Previous Article | Next Article ![]()
Applied and Environmental Microbiology, November 2006, p. 7339-7344, Vol. 72, No. 11
0099-2240/06/$08.00+0 doi:10.1128/AEM.01324-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
,
Department of Chemistry and Ilse Katz Center for Nanotechnology,1 Department of Immunology, Ben Gurion University of the Negev, Beer Sheva 84105, Israel2
Received 9 June 2006/ Accepted 13 September 2006
|
|
|---|
|
|
|---|
Various membrane-active compounds are released by bacteria to their environments (2, 30), a process that often has an essential functional role as a means for overcoming host defense mechanisms, allowing colony proliferation, and facilitating bacterial communication (17). Membrane-active peptides and toxins, in particular, are produced by bacteria, for example, pneumolysins secreted by streptococci (27) and
-toxin, which is the major cytolysin emitted by Staphylococcus aureus (1). Secretion of pore-forming exotoxins by bacteria is abundant, and endotoxins, such as lipopolysaccharides, which are often released by gram-negative bacteria, strongly interact with membrane components of the host cell (21). Here, we describe a new bacterial detection platform generating dramatic visible color changes accompanied by intense fluorescence emission that are induced by membrane-active molecules secreted by proliferating bacteria.
|
|
|---|
The bacteria used in the studies were Salmonella enterica serovar Typhimurium 1a (strain CS093 [22]), Bacillus cereus, Escherichia coli K-12 strains C600, C600 pMRInv, and MC4100 (provided by Ehud Gazit, Tel Aviv University), and E. coli BL (provided by Dudy Bar-Zvi, Ben Gurion University).
Bacterial growth.
Salmonella enterica serovar Typhimurium 1a, E. coli BL, and E. coli K-12 C600 were grown to saturation at 37°C and streaked. E. coli K-12 MC4100 was grown in LB medium supplemented with streptomycin (10 µg/ml). E. coli K-12 C600 pMRInv was grown in LB medium supplemented with kanamycin (30 µg/ml). Bacterial concentrations were determined from absorption at 600 nm on a Jasco V-550 spectrophotometer.
Vesicle preparation.
Vesicles containing DMPC and 10,12-tricosadiynoic acid (2:3 molar ratio) were prepared at a concentration of 1 mM. The lipids were dried together in vacuo. Following evaporation, distilled water was added and the suspension was probe sonicated at 70°C. The resultant vesicle solution was cooled at 4°C overnight and then polymerized by irradiation at 254 nm for 0.5 min.
Visible spectroscopy.
Samples were prepared by adding 30 µl bacterial samples (at the different concentrations tested in the experiments) to 30 µl of 1 mM vesicle solutions, followed by the addition of 30 µl Tris base at 50 mM, pH 8, and dilution in water to 1 ml prior to spectral acquisition. All measurements were carried out at room temperature on a Jasco V-550 UV/VIS spectrophotometer.
Preparation of chromatic lipid-PDA agar matrix.
Unpolymerized DMPC-polydiacetylene (PDA) vesicles at a concentration of 5 mM were added right after the sonication stage to hot LB agar. The mixture was then cooled to room temperature and poured into a six-well plate (Cellstar, Greiner Bio-one). After solidification of the agar, the plate was kept at 4°C for 2 days and polymerized by irradiation (254 nm, 40 s) in a UV cross-linker (UV-8000; Stratagene, California).
Multiwell fluorescence spectroscopy.
On the surface of the chromatic DMPC-PDA agar (in each well), we spotted 10-µl diluted bacterial samples (107 particles), placed the plate in a multiwell fluorescence plate reader (Fluoroscan Ascent; Thermo, Finland), and kept it at 26°C or 29°C. All measurements were carried out using 485-nm excitation and 555-nm emission, using LP filters with normal slits. Acquisition of data was automatically performed every 30 min.
Color scanning.
Scanned images were recorded on an Epson Perfection 4990 photo scanner.
All experiments were repeated at least three times.
|
|
|---|
![]() View larger version (26K): [in a new window] |
FIG. 1. Chromatic sensor concept. (Left) Agar scaffold containing vesicular nanoparticles composed of phospholipids (black) and PDA (blue). (Right) Bacterial proliferation (green oval) on the agar surface causes blue-to-red transformation of embedded vesicles due to bacterially secreted compounds that diffuse through the agar.
|
The new bacterial sensor exploits the chromatic response induced upon interaction of lipid-PDA assemblies with membrane-active substances as the underlying mechanism for bacterial detection. Previous studies have employed chemically derivatized PDA for bacterial sensing; however, those systems relied mostly on the covalently displaying specific recognition units on the PDA surface, producing signals only through actual binding of bacterial proteins to the PDA matrix. Such assemblies require chemical modifications of the PDA units, and their detection levels are determined by the number of bacterial particles attaching to the PDA-displayed receptors (15, 23). A previous study employing PDA for bacterial sensing exhibited a detection limit of 108 bacterial particles (16). The concept we describe here is simpler and more generic and features higher sensitivity. Essentially, we show that bacterially released substances diffuse through the agar matrix and interact with the incorporated phospholipid-PDA nanoparticles, consequently inducing the colorimetric/fluorescence transitions that result from bacterial presence.
Color detection of bacteria.
Fig. 2 depicts a representative experiment showing the color transitions induced by bacteria in DMPC-PDA agar. Specifically, Fig. 2B and C show scanned images of DMPC-PDA agar plates onto which we transferred colonies of the gram-positive Bacillus cereus and the gram-negative Escherichia coli BL strains, respectively, and incubated them at 26°C. The pictures in Fig. 2B and C clearly show that red hallows form around the bacterial colonies following incubation (note that the apparent "doublets" in Fig. 2B are due to the reflection of the scanner light). The blue-to-red transformation of the matrix was directly related to bacterial proliferation; each colony was surrounded by an area in which the blue agar matrix changed color to red, while the remaining blue agar matrix stayed blue (Fig. 2B and C). The dispersion of red regions under and around the bacterial colonies indicates that the color transitions were due to diffusion of substances released by the bacteria into the surrounding matrix.
![]() View larger version (40K): [in a new window] |
FIG. 2. Color transitions induced by bacterial colonies. Scanned images of DMPC-PDA agar plates prior to bacterial growth (A), 18 h after three colonies of Bacillus cereus were transferred (B), and 18 h after three colonies of E. coli BL were transferred (C). The plates were incubated at 26°C.
|
![]() View larger version (37K): [in a new window] |
FIG. 3. Reactions of bacteria with lipid-PDA vesicles. (A) Colors observed after mixing of DMPC-PDA vesicles with the resuspended pellet (left) and supernatant (right) after Salmonella enterica serovar Typhimurium 1a was grown overnight at 37°C (number of bacteria, 109). (B) Relationship between the bacterial growth curve and the blue-to-red transformation of PDA. The bacterial growth curve (black) and blue-to-red transition (blue) of a solution containing Salmonella enterica serovar Typhimurium 1a in growth medium mixed with DMPC-PDA vesicles (bacteria grown at 26°C) are shown. The extent of the blue-to-red transformation is reflected in the ratio of the intensity of the peak at 500 nm (the red peak in the visible spectrum) to that at 640 nm (the blue peak). Experiments were repeated at least three times. The variability of color response was between 10% and 15%.
|
Bacterial detection through fluorescence.
In addition to color transformations, the fluorescence properties of PDA can be exploited as well for bacterial detection using the new assay. Figure 4 shows three-dimensional fluorescence "topographic maps" recorded in a conventional multiwell enzyme-linked immunosorbent assay reader (excitation, 485 nm; emission, 555 nm). In the experiment whose results are depicted in Fig. 4, we compared the fluorescence responses of conventional agar (used as a control) (Fig. 4A and B) and DMPC-PDA agar (Fig. 4C and D) following bacterial proliferation. Specifically, we transferred microscopic colonies of Salmonella enterica serovar Typhimurium, extracted from a fresh agar plate, onto the agar surfaces and recorded the fluorescence intensities at different time points using a multiwell fluorescence reader, with the temperature maintained at 26°C.
![]() View larger version (43K): [in a new window] |
FIG. 4. Fluorescence induced by bacterial colonies. Three-dimensional fluorescence intensity distributions, recorded using a multiwell fluorescence reader, in conventional bacterial growth agar plates used as a control (A and B) and DMPC-PDA agar plates (C and D) are shown. The z axes depict the fluorescence intensity (arbitrary units), while the x and y planes correspond to the well surface area. (A and C) Initial measurements (time zero). (B and D) Measurements 6 h after Salmonella enterica serovar Typhimurium 1a colonies were streaked on the agar surface. Growth was carried out at 26°C.
|
Practical applications.
Fig. 5 and 6 depict the results of experiments in which the new assay was used for evaluation of bacterial resistance to antibiotics (Fig. 5) and for detection of bacteria in foods (Fig. 6). In the experiment whose results are shown in Fig. 5, we streaked E. coli strains exhibiting different antibiotic resistances onto plates containing the lipid-PDA agar matrix that further incorporated antibiotic compounds and incubated the plates at 29°C. Figure 5B features fluorescence distribution maps of the plates recorded 8 hours after streaking that clearly reveal which of the bacteria were resistant to the antibiotic compound present in the agar matrix. For example, fluorescence emission appeared in a plate containing kanamycin-DMPC-PDA agar onto which the kanamycin-resistant E. coli C600 pMRInv strain (7) was streaked (Fig. 5B, top left), while no fluorescence change occurred when the E. coli K-12 C600 strain, a bacterium that cannot grow on kanamycin-containing substrates, was streaked (8). Similarly, fluorescence induction was recorded only in a streptomycin-DMPC-PDA agar plate onto which the streptomycin-resistant E. coli K-12 MC4100 strain was streaked (Fig. 5B, bottom left). Importantly, the fluorescence transitions could be detected earlier than the bacterial colonies, highlighting the usefulness of the new assembly as a bacterial screening tool. The striking color transitions induced by proliferation of the antibiotic-resistant strains (Fig. 5C), which appear exactly in the regions where fluorescence was recorded, further emphasize the practicality of the new assay for visual screening of bacterial antibiotic resistance.
![]() View larger version (74K): [in a new window] |
FIG. 5. Chromatic screening of bacterial antibiotic resistance. A portion of a multiwell plate containing the DMPC-PDA agar matrix, further incorporating the antibiotic compounds kanamycin (top rows) and streptomycin (bottom rows), is shown. The bacterial strains (1 x 107 bacterial particles) streaked in the wells were (top left) E. coli K-12 C600 pMRInv (kanamycin resistant), (bottom left) E. coli K-12 MC4100 (streptomycin resistant), and (bottom and top right) E. coli K-12 C600 (not resistant to either antibiotic). (A and B) Area distributions of fluorescence intensities recorded using a multiwell fluorescence reader (color key for relative intensity units shown on left). (A) Measurements at time zero. (B) Measurements 8 hours after streaking. (C) Scanned color image of the plate after an 18-h incubation. Growth was carried out at 26°C.
|
![]() View larger version (38K): [in a new window] |
FIG. 6. Inspection of bacterial contamination in food. DMPC-PDA agar plates incubated overnight at 26°C are shown. (A) Control. (B) Plate contacted with exposed liver. (C) Plate contacted with exposed liver that was boiled for 3 min.
|
|
|
|---|
The sensitivity of the new assay is essentially determined by the number of bacteria and the time of detection; even a single bacterium can be detected following its proliferation on the agar and reaction with PDA. This feature distinguishes this sensor from existing bacterial sensing approaches and PDA-based assays that rely mostly on specific recognition elements displayed by the bacteria.
The reported experiments demonstrate that bacterially secreted molecules accumulate and diffuse within the porous agar scaffolding, subsequently inducing the color and fluorescence transformations of lipid-PDA. The chromatic sensing capabilities of lipid-PDA assemblies have previously been demonstrated in various applications; in particular, the vesicles were previously shown to constitute a useful platform for detection of membrane-active and amphiphilic molecules (28). A recent study demonstrated the induction of colorimetric transitions in lipid-PDA vesicles by a pore-forming bacterial toxin (14). In the context of this work, the lipid-PDA nanoparticles constitute the reporter element within the sensing assembly, emitting signals in response to bacterial presence.
Utilization of either visual color changes observed by the naked eye or fluorescence emission as a viable detection method is an important advantage of the new bacterial sensing system. While the pronounced color changes should facilitate bacterial detection by nonexperts in diverse settings, the fluorescence properties of the chromatic matrix could additionally open the way for employing the platform for high-throughput screening applications requiring high sensitivity, large sample quantities, or automation. Indeed, the fluorescence images in Fig. 3 and 4 underlie the enhanced sensitivity of the platform to bacterial proliferation and the possibility of detecting bacteria in shorter times than with conventional microbiology methods.
Microbe selection was designed to demonstrate the general applicability of the assay, and the choice of bacterium was also related to the type of experiment presented. Specifically, the experiments aimed to show the fundamental features of the technique (Fig. 2 to 4) were carried out with widely used conventional bacterial species: E. coli (gram negative), B. cereus (gram positive), and S. enterica serovar Typhimurium (representing common pathogenic bacteria). In the experiments depicting specific applications of the technique for antibiotic screening (Fig. 5), we have used E. coli genotypes containing the desired antibiotic resistance.
The chromatic platform could be employed for diverse practical applications. Utilizing the assay for screening for antibiotic resistance of bacteria is feasible through incorporation of antibiotic compounds within the chromatic matrix. Figure 5, for example, demonstrates that evaluation of bacterial sensitivity to antibiotic substances can be carried out in such matrixes by using a conventional fluorescence multiwell reader. The experiment whose results are depicted in Fig. 5 also points to applications of the chromatic platform for rapid screening of putative antibacterial properties of compound libraries. The color/fluorescence sensitivity of the chromatic assembly and the amenability to conventional high-throughput screening formats are attractive features of the new system for diagnostic and industrial utilization. The facile blue-to-red changes following bacterial proliferation could be useful in other applications, such as monitoring food freshness. Figure 6, for example, demonstrates that visible color transformations provide a direct alert on the presence of bacteria in a contaminated food specimen.
The new bacterial detection assay is not capable, at this stage, of differentiating among bacteria. However, the generality of the detection concept and the nonspecificity of the chromatic platform can, in fact, be advantages in applications in which reporting on the presence of any type of bacteria is required, for example, monitoring sterile environments, evaluating food freshness, screening for bacterial resistance to existing and new antibiotic compounds, and others. Research currently being carried out in our laboratory is exploring avenues for facilitating specificity of detection among bacterial species by varying the lipid composition within the chromatic platform.
Published ahead of print on 22 September 2006. ![]()
Supplemental material for this article may be found at http://aem.asm.org/. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»