Previous Article | Next Article ![]()
Applied and Environmental Microbiology, May 2009, p. 2931-2935, Vol. 75, No. 9
0099-2240/09/$08.00+0 doi:10.1128/AEM.02108-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

The Ohio State University, Columbus, Ohio 43210,1 Corning Incorporated, Corning, New York 14831,2 Pacific Northwest National Laboratory, Richland, Washington 99352,3 Johannes Kepler University of Linz, A-4040 Linz, Austria,4 Ecole Polytechnique Fédérale de Lausanne, CH-1015 Lausanne, Switzerland,5 Umeå University, SE-901 87 Umeå, Sweden6
Received 11 September 2008/ Accepted 27 February 2009
|
|
|---|
|
|
|---|
In the present study, we used atomic force microscopy (AFM) to probe the surface of live S. oneidensis MR-1 cells, using AFM tips that were functionalized with cytochrome-specific polyclonal antibodies (i.e., anti-OmcA or anti-MtrC). This technique, termed antibody recognition force microscopy (Ig-RFM), detects binding events that occur between antibodies (e.g., anti-OmcA) on an AFM tip and antigens (e.g., OmcA) that are exposed on a cell surface. While this is a relatively new technique, Ig-RFM has been used to map the nanoscale spatial location of single molecules in complex biological structures under physiological conditions (5, 9, 11, 13).
Anti-MtrC or anti-OmcA molecules were covalently coupled to silicon nitride (Si3N4) cantilevers (Veeco or Olympus) via a flexible, heterofunctional polyethylene glycol (PEG) linker molecule. The PEG linker consists of an NHS (N-hydroxysuccinimide) group at one end and an aldehyde group at the other end (i.e., NHS-PEG-aldehyde). AFM tips were functionalized with amine groups, using ethanolamine (6, 7). The active NHS ester of the NHS-PEG-aldehyde linker molecule was then used to form a covalent linkage between PEG-aldehyde and the amine groups on the AFM tips (6, 7). Next, anti-MtrC or anti-OmcA molecules were covalently tethered to these tips via the linker molecule's aldehyde group. This was accomplished by incubating the tips with antibody (0.2 mg/ml) and NaCNBH3 as described previously (7). The cantilevers were purchased from Veeco and had spring constant values between 0.06 and 0.07 N/m, as determined by the thermal method of Hutter and Bechhoefer (12).
Prior to conducting the Ig-RFM experiments, the specificity of each polyclonal antibody (i.e., anti-OmcA and anti-MtrC) for OmcA or MtrC was verified by Western blot analysis as described previously (24, 28). Proteins were resolved by both denaturing and nondenaturing polyacrylamide gel electrophoresis (PAGE). Briefly, 2.5 µg of purified OmcA or MtrC (23) was resolved by sodium dodecyl sulfate-PAGE or native PAGE, transferred to a polyvinylidene difluoride membrane, incubated with either anti-OmcA or anti-MtrC, and then visualized using the Amersham ECL Plus Western blotting detection kit. Anti-OmcA bound exclusively to OmcA, anti-MtrC bound exclusively to MtrC, and neither antibody showed cross-reactivity with the other cytochrome. Antibody specificities of anti-OmcA and anti-MtrC were also validated by immunoblot analysis of S. oneidensis whole-cell lysate (28).
To determine if MtrC or OmcA was expressed on the external surface of live bacteria when Fe(III) served as the TEA, Ig-RFM was conducted on wild-type versus
omcA
mtrC double mutant cells. For these experiments, bacteria were cultivated anaerobically with Fe(III), in the form of Fe(III) chelated to nitrilotriacetic acid (NTA), serving as the TEA (19, 23). Growth conditions have been described elsewhere (3, 15) and were based on previous studies (3, 15, 16, 18) that suggest that S. oneidensis MR-1 targets OmcA and MtrC to the cell surface when Fe(III) serves as the TEA.
An Asylum Research MFP-3D-BIO AFM or a Digital Instruments Bioscope AFM (16, 17) was used for these experiments. The z-piezoelectric scanners were calibrated as described previously (17). Cells were deposited on a hydrophobic glass coverslip and immersed in imaging buffer (i.e., phosphate-buffered saline [pH 7.4]). The hydrophobic glass coverslips were made as described previously (17) using a self-assembling silane compound called octadecyltrichlorosilane (OTS; Sigma-Aldrich). S. oneidensis MR-1 cells readily adsorbed onto OTS glass coverslips and remained attached to the coverslips during the entire experiment. No lateral cell movement was observed during the experiment, consistent with previous studies that used OTS glass to immobilize bacteria (15, 17, 18, 27).
The AFM tip was brought into contact with the surface of a bacterium, and the antibody-functionalized tip was repeatedly brought into and out of contact with the sample, "fishing" for a binding reaction with cytochrome molecules that were exposed on the external cell surface. Binding events were observed upon separating anti-OmcA- or anti-MtrC-functionalized tips from wild-type S. oneidensis MR-1 cells (Fig. 1). For the wild-type cells, we observed both nonspecific and specific interactions (Fig. 1).
![]() View larger version (26K): [in a new window] |
FIG. 1. Retraction force curves for anti-MtrC-functionalized tips (A) and anti-OmcA-functionalized tips (B) that are being pulled away from the surface of living omcA mtrC double mutant (gray dotted line) or wild-type (solid black line) S. oneidensis MR-1. These bacteria were adsorbed onto OTS glass coverslips. (C) Retraction curves exhibiting nonspecific binding, specific binding, or no binding between the AFM tip and the cell surface.
|
Figure 2 summarizes the frequency or probability of observing a binding event for both anti-OmcA and anti-MtrC tips. Each bar in Fig. 2 represents one experiment in which 500 to 1,000 force curves were collected between one AFM tip and two to four live bacterial cells. This figure does not make a distinction between specific and nonspecific binding. It simply shows the frequency of observing an attractive interaction as the antibody-functionalized tip was pulled away from the surface of S. oneidensis MR-1. Binding events occurred with roughly the same frequency when wild-type S. oneidensis MR-1 cells were probed with anti-MtrC-functionalized tips as when they were probed with anti-OmcA-functionalized tips (Fig. 2).
![]() View larger version (19K): [in a new window] |
FIG. 2. Histograms showing the frequency of observing a binding event for anti-MtrC-functionalized (blue) or anti-OmcA-functionalized (red) AFM tips on live wild-type S. oneidensis MR-1 (solid bars) or omcA mtrC double mutant (diagonally hatched bars) cells. The downward arrows designate injection of free antibody into the imaging buffer. The solid gray bars correspond to results obtained with unbaited AFM tips.
|
Second, we performed force measurements on
omcA
mtrC double mutant S. oneidensis MR-1 cells. This mutant is deficient in both OmcA and MtrC (19, 23, 24) but produces other proteins native to the outer surface of S. oneidensis MR-1. The resulting force spectra showed a noticeable reduction in binding events for the
omcA
mtrC double mutant cells (Fig. 2). The binding events that were observed for the double mutant were only nonspecific in nature (Fig. 1). This indicates that the antibodies on the tip do not participate in specific interactions with other proteins on the surface of S. oneidensis MR-1 cells.
As a final control experiment, force measurements were conducted on wild-type S. oneidensis MR-1 cells, using Si3N4 tips conjugated with the PEG linker but not functionalized with polyclonal antibody (unbaited tips). Like the results with the double mutant, the unbaited tips were largely unreactive with the surface of the bacteria (Fig. 2). Those binding events that were observed were nonspecific in nature. Taken together, these results demonstrate that the antibody-coated tips have a specific reactivity with OmcA and MtrC molecules. Furthermore, these force measurements show that MtrC and OmcA are present on the external cell surface when Fe(III) serves as the TEA.
To map the distribution of cytochromes on living cells, Ig-RFM was conducted on living S. oneidensis MR-1 cells that were growing on a hematite (
-Fe2O3) thin film. The conditions for these experiments were as follows. A hematite film was grown on a 10-mm by 10-mm by 1-mm oxide substrate via oxygen plasma-assisted molecular beam epitaxy (14, 16). The cells were grown anaerobically to mid-log phase with Fe(III)-NTA serving as the TEA. Cells were deposited onto the hematite thin film along with anaerobic growth medium that lacked Fe(III)-NTA. The cells were allowed to attach to the hematite surface (without drying) overnight in an anaerobic chamber. The following day, the liquid was carefully removed and immediately replaced with fresh anaerobic solution (pH 7.4). Ig-RFM was performed on the cells by raster scanning an antibody-functionalized AFM tip across the sample surface, thereby creating an affinity map (1). Force curves were collected for a 32-by-32 array. The raw pixilated force-volume data were deconvoluted using a regularized filter algorithm. The total time to acquire a complete image was approximately 20 min.
As noted above, attractive interactions between an antibody tip and cell resulted in relatively short-range, nonspecific and longer-range, specific adhesive forces (Fig. 1C). To distinguish between these two interactions, we integrated each force curve beginning at >20 nm and ending at the full retraction of the piezoelectric motor (
1,800 nm). This integration procedure quantifies the work of binding, measured in joules, between the antibody tip and a particular position on the sample. While this integration procedure does not totally exclude nonspecific binding, it does select for those events associated primarily with specific antibody-antigen binding. Figure 3 is the antibody-cytochrome recognition images for MtrC and OmcA. The corresponding height (or topography) images of the bacterial cells are also shown in Fig. 3.
![]() View larger version (97K): [in a new window] |
FIG. 3. Ig-RFM of live S. oneidensis MR-1 cells deposited on a hematite ( -Fe2O3) thin film. Height image (A) and corresponding Ig-RFM image (B) for a bare unfunctionalized Si3N4 tip. Height and corresponding Ig-RFM image for a tip functionalized with anti-MtrC (C and D) or anti-OmcA (E and F). Each panel contains a thin white oval showing the approximate location of the bacterium on the hematite surface. A color-coded scale bar is shown on the right (height in micrometers [µm], and the work required to separate the tip from the surface in attojoules [aJ]).
|
It is important to note that these affinity maps were collected on only a few cells because it so challenging to produce large numbers of quality images. Future work should be conducted on a population of cells. Until this time, these affinity maps can be used to provide a crude, lowest-order estimate of the number of cytochromes on the outer surface of living S. oneidensis MR-1. For example, there were 236 force curves collected on the bacterium shown in Fig. 3D. Thirty-eight of these curves exhibited a distinct, sawtooth-shaped, antibody-antigen binding event. In other words, MtrC molecules were detected in one out of every six force curves (16%) that were collected on the cell surface.
This probability can be compared to other independent studies that estimated the density and size of MtrC and OmcA molecules from S. oneidensis MR-1. Lower et al. (16) estimated that S. oneidensis has 4 x 1015 to 7 x 1015 cytochromes per square meter by comparing AFM measurements for whole cells to force curves on purified MtrC and OmcA molecules. Wigginton et al. (25) used scanning tunneling microscopy to determine that the diameter of an individual cytochrome is 5 to 8 nm. These values can be used to create a simple, geometric, close-packing arrangement of MtrC or OmcA molecules on a surface. Using this approach, cytochromes could occupy 8 to 34% of the cell surface.
This estimate is consistent with the observed number of putative MtrC molecules shown in Fig. 3D. Therefore, it appears that these affinity maps can be used as a lowest-order estimate for the number of cytochromes on S. oneidensis MR-1 even though we do not know a priori the exact configuration of the antibody tip (e.g., the concentration of antibody on the tip, the exact shape of the tip, the binding epitopes within the antibody).
In summary, the data presented here show that S. oneidensis MR-1 localizes OmcA and MtrC molecules to the exterior cell surface, including an EPS, when Fe(III) is the TEA. Here, the cytochromes presumably serve as terminal reductases that catalyze the reduction of Fe(III) through direct contact with the extracellular iron-oxide mineral.
We are grateful to J. K. Fredrickson and J. M. Zachara for helpful discussions, the suggestions of the anonymous reviewers, and the efforts and comments of the editor. We are forever grateful for the advice given to us by Terry J. Beveridge.
Published ahead of print on 13 March 2009. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»