The Journal of
Physical Chemistry
0 Copyright, 1992, by the American Chemical Society
VOLUME 96, NUMBER 25 DECEMBER 10,1992
LETTERS Measurement of Bound Water in an Aqueous DNA Solution Using Nuclear Magnetic Resonance and lime Domain Reflectometry Minoru Fukuzaki,* Toshihiro Umebara, Daisaku Kurita, Sumie Shioya, Munetaka Haida, and Satoru Mashimo NMR Laboratory, Electro- Chemical and Cancer Institute, Chofu, Tokyo 182; Department of Physics, Tokai University, Hiratsuka, Kanagawa 259-1 2; and Department of Internal Medicine, Tokai University School of Medicine, Isehara. Kanagawa 259- I I , Japan (Received: June 29, 1992; In Final Form: October 14, 1992)
Nuclear magnetic resonance (NMR) and time domain reflectometry (TDR) were employed to investigate the bound water of deoxyribonucleicacid (DNA) in aqueous solution. The NMR transverse magnetization of the water protons decays in the manner described by a monoexponential function which gives a single magnetic relaxation time (T2).However, this fact does not mean that the correlation times for rotational reorientation of both bound and free water molecules are identical. On the other hand, two different dielectric relaxation times ( r d )associated with bound and free water molecules were observed in the same solution. Substituting the dielectric relaxation times into the fast exchange model for the NMR relaxation, we could evaluate the amount of bound water as 0.3 g of water/g of DNA.
Biological water-such as bound waterplays a significant role in maintaining the structure of macromolecules in a variety of biological systems.'.2 It is known that the amount and stability of bound water are affected by physiological changes in cell or tissue systems. These changes could be detected by NMR relaxation time or TDR relaxation time measurements.'J With NMR relaxation, fast exchange between bound and free water typically prevails: and the NMR relaxation time is a weightedaverage of the relaxation times of the bound and free water molecula. On the other hand, two relaxation peaks,due to bound and free water, respectively, will be observed in the dielectric relaxation measurement. We obtained 1-ns and 10-ps dielectric *Towhom correspondence should be addresred at the Electro-Chemical and Cancer Institute. 0022-3654/92/2096-10087$03.00/0
relaxation times associated with bound and free water, respecti~ely.~ The purpose of this paper is to illustrate the quantitative interpretation of NMR relaxation time in terms of the amount of bound water, using the dielectric relaxation times obtained for DNA in an aqueous solution. Experimental Section DNA from Salmon testes was purchased from Sigma Chemical Co. Ltd.(D-1626). The DNA was digsolved in 10 mM Tris buffer (pH 7.5) with 1 mM EDTA and adjusted to 10% DNA concentration. We used a 'H 90 MHz, FT-NMR, EX-90(JEOL)instrument for measurements of the water proton relaxation time T2. Hahn's spin echo pulse sequence (SE,9Oo7-18O01-) was employed to measure the water proton magnetization decay.6 The T2value Q 1992 American Chemical Society
10088 The Journal of Physical Chemistry, Vol. 96, No. 25, 1992 TABLE I: Experimental NMR Relaxation Time
Letters
T2and Dielectric Rclwtim P a "
from TDR TDR
temp, OC 0 25
T2, 72.3 & 1.8 152.7 f 1.9
rdb, ns 1.49 & 0.055 0.89 f 0.033
Acb 19.15 & 0.27 18.25 & 0.26
A% 72.92 & 0.00 64.10 & 0.00
?dh,
15.5 f 0.098 8.66 f 0.054
'Mean & SD.
80 w
60
b
0
500
2.r
I I
I
I
I I
7
8
9
10
I
1000
(msec)
Figure 1. Logarithmic plot of the magnetization M a s a function of 27, at 0 and 25 "C,using Hahn's spin-echo method. is calculated from the slope of a logarithmic plot of the magnetization M as a function of 27, with M given by the following equatiod W27) = Mo[exp(-(27/T2) - Df)l
(1)
where Dfis given by
6 Df = 2/3y2G2Dr3
(2)
Mois the magnetization intensity at 7 = 0; y is the gyromagnetic ratio; G is the magnetic field gradient due to an inhomogeneous field; D is the self diffusion constant. The TDR method' was used to measure the dielectric relaxation spectrum for the DNA solution. The time domain reflected signal following a step impulse voltage was accumulated in a digital oscilloscope (Hewlett-Packard, HP5412) and was Fourier transformed (FT)for analysis to a frequency domain ranging from lo6 to 10'O Hz. The complex permittivity (e*) obtained from this analysis is given by the following equation:*v9 (3)
where 7 b and 7 h are the dielectric relaxation times for bound and free water, respectively; Atb and ACh are the corresponding relaxation peak intensities; w is the angular frequency; em is the high-frequency dielectric constant. The subscripts b and h denote bound and free water, respectively. The relaxation parameters in eq 3 were calculated by the least-squares curve fitting. The sample temperatures were fixed at 0 and 25 "Cfor both the NMR and the TDR experiments. Figure 1 shows the logarithmic plot of the water proton magnetization decay of the DNA solution at 0 and 25 "C. The magnetization was assumed to a monoexponential function as shown in eq 1. Figure 2 shows the dielectric dispersion and absorption curves of the same solution at 0 and 25 O C . The values of NMR relaxation time T2and the dielectric parameters, rd and Ae, are summarized in Table I. WuMiOa
In general, the relaxation time T2of two protons at distance
log f ( Hz) Figure 2. Dielectric dispersion and absorption curves at 0 and 25 "C. TABLE II: CompuisOn of the R o t a t i d CMeLtion Time (2,) O b M Directly from T3and tbe Dkktric ReLxlrtion Timer (2,) temp, "C ns 7db/3? ns ~ & / 3 , 0ns 0 0.196 0.5 0.0052 25 0.0912 0.30 0.0029
"The dielectric relaxation time was divided by 3 for comparison with the NMR data. The subscripts b and h denote bound and free water, respectively.
r experiencing magnetic dipoledipole interaction is given by the following equation+ r
where B is 2y4h2Z(Z+ 1) B=
80dP
(5)
w is the Larmor frequency; rc is the rotational correlation time of the two rigidly bound protons; y is the gyromagnetic ratio of
the proton; h is Planck's constant; I is the nuclear spin; r is the distance between dipolarly coupled protons. Since only the rotational motion is allowed for the water molecules bound to the DNA,rCrepresents its rotational correlation time. Intermolecular interaction is also modulated by translational diffusion of the free water. If the interaction for the translational motions is similar to that for the rotational motions, both types of interactions are represented as the same form of relaxation.lo Because the strength of dipole-dipole interaction is inversely proportional to # as shown in eq 5, the
J . Phys. Chem. 1992, 96, 10089-10092 TABLE JIk Molar Fraction Fb of B o d Water and the NMR Relaxation Parameters (Cdculated As Explained in tbe Text)' OC
ob
T2, bound
free
Fb
0 25
72.3 152.7
30.51 48.41
2685 4806
0.42 0.31
temp,
g of
Bw,
water/g of DNA 0.37 0.28
'T2(ob): observed NMR relaxation time. T2(bound): calculated value for bound water using Tdb. T2(free): calculated value for free water using f h Fb: molar fraction of bound water. Bw: the relative amount of bound water. interaction between two nearest protons prevails in the free water. Therefore, the rotational correlation time stands for the correlation time of the dipoledipole interaction of free water. If we use r = 1.51 X lo-*cm' 1*12 for the interproton distance in HzO, we calculate the rotational correlation times T, listed in Table 11. The values of T , obtained from eq 4 using the experimental value of T2and the dielectric relaxation times obtained from the TDR measurements are given in Table 11. Since the rotational correlation time is theoretically related to the dielectric relaxation time:" we can compare the T, values obtained from the NMR with that obtained from the TDR. As shown in Table 11, this value T, is between the Td/3 values for bound and free water. This suggests that fast exchange between bound and free water takes place on the NMR time scale. In this case, the NMR relaxation time T2 is described as f01lows:~ (7)
10089
where T2 is the observed relaxation time, T2band TZ are NMR relaxation times of bound and free water, respectively, and Fb is the molar fraction of bound water. The relaxation times T z and ~ T2fcan be calculated by sub/ 3 in Table I1 into T, in eq 4. Thus, the stituting the ~ ~ values molar fraction of bound water is calculated from eq 7 and are given in Table 111. Our results are comparable to the solvation sphere required to maintain the B form of DNA in solution.2" Our results demonstrate that the exchange between bound and free water in the DNA solution takes place as fast as experimental NMR time scales,and therefore the observed relaxation time T2 is a weighted average. On the other hand, the rate of this proceas is slow on the TDR time scale, and therefore separate measurements associated with bound and free water are possible.
References and Notes (1) Kuntz, I. D., Jr.; Kauzman, W. Adv. Protein Chem. 1974, 28, 239. (2) Umehara, T.; Kuwabara. S.;Mashimo, S.;Yagihara, S . Biopolymers 1990,30,649.
(3) Mashimo, S.; Ota, T.; Shinyashiki, N.; Tanaka, S.;Yagihara, S. Macromolecules 1989, 22, 1285. (4) Woessner, D. E.; Zimmerman, J. R. J . Phys. Chem. 1%3,67, 1590. (5) Mashimo,S.; Umehara, T.; Kuwabara, S.;Yagihara, S.J. Phys. Chem. 1989, 93, 4963. (6) Farrar, T. C.; Becker, E. D.Pulse and Fourier Transform N M R Academc Press: New York, 1971; Chapters 2 and 4. (7) Mashimo, S.; Umehara, T.; Ota, T.; Kuwabara, S.;Shinyashiki, N.; Yagihara, S.J. Mol. Liq. 1987, 36, 135. (8) Cole, K. S.; Cole, R. H. J . Chem. Phys. 1941, 9, 341. (9) Davidson, D. W.; Cole, R. H. J . Chem. Phys. 1951, 19, 1484. (10) Lynch, L. J. Magnetic Resonance in Biology; Cohen, J. S., Ed.; John Wiley & Sons: New York, 1983; Vol. 2, Chapter 5. (1 1) Blinc, A.; Lahjnar, G.; Blinc, R.; Zidansek, A.; Sepe, A. Magn. Reson. Med. 1990, 14, 105. (12) Bloembergen, N.Nuclear Magnetic Relaxation; W. A. Benjamin: New York, 1961; Chapter 4.
Controlled Nanofabrication of Highly Oriented Pyrolytic Graphite with the Scanning Tunneling Microscope Robin L. McCarleyJ Samuel A. Hendricks, and Allen J. Bard* Department of Chemistry and Biochemistry, The University of Texas at Austin, Austin, Texas 78712 (Received: July 13, 1992; In Final Form: October 8, 1992)
The formation and selective etching of recessed features of various shapes in highly oriented pyrolytic graphite (HOPG) in air were accomplished by use of the scanning tunneling microscope with positive substrate bias voltages of +1500 to +3000 mV and tunneling currents of less than 2 nA. Etching of the surface was restricted to the scan area and only occurred with positive biases, which suggests that etching occurs by an electrochemical mode of oxidation of the graphite under the tip. Lines with widths as small as 10 nm and squares 25 X 25 nm could be formed with monolayer depth (0.34 nm) in the HOPG by varying the time that the tip was rastered over a designated area.
Introductioa We describe here a highly controlled method for the fabrication of nanometer size structures on highly oriented pyrolytic graphite (HOPG) using the scanning tunneling microscope (STM).' Recessed lines, squares, and disks could be formed on the basal plane of HOPG by using the STM to etch the surface. Bias voltages (vb)of +2800 mV (substrate vs tip) with tunneling currents (it)of 0.06-2 nA caused layer-by-layer removal of the carbon atoms directly under the tip. Selective etching of existing defects (exposed edgeplane graphite) occurred at bias potentials greater than or equal to +1500 mV. The rate of etching depended upon the magnitude of the bias but was controlled best with +1500 Prcscnt address: Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803.
mV I vb 5 +2400 mV. Negative substrate biases of the same magnitude did not cause etching of the HOPG. Low biases (5k500 mV) and high tunneling currents (1-4 nA), which bring the tip closer to the HOPG, did not cause removal of material, indicating that etching is not due to physical interaction of the tip with the HOPG. Thew results suggest that the etching proass is probably electrochemical in nature. The advent of the STM has led to many advances in the -C ' ticm of conducting and semiconducting surfaces. There have also been several applications to "nanofabrication" of surfacesa2 Such studies are motivated by a desire to understand different approaches to the controlled modification of surfaces at the atomic level with ultimate applications to devices or structures like masks for X-ray lithography. In these the tip voltage is usually pulsed to higher bias voltages to remove or place
0022-3654/92/2096- 10089$03.00/0 Q 1992 American Chemical Society
