Arabidopsis cell culture for comparable physiological and genetic studies

ABSTRACT

Cell cultures established from various plant species have been used for a range of physiological and biochemical studies. Homogeneity of cell types and size of clusters in the cell culture often gave a clearer and simpler results compared to those obtained with the whole plant. On the other hand, possible variability of physiological conditions and responsiveness to external stimuli between the cell lines could be problematic for comparative studies. Aiming at combining the usefulness of plant cell culture with the rich information and genetic resources of Arabidopsis, we systemically examined the methods/conditions to establish cell lines for comparative studies, which could be applicable to a variety of genetic resources. Arabidopsis cell lines thus established from the meristem of mature seeds showed reproducible and comparable MAMP responses such as ROS generation and defense-related gene expression. MAMP responses of the cultured cells showed the specificity depending on the presence/absence of the corresponding MAMP receptor. Pharmacological study with a protein kinase inhibitor, K252a, also showed the usefulness of the cell culture for such studies. These results indicated the usefulness of the method to establish Arabidopsis cell lines, which are useful for comparative studies between genetic resources.

Plant cell culture is a useful tool for plant research. Because of the relative homogeneity of cell types/size of clusters and penetrability for reagents, results obtained with the cultured cells are often clearer and simpler than those obtained with whole plant. Rapid growth of some cultured plant cells also benefits their use in the laboratory. Among the cultured plant cells so far established, tobacco BY-2 cell line is well known and has been extensively studied. BY-2 cells contributed to various studies in plant biology, e.g. those on cell cycle, cytoskeletons, and defense responses.^{1, 2} Transcriptome analysis of BY-2 cells was reported and the EST clones can be obtained from RIKEN BRC.³ Furthermore, gene transformation of the BY-2 cells can be easily performed by Agrobacterium-mediated transformation¹. Cultured plant cells also contributed to the study of morphogenesis. For example, zinnia cells have been a preferred model system for xylem formation, by which a xylem differentiation protein (xylogen) was identified.⁴ In rice, callus cells can be easily induced from the embryo cells of the seeds with various genetic backgrounds, enabling the application for the studies on gene function, cell biology, signaling, etc. Rice cell culture system, in combination with pharmacological studies, contributed to the study of elicitor responses and signaling mechanisms.⁵

Combining such advantages of cell culture system with the richness of genetic information and resources in Arabidopsis, where most studies have been performed with whole plant system, would benefit related research. However, established cell lines in Arabidopsis are still limited. While several cell lines derived from wild-type Arabidopsis have been used for various physiological studies,^6-9 it seems difficult to compare the results obtained with these cell lines directly because they were sometimes established by different methods for callus induction and also from different tissues/organs. Ikeuchi et al. previously observed wound-induced callus formation in Arabidopsis and discussed about the possible factors affecting the callus formation.¹⁰ However, the authors did not compare the physiological properties of the established cultured cells for further application. To establish Arabidopsis cell cultures useful for comparative physiological/biochemical studies from various mutants or transgenic plants, such cell lines should be established from the same tissue of these plants by using the same methodology. We here examined the method/conditions to establish Arabidopsis cell lines compatible for MAMP (Microbe-Associated Molecular Pattern) response study from genetically different seeds and evaluated the established cell lines for their comparability and applicability.

To establish Arabidopsis cell lines, we induced callus from a mature seed based on the rice callus inducing method.¹¹ Each seed was sterilized by chlorine gas for 18 h and absorbed water for several minutes. Seed coat was removed using tweezers under the microscope and embryo was put on solid gel prepared with 0.8% gellan gum in CIM medium (2% glucose, 0.5 g/L MES (pH 5.7), B5 basal salt mix, vitamin mix, 0.5 mg/L 2,4- dichlorophenoxyacetic acidand 0.05 mg/L Kinetin; Cold Spring Harbor Protocols, 2006; doi:10.1101/pdb.rec10814), which was one of the popular Arabidopsis culture medium for generation of the callus (Figure 1a). After the seed was incubated for 1 week at 22°C under constant dark condition, small white callus was generated in the meristem region (Figure 1b). The callus was transferred on a new solid medium plate and grown for several weeks. When the volume of callus grew up to the diameter of 0.5 cm, they were transferred into 30 ml of CIM in a 100 ml conical flask. Cultured cells were grown on a rotary shaker at 22°C under the same constant dark condition. A part of the cells were transferred to the fresh medium every 2 weeks for maintaining the culture. The cultured cells for experiments of the reactive oxygen species (ROS) generation and gene induction assays were used 1 week after transferred to the new fresh medium. As the soluble fragments of typical MAMPs, a chitin fragment, (GlcNAc)₇, and a flagellin peptide, flg22, were used for these studies.

Figure 1.

Establishment of comparative Arabidopsis cell lines. (a, b) Pictures of an Arabidopsis seed after removal of the seed coat and induction of callus. Arrowhead indicated the position of callus. Scale bar = 1 mm. (c) Total RNA was extracted from the cultured cells. The methods of cDNA synthesis and real-time PCR were previously described.¹² Actin forward primer (5ʹ-CTTGCACCAAGCAGCATGAA-3ʹ) and reverse primer (5ʹ-CCGATCCAGACACTGTACTTCCTT-3ʹ) were used for evaluation. (d-g) Forty mg of cultured cells were transferred to 500 µl of each medium in a 2 ml centrifuge tube and pre-incubated for 30 min at 22°C on the shaker. Three replicates were prepared for each treatment. After (GlcNAc)₇ treatment (GN7) or control treatment (water), 10 µl aliquots of medium were used for 96-well plate-based chemiluminescence assay. Detailed method of luminol assay was also previously described.¹² The amount of ROS was estimated by using a standard curve for hydrogen peroxide. The data were given as means ± SD of three replicates. (c, d) a and b were indicated independent cell lines, established from Col-0 seeds. (e) The numbers showed the size of steel mesh to filter the cultured cells.

To evaluate the comparability of the data obtained with different cell lines, basal level of gene expression and MAMP responsiveness of two independent wild-type cell lines derived from different seeds were compared. The mRNA expression level of ACTIN housekeeping gene of these cell lines was analyzed as an index of gene expression level in the basal state. The result indicated that these cell lines expressed the same level of ACTIN mRNA, suggesting the comparability of these independent cell lines at least for the expression of housekeeping genes (Figure 1c). These cell lines also showed the same level of (GlcNAc)₇-induced ROS generation, indicating the similar level of responsiveness to MAMP treatment (Figure 1d). These data indicated that the system described here is useful for establishing cell cultures applicable to comparative studies.

Effect of the experimental conditions on the responsiveness of the established cell line was also evaluated. Since the penetration efficiency of the applied MAMPs through the cell cluster could be an important factor to induce defense responses, the effect of the size of cell cluster on the (GlcNAc)₇-induced ROS generation was evaluated. To obtain the cell clusters of different sizes, cultured cells were filtered through the 0.5 mm or 1 mm steel mesh and then transferred to the fresh medium. The results obtained with these cell clusters showed that the cultured cells filtered through 0.5 mm mesh showed a higher ROS response than those filtered through 1 mm mesh (Figure 1e). These data suggested that the smaller cell cluster facilitates more efficient MAMP penetration. To survey the suitable medium condition for the generation of (GlcNAc)₇-induced ROS in suspension-cultured Col-0 cells, we tested three kinds of mediums (CIM, 1% sucrose in half-strength MS medium and 1% sucrose in half-strength MGRL medium).^13,14 The results showed that the cells cultured with half-strength MGRL showed the highest response compared to the other two mediums (figure 1f). These results also well corresponded with those of Arabidopsis seedling assay, where the incubation with half-strength MGRL resulted in a better response than MS medium.¹⁵ Dose dependency of (GlcNAc)₇ for ROS generation of wild-type Col-0 cells showed that the ROS generation was saturated at 1 µM (GlcNAc)₇ (Figure 1g), which is much lower compared to those required for seedling and leaf disk assays.^12,15

Two different cell lines, wild-type (Col-0) cells and chitin elicitor receptor-like kinase 1 (CERK1) knockout cells (cerk1-2), were compared for ROS generation induced by two typical MAMPs, (GlcNAc)₇ and flg22. Although both cultured cells showed flg22-induced ROS generation, (GlcNAc)₇-induced ROS generation was only observed in the wild-type cells (Figure 2a), which coincided with the lack of chitin receptor in the cerk1 mutant. While the cerk1-2 cells complemented with CERK1 (CERK1/cerk1-2) recovered the (GlcNAc)₇-induced ROS accumulation, those complemented with the kinase-inactive CERK1_D441V (CERK1_D441V/cerk1-2) did not recover the (GlcNAc)₇-induced ROS generation (Figure 2b). These results were well corresponding to the previous results with Arabidopsis seedlings^16,17 and indicated the applicability of the established cell cultures for genetic studies.

Figure 2.

Applicability of the established cultured cells. (a, b) MAMP-induced ROS generation in the wild type (Col-0), CERK1-KO mutant (cerk1-2; GK_096F09)¹⁶ and CERK1 complemented plant¹⁷ were measured. (c) K252a (1 µM) was added in a stepwise manner at −10, 30 and 60 min after the addition of (GlcNAc)₇ to ensure the inhibition of protein kinases. (d) Total RNAs were extracted from Col-0 and cerk1-2 mutant cells treated with each MAMP for 120 min. Expression levels of FRK1 (Forward prime, 5ʹ-TGCACTTACCCTCCTTCG-3ʹ; Reverse primer, 5ʹ-GACAGTAGAAGCCGGTTGGT-3ʹ) and PAD3 (Forward prime, 5ʹ-TGCTCCCAAGACAGACAATG-3ʹ; Reverse primer, 5ʹ-GTTTTGGATCACGACCCATC-3ʹ) as determined by qPCR were shown. Actin was used as an internal control. Asterisks indicate statistically significant differences from the water treatment controls by Student’s t-test (P < 0.05). Each data was given as means ± SD of three replicates. GN7, 1 µM (GlcNAc)₇ treatment; flg22, 1 µM flagellin peptide treatment; water, control treatment.

Pharmacological study with a protein kinase inhibitor, K252a, showed that K252a almost completely inhibited the chitin-induced ROS generation of the Col-0 cultured cells (Figure 2c), as similar to the previous observation for cultured rice cells and Arabidopsis seedlings,^15,18 supporting the usefulness of the cultured cells for pharmacological analyses. MAMP-induced expression of defense-related genes, FRK1 and PAD3, in the wild-type Col-0 and cerk1-2 mutant cells were also compared. After the treatment with (GlcNAc)₇ or flg22 for 2 h, the Col-0 cells showed the up-regulation of these genes for both MAMP treatment but the cerk1-2 cells only responded to flg22 again (Figure 2d).

In summary, we propose a method to establish cultured cell lines suitable for physiological/biochemical studies, such as MAMP response study, from Arabidopsis seeds with different genetic backgrounds. Established cell lines showed reproducible and comparable MAMP responses depending on the genetic background. Advantages of cell culture system, discussed in the introduction and partly confirmed in the present study, would also contribute to expand the use of cell culture system, especially in combination with the richness of genetic resources in this model plant. We hope the development of efficient MAMP response assay with Arabidopsis seedlings,¹⁵ leaf disks¹⁹ and cell culture described here all contribute to the development of related fields in plant research.

Funding Statement

This work was supported in part by Grants-in-Aid for Scientific Research (No. 20K06058 to Y.D., 18H02208 to H.K.), Grants-in-Aid for Young Scientists (No. 17K15231 to Y.D.) and MEXT-Supported Program for the Strategic Research Foundation at Private Universities 2014-2018 (S1411023) to H.K. from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.