Plant materials and chemicals
The explants of Gloriosa superba L. were collected and selected for further studies. The chemicals used for the experiment were of analytical grade quality and were obtained from SD Fine, India, and Himedia, India. The Silica gel 60F254 (20 cm × 10 cm) HPTLC plate used for the chromatographic studies was obtained from E. Merck (Darmstadt, Germany). The standard Colchicine was procured from Natural Remedies, Bangalore, India.
In vitro tuberization
For in vitro studies, terminal shoot tips were used as explants. The explants were washed with running tap water for 5–10 min. Then, all the selected explants were surface sterilized by dipping with 0.1% (w/v) mercuric chloride for 0.5–1 min and at last rinsed thoroughly 2–3 times with sterile double distilled water. Sterilized explants were then inoculated into culture media under laminar airflow. The full-strength Murashige and Skoog (MS)  medium was prepared by supplementing various concentrations and combinations of plant growth regulators (PGRs) like BA (N6-benzyl adenine) alone (1–3 mg/l), in combination with α-naphthalene acetic acid (NAA). (NAA 0.2 mg/l), combination of Kinetin + 2,4-Dichloro phenoxy acetic acid (2,4-D) (0.2 + 0.5, 0.5 + 1 and 1 + 1.5 mg/l). The treatment without any PGRs was considered as a control. The pH of the medium was adjusted to 5.8 with NaOH/HCl and dispensed into culture tubes and phytajars (15–20 ml). The media was sterilized by autoclaving at 121 °C and 1.063 kg/cm3 pressure for 20 min. The cultures were maintained at 28 ± 2 °C under 16:8 h light/dark photoperiod under cool white fluorescent lamps at a photosynthetic photon flux density of 70 μmol m−2 s−1. All the periodic data were collected for each parameter.
In vivo tubers of nearly two months were collected from the natural plant, whereas in vitro tubers have been developed through tissue cultures were harvested at different intervals of 2, 4, 6, 8, 10, and 12 weeks of culture. All the samples were thoroughly washed with double distilled water and dried under shade at room temperature. Samples were powdered with a mechanical grinder and further extracted through a simple maceration method at room temperature. For extraction, 1 g powdered plant samples were mixed with 10 ml of methanol and incubated for 24 h in the shaker at 200 rpm. After 24 h, centrifugation was done and the supernatant was collected and filtered through Whatman filter paper No. 41. The obtained filtrate was concentrated and was used for further analysis [19, 20]. A common stock solution of colchicine was prepared by dissolving it in methanol at the concentration of 1 mg/ 1 ml. The stock solution was then further diluted in methanol to obtain a solution containing the concentration of 100 ng/µl and was used as a working standard for the HPTLC study.
Chromatographic studies performed on a CAMAG HPTLC unit consist of Linomat-V automatic sample applicator, TLC scanner III, and integrated software WINCATS (CAMAG, Switzerland). The thin-layer chromatography was performed on a pre-coated silica gel HPTLC 60F254 (20 cm × 10 cm) plate of 0.20 mm layer thickness, used as a stationary phase for the quantification of Colchicine. The plant samples and the standards were spotted on the TLC plate at 8-mm-wide bands with the help of the Linomat applicator V equipped with a 100 µl syringe (Hamilton, Switzerland) under a continuous flow of nitrogen gas at a constant application rate of 150 nl s−1. Before placing the plate, Camag twin trough chamber (20 cm × 10 cm) was saturated with 20 ml mobile phase with chloroform/methanol: formic acid (20:1:0.2) for 20 min, under room temperature (28 ± 2 °C) and 55 ± 5% relative humidity. The spotted plate was then placed into a pre-saturated chamber and the chromatogram was run at 8 cm from the base. After that, the plate was dried with an air dryer and further quantitative evaluation was performed by scanning with TLC scanner III in the absorption–reflection mode at 254 nm. The slit dimension was set at 6 mm × 0.45 mm, with a data resolution of 100 nm s−1 and scanning speed of 20 mm s−1.
The present method was validated as per the ICH (International Council of Harmonization) guidelines . For the linearity study, standard solutions at 100–1000 ng/spot were analyzed. The calibration curve was generated by plotting the peak areas against the concentrations with the help of regression analysis. For the measurement of detection limit (LOD) and quantification limit (LOQ), the formula suggested by Alam et al.  was used. To measure the recovery study, the known samples were spiked with an extra 80, 100, 120% of the standard colchicine (standard addition method) and the mixtures obtained by this method were analyzed. The experiment was performed three times at each level.
Precision, reproducibility, and specificity
The system precision was estimated by using repeatability of the sample application, and peak areas were measured by using eight replicates of the same band (600 ng per spot of colchicine). The intraday and inter-day precision for colchicine was carried out by spotting eight samples at a concentration of 600 ng/spot that were analyzed on the same day and three different days, respectively. The precision was calculated in terms of relative standard deviation (%RSD) of obtained peak area. The specificity of the present method was determined by comparing the obtained Rf values from TLC and the spectrum of the colchicine band from a plant sample with those of a standard.
Quantitative determination of colchicine from in vivo and in vitro tubers
Extracts of plant samples and standard solutions (2,4,6,8,10 µl) were spotted on silica gel60 HPTLC plates of size 20 cm × 10 cm and analyzed as mentioned above. The peak area of standard colchicine was determined from the calibration plot of peak area against the amount of colchicine (standard). The standard compound present in the different samples was determined by using a calibration plot, and data were recorded.
Completely randomized design (CRD) was used for performing tissue culture experiments. Each experiment was conducted with three replicates by using 10 samples. The recorded data were analyzed statistically using SPSS (version 20, SPSS Inc. Chicago, USA) software. The one-way analysis of variance (ANOVA) was used to determine the effect of treatment. Mean values of data were compared by Duncan multiple range test at P ≤ 0.05.