All the materials and reagents for bacterial cultures and TLC, were purchased from Merck (E. Merck, Darmstadt, Germany).

Water samples were obtained from tannery wastewater of a company in Qom, Iran. Approximately 1 ml of each sample was poured in 20 ml of nutrient broth medium with 20 % solar salt. The pH was adjusted to 7.2. The flasks were incubated on a shaker at 37 °C until turbidity appeared. The turbid cultures were streaked on nutrient agar 20 % with solar salt and incubated at 37 °C. After 14 day of incubation, yellow colonies were selected and purified (Asker and Ohta, 1999[2]). Amongst 15 isolated halophilic strains, QWT-12, a bacterial strain with sharp yellow colonies were selected for further experiments. Salt tolerance assays for QWT-12 strain were performed by growing it in nutrient broth plus various concentrations of NaCl (0 to 15 % w/v) at 37 °C. The bacterial growth was monitored at temperature range of 10-50 °C with intervals of 5 °C and over a pH values range from 5.0 to 10.0. Sodium acetate buffer was used for pH 6 and lower while Tris-HCl buffer was used for pH higher than 6.

Morphological and physiological characteristics of the strain were either studied on nutrient agar or in nutrient broth plus 3 % (w/v) NaCl. Gram reaction, shape and color of colony, catalase, indole formation and oxidase activities, nitrate reduction and tween 80 hydrolyses were checked as recommended by Smibert and Krieg (1994[27]). Carbon and nitrogen sources utilization were performed according to Ventosa et al. (1982[31]).

Genomic DNA from strain QWT-12 was prepared using the method described by Marmur (1961[14]). The 16S rRNA gene was amplified by PCR with the forward primer 16F27 and the reverse primer 16R1488 as described by Mellado et al. (1995[16]). Direct sequence determination of the amplified DNA was carried out using an automated DNA sequencer (ABI 3130 XL; Applied Biosystems) at the SeqLab Laboratory (Göttingen, Germany). The 16S rRNA molecule phylogenetic analysis was performed using the software package MEGA version 5 (Tamura et al., 2011[28]) after obtaining multiple alignments of data available from public databases (Thompson et al., 1997[30]). Clustering was performed using the neighbour-joining (Saitou and Nei, 1987[25]), maximum-parsimony and minimum-evolution methods. Bootstrap analysis was used to evaluate the tree topology of the neighbour-joining data by performing 1000 resamplings (Felsenstein, 1985[7]).

Five culture media including TSB, NB, HM, SW and MH mediums were used to find the proper medium for pigment production. The cells were grown at 37 ºC for 24 hours in each medium. After evaluation of optical density and optical pigmentation, TSB medium culture was selected as proper medium to optimize carotenoid production. Additionally, gradient of NaCl, NaNO₃, Na₂SO₄ and KCl in TSB medium were also evaluated to optimize cell growth and pigmentation. Growth curve of strain QWT-12 was drawn in rational to OD and cell pigmentation to find proper time point for pigment extraction.

In order to pigment extraction, the yellow pigmented bacterial isolates were grown in TSB with 3 % (w/v) of salt, in a rotary shaker at 180 rpm in 37 °C for three days. After three days, cells were harvested by centrifugation at 8,000 g for 15 min. Then the pellet was washed with sterile distilled water and spin for 15 min at 4,000 g. Five ml of methanol was added to the pellet and suspended it. Then it was incubated in a water bath at 60 °C for 15 min until all visible pigments were extracted and centrifuged at 4,000 g for 15 min. The colored supernatant was separated, then it was filtered through Whatman no. 1 filter paper. The carotenoid extracts were analyzed by scanning the absorbance in the wavelength region of 450 nm using the spectrophotometer. The total carotenoid content in the methanol extract was estimated by measuring the absorbance at λmax (450 nm) (Ravindar et al., 2014[23]). For MTT assay after evaporating of methanol, the carotenoids were dissolved in ethanol (Abbes et al., 2013[1]).

Human cancer cell lines that were used in this project were DU145, PC3, LNCaP for prostate cancer and MDA-MB-468, MDA-MB-231, MCF-7 for breast cancer and A549 for lung cancer. As control cell line we used human skin fibroblast cell line Hu02. All cell lines were purchased from IBRC Cell bank. The cells were grown as a monolayer in an RPMI 1640 medium (Gibco) supplemented with 10 % (v/v) fetal bovine serum (FBS) (Gibco), 1.0 mM sodium pyruvate (Sigma) and 2 mM L-glutamine (Sigma) at 37 °C in a humidified atmosphere of 95 % air and 5 % CO₂. After 24 h of growth, the cells were transferred into a 96-well plate (2000 cells/well) and treated with the carotenoid extracts for 48 h. The carotenoid extracts were dissolved in RPMI medium and served as a stock solution that was later diluted to have a final solvent concentration.

The MTT-assay was used to evaluate the carotenoid extract effect on mentioned cell line's cell viability, and the results were expressed as viable cell percentages with respect to the control. Approximately 2000 cells/well were seeded onto a 96-well plate and allowed to adhere for 24 h. Three replicates of each plate were incubated with different concentrations of carotenoid extracts (0.5, 1, 2, 4, 8 mg/ ml), and viability was recorded at 48 h. After treatment, the medium was removed, and 20 μl of 5 mg/ml solution of MTT (Sigma) in PBS were added to each well. The plate was then incubated for 3 h at 37 °C. Finally, the medium was removed, and 200 μl of the DMSO (Merck) were added in each well to solubilize the blue formazan. Dye absorbance was measured at 560 nm (Abbes et al., 2013[1]).

The extracted total carotenoids were subjected to TLC using following system: Plate: silica gel 60 F254, Merck; Solvent system: acetone: hexane (40+60); Comparing solutions: regarding to online pigment libraries, stock solution of Neoxanthin, Violaxanthin and Neurosporene were considered to molecular structure alignment (Weber and Davoli, 2003[32]).

A TLC plate, on which the pigment was developed, was sprayed with an absolute chloroform solution of 20 % antimony chloride (III). Mass spectrophotometry test has been carried out in order to align results with online library of similar carotenoids and finally determining the exact structure (Weber and Davoli, 2003[32]).

According to phenotypic characterization, strain QWT-12 was a Gram-stain-positive, non-motile, coccus, non-spore forming, facultative aerobic bacterium. Figure 1(Fig. 1) shows the colony of strain QWT-12 on nutrient agar with 3 % salt and optical microscope respectively. The phylogenetic position of strain QWT-12 was determined based on 16S rDNA gene sequencing. 1445 bp of 16S rDNA gene of strain QWT12 with accession no. KX073818 was determined. Alignment of this sequence with described species showed this strain belonged to the genus Kocuria with more than 99.8 % similarity to Kocuria marina (Figure 2(Fig. 2)). Biochemical characterization results of QWT-12 strain in comparison with Kocuria marina KMM 3905^T have been exhibited in Table 1(Tab. 1).

Kinetics of bacterial growth and pigment production were investigated in the nutrient broth plus 3 % (w/v) NaCl. Lag phase of the bacterial growth was 15 h and after 23 h, growth reached stationary phase. Pigment production activity was undetectable during the early- and mid-exponential growth phase but started from the end of the exponential phase and continued in the stationary phase (Figure 3(Fig. 3)). To determine the optimal growth conditions yielding the highest pigment production activity, the effect of various culture media was analyzed.

In Figure 4(Fig. 4), comparative effectiveness of each culture medium for cell growth and cell pigmentation were exhibited elaborately. Of the five culture media investigated TSB medium was the best for growth and pigment production. Effects caused by gradient of NaCl, NaNO₃, Na₂SO₄ and KCl in TSB medium on cell pigmentation and OD were shown in Figure 5(Fig. 5). According to these results the proper concentration of different salts for pigment production and growth was 0.1 M for NaNO₃, Na₂SO₄ and KCl and 0. 5 M for NaCl.

In Figure 6(Fig. 6) viability percentage results gained from using five gradient concentration levels of Kocuria carotenoid against seven cancer cell lines have been shown. Abbreviations located beside histogram are due to cell lines while Hu02 has been considered negative control. After calculation for the chi-square test significant IC_50s of 1, 4 and 8 mg ml of extracted pigments against MCF-7, (A-549 and MDA-MB-468) and MDA-MB-231 cancer cell lines were obtained respectively. The extracted pigments have no effect on viability of fibroblast cell line and only 20- 30 percent decrease the viability of prostate cancer cell lines.

Extracted pigments were separated into several spots by TLC. The orange colored pigment which had the highest affinity for the mobile (solvent) phase migrated towards the top of the TLC, and had Rf equal to the obtained Rf from stock solutions. Two other extracted pigments were also present, including the yellow pigments which have the higher affinity for the stationary (silica) phase. The carr-price test for all spots confirming that these pigments are carotenoids. The UV/light absorption spectra of two yellow pigments were identical to those violaxanthin and neurosporene (Figure 7(Fig. 7)). The absorption maxima of extracted pigments in different solvents including chloroform, ethanol and petroleum ether in comparison with neoxanthin, violaxanthin and neurosporene were exhibited in Table 2(Tab. 2). Number of conjugated double bonds (N) can be estimated following the formulae obtain from (Rivera et al., 2014[24]).

In ethanol:

For both two pigments, it was found that N=9 and according to chemical structure of neoxanthin (N=8), violaxanthin (N=9) and neurosporene(N=9) we could estimate that these extracted pigments are violaxanthin or neurosporene.

To more structure elucidation of the extracted pigments, mass spectrophotometry was done for the yellow pigment, as it seems to be the major part of the pigment extraction. Figure 8(Fig. 8), exhibited that the molecular weight (MW) of yellow pigment (yielded from TLC plate) is 538 and this has more similarity to neurosporene (MW=538.8) than violaxanthin and neoxanthin (MW=600). According to all above, it could be estimated that the principle pigment extracted form Kocuria sp. strain QWT-12 is neurosporene.