Functional characterization of a serine-threonine protein kinase from Bambusa balcooathat implicates in cellulose overproduction and superior quality fiber formation

Analysis of the BbKstgene architecture

The full length BbKst gene (7375 bp) was isolated using RACE and genome walking. A representative gel documenting RACE products and genome walked PCR products are shown in Additional file 1: Figure S1 Ai, Aii. The BbKst sequence (GenBank ID JQ432560) was subjected to in silico analysis, which showed a 3342 bp long coding region with three exons of 443 (E1), 268 (E2) and 2631 (E3) bp sizes, separated by two introns (1900 and 684 bp) as shown in Figure 1. The putative start (ATG) and stop (TGA) codons were positioned at 265 and 7033 bp, from the transcription start site, respectively, and a poly-adenylation motif AATAAA was found at the end of the 3′ UTR sequence. A schematic diagram of the BbKst sequence organization is shown in Figure 1A. A putative promoter sequence, 840 bp in length, was analysed by using PlantCare (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) and PlantPAN (http://plantpan.mbc.nctu.edu.tw/gene_group/index.php), and contains several specific structural elements including a putative TATA box (-34 to -29), CAAT box (–81 to –78), Kozak sequence (GCACCATGG) at the positions -5 to +3; with a cluster of important plant specific key regulatory cis-elements (Figure 1B; Additional file 2: Table S1). DNAMAN based analysis of 3951 bp long cDNA sequence revealed that it contains 267 and 342 bp long UTR regions at the 5′ and 3′ ends, respectively. Further analysis of BbKst gene using NCBI-CDS software has clearly shown that the predicted BbKst protein has distinct PKC domain and LRR regions. Exact orientation of the kinase domain and LRR repeats were recognized with a Pfam search (Figure 1C). Positions of six LRR repeat regions, ATP binding site, active sites and nucleotide phosphate-binding region were determined by PROSITE and results are depicted in Figure 2. Domain analysis also indicates that BbKst protein has no extracellular domain indicating that it belongs to RLCK subfamily.

Analysis of biochemical properties of BbKst-kinase

The kinase domain of the BbKst protein was expressed in E. coli BL21 cells as described in the methods section. Expression of BbKst-kinase was found in both soluble and insoluble fractions of total cell lysate (Figure 3A). Subsequently, the purified protein was confirmed by western blot (Figure 3B).

BbKst-kinase phosphorylates all the substrates tested in the present study viz., casein, myelin basic protein (MBP), histone (H1), ‘LRRKtide’ and ‘GS’ peptide (Figure 4Ai). The respective phosphorylated products were electrophoresed and the auto-radiographed image is shown in Figure 4Aii. Among these substrates, the protein showed higher specific activity with casein and MBP than others (Figure 4Ai). Therefore, we used casein as a substrate in the subsequent experiments to study the biochemical properties of BbKst-kinase. Casein phosphorylation was determined over a pH range from 3 to 11 and subsequent specific activity showed a bell shaped pattern within the above said pH range (Figure 4Bi). The BbKst kinase showed highest activity at pH 7.0 when casein was used as the in vitro substrate (Figure 4Bii). The effect of temperature on casein phosphorylation by the BbKst-kinase was also determined. The substrate phosphorylation activity of the BbKst-kinase showed a gradual increase in activity from 4 to 15°C, followed by an optimum stationary stage up to 40°C and then gradually decreased with increasing temperature (Figure 4Ci, Cii).

BbKst-kinase showed highest activity in presence of Mg²⁺. It was found that casein phosphorylation by BbKst-kinase increased along with increased concentration of Mg²⁺ from 0.5 or 5.0 mM, while, in the case of Mn²⁺ a weak phosphorylation of casein was noted only at 2.5 mM (Figure 4D) indicating that BbKst-kinase prefer Mg²⁺ over Mn²⁺ for its enzyme activity. Phosphoamino acid analysis of autophosphorylated BbKst revealed that BbKst phosphorylated only serine and threonine residues but not tyrosine (Figure 5).

Enzyme kinetics of BbKst-kinase

BbKst exhibited Michaelis-Menten kinetics with respect to casein as substrate. The Vmax and Km value of BbKst-kinase for casein determined from Eadie-Hofstee plots of V₀ versus V₀/[S] were 537.8 ± 65.08 pmol/min/μg and 173.4 ± 48.78 μM, respectively (Figure 6A). The Kcat (ratio of Vmax to Km) was determined and the value was found to be ~ 3.01 min-1. Casein-phosphorylation of BbKst-kinase showed a steady increase along with time upto 3 min, which remain steady upto 5 min. The dissociation constant (Kd) for substrate phosphorylation was computed as 2.3 ± 0.7 pmol (Figure 6B).

Effect of protein kinase inhibitors on specific activity of BbKst

None of the three inhibitors (quercetin, H-89 and staurosporin) tested could inhibit completely the activity of BbKst-kinase (Figure 7A). Of these three, quercetin showed significantly higher inhibition than H-89 and staurosporine. The IC50 value for quercetin was estimated at 7 nM. ANOVA results showed there is significant difference (P < 0.0001) in IC50 values of these three kinase inhibitors. The products were electrophoresed in SDS-PAGE and a corresponding autoradiogram of the gel is presented in Figure 7B.

Spatial expression of BbKst transcripts in B. Balcooa

Differential expression of the BbKst transcripts in the stem, internode, rhizome and leaf of B. balcooa was noted and confirmed through three independent RNA slot blot hybridization as described in the method section. Results obtained were found to be highly reproducible. The strongest expression of the BbKst transcripts was found in mature stem while weaker expression was observed in rhizome. Leaf and internodal tissue showed moderate expression of BbKst transcript (Figure 8Ai). All the above findings were further confirmed by qPCR analysis, wherein a weaker expression of the transcript in the rhizome was detected (Figure 8Aii).

Differential expression of BbKst transcript at different fiber developmental stages of B. balcooa

Expression analyses of the BbKst transcript at different internodal tissue (2^nd, 5^th, 8^th, 10^th, 12^th, and 15^th) revealed highest expression of BbKst in the 10^th internode followed by 8^th (Figure 8Bi). Expression level gradually decreased with fiber maturity and lowest expression level was noted in the 15^th internode. Results obtained were consistent in three independent experiments. The expression level of the BbKst was found to be 2.8 fold higher during fiber elongation (8^th internode) and 3.85 fold higher during fiber maturation (10^th internode) than in the fiber initiation stage. All the above findings were further confirmed by qPCR analysis (Figure 8Bii).

Analysis of transient expressions of the BbKstin plant cell

Transient expression of a BbKst-GFP fusion was analysed using onion peels bombarded with plant expression vector pCambia 1304 carrying total cDNA sequence of BbKst, designated as Js1304 (Figure 9). Onion epidermal cells exhibited expression of the GFP-fusion protein in the cytoplasmic region of the Js1304 transformed onion cell after 24 hr (Figure 10). Expression was not found in the non-transformed tissue (Figure 10). This observation indicates constitutive expression of GFP reporter in the cytoplasm.

Development of transgenic tobacco plants with BbKstgene

Tobacco leaf discs were transformed with pCambia 1304 carrying Js1304 and transformants were selected under hygromycin containing medium. Putative transgenic plants were then transferred to soil and T₀ plants took nearly 6 months to produce flowers and seeds. The comparative phenotypic features of empty-vector transformed and transgenic plants are shown in Figure 11A-F. It was noted that empty vector-transformed plants were gradually stooping from one month onwards after soil transfer (Figure 11C), but the transgenic plants were strong and upright (Figure 11D, F). The stem of vector-transformed plants were comparatively weak than the transgenic plants containing BbKst (compare Figure 11E with Figure 11F). Similar observations were also noted in case of T₁ transgenic tobacco plants (Additional file 3: Figure S2 A-F).

Selection of GUS positive plants

GUS expression was studied in thirty day old transgenic tobacco seedlings (T₁) in presence of X-gluc as stated in Method section. Prominent blue colouration was observed in transgenic seedlings (Additional file 4: Figure S3 A-C). Blue colour did not appear in empty vector-transformed seedling (Additional file 4: Figure S3 D).

Copy number determination of BbKst

In most of the T₁ events the transgene was found to be stably integrated into the genome as a single copy, confirming perfect transgene transmission in the next generation (Additional file 5: Figure S4).The GUS expressive and single copy BbKst integrated lines were selected for further study.

Heterologous expression of BbKstin transgenic tobacco

qPCR detected the expression level of BbKst in different T₁ progeny lines. Different transgenic lines exhibited 150-200 fold enhanced expression level than that of the empty vector control tobacco plants. Among five lines of transgenic BbKst plants, line 3 on an average showed maximum expression followed by line 4 and 2 (Figure 12).

Physical and anatomical characteristics of fibers of transgenic plants

Stem diameters of five transgenic plants (S1 to S5) expressing BbKst gene and empty vector-transformed tobacco plants (control) were measured and data are represented in Figure 13A. Stem thickness of independent transgenic plants varied from 1.0 cm to 1.04 cm with an average diameter of 1.0 cm. No significant increase in stem diameter of transgenic plants expressing BbKst gene over vector-transformed plants was noted. A comparative account between xylem width and stem width of five independent transgenic events at the T₁ generation is presented in Figure 13B.

Anatomical changes concomitant with transgene expression in different transgenic lines (T₁) revealed significant increase in the number of xylary fibers of vascular bundles compared to the empty-vector transformed tobacco stems (Table 1).

Heterologous expression of BbKstin transgenic tobacco plants

Stem cross sections of transgenic tobacco plants carrying BbKst were subjected to Congo-red staining as described in Methods and subsequently the stained sections were studied under CLSM. Enhanced level of cellulose depositions in the walls of xylary fibers in the stems of transgenic plants was evident. The vascular zone was more pronounced in transgenic than empty vector-transformed T₀ plants (Additional file 6: Figure S5). Similar observations were also noted in T₁ transgenic tobacco plants (Figure 14).

Biochemical quantification of cellulose contents

Cellulose contents of five independent transgenic plants (S1 to S5) expressing BbKst gene and empty vector transformed tobacco plants were analysed as stated in method and the quantitative data are represented in Table 1. Cellulose content of these transgenic lines varied from 13.17-16.58% fresh wt. with an average of 14.82% fresh wt, whereas 11.9% cellulose content was noted in the vector-transformed tobacco plants. Overall a significant (P ≤ 0.05) 1.25 fold increase of cellulose contents in transgenic tobacco plants expressing BbKst gene was noted, compared to that of vector-transformed plant (Table 1).

Plant material

Young leaf and internodes of Bambusa balcooa (Roxb.) were collected from natural strands of Bhadreswar (22° 49′ N 88° 21′ E), West-Bengal, India.

RNA ligase mediated rapid amplification of cDNA ends (RLM RACE)

Total RNA was isolated from the fifth internodal tissue of bamboo following protocol adapted by Rai et al. [32]; RACE was performed using First Choice RLM RACE kit (Invitrogen, USA) following manufacturer’s instructions. Primers used in 5′ RACE were designed using the information of 5′ end sequence of the marker, Balco₁₁₂₈. Similarly 3′ end sequence information was used to design primers for 3′ RACE (Additional file 8: Table S2). Products obtained from RACE were gel purified using Qiagen gel extraction kit (Cat No 28704; QIAGEN; USA) and sequenced.

Genomic DNA isolation and genome walking

The PCR based genome walking library was constructed by Genome Walker Universal Kit (Cat no. 638904; Clonetech, U.S.A). Briefly, high quality genomic DNA was isolated from surface sterilized tissue with 250 μl of warm CTAB extraction buffer [33], and digested with four different restriction enzymes namely, DraI, EcoRV, PvuII, StuI for overnight at 37°C. The digested DNA was purified, and ligated with Blunt end adaptors (25 μM, provided with the kit) for overnight at 16°C. Adaptor ligated DNA was used as template for first round genome walking PCR employing both adaptor specific (AP1, provided with the kit) and gene specific primers (Additional file 9: Table S3). The longest product was eluted from the 1% agarose gel and sequenced with BigDye Terminator Reaction Mix (Applied Biosystems, USA) using an automated DNA sequencer (ABI Prism 3100) following manufacturer’s instruction. Four rounds of walking were required to obtain the full length BbKst clone.

In-silicosequence analysis

Both the sequences obtained from genome walking and RACE were aligned by NCBI blast alignment using Bioedit software. Positions of promoter sequence, transcription start site, introns and exons were determined by NCBI- blast and by multiple alignments with ClustalW (http://www.ebi.ac.uk/Tools/msa/clustalw2/). The in silico translated amino acid sequence derived from the BbKst gene was further analyzed for conserved domains and motifs using ProSite and Inter-Proscan.

Bacterial expression of BbKst- kinase cDNA

In vitrokinase assays

Substrate phosphorylation experiments were carried out employing 10 μg of BbKst-kinase in 50 μl of reaction mixture (25 mM Tris-HCl; pH 7.5, 1 mM DTT; 5 mM MgCl₂ and 10 μM ATP) in presence of 10 μM [γ ^-32P] ATP (3000 cpm pmol^-1) at 30ºC with 250 mM of different substrates viz. casein, myelin basic protein (MBP), histone (H1), LRRKtide and GS peptide. Phosphorylated products were analysed by SDS-PAGE and autoradiographed. Activity of BbKst-kinase was determined by plotting its specific activities with varying concentrations of casein as substrate. Steady-state kinetic parameters were evaluated using Michaelis-Menten kinetics. The V _max and K_m value of BbKst-kinase using casein as a substrate following Eadie-Hofster plots of V₀ versus V₀/[S]. The turnover number was calculated from K_cat at the optimal substrate concentration per active site and the average time required to hydrolyze one molecule of casein was determined.

Effect of protein kinase inhibitors of BbKst-kinase activity

Several potent inhibitors of protein kinases like heparin, H-89, quercetin and staurosporine were tested on BbKst-kinase following recommended methods of Davies et al. [34].

Phospho-amino acid analysis

The purified BbKst protein was labeled in vitro with [γ-³²P] ATP and autophosphorylated protein electroblotted onto a polyvinylidine difluoride membrane. Radioactive band of interest was excised and hydrolyzed in 6 N HCl for 2 hr at 110°C [35]. The hydrolysate was concentrated and spotted on TLC plate and analysed along with Phosphoamino acid standards such as phospho- Ser, phospho-Thr, and phospho-Tyr (25 nM each; Sigma). TLC plates was then exposed to X-ray film and subsequently developed, phosphoamino acids standards were visualized by ninhydrin spraying over the TLC plate.

RNA slot blot hybridization

Different intermodal tissues (viz. I2, I5, I8, I10, I12, I15) were collected from three independent clumps of B. balcooa separately and processed for extracting total RNA as described earlier (35). Twenty μg RNA of each sample were loaded in formaldehyde denatured agarose gel and electrophoresed at a constant voltage of 50 V. The RNA was transferred to Hybond-N + membrane (Amersham-Pharmacia Biotech) with Bio-Rad Trans-Blot SD Semi-Dry Electrophoretic Transfer Cell at a constant current of 400 mA for 45 min. 1265 bp long cDNA fragment was radiolabelled with [α-³²P]dCTP using Prime-a-gene labeling system kit (Promega, USA) as per the manufacturer’s instructions. The membrane was incubated with the ³²P labeled gene specific probe at 65°C overnight in hybridization buffer (Sigma). After hybridization the membrane was washed twice with 2× SSC containing 0.1% SDS for 30 min each at 65°C followed by washing for 30 min (2 times) in 1× SSC and 0.5× SSC containing 0.1% SDS at 65°C. The membrane was exposed to X-Ray films for 24 hr. with intensifying screen and films were developed. The same procedure was followed for GAPDH as control. All experiments were repeated three times.

qPCR analyses of BbKsttranscripts in bamboo

cDNA was synthesized from the total RNA of leaf, stem, different internodal tissues (2^nd, 5^th, 8^th, 10^th, 12^th and 15^th from the tip) and rhizome of B. balcooa using First strand cDNA synthesis Kit (Cat # K1622. Fermentas). qPCR reactions were performed in presence of gene specific primers using a LightCycler PCR System qPCR instrument (MJ Research, Bio-Rad; Model CFD-3220). Fold differences in the BbKst transcript level were quantified using the 2^ΔΔC_T method [36, 37], considering fold difference of GAPDH as 1.0.

Construction of plant expression vector and Agrobacteriamediated transformation

The full length 3339 bp cDNA of BbKst was inserted between Spe1 and Nco1 restriction sites in a sense orientation in pCambia 1304. Positive construct was designated as Js1304 (Figure 9). Agrobacterium tumefaciens strain C58C1 was transformed with Js1304 by heat shock method. Js1304 integrated-Agrobacteria mediated transformation of tobacco was performed following the protocol of Kumar et al. [38]. Ten independent transgenic tobacco lines were generated with the Js1304 construct and maintained in the green house until seeds were mature. Seeds were harvested and germinated in presence of 30 ug^-ml hygromycin and segregation ratio was analyzed.

Transient expression of the BbKst

Gold-spermidine mixture (25 mg of 1 μm gold particles + 100 μl spermidine) was added to 25 μg of Js1304 plasmid DNA and mixed prior to the addition of 100 μl of 1 M CaCl₂. The onion peels were placed on the wet filter paper and kept in a Petridish. The gold particles were fired at 200 psi from a 3 cm distance of a Particle Gun (Helios Gene Gun, Bio-Rad make). Each onion peel was put in the Petri dish keeping the epidermal side in contact with the Murashige and Skoog’s (MS) medium [39]. Petridishes were sealed with parafilm and kept at 23˚C for 24 hr. Each onion piece was taken on a glass slide and GFP fluorescence was visualized and image captured in confocal laser scanning microscope (CLSM) using LAS AF (Leica Application Suite Advanced Fluorescence, 1.8.1) software with a confocal pinhole set at “Airy” 5.50 and zoomed to a factor of 1.4× for improved 8- bit resolution. Materials were excited at 488 nm wave length and emission was at 518 nm.

GUS histochemical assay

Histochemical GUS assay was performed following the method of Jefferson [40] in order to monitor the transient gene expression in 21 days old T₁ plants. Tissues were incubated in 1 mM chromogenic substrate XGlcA (5-bromo-4-chloro-3-indolyl-β-D-glucuronic acid) containing GUS substrate solution for overnight at 37°C. Then tissues were fixed in the fixative for at least 4 hrs followed by treatment with 50% and 100% ethyl alcohol for complete decolorization. Finally tissues were transferred to GUS fixative solution and examined under light microscope and photographed.

Southern hybridization with partial kinase domain of BbKstas probe

Total genomic DNA from young leaves of GUS-positive transgenic plants and vector-transformed plants was isolated following the protocol mentioned above. For Southern analysis, approximately 10 μg of genomic DNA was digested with 100 units (about 10 U/μl) of two enzymes Spe1 and Nco1, electrophoresed in a 0.8% agarose gel and subsequently transferred to a Hybond-N + membrane (Amersham-Pharmacia Biotech). DNA fragment of 1265 bp was amplified with JNWF1 (5′-CTAAAAGGCACCCTGATATGGACAGCATC-3′; Additional file 9: Table S3); and JF3R (5′-GCAAGAAGCATGTACCTATTGAC-3′) primers and then eluted from the gel, 25-50 ng of the eluted DNA was used as a probe after labelling with [α-³²P]-dCTP using Prime-a-gene labeling kit (Promega, USA) as per manufacturer’s instructions. Hybridization was carried out overnight at 65°C. After hybridization the membrane was washed twice with 2× SSC buffer containing 0.1% SDS for 30 min each at 65°C followed by washing for 30 min (2 times) in 1× SSC and 0.5× SSC buffers containing 0.1% SDS at 65°C. The membrane was covered with the saran-wrap, exposed to X-ray film (Kodak) with intensifying screen and kept at -80 ºC for 48 hr.

Expression analysis of BbKstin transgenic tobacco plant

Expression analyses of BbKst in transgenic tobacco plants and empty vector transformed plants were carried out in Light-Cycler PCR System qPCR instrument (MJ Research, Bio-Rad; Model CFD-3220). Primers used for qPCR were 5′-CCTCGGTGTTCCTTCCCTTCTGTC-3′ and 5′-GTCAATAGGTACATGCTTCTT GC-3′. The reaction conditions were as follows: initial denaturation at 95°C for 5 min, followed by 40 cycles of denaturation at 94°C for 45 sec, annealing at 56°C for 30 sec, and extension at 72°C for 1 min, final elongation step at 72°C for 10 min. Melting curve was analysed by continuous monitoring of fluorescence between 60°C and 95°C with 0.5°C increment after every 30 sec. Tobacco 18S rRNA house keeping gene was used as internal control. Fold differences in the BbKst transcript level were quantified using the 2^-ΔΔC_T method [36, 37], considering fold difference of 18S rRNA as 1.0.

Measurement of stem thickness of transgenic tobacco plants expressing BbKstgene

The stem thickness (diameter) of 5 month old transgenic tobacco plants and empty vector-transformed control tobacco plants grown in green house conditions (28°C and 78% humidity) were measured with slide calipers. Measurements of 7^th to 20^th internodes from tip of these plants were taken. Three independent measurements from each internode were noted and the mean value was scored. Width of the xylem zone (xylem vessels and xylary fibers) of independent transgenic and control plants were measured under the microscope from at least 5 samples and the mean value was tabulated. Width of xylem zone (containing xylary fibers and vessels) of five independent transgenic lines was compared with respective stem width. Similar comparison was also made for one vector-transformed tobacco plant.

Physical characteristics of fiber cells

Small slivers were prepared from 19^th internode (from the tip) by macerating the tissue with 67% nitric acid, boiled in a water bath for 10 min and subsequently washed thoroughly in distilled water following the protocol of Bhattacharya et al. [26]. Measurements of fiber diameter, cell wall thickness and lumen diameter (i.e. fiber diameter - cell wall thickness) of 25 intact fiber cells isolated from five T₁ transgenic plants were taken. The following equations were used to derive the physical traits: slenderness ratio = fiber length/fiber diameter, flexibility coefficient = (fiber lumen diameter/ fiber diameter) X 100 and Runkel ratio = (2 X fiber cell wall thickness)/lumen diameter [26]. Duncan’s Multiple Range Test (DMRT) was performed to analyze the results proximity with each other, the test was made in SPSS statistical software.

Histological analysis of cellulose contents of fiber cells

Transverse hand sections were taken from the 18^th internodes (from tip of the stem) of vector- transformed and transgenic tobacco plants. Sections were stained with 0.2% aqueous solution of Congo Red (CR; Sigma, C6767) overnight in the dark following the standardized protocol prescribed by Bhattacharya et al. [26]. Briefly, the stained sections were washed with deionized water, and images were acquired using CLSM with a confocal pinhole set at ‘Airy’ 1.0. A 30% Argon laser with Acousto-Optical Tunable Filter (AOTF) for 514 nm (20 mW) was used for exciting CR-labeled samples, and red fluorescence emissions were acquired between 560 and 800 nm with the PMT detector gain set at 1100 V. Fluorescence was recorded after passage through a triple dichroic filter 458/514/594. Images were captured using a HCX PL APO objective (63X/NA 1.40) under the above stated conditions.

Estimation of cellulose contents in transgenic plants

The α-cellulose content was estimated following the colorimetric method based on anthrone reagent [26, 41]. Approximately 1.0 g of 3^rd internodal tissues from base were crushed, mixed with 3 ml of nitric acid and acetic acid solution (1:8 v/v) and refluxed in boiling water for 30 min. Lignin, hemicellulose and xylosans were removed through successive three washing followed by centrifugation at 10000 rpm for 10 min. The resultant pellet was dissolved in 67% sulphuric acid (v/v), mixed with chilled anthrone reagent (HiMedia Laboratories, India), and incubated for 20 min in a boiling water bath. Cellulose estimation was made spectrophotometrically at 620 nm (Beckman-Coulter, DU-520). The α-cellulose content was estimated from the standard curve prepared using different known concentrations of cellulose (Merck, Germany) and was expressed as the gram percentage of fiber fresh weight.