|Home | About | Journals | Submit | Contact Us | Français|
Information obtained by atomic force microscopy (AFM) depends strongly on the kind of probe or tip used; therefore, probe and tip effects have to be taken into account when verifying or interpreting the data acquired. In many papers, double-tip effects have been mentioned while other research was done; however, there are only a few special reports on double- or triple-tip effects, especially double-probe effects. In our paper, metaphase chromosomes of Chinese hamster ovary (CHO) cells, aggregates of pectin molecules, membrane surface of mouse embryonic stem cells, and R-phycoerythrin-conjugated immunoglobulin G complexes were imaged by AFM with high-quality probes, double-probe cantilever, and double-tip and triple-tip probes, respectively, in order to determine double-probe, double-tip, and triple-tip effects during AFM scanning. We found that the double-probe, double-tip, and triple-tip effects share the same principle, and that these effects correlate with distance and height differences between probes of double-probe cantilever or tips of double-tip or multiple-tip probes. Since many other factors influence double-probe or double-tip effects, more in-depth studies must be undertaken. However, this initial research will make all users of AFM techniques aware of double-probe and double-tip or triple-tip effects during AFM scanning and aid in verifying or interpreting the data acquired.
Since its invention (Binnig et al. 1986), the atomic force microscope (AFM) has been widely used in chemistry, physics, surface science, biology, medical science, and so forth. With AFM’s popularization, more and more scientists have begun to apply the novel technique for studying their research. Therefore, it has become increasingly important, as a first step in using this technique, to characterize the quality of AFM probes and their effect on image quality.
Atomic force microscope images are usually affected by different kinds of artifacts, either because of the microscope design and operational mode or because of external environmental factors. These factors mainly include rigidity of sample surface (Chen et al. 1998); tilt or drift of the scanner; optical interference between the laser light reflected off the top of the cantilever and the light being scattered by the surface in the same direction (Méndez-Vilas et al. 2002); asymmetric cantilever tips (Kaupp et al. 1995); surface topography of scanning probe (Kitching et al. 1999); double-probe cantilever, double-tip, or multiple-tip probes; as well as the skill of the AFM instrument operator.
Many factors induce the formation of double-probes and double-tip and triple-tip probes. At present, AFM probes are commercially manufactured. However, poor quality probes, including double-probe cantilever, double-tip or multiple-tip probes, and so forth, may be produced before leaving the factory. Because of long-term storage, dust falling on the tip surface of probes will also produce these poor quality, altered probes. During AFM scanning, repeated scanning can create particles at the nanometer scale attaching on the tip surface of probes from substrates. In addition, probes scrape samples and form double-tip or multiple-tip probes (Jass et al. 2000). In experiments, the last phenomenon is the most common and is often overlooked.
Relative to double-tip or multiple-tip effects, double-probe effects are not as common and are not identified, especially when scanning samples at a small scale. Therefore, researchers have paid little attention to double-probe effects, and few reports studying these effects are found in the literature. Up until now, although there have been many publications on double-tip effects, most only described the phenomena of double-tip effects. Usually, when observing these phenomena, it became obvious that each of the randomly located features contained an identical substructure of an oval and an adjacent smaller circle, and that the tip shape will induce an artificial preferential direction (Schwarz et al. 1994). In recent years, only a few published reports have focused on double-tip effects; however, most only mentioned double-tip effects while performing other research. In this study, metaphase chromosomes of Chinese hamster ovary (CHO) cells, aggregates of pectin molecules, membrane surfaces of mouse embryonic stem cells, and R-phycoerythrin-conjugated immunoglobulin G molecules were observed by AFM with unaltered probe, double-probe cantilever, and double-tip and triple-tip probes, respectively. Not only the phenomena of these effects but also their principles are discussed, and some situations in which these effects are difficult to identify are introduced simply to provide a reference for researchers using AFM.
After 72 h incubation, CHO cells were arrested with 0.07 µg/ml colcemid (Demecolcine, Wako, Wako Pure Chemical Industries Ltd., Tokyo, Japan) for 60 min. The cell suspension was exposed to 0.075 M KCl at 37°C for hypotonic treatment for 20 min and fixed with methanol-acetic acid (3:1). Spreads of metaphase chromosomes were made by dropping the cell suspension onto coverslips, followed by air drying.
Phycoerythrin conjugated immunoglobulin G 100 mg/l (PE-conjugated IgG, Becton Dickinson Biosciences Pharmingen, San Diego, Calif., USA) and pectin powders (Sigma Aldrich Corp., St. Louis, Mo., USA) were diluted into 0.1 mg/l and 0.05 mg/ml with distilled water, respectively. Of these biomolecule samples, 5 µl were introduced onto freshly cleaved mica surfaces and then air fixed rapidly by vigorous waving of the slide. For a cell sample, cell suspension without hypotonic treatment was dropped onto coverslips and then fixed with 1% glutaraldehyde solution, followed by washing and air drying. All these samples on substrates were imaged by AFM.
Atomic force microscopy (Autoprobe CP, Thermomicroscopes, Veeco Digital Instruments, Santa Barbara, Calif., USA) was performed in air in tapping mode. The coverslips or micas carrying chromosome or cell or biomolecule samples were mounted onto the XY stage of the AFM, and an integral video camera was used to locate the regions of interest. Microfabricated silicon nitrite cantilevers (Park Scientific Instrument’s Veeco Digital Instruments) with a force constant of approximately 25 N/m were used. Repeated scanning of the samples confirmed that no physical damage occurred to them during imaging. The images were reproducible during repeated scanning. The AFM images were planar-leveled, using the software provided with the instrument (Thermomicroscopes Proscan Image Processing Software Version 2.1 Veeco Digital Instruments). The quantification of images was carried out as routine. The width and height of the samples were determined using the line-analytical function of the software.
To discuss the relationship between the relative position of double probes on double-probe cantilever with the relative position of real image and ghost image, an observation of the double-probe cantilever was performed by optical microscope (Fig. 1). In this figure, a black arrow points to the end of the triangle cantilever. The position of the original probe on the double-probe cantilever is shown by a white circle, and the other added one by a black arrow. The two probes on the double-probe cantilever are far from each other, about 10 µm. Figure 2 is the AFM images of the same metaphase genome of CHO cells scanned by the single-probe (Fig. 2a) and double-probe (Fig. 2b) cantilevers, respectively. There are around fifteen chromosomes in the genome, and some chromosomes have been lost during sample preparation. Comparing Figure 2b with a, it was found that, when imaged by the double-probe cantilever, ghost chromosomes (purple arrows) with the same shapes and tinged color as the corresponding real chromosomes (red arrows) appeared in positions several microns away from the corresponding real chromosomes, and the artificial preferential direction was induced by the double-probe effect (white long arrow in Fig. 2b). Moreover, the distances between real and ghost chromosomes in×and y directions (green coordinates in Fig. 2b) are coincident with those between the two probes on the double-probe cantilever (green coordinates in Fig. 2c). It was concluded preliminarily that the horizontal distance between the two probes determines the horizontal distance between the real and ghost chromosomes. Figure 3 shows the AFM topographic images and their cross sections’ height profiles of the enlargement of the regions surrounded by black and white squares in Fig. 2b, in which the real and ghost chromosomes were shown, respectively. According to Figure 3, it was found that the real chromosome is over 100 nm higher and a little rougher than its ghost chromosome, and that some lower parts at the edge of the chromosome disappeared in the ghost image (just as shown by a white arrow in topographic images and a black arrow in height profiles). When scanning the region of the ghost chromosome with a single-probe cantilever, only topographic images of the coverslip surface were obtained.
Figures 4 and and55 are AFM images and height profiles of pectin aggregates and cell membrane surface induced by double-tip effect, respectively. In Figures 4a and and5a,5a, there were lower and narrower ghost images alongside the real pectin aggregates on mica and also of particles on the cell membrane surface. The horizontal distance between the real and ghost objects is the same and is distributed in the artificial preferential direction induced by the double-tip effect. In Figures 2b, ,4,4, and and5,5, the horizontal distances between the real and ghost objects were around 10 microns, 70 nm, and 100 nm, respectively, and the vertical distances between the real and ghost objects were about 100 nm and several nanometers, respectively. Especially in Figure 5, some real objects of larger size overlapped with their ghost objects (black long arrow), and some of smaller size were separated from their ghost objects (black short arrows). A few studies on multiple-tip effects have also been reported previously (Ohnesorge et al. 1997). Figure 6 shows AFM images of PE-conjugated IgG complexes induced by triple-tip and multiple-tip effects. The same phenomena in these figures as in the figures of chromosomes and pectin aggregates were observed. It can be presumed that triple-tip or multiple-tip effects may have the same principle as double-probe or double-tip effects.
Just as in the double-probe cantilever, the relative positions between the real and ghost images are presumed to be determined by the relative positions between the two tips on the double-tip probe, although directed morphologic observation of the double-tip probe was not performed. In our experiment, we used a single-probe and single-tip conical probe with a height of approximate 4 microns, a basal diameter of 1–2 microns, and a curvature diameter of 10 nm. Figure 7 is the schematic representation of a single-probe and single-tip cantilever, a double-probe cantilever used for chromosome imaging, and a double-tip probe for imaging of pectin aggregates and cell membrane surface, respectively.
In the experiment, samples were imaged by AFM in the constant-height mode, and the feedback was dependent on interaction between probe and sample. Take a double-probe cantilever for example (Fig. 8a): suppose the probe a (gray) is higher than the probe b (white); when probe a reaches positions 2 and 4, the real images of substrate and objects are obtained, because of the interaction between probe a and substrate or objects. When probe a reaches position 6 and, at the same time, probe b is in the position in which the object lies, an interaction between probe b and the object induces an ghost image apart from its real image on position 6, because the vertical distance between probe b and the object is shorter than that between probe a and the substrate, and feedback is dependent on probe b but not probe a. When probe a reaches position 8, the result is the same as in position 2. In the case of a double-tip probe, if the horizontal distance between the two tips is larger than the diameter of the object, the double-tip effect and principle are similar to the double-probe effect and principle, and an ghost image separated from its real image can be obtained. If the horizontal distance between the two tips (when tip c is higher than tip d) is smaller than the diameter of the object (Fig. 8b), then, when tip c almost reaches the far-end edge of the object (shown as red lines), tip d begins to interact with the surface of the object, and then the ghost image overlapping with its real image is obtained, because feedback is dependent on tip d and not tip c at this time.
Double-tip and double-probe effects have the same principle, and the difference is whether the distances are far or near and whether or not they are overlapping between ghost images and their real images caused by horizontal distances between two tips or two probes. Previously, many papers have discussed enlargement effects of AFM probes on samples using the deconvolution theory (Williams et al. 1996). When tips interact with samples, images larger than real objects will be obtained because of tip self-imaging and back-calculation, and the tip artifacts are relative to tip shape, decentralization, and cumulus of objects, and so forth. According to our results, double-tip effects can also induce distortion of real objects (Fig. 8b), that is, images obtained by double-tip probe are larger than their real objects. The principle of triple- or multiple-tip effects probably is same as that of double-tip effects, but more research is needed.
The examples of double-probe and double-, triple-, and multiple-tip effects here are easy to distinguish, but for some special samples or under some special conditions, these effects are concealed and easily ignored. For instance, when objects scanned have the same shape and a different height, the number of objects in AFM images will increase two-fold, and this is difficult to recognize. When a scan size is smaller than the horizontal distance between two tips of double-tip probe or two probes of double-probe cantilever, ghost images of objects outside the scan area will appear in the scan region and we cannot distinguish them. When objects are dense in the scan area, such as Langmuir-Blodgett (LB) films, cell membrane surface (as shown in Fig. 5), biomolecules in high concentration, and so forth, some ghost or real objects will overlap with other ghost or real objects. Therefore, in order to avoid these concealed artifacts, examination of probes is needed by scanning the samples whose artifacts are easy to distinguish before scanning those special samples.
During AFM scanning by double-probe cantilever, double-probe effects as follows will be observed. For larger objects at micron size, ghost images will be observed separated from the real objects whose height and width are equal to or larger than the ghost objects. The number of objects in those AFM images will increase two-fold in the scan region, but the horizontal distance between the ghost and real objects is at micron scale. Furthermore, the distribution of ghost and real objects is in a fixed direction and at a distance induced by the double-probe effect. For smaller objects of nanometer size, when the scan area is larger than the horizontal distance between the two probes of the double-probe cantilever, ghost images will also be observed separated from the real objects whose height and width are equal to or larger than those of the ghost objects, and the number of objects in the AFM images will increase two-fold in the scan region. When the scan area is smaller than the horizontal distance, ghost images whose real objects are outside the scan area will appear in the scan region, and at the same time, the number of objects in the images will also increase two-fold.
During AFM scanning by double-tip probe, double-tip effects as follows will be observed. When the horizontal distance between the two tips of the double-tip probe is larger than the diameter of objects observed, ghost images will also be observed separated from the real objects whose height and width are equal to or larger than those of the ghost objects, but the horizontal distance between ghost and real objects is at the nanometer scale. When the horizontal distance between the two tips is smaller than the diameter of the objects, the ghost images will overlap with their real objects, which makes the object’s shape in the AFM images to be deformed in an artificial preferential direction.
According to our results, double-probe, double-, triple-, and multiple-tip effects maybe share the same phenomena and principle, but they are also relative to the vertical distance and the relative position of the two probes on the double-probe cantilever, or the two tips on the double-tip probe, scan direction, scan size, shape and size of the objects scanned, among other factors; therefore, additional research is indicated.
This work was supported by grants from the National 973 Programs of China (No. 2001CB510101), National Natural Science Foundation of China (No. 62078014), the Key Program of National Natural Science Foundation of China (No. 30230350), the NIH of USA (HL64560), and the Doctoral Start-up Foundation of Jinan University.
The authors would like to thank Prof. Michael Jackson, Department of Food Science and Engineering at Jinan University, for his helpful corrections and suggestions for improving the manuscript.
PACS: 07.79.-v, 07.79.Lh
Yong Chen, Department of Chemistry, Jinan University, Guangzhou, Guangdong, P. R. China.
Jiye Cai, Department of Chemistry, Jinan University, Guangzhou, Guangdong, P. R. China.
Meili Liu, Department of Chemistry, Jinan University, Guangzhou, Guangdong, P. R. China.
Gucheng Zeng, Department of Chemistry, Jinan University, Guangzhou, Guangdong, P. R. China.
Qian Feng, Department of Chemistry, Jinan University, Guangzhou, Guangdong, P. R. China.
Zhengwei Chen, Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA.