Mechanical Properties of DNA Films S. G. GEVORKIAN and E. E. KHUDAVERDIAN
Yerevan Physics Institute, Markarian St. 2, 375036 Yerevan, Armenia, USSR
SYNOPSIS
The Young’s dynamical modulus ( E ) and the DNA film logarithmic decrement ( 9 ) at frequencies from 50 Hz to 20 kHz are measured. These values are investigated as functions of the degree of hydration and temperature. Isotherms of DNA film hydration at 25OC are measured. The process of film hydration changing with temperature is studied. It is shown that the Young’s modulus for wet DNA films ( E = 0.02-0.025 GN m-’) strongly increases with decreasing hydration and makes E = 0.5-0.7 GN m-’. Dependence of E on hydration is of a complex character. Young’s modulus of denatured DNA films is larger than that of native ones. All peculiarities of changing of E and 9 of native DNA films (observed at variation of hydration) vanish in the case of denatured ones. The native and denatured DNA films isotherms are different and depend on the technique of denaturation. The Young’s modulus of DNA films containing >1 g H 2 0 / g dry DNA is found to decrease with increasing temperature, undergoing a number of step-like changes accompanied by changes in the film hydration. At low water content ( < 0.3 g H,O/g dry DNA), changing of E with increasing temperature takes place smoothly. The denaturation temperature is a function of the water content.
INTRODUCTION The DNA rigidity problem is a n important one in the physics of biopolymers. This parameter is of great importance for quantitative understanding of the microscopic properties of DNA, its packing in chromosomes and virus particles, and properties of circular closed and superhelical DNAs.’-~A number of experimental and theoretical works4-13 are devoted to the study of this problem. In DNA rigidity calculations based on experimental data, the rigid stick model is used and, consequently, the problem of persistent length arises. In the experiments on the study of mechanical properties of DNA in solution, one of the conformations of DNA, the Bform, is mainly observed. I t is also known, that in the solid phase (films, oriented fibers), depending on the environment (hydration, temperature), DNA may be in various conformational states. The works using the DNA microwave absorption are of special importance in the experiments on the study of DNA
0 1990 J o h n U’iley & Sons, Inc. CCC OOOS-3525/90/3-40279-07 $04.00 Biopolymers, Vol. 30, 279-285 (1990)
rigidity.14,15In one of the recent works of this series, l5 the author spoke of a frequency dependence of the mechanical properties, stating that it was “paradoxical” that a t high frequencies water acquired properties of solids. But is it solely water that changes its properties a t high frequencies? We think it is not. Frequency dependence of the mechanical properties of many polymers has been studied in detail.16 This phenomenon is also studied in case of biopolymers, e.g., protein^.'^ One could suppose there was such dependence for DNA. T h e purpose of this paper is to study the DNA viscoelastic properties in the frequency range from 50 Hz to 20 kHz. A t such frequencies free water can have no “paradoxical” mechanical properties. As the DNA films are compounds consisting of DNA and water, we have also investigated their hydration isotherms.
MATERIALS A N D METHODS T o measure the Young’s dynamical modulus E and the logarithmic decrement 8 of DNA films, V. N. Morozov’s micromethod was used.l8,l9The method 279
280
GEVORKIAN AND KHUDAVERDIAN
is based on the analysis of resonance transverse vibrations of plates supported as cantilevers. A method and an experimental chamber for measuring E and 6 as functions of temperature are described in Ref. 17. A hydration isotherm measuring method allowing the study of microobjects weighing up to 0.02 mg is described in Ref. 20. Amorphous films of cattle spleen and calf thymus DNAs were investigated. The films were prepared by slow drying of DNA saturated water solution on a teflon sheet at 10°C. Films 5- to 20-pm thick were obtained that were cut to form 0.5-2.0-mm long and 0.05-0.1-mm wide rectangular plates. Samples of that size were supported as cantilevers, and the resonance frequencies of their transverse vibrations were in the range from 50 Hz to 20 kHz. One can change the resonance frequency by changing the sample's length. These procedures are comprehensively described in Ref. 17. The relative humidity from 97 to 32% was supported by CaC12 water solution of different concentrations. The relative humidities of 15 and 10% were obtained with the help of LiCl and ZnC12 saturated water solutions, respectively. For heating, hot air was blown into the chamber radiator. The heating rate was 1deglmin. The chamber temperature was measured by a copper-constantan thermocouple. The measurement accuracy was 0.1OC.
FREQUENCY DEPENDENCE OF E AND 0 Dependencies of E and 6 on frequency were measured. The frequency was changed by fractional
cuI
i
Ir-
100 60 20 RELATIVE HUMIDITY A.% Figure 2 The Young's modulus E (- 0 - 0 -) and the logarithmic decrement I9 ( * 0 0 ) of calf thymus DNA films as functions of relative humidity, A%.
- -- --
shortening of the samples. Thus we succeeded in changing the transverse vibrations resonance frequency within 50 Hz and 20 kHz. The experiments have shown that it is only a t high humidities ( A 85%) that E increases with the frequency. The lower the humidity, the weaker the frequency dependence. At frequencies lower than 200 Hz, a t A 3 85%, a strong decrease of E and an increase in 6 can be observed. (See Figure 1.) All the further investigations were carried out a t frequencies > 200 Hz,where E depends on the frequency weakly.
DEPENDENCE OF E AND 0 ON HYDRATION
i
U
u-3
N
0
4
6, 8
10
Figure 1 E and I9 as functions of frequency at different humidities. ( a ) - 0 - 0 -, E a t A = 85%;- - 0 - - 0 - -I9,at A = 85%. (b) - V - V -, E and I9 at A = 50% (the horizontal curve is E) .
In Figure 2 the Young's modulus and the logarithmic decrement of an amorphous film of calf thymus DNA are shown as functions of relative humidity at 25"C, A = 95%, E = 0.02 GN m-2. For comparison note that for protein amorphous films E = 1GN m-' ( a t the same relative humidity) ,I9 i.e., two orders of magnitude higher than it is for DNA. As can be seen from Figure 2, the Young's modulus essentially increases with decreasing relative humidity. This increase takes place through a number of transitions followed by plateaus. According to the literature, in the solid state, a t A 3 95%, the B form of DNA is observed. The B-A transition region is not a monotonic one and, probably, this transition does not take place smoothly. The region of A < 60% is also char-
MECHANICAL PROPERTIES OF DNA FILMS
acterized by a monotonic behavior of E. This is the case with all the samples (8experiments). The logarithmic decrement, which characterizes the internal friction in ~ a m p l e s ' ~at , ' ~the sites of these transitions, has marked extrema. E and 3 of cattle spleen DNA films were also investigated (Figure 3 ) . It can be seen that here too the rigidity strongly increases with decreasing hydration. Despite the fact that at high humidity (A = 90%) these films are somewhat softer than the preceding ones, at low humidity ( A = 30%) they become more rigid. The monotonic behavior of these samples is somewhat different from the preceding ones. The reversibility of dehydration was also checked (Figure 4 ) . In the counterprocess (hydration) a hysteresis can be observed- E is larger than it is at dehydration, but the peculiarities of a direct process recur.
EFFECT OF THERMAL DENATURATION As is known, the DNA helicity stability is specified by different intra- and intermolecular interactions. These interactions are affected by hydration and temperature." We have also studied the effect of denaturation on the viscoelastic properties of DNA films. As hydration of DNA films depends on the temperature, salt content, and molecular ratio of anions to cations, we have also investigated the DNA films hydration isotherms. We have earlier2' developed a micromethod al-
5
I
1
100 60 20 RELATIVE HUMIDITY A% Figure 3 E ( - - 0 - - 0 - -and ) 9 ( - -0. .O- - ) of cattle spleen DNA films as functions of humidity.
QJ
281
Gr
'zJ i. U u-3 cv
-
0
I00 60 20 RELATIVE HUMIDITY A,% Figure 4 A cattle spleen DNA film: reversibility of changes in the Young's modulus E (- 0 - 0 -) dehydration -0. - ) hydration. and ( - -0-
lowing the study of the hydration of samples weighing from 0.1 to 0.01 mg with an accuracy of 0.1%. The method is based on the analysis of the transverse resonance vibrations of a microrod supported as a cantilever; the sample is fastened to its free end. Bronze rods with rectangular cross section were used in this work. Such rods, of 10 pm constant thickness, 200 pm width, and 1-2 mm length, have 1-2 kHz intrinsic resonance frequency of transverse vibrations. Test DNA films were prepared directly on the free end of the rod by drying the DNA-saturated solution on it. Isotherms of hydration of native DNAs denatured in solution and in solid phase were studied. For denaturation in solution the DNA water solution was boiled for 30 minutes. For denaturation in solid phase, the native DNA film was kept at t = 150°C for 30 min. All the experiments were started at high humidity, A = 95%,which then was gradually lowered. For complete removal of water ( A = 0 % ) the sample was heated up to 15OOC and kept at that temperature for one hour. The water c o n t p t estimate was made relative to the point of A = 0. The counterprocess (from A = 0% to A = 95%) was not studied, because a t that temperature the sample was denatured. Figure 5 shows the isotherms of hydration of cattle spleen DNA films. As it is seen, the isotherm of hydration of DNA denatured in solution strongly differs from the other two ones. It can be explained by the fact that in solution DNAs are completely denatured as they roll up into coils and this, in its turn, leads to their tighter packing up in
282
GEVORKIAN AND KHUDAVERDIAN
(v
‘c5
3
r- I L
m LlJ I---
u-
Q6
I
+ 0.2 1c
Yerevan Physics Institute, Markarian St. 2, 375036 Yerevan, Armenia, USSR
SYNOPSIS
The Young’s dynamical modulus ( E ) and the DNA film logarithmic decrement ( 9 ) at frequencies from 50 Hz to 20 kHz are measured. These values are investigated as functions of the degree of hydration and temperature. Isotherms of DNA film hydration at 25OC are measured. The process of film hydration changing with temperature is studied. It is shown that the Young’s modulus for wet DNA films ( E = 0.02-0.025 GN m-’) strongly increases with decreasing hydration and makes E = 0.5-0.7 GN m-’. Dependence of E on hydration is of a complex character. Young’s modulus of denatured DNA films is larger than that of native ones. All peculiarities of changing of E and 9 of native DNA films (observed at variation of hydration) vanish in the case of denatured ones. The native and denatured DNA films isotherms are different and depend on the technique of denaturation. The Young’s modulus of DNA films containing >1 g H 2 0 / g dry DNA is found to decrease with increasing temperature, undergoing a number of step-like changes accompanied by changes in the film hydration. At low water content ( < 0.3 g H,O/g dry DNA), changing of E with increasing temperature takes place smoothly. The denaturation temperature is a function of the water content.
INTRODUCTION The DNA rigidity problem is a n important one in the physics of biopolymers. This parameter is of great importance for quantitative understanding of the microscopic properties of DNA, its packing in chromosomes and virus particles, and properties of circular closed and superhelical DNAs.’-~A number of experimental and theoretical works4-13 are devoted to the study of this problem. In DNA rigidity calculations based on experimental data, the rigid stick model is used and, consequently, the problem of persistent length arises. In the experiments on the study of mechanical properties of DNA in solution, one of the conformations of DNA, the Bform, is mainly observed. I t is also known, that in the solid phase (films, oriented fibers), depending on the environment (hydration, temperature), DNA may be in various conformational states. The works using the DNA microwave absorption are of special importance in the experiments on the study of DNA
0 1990 J o h n U’iley & Sons, Inc. CCC OOOS-3525/90/3-40279-07 $04.00 Biopolymers, Vol. 30, 279-285 (1990)
rigidity.14,15In one of the recent works of this series, l5 the author spoke of a frequency dependence of the mechanical properties, stating that it was “paradoxical” that a t high frequencies water acquired properties of solids. But is it solely water that changes its properties a t high frequencies? We think it is not. Frequency dependence of the mechanical properties of many polymers has been studied in detail.16 This phenomenon is also studied in case of biopolymers, e.g., protein^.'^ One could suppose there was such dependence for DNA. T h e purpose of this paper is to study the DNA viscoelastic properties in the frequency range from 50 Hz to 20 kHz. A t such frequencies free water can have no “paradoxical” mechanical properties. As the DNA films are compounds consisting of DNA and water, we have also investigated their hydration isotherms.
MATERIALS A N D METHODS T o measure the Young’s dynamical modulus E and the logarithmic decrement 8 of DNA films, V. N. Morozov’s micromethod was used.l8,l9The method 279
280
GEVORKIAN AND KHUDAVERDIAN
is based on the analysis of resonance transverse vibrations of plates supported as cantilevers. A method and an experimental chamber for measuring E and 6 as functions of temperature are described in Ref. 17. A hydration isotherm measuring method allowing the study of microobjects weighing up to 0.02 mg is described in Ref. 20. Amorphous films of cattle spleen and calf thymus DNAs were investigated. The films were prepared by slow drying of DNA saturated water solution on a teflon sheet at 10°C. Films 5- to 20-pm thick were obtained that were cut to form 0.5-2.0-mm long and 0.05-0.1-mm wide rectangular plates. Samples of that size were supported as cantilevers, and the resonance frequencies of their transverse vibrations were in the range from 50 Hz to 20 kHz. One can change the resonance frequency by changing the sample's length. These procedures are comprehensively described in Ref. 17. The relative humidity from 97 to 32% was supported by CaC12 water solution of different concentrations. The relative humidities of 15 and 10% were obtained with the help of LiCl and ZnC12 saturated water solutions, respectively. For heating, hot air was blown into the chamber radiator. The heating rate was 1deglmin. The chamber temperature was measured by a copper-constantan thermocouple. The measurement accuracy was 0.1OC.
FREQUENCY DEPENDENCE OF E AND 0 Dependencies of E and 6 on frequency were measured. The frequency was changed by fractional
cuI
i
Ir-
100 60 20 RELATIVE HUMIDITY A.% Figure 2 The Young's modulus E (- 0 - 0 -) and the logarithmic decrement I9 ( * 0 0 ) of calf thymus DNA films as functions of relative humidity, A%.
- -- --
shortening of the samples. Thus we succeeded in changing the transverse vibrations resonance frequency within 50 Hz and 20 kHz. The experiments have shown that it is only a t high humidities ( A 85%) that E increases with the frequency. The lower the humidity, the weaker the frequency dependence. At frequencies lower than 200 Hz, a t A 3 85%, a strong decrease of E and an increase in 6 can be observed. (See Figure 1.) All the further investigations were carried out a t frequencies > 200 Hz,where E depends on the frequency weakly.
DEPENDENCE OF E AND 0 ON HYDRATION
i
U
u-3
N
0
4
6, 8
10
Figure 1 E and I9 as functions of frequency at different humidities. ( a ) - 0 - 0 -, E a t A = 85%;- - 0 - - 0 - -I9,at A = 85%. (b) - V - V -, E and I9 at A = 50% (the horizontal curve is E) .
In Figure 2 the Young's modulus and the logarithmic decrement of an amorphous film of calf thymus DNA are shown as functions of relative humidity at 25"C, A = 95%, E = 0.02 GN m-2. For comparison note that for protein amorphous films E = 1GN m-' ( a t the same relative humidity) ,I9 i.e., two orders of magnitude higher than it is for DNA. As can be seen from Figure 2, the Young's modulus essentially increases with decreasing relative humidity. This increase takes place through a number of transitions followed by plateaus. According to the literature, in the solid state, a t A 3 95%, the B form of DNA is observed. The B-A transition region is not a monotonic one and, probably, this transition does not take place smoothly. The region of A < 60% is also char-
MECHANICAL PROPERTIES OF DNA FILMS
acterized by a monotonic behavior of E. This is the case with all the samples (8experiments). The logarithmic decrement, which characterizes the internal friction in ~ a m p l e s ' ~at , ' ~the sites of these transitions, has marked extrema. E and 3 of cattle spleen DNA films were also investigated (Figure 3 ) . It can be seen that here too the rigidity strongly increases with decreasing hydration. Despite the fact that at high humidity (A = 90%) these films are somewhat softer than the preceding ones, at low humidity ( A = 30%) they become more rigid. The monotonic behavior of these samples is somewhat different from the preceding ones. The reversibility of dehydration was also checked (Figure 4 ) . In the counterprocess (hydration) a hysteresis can be observed- E is larger than it is at dehydration, but the peculiarities of a direct process recur.
EFFECT OF THERMAL DENATURATION As is known, the DNA helicity stability is specified by different intra- and intermolecular interactions. These interactions are affected by hydration and temperature." We have also studied the effect of denaturation on the viscoelastic properties of DNA films. As hydration of DNA films depends on the temperature, salt content, and molecular ratio of anions to cations, we have also investigated the DNA films hydration isotherms. We have earlier2' developed a micromethod al-
5
I
1
100 60 20 RELATIVE HUMIDITY A% Figure 3 E ( - - 0 - - 0 - -and ) 9 ( - -0. .O- - ) of cattle spleen DNA films as functions of humidity.
QJ
281
Gr
'zJ i. U u-3 cv
-
0
I00 60 20 RELATIVE HUMIDITY A,% Figure 4 A cattle spleen DNA film: reversibility of changes in the Young's modulus E (- 0 - 0 -) dehydration -0. - ) hydration. and ( - -0-
lowing the study of the hydration of samples weighing from 0.1 to 0.01 mg with an accuracy of 0.1%. The method is based on the analysis of the transverse resonance vibrations of a microrod supported as a cantilever; the sample is fastened to its free end. Bronze rods with rectangular cross section were used in this work. Such rods, of 10 pm constant thickness, 200 pm width, and 1-2 mm length, have 1-2 kHz intrinsic resonance frequency of transverse vibrations. Test DNA films were prepared directly on the free end of the rod by drying the DNA-saturated solution on it. Isotherms of hydration of native DNAs denatured in solution and in solid phase were studied. For denaturation in solution the DNA water solution was boiled for 30 minutes. For denaturation in solid phase, the native DNA film was kept at t = 150°C for 30 min. All the experiments were started at high humidity, A = 95%,which then was gradually lowered. For complete removal of water ( A = 0 % ) the sample was heated up to 15OOC and kept at that temperature for one hour. The water c o n t p t estimate was made relative to the point of A = 0. The counterprocess (from A = 0% to A = 95%) was not studied, because a t that temperature the sample was denatured. Figure 5 shows the isotherms of hydration of cattle spleen DNA films. As it is seen, the isotherm of hydration of DNA denatured in solution strongly differs from the other two ones. It can be explained by the fact that in solution DNAs are completely denatured as they roll up into coils and this, in its turn, leads to their tighter packing up in
282
GEVORKIAN AND KHUDAVERDIAN
(v
‘c5
3
r- I L
m LlJ I---
u-
Q6
I
+ 0.2 1c
