Genome wide identification and biochemical characterization of Calcineurin B-like calcium sensor proteins in Chlamydomonas reinhardtii

Calcium (Ca 2+ ) signalling is involved in the regulation of diverse biological functions through association with several proteins that enable them to respond to abiotic and biotic stresses. Though Ca 2+ -dependent signalling has been implicated in the regulation of several physiological processes in Chlamydomonas reinhardtii, Ca 2+ sensor proteins are not characterized completely . Chlamydomonas reinhardtii has diverged from land plants lineage, but shares many common genes with animals, particularly those encoding proteins of the eukaryotic flagellum (or cilium) along with the basal body. Calcineurin, a Ca 2+ /calmodulin-dependent protein phosphatase, is an important effector of Ca 2+ signalling in animals, while calcineurin B-like proteins (CBLs) play an important role in Ca 2+ sensing and signalling in plants. The present study led to the identification of 13 novel CBL-like Ca 2+ sensors in Chlamydomonas reinhardtii genome . One of the archetypical genes of the newly identified candidate, CrCBL-like1 was characterized. The ability of CrCBL-like1 protein to sense as well as bind Ca 2+ were validated using two-step Ca 2+ -binding kinetics. The CrCBL-like1 protein localized around plasma membrane, basal bodies and in flagella, and interacted with voltage gated Ca 2+ channel protein (VGCC) present abundantly in the flagella, indicating its involvement in the regulation of the Ca 2+ concentration for flagellar movement. The CrCBL-like1 transcript and protein expression was also found to respond to abiotic stresses, suggesting its involvement in diverse physiological processes. Thus, the present study identifies novel Ca 2+ sensors and sheds light on key players involved in Ca 2+ signalling in Chlamydomonas reinhardtii, that could further be extrapolated to understand the evolution of Ca 2+ mediated signalling in other eukaryotes. in a CD were recorded using average of three scans and data was collected and IGOR software version 3.14. To identify calcineurin B-like Ca 2+ sensor proteins in C. reinhardtii, a multiple sequence alignment of Arabidopsis CBLs (AtCBLs) was performed and used as an input to generate HMM profile. This profile was used to search for CBL orthologs in Chlamydomonas proteome. A total of 35 proteins


INTRODUCTION
Ca 2+ binding property, expression pattern and interacting partners. This is the first report that demonstrates the presence of CBL-like Ca 2+ sensors in C. reinhardtii and also provides functional relevance of one of the archetypical genes. In future, detailed functional characterization of the CrCBLlike proteins will pave the path for understanding their significance in different physiological as well as developmental processes of the green algae.

In silico analysis for identification of putative homologs of CBL in Chlamydomonas
Multiple sequence alignment of Arabidopsis CBLs was performed and used as an input to generate HMM profile using HMMER software (36). The generated profile was used with default parameters in the HMM search program of the HMMER package against Chlamydomonas reinhardtii proteome (Uniprot id: UP000006906). All the hits with significant positive scores were selected and then examined individually for the accessory domains. These hits were individually confirmed by BLASTP.
The sequence analysis and motif detection were carried out using Geneious Version 7 (https://www.geneious.com) software with default parameters. The structural modeling of the CrCBL-like1 protein was performed using web based SWISS-MODEL program (37) and the structure was visualized using PyMOL X11Hybrid (Ver 2) (https://pymol.org).

Chlamydomonas reinhardtii cell culture
Chlamydomonas reinhardtii wild type strain CC-124 was procured from Chlamydomonas resource center (www.chlamycollection.org). Culture was grown using TAP (tris-acetate-phosphate media, pH 7.4 and supplemented with Hutner trace elements) at 25˚C in a shaker incubator (120 rpm) in a synchronized manner (14 hrs light and 10 hrs dark condition). Light exposure was provided using white fluorescent light (30-40 µmoles/m 2 /s).

Cloning, Expression and Purification of the CrCBL-like1 protein
The ORF of Cre08.363750 (referred to as CrCBL-like1 in this study) was cloned into pET21a vector and the sequenced construct was transformed into E. coli BL21λDE3 cells. The transformed cells were grown at 37˚C until OD of 0.5-0.6 after which the temperature was reduced to 28˚C and 0.1 mM IPTG (isopropyl thiogalactosidase) was used for induction. The cells were pelleted after 6 to 7 hrs by centrifuging at 5000g for 10 minutes and stored at -80˚C. The stored cells were thawed and Triton X-100 for 30 minutes at 4˚C and centrifuged at 20,000g for 30 minutes. The supernatant was incubated with 1 ml of Ni-NTA beads (Qiagen) for 90 minutes. Further purification was performed according to Qiagen protocol with the following buffers-Wash buffer 1: Resuspension buffer with additional 150 mM NaCl; Wash buffer 2: Re-suspension buffer + 20mM Imidazole; Elution buffer: Resuspension buffer + 350 mM Imidazole). 1-2 µg of protein was visualized on 12% SDS PAGE and 1 µg of protein was used for western blot analysis.

Antibody against CrCBL-like1 protein
Affinity purified CrCBL-like1 protein was used for raising antibody in rabbit using commercial facility provided by Genei Bangalore, India. The antibody was used for CrCBL-like1 detection.

Native PAGE, SDS PAGE and Western Blotting
For protein expression analysis, the frozen cells were re-suspended in 1X PBS supplemented with 1mM PMSF and sonicated at 40% amplitude for 2 min. with 30 sec. on/off cycle. The prepared cell lysate was centrifuged at 10,000 rpm for 5 min. The clear supernatant containing the total cellular protein was dispensed and its concentration was estimated via nanodrop as well as Bradford assay.
30µg protein was mixed with 6X Laemmli buffer and then resolved on 12% SDS PAGE gel. Protein expression profile was visualized via western blot using the antibodies as mentioned previously.
Anti-His antibodies and anti-mouse alkaline phosphatase (AP) conjugated secondary antibodies were used for detecting the recombinant proteins at a dilution of 1:1000 (Sigma H1029) and 1:10,000, respectively. For detecting endogenous CrCBL-like1 protein, rabbit anti-CrCBL-like1 antibody was used at a dilution of 1:3000 and anti-rabbit horseradish peroxidase (HRP) conjugated secondary antibody at a dilution of 1:5000, was used (Sigma, USA). Tubulin was visualized using mouse antitubulin antibody (Sigma) at a dilution of 1:5000 and anti-mouse HRP conjugated secondary antibody at a dilution of 1:5000. Western blot was performed according to the standard protocols. The blots were quantified using ImageJ tool (https://imagej.nih.gov/ij/). Excitation wavelength -280 nm, emission range -285-400 nm, excitation bandwidth -5 nm and emission bandwidth -10 nm were used. For Ca 2+ binding saturation curve, different concentrations of Ca 2+ starting from 0.01 µM-3 mM were used. The saturation curve was plotted using fluorescence intensity measured at the wavelength 304.9 nm at Y-axis and free Ca 2+ concentration in the X-axis. The data was fitted using non-linear regression in Graphpad Prism 5 and apparent Kd was calculated manually as midpoint of the sigmoidal curve assuming that 50% of CrCBL-like1 is bound to Ca 2+ . The experiment was performed thrice for statistical significance.

Gel Exclusion Chromatography
All the eluted fractions of proteins were pooled and concentrated to 20 mg/ml using 3 kDa Millipore filters. The protein was passed through Superdex 16/600 GL 75 Pg column at a 1-ml/min flow rate and the fractions containing the proteins were collected. Ca 2+ binding was also observed via gel exclusion chromatography. 1.2 mg of CrCBL-like1 was incubated with 2 mM CaCl 2 and 5 mM EGTA for 30 minutes and run on Superdex 200 (10/300 GL) column. Gel filtration buffer: 50 mM Tris/HCl pH8 and 150 mM NaCl. During gel filtration (GF) respective ligands (Ca 2+ /EGTA) were also added in the GF buffers. Chlamydomonas reinhardtii was grown in 50 ml medium until mid log phase i.e., OD of 0.5. Different stress conditions including cold (4°C), oxidative stresses i.e., methyl viologen (1mM) and hydrogen peroxide (1µM), heat (37°C) and salt (200mM), were subjected by incubating the mid log phase culture from 30 min. to 4hrs. Samples were harvested at different time points and cells were freezed in liquid nitrogen and stored at -80˚C. RNA was extracted via hot phenol method. The frozen cells were resuspended into 300 µl of TES buffer (heated at 65 ˚C) and then aliquoted into 1.5 ml micro-centrifuge tube. Magna lyser beads (about 100 µl) were added to lyse the cells using hand held tissue grinder.

RNA extraction, cDNA preparation and Real Time Quantitative PCR
Subsequently, 300 µl of hot phenol (1:1 ratio of acidic phenol pH 4.5 and chloroform) was added to the tube, mixed well by vortexing and incubated at 65˚C for 30 minutes with constant shaking on dry bath heater. After 30 minutes, the tube was centrifuged at room temperature for 15 minutes at 12,000 rpm.
The aqueous phase obtained after centrifugation was transferred to a new tube containing 240 µl of chilled isopropanol and 30 µl of sodium acetate (pH 5.2). The tubes were incubated at -80 ˚C for [15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30] minutes. Subsequently, the tubes were centrifuged at 15,000 rpm for 30 minutes, supernatant was discarded and the pellet was washed with 75% ethanol for 5 minutes. The pellet was air dried for approximately 30-40 minutes until residual ethanol evaporated and then, resuspended in 50 µl of nuclease free water. The integrity of the RNA was visualized via agarose gel electrophoresis. 1µg of RNA was used for cDNA preparation (High-capacity cDNA Reverse Transcription kit, Thermo Fischer) and subsequently qPCR (USB HotStart-IT SYBR Green qPCR Master Mix, Affymetrix) was performed according to manufacturer's protocol. qPCR of CrCBL-like1 under salt and heat stress was performed as described above and data was evaluated to calculate the fold change in gene expression.
The house keeping gene, Guanine nucleotide-binding protein subunit beta-like protein (Crgblp), was used as an internal control (38).

RNA-seq data analysis
The previously generated RNA-seq data were downloaded from chlamynet (http://viridiplantae.ibvf.csic.es/ChlamyNet/ChlamyNet.html) and used to analyze expression patterns of CrCBL-like orthologs from Chlamydomonas. The expression values of different genes were plotted as heat maps using complex heat map package of Bioconductor using R software. Purified recombinant CrCBL-like1 was dialyzed in 10 mM sodium phosphate buffer of pH 7, which was compatible for CD spectropolarimeter. 2 mM CaCl 2 /EGTA was added to buffer solution containing 0.5 mg/mL of CrCBL-like1 and spectra were collected on a J-815 (Jasco, Japan) CD spectropolarimeter continuously purged by N 2 and equipped with a temperature control system. Blank was set with 10 mM sodium phosphate buffer, pH 7. Spectral measurements were performed in far-UV (188-260 nm) range using quartz cell of path length 1 nm in a thermostatic cell holder. CD spectra were recorded using average of three scans and data was collected and plotted using IGOR software version 3.14.

Immunofluorescence
Cells in early log phase were harvested and re-suspended in 1X PBS. 200µL aliquot of cell suspension was seeded on acid washed cover slips that were coated with poly-L-lysine. Cells were fixed with 3.7% paraformaldehyde in 1X PBS and permeabilized by submerging in cold 100% ethanol at -20°C for 10 min. Cells were then washed with 1X PBS containing 0.25 M NaCl at room temperature for 5 min.
Samples were then incubated with FITC conjugated anti-rabbit IgG antibody (Invitrogen), at 1:1000 dilutions. After three rounds of brief washings, cover slips were mounted on slides by applying antifade reagent (Slow Fade Gold, Molecular Probes). Slides were visualized by Leica TCS SP5 confocal microscope.

Statistical analysis
If not otherwise indicated, data were obtained from at least three independent experiments (n = 3).
Mean values were calculated and used for the analysis of standard deviation (SD) or standard error (SEM). Data analysis was performed with Prism 6.0 software (GraphPad, La Jolla, California).

Identification of putative CBL orthologs in Chlamydomonas reinhardtti
To identify calcineurin B-like Ca 2+ sensor proteins in C. reinhardtii, a multiple sequence alignment of Arabidopsis CBLs (AtCBLs) was performed and used as an input to generate HMM profile. This profile was used to search for CBL orthologs in Chlamydomonas proteome. A total of 35 proteins proteins were identified as AtCBL orthologs and confirmed by BLASTP. Structural similarities and differences between these proteins were further analyzed. Multiple sequence alignments revealed sequence homology amongst these proteins especially at their EF-hand motifs, which is one of the major characteristics of the Ca 2+ sensors ( Figure 1A). Majority of AtCBLs studied so far, contained four EF-hand motifs and a similar trend was observed in C. reinhardtii. Figure 1A shows

Expression and purification analysis of CrCBL-like1 protein
CrCBL-like1 was expressed and purified (Materials and methods section) through nickel NTA chromatography and subsequently through gel exclusion chromatography. A single highly pure fraction was obtained at the size of approximately 17 kDa. The identity of the protein was confirmed through western blotting using anti-Histidine antibody (Figure 2A). The oligomerisation of CrCBL-like1 was explored using gel exclusion chromatography. Protein eluted mostly as a dimer; however higher order oligomers were also visible but at lower concentrations ( Figure 2B). Ca 2+ binding property of CrCBL-like1 and the associated conformational change was explored through multiple methods. We initially, confirmed the Ca 2+ binding through native-PAGE and SDS-PAGE. Figure 2C demonstrates that CrCBL-like1 shows a shift in its position upon binding to Ca 2+ and EGTA in native-PAGE confirming the conformational transitions and hence, indicating Ca 2+ binding. These conformational differences were also visible on SDS-PAGE ( Figure 2D). Glutathione-S-Transferase (GST) protein was used as a negative control, which didn't show any Ca 2+ and EGTA mediated mobility shift on both native-and SDS-PAGE (Suppl. Figure S1). Additionally, distinct conformation adopted by Ca 2+ bound and unbound states was confirmed through gel exclusion chromatography, which aligned well with native-PAGE data. Ca 2+ bound and unbound CrCBL-like1 proteins eluted at different elution volumes, with bound protein eluting earlier than the unbound protein suggesting that it was in an extended conformation when bound to Ca 2+ ( Figure 2G,H). Effect of Ca 2+ binding on the secondary structure of CrCBL1 was analyzed by far UV circular dichroism spectroscopy. CrCBL-like1

CrCBL-like1 protein undergoes conformational change upon binding to calcium
CD spectra showed intense negative ellipticity at 208 nm and 222 nm depicting its significant alpha helical content, which is characteristic of most Ca 2+ binding proteins. In the presence of Ca 2+ ions, a decrease in helical content was observed as the peaks at 208 and 222 nm shift to lower negative values as compared to the control ( Figure 2E). The conformational change in CrCBL-like1 protein due to Ca 2+ binding was also investigated using Tyr fluorescence assay. Figure 2F shows that the Tyr fluorescence is altered in the Ca 2+ bound and unbound complexes, strengthening the conclusion that CrCBL-like1 protein binds Ca 2+ leading to conformational changes. In order to estimate the Ca 2+ binding affinity of CrCBL-like1 protein, a Ca 2+ titration assay using fluorescence spectroscopy was performed. A two-step Ca 2+ binding saturation curve of CrCBL-like1 protein was observed (Figure 3

In vivo expression of CrCBL-like1 and light induced change in expression pattern
CBLs have been extensively studied in plant systems with involvement in several physiological and developmental processes (39). Most of the characterized CBLs are localized in plasma membrane and tonoplast where they regulate various proteins involved in sodium compartmentalization, potassium homeostasis and production of reactive oxygen species (39)(40)(41)(42)(43). Since CrCBL-like1 protein was identified using A. thaliana CBLs as template, therefore; it may be speculated that CrCBL-like1 protein might also localize in plasma membrane and perform similar functions. Immunostaining with CrCBL-like1 specific primary antibodies followed by FITC labeled secondary antibodies revealed the expression of this protein around plasma membrane as well as near basal bodies ( Figure 4).
The presence of any signal sequence in the protein might be responsible for trafficking of the protein near cell membrane. It has been reported that dual fatty acyl modification via N-myristoylation and Sacylation determines the plasma membrane associated targeting of CBLs and CIPKs in Arabidopsis (44). Therefore, the protein sequences were analysed for the presence of these motifs. Interestingly, a myristoylation site with very high confidence was detected, however, palmitolyation site was not reinhardtii cells ( Figure 4B).

CrCBL-like1 interacts with Voltage Gated Calcium Channel (VGCC)
Ca 2+ signaling is essential for flagellar movement and therefore, several voltage-gated Ca 2+ -channels (VGCCs) have been reported to be present along the flagellar membrane of C. reinhardtii (46) CrCBL-like1 protein with VGCC was also analysed. Indeed, the CrCBL-like1 protein co-localized with VGCC in the flagella during the mid-phase of light cycle ( Figure 4C) as well as interacted with each other as established by co-immunoprecipitation analysis ( Figure 4D).

CrCBL-like genes are involved in stress response
In order to understand the functional role of CrCBL-like genes in response to environmental stresses, the gene expression analysis of publically available RNA Seq data of C. reinhardtii under multiple stress conditions was analyzed. The heatmap revealed varying expression patterns for most of the genes, except Cre03.g178150, which showed consistently higher expression in all condition. As noted, CrCBL-like1 as well as Cre15.g641250 expressed similar patterns, showing higher expression in various nutrient deprived conditions, suggesting that these genes respond transcriptionally to extrinsic cues (Suppl. Figure S2). The changes in CrCBL-like1 expression in response to additional stress conditions, such as heat, cold, salt, H 2 O 2 and methyl viologen, (not included in the RNA seq data) were also analysed. The CrCBL-like1 transcript as well as protein expression were tested by performing real-time PCR and western blot analyses, respectively. In all stress conditions, a decrease in protein expression was observed when compared to untreated sample except in cold condition(s) ( Figure 5A, B, Suppl. Figure S3). The gene expression analysis of CrCBL-like1 under the heat and salt stress conditions was performed and the results corroborated with the protein expression profiles ( Figure 5C).

DISCUSSION
Ca 2+ is considered a second messenger in eukaryotes. The temporally and spatially defined "Ca 2+ signatures" are involved in signal transduction pathways regulating a number of biological processes.
Ca 2+ -mediated signalling is widely employed in different physiological functions such as photobehavioural responses (phototaxis and photo-stop response), motility, chemotaxis and mating in C. reinhardtii, (10,12,13,50). A small variation in the intracellular Ca 2+ concentration in the flagella of C. reinhardtii, leads to major changes in its flagellar beating patterns and hence, the photomovement (51). Based on these variations in intraflagellar Ca 2+ , C. reinhardtii confronts flagellar biogenesis or sensor proteins in C. reinhardtii using Arabidopsis CBLs as baits through genome-wide analysis. Insilico analysis revealed the presence of 13 orthologs of CBLs in C. reinhardtii ( Figure 1A). Most of the characterized CBLs in other plant species contain typical Ca 2+ -binding motif known as EF-hands. They are present mostly in two pairs. The identified proteins contained 4 EF-hands like motifs as observed by multiple sequence alignment. Structural modeling of the potential EF-hand like motifs revealed similar pentagonal-bipyramidal geometry with conserved Ca 2+ binding residues ( Figure 1B). In plants, CBLs have been reported to bind Ca 2+ and regulate several physiological and developmental processes by interacting with its cognate kinases, known as CBL-interacting protein kinases (CIPKs) (28). CIPKs are specific kinases containing the NAF/FISL domains, along with kinase and phosphatase interaction motif. We did not find any CIPK orthologs in C. reinhardtii, which is in line with other reports (29,30). Given that, we identified these putative Ca 2+ sensor proteins using CBLs as input, it's intriguing to find the functioning of these identified proteins in the absence of CIPKs. However, CBL family members are reported to interact with other proteins as well, in addition to CIPKs. CBL3 interacts with AtMTAN, in addition to many CIPK family members through different regions, in Arabidopsis (52,53). CBL10 also interacts directly with AKT1 that leads to inhibition of inward movement of K + (54). CBL10 was also shown to be involved in reproductive development without the involvement of any of the CIPKs in Arabidopsis (55). These reports indicate that the interactions of the CBL or CBL-like proteins are not limited to CIPK members, suggesting that the identified proteins might have CIPK independent novel physiological functions in C. reinhardtii.
The CrCBL-like1 protein was further characterized based on structural modeling. Its EF-hand motif showed similarity to that of calcineurin-B variant from Coccidioides immitis (PDB Id: 5b8i) as well as CreinCBL8 in C. reinhardtii (56). CrCBL-like1 structural model showed the typical EF-hand fold with Ca 2+ coordinated in a pentagonal-bipyramidal geometry as described in literature. The binding of Ca 2+ to the EF-hand motifs is a functional feature of CBL proteins for relaying Ca 2+ signals as shown in Figure 2. The binding of Ca 2+ to CBLs using different methods like gel electrophoresis and mobility shift assays is well documented. The effect of Ca 2+ binding on the mobility of the protein was assessed using native-and SDS-PAGE. Under both denaturing and native conditions, the Ca 2+ bound and unbound protein showed differences in their mobility suggesting a change in the conformation of the protein. In the native condition, the Ca 2+ -saturated protein had a larger apparent molecular mass compared to the Ca 2+ depleted form, which suggests that Ca 2+ binding leads to a more extended protein conformation ( Figure 2C, D). This aligns well with the structural changes observed in most of the EF-hand containing proteins upon Ca 2+ binding. It is well documented that the EF-hand motifs adopt an open conformation when bound to Ca 2+ , with both the alpha helices lying perpendicular to each other. In contrast, it adopts a closed conformation, with both the helices lying parallel to each other in Ca 2+ free state.
The Ca 2+ -induced conformational changes in the protein observed using gel exclusion chromatography, were in agreement with the PAGE results. The data correlated well with the gel exclusion chromatography data wherein Ca 2+ bound protein eluted earlier compared to the unbound protein.
However, under the denaturing conditions, the Ca 2+ bound protein migrated faster compared to EGTA bound protein suggesting vice versa ( Figure 2E, F). Such faster movement has been observed in a few Ca 2+ -binding proteins when associated with Ca 2+ , in comparison to Ca 2+ depleted proteins (57)(58)(59)(60). Since EGTA has a higher molecular mass compared to Ca 2+ , therefore, the net mass of the protein when bound to EGTA is higher, hence its mobility is retarded. Ca 2+ binding often leads to change in the secondary structure of the protein.  This work opens a new avenue to study stress signalling in Chlamydomonas, considering that CrCBL-like1 could be used as a tool to study the role of Ca 2+ in stress signalling as well as to explore the diversity of lesser-known CIPK-independent functions of CBL-like proteins.