Bacterial strains
The bacteria Klebsiella pneumoniae was provided by the Department of Microbiology, Faculty of Medicine, Brawijaya University. They were isolated from a patient with severe pneumonia, and susceptibility test against several antibiotics using Vitek@2Compact (BioMיrieux, France) indicated that the K. pneumoniae was resistant to ciprofloxacin and trimethoprim/sulfame-thoxazole, cefazolin, ceftazidime, ceftriaxone, cefepime, and aztreonam. Lactobacillus strains were isolated from goat milk kefir that was purchased from a small dairy in Malang, Indonesia. One gram of kefir grains as well as kefir was homogenized in 9 mL sterile H₂O, and serially diluted from 10^–1 to 10^–4. Of each dilution, 100µL was spread on de Man Rogosa and Sharpe (MRS) agar (Merck Millipore) and anaerobically incubated at 37°C for 48 h¹¹. The colonies were counted, and each colony was streaked on MRS medium. Individual colonies were inoculated into milk yeast extract (MYE) broth (10% skim milk, 15% glucose, 0.5% yeast extract; HiMedia) with 5% glycerol, and the stocks were stored at –20°C.

Biofilm assay

Lactic acid bacteria that have been successfully grown from whole kefir as well as kefir grain were sub-cultured in 15 ml tubes containing 5 ml of MRS broth and incubated for 16 hours at 37°C. After incubation, LAB cultures were centrifuged at 12,000 rpm for 15 min, and the cell free supernatants (CFS) and pellets were separated. K. pneumoniae isolates were sub-cultured in Luria Bertani (LB) broth and incubated at 37°C for 18 h. Next day, after ascertaining confluency at OD₆₀₀ 40.5, 200µL of the culture was dispensed per well in a 48-well plate (Corning-Costar®). This was followed by the addition of 200µL LAB CFS or pellet (diluted with MRS medium), or MRS medium as the negative control. The plates were incubated for 6 h at 37°C¹¹.

Biofilm inhibition was tested using crystal violet¹¹. The cultures were discarded, and the plates were washed thrice by submerging in water. To each well, 125µL of 0.1% crystal violet solution (Bioworld®) was added and incubated for 15 min at room temperature (RT). The staining solution was discarded, and the plate was washed thrice with sterile water to remove the excess stain. The resulting biofilms were diluted with 125µL 30% acetic acid per well, and after incubation for 15 min at RT, the OD of the supernatant was measured at 570nm using ELISA reader (iMark™ Microplate Absorbance Reader, BioRad). Biofilm formation (and inhibition) was quantified after removing the planktonic cells on the basis of OD₅₇₀ values: the isolates with OD₅₇₀ < 0.120 were considered as weak biofilm producers, those with OD₅₇₀ 0.120–0.240 had biofilm formation ability, and isolates with OD₅₇₀ > 0.240 were classified as potent biofilm producers¹². The supernatants and pellets from two LAB isolates resulted in the lowest OD₅₇₀ values, indicating best inhibitory effects, and were chosen for further analysis. Each sample was analyzed in triplicate.

MTT assay of Lactobacillus cell-free supernatant against K. pneumoniae

The four LAB isolates with the lowest OD i.e. K10 and K15 (from milk kefir) and G24 and G25 (from kefir grain) were each sub-cultured in 1 mL MRS broth and incubated anaerobically for 16 h at 37°C. The cultures were centrifuged at 12,000 rpm for 15 min and the CFS and pellets were separated. The K. pneumoniae were sub-cultured in LB medium at 37°C for 18 h. After reaching confluence (OD₆₀₀ 0.5), 100µL of each culture was dispensed per well in a 96-well plate (Corning-Costar®), followed by addition of 100µL LAB CFS or pellet, or MRS medium (negative control). The co-cultures were incubated at 37°C for 6 h.

The antibacterial activity of LAB CFS on K. pneumoniae was determined using the MTT assay¹³. Briefly, 5µL MTT solution (Cell Quanti-MTT^TM Cell Viability Assay Kit) was added to each well followed by 1 h incubation at 37°C away from light. The cultures were discarded, and the wells were washed with 250µL PBS each. To each well, 100µL dimethyl sulfoxide (DMSO; Merck) was added and incubated for 10 min at 37°C away from light followed by 10 min agitation. The OD of the supernatants was measured at 570 nm using an ELISA reader. The procedure was repeated three times for each sample. OD₅₇₀ values were converted to percentage of cell viability (% viability) relative to the control group by using the following equation:

% activity = (Abs_sample/Abs_control) x 100,

where Abs _sample refers to sample treated with LAB CFS, and Abs_control refers to the untreated control. High OD₅₇₀ indicated high cell viability¹⁴.

Microscopic determination of biofilm inhibition using scanning electron microscopy (SEM)

Biofilm formation on the surface of plastic coverslips (diameter 13 mm) (Thermanox®) was monitored as per the manufacturer’s instructions. Briefly, a coverslip was placed inside each well of a 24-well plate containing 200µL K. pneumoniae in LB broth and 200µL LAB CFS or pellet from soluble MRS and incubated for 6 h. The culture media was removed, and the wells were washed several times with water. The coverslips were air dried and placed onto larger (22 x 22 mm) glass coverslips for imaging using FEI Verios 460 scanning electron microscope (FEI, Oregon, USA).

Identification of Lactobacillus
The LAB isolates were first identified based on colony morphology, physiology, and biochemical reactions as per the criteria of Bergey’s Manual of Systematic Bacteriology. Lactobacillus was characterized by cell morphology, Gram reaction, catalase and oxidase activity, sporulation, and cell motility. The final 4 LAB isolates with potent anti-biofilm activity were identified to the species level by using 16S ribosomal RNA gene sequencing. Total genomic DNA was isolated from 5 mL overnight cultures grown in MRS broth at 37°C using spin column method according to manufacturer’s instructions (PureLink® Genomic DNA, Life Technologies Corp., Carlsbad, USA). Targeted genes of each isolate were amplified using Lac16f (5′ -AGAGGTTTGATCCTGGCTCAG-3′) and Lac16r (5′ -CTACGGCTACCTTGTTACGA-3′) primers¹⁵. A 50µL PCR reaction mix was prepared for each sample containing 25µL 2X AmpMaster®Taq (GeneAll Biotechnology, Songpa-gu, Korea), 100 ng template DNA, and 10 pmol/µL of each primer. The PCR conditions were as follows: initial denaturation at 95°C for 2 min, followed by 35 cycles of 2 min denaturation at 95°C, 20s annealing at 49°C and 1 min extension at 72°C, and a final extension at 72°C for 5 min. The PCR products were visualized with DNA gel electrophoresis Mini-Sub® Cell GT Systems (Bio-Rad Laboratories, Singapore), and DNA fragments corresponding to the correct length of 16S rRNA were purified using Gel/PCR DNA Fragments Extraction Kit® (Geneaid Biotech, Taiwan). The purified products were sent for sequencing to Macrogen Corporation (Seoul, South Korea). The nucleotide sequences of the 16S rRNA gene were reconstructed using Unipro UGENE software version 1.24.0, and the identities of the isolates were determined on the basis of the highest BLAST score (http://www.ncbi.nlm.nih.gov.blast) against the GenBank DNA database (www.ncbi.nlm. nih.gov/Genbank/).

QS assay via inactivation of AHL by LAB

To determine whether the inhibitory effect of LAB on K. pneumoniae biofilm formation was related to the inactivation of AHL-mediated QS, AHL inactivation was qualitatively assessed with the help of a biosensor agent, Agrobacterium tumefaciens NT1 (tra::lacZ749) (ATCC®)¹⁶ (kindly provided by Dr. Ating, Faculty of Fishery, UB). The reporter strain was grown in LB medium supplemented with 25µg/mL gentamicin at 28°C until it reached an OD₆₀₀ 1.5. LAB isolates were inoculated in MRS broth medium and incubated overnight at 28°C under microaerophilic conditions. The cultures were centrifuged for 5 min at 13000 rpm, and then 50µL supernatant was dispensed into each well of a 96-well plate containing 50µL of 10µM 3OC8HSL per well (Sigma-Aldrich). The plates were incubated at 28°C for 5 h.

Minimal agar medium plates supplemented with 50µg/mL X-gal were prepared, and the agar was cut into 6 separate bars 1-cm wide using a sterile scalpel by removing 2-3mm strips between the bars. Starting from one end of each bar, 5µL of the LAB CFS and 3OC8HSL mix were spotted along its length, followed by 0.5µL drops of the A. tumefaciens culture; MRS broth medium and 3OC8HSL without LAB served as negative control. The plates were tightly sealed using plastic wrap and incubated for 24 h at 28°C. The biosensor spots turned to blue in the presence of detectable AHL signal.

Statistical analysis
One-way ANOVA was used to evaluate the effect of the CFS and pellet of LAB on in vitro biofilm inhibition test and viability test (optical density). Following ANOVAs, Tukey’s test was used for multiple pairwise comparisons among treatment levels. Significance level was set at P<0.05 for all comparisons. All statistical analyses were performed using SPSS version 16.