Elsevier

Micron

Volume 43, Issue 12, December 2012, Pages 1390-1398
Micron

Scanning ion conductance microscopy for imaging biological samples in liquid: A comparative study with atomic force microscopy and scanning electron microscopy

https://doi.org/10.1016/j.micron.2012.01.012Get rights and content
Full text access

Abstract

The present study was designed to show the applicability of scanning ion conductance microscopy (SICM) for imaging different types of biological samples. For this purpose, we first applied SICM to image collagen fibrils and showed the usefulness of the approach-retract scanning (ARS)/hopping mode for such samples with steep slopes. Comparison of SICM images with those obtained by AFM revealed that the ARS/hopping SICM mode can probe the surface topography of collagen fibrils and chromosomes at nanoscale resolution under liquid conditions. In addition, we successfully imaged cultured HeLa cells, with 15 μm in height by ARS/hopping SICM mode. Because SICM can obtain non-contact (or force-free) images, delicate cellular projections were visualized on the surface of the fixed cell. SICM imaging of live HeLa cells further demonstrated its applicability to study the morphological dynamics associated with biological processes on the time scale of minutes under liquid conditions. We further applied SICM for imaging the luminal surface of the trachea and succeeded in visualizing the surface of both ciliated and non-ciliated cells. These SICM images were comparable with those obtained by scanning electron microscopy. Although the dynamic mode of AFM provides better resolution than the ARS/hopping mode of SICM in some samples, only the latter can obtain contact-free images of samples with steep slopes, rendering it an important tool for observing live cells as well as unfixed or fixed soft samples with complicated shapes. Taken together, we demonstrate that SICM imaging, especially using an ARS/hopping mode, is a useful technique with unique capabilities for imaging the three-dimensional topography of a range of biological samples under physiologically relevant aqueous conditions.

Highlights

► We show the applicability of scanning ion conductance microscopy (SICM) to biology. ► Collagen fibrils are imaged by SICM. ► Human chromosomes are imaged by SICM. ► Fixed and live cultures cells are imaged by SICM. ► The luminal surface of the trachea is observed by SICM.

Abbreviations

AC mode
alternating current mode
AFM
atomic force microscopy
ARS/hopping mode
approach-retract scanning, or hopping mode
SEM
scanning electron microscopy
SICM
scanning ion conductance microscopy
SPM
scanning probe microscopy

Keywords

Scanning ion conductance microscopy
Atomic force microscopy
Collagen fibrils
Chromosomes
HeLa cells

1. Introduction

Scanning probe microscopy (SPM) refers to a family of microscopy techniques that scan a probing tip over a sample surface to provide information about the local characteristics of solid samples. Among various SPM techniques, atomic force microscopy (AFM) has been widely utilized for biological studies since its invention (Binnig et al., 1986), because it can obtain three-dimensional information on sample topography at resolutions ranging from the micrometer to nanometer scales (Casuso et al., 2011, Allison et al., 2010, Ushiki and Kawabata, 2008, Hörber and Miles, 2007, Ushiki et al., 1996). In particular, since the AFM works in various environments (vacuum, air and liquid), many biologists have been interested in imaging soft, biological samples in physiologically relevant aqueous conditions (Braet et al., 2001, Le Grimellec et al., 1998, Schoenenberger and Hoh, 1994, Butt et al., 1990). We have also utilized AFM to characterize various biological samples such as cultured cells and chromosomes (Ushiki et al., 2008, Ushiki et al., 2000, Ushiki et al., 1999, Hoshi et al., 2006, Hoshi et al., 2004). Furthermore, recent advances in AFM have enabled high-speed imaging of soft samples (Picco et al., 2008, Picco et al., 2007, Ando et al., 2001), allowing the direct visualization of biomolecules and their movement in liquid conditions (Uchihashi and Ando, 2011, Kodera et al., 2010). However, it has remained difficult to image the nanoscale topography of larger meso- and microscale biological samples by AFM, because there may be artifacts due to the force between the tip and sample, as well as other factors such as tip geometry and lateral forces (Zhang et al., 2011, Braet et al., 2001).
To overcome these technical difficulties, some previous investigators have been interested in scanning ion conductance microscopy (SICM). SICM is another type of SPM technique that was first introduced by Hansma et al. (1989), and is based on a glass micropipette that serves as a sensitive probe. The signal is modulated by an ion current that flows between an electrode located within the pipette and a bath electrode for feedback control of the pipette-sample distance (Fig. 1). The distance is maintained at the radius of the pipette during scanning, which allows noncontact (or contact-free) imaging of the sample topography under liquid conditions. Biological application of SICM was first reported by Korchev et al. (1997), who observed live cultured cells under liquid conditions. Since then, SICM has been used to image the surfaces of different cultured cells such as neurons and cardiomyocytes (Gorelik et al., 2006, Gorelik et al., 2004, Happel et al., 2003, Pastré et al., 2001, Korchev et al., 2000). Recent advances in SICM have also provided a hopping mode (also called a backstep mode or approach-retract scanning mode), in which the pipette is moved vertically and repeatedly approaches and retracts from the sample surface (Novak et al., 2009, Happel and Dietzel, 2009). This mode was especially useful for non-contact imaging of the cell surface of cultured rat hippocampal neurons with a resolution better than 20 nm (Novak et al., 2009, Klenerman et al., 2011). These studies provided us with the idea that SICM can be widely used not only for imaging cultured cells but also for various other biological samples at resolutions comparable to that of AFM under liquid conditions.
  1. Download: Download high-res image (195KB)
  2. Download: Download full-size image

Fig. 1. Schematic drawing of the operating principles of scanning ion conductance microscopy (SICM).

In the present study, we introduce our results on SICM imaging of different biological samples and demonstrate the technique's excellent imaging capabilities by comparative analysis of SICM images with those obtained by AFM and/or scanning electron microscopy (SEM).

2. Instrumentation

2.1. Scanning ion-conductance microscopy (SICM)

SICM images were obtained with a commercial SPM (XE-Bio system, Park Systems Corp, Suwon, Korea). This system has an SPM system on the stage of an inverted optical microscope (Eclipse Ti, Nikon Corp., Tokyo, Japan). The SPM system has an xy flat scanner (100 μm × 100 μm) and a separate SICM or AFM head with a 25 μm z scanner. The SICM probe consists of a glass pipette filled with electrolyte with an Ag/AgCl electrode plugged into it. The glass pipette was fabricated from borosilicate capillaries (inner diameter 0.6 mm, outer diameter 1.0 mm, Narishige, Tokyo, Japan) using a CO2-laser-based micropipette puller (P-2000, Sutter Instruments, Novato, CA, USA); the inner and outer diameters of the pipette were about 100 nm and 200 nm, respectively. There are several operating modes for SICM. In the direct current (DC) mode, DC ion current is recorded during scanning, while, in the alternating current (AC) mode, the amplitude of the AC ion current modulated by oscillating the pipette probe is detected by using a lock-in amplifier (Pastré et al., 2001). In the hopping or backstep mode, ion current is recorded while the pipette is moved vertically and repeatedly approaches and retracts from the sample surface (Novak et al., 2009, Happel and Dietzel, 2009). Thus, this mode is also referred to as approach-retract scanning (ARS) mode. In the present study, images were primarily obtained in the ARS/hopping mode, although the AC mode was also used in some cases. In the ARS/hopping mode, the probe approached towards an insulating surface until the resistance exceeded a predefined threshold. A resistance increase of 2% with respect to the basal resistance was used in this experiment to stop the approach.

2.2. Atomic force microscopy (AFM)

In some cases, AFM images were obtained with the same SPM platform after the head was exchanged for AFM imaging. This microscope was operated in dynamic mode in liquid conditions. Reduction of the oscillation amplitude was used as the feedback parameter by a slope detection technique. The cantilevers used were DNP-S (triangle cantilevers with a nominal spring constant of 0.32 N/m and resonance frequency of about 56 kHz, Veeco, Metrology, Santa Barbara, CA, USA) or PNP-TR (triangle cantilever with a nominal spring constant of 0.32 N/m and resonance frequency of about 67 kHz, Nanotools, München, Germany).

2.3. Scanning electron microscopy (SEM)

After SICM imaging, some of the specimens were prepared for SEM imaging. Briefly, the samples were made electron-conductive by staining with tannic acid and osmium based on the method of Murakami (1973). They were then dehydrated with an ascending ethanol series, transferred to isoamyl acetate, and dried in a critical point dryer using CO2 (HCP-2, Hitachi High-technologies, Tokyo, Japan). This was followed by mounting on aluminum stubs, coating with platinum–palladium in an ion coater, and observation in a field emission SEM (S-4300SE/N, Hitachi High-technologies,).

3. Results and discussion

3.1. Collagen fibrils

As AFM imaging of collagen fibrils has been reported by several investigators (Yamamoto et al., 2002, Yamamoto et al., 2000, Yamamoto et al., 1997, Baselt et al., 1993), we attempted to examine the applicability of SICM for imaging collagen fibrils. For this purpose, collagen fibrils were obtained from the tail tendon of adult Wistar rats (Japan SLC, Inc., Hamamatsu, Japan) and stored in physiological saline with 1–10% tymol (2-isopropyl-5-methylphenol) at 4 °C for a minimum of 1 day. A small piece of the tendon was then stretched on the glass surface and air dried overnight, followed by immersion of the sample in physiological saline or phosphate buffer saline (PBS) before AFM and/or SICM imaging.
At first, we compared the image quality obtained by SICM imaging in AC mode versus ARS/hopping mode. Fig. 2 is an example of the comparison images of the two modes, in which AC mode imaging preceded the ARS/hopping mode imaging for this sample. In the AC mode, no collagen fibril was imaged in this portion of the sample, whereas there was a collagen fibril, about 530 nm in height, in the ARS/hopping mode. These two images indicate that the collagen fibrils likely moved away during scanning in AC mode as a result of fibril deformation by the probing pipette, but were well-visualized in the ARS/hopping mode. On the other hand, spatial resolution is greater in the AC mode than in the ARS/hopping mode in our experiment, because structures in the range of 50–60 nm were clearly visualized only in AC mode.
  1. Download: Download high-res image (830KB)
  2. Download: Download full-size image

Fig. 2. Comparison of SICM imaging in AC and ARS/hopping modes. Collagen fibrils were attached on the glass slide and observed with SICM in AC mode (above left) and then in ARS/hopping mode (above right). Arrows indicate corresponding structures of the AC and ARS/hopping mode images. Line profile analysis is based on the broken line indicated in the ARS/hopping mode image.

Next, we compared the dynamic mode of AFM with the ARS/hopping mode of SICM. Fig. 3 is an example of AFM and SICM images on the same collagen fibrils. In these images, the height of the collagen fibrils was 1.4 times greater in the SICM image than in the AFM image, suggesting that SICM is better suited for non-contact imaging than AFM; the height of the collagen fibrils are underestimated by AFM (here by about 30%, but this difference may depend on the AFM imaging force) compared with those obtained by SICM. On the other hand, the width of collagen fibrils was 0.7 times smaller in the SICM image than that of the AFM image. Because collagen fibrils are cylindrical, the width is likely influenced by the shape of the probing chip. Our finding suggests that the shape of the collagen fibrils is well-delineated in the SICM image even though the inner radius of the pipette (about 100 nm) is much larger than the radius of the AFM probing tip. In the AFM image, distortions are inherent due to the effect of a convolution between the pyramidal tip and sample (Meyer et al., 2004, Baselt et al., 1993).
  1. Download: Download high-res image (839KB)
  2. Download: Download full-size image

Fig. 3. Comparison of SICM and AFM imaging capabilities under liquid conditions. Collagen fibrils on a glass slide were first imaged by AFM (upper left image), and then by SICM (upper right image). The section profiles 1 and 2 are based on the lines labeled “profile 1” and “profile 2” in the upper two images.

To examine the applicability of the ARS/hopping mode for imaging biological samples with greater height gaps, dense collagen fibril networks were examined by SICM. Fig. 4 is an example of an SICM image of collagen fibril networks. The width of the individual fibrils varied from 50 to 470 nm. The section profile also showed that the height difference between the glass surface and fibril (328 nm thickness), as indicated in Fig. 4, was 1.549 μm, suggesting that some fibrils were suspended over the glass substrate.
  1. Download: Download high-res image (490KB)
  2. Download: Download full-size image

Fig. 4. Dense collagen fibril networks imaged by the ARS/hopping SICM mode. Line profile analysis is based on the red line indicated in the SICM image. The width of the collagen fibril, as indicated by the two white arrowheads, is 328 nm. The height gap between the top of the collagen fibril and the substrate is 1.549 μm.

Because collagen fibrils with a width of 50 nm can be observed with SICM imaging, the horizontal resolution is estimated to be in the range of 50 nm. However, grooves produced by D periodicity (60–67 nm) in the collagen fibrils were faint in SICM images, probably because the depth of the groove (2 nm or less) in liquid environment is beyond the vertical resolution of the pipette used for this SICM experiment.

3.2. Chromosomes

The higher order structure of chromosomes has been widely studied by transmission and scanning electron microscopy, and recently by AFM (Daban, 2011, Ushiki and Hoshi, 2008, Ushiki et al., 2002, de Grooth and Putman, 1992, Summer, 1991, Earnshaw and Laemmli, 1983). The advantage of AFM is its ability to image at high resolution in a liquid environment, which is also expected by SICM. Thus, we applied SICM to imaging of chromosomes. Chromosomes of human lymphocytes were prepared according to the standard air-drying method for light microscopy (Ushiki and Hoshi, 2008). Briefly, human lymphocytes after cultivation were arrested in metaphase by adding colcemid to the culture medium at a final concentration of 0.05 μg/ml for 1 h, exposed to 75 mM KCl for 30 min at room temperature and fixed with Carnoy's solution (methanol: acetic acid = 3:1, v/v). Chromosome spreads were then formed by dropping the cell suspension onto glass slides, briefly dried in air in order to fix them onto the glass slides. They were immersed in PBS and observed by SICM and AFM.
Fig. 5 shows the SICM image of human chromosomes in the liquid environment. The image quality appears qualitatively similar to that obtained previously by the dynamic mode of AFM (Ushiki and Hoshi, 2008, Hoshi et al., 2006, Hoshi et al., 2004). In order to compare the quality of AFM and SICM images more precisely, chromosomes were imaged first by the dynamic mode of AFM, then by the ARS/hopping mode of SICM and again by AFM. In Fig. 6, the height profiles obtained by sequential AFM, SICM and AFM measurements were 473, 746 and 375 nm, respectively. These findings indicate that the height of the chromosome determined by SICM (c.a. 750 nm) is much greater than that of AFM (roughly 400–500 nm depending on the AFM imaging force). At higher magnification, globular or fibrous structures about 50 nm were observed by SICM (micrograph not presented), suggesting that SICM is applicable for studying higher order structures of very soft chromosomes in liquid environments.
  1. Download: Download high-res image (221KB)
  2. Download: Download full-size image

Fig. 5. Human chromosomes imaged by the ARS/hopping SICM mode.

  1. Download: Download high-res image (637KB)
  2. Download: Download full-size image

Fig. 6. Comparison of AFM and SICM images on the same human chromosomes. Chromosomes immersed in an aqueous solution were first imaged by AFM, then by SICM, and finally by AFM again. Line profile analysis is based on the red line indicated in the corresponding image.

3.3. Cultured cells

SICM imaging of cultured cells was first reported by Korchev et al. (1997). Since then, multiple studies have reported SICM imaging of cultured cells (Gorelik et al., 2004, Rheinlaender et al., 2011). These previous findings indicate that SICM is very useful for imaging the surface topography of living and fixed cultured cells at nanoscale resolution under liquid. To examine the image quality of SICM for cultured cells, fixed and live HeLa cells were imaged by ARS/hopping SICM in this study.
At first, HeLa cells were grown on cover slips in culture dishes (50 mm in diameter) for 24–48 h in a CO2 incubator at 38 °C, fixed with 1% glutaraldehyde in 0.1 M phosphate buffer (PB) at pH 7.4 for overnight at 4 °C, and observed by SICM. The height of HeLa cells was estimated to be about 15 μm in the SICM image of Fig. 7. Microprojections on the cell surface were visualized on the cell surface, suggesting that the ARS/hopping mode of SICM can obtain high-resolution topographic image of cultured cells with minimal deformation in liquid conditions. By contrast, the steep slope of the cell morphology renders it difficult to observe such soft structures by AFM. Fig. 8 is a higher magnification image of the HeLa cell surface, in which microprojections with various irregular shapes are observed on the cell surface; the height and thickness of these projections are roughly 0.5–1 μm and 100–150 nm, respectively.
  1. Download: Download high-res image (659KB)
  2. Download: Download full-size image

Fig. 7. HeLa cells imaged by the ARS/hopping SICM mode. The sample was fixed with 1% glutaraldehyde before SICM imaging. Line profile analysis is based on the line indicated in the upper right image.

  1. Download: Download high-res image (350KB)
  2. Download: Download full-size image

Fig. 8. Projections on the surface of HeLa cell. Lamellar projections with a range of topographies are clearly observed on the surface of the HeLa cell. Image is a magnified view of Fig. 7.

We further tried to obtain SICM images of live cells. Fig. 9 shows sequential time-lapsed SICM images of live HeLa cells. The data acquisition time per image was about 10 min in Fig. 9, but we found that the time can be reduced to about 6 min when 128 by 128 pixel images were appropriately collected with our instrument. The time resolution is thus comparable with AFM imaging of living cells (Ushiki et al., 1996, Ushiki et al., 2000), even though high-speed AFM imaging is also introduced for specimens such as chromosomes (Picco et al., 2008). In addition, the ARS/hopping mode has the advantage of minimizing deformation of cellular structures by the probing tip compared with AFM. Thus, the ARS/hopping mode of SICM is expected to be widely useful for live cell imaging, especially in relation to the movement of cellular processes on a time scale of minutes under liquid conditions.
  1. Download: Download high-res image (863KB)
  2. Download: Download full-size image

Fig. 9. Repeated observation of live HeLa cells by the ARS/hopping SICM mode. The scan time of each image (128 by 128 pixels) was 10 min. Note the movement of cellular processes indicated by arrows and arrowheads.

3.4. Bulky tissues and tissue blocks

Because the ARS/hopping mode of SICM can obtain images of samples with surface irregularities, we examined the applicability of SICM on a variety of samples, including the imaging of bulky tissue blocks with irregular surfaces. For this purpose, trachea was obtained from Wistar rats, which were transcardially fixed with 2% glutaraldehyde in PB. A small piece of the trachea was then mounted on the glass slide with an adhesive (Aron Alpha cyanoacrylate instant adhesives, Toagosei Co. Ltd., Tokyo, Japan) and observed by the ARS/hopping mode of SICM.
Fig. 10 shows an SICM image of the luminal surface of the trachea. This image shows that both the ciliated and non-ciliated cells can be imaged by SICM, even though the height difference is at least 8 μm in this sample. The short cellular projections are clearly discernible on the surface of non-ciliated cells. The thickness of cilia is about 100–150 nm in the SICM image. To compare the SICM and SEM images, the samples were observed by SICM followed by observation of the same portion by SEM after sample preparation. Fig. 11 presents SICM and SEM images of the same portion of the trachea. These images indicate that SICM images are comparable with SEM images, although the thickness of cilia is somewhat smaller in SICM images (c.a. 100 nm) than in SEM images (c.a. 200 nm). The difference in thickness of cilia between the SICM and AFM images is not clear but it may be partially explained by the drift of cilia during scanning. Further studies are needed to clarify this point.
  1. Download: Download high-res image (659KB)
  2. Download: Download full-size image

Fig. 10. Luminal surface of rat trachea imaged by the ARS/hopping SICM mode. Line profile analysis is based on the line indicated in the upper right image.

  1. Download: Download high-res image (2MB)
  2. Download: Download full-size image

Fig. 11. Comparison of SICM and SEM images of the luminal surface of rat trachea. After SICM imaging by the ARS/hopping mode, the samples were made electron-conductive, dehydrated, subjected to critical point drying, coated with metal, and observed by SEM.

4. Conclusion

The present study shows our results to test the applicability of SICM for imaging different types of biological samples. SICM imaging of live and fixed cultured cells has been repeatedly reported, since Korchev and his colleagues first applied SICM for this purpose (Rheinlaender et al., 2011, Novak et al., 2009, Gorelik et al., 2004, Korchev et al., 2000, Korchev et al., 1997). However, there has been no report on the application of SICM to image other types of biological samples. In this paper, we show for the first time that SICM can be used to image collagen fibrils and chromosomes as well as cultured cells. We also introduce the applicability of SICM for studying the surface topography of thick tissue samples. Although Novak et al. (2009) reported SICM imaging of auditory hair cells in the cultured organ of Corti explants, this is the first report that shows the applicability of SICM for imaging tissue blocks with irregular surfaces.
Comparison of SICM with AFM images also revealed that SICM can be utilized for obtaining contact-free images of the surface topography of biological samples. The ARS/hopping mode of SICM is especially powerful for imaging samples with steep slopes (Novak et al., 2009). We showed that the lateral resolution of the ARS/hopping SICM mode is about 50 nm, which is comparable to the finding by previous investigators (Novak et al., 2009, Happel and Dietzel, 2009). However, the actual resolution of SICM imaging is still a matter of debate (Rheinlaender and Schaffer, 2009). Theoretical studies of SICM will be required to consider the resolution in relation to the pipette radius and other parameters, although there are a few papers already in this field (Edwards et al., 2009).
While it may be true that image resolution is better in the dynamic mode of AFM than in the ARS/hopping mode of SICM in some samples, only the latter can obtain contact-free images. As such, SICM is an excellent tool for observing live cells and unfixed soft samples. Taken together, SICM imaging, especially by ARS/hopping mode, is expected to be useful for obtaining images on the three-dimensional topography of biological samples.

Acknowledgements

This study was supported in part by Grant-in-Aid for Scientific Research (B) (no. 21390051 for TU) by Japan Society for the Promotion of Science (JSPS). The part of development of instrument was also sponsored by the Industrial Source Technology Development Programs (ISTDP10033633) in Ministry of Knowledge Economy in Korea. HeLa cells were kindly provided by Dr. Hirota, Department of Experimental Pathology, Cancer Institute of the Japanese Foundation for Cancer Research, Tokyo, Japan The authors are also grateful to Dr. Koga, Division of Microscopic Anatomy, Niigata University Graduate School of Medical and Dental Sciences, Niigata Japan for his technical assistance.

References

View Abstract