Ultramicroscopy 42-44 (1992) 1101-1106 North-Holland
Imaging isolated strands of DNA molecules by atomic force microscopy T. Thundat
1,
D.P. Allison, R.J. Warmack 1 and T.L. Ferrell 1
Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA Received 12 August 1991
W e have employed an atomic force microscope (AFM) to image in air isolated strands of pBS + plasmid D N A adsorbed onto freshly cleaved mica. At a D N A concentration below 0 . 3 / ~ g / m l isolated strands of the plasmid D N A are usually seen, while for concentrations higher than 3/.Lg/ml a uniform coverage of interconnected D N A strands was observed. We found that the contrast and the width of D N A were dependent upon humidity. W h e n the relative humidity exceeds 60%, negative contrast images with strand widths 20 times the width of D N A are found, while positive contrast images with 7 to 10 times the width of D N A are found when the humidity is below 30%. By placing the A F M in an environment where the humidity could be controlled, we were able to switch between positive and negative contrasts.
1. Introduction The demonstrated ability of scanning probe microscopies to obtain high-resolution images of biopolymers in air and liquids has established them as potential tools for studying DNA structure and sequences. Since the pioneering work of Binnig and Rohrer [1], many groups have published STM [2-7] and AFM [8,9] images of uncoated DNA obtained with unprecedented resolution. It has been suggested that the sequencing of DNA using scanning probe techniques may be at least two orders of magnitude faster than techniques based on conventional gel electrophoresis [10]. It has also been proposed that the AFM may prove to be more useful than the STM in imaging biopolymers because most biopolymers are insulators. Also, DNA apparently binds more easily to insulators (glass, for example) than to conducting substrates commonly used for STM such as graphite (HOPG) and gold. This is an advantage 1 AlSO at: University of Tennessee, Knoxville, TN 37994, USA.
since both the STM and AFM exert considerable contact forces [11-13] on adsorbed macromolecules and as a result the tip tends to sweep DNA or any weakly bonded materials away [14,15]. Therefore, a strong bonding of molecules to the substrate is required for successful imaging. For this reason many techniques such as electrochemical deposition [16-20], adhesion to a chemically modified surface [7,8,21,22], the spreading technique [23], pulse deposition [24] and electrostatic spraying [25] of biopolymers have been explored. Although both the STM and AFM have been used successfully in obtaining high-resolution images of DNA, reproducibility in obtaining images appears to be a problem mainly due to sample preparation. In this paper, we report AFM imaging of isolated strands of pBS ÷ plasmid DNA molecules adsorbed on mica. We are able to routinely obtain very stable images. However, the widths of the DNA molecules measured from the topographs are larger than expected by a factor of 7. We also report on the effect of humidity on the contrast and the width of DNA observed with the AFM.
0304-3991/92/$05.00 © 1992 - Elsevier Science Publishers B.V. All rights reserved
1102
T. Thundat et al. / Imaging isolated strands of DNA by AFM
2. Experiment
2.1. Preparation of DNA Plasmid D N A (pBS + from Stratagene, 3200 bp) was prepared from cultures of E. coli strain MV1190 grown in 2X Y T media. Cells were ruptured by alkaline lysis [26] and the supercoiled plasmid D N A purified by cesium chlorideethidium bromide equilibrium density gradient centrifugation, ethanol precipitated two times, and resuspended in 0.01M Tris-HC1 containing 0.001M E D T A p H 7.5 (TE) at a concentration 500 /xg/ml. Relaxed circular plasmid D N A was obtained by single-stranded nicking of supercoiled molecules via X-ray radiation (70 000 rad). Linear plasmid D N A was prepared by using restriction endonuclease Sinai (New England Biolabs) at the manufacturer's recommended conditions: 100/xm of plasmid was incubated with 100 units of Sinai in a 200 /xl reaction mixture overnight at 25°C. The reaction mixture was extracted with phenol, precipitated with ethanol and resuspended in TE. The purity of D N A was determined by agrose gel electrophoresis. The concentration of D N A used in this study varied from 0.35 to 10 / x g / m l in 0.01M ammonium acetate ( N H a O A c ) which had been previously adjusted to a p H of 7.2.
2.2. AFM imaging D N A was deposited onto freshly cleaved muscovite mica (New York Mica Co.) by either placing a 10/xl drop on mica and letting it air dry or by spraying D N A solution onto mica substrate using a nebulizer. All of the images shown in this paper were obtained from air-dried samples of DNA. In some cases the substrates were heated to 60°C by a sun lamp prior to A F M experiments to volatilize the ammonium acetate buffer. However, heating of the sample did not improve the images appreciably. A Nanoscope II (Digital Instruments, Santa Barbara, CA) was used in this study. The cantilevers used were 200 t~m long with a spring constant of 0.12 N / m (Nanoprobes) and images were recorded in a constant-force mode. For
experiments under water, D N A was first adsorbed on mica, as in air experiments, and then transferred to the liquid cell. Prior to A F M imaging, nanopure water was injected into the liquid cell. Relative humidity was controlled when operating the A F M in air by enclosing the A F M in a bell jar and purging the system with dry nitrogen.
3. Results and discussions
3.1. A F M images When a droplet of D N A solution was allowed to air-dry onto a mica surface, images of D N A could be obtained in all of the area scanned. If the D N A was deposited onto the surface by a nebulizer spray, D N A images were obtained at a frequency that depended upon the surface coverage of the nebulized droplets. The D N A molecules observed with the A F M were stable for 5 to 10 scans but repeated scanning tended to break the strands. When a control solution of buffer was deposited onto the mica surface, either by droplet or by nebulizer, no structures were found on the surface. The A F M scans carried out under buffer solution on DNA-covered mica show a clean surface without any indication of D N A adsorption. When the concentration of D N A applied to the mica was below 0.3 ~ g / m l , isolated molecules (fig. 1) are most frequently seen. The lengths of individual D N A molecules measured from the topographs were in good agreement (within 10%, averaged over 20 isolated strands) with the expected molecular length (slightly larger than 1 ~ m ) of the 3200 bp plasmid pBS +. However, the molecular width ranged from 7 to 10 times (7 was the lower limit) larger than the 2 nm width of B-DNA. Although the preparation of circular plasmid D N A used in these experiments was approximately 80% open circles, no completely open circles (fig. 1) were found in our images. Higher-resolution images (see fig. 2) of a representative circular molecule show apparent supercoiling of the molecules. We also found that the images of plasmid D N A that had been linearized showed a higher degree of supercoiling, which we
T. Thundat et aL / Imaging isolated strands of DNA by AFM
attribute to stresses introduced as a result of drying. We propose to test this hypothesis on each circular and linearized plasmid D N A by freeze-drying samples onto mica surfaces. When the concentration of D N A was increased to between 3 t z g / m l and the maximum concentration of 1 0 / z g / m l that we used in these experiments, instead of individual molecules, the D N A formed a network of interconnecting molecules that extended uniformly over the entire area of mica wetted by the DNA. Fig. 3 shows the beginning of this interconnecting network observed for a concentration 0.3 /xg/ml while fig. 4b shows a more interconnected network at a concentration of 3 / ~ g / m l . The density of D N A strands observed on the surface was directly proportional to the amount of D N A in the droplet applied. 3.2. Contrast o f images
During these experiments the contrast of the D N A images could be either positive, where D N A strands appear to be higher than the substrate, or negative, where the D N A strands appear to be
Fig. 1. 2 tzmx2 /,m AFM image showing isolated strands of pBS + DNA (concentration 0.3/*g/ml) on mica. The tip force was of 3 nN (repulsive mode). Details are enhanced using an artificial light source illumination.
1103
Fig. 2. 600 nm×600 nm AFM topograph of few strands of circular DNA showingsigns of supercoiling taken with a force of 2 nN. Black (low) to white (high) corresponds to 30 nm for the entire image. The height of DNA above the substrate varies from 1.5 to 3.1 nm.
lower than the substrate. The A F M topographs obtained with freshly prepared D N A / m i c a samples often showed a negative contrast. In addition, the D N A strands were also found to move around due to scanning. However, drying the sample for a long length of time or heating the sample often resulted in a stable positive contrast. Fig. 4a shows an A F M image of linear D N A on mica with a negative contrast taken in 64% humidity. However, a sample that gave a positive contrast one day might give a negative contrast the next day. We have investigated this phenomenon in detail and found a direct relationship between contrast and humidity. D N A images obtained on humid days (humidity over 60%) always had a negative contrast while those obtained on dry days (humidity below 30%) produced positive contrast images. On humid days it was possible to isolate the A F M in a bell jar and change the contrast from negative to positive by passing dry nitrogen into the bell jar housing the AFM. It was also possible to change the positive contrast back to a negative contrast by switching off the nitrogen and letting humid air back into
1104
T. Thundat et al. / Imaging isolated strands of DNA by AFM
Fig. 3. 2 /zmx2 /zm area of DNA on mica initial stages of formation interconnecting networks (DNA concentration 0.3 p,g/ml). Black (low) to white (high) corresponds to 46 nm. The height of individual strands varies from 1.5 to 3.1 nm. Network of DNA strands becomes more clear as the concentration of DNA is increased (see fig. 4b).
the chamber. W e have r e p r o d u c e d in this result on 20 different samples. Thus, the effect is completely reversible. Fig. 4b is a positive contrast image of the same sample as in fig. 4a t a k e n
u n d e r dry n i t r o g e n a m b i e n t . T h e D N A strands a p p e a r to be b r o a d e n e d due to coiling. T h e force curves o b t a i n e d in h u m i d air (64%) on a D N A - c o v e r e d mica surface show a rather large attractive force o n the tip before it breaks away from the surface. This b r e a k - a w a y force can be as large as 100 n N d e p e n d i n g on the humidity. However, as dry n i t r o g e n is i n t r o d u c e d into the c h a m b e r , the force can be decreased to 10 nN, or occasionally, smaller. This h u m i d i t y - d e p e n d e n t force is p r e s u m e d to be the capillary force from the liquid bridge formed b e t w e e n the a d s o r b e d water on the substrate a n d o n the sample [27-29]. T a k i n g the accepted value of surface t e n s i o n of water as 72 x 10 -3 N / m [29], the capillary forces can be as high as 100 n N d e p e n d i n g on the area of the water bridge formed b e t w e e n the tip a n d the substrate. T h e negative contrast can be att r i b u t e d to cantilever buckling d u e to an increased capillary force a r o u n d the D N A molecule [301. 3.3. Width o f D N A images T h e a p p a r e n t width of D N A o b t a i n e d from positive contrast A F M t o p o g r a p h s was 7 to 10
a
Fig. 4. Humidity-dependent positive and negative contrast of AFM images of linear DNA (concentration 3 ~zg/ml). Black (low) to white (high) corresponds to 5.2 nm. (a) 2 jzm x 2 izm on mica taken in 60% humidity. The DNA strands show negative contrast. (b) Same area of the sample in dry nitrogen ambient. The DNA strands show positive contrast.
T. Thundat et al. / Imaging isolated strands of DNA by AFM times larger while widths from the negative contrast i m a g e s w e r e up to 20 times g r e a t e r t h a n the 2 n m k n o w n width of D N A T h e s e differences can be explained by the capillary forces exerted on the tip u n d e r high h u m i d i t y c o n d i t i o n s that produce negative images. T h e widths of positive contrast D N A images were f o u n d to c h a n g e from tip to tip. O n e p r o b a b l e r e a s o n for this c o n c e r n s the c o n v o l u t i o n of the tip geometry with the D N A molecule. F r o m a simple geometrical a r g u m e n t , the d i a m e t e r of a cylinder of radius r w h e n imaged with a tip of radius R will a p p e a r as 4(rR) 1/2. A n observed D N A width of 20 n m will t h e n lead to a tip radius of 25 n m which is in a g r e e m e n t with typical tip sizes as m e a s u r e d by SEM. In some cases it is possible to have a s h a r p e r tip, either by c o n t a m i n a t i o n or by accid e n t which can p r o d u c e D N A images with a smaller width. T h e smallest width of D N A we have m e a s u r e d with a n A F M t o p o g r a p h was 7 nm, b u t even with this exceptional tip it was not possible to resolve the 3.4 n m helical r e p e a t of DNA.
4. Conclusion W e have imaged isolated strands of D N A reproducibly by a d s o r b i n g t h e m on a freshly cleaved mica. R e p e a t e d l y s c a n n i n g the same area or h i g h - m a g n i f i c a t i o n scans t e n d to b r e a k the D N A strands. T h e observed width a n d contrast of the D N A s t r a n d s was f o u n d to d e p e n d o n humidity. T h e images of D N A o b t a i n e d at 60% or higher h u m i d i t y show negative c o n t r a s t while those obt a i n e d in dry n i t r o g e n or low h u m i d i t y show positive contrast. M o r e e x p e r i m e n t s are presently underway to m a p out the exact d e p e n d e n c e of the contrast of D N A o n humidity. T h e width of D N A observed with the A F M was also f o u n d to d e p e n d o n the tip conditions. It may be possible to improve A F M r e s o l u t i o n o n D N A by using s h a r p e r tips u n d e r a dry a t m o s p h e r e .
Acknowledgements W e are grateful to K.B. J a c o b s o n a n d L.A. Bottomley helpful discussions. This research was
1105
s p o n s o r e d by the D i r e c t o r ' s R e s e a r c h a n d Develo p m e n t F u n d at O R N L a n d the Office of H e a l t h a n d E n v i r o n m e n t a l Research, U.S. D e p a r t m e n t of E n e r g y u n d e r contract D E - A C 0 5 - 8 4 O R 2 1 4 0 0 with M a r t i n M a r i e t t a E n e r g y Systems, Inc.
References [1] G. Binnig and H. Rohrer, in: Trends in Physics, Eds. J. Janata and J. Pantoflicek (European Physical Society, The Hague, pp. 38-46. [2] T.P. Beebe, T.E. Wilson, D.F. Ogletree, J.E. Katz, R. Balhorn, M.B. Salmeron and W.J. Siekhaus, Science 243 (1989) 370. [3] P.G. Arscott, G. Lee, V.A. Bloomfield and D.F. Evans, Nature 319 (1989) 484. [4] S.M. Lindsay, T. Thundat, L.A. Nagahra, U. Knipping and R. Rill, Science 244 (1989) 1063. [5] D.D. Dunlap and C. Bustamante, Nature 342 (1989) 204. [6] A. Cricenti, S. Selci, A.C. Felici, R. Generosi, E. Gori, W. Djaczenko and G. Chiarotti, Science 245 (1989) 1226. [7] R.J. Driscoll, M.G. Youngquist and J.D. Baldeschwieler, Nature 346 (1990) 294. [8] A.L. Weisenhorn, H.E. Gaub, H.G. Hansma, R.L. Sinsheimer, G.L. Kelderman and P.K. Hansma, Scanning Microsc. 4 (1990) 511. [9] A.L. Weisenhorn, M. Egger, F. Ohnesorge, S.A.C. Gould, S.P. Heyn, H.G. Hansma, R.L. Sinsheimer, H.E. Gaub and P.K. Hansma, Langmuir 7 (1991) 8. [10] H.G. Hansma, A.L. Weisenhorn, S.A.C. Gould, R.L. Sinsheimer, H.E. Gaub, G.D. Stucky, C.M. Zaremba and P.K. Hansma, J. Vac. Sci. Technol. B 9 (1991) 1282. [11] J.H. Coombs and J.W. Pethica, IBM J. Res. Dev. 30 (1986) 455. [12] H.J. Mamin, E. Ganz, D.W. Abraham, R.E. Thompson and J. Clarke, Phys. Rev. B 34 (1986) 9015. [13] U. Durig, J.K. Gimzewski and D.W. Pohl, Phys. Rev. Lett. 57 (1986) 2043. [14] M.B. Salmeron, T.P. Beebe, J. Odriozola, T.E. Wilson, D.F. Ogletree and W.J. Siekhaus, J. Vac. Sci. Technol. A 8 (1990) 635. [15] M.B. Salmeron, D.F. Ogletree, C. Ocal, H.-C. Wang, G. Neubuer, W. Kolbe and G. Meyers, J. Vac. Sci. Technol. B 9 (1991) 1347. [16] S.M. Lindsay and B. Barris, J. Vac. Sci. Technol. A 6 (1988) 544. [17] S.M. Lindsay, T. Thundat and L.A. Nagahara, J. Microscopy 152 (1988) 213. [18] T. Thundat, L.A. Nagahara, P. Oden and S.M. Lindsay, J. Vac. Sci. Technol. A 8 (1990) 645. [19] R.W. Keller, D. Bear and C. Bustamante, J. Vac. Sci. Technol. B 9 (1991) 1291. [20] J.A. Derose, S.M. Lindsay, L.A. Nagahara, P.I. Oden and T. Thundat, J. Vac. Sci. Technol. B 9 (1991) 1166.
1106
7~ Thundat et al. / Imaging isolated strands of DNA by AFM
[21] W.M. Heckl, K.M.R. Kallury, M. Thompson, C. Gerber, H.J.K. Horber and G. Binnig, Langmuir 5 (1989) 1433. [22] Y.L. Lyubchenko, S.M. Lindsay, J.A. DeRose and T. Thundat, J. Vac. Sci. Technol. B 9 (1991) 1288. [23] A. Cricenti, S. Selci, G. Chiarotti and F. Amaldi, J. Vac. Sci. Technol. B 9 (1991) 1285. [24] M.J. Allen, R.J. Tench, J.A. Mazrimas, M. Balooch, W.J. Siekhaus and R. Balhorn, J. Vac. Sci. Technol. B 9 (1991) 1272. [25] T. Thundat, R.J. Warmack, D.P. Allison and T.L. Ferrell, Ultramicroscopy 42-44 (1992) 1082.
[26] J. Sambrook, E.F. Fritsch and T. Maniat, Molecular Cloning, 2nd ed. (Cold Spring Harber Press, 1989), [27] J.T. Woodward, J.A.N. Zasadzinski and P.K. Hansma, J. Vac. Sci. Technol. B 9 (1991) 1231. [28] J.N. Israelachvilli, Intermolecular and Surface Forces (Academic Press, New York, 1985). [29] A.W. Adamson, Physical Chemistry of Surfaces, 5th ed. (Wiley Interscience, New York, 1990). [30] T. Thundat, R.J. Warmack, D.P. Allison, T.L. Ferrell and L.A Bottomley, J. Vac. Sci. Technol., in press.
Imaging isolated strands of DNA molecules by atomic force microscopy T. Thundat
1,
D.P. Allison, R.J. Warmack 1 and T.L. Ferrell 1
Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA Received 12 August 1991
W e have employed an atomic force microscope (AFM) to image in air isolated strands of pBS + plasmid D N A adsorbed onto freshly cleaved mica. At a D N A concentration below 0 . 3 / ~ g / m l isolated strands of the plasmid D N A are usually seen, while for concentrations higher than 3/.Lg/ml a uniform coverage of interconnected D N A strands was observed. We found that the contrast and the width of D N A were dependent upon humidity. W h e n the relative humidity exceeds 60%, negative contrast images with strand widths 20 times the width of D N A are found, while positive contrast images with 7 to 10 times the width of D N A are found when the humidity is below 30%. By placing the A F M in an environment where the humidity could be controlled, we were able to switch between positive and negative contrasts.
1. Introduction The demonstrated ability of scanning probe microscopies to obtain high-resolution images of biopolymers in air and liquids has established them as potential tools for studying DNA structure and sequences. Since the pioneering work of Binnig and Rohrer [1], many groups have published STM [2-7] and AFM [8,9] images of uncoated DNA obtained with unprecedented resolution. It has been suggested that the sequencing of DNA using scanning probe techniques may be at least two orders of magnitude faster than techniques based on conventional gel electrophoresis [10]. It has also been proposed that the AFM may prove to be more useful than the STM in imaging biopolymers because most biopolymers are insulators. Also, DNA apparently binds more easily to insulators (glass, for example) than to conducting substrates commonly used for STM such as graphite (HOPG) and gold. This is an advantage 1 AlSO at: University of Tennessee, Knoxville, TN 37994, USA.
since both the STM and AFM exert considerable contact forces [11-13] on adsorbed macromolecules and as a result the tip tends to sweep DNA or any weakly bonded materials away [14,15]. Therefore, a strong bonding of molecules to the substrate is required for successful imaging. For this reason many techniques such as electrochemical deposition [16-20], adhesion to a chemically modified surface [7,8,21,22], the spreading technique [23], pulse deposition [24] and electrostatic spraying [25] of biopolymers have been explored. Although both the STM and AFM have been used successfully in obtaining high-resolution images of DNA, reproducibility in obtaining images appears to be a problem mainly due to sample preparation. In this paper, we report AFM imaging of isolated strands of pBS ÷ plasmid DNA molecules adsorbed on mica. We are able to routinely obtain very stable images. However, the widths of the DNA molecules measured from the topographs are larger than expected by a factor of 7. We also report on the effect of humidity on the contrast and the width of DNA observed with the AFM.
0304-3991/92/$05.00 © 1992 - Elsevier Science Publishers B.V. All rights reserved
1102
T. Thundat et al. / Imaging isolated strands of DNA by AFM
2. Experiment
2.1. Preparation of DNA Plasmid D N A (pBS + from Stratagene, 3200 bp) was prepared from cultures of E. coli strain MV1190 grown in 2X Y T media. Cells were ruptured by alkaline lysis [26] and the supercoiled plasmid D N A purified by cesium chlorideethidium bromide equilibrium density gradient centrifugation, ethanol precipitated two times, and resuspended in 0.01M Tris-HC1 containing 0.001M E D T A p H 7.5 (TE) at a concentration 500 /xg/ml. Relaxed circular plasmid D N A was obtained by single-stranded nicking of supercoiled molecules via X-ray radiation (70 000 rad). Linear plasmid D N A was prepared by using restriction endonuclease Sinai (New England Biolabs) at the manufacturer's recommended conditions: 100/xm of plasmid was incubated with 100 units of Sinai in a 200 /xl reaction mixture overnight at 25°C. The reaction mixture was extracted with phenol, precipitated with ethanol and resuspended in TE. The purity of D N A was determined by agrose gel electrophoresis. The concentration of D N A used in this study varied from 0.35 to 10 / x g / m l in 0.01M ammonium acetate ( N H a O A c ) which had been previously adjusted to a p H of 7.2.
2.2. AFM imaging D N A was deposited onto freshly cleaved muscovite mica (New York Mica Co.) by either placing a 10/xl drop on mica and letting it air dry or by spraying D N A solution onto mica substrate using a nebulizer. All of the images shown in this paper were obtained from air-dried samples of DNA. In some cases the substrates were heated to 60°C by a sun lamp prior to A F M experiments to volatilize the ammonium acetate buffer. However, heating of the sample did not improve the images appreciably. A Nanoscope II (Digital Instruments, Santa Barbara, CA) was used in this study. The cantilevers used were 200 t~m long with a spring constant of 0.12 N / m (Nanoprobes) and images were recorded in a constant-force mode. For
experiments under water, D N A was first adsorbed on mica, as in air experiments, and then transferred to the liquid cell. Prior to A F M imaging, nanopure water was injected into the liquid cell. Relative humidity was controlled when operating the A F M in air by enclosing the A F M in a bell jar and purging the system with dry nitrogen.
3. Results and discussions
3.1. A F M images When a droplet of D N A solution was allowed to air-dry onto a mica surface, images of D N A could be obtained in all of the area scanned. If the D N A was deposited onto the surface by a nebulizer spray, D N A images were obtained at a frequency that depended upon the surface coverage of the nebulized droplets. The D N A molecules observed with the A F M were stable for 5 to 10 scans but repeated scanning tended to break the strands. When a control solution of buffer was deposited onto the mica surface, either by droplet or by nebulizer, no structures were found on the surface. The A F M scans carried out under buffer solution on DNA-covered mica show a clean surface without any indication of D N A adsorption. When the concentration of D N A applied to the mica was below 0.3 ~ g / m l , isolated molecules (fig. 1) are most frequently seen. The lengths of individual D N A molecules measured from the topographs were in good agreement (within 10%, averaged over 20 isolated strands) with the expected molecular length (slightly larger than 1 ~ m ) of the 3200 bp plasmid pBS +. However, the molecular width ranged from 7 to 10 times (7 was the lower limit) larger than the 2 nm width of B-DNA. Although the preparation of circular plasmid D N A used in these experiments was approximately 80% open circles, no completely open circles (fig. 1) were found in our images. Higher-resolution images (see fig. 2) of a representative circular molecule show apparent supercoiling of the molecules. We also found that the images of plasmid D N A that had been linearized showed a higher degree of supercoiling, which we
T. Thundat et aL / Imaging isolated strands of DNA by AFM
attribute to stresses introduced as a result of drying. We propose to test this hypothesis on each circular and linearized plasmid D N A by freeze-drying samples onto mica surfaces. When the concentration of D N A was increased to between 3 t z g / m l and the maximum concentration of 1 0 / z g / m l that we used in these experiments, instead of individual molecules, the D N A formed a network of interconnecting molecules that extended uniformly over the entire area of mica wetted by the DNA. Fig. 3 shows the beginning of this interconnecting network observed for a concentration 0.3 /xg/ml while fig. 4b shows a more interconnected network at a concentration of 3 / ~ g / m l . The density of D N A strands observed on the surface was directly proportional to the amount of D N A in the droplet applied. 3.2. Contrast o f images
During these experiments the contrast of the D N A images could be either positive, where D N A strands appear to be higher than the substrate, or negative, where the D N A strands appear to be
Fig. 1. 2 tzmx2 /,m AFM image showing isolated strands of pBS + DNA (concentration 0.3/*g/ml) on mica. The tip force was of 3 nN (repulsive mode). Details are enhanced using an artificial light source illumination.
1103
Fig. 2. 600 nm×600 nm AFM topograph of few strands of circular DNA showingsigns of supercoiling taken with a force of 2 nN. Black (low) to white (high) corresponds to 30 nm for the entire image. The height of DNA above the substrate varies from 1.5 to 3.1 nm.
lower than the substrate. The A F M topographs obtained with freshly prepared D N A / m i c a samples often showed a negative contrast. In addition, the D N A strands were also found to move around due to scanning. However, drying the sample for a long length of time or heating the sample often resulted in a stable positive contrast. Fig. 4a shows an A F M image of linear D N A on mica with a negative contrast taken in 64% humidity. However, a sample that gave a positive contrast one day might give a negative contrast the next day. We have investigated this phenomenon in detail and found a direct relationship between contrast and humidity. D N A images obtained on humid days (humidity over 60%) always had a negative contrast while those obtained on dry days (humidity below 30%) produced positive contrast images. On humid days it was possible to isolate the A F M in a bell jar and change the contrast from negative to positive by passing dry nitrogen into the bell jar housing the AFM. It was also possible to change the positive contrast back to a negative contrast by switching off the nitrogen and letting humid air back into
1104
T. Thundat et al. / Imaging isolated strands of DNA by AFM
Fig. 3. 2 /zmx2 /zm area of DNA on mica initial stages of formation interconnecting networks (DNA concentration 0.3 p,g/ml). Black (low) to white (high) corresponds to 46 nm. The height of individual strands varies from 1.5 to 3.1 nm. Network of DNA strands becomes more clear as the concentration of DNA is increased (see fig. 4b).
the chamber. W e have r e p r o d u c e d in this result on 20 different samples. Thus, the effect is completely reversible. Fig. 4b is a positive contrast image of the same sample as in fig. 4a t a k e n
u n d e r dry n i t r o g e n a m b i e n t . T h e D N A strands a p p e a r to be b r o a d e n e d due to coiling. T h e force curves o b t a i n e d in h u m i d air (64%) on a D N A - c o v e r e d mica surface show a rather large attractive force o n the tip before it breaks away from the surface. This b r e a k - a w a y force can be as large as 100 n N d e p e n d i n g on the humidity. However, as dry n i t r o g e n is i n t r o d u c e d into the c h a m b e r , the force can be decreased to 10 nN, or occasionally, smaller. This h u m i d i t y - d e p e n d e n t force is p r e s u m e d to be the capillary force from the liquid bridge formed b e t w e e n the a d s o r b e d water on the substrate a n d o n the sample [27-29]. T a k i n g the accepted value of surface t e n s i o n of water as 72 x 10 -3 N / m [29], the capillary forces can be as high as 100 n N d e p e n d i n g on the area of the water bridge formed b e t w e e n the tip a n d the substrate. T h e negative contrast can be att r i b u t e d to cantilever buckling d u e to an increased capillary force a r o u n d the D N A molecule [301. 3.3. Width o f D N A images T h e a p p a r e n t width of D N A o b t a i n e d from positive contrast A F M t o p o g r a p h s was 7 to 10
a
Fig. 4. Humidity-dependent positive and negative contrast of AFM images of linear DNA (concentration 3 ~zg/ml). Black (low) to white (high) corresponds to 5.2 nm. (a) 2 jzm x 2 izm on mica taken in 60% humidity. The DNA strands show negative contrast. (b) Same area of the sample in dry nitrogen ambient. The DNA strands show positive contrast.
T. Thundat et al. / Imaging isolated strands of DNA by AFM times larger while widths from the negative contrast i m a g e s w e r e up to 20 times g r e a t e r t h a n the 2 n m k n o w n width of D N A T h e s e differences can be explained by the capillary forces exerted on the tip u n d e r high h u m i d i t y c o n d i t i o n s that produce negative images. T h e widths of positive contrast D N A images were f o u n d to c h a n g e from tip to tip. O n e p r o b a b l e r e a s o n for this c o n c e r n s the c o n v o l u t i o n of the tip geometry with the D N A molecule. F r o m a simple geometrical a r g u m e n t , the d i a m e t e r of a cylinder of radius r w h e n imaged with a tip of radius R will a p p e a r as 4(rR) 1/2. A n observed D N A width of 20 n m will t h e n lead to a tip radius of 25 n m which is in a g r e e m e n t with typical tip sizes as m e a s u r e d by SEM. In some cases it is possible to have a s h a r p e r tip, either by c o n t a m i n a t i o n or by accid e n t which can p r o d u c e D N A images with a smaller width. T h e smallest width of D N A we have m e a s u r e d with a n A F M t o p o g r a p h was 7 nm, b u t even with this exceptional tip it was not possible to resolve the 3.4 n m helical r e p e a t of DNA.
4. Conclusion W e have imaged isolated strands of D N A reproducibly by a d s o r b i n g t h e m on a freshly cleaved mica. R e p e a t e d l y s c a n n i n g the same area or h i g h - m a g n i f i c a t i o n scans t e n d to b r e a k the D N A strands. T h e observed width a n d contrast of the D N A s t r a n d s was f o u n d to d e p e n d o n humidity. T h e images of D N A o b t a i n e d at 60% or higher h u m i d i t y show negative c o n t r a s t while those obt a i n e d in dry n i t r o g e n or low h u m i d i t y show positive contrast. M o r e e x p e r i m e n t s are presently underway to m a p out the exact d e p e n d e n c e of the contrast of D N A o n humidity. T h e width of D N A observed with the A F M was also f o u n d to d e p e n d o n the tip conditions. It may be possible to improve A F M r e s o l u t i o n o n D N A by using s h a r p e r tips u n d e r a dry a t m o s p h e r e .
Acknowledgements W e are grateful to K.B. J a c o b s o n a n d L.A. Bottomley helpful discussions. This research was
1105
s p o n s o r e d by the D i r e c t o r ' s R e s e a r c h a n d Develo p m e n t F u n d at O R N L a n d the Office of H e a l t h a n d E n v i r o n m e n t a l Research, U.S. D e p a r t m e n t of E n e r g y u n d e r contract D E - A C 0 5 - 8 4 O R 2 1 4 0 0 with M a r t i n M a r i e t t a E n e r g y Systems, Inc.
References [1] G. Binnig and H. Rohrer, in: Trends in Physics, Eds. J. Janata and J. Pantoflicek (European Physical Society, The Hague, pp. 38-46. [2] T.P. Beebe, T.E. Wilson, D.F. Ogletree, J.E. Katz, R. Balhorn, M.B. Salmeron and W.J. Siekhaus, Science 243 (1989) 370. [3] P.G. Arscott, G. Lee, V.A. Bloomfield and D.F. Evans, Nature 319 (1989) 484. [4] S.M. Lindsay, T. Thundat, L.A. Nagahra, U. Knipping and R. Rill, Science 244 (1989) 1063. [5] D.D. Dunlap and C. Bustamante, Nature 342 (1989) 204. [6] A. Cricenti, S. Selci, A.C. Felici, R. Generosi, E. Gori, W. Djaczenko and G. Chiarotti, Science 245 (1989) 1226. [7] R.J. Driscoll, M.G. Youngquist and J.D. Baldeschwieler, Nature 346 (1990) 294. [8] A.L. Weisenhorn, H.E. Gaub, H.G. Hansma, R.L. Sinsheimer, G.L. Kelderman and P.K. Hansma, Scanning Microsc. 4 (1990) 511. [9] A.L. Weisenhorn, M. Egger, F. Ohnesorge, S.A.C. Gould, S.P. Heyn, H.G. Hansma, R.L. Sinsheimer, H.E. Gaub and P.K. Hansma, Langmuir 7 (1991) 8. [10] H.G. Hansma, A.L. Weisenhorn, S.A.C. Gould, R.L. Sinsheimer, H.E. Gaub, G.D. Stucky, C.M. Zaremba and P.K. Hansma, J. Vac. Sci. Technol. B 9 (1991) 1282. [11] J.H. Coombs and J.W. Pethica, IBM J. Res. Dev. 30 (1986) 455. [12] H.J. Mamin, E. Ganz, D.W. Abraham, R.E. Thompson and J. Clarke, Phys. Rev. B 34 (1986) 9015. [13] U. Durig, J.K. Gimzewski and D.W. Pohl, Phys. Rev. Lett. 57 (1986) 2043. [14] M.B. Salmeron, T.P. Beebe, J. Odriozola, T.E. Wilson, D.F. Ogletree and W.J. Siekhaus, J. Vac. Sci. Technol. A 8 (1990) 635. [15] M.B. Salmeron, D.F. Ogletree, C. Ocal, H.-C. Wang, G. Neubuer, W. Kolbe and G. Meyers, J. Vac. Sci. Technol. B 9 (1991) 1347. [16] S.M. Lindsay and B. Barris, J. Vac. Sci. Technol. A 6 (1988) 544. [17] S.M. Lindsay, T. Thundat and L.A. Nagahara, J. Microscopy 152 (1988) 213. [18] T. Thundat, L.A. Nagahara, P. Oden and S.M. Lindsay, J. Vac. Sci. Technol. A 8 (1990) 645. [19] R.W. Keller, D. Bear and C. Bustamante, J. Vac. Sci. Technol. B 9 (1991) 1291. [20] J.A. Derose, S.M. Lindsay, L.A. Nagahara, P.I. Oden and T. Thundat, J. Vac. Sci. Technol. B 9 (1991) 1166.
1106
7~ Thundat et al. / Imaging isolated strands of DNA by AFM
[21] W.M. Heckl, K.M.R. Kallury, M. Thompson, C. Gerber, H.J.K. Horber and G. Binnig, Langmuir 5 (1989) 1433. [22] Y.L. Lyubchenko, S.M. Lindsay, J.A. DeRose and T. Thundat, J. Vac. Sci. Technol. B 9 (1991) 1288. [23] A. Cricenti, S. Selci, G. Chiarotti and F. Amaldi, J. Vac. Sci. Technol. B 9 (1991) 1285. [24] M.J. Allen, R.J. Tench, J.A. Mazrimas, M. Balooch, W.J. Siekhaus and R. Balhorn, J. Vac. Sci. Technol. B 9 (1991) 1272. [25] T. Thundat, R.J. Warmack, D.P. Allison and T.L. Ferrell, Ultramicroscopy 42-44 (1992) 1082.
[26] J. Sambrook, E.F. Fritsch and T. Maniat, Molecular Cloning, 2nd ed. (Cold Spring Harber Press, 1989), [27] J.T. Woodward, J.A.N. Zasadzinski and P.K. Hansma, J. Vac. Sci. Technol. B 9 (1991) 1231. [28] J.N. Israelachvilli, Intermolecular and Surface Forces (Academic Press, New York, 1985). [29] A.W. Adamson, Physical Chemistry of Surfaces, 5th ed. (Wiley Interscience, New York, 1990). [30] T. Thundat, R.J. Warmack, D.P. Allison, T.L. Ferrell and L.A Bottomley, J. Vac. Sci. Technol., in press.
