Identification of Bacterial strains
Totally, 20 bacterial strains were previously isolated from the wastewater. Potential strain showing highest flocculating activity were screened for further research. Test bacteria were recognized as Proteus mirabilis through partial sequencing of 16S rRNA genes. Nucleotide sequences were finally deposited in GenBank (MG996004).

Production and partial purification of bioflocculant
Bioflocculant producing media was prepared based on the methods of Zhang et al.²⁰. Components of the media such as KH₂PO₄ (2.0 g), Glucose (15.0 g), (NH₄)₂SO₄ (0.2 g), Yeast extract (0.5 g), K₂HPO₄ (5.0 g), MgSO₄ · 7H₂O (0.2 g), NaCl (0.1 g), were dissolved in 1L of distilled water (pH=7) under optimum conditions. 0.5 ml culture of P. mirabilis isolated from wastewater, inoculated into the medium and kept it for 72 hrs incubation at 30°C, 160 rpm/min.  The broth was centrifuged (8000׳g, 20 min) to pelletize the cells. The upper portion of the bioflocculant were added with thrice the volume of ethanol. This washing step was repeated several times to obtain the precipitate. Thereafter, they were assayed for flocculation activity.

Characteristics of the bioflocculant
Carbohydrate content of the bioflocculant using glucose as the standard solution were measured using Phenol-sulfuric acid method²¹.  Lowry method was assessed in measuring the protein content of the bioflocculant²². The monomeric unit of the bioflocculant was obtained by hydrolysis of the bioflocculant with HCl at 121°C for 2 hrs. They were further subjected to the analysis of FT-IR. The scanning electron microscopy (SEM: SUPRA 55 SAPPHIRE; Germany) image of the bioflocculant was obtained.

Gas Chromatograph-Mass Spectrometer
The extracted samples were subjected to GC/MS analysis (TQ8030, Shimadzu, Japan). For identifying sugars components, carrier gas used was helium the with a flow rate of 1.0 mL min-1, maintaining the temperature of 300°C with an electron ionization mode (70 eV). Compounds were identified by the NIST1 1library database using chromatograms.

Heavy metal removal rate
The bioflocculant (0.8mg/100ml) was applied for the bioflocculation test against heavy metal solutions containing Cd²⁺, Zn²⁺, Cu²⁺, and Hg²⁺ at the concentrations of 0-50 mg L-1, respectively. The amount of heavy metals were identified by inductively coupled plasma atomic emission spectroscopy (ICP-AES; iCAP 7000, Thermo Fischer Scientific, Bremen, Germany). All experiments were repeated thrice. The percentage of heavy metal removal was measured based on the following equation:

% R= (C_i –C_f) /m

Where, Ci – amount of heavy metal of test solution initially; C_f – amount of heavy metal of test solution finally; m (g) is the weight of bioflocculant.

Antibiofilm assay
Crystal violet assay was performed which was examined to quantify the biofilm inhibition percentage²³.

Biofilm of Pseudomonas aeruginosa and Staphylococcus aureus grown separately on glass slides in well plates were added with the bioflocculant. They kept the incubation for 24 hrs at 37°C and washed with phosphate buffered saline (0.1 M; pH 7.4) to discard non-adherent cells. In the same way, control were also subjected to the same procedure without the bioflocculant. The glass slides were observed under confocal microscopy (CLSM, LSM 510, Carl Zeiss, and Germany) after the Acridine orange staining. All the experiments were conducted thrice.

Production of bioflocculant
Totally, 20 bacterial strains were isolated from the sewage water. In which, strain showing highest flocculating activity were selected for bioflocculant production. The strain were identified as P. mirabilis (MG996004) through 16S rRNA sequence²¹. Phylogenetic tree based on 16S rRNA sequences were constructed (Fig. 1). Morphological structure of P. mirabilis were identified (Fig. 2). Totally, 2 g of bioflocculant was obtained from 1L fermentation broth under optimal conditions²⁰. The production of bioflocculant was mainly controlled by the factors, such as the components of the culture medium and the cultivation conditions²⁴. The key factors, such as the concentration of carbon sources, temperature, shaking speed, and nitrogen sources, pH, and inoculum size have to be focused on the flocculating activity of the bioflocculant by Proteus mirabilis²⁵.

Fig. 1. Phylogenetic tree based on 16S rRNA sequences

Fig. 2. a) Image of Gram negative bacteria Proteus mirabilis observed under the confocal microscope; b) SEM image of Gram negative bacteria Proteus mirabilis.

Characteristics of bioflocculant
The bioflocculant produced from Proteus mirabilis (MG996004) included 68% (wt/wt) carbohydrate, and 15% (wt/wt) protein. These components were confirmed by FTIR spectroscopy by identifying the functional group of the bioflocculant (Fig. 3). The broad peak at 3433 cm^-1, which was confirmed it as hydroxyl and amino group²⁶. Sharp peak observed at 1635 cm^-1 was attributed to the characteristic vibration of amide group. Weak peak at 1,358 cm^-1 was correlated to the carboxylate group of the bioflocculant²⁷. The adsorption band at 672 cm^-1 could be related  to the binding between sugar monomers²⁸. Finally, spectrum revealed the peaks for polysaccharides, protein and identified the presence of the functional groups: hydroxyl (-OH), amide (-CO-NH) and carboxyl (-COO-)²⁵. These functional groups act as a binding site for divalent cations, mainly involved in flocculating processes.

Fig. 3. FTIR spectra of bio occulant showing presence of hydroxyl, amino, and carboxylic group.

GC/MS analysis of bioflocculant was found the monomers, viz., D-Mannose, D-fructose, D-Sucrose, D-Galactose (retention times of 14.18, 15.98, 23.18, and 20.37 min, respectively) and compared with NIST Library (Fig. 4). The results indicated that the bioflocculant made up of four sugar monomers^{29, 30}. HR-SEM showed the outer structure of the bioflocculant. Representative HR-SEM image showed well-dispersed, irregular shaped bioflocculant (Figs. 5(a)) with negatively charged zeta potential possessing best flocculating property²². The configuration of the bioflocculant were major factor involved in flocculating activity²¹. During flocculation, they bonded tightly and intertwined each other due to intermolecular interaction in between them^22,31.  The elemental analysis for the bioflocculant were shown tthe presence of C, K, Na, Mg, Cl, and O at particular proportions ²⁹ (Fig.s 5(b)).

Fig. 4. GC/MS chromatogram of monomeric sugar units identified in the bioflocculant
a) D-Fructose, b) D-Sucrose, c) D-Mannose, d) D-Galactose.

Fig. 5. (a) Irregular shape of the bio occulant; (b) SEM-EDAX spectra depicting the presence of C, K, O, Na, Ca, Mg, and Cl in the bio occulant.

Heavy metals removing efficiency
In the study, the metal ions such as Zn²⁺, Cd²⁺, Cu²⁺, and Hg²⁺ within the arrange of 0–50 mg L-1 in aqueous solutions could be effectively flocculated by the bioflocculant. The heavy metal removing percentage achieved by the precipitated bioflocculant was 82.63 ± 1.20, 72.076 ± 0.42, 57.36 ± 1.05, and 44.7 ± 1.053%, respectively for Zn²⁺, Cd²⁺, Cu²⁺, and Hg²⁺ at  optimum conditions. When compare the treated sample with untreated sample, concentration of metals in treated sample was completely reduced that showed adsorption and bioflocculating ability of bioflocculant. In addition, concentration of bioflocculant reported was very less on comparing other reported literature^32,33. It was interpreted that the electro-negative properties of the biological material and the metals regulated the metal-removal capacity. In addition, the interactions between metal cations and negatively charged functional group of bioflocculant such as carboxylic group, and the phosphate group found in bioflocculant (Fig. 6) were highly enhanced under constant optimum conditions³³.

Fig. 6. Removal percentage of  Cu²⁺, Zn²⁺, Hg²⁺, Cd²⁺ using bioflocculant. The given data represents the mean value of three independent experiments. Error bars indicate mean ± SD.

Antibiofilm assay
Crystal violet staining (Biofilm assay) assay are used to quantify the measurement of biofilm. Bioflocculant were showed excellent biofilm inhibition against P. aeruginosa where maximum reduction (73%) were achieved at the amount of 80µg/ml dose. Similarly, maximum reduction (80%) were achieved at the amount of 70µg/ml dose for S. aureus³⁴.  Control (without bioflocculant) did not show any inhibiting percentage exhibited dense bacterial population (Fig. 7). High inhibition percentage and minimum dosage requirement were showed that, bioflocculant was acting as a best antibiofilm agent. They were further visualized by confocal microscope^35,36. Previously reports suggested that, bioflocculant reduced cell –cell interaction by inhibiting the attachment of bacterial cells³⁷ and also polysaccharides found in the bioflocculant might be responsible for biofilm inhibition³⁸.

Fig. 7. (a) CLSM images of P. aeruginosa biofilm formation; (b) CLSM images of S. aureus biofilm formation; (c) CLSM images of P. aeruginosa biofilm formation treated with bioflocculant (80 µg/ml);  (d) CLSM images of S. aureus biofilm formation treated with bioflocculant (70 µg/ml); Each experiment was replicated thrice.