Materials and methods

2.1. Biological material and growth conditions

In this work, we used the Arabidopsis thaliana Col-0 (N1092) and Col-3 (N28171) backgrounds as references. We obtained the mutant allelescipk13-1 (SALK_124748C) and cipk13-2 (SALK_130671), as well as cbl7-1 (SAIL_201_A01) from the Nottingham Arabidopsis Stock Center (NASC). T-DNA insertion lines were genotyped as previously described (Alonso et al., 2003), using the primer pairs listed inSupplemental Data Sheet 1 . The T-DNA insertion mutantscbl7-2 and hak5 have previously been described (Ma et al., 2015; Ragel et al., 2015). Plants were grown on Petri dishes containing solidified ½ MS medium supplemented with 1% sucrose (w/v) (Murashige and Skoog, 1962). Plant growth proceeded in a growth chamber under strictly controlled environmental conditions (16 h light, 8 h darkness, constant temperature of 22 ºC, 100 to 105 µmol photons m-2 s-1photosynthetically active radiation). In addition, we usedSerendipita indica strain DSM 11827, which was obtained from the German Collection of Microorganisms and Cell Cultures (DSZM) in Braunschweig, Germany. The fungus was grown at 28 ºC in darkness on solidified arginine phosphate (AP) medium (Rodríguez-Navarro and Ramos, 1984) and refreshed weekly.

2.2. Root growth promotion assay

Surface-sterilized Arabidopsis seeds were plated on vertical ½ MS plates. After stratification (2 days at 4 ºC), the plates were transferred to the growth chamber and the seedlings were grown vertically for one week. Thereafter, four to five seedlings were transferred to Petri dishes containing solidified Plant Nutrition Medium (PNM) supplemented with 50 mM NaCl (Johnson et al., 2013). Each seedling was then associated with a 5 mm Ø medium plug extracted from either sterile AP plates (control) or from AP plates harboring a one-week-oldS. indica mycelium (co-cultivation). The PMN plates with the control seedlings and the seedlings co-cultivated with the fungus were further kept in a growth chamber maintained at 23.5 ºC, 16/8 h photoperiod, 100 µmol photons m-2 s-1 light intensity for another fourteen days. After that time, the plants were photographed for the further analysis of the root system and the plant material was either used for RNA extraction or the determination of the fresh weight (fw).

2.3. Quantitative analysis of root system architecture traits

The stimulation of root growth is a well-described trait in the interaction of S. indica with its host plant. With the aim to quantify the effect of S. indica in the different genotypes and treatments, respectively, photographs of the plates were captured with a digital camera at a fixed distance of 29 cm. Using Adobe Photoshop CC, the images were cropped to a height of 14 cm maintaining only the part containing the root system and converted the pictures to black and white images. The root network traits of the plants in the prepared images were then analyzed using the GiA Roots software (Galkovskyi et al., 2012). Further processing of the images included their segmentation employing global thresholding (Binary_inverted) and Gaussian adaptive thresholding. For the comparative analysis of alterations of the root system architecture, the total network area and total network length was used as readout. Taking the biological variability of the system into account, 24 individual plants per genotype and growth condition were analyzed, respectively.

2.4. Total RNA extraction, library construction, and RNA-seq analysis

To study transcriptional alterations provoked by either the co-cultivation of Arabidopsis roots with S. indica or the functional knockout of CBL7 , total RNA from 100 mg plant roots was extracted as previously described (Oñate-Sánchez and Vicente-Carbajosa, 2008). The quality and concentration of the extracted RNA was tested by absorbance analysis using a Nanodrop® ND-1000 spectrophotometer (ThermoFisher). After an additional confirmation of the RNA sample integrity on a Bioanalyzer 2100 (Agilent) by the Novogene Genomics Service (Cambridge, UK), the service laboratory proceeded with the library construction and RNA sequencing (PE150) on Illumina NovaSeq™ 6000 platforms. The Novogene Genomics Service also provided basic data analysis applying their RNAseq pipeline. This included data filtering and sequence alignment using HISAT2 v2.0.5 with default parameters (Kim et al., 2019), transcript quantification with HTSeq v0.6.1 with -m union parameter (Anders et al., 2014), and differential gene expression analysis employing the DESeq2 v1.22.2 algorithm with a cut-off value of an adjusted p -value of < 0.05 (Love et al., 2014). For each genotype and treatment, respectively, three biological replicates were processed. The resulting p -values were adjusted for multiple testing using the Benjamini–Hochberg correction (Benjamini and Hochberg, 1995). Along with the adjusted p -value (FDR) of < 0.05 an absolute differential expression of log2 fold change (FC) ≥ 1.25 was chosen to select differentially expressed genes (DEGs). The functional classification of DEGs was performed using the MapMan v3.6 software (Thimm et al., 2004), paying special attention to DEGs related with plant defense and nutrient assimilation. Furthermore, functional relationships between the DEGs were investigated using the applications stringApp v1.7 (Doncheva et al., 2019), MCODE v2.0 (Bader and Hogue, 2003), EnrichmentMaps v3.3.3 (Merico et al., 2010), and ClueGO v2.5.8 (Bindea et al., 2009) in Cytoscape v3.9.0 (Shannon et al., 2003).

2.5. qPCR analysis

Real-time quantitative RT-PCR was conducted as previously described (Pérez-Alonso et al., 2021). In brief, total RNA from three different biological samples was converted into cDNA using M-MLV reverse transcriptase and oligo(dT)15 primer. Two nanograms of cDNA was then used as template for the qPCR reactions, which were conducted in triplicate (technical replicates). The oligonucleotide pairs used in the experiments are given in Supplementary Data Sheet 1 . The reactions were monitored on a Lightcycler 480 Real-time PCR system (Roche Diagnostics). Differential gene expression in Arabidopsis was analyzed by using the comparative 2-∆∆CT method (Livak and Schmittgen, 2001) withADENINE PHOSPHORIBOSYL TRANSFERASE 1 (APT1 , At1g27450) as reference gene (Jost et al., 2007). Root colonization withS. indica was monitored with a primer pair for fungal translation elongation factor EF-1α (SiTEF1 ) (Bütehorn et al., 2000). The fungal SiTEF1 cDNA levels were expressed relative to the plantGLYCERINALDEHYDE-3-PHOSPHATE DEHYDROGENASE C2 (GAPC2 , At1g13440) cDNA levels. To exclude that the amplified DNA fragments stem from DNA of dead fungal tissues within the roots, all data presented here derived from cDNA libraries generated from RNA of colonized roots.

2.6. Trypan blue staining of fungal hyphae and spores

To visually inspect root colonization, 10-12 small root samples from control and co-colonized plants were employed. First, the root samples were thoroughly washed with deionized water. Next, the root samples were cut in 1 cm long pieces and incubated overnight in 10 N KOH. The root samples were then rinsed 5 times with sterile H2O, before they were incubated for 5 min in 0.1 N HCl. Finally, the samples were incubated in a 0.05% trypan blue solution (w/v), before they were partially decolorized with lactophenol over ten minutes. Before the specimen were mounted on glass slides and examined by microscopy, they were washed once with 100% ethanol and thrice with sterile H2O and stored in 60% glycerol (v/v).

2.7. Yeast two-hybrid analysis

In order to examine the physical interaction between CBL7 and CIPK13, total RNA from four weeks-old Arabidopsis seedlings (Col-0) was extracted (Oñate-Sánchez and Vicente-Carbajosa, 2008) and first-strand synthesis was performed according to the supplier’s instructions, using M-MLV reverse transcriptase and oligo(dT)15 primer (Promega). For the cloning of CBL7 and CIPK13 , the corresponding cDNA fragments were amplified by PCR using specific primer pairs listed in Supplementary Data Sheet 1 . The resulting PCR fragments were inserted into the vector pGEM® -T Easy (Promega). Sequence integrity of the obtained products was confirmed by commercial sequencing (StabVida). Subsequently, the fragments were introduced into the vector pENTR-3C (Thermo Fisher) using the Eco RI andSal I restriction sites included in the primer sequences. The obtained pENTR-CBL7 and pENTR-CIPK13 vectors were used in Gateway LR-recombination reactions with the destination vectors pDEST-22 and pDEST-32 according to the manufacturer’s instructions. The resulting destination vectors were then used to transformSaccharomyces cerevisiae strain HF7c. Transformants were plated on SD (simple drop-out) medium /-Leu/-Trp/+Ade and incubated at 28ºC for 2 days. Protein interaction was tested by growing transformants on SD medium /-Leu/-Trp/-His/+Ade. The plates were incubated at 28ºC for up to one week. To investigate the interaction of CBL7 with two additional putative interaction partners, CIPK9 and CIPK24, the yeast two-hybrid vectors PDEST-AD092F08 and PDEST-AD107D03 for CIPK9 and CIPK24, respectively, were obtained from the Arabidopsis Biological Resource Center (ABRC).

2.8. K+ ion content quantification

The analysis of endogenous cation contents of infected and control plants was performed in fractions of root and shoot samples. To avoid carry overs from the medium, roots were thoroughly rinsed with 10 mM MES-Ca2+ pH 6.5. Next, root and shoot samples were dried, weight and extracted with 1 M HNO3. The K+ contents of the supernatants were then determined by atomic emission spectroscopy. The results are given as the means and their standard errors of three independent experiments.

2.9. Statistical analysis

For statistical data assessment and the generation of box plots, JASP v0.16 was employed. The box plots show the median, quartiles, and extremes of the compared experimental values. One-way anova followed by Tukey’s post-hoc test or Student’s t -test were performed to statistically analyze the data. Sample sizes (n) for each experiment are given in the respective figure legends. Hierarchical clustering and heatmaps of selected gene expression levels across the different experiments was conducted using the Instant Clue software v0.10.10 (Nolte et al., 2018).