in its entirety.
envelope markers in meristematic cells during main root growth inside a PDMS
microfluidic device. The scans were performed at regular intervals after oryzalin
treatment using low concentration (200 nM) for 10 hours. The nuclear envelope
walls (also in red). Numbers highlight connected cells which can be followed
throughout the scanning, showing progressive nuclear deformation and changes in the
32