Ultramicroscopy 1 (1975) 7-14 © North-Holland Publishing Company
ELECTRON MICROSCOPE IMAGES OF MERCURY ATOMS BOUND TO DNA FILAMENT M. TANAKA*, S. HIGASHI-FUJIME** and R. UYEDA* Nagoya University, Nagoya, Japan Received 14 January 1975 Ultra-fine grids with holes of a few hundred angstroms in diameter were made from tropomyosin for specimen supporting media. DNA was mounted on the grid. Microscope images were obtained demonstrating one double-stranded DNA filament bridging a hole. Using mercurated DNA a few filamentous images were observed in a bridge, and series of faint fine spots were recognized along an edge of a hole. These spots were interpreted as the images of single mercury atoms bound to the filament. The control experiment was made on unstained DNA, some images of which showed only uniform intensity over the bridge, whereas others were disturbed to some extent by noise presumably due to extraneous molecules coating the filament.
1. Introduction Beer and Moudrianakis [1 ] studied by electron microscopy DNA stained with uranyl acetate and observed in the bright field a row o f dark specks, each of which was about 10 A in size. They attributed the speck to 3~6 uranyl ions. Crewe, Wall and Langmore [2] also observed similar images and interpreted the speck as a cluster o f 2~3 uranyl ions. In higher resolution Whiting and Ottensmeyer [3] studied DNA selectively stained with osmium tetroxide. However, the images of the filament were seriously disturbed by the noise from the support film of carbon. Since 1970 several authors [ 4 - 1 7 ] have claimed that they imaged single heavy atoms in smaller molecules with the transmission electron microscope. Since all o f them, except for Hashimoto et al. [9,10], used evaporated carbon or other amorphous materials for support films, most o f their images were heavily disturbed by the noise characteristic of amorphous films. For this reason, there are still many arguments among electron microscopists on the interpretation o f the images. In any event no clear-cut image of a single atom has yet been obtained. * Department of Applied Physics, Faculty of Engineering. Institute of Molecular Biology, Faculty of Science, Nagoya University, Nagoya, Japan.
**
In the present work, mercurated DNA filaments were mounted across a hole of newly developed ultrafine grids so that the image o f single mercury atoms could be observed free from the noise due to a support film. The result showed that most of the images were still disturbed more or less b y the noise which was due to extraneous molecules on the filaments. In one o f the best micrographs series of faint fine spots could be recognized. Each o f the spots was interpreted as the image of single mercury atom. 2. Experimental Fig. 1 shows an example of electron micrographs o f the ultra-fine grid. The diameter of holes ranges from about 100 to 1000 A, which is about one percent of that o f the conventional holey plastic film. The grid was made from tropomyosin, one o f the muscle proteins, in fibrous polymer as follows: a drop of 0.2 mg/ml tropomyosin in the standard solu" tion (10 mM sodium acetate buffer pH = 5.6 and 0.2 M KCI) was mounted on a holey plastic Film on a 150 mesh copper grid. After about half a minute the excess protein was washed ,off in the standard solution. The preparation was then fixed with OsO 4 (0.5% in the standard solution), rinsed in water, dried and followed b y carbon coating. The ultra-f'me grid
8
M. Tanaka et al. / Hg atoms bound to DNA filament ~ ",~,~- ...
•
.
.,.&:: . , - -:~: .- k, , ;' ~" ~¢ ~, : ' "-.
r' ~%~T~,t~,~;: , ~ . ,~"
"%*' . . . .
~,?:~'-': ~ :~" ~,,~ . ~:~, ,-~,
't'
~,~':, .¢
~
~ • %.,.~
(
~-~........ ~ -
"v.
~.
•
,--~
.-
'..
:-- .
r
,i~'~,x~,.,,~", ~, ~.~..~ ~: .~
.~
~
~t~"~• ~;S .~.'-'~"~ . g
'~ .:: . :~ ~--~.~ ..~,. ~ ~ • ;~;~, /~ ~J..~:. '~:
~
.~"
k..,A,'~,-~" ,,~,.,,~.
x,~
-~i
•
,
'.'..
~..
~,~
" .~.
".~'~'~. ~_~ ~.', ~ ~,~, ;_
' "~-,-' . ~ ' , , ~ . ' ¢ ~ .- ~. ~' ; ~, ~ ~ ,. .... .. , ~. .~, .. , ~ " ~ : ,':, # . x . ~--,-~" "
:7'
.-~. ~ ~ '
~ ,,~.~ ~'-,' --~
F i g . 1. U l t r a - f i n e g r i d m a d e f r o m t r o p o m y o s i n .
thus obtained proved to be as stable as the holey plastic film against electron irradiation. The grid was made hydrophilic just before mounting the DNA specimen by glow-discharge [18] in air at 0.28---0.30 torr under an AC field of 60 V/cm for 20 sec. Well purified ;k-phage DNA was supplied by Prof. R. Okazaki o f Nagoya University. In the preliminary experiment the original DNA was mounted on the conventional carbon film or on the ultra-fine grid, and then stained with 0.3% uranyl acetate. The main experiment was carried out with the DNA complexed by mercuric chloride, i.e. mercurated DNA. It is known that mercuric ions are specifically bound to DNA molecules. In the present experiment the mercuration was carried out after Nandi, Wang and Davidson [19] except for the use of ammonium acetate instead of NaCIO 4 and Na2B407. Since ammonium acetate is easy to evaporate, DNA molecules become cleaner when they are kept under vacuum. The mercuration of DNA was checked by taking ultraviolet absorption spectra. Fig. 2 is an absorption spectrum, which shows that the peak shifts from 258 rn/a to 270 m/a through the mercuration. This shift indicates that one mercuric ion has been bound to every base pair according. to Yamane and Davidson [20]. A 3/al droplet of 30/aM DNA solution containing 3.2 mM ammonium acetate (pH = 6.7) and 15/aM HgCI2 was mounted on
an ultra-fine grid. This preparation was dried in a belljar of 10 -3 torr and then kept for about 3 hours in high vacuum of 10 - 6 torr. The original DNA, neither stained nor mercurated, also was observed on the grid for comparison. In all of the above preparations DNA was not denatured. High-resolution micrographs were taken on a JEM 100B with a pointed filament at I00 kV. The objective aperture was of 8.3 × 10 -3 rad in half angle. O.D. 270
0.2-
258
1
L
o
o • •
o
o
(i
o
o •
o
o • o o
o
o
•
0,'
o
•
o
•
O •
o o
,•0 I
240
I
260
I
280
I
300
o I
320
o I
340
(mj J)
wavelength F i g . 2. U l t r a v i o l e t a b s o r p t i o n
spectra of h-phage DNA. Open
circles: with mercuric ions; closed circles: without mercuric ions.
M. Tanaka et al. / Hg atoms bound to DNA filament
Fig. 3. DNA filaments on carbon film, stained with 0.3% uranyl acetate. (a) bright field, and (b) daxk.field.
9
10
M. Tanaka et al. / Hg atoms bound to DNA filament
Dark field images were obtained by tilting the incident beam outside the objective aperture which was fixed on the optic axis, the tilting angle being 1.4 X 10 -2 rad. The electron optical magnification was generally 2.23 X 105. The anticontamination device cooled by liquid nitrogen was the standard type. The specimen chamber was kept under the pressure of about 3 X 10 -6 torr.
3. Results Fig. 3 shows a pair of bright and dark field images of stained DNA filaments on a carbon film. The thinnest filament is seen as a row of nearly equally spaced (~35 A) dark specks in the bright field image, while thicker ones are irregular. These images resemble those reported by Beer et al. [1] or Crewe et al. [2]. Since natural DNA was used in the present experiment while denatured DNA was used in the previous work, the row in fig. 3 corresponds to one double-. stranded DNA filament.
Fig. 4a shows a bright field image of a bridge of DNA across a hole of the ultra-fine grid. The specimen was stained after the bridge of original DNA was spanned and dried up. On the bridge a row of dark specks is seen, whose size and spacing are much the same as those in fig. 3a. This implies that the bridge is made of one double-stranded DNA filament. Fig. 4b shows the corresponding dark field image. A group of a few fine bright spots is recognizable at each place where a dark speck is seen in the bright field image. No exact explanation has yet been given to the specks, nor to the fine bright spots. The tlfickness of the bridge is about 45 A, which is much larger than the value of about 20 A proper to a doublestranded DNA, and the image shows such granular noise as is seen for amorphous carbon fdms. These facts indicate that the filament is coated with extraneous molecules. Since the DNA used in the present work was well deprived of proteins, the extraneous molecules must be the contaminations deposited mostly in the specimen preparation and, to some extent, in electron microscope observation. Since the
!1 ,I00,~ Fig. 4. Bridge of DNA filament stained with uranyl acetate. (a) bright field, and (b) dark field.
Fig. 5. Bridges of original DNA without any stain (a) bright field, taken at 700 A under-focus, and (b) dark field.
i" : ~I-,~ •: ~ ',~ Fig. 6. Bridge of mercurated DNA. (a) bright field, taken at 820 A under-focus, and (b) dark field.
12
M. Tanaka et al. / Hg atoms bound to DNA filament
P~T • lt
," , .,, - ~
,
•
" Fig. 7. Enlargement of the bridge seen in fig. 6(b). Arrows indicate fine spots.,
former contamination was thought attributable to ions in the ordinary distilled water used in the preparation, water distilled in quartzware was used in the later part of the experiment. The latter contamina-
tion could be practically eliminated by shortening the observation duration, as it was found negligible within several minutes under the actual working conditions of the electron microscope. With these improvements the micrographs in figs. 5 and 6 were taken. Fig. 5 shows the image of bridges of original DNA. In the dark field image the thicker bridge (A) shows granular noises which may be due to contamination. On the other hand, the thinner one (B) shows weak uniform intensity without any granular noise. It is worth noting that a bridge as thin as 30 A has little contamination to give granular noise. Fig. 6 shows a pair o f images of DNA with mercuric ions added. A bridge about 30.8, thick and 150,8, in its span is seen across the lower hole. In the dark field image two filaments can be seen in this bridge. In the enlargement of the filaments, fig. 7, a few faint fine spots can be recognized along one of them, although one can hardly say further about
Fig. 8.Enlargement of the marked region in fig. 6Co). Faint fine spots as sketched'in the inset are interpreted as images of single mercury atoms.
M: Tanaka et al. / Hg atoms bound to DNA filament
these spots because they are not very clear. On the other hand, along the edge of the adjacent small hole something like two pieces of thread, partly tangled each other, can be seen as indicated by arrows. These threads must be DNA filaments which happened to lie along the edge. They are seen in a more highly magnified print, fig. 8, to be series of very faint fine spots as illustrated in the inset. The intensity of some of them could be measured with a microdensitometer and was found to be between 0.35 and 0.11% of the illumination intensity when subtracted from the background. The indefiniteness in intensity was due to the difficulty in estimating the illumination intensity of the pointed filament at the moment of taking the dark field image. The intensity of the fine spots is lower than that of fine bright spots which are seen in the dark field image of.amorphous carbon films [21]. Two facts should be pointed out about the threads. First, the fine spots are aligned in series, the spacing between the two indicated being about 3.6 A, which is in good accord with that between two adjacent base pairs of DNA. Second, no image of the same sort could be observed in the case of DNA without mercury. These faint f'me spots, therefore, are presumably the images of single mercury atoms bound to DNA.
13
with their transmission scanning microscope and reported that they imaged single mercury atoms. However, the number of observed mercury atoms was less than 20% of the expected total. They suggested the possibility of chemical decomposition of the mercurated DNA on the carbon film. On the other hand, according to Yamane and Davidson [20] mercuric ions bound to DNA are not easily removed without a sufficiently high concentration of chloride or cyanide ions. The present result seems to be in accord with the latter.
5. Conclusion Mercurated DNA filaments bridging a hole in the newly-developed ultra-free grid showed filamentous images in the dark field image, although further details in the images were not clear. This result irfiplies that the contamination coating the filament masked the image of mercury atoms since no support f'tim existed beneath the filaments. However, images of single mercury atoms were obtained, by chance, which were bound to DNA filaments lying along an edge of a hole. From their intensity and spacing as well as the control experiment it was concluded that they were images of single mercury atoms.
4. Discussibn Acknowledgements
Series of fine spots were observed along an edge of a hole and interpreted as the images of single mercury atoms. They are less disturbed by noise than any other reported atom images. It is desirable that clearer images be obtained consistently to establish the ability to image single atoms. To this aim the attempt to image mercury atoms bound to ffflamentous molecules bridging a hole is quite new. In this way single atoms must be imaged free from the noise due to a support film. At the present stage, however, theimage obtained is not as good as that observed along the edge. The contamination due to extraneous molecules in the specimen preparation seems to form the main difficulty. In order to reduce the contamination it is worthwhile to try decreasing the concentration of ammonium acetate insofar as both good dispersion and mercuration of DNA are ensured. Recently Crewe et al. [7] studied mereurated DNA
The authors would like to express their sincere thanks to Professor Reiji Okazaki for kindly providing h-phage DNA. The support of the Japan Society for Promotion of Science for one of the authors (M.T.) is gratefully acknowledged.
References
[1] M. Beer, and E.N. Moudfianakis, Prec. Nat. Acad. Sci. 48 (1962) 409-416.
[2] A.V. Crewe, J. Wall, and J. Langmore, MicroscopicElec tronique, Grenoble, 1, (1970) 467-468. [3] R.F. Whiting, and F.P. Ottensmeyer, J. MoL Bi01. 67
(1972) 173-181. [4] A.V. Crewe, J. Waft and J. Langmore, Science 168 (1970) 1338-1340. [5] A.V. Crewe, l~oc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 640-642.
14
M. Tanaka et al. / H g atoms bound to DNA filament
[6] A.V. Crewe, J. Langmore, M. Isaacson, and M. Retsky, Electron Microscopy, Canberra, 1, (1974) 260-261. [7] A.V. Crewe, J.P. Langrnore, M.K. Lamvik, and J. Pullman, Electron Microscopy, Canberra, 2, (1974) 162-163. [8] H. Formanek, M. Mfiller, M.H. Hahn, and T. Koller, Naturwiss. 58 (1971) 339-344. [9] H. Hashimoto, A. Kumao, K. Hino, H. Yotsumoto, and A. OnocJapan. J. Appl. Phys. 10 (1971) 1115-1116. [10] H. Hashimoto, A. Kumao, K. Hino, H. Endoh, H. Yotsumoto, and A. Ono, J. Electron Microscopy 22 (1973) 123-134. [11 ] R.M. Henkelman, and F.P. Ottensmeyer, Proc. Nat. Acad. Sei. USA, 68 (1971) 3000-3004. [12] J.R. Parsons, H.M. Johnson, C.W. Hoelke, and R.R. Hosbons, Proc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 646-647; Phil. Mag. 27 (1973) 1359-1368. [13] V.A. Phillips, A.J. Chalk, and J.A. Hugo, J. Electron Microscopy 21 (1972) 323-324.
[14] E.B. Prestridge, and D.J.C. Yates, Nature, 234 (1971) 345 -347. [15] F. Thon, and D. Willasch, Optik 36 (1972) 55-58; Proc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 650-651. [16] H.J. Vollenweider, T. Koller, and O. Kiibler, J. Microscopie, 16 (1973) 247-256. [17] D. Dorignac, Electron Microscopy, Canberra, 1, (1974) 270-271. [18] J. Dubochet, M. Ducommun, M. Zoilinger, and E. Kellenberger, J. Ultrastructure Research 35 (1971) 147-167. [19] U.S. Nandi, J.C. Wang, and N. Davidson, Biochemistry 4 (1965) 1687-1696. [20] T. Yamane, and N. Davidson, J. Am. Chem. Soc. 83 (1961) 2599-2607. [21] M. Tanaka, K. Mihama, and R. Uyeda, J. Electron Microscopy 22 (1973) 221-222.
ELECTRON MICROSCOPE IMAGES OF MERCURY ATOMS BOUND TO DNA FILAMENT M. TANAKA*, S. HIGASHI-FUJIME** and R. UYEDA* Nagoya University, Nagoya, Japan Received 14 January 1975 Ultra-fine grids with holes of a few hundred angstroms in diameter were made from tropomyosin for specimen supporting media. DNA was mounted on the grid. Microscope images were obtained demonstrating one double-stranded DNA filament bridging a hole. Using mercurated DNA a few filamentous images were observed in a bridge, and series of faint fine spots were recognized along an edge of a hole. These spots were interpreted as the images of single mercury atoms bound to the filament. The control experiment was made on unstained DNA, some images of which showed only uniform intensity over the bridge, whereas others were disturbed to some extent by noise presumably due to extraneous molecules coating the filament.
1. Introduction Beer and Moudrianakis [1 ] studied by electron microscopy DNA stained with uranyl acetate and observed in the bright field a row o f dark specks, each of which was about 10 A in size. They attributed the speck to 3~6 uranyl ions. Crewe, Wall and Langmore [2] also observed similar images and interpreted the speck as a cluster o f 2~3 uranyl ions. In higher resolution Whiting and Ottensmeyer [3] studied DNA selectively stained with osmium tetroxide. However, the images of the filament were seriously disturbed by the noise from the support film of carbon. Since 1970 several authors [ 4 - 1 7 ] have claimed that they imaged single heavy atoms in smaller molecules with the transmission electron microscope. Since all o f them, except for Hashimoto et al. [9,10], used evaporated carbon or other amorphous materials for support films, most o f their images were heavily disturbed by the noise characteristic of amorphous films. For this reason, there are still many arguments among electron microscopists on the interpretation o f the images. In any event no clear-cut image of a single atom has yet been obtained. * Department of Applied Physics, Faculty of Engineering. Institute of Molecular Biology, Faculty of Science, Nagoya University, Nagoya, Japan.
**
In the present work, mercurated DNA filaments were mounted across a hole of newly developed ultrafine grids so that the image o f single mercury atoms could be observed free from the noise due to a support film. The result showed that most of the images were still disturbed more or less b y the noise which was due to extraneous molecules on the filaments. In one o f the best micrographs series of faint fine spots could be recognized. Each o f the spots was interpreted as the image of single mercury atom. 2. Experimental Fig. 1 shows an example of electron micrographs o f the ultra-fine grid. The diameter of holes ranges from about 100 to 1000 A, which is about one percent of that o f the conventional holey plastic film. The grid was made from tropomyosin, one o f the muscle proteins, in fibrous polymer as follows: a drop of 0.2 mg/ml tropomyosin in the standard solu" tion (10 mM sodium acetate buffer pH = 5.6 and 0.2 M KCI) was mounted on a holey plastic Film on a 150 mesh copper grid. After about half a minute the excess protein was washed ,off in the standard solution. The preparation was then fixed with OsO 4 (0.5% in the standard solution), rinsed in water, dried and followed b y carbon coating. The ultra-f'me grid
8
M. Tanaka et al. / Hg atoms bound to DNA filament ~ ",~,~- ...
•
.
.,.&:: . , - -:~: .- k, , ;' ~" ~¢ ~, : ' "-.
r' ~%~T~,t~,~;: , ~ . ,~"
"%*' . . . .
~,?:~'-': ~ :~" ~,,~ . ~:~, ,-~,
't'
~,~':, .¢
~
~ • %.,.~
(
~-~........ ~ -
"v.
~.
•
,--~
.-
'..
:-- .
r
,i~'~,x~,.,,~", ~, ~.~..~ ~: .~
.~
~
~t~"~• ~;S .~.'-'~"~ . g
'~ .:: . :~ ~--~.~ ..~,. ~ ~ • ;~;~, /~ ~J..~:. '~:
~
.~"
k..,A,'~,-~" ,,~,.,,~.
x,~
-~i
•
,
'.'..
~..
~,~
" .~.
".~'~'~. ~_~ ~.', ~ ~,~, ;_
' "~-,-' . ~ ' , , ~ . ' ¢ ~ .- ~. ~' ; ~, ~ ~ ,. .... .. , ~. .~, .. , ~ " ~ : ,':, # . x . ~--,-~" "
:7'
.-~. ~ ~ '
~ ,,~.~ ~'-,' --~
F i g . 1. U l t r a - f i n e g r i d m a d e f r o m t r o p o m y o s i n .
thus obtained proved to be as stable as the holey plastic film against electron irradiation. The grid was made hydrophilic just before mounting the DNA specimen by glow-discharge [18] in air at 0.28---0.30 torr under an AC field of 60 V/cm for 20 sec. Well purified ;k-phage DNA was supplied by Prof. R. Okazaki o f Nagoya University. In the preliminary experiment the original DNA was mounted on the conventional carbon film or on the ultra-fine grid, and then stained with 0.3% uranyl acetate. The main experiment was carried out with the DNA complexed by mercuric chloride, i.e. mercurated DNA. It is known that mercuric ions are specifically bound to DNA molecules. In the present experiment the mercuration was carried out after Nandi, Wang and Davidson [19] except for the use of ammonium acetate instead of NaCIO 4 and Na2B407. Since ammonium acetate is easy to evaporate, DNA molecules become cleaner when they are kept under vacuum. The mercuration of DNA was checked by taking ultraviolet absorption spectra. Fig. 2 is an absorption spectrum, which shows that the peak shifts from 258 rn/a to 270 m/a through the mercuration. This shift indicates that one mercuric ion has been bound to every base pair according. to Yamane and Davidson [20]. A 3/al droplet of 30/aM DNA solution containing 3.2 mM ammonium acetate (pH = 6.7) and 15/aM HgCI2 was mounted on
an ultra-fine grid. This preparation was dried in a belljar of 10 -3 torr and then kept for about 3 hours in high vacuum of 10 - 6 torr. The original DNA, neither stained nor mercurated, also was observed on the grid for comparison. In all of the above preparations DNA was not denatured. High-resolution micrographs were taken on a JEM 100B with a pointed filament at I00 kV. The objective aperture was of 8.3 × 10 -3 rad in half angle. O.D. 270
0.2-
258
1
L
o
o • •
o
o
(i
o
o •
o
o • o o
o
o
•
0,'
o
•
o
•
O •
o o
,•0 I
240
I
260
I
280
I
300
o I
320
o I
340
(mj J)
wavelength F i g . 2. U l t r a v i o l e t a b s o r p t i o n
spectra of h-phage DNA. Open
circles: with mercuric ions; closed circles: without mercuric ions.
M. Tanaka et al. / Hg atoms bound to DNA filament
Fig. 3. DNA filaments on carbon film, stained with 0.3% uranyl acetate. (a) bright field, and (b) daxk.field.
9
10
M. Tanaka et al. / Hg atoms bound to DNA filament
Dark field images were obtained by tilting the incident beam outside the objective aperture which was fixed on the optic axis, the tilting angle being 1.4 X 10 -2 rad. The electron optical magnification was generally 2.23 X 105. The anticontamination device cooled by liquid nitrogen was the standard type. The specimen chamber was kept under the pressure of about 3 X 10 -6 torr.
3. Results Fig. 3 shows a pair of bright and dark field images of stained DNA filaments on a carbon film. The thinnest filament is seen as a row of nearly equally spaced (~35 A) dark specks in the bright field image, while thicker ones are irregular. These images resemble those reported by Beer et al. [1] or Crewe et al. [2]. Since natural DNA was used in the present experiment while denatured DNA was used in the previous work, the row in fig. 3 corresponds to one double-. stranded DNA filament.
Fig. 4a shows a bright field image of a bridge of DNA across a hole of the ultra-fine grid. The specimen was stained after the bridge of original DNA was spanned and dried up. On the bridge a row of dark specks is seen, whose size and spacing are much the same as those in fig. 3a. This implies that the bridge is made of one double-stranded DNA filament. Fig. 4b shows the corresponding dark field image. A group of a few fine bright spots is recognizable at each place where a dark speck is seen in the bright field image. No exact explanation has yet been given to the specks, nor to the fine bright spots. The tlfickness of the bridge is about 45 A, which is much larger than the value of about 20 A proper to a doublestranded DNA, and the image shows such granular noise as is seen for amorphous carbon fdms. These facts indicate that the filament is coated with extraneous molecules. Since the DNA used in the present work was well deprived of proteins, the extraneous molecules must be the contaminations deposited mostly in the specimen preparation and, to some extent, in electron microscope observation. Since the
!1 ,I00,~ Fig. 4. Bridge of DNA filament stained with uranyl acetate. (a) bright field, and (b) dark field.
Fig. 5. Bridges of original DNA without any stain (a) bright field, taken at 700 A under-focus, and (b) dark field.
i" : ~I-,~ •: ~ ',~ Fig. 6. Bridge of mercurated DNA. (a) bright field, taken at 820 A under-focus, and (b) dark field.
12
M. Tanaka et al. / Hg atoms bound to DNA filament
P~T • lt
," , .,, - ~
,
•
" Fig. 7. Enlargement of the bridge seen in fig. 6(b). Arrows indicate fine spots.,
former contamination was thought attributable to ions in the ordinary distilled water used in the preparation, water distilled in quartzware was used in the later part of the experiment. The latter contamina-
tion could be practically eliminated by shortening the observation duration, as it was found negligible within several minutes under the actual working conditions of the electron microscope. With these improvements the micrographs in figs. 5 and 6 were taken. Fig. 5 shows the image of bridges of original DNA. In the dark field image the thicker bridge (A) shows granular noises which may be due to contamination. On the other hand, the thinner one (B) shows weak uniform intensity without any granular noise. It is worth noting that a bridge as thin as 30 A has little contamination to give granular noise. Fig. 6 shows a pair o f images of DNA with mercuric ions added. A bridge about 30.8, thick and 150,8, in its span is seen across the lower hole. In the dark field image two filaments can be seen in this bridge. In the enlargement of the filaments, fig. 7, a few faint fine spots can be recognized along one of them, although one can hardly say further about
Fig. 8.Enlargement of the marked region in fig. 6Co). Faint fine spots as sketched'in the inset are interpreted as images of single mercury atoms.
M: Tanaka et al. / Hg atoms bound to DNA filament
these spots because they are not very clear. On the other hand, along the edge of the adjacent small hole something like two pieces of thread, partly tangled each other, can be seen as indicated by arrows. These threads must be DNA filaments which happened to lie along the edge. They are seen in a more highly magnified print, fig. 8, to be series of very faint fine spots as illustrated in the inset. The intensity of some of them could be measured with a microdensitometer and was found to be between 0.35 and 0.11% of the illumination intensity when subtracted from the background. The indefiniteness in intensity was due to the difficulty in estimating the illumination intensity of the pointed filament at the moment of taking the dark field image. The intensity of the fine spots is lower than that of fine bright spots which are seen in the dark field image of.amorphous carbon films [21]. Two facts should be pointed out about the threads. First, the fine spots are aligned in series, the spacing between the two indicated being about 3.6 A, which is in good accord with that between two adjacent base pairs of DNA. Second, no image of the same sort could be observed in the case of DNA without mercury. These faint f'me spots, therefore, are presumably the images of single mercury atoms bound to DNA.
13
with their transmission scanning microscope and reported that they imaged single mercury atoms. However, the number of observed mercury atoms was less than 20% of the expected total. They suggested the possibility of chemical decomposition of the mercurated DNA on the carbon film. On the other hand, according to Yamane and Davidson [20] mercuric ions bound to DNA are not easily removed without a sufficiently high concentration of chloride or cyanide ions. The present result seems to be in accord with the latter.
5. Conclusion Mercurated DNA filaments bridging a hole in the newly-developed ultra-free grid showed filamentous images in the dark field image, although further details in the images were not clear. This result irfiplies that the contamination coating the filament masked the image of mercury atoms since no support f'tim existed beneath the filaments. However, images of single mercury atoms were obtained, by chance, which were bound to DNA filaments lying along an edge of a hole. From their intensity and spacing as well as the control experiment it was concluded that they were images of single mercury atoms.
4. Discussibn Acknowledgements
Series of fine spots were observed along an edge of a hole and interpreted as the images of single mercury atoms. They are less disturbed by noise than any other reported atom images. It is desirable that clearer images be obtained consistently to establish the ability to image single atoms. To this aim the attempt to image mercury atoms bound to ffflamentous molecules bridging a hole is quite new. In this way single atoms must be imaged free from the noise due to a support film. At the present stage, however, theimage obtained is not as good as that observed along the edge. The contamination due to extraneous molecules in the specimen preparation seems to form the main difficulty. In order to reduce the contamination it is worthwhile to try decreasing the concentration of ammonium acetate insofar as both good dispersion and mercuration of DNA are ensured. Recently Crewe et al. [7] studied mereurated DNA
The authors would like to express their sincere thanks to Professor Reiji Okazaki for kindly providing h-phage DNA. The support of the Japan Society for Promotion of Science for one of the authors (M.T.) is gratefully acknowledged.
References
[1] M. Beer, and E.N. Moudfianakis, Prec. Nat. Acad. Sci. 48 (1962) 409-416.
[2] A.V. Crewe, J. Wall, and J. Langmore, MicroscopicElec tronique, Grenoble, 1, (1970) 467-468. [3] R.F. Whiting, and F.P. Ottensmeyer, J. MoL Bi01. 67
(1972) 173-181. [4] A.V. Crewe, J. Waft and J. Langmore, Science 168 (1970) 1338-1340. [5] A.V. Crewe, l~oc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 640-642.
14
M. Tanaka et al. / H g atoms bound to DNA filament
[6] A.V. Crewe, J. Langmore, M. Isaacson, and M. Retsky, Electron Microscopy, Canberra, 1, (1974) 260-261. [7] A.V. Crewe, J.P. Langrnore, M.K. Lamvik, and J. Pullman, Electron Microscopy, Canberra, 2, (1974) 162-163. [8] H. Formanek, M. Mfiller, M.H. Hahn, and T. Koller, Naturwiss. 58 (1971) 339-344. [9] H. Hashimoto, A. Kumao, K. Hino, H. Yotsumoto, and A. OnocJapan. J. Appl. Phys. 10 (1971) 1115-1116. [10] H. Hashimoto, A. Kumao, K. Hino, H. Endoh, H. Yotsumoto, and A. Ono, J. Electron Microscopy 22 (1973) 123-134. [11 ] R.M. Henkelman, and F.P. Ottensmeyer, Proc. Nat. Acad. Sei. USA, 68 (1971) 3000-3004. [12] J.R. Parsons, H.M. Johnson, C.W. Hoelke, and R.R. Hosbons, Proc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 646-647; Phil. Mag. 27 (1973) 1359-1368. [13] V.A. Phillips, A.J. Chalk, and J.A. Hugo, J. Electron Microscopy 21 (1972) 323-324.
[14] E.B. Prestridge, and D.J.C. Yates, Nature, 234 (1971) 345 -347. [15] F. Thon, and D. Willasch, Optik 36 (1972) 55-58; Proc. 5th Europ. Cong. on Electron Microscopy, Manchester, (1972) 650-651. [16] H.J. Vollenweider, T. Koller, and O. Kiibler, J. Microscopie, 16 (1973) 247-256. [17] D. Dorignac, Electron Microscopy, Canberra, 1, (1974) 270-271. [18] J. Dubochet, M. Ducommun, M. Zoilinger, and E. Kellenberger, J. Ultrastructure Research 35 (1971) 147-167. [19] U.S. Nandi, J.C. Wang, and N. Davidson, Biochemistry 4 (1965) 1687-1696. [20] T. Yamane, and N. Davidson, J. Am. Chem. Soc. 83 (1961) 2599-2607. [21] M. Tanaka, K. Mihama, and R. Uyeda, J. Electron Microscopy 22 (1973) 221-222.
