2.1 Bacteria strains and culture
S. mutans GS5 strains were used in this study. Bacteria were grown and
maintained in brain-heart infusion (BHI) broth (BD, Bacto) and agar plate (BD,
Difco). Strains were cultured at 37 °C under an anaerobic atmosphere with 95% N2
and 5% CO2 for 16-18 hours. For confocal laser scanning microscopy observation,
GFPuv-tagged GS5 strains were generated by transformed with the GFPuv-contained
shuttle plasmid (pPDGFPuv) and selected by spectinomycin (500 μg ml-1).
2.2 Preparation of human platelets
Whole blood was collected from health donors in the laboratory by using venous
blood collection tubes containing sodium citrate 3.2 % (BD Vacutainer® citrate tube).
2.2.1 Platelet-rich plasma (PRP)
For preparation of platelet-rich plasma (PRP), whole blood was manipulated by
centrifugation at 1200 rpm for 12 min at 25°C and collected the upper layer. This PRP
would be used in biofilm formation and bacteria-platelet interacting assay to mimic
the condition in vivo.
13
2.2.2 Platelet-poor plasma (PPP)
The platelet-poor plasma (PPP), used as blank in platelet aggregation assay, was
collected by centrifugation at 3000 rpm for 10 min at 25°C. The upper layer would be
clearer than PRP.
2.2.3 Platelet suspension (PS)
The platelets suspension (PS) would be prepared for analyzing the direct
interaction between platelets and bacteria. Briefly, PRP was added with heparin (10 U
ml-1; B. Braun Melsungen AG) and Prostaglandin E1 (1 μM; Sigma-Aldrich) and
centrifuged at 3000 rpm for 6-8 min. Collect the pellet and resuspended gently with
Tyrode solution (11.2 mM glucose, 136.8 mM NaCl, 11.9 mM NaHCO3, 2.8mM KCl,
1.1 mM MgCl2, 0.33 mM NaH2PO4, 1.0 mM CaCl2, and 3.5 mg bovine serum
albumin per ml; pH 7.35~7.4). Heparin (6.4 U ml-1) and PGE1 (1 μM) were further
added to the suspension, and again, centrifuged at 2500 rpm for 5-7 min. After that,
the pellet was repeatedly resuspended with Tyrode solution. The concentration of PS
would be suggested at 3 x 108 ~5 x 108 platelets per ml.
2.3 Preparation of small molecules
A range of small molecules were kindly provided from Associate Professor
14
Ling-Wei Hsin (Graduate institute of Pharmaceutical Sciences, College of Medicine,
NTU). These 76 different small molecules were dissolved in dimethyl sulfoxide
(DMSO; Sigma-Aldrich) at 10 mg ml-1. The test concentrations in vitro were 100, 20,
10 μg ml-1 diluted with double-distilled water. For experiment in animal model, the
concentration was used at 6 mg kg-1.
2.4 Biofilm formation assay
To screen the targeted small molecules, the inhibiting effect would be test by
using bacterial biofilm formation assay in 96-well U-bottom polystyrene microtiter
plates (Greiner bio-one, NO. 650101) and condition was set in normal nutrient broth
(BHI) or PRP. S. mutans GS5 strain, cultured overnight, was centrifuged (3000rpm,
10 minutes) for removing the upper cultured medium. Pellet was washed by 1X
phosphate buffer saline (PBS) and suspended by sonication (40W) for 5 minutes.
After that, bacteria suspension was prepared as the concentration of 109 CFU ml-1
(optical density: A550 adjusted to 1.5) in 1X PBS and then seeded into wells as a ratio
of 1 to 100 in BHI supplemented 1 % glucose or 1 to 10 in PRP. Simultaneously,
small molecules with (or without) antibiotics (gentamicin 20 μg ml-1; penicillin 0.5 μg
ml-1) were added with. The incubation time was 18-20 hours at 37°C. After incubation,
15
the upper liquid was removed and washed twice with distilled water. The 96-well
plates were dried for 45 minutes and then stained with crystal violet (0.05%) for 5
minutes. After rinsed twice, the plates were distained with acetone-ethanol (ratio of 1
to 4) for 1 hour. By detecting the absorbance at 550 nm using a MicroELISA reader
(Dynatech Corp), the inhibition ability on biofilm formation would be semi-quantified.
Each experiment would be repeated three times in triplicate. For the confocal
microscopy observation, the biofilms would be cultured in 24-well plate which a
round glass coverslip was put in individual wells. After incubation, the coverslips
were washed by 1X PBS and fixed with 2 % paraformaldehyde.
2.5 Bacteriostatic analysis
To examine whether targeted small molecules could inhibit bacterial growth, it was necessary to quantify the bacterial population size and plot the values as a growth
curve. As mentioned before, bacteria suspension was prepared and inoculated into
BHI at 107 CFU ml-1. The inoculation was set as the first timepoint (“0 hour”), and
optical density was determined by measuring the absorbance at 550 nm. Repeatedly,
record the absorbance every hour until the 8th timepoint, and the 9th data was
measured at the 24th hour. All the records would be display as a curve by using
16
GraphPad Prism 5.
2.6 Platelet aggregation test
Platelets aggregation test could examine the ability of platelets activation. After
adding the agonist such as adenosine diphosphate (ADP) into PRP, platelets were
activated and sequencely underwent aggregation. Once platelets aggregated, the level
of light transmission would be increased. By measuring the difference of light
transmission level between pre- and post-stimulation, it could determine the ability of
platelets activation.
In this study, both S. mutans GS5 strain and ADP were used as agonists. The
concentration of bacterial treatment was 1010 CFU ml-1 and ADP was 10 μM.
Un-stimulated PRP was set as 0 % of light transmission and PPP was as 100 %. The
analysis was performed as following. PRP incubated with small molecules (100 and
20 μg ml-1) was pre-warmed for 3 minutes and stirred at 900 rpm in a test cuvette at
37 °C. After adding the agonist (bacteria or ADP), the level of platelet activation
would be monitored continually by the photocell in Lumi-Aggregometer (Payton
Scientific)43. If targeted small molecules have the inhibiting effect on
bacterial-inducing aggregation, light transmission level would not change for 25-30
17
minutes at least after addition of bacteria.
2.7 Bacteria-platelet interacting assay
For investigating whether targeted small molecules could interfere with the
interaction between bacteria and platelets, the experiment would be used as a
quantification assay. Bacterial cultured overnight were centrifuged at 3000 rpm for 10
minutes to remove the cultured medium, and the pellet was washed by 1X PBS. Again,
bacterial suspension was centrifuged and suspended by sonication (40 W) with
ELISA coating buffer instead. Next, bacteria were prepared as the concentration of
109 CFU ml-1 (optical density: A550 adjusted to 1.5) and then coated onto the 96-well
plates (Nunc MaxiSorp® flat-bottom 96 well plate) at 4 °C overnight. The upper
liquid part was removed and the plats were rinsed 1X PBST (1X PBS supplemented
with 1% Tween-20) once. Additionally, the wells were blocked by 1% bovine serum
albumin (BSA). The targeted small molecules were diluted with PRP
(indirect-binding assay) or PS (direct-binding assay), and then these samples were
added into wells which had coated with bacteria. The plates were incubated at 37°C
for 2 hours and sequencely washed by 1X PBST once again. Next, phosphatase
substrate (1 tablet per 5 ml buffer; Sigma-Aldrich) dissolved in specific buffer (0.1M
18
Na-acetate addition of 0.1% Triton-X100; pH 5.5) was added into wells, and the
plates were incubated at 37°C. If platelets bind to bacteria, phosphatase substrate
would interact with phosphatase, which located inside platelets, and undergo
chromogenic reaction (Fig. 6). By detecting the absorbance at 405 nm, the inhibition
effect on the bacteria-platelet interaction would be quantified.
2.8 Experimental streptococcal endocarditis rat model
All animal experiments in this study were approved by National Taiwan
University Institutional Animal Care and Use Committee.
2.8.1 Experimental schedule design
At the first day, rats were operated with cardiac catheterization. After 24 hour,
small molecules (I.V.) or Aspirin (I.P.) would be injected into these animals. 30
minutes later, these rats were treated with bacteria. For Aspirin (25mg kg-1;
Sigma-Aldrich treatment), there would be two more injections at 30 minutes and 3
hours after bacterial infection. To analyze the effect of targeted small molecules on
bacteria survive in circulation, blood was collected at three timepoints: 30 minutes, 3
hours and 24 hours after bacteria treatment. At the third day, these rats were sacrificed
and their hearts were removed (Fig. 8A).
19
2.8.2 Preparation for bacteria, small molecules and anesthetic
Preparation for bacteria (GFP-tagged or normal GS5 strain) and small
molecules were described as previously. The amount of bacterial infection was 109
CFU and the concentration of targeted small molecules was 6 mg kg-1. The anesthetic
(“Zoletil 50”, Virbac) were supplemented with muscle relaxant (“Rompun”, Bayer
HealthCare) and water (for injection only) as a ratio of 500:280:220 respectively.
2.8.3 Cardiac catheterization
8-week-old male rats (Wistar) were used as ideal endocarditis rat model. A
stainless steel (10 cm) embedded into a polyethylene tube (8 cm; I.D. 0.28mm and
O.D. 0.61mm; Becton Dickinson, PE10) as a catheter which was further bent as 1/4 of
circle by a tweezers. This catheter was then inserted into carotid artery through an
incision on the chest and moved on to the left ventricle along the vessel. The catheter
would tremble more and more intensely with heartbeat when it was closed to the
aortic valves. When it no longer went forward by the blood flow resistance, the
catheter remained right there and was fixed by sutures.
2.8.4 Quantification and observation
For quantification of the effect on bacteria survive in circulation, blood was drew
and lysed by 1% Triton-X100 (in 1X PBS). The cell lysate was plated on BHI agar
20
with optimized dilutions. To determine the density of bacteria colonized in
vegetations, the vegetations were removed and their masses were weighed. By
sonication in 1X PBS, the homogenous suspension was further plated on Mitis
Salivarius agar (MS agar; BD, Difco) added with 20% sucrose. Both BHI and MS
agars were incubated at 37 °C for 2 days. The biofilms inside vegetations were also
visualized by confocal microscopy and their thicknesses were measured by vertical
section quantification.
2.8 Confocal laser scanning microscopy (CLSM) analysis
By using confocal laser scanning microscopy (CLSM; Leica TCS SP5), the
detailed composition of biofilm formation in wells and inside vegetation would be
observed. Biofilms cultured in wells were washed with 1X PBS and fixed with 2 %
paraformaldehyde. GFP-tagged bacteria were visualized by detecting emission spectra
of GFP. For platelets observation, on the other hand, samples which were incubated
with 1% Triton-X100 for 5 minutes and stained with rhodamine-conjugated phalloidin
(1 to 500 dilution; Invitrogen) were traced by detecting the emission wavelength at
565 nm. For the vegetation observation, it was performed as describes previously.
21