Abiotic stress-related research in plants has considerably increased in recent years as a result of the constant change in the global climate conditions [1
]. Plant growth under stress conditions is generally phenotyped and visualized by macroscale parameters [3
], which requires dedicated greenhouse space, labour, a great deal of test sample and consumables, thus, limits the number of parallel experiments. Also, conventional plant growth techniques are not always compatible with state-of-the-art characterization tools, such imaging tools due to the optical transparency issues of the soil pots preventing microscopic analyses at high-resolution, and the out of plane growth on agar plates hindering imaging on a single plane of focus. Engagement of microfabricated fluidic systems with plant biology research has paved the way for large-scale plant growth studies and precise morphological and physiological analyses at microscale with reduced cost and labour [4
]. Those pioneering devices have allowed miniaturisation of the individual experiments and related costs while providing automated parallel assays to achieve accurate as well as high-throughput quantitative data [4
]. Arabidopsis thaliana [6
], Camellia Japonica [14
], Oryza sativa [18
], Nicotiana tabacum [19
], Phalaenopsis Chiada Pioneer [21
], and Physcomitrella patens [22
] were the plants that have been employed in various microfluidic platforms for in-depth optical analysis of the dicot seed germination, leaf development, cell phenotypes, protoplasts, pollen tube development and dynamics, shoot and the root growth. As summarized in Table 1
, these studies have primarily been carried out in dicot plants and studies on monocot plants are still to be explored.
Roots are responsible for water and mineral nutrient uptake from the soil. They offer structural stability to the plant and affect the growth and development of the plant organs above the soil. Characterization of the root behaviour at different developmental stages and under various environmental conditions is of great importance to reveal the plant tolerance mechanisms and dynamic changes, especially the ones taking place in the root systems of important food cereal crops such as wheat, rice, maize, and barley. However, conventional techniques for root investigations are usually conducted at the macro scale and do require relocation of the plants for microscopic analyses, causing dehydration or damage, thus, the data shortage. Also, the conventional tools do not allow real-time observation of the changes in the root systems that are exposed to different stress conditions such as drought, salt, growth factors, drug or nanomaterials.
To analyze temporal changes occurring under osmotic pressure and to maintain continuous imaging data during the stress application in developing seedlings, the long-term analysis in a proper setup is a requisite. To allow such imaging, the specimen must be under continuous nutrient supply while it is simultaneously accessible for visualization. While several studies of young Brachypodium roots have been carried out [23
] the real-time optical imaging of developing roots under stress has not been reported so far, due to the technicalities mentioned above. Polydimethylsiloxane devices offer several advantages to create platforms for manipulating thick tissues to subsequently allow live imaging of cellular dynamics. These platforms can be created into desirable designs, size and tailor-made structures for the sample under observation to allow high throughput imaging data. These PDMS chips can be made into high-resolution microfluidic chips to visualize cellular dynamics or molded into larger organ-on-a-chip for tissue and plant-on-a-chip for whole plant growth analysis [12
]. To capture long-term adaptations of plant roots to different microenvironments several root chip microfluidic platforms have also been introduced.
Recently, several chip platforms have been reported for tip-growing cells, such as in root and pollen tube growth, microenvironment investigation mostly in ornamental dicotyledonous plants [14
]. Although all these systems attempt to mimic the physical microenvironment and provide appropriate designs for analysing spherical seeds or pollen tube elongation, there exists a need for a platform capable of measuring the elongation and growth dynamics of a model monocot plant which differs considerably in its seed architecture and germination behaviour at the cellular level. The application of abiotic stress conditions at the microscale to the monocot seeds may allow phenotyping of the most important staple food crops (having elliptical slender long grains) and be a valuable resource for a better understanding of the crop adaptation with high precision. While the analyses of several plant organs, organogenesis, pattern formation, growth dynamics have been observed in microenvironments, the structures observed have been at the cellular level with few dicotyledonous species observed Table 1
. There has been little manipulation of monocot grain seedlings in such systems partly due to the size, scale and morphological intricacies of most grasses. Monocotyledonous seeds are usually elliptical, long slender grains, with embryo polarity which makes the germination behavior at the tissue and cellular level distinct from dicotyledons.
Here we report the growth of monocot seeds from model plant Brachypodium distachyon [3
] in a polydimethylsiloxane (PDMS) based microfluidic channel in which the effects of abiotic stress on the root development were investigated in real time with various microscopy studies. The Brachypodium root has emerged as a feasible model system to study cereals organogenesis since grain grasses have either huge roots, e.g., maize, multiple roots, e.g., wheat or specialized water conditions, e.g., rice [30
]. It offers various advantages for live imaging such as its simple architecture with a single primary axile root until 3 leaf stage, its moderate transparency, small size and steady growth rate [24
]. The microfluidic channel system allowed the positioning of monocot Brachypodium seeds at the serially arranged microchannels where the root-cell microenvironment can be precisely controlled, watered, visualised in real-time, and desired stress conditions can be established. Earlier microscopic studies have been done on the morphology [32
], growth [34
] and development [35
] of Brachypodium and our study focuses on the real-time growth dynamics and drought conditions in young seedlings.
Growth, directionality and compatibility were observed for Brachypodium seeds on all three PDMS punched molds, and the results were in line with the previous reports conducted with Nicotiana and Arabidopsis [12
]. After several experiments, we concluded that after 4 days of vernalization and 2 DAG seedling stage, the seedling had to be inserted in the correct orientation in the chip to make it grow along the length of the narrow 1 mm channel. Growth was observed with the root penetrating the length of the microchannel with a slight curvature and bending.
A study on young wheat seedlings shows cell wall expansion in the maturation zone upon a low water potential around the roots and the authors suggest the accumulation of some solutes within the elongation and maturation zones in order to maintain the turgor pressure, resulting in an increase in the root diameter [46
]. Although not seen in maturation zone cells, but a similar swelling behaviour of cells at the root apical meristem zone upon treatment with 5% PEG was previously reported for Brachypodium as well as wheat, rice, soybean, and maize [47
], suggesting a collective response by root tissues of different plants to surmount the osmotic stress. The observation of Casparian strips in the endodermis is in line with results on waterlogging in Brachypodium since both stagnant water and dehydration with PEG are osmotic pressures on the cells [23
Our results show the ease of visualization and utilization of neutral red as a promising dye for monocot PEG-mediated stress without tissue processing. Since Neutral red stains casparian tubes it is an excellent vital stain to be used in monocots for in situ analysis without processing the tissue or sample preparation.
The Di19 protein was first reported in Oryza sativa as OsDi19-1 through RNA-Seq data. In Triticum aestivum, it was known to have a C terminal domain. At the N terminus is a zinc finger zf-Di19. It was reported to be involved in environmental stress, i.e., cold, drought, osmotic stress, and salinity. The protein was induced by high levels of abscisic acid and ethylene [51
]. Though spatiotemporal expression analysis the stem root and leaf overall showed an increase in expression of OsDi19 as compared to all other tissues [52
]. Under flooding and osmotic pressure, the expression also increased compared to other stress conditions such as salinity, cold, dryness, cadmium and hormones [53
]. BdDi19 is thus a little-known gene which has a profound expression during short-term osmotic stress in young seedlings. Further studies on this gene and protein in Poaceae could be useful in understanding dehydration and osmotic stress in relation to development.
This is the first study to incorporate neutral red for Brachypodium morphological analysis; this is also the first study to analyse PEG-mediated stress with neutral red dye. Moreover, this is the first study to experimentally validate Dehydration Induced 19 protein and analyse the expression levels of BdDi19 under drought stress.
We propose that a follow-up study on Brachypodium seedlings in automated microfluidics or bioMEMS devices can be built upon our observations. Organ growth dynamics, root elasticity, microfluidic flow analysis and root hair dynamics under real time are a few of several areas which are desirable to be undertaken to unravel yet unknown physical patterns of growth and growth cessation and adaptation in favourable and stresses conditions.
4. Materials and Methods
4.1. Device Fabrication
Rectangular PDMS pieces with a scale of 65 × 20 × 10 mm single, double and triple punched with 5 mm diameter punchers were initial seed growing reservoirs at different volumes to check biocompatibility. Acetone cleaned glass slides, and PDMS pieces were plasma treated and bonded to get the final devices, which were used to test the compatibility of Brachypodium seeds with PDMS. A mold for the plant chip was designed with SOLIDWORKS Software, reproduced onto ABS 3D material, and 3D printed. The mold dimensions were 10 mm height, 9.5 mm channel length, 1 mm outlet diameter, and each seed channel 4 mm in diameter. The channel height was fixed at 1 mm to ensure the growth of the root to remain in one plane and not be out of focus in the Z-axis under microscopy as was earlier observed for 2 mm channel. For the construction of the device, PDMS and curing agent were mixed in 10:1 ratio and poured into the mold in a 100-mm diameter Petri dish, degassed in a desecrator, and cured at 75 °C for 60 min in an oven. The PDMS pieces were cut and gently peeled off from the mold on the Petri dish. The constructed device was submerged in Murashige and Skoog media overnight to ensure the hardening of the device. 0.17 mm coverslips and the PDMS pieces were plasma treated and bonded to get the final devices. Coverslips were used instead of the glass slides to facilitate fluorescent imaging. This setup was fixed with an adhesive to the Petri plate cover. Each channel was filled with MS media.
4.2. Preparation of Seeds and Measurement of Growth
Brachypodium wild-type seed line Bd21-3 was used in this study. The seeds were dehusked then soaked in water for 10 min. They were sterilised for 1 min with 70% ethanol in a sterile Petri dish. The ethanol was drained, and the seeds were rinsed with sterile deionized water. 20 mL of 1.3% NaOCl solution was poured into the Petri dish and rotated for 5 min. The seeds were then rinsed thrice with sterile deionized water. Ten seeds were placed in between two layers of sterile filter papers soaked in sterile water. It was observed that incubation at 4 °C for 7 days synchronised germination and promoted rapid growth as compared to 2d or 4d vernalization.
Media prepared was Murashige and Skoog 4.43 g, MES Monohydrate 0.5 g, Sucrose 30 g, and BAP 2.5 mg/L. After germination, the seeds were transferred to agar media and allowed to grow for 48 h at 22 °C with a 16 h photoperiod and high relative humidity at 57%. Finally, the seedlings were transferred to the device.
Epson perfection v700 photo scanner was used to visualise the full length of the seedlings grown in the microfluidic device and standard agar environment. WinRHIZO software (Regent Instruments, Quebec, Canada) was used to analyse the shoot and the root scan images.
For each channel, a seedling was grown in conventional Petri plates with Murashige and Skoog solid media in closed and sterile conditions. The channel was designed to restrain the root growth to a horizontally narrow path (1 mm diameter) which was optically transparent (0.17 mm glass coverslips). 4-days of vernalization and two days are post-germination synchronously growing Brachypodium seedlings were inserted into the wells vertically at around 75–55° angle, with the scutellum facing slightly upwards and radicula facing downward to allow growth imaging of the roots in the narrow horizontal channels. The anterior end was immersed in the well, and the posterior end was entirely out of the well, with the emerging leaf facing outwards. This provided gas exchange and illumination for the leaves. The potential of a plant-on-a-chip setup for Brachypodium seeds was shown in Figure S2
4.3. Osmotic Stress Application
Osmotic stress was applied with 20% PEG 6000 that was poured into the MS agar media. 270 uL of the prepared stress media was placed in the 3-punch microchannel. The seedlings were initially grown in the microfluidic device for three weeks when the plant reached a three-leaf stage; then they were transferred into the 3-punch microchannel device filled with the stress media to ensure the roots were fully immersed in the PEG-MS.
For 6 h and 24 h osmotic stress analyses, the seedlings were first stained with neutral red for 20 min and then transferred to the preliminary microchannel device containing 20% PEG-MS and visualized under a fluorescence microscope.
4.4. Imaging Setups
The seedlings were selected 2 days after the germination for microscopy studies. For standard visualisation of the control samples and the samples under the osmotic stress, the device setup for top imaging was used. MS media with PEG-supplement was used for samples under the osmotic stress. The top imaging was performed using Nikon and Olympus stereo microscopes from Japan.
For bottom imaging of the samples with fluorescent, a stock solution of 4 µM neutral stain was prepared with 0.2X MS medium supplemented with 20 mM potassium phosphate buffer at 8.0 pH, according to the procedure reported earlier [43
]. The samples with and without abiotic stress were stained for 15–20 min following the removal of PEG-supplemented MS media. The staining procedure made the cells slightly more visible and enabled fluorescent visualisation of the seedlings. The cross-section samples were prepared according to the protocol described online by Schiefelbein Lab [54
]. Fluorescence imaging was performed with an Axio Vert.A1 inverted microscope by Carl Zeiss (Germany), using the microfluidic device with the bottom imaging setup.
The confocal microscopy was performed with Carl Zeiss LSM 710, Germany and images recorded with Zen software (Carl Zeiss Microscopy GmbH, Jena, Germany). A single channel was used for visualisation with neutral red. The images were taken in 20X objective lens. Three-week seedlings were selected which were already pre-stained with neutral red at the 2-day seedling stage (2 DAG-days after germination) (stained as mentioned previously in the article). These seedlings were given drought stress for 6 h in Murashige and Skoog media with 20% polyethylene glycol 6000. These were then embedded in agarose (as described for the fluorescent microscope staining) to enable section slicing as thin as possible. Cut sections ~0.5–0.9 mm were achieved. The maturation zone of the plant was selected. Transverse sections were removed from the agarose molds and placed separately on acetone-ethanol cleansed coverslips and glass slides. The coverslips were sealed securely with clear nail polish.
4.5. RNA Isolation, DNase Treatment and qRT-PCR
RNA isolation was done from the whole plantlets with a Zymo research kit MiniPrep RNA isolation kit. Briefly, the tissue was homogenized in trizol reagent and transferred to column tubes. Flow through was DNase treated in the column, followed by prep buffer and wash buffers with intermittent centrifugations. The final eluted RNA was 20 uL. The concentration of RNA was verified by nanodrop spectrometer. A bleaching gel was used to analyse the integrity of the RNA. A 1% bleach gel was prepared in 1X TBE. 10X RNA loading buffer was used, and a low molecular weight RNA ladder was used.
From the plant genome database (http://www.plantgdb.org/prj/GenomeBrowser/
) the Brachypodium genome was used to search for upregulated genes in drought. drought responsive family protein 19 (now renamed Dehydration Induced19) DI19, Late Embryogenesis abundant protein Lea5, sequences were downloaded, and primers designed. Primers for DREB2A were taken from Feng et al. 2015. Two downregulated genes BdNAC054 and BdNAC092 were selected, and primer sequences were taken from You et al. [26
]. Ubiquitin BdUBC18 was taken as internal Control, and the primers used for it were also from You et al. [26
]. Primer sequences were also presented in Table S1
After confirmation of the integrity of the RNA samples, cDNA synthesis was performed with RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Waltham MA, USA) following the manufacturer’s instructions. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed by Light Cycler 480 II Instrument (Roche Diagnostics GmbH, Mannheim, Germany) using Perfecta SYBR Green SuperMix from Quanta Biosciences, Gaithersburg, MD, USA. Amplification was performed in a total reaction volume of 10 µL containing 5 µL of 1X SYBR Green Super Mix, 300 nM of each primer, and 100 ng of the template cDNA. The PCR thermal cycling parameters were set at 95 °C for 10 min to activate the SYBR green followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. For each sample, three technical replicates were made. Plant samples without stress treatment (0 h) was used for normalization and fold change calculation. The ΔΔCt Pfaffel method was used to analyse the relative gene expression of the qPCR results.