close

Вход

Забыли?

вход по аккаунту

?

DNA Supercoiling Imaged in Three Dimensions by Scanning Force Microscopy.

код для вставкиСкачать
order to minimize electrostatic repulsion between the positively charged reactants. In fact, if the cyclophane is centered
over the dialkoxybenzene subunit in the RIP it would require
only about 10 ns for it to migrate coherently to the nonoxidized ferrocene subunit.["] In this respect, 3 operates as a
light-induced molecular shuttle, since absorption of light induces migration of the cyclophane along the thread before
reverse electron transfer restores the original conformation.
[16] P. R. Ashton. M. R. Johnson, J. F. Stoddart, M. S. Tolley, J. W. Wheeler.
J. Chem. Soc. Chem. Commun. 1992. 1128.
[17] N. S. Hush, Prog. Inorg. Chem. 1967,8, 391.
[I81 According to this analytical treatment. both AGO and i.
are obtained from
the absorption and emission peak maxima, while V is calculated by fitting
the charge-transfer absorption band to a Gaussian profile. Modern theory
relates the rate of electron transfer to these three parameters.
[I91 A. Masad. H. Huppert, E. M. Kosower. Chem. Phys. 1990, 144. 391
[20] R. A. Marcus, J. Chem. Phys. 1965, 43, 679.
[21] Calculated according to d = (2 Dr)'!' where dis the one-dimensional translation distance (10 A), D is the diffusion coefficient of the cyclophane
( = 5 x lo-' cm's-I). and I is the time required for the migration.
Experimental Procedure
1 : To a vigorously stirred solution of 1,4-bis[2-{2-(2-hydroxyethoxy)ethoxy]ethoxylbenzene 1121 (1.2 g. 3.2 mmol) in dried CH,CI, (50 m i ) containing triethylamine (1 g, 9.9 mmol) was added dropwise a solution of ferrocenyl carboxylic chloride 1131 ( 2 g, 8.0 mmol) i n dried CH,CI, (25 m i ) . The mixture was
stirred for a further 48 h before addition of water (20 mL). The organic layer
was 5eparated. washed with water. dried, and evaporated to dryness under
reduced pressure to give a red oil. The oil was purified by chromatography on
silica (60 200 mesh) with CH,CI, as eluent to remove unchanged starting
materials and subsequently with CH,CI,/CH,OH (95:'5) to elute compound 1.
Removal of the solvent gave I as a yellow solid. Yield: 1.6 g (62%). 'H NMR
(250 MHz. CDCI,. 25 C): 6 = 3.71-3.85 (m. 16H). 4.04-4.08 (t. 4H.
J = 5 H ~ ) : 4 . 2 0 ( s .lOH).4.37-4.41 (m.XH).4.81-4.83(m,4H):6.82(~.4H).
I3C NMR (63 MHz. CDCI,, 25°C): 6 = 63.23. 68.00, 69.44. 69.79, 69.87,
70.17. 70.64, 70.79. 70.88, 71.30. 115 52. 153.03, 170.90, FAB-MS (nitrobenzyl
[nm]
, (6:
alcohol matrix): 798 [MI' ; 585 [ M FcCO]+,U V j V t S (CH,CN): i.,,
[ K l c m - ' ] ) = 442 (500).
DNA Supercoiling Imaged in Three Dimensions
by Scanning Force Microscopy **
By Bruno Samori,* Carmelo Nigro, Valerio Armentano,
Salvatore Cimieri, Giampaolo Zuccheri,
and Carla Quagliariello
Scanning force microscopy (SFM) is being employed ever
more extensively in organic['] and biological chemistry.['. 31
With S F M molecules can be imaged both in water14] and
organic solvents.12'',61 Though this technique is still being
3: A mixture of 2 [12] (180 mg. 0.25 mmol), 1.4-bis(bromomethyl)benzene
developed, there is no question about the profound impact it
(66 mg. 0.25 mmol). 1 (600 mg, 0.75 mmol), and AgPF, (1 58 mg, 0.63 mmol) in
will have on the study of molecular structure and behavior.
dried CH,CN (50 mL) was stirred for 7 days at room temperature. SubsequentIn essentially all organisms the D N A double strand is
ly, the solution was filtered and evaporated to dryness under reduced pressure.
coiled in a right-handed fashion termed negative superThe residue wits washed with CH,CI, (15 mL) and centrifuged. After removal
of the supernatant liquid. the residue was washed again with CH,CI,
coiling.''] Supercoiled and relaxed circular D N A molecules
( 2 x 15 mL) and centrifuged. The solid was dissolved in a mixture of CH,NO,
are now routinely imaged by SFM.[2.4s6'U p until now this
(15 mL) and CH,CI, (10 mL) and centrifuged once more. The liquid phase was
kind of imaging provided, like electron microscopy (EM),"]
separated. and the solid residue was extracted with a mixture of CH,NO,
only a planar projection of the molecular shape of super(10 mL) and CH,CI, (10 mL). The liquid phases were combined. and the solvent removed under vacuum without heating to give a brown oil. Chromatogcoiled circular
'] However, these SFM images also
raphy o n silica with CH,OH/aq. NH,CI ( 2 M)/CH,NO, (71211) aseluent yieldcontain surface height information for the D N A molecules
ed the required fraction ( R , = 0.37), from which the solvent was removed by
examined, although the vertical resolution attained is not
freeze-drying to give a brown solid. This material was dissolved in water
high enough to allow the complete three-dimensional path of
(20 m i ) . and the required product was precpitated by addition of aqueous
NH,PF,. Yield: 31 mg (6%). 'H NMR (250 MHz. [D,]-dimethylsulfoxide.
the chains to be observed directly. The image in Figure 1
2 5 Y 3 . 6 =3.32-3.66(m,20H);3.89-3.93(m.8H),4.21(~,10H),4.39(t,2H,shows, however, that this can indeed be achieved.
J = 5 H z ) , 4.48 (t. 2H. J = 5 H z ) . 4.61 (t, 2H. J = 5 H z ) , 4.72 (I. 2H.
The number of turns (two) of the DNA molecule chain
J = 5 Hz), 5.75(s. XH),7.88(~,8H),821-8.24(d, XH, J = 6.5 Hz).9.23-9.26
about the superhelical axis and the right-handedness of the
(d, 8H. J = 6.5 Hz). FAB-MS (nitrobenzyl alcohol matrix): 1898 [MI': 1753
=
[M - PF,]+; 1608 [M - 2PFJ. UVjVIS (CH,CN): i.,,,[nm](E[M-'cm-'])
winding are well distinguished. We can thus fully and directly
452 (700).
characterize the supercoiling of this molecule because we
"see" it. This high resolution has so far been obtained only
Received: May 10.1993 [Z 6073 IE]
for molecules with two o r three turns.['] In the other cases
German version: Angetv. Chem. 1993, 105, 1553
stratigraphic analysis can be applied16] to obtain the thirddimension data contained in S F M images of more tightly
[l] T. Yabe, J. Kochi, J. Am. Chem. SOC.1992. 114, 4491.
supercoiled molecules, and the chirality of the supercoiling
[2] T. Asahi. N. Mataga, J. Phys. Chem. 1989, 93, 6575.
can be identified for most well-detected loops.
[3] 1. R. Gould. R. Moody, S. Farid. J. Am. Chen?.Soc. 1988. 110. 7242.
141 N. Mataga. Y. Kanda. T. Asahi, H. Misasaka, T. Okada. T. Kakitani,
This display of the complete three-dimensional path of the
Chrm P h w 1988, 127. 239.
axis of the D N A double strand opens up several important
[51 C. Zhou. J. B. Miers, R. M. Ballew, D. D. Dlott, G. B. Schuster, J. Am.
prospects:
Chm7. Sot. 1991, 113, 7823.
1. Experiments can now be designed so that one can di[6] H.. Segawa. C. Takehara. K. Honda. T. Shimidzu, T. Asahi. N. Mataga, J.
Pi7y.c. Chem. 1992, 96. 503.
rectly monitor the local modulations of supercoiling, which
[7] A. M. Brun, A. Harriman, S. M. Hubig, J. Ph.vs. Chem. 1992, 96, 254.
facilitate and drive many important biological processes
[XI A. C. Benniston, A. Harriman, D. Philp, J. F. Stoddart, J. Am. Chem. SOC.
such as D N A transcription['and polymorphism.[' 'I Su1993, 115. 5298.
percoiling stabilizes H-triple helices and cruciform and left191 A. C. Benniston, A. Harriman, Synlert 1993, 223.
~
1101 D. Gust, T. A. Moore, Science 1989, 244, 35.
[ I l l M. R. Wasielewski, Chem. Rev. 1992, 92, 435.
1121 P:L. Anelli, P. R. Ashton, R. Ballardini. V. Balzani, M. Delgado, M. T.
Gandolfi, T. T. Goodnow. A. E. Kaifer, D. Philp. M. Pietraszkiewicz. L.
Prodi. M. V. Reddington. A.M. Z. Slawin, N. Spencer, J. F. Stoddart. C.
Vicent. D. J. Williams, J. Am. Chem. Soc. 1992. 114, 193.
[I31 H. J. von Lorkowski. R. Pannier, A. Wende. J. Prukr. Chem. 1967,35,149.
[I41 B. Odell. M. V. Reddington. A.M. Z. Slawin, N. Spencer, J. F. Stoddart.
D. J. Williams, Angew. Chern. 1988, 100.1605; Angen. Chem. Inr. Ed. Engl.
1988.27. 1547.
[I51 P. R. Ashton, D. Philp. M. V. Reddington. A. M. Z . Slawin. N. Spencer,
1. F. Stoddart. D. J. Williams, J. Chem. SOC.Chem. Commun. 1991, 1680.
Angrw. Chrm. Int. Ed. Engl. 1993, 32, No. I0
0 VCH
[*] Prof. B. Samori, Dr. C. Nigro. V. Armentano. s. Cimieri, G. Zuccheri
Dipartimento di Chimica. Universiti della Calabria
ArCdVdcata di Rende (CS) 1-87030 (Italy)
Telefax: Int. code (984)492044
Prof. C. Quagliariello
Dipartimento di Biologia Cehiare. Universitd della Cdhbrbd
+
[**I This work was supported by CNR (Rome), Progetto Finalizzdto Btotecnologie Biostrumentazione. Regione Emilia Romagna, and Digital Instruments (Santa Barbara, CA, USA).
l4~rlagsgeseIlschufimbH. 0-69451 Weinheim, 1993
o570-0X33~93ji010-1461$ iO.00f .25/O
1461
Fig. 1. SFM image o f a supercoiled pBR322 DNA molecule. The three-dimensional
shape of a circular DNA molecule is clearly displayed by the
unfiltered image (left) in which
the height information is coded by shades of purple (the
lighter. the higher). In the line
plot (right). which IS obtained
by plotting each scan line on
an x.i’.: scale, the path of the
DNA double strand is even
more distinct. The two turns of
the DNA chains about the
superhelical axis and the righthandedness of the supercoiling
can be directly recognized by
inspection.
100 nrn
handed Z-DNA structures.“ ‘I The conversion from negative
to positive supercoiling is reported as characterizing highly
transcribed genes.“ The modulations of supercoiling depend upon local sequences, the position of promoters, and
the vectorial tracking of macromolecular complexes along
the DNA.l9
2. The writhing number (W,) of coiled molecule^,[^^ 13] the
most complete and accurate parameter for describing supercoiling and the overall hydrodynamic shape of D N A molecules, can now be determined experimentally. The most
widely used descriptor of supercoiling has thus far been the
“linking deficit” (ALk = L, - L,,), which reduces the threedimensional intertwining of D N A strands to a single number
that can be measured by two-dimensional gel electrophoresis. The linking number L, is the sum of the signed
indexes associated with the crossing between either D N A
strand and the D N A axis.[31These nodes can arise in either
of two ways: from the winding of the strands about the axis
of the double helix, which is related to the twisting number
(T,), or from the crossings of the helix axis itself, which are
described by the writhing number (WT).L,, is the reference
linking number of the relaxed circular D N A molecule.
There are at least two reasons to regard the linking deficit
as a poor descriptor of supercoiling. It is a topological property defined independently of D N A geometry, whereas
supercoiling is a geometric concept. Then, too, it is formally
possible to have a nonzero linking deficit ALk for a perfectly
planar D N A molecule.[131
By contrast, W, is a truly geometric property of the D N A
axis. It is directly proportional to the number of times the
D N A winds round the supercoiling axis and equal to zero for
relaxed DNA. The use of W, in place of ALk as a descriptor
of supercoiling has been very limited heretofore, as W, could
not be experimentally determined. The W, of an ideal model
of the supercoiled helical winding was indirectly computed
from EM-derived D N A geometric parameters.[’41 Resolution like that of the image in Figure 1 allows the direct determination of the W, of the imaged molecules as the Gauss
integral of the complete three-dimensional path of the
chain.[i31In simple cases like that in Figure 1 it is also possible to estimate W, directly from the image. For this molecule deposited on mica the contribution from planar looped
pieces is zero. Because of the tight interwinding of the chain
in the central part of the molecule, the number of negative[’]
-
crossings is two for almost all projections. W, of the molecule
in Figure 1 is therefore very close to -2.
3. The partition of ALk between the changes in the supercoiling state (W,) and the change in the twist of the D N A
strand (T,) can now be experimentally determined too. In
fact, the measurement of W, in plasmid samples of defined
AL, yields ATw.
4. SFM makes it possible to study the spatial architecture,
intertwining, and coiling of polymers other than DNA.
In conclusion, images like those in Figure 1 demonstrate
that it is now possible to monitor local modulations in the
supercoiling of single D N A molecules directly and to fully
characterize their topology.
Experimental Procedure
10 pLofasolutionofpBR322pIasmidDNA(lO p g m L - ’ ) i n T E buffer(l0mM
Tris. 1 mM EDTA, pH 8.0) was deposited on the surface of freshly cleaved mica.
After 2 min the excess solution was blotted offwith filter paper. The sample was
immediately rinsed with 15 NL of water and then dried for 5 min under a flow
of nitrogen. The sample was imaged with the NANOSCOPE I11 AFM (Digital
Instruments) under 1-propanol with silicon nitride tips (”nanoprobes”, Digital
Instruments). PBR322 plasmid DNA was obtained from transformed DH5
E. coli cells. Large-scale plasmid purification was performed using a method
employing thermal lysis, followed by two cycles of density-gradient centrifugation with ethidium bromide/CsCl toequilibrium [15] The sample showed three
bands in its agarose gel electrophoresis pattern: the fastest has been attributed
to supercoiled forms. the second to nicked and relaxed, and the slowest to high
molecular weight DNA.
Received: March 28. 1993 [ Z 5957 IE]
German version: AnRew. Chem. 1993, 105. 1482
[I] J. Frommer. Angew. Chmn. 1992. 104, 1325; Angew. Chum. Int. Ed. Engl.
1992.3l. 1298 - 1328.
[2] H. G. Hansma, J. Vesenka, C. Siegerist. G. Kelderman. H. Morrett, R. L.
Sinsheimer. V. Elings. C. Bustamante, P. K. Hansma, Science 1992, 256,
1180- 1184.
131 A. Engel, Annu. Rev. Biophys. Chrm. 1991, 20, 79-108.
(41 H. G. Hansma. M. BezdnilkI. F. Zenhausern, M. Adrian, R. L. Sinsheimer, Nuclric Acids Res., 1993, 21, 505-512.
151 F. Zenhausern, M. Adrian. B. ten Heggeler-Bordier. R. Emch, M. Jobin.
M. Taborelli. P. Descouts. J. Slrur.1. B i d 1992. 108, 69-73.
[6] B. Samori. G. Siligardi, C. Quagliariello, A. L. Weisenhorn, J. Vesenka,
C. J. Bustamante, Proc. Null. Acad. Sci. U S A 1993, 90, 3598-3601.
171 W. R. Bauer. F. H. C. Crick. J. H. White, Scr. A m . 1980. 243, 100-1 13.
[8]The three-dimensional resolution of the complete path of very tightly
intertwined chains appeared to be hardly achievable with the commercial
tips we used in this investigation. This is likely to be due both to the
decrease of the superhelix radius with the increase of the supercoiling [14]
and the tip’s shape and profile. In fact, in S F M images the horizontal
resolution depends on the end radius of curvature of the tip (M. J. Allen,
N. V. Hud, M. Balooch. R. J. Tench. W. J. Siekhaus. R. Balhorn, Ultrumiw m w p 1992. 42-44, 1095; J. Vesenka. M . Guthold, C. L. Tang, D.
Keller. E. Delaine. C. Bustamante, hid. 1992. 42-44, 1243). whereas the
vertical resolution, in other words, how well steep features are visualized,
is determined by the opening angle of the tip (D. Keller, D. Deputy, A.
Alduino. K. Luo. U/trumicro.scopr 1992, 42 44, 1481). A chain crossing
profile can thus be resolved in three dimensions only if its opening angle
is greater than that ofthe tip. The unusually large width and double-traced
path of the DNA chain in Figure 1 suggest that this image was traced by
a probe with two very sharp and close tips at its end. Under these conditions unusually high vertical resolutions are frequently obtained a h imaging straight pieces of D N A .
[Y] L. F. Liu. J. C. Wang. Proc. Null. Acud. Sci. U S A 1987, 84, 7024- 7027.
[lo] M . S. Lee. T. W. Garrard, Proc. Nurl. Acud. Sci. USA 1991. 88. 9675- 9679.
[ l l ] M D. Frank-Kamenetskii in D N A Tupulogy und its Biu/ogicu/ E/ficts
(Eds.: N. R. Cozzarelli, J. C. Wang). Cold Spring Harbor Laboratory
Press, New York, 1990, pp. 185-215.
[12] H. Zhang, C. B. Jessee, L. F. Liu. Proc. Null. Acad. Sci. U S A 1990, 87.
9078-9082.
[13] N. R. Couarelli, T. C . Boles, J. H. White in D N A Topotugy und its 5i010gM U / E/fe<./s(Eds.: N. R. Cozzarelli, J. C. Wang), Cold Spring Harbor Laboratory Press, New York, 1990, pp. 139 - 184.
[14] T. C Boles. J. H. White, N. R. Cozzarelli. J. Mol. Biol. 1990,213,931-951.
[15] E F. Fritsch. J. Sambrook, T. Maniatis. Moleculur Cloning. u Luhorutory
Munirul. 2nd edition. Cold Spring Harbor Laboratory Press, New York,
1989. Vol. 1. pp. 34-35 and 42-43.
~
the ideal case a one-dimensional infinite helical complex, in
two steps: an initial formation of a helical subunit, which is
followed by an association to give an extended helical structure. We report here the realization of this strategy to give a
one-dimensional infinite double-helical copper(1) complex.
Some years ago we reported the synthesis and structure of
the simple double-helical complex [Cu2(l),12' (3).['] A feature of 3, which is also found in many other helical complexes,
is the intramolecular stacking between the aromatic planes
of the heterocyclic ligands. In order to study the importance
of this stacking we have prepared the ligand 2,6-bis( 1methyl-imidazol-2-y1)pyridine (2). in which molecular models suggested that the intramolecular stacking interactions
would be greatly reduced.
I
Structure of a One-Dimensional Infinite
Double-Helical Copper(1) Complex**
By Riccardo E Carina, Gerald Bernardinelli,
and Alan E Williams*
Dedicuied to Professor Erwin Parthe
on the occasion of his 65th birthday
Supramolecular chemistry, the chemistry of complicated
structures formed by the controlled association of simple
molecular units, is currently a n area of considerable interest.''] In supramolecular coordination chemistry the structure is assembled by the coordination of suitably structured
ligands about a metal ion, and the assembly may be controlled by the disposition of binding sites within the ligand
structure, and by the stereochemical preferences of the metal
ion.[21The synthesis of helical complexes has recently attracted particular attention : double-helical complexes containing
up to five Cu' ions have been reported by Lehn et aLr31by
using oligobipyridyl ligands; Constable has prepared helical
complexes with a variety of metal ions by using polypyridyl
l i g a n d ~ , ' ~while
'
we have previously reported triple-helical
complexes of six-coordinate C O ' ' ~ ' ~and nine-coordinate
Eu"'.['] A general feature of these complexes is that the
length of the helix is determined by the number of binding
sites of the ligand; thus, a ligand with three bipyridyl binding
sites can bind u p to three cations, and will form a helical,
trinuclear complex. If it is desired to extend the helix, it will
be necessary to design a new ligand with additional binding
sites. An alternative strategy is to form an extended helix, in
[*] Dr. A. F. Williams. Dr. R. F. Carina
Departement de Chimie Minerale Analytique
et Appliquie Universite de Geneve
30 quai Ernest Ansermet, CH 1211 Geneve 4 (Switzerland)
Telefax: Int. code (22)329-6102
Dr. G . Bernardinelli
Laboratoire de Cristallographie aux Rayons-X
Universite de Geneve, CH 121 1 Geneve 4 (Switzerland)
[**I This work was supported by the Swiss National Science Foundation
(Grant no. 20.30129.90)
+
Angat
Clirm. Inr Ed. EngI. 1993. 32, Nu. 10
$3 VCH
2
The X-ray crystal structure of [Cu,(2),](CI04), (4) shows
that the basic [Cu2LZI2+unit found in 3['l is also observed
with ligand 2. The ligands bind in a bis-monodentate coordination mode, and form the strands of the helix that twist
around the helical axis on which the copper ions lie. Each Cu'
center is essentially linearly coordinated by two imidazole
units, and is slightly bent towards the second Cu' center of
the [CU,L,]~+unit; the interaction with the two bridging
pyridines is weak. The asymmetric unit is composed of two
independent half molecules, and a crystallographic twofold
symmetry axis passes through the Cu ions so that the two
ligand strands are symmetry related. Relevant bond lengths
and angles of 4 are given in Table 1, and Figures 1 to 3 show
different views of the complex cations. The Cu-Nimtdazole
dis-
Fig. 1. ORTEP [13] stereoview of two [Cu,(Z),]" units. The crystallographic
twofold axis passes through all four copper ions, and the two molecules of
complex are crystallographically nonequivalent.
tances in 4 are slightly shorter than the Cu-Nbenzimidarole
distances in 3; the Cu -Npyridine
distances, however, are significantly longer in 4 than in 3, although the Cu-Cu distances
(2.821(5), 2.828(5) A) are 0.03A shorter. The dihedral angles
between the pyridine and the imidazole groups, which describe the twisting of the ligand, are in 4 slightly smaller than
the average value for 3, but more significantly, they show
much less variation than for 3 where the variation was from
18 to 44". As expected, the intramolecular stacking of the
Ver/ug.~~~~.~ell.vcliurfl
mhH, 0-69451 Wi~inhPim,1993
OS7U-O833/93//0l(l-1463$ lU.OO+ .25 0
1463
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 043 Кб
Теги
dimensions, supercoiling, image, microscopy, dna, scanning, three, force
1/--страниц
Пожаловаться на содержимое документа