Volume Phase Transition of Bovine Vitreous Body in Vitro and

Biomacromolecules 2004, 5, 1296-1302

1296

Volume Phase Transition of Bovine Vitreous Body in Vitro and Determination of Its Dynamics Toyoaki Matsuura,† Yoshiaki Hara,† Futoshi Taketani,† Eiichi Yukawa,† Shinzi Maruoka,† Kensuke Kawasaki,† and Masahiko Annaka*,‡ Department of Ophthalmology, Nara Medical University, 840 Shijyo-cho, Kashihara-shi, Nara 634-8522, Japan, and Department of Chemistry, Kyushu University, 6-10-1, Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan Received January 17, 2004; Revised Manuscript Received March 22, 2004

The phase equilibrium property and structural and dynamical properties of bovine vitreous body was studied by macroscopic observation of swelling behavior and dynamic light scattering under various conditions. It was found that the vitreous body collapses into a compact state isotropically or anisotropically depending on the external conditions. The vitreous body collapses while maintaining the shape when the pH (e 4) and the concentration of calcium ions (e 1 mol/L) are changed, whereas it collapses along the orbital axis in a mixed solvent of methanol and water. From observations of the dynamics of light scattered by the vitreous body, intensity autocorrelation functions that revealed two independent diffusion coefficients, D(fast) ) 7.8 ( 1.5 × 10 - 8 cm2/s and D(slow) ) 3.8 ( 0.60 × 10 -9 cm2/s, were obtained. The diffusion coefficients were found to be statistically independent of position within a focal depth range of 1-1.5 mm. Divergent behavior in the measured total scattered light intensities and diffusion coefficients was observed as the concentration of calcium ion approached the critical threshold, 1 mol/L. Namely, a slowing down of the dynamic modes accompanied by increased “static” scattered intensities was observed. The divergent behavior in the scattered light intensities and diffusion coefficients was reversible. This is indicative of the occurrence of a phase transition upon calcium ion concentration. 1. Introduction Phase transition and critical phenomena observed in synthetic polymer gels are considered to be universal to all gels.1,2 Regardless of their physicochemical composition, all polymer networks should then exhibit phase transitions and critical phenomena under appropriate conditions. A simple confirmation of their assertion was demonstrated by including phase transition in gels that were made up of cross-linked biopolymers.3 The “bio-gels” were formed from artificially cross-linked biopolymers, such as DNA, agarose, and gelatin. However, there have been few studies detailing the phase transitions and critical phenomena of gels formed in nature. The vitreous body is a tenuous gel that contains collagen and hyaluronic acid.4 The fraction of the polymer network is only about 1-2%, and the remaining is water. The vitreous body is located between the lens and the retina, and comprises 80% of the overall volume of eye. The functions of the vitreous body are supposed to keep the shape of the eyeball, to absorb the external mechanical shock, to maintain the homoeostasis of the eye, and to regulate the position of the lens. The appearance of fresh vitreous body is transparent, and hence, the vitreous body is considered a uniform tissue. Many studies performed to date have suggested that hyaluronic acid, which has a coil shape, is uniformly distributed * To whom correspondence should be addressed. Fax: +81-92-6422594. E-mail: [email protected]. † Nara Medical University. ‡ Kyushu University.

throughout the three-dimensional network of collagen fibers that form the triple helix in the vitreous body.5 Essentially, no investigations on the structural, dynamics, and phase equilibrium properties of the vitreous body, however, have been performed to verify indisputably that the vitreous body is indeed a gel network. In investigations of the vitreous body, it had often been alluded to as a gel network. If indeed the vitreous body were a gel network, well-established theories would predict that the vitreous body should exhibit phase transition and critical phenomena in response to varied external conditions. The present research places the emphasis on the observation the volume phase transition behaviors, and the measurements of scattered light intensities and dynamics from bovine vitreous body in vitro, subjected to the changes in the external conditions. These studies will promote better understanding of the function and the mechanism of diseases of the vitreous body. 2. Experimental Section 2.1. Materials. Bovine vitreous bodies were isolated from sclera of the eyeball. The choroid membrane was also carefully removed by the standard method.6 The samples were excised within 8 h after extraction of the eye at a local slaughterhouse. The approximate age of the calves was 1 year. 2.2. Swelling Experiments. The swelling ratio of the vitreous body was determined by weighing in a gel in the

10.1021/bm049954y CCC: $27.50 © 2004 American Chemical Society Published on Web 05/04/2004

Bovine Vitreous Body

Biomacromolecules, Vol. 5, No. 4, 2004 1297

Figure 1. Schematic illustration of the microscope laser light scattering spectroscopy (MLLSS).

equilibrium state (Sartorius 2474, Sartorius-Werke Gmbh, Go¨ttingen Germany). Both the weight of the sample immediately after preparation, Wi, and the weight of sample in the equilibrium state, We, were measured. Then the swelling ratio was calculated from the ratio {We}/{Wi}. The solvent attached to the gel was carefully wiped to minimize the error. The effects of pH and calcium ion on the swelling ratio of the vitreous body were studied. The pH was changed using hydrochloric acid (Nacalai tesque, guaranteed reagent) and sodium hydroxide (Nacalai tesque, guaranteed reagent). The concentration of calcium ions was changed from 1.0 × 10-7 to 1 mol/L using calcium chloride (Nacalai tesque, guaranteed reagent). The swelling behavior of bovine vitreous body was also determined to be a function of the concentration of methanol in a methanol-water mixture. The concentration of methanol in the mixed solvent of methanol and water was changed from 0 to 100% (v/v) by mixing methanol (Nacalai tesque, guaranteed reagent) and distilled and deionized water at the desired ratio. Since the size of the bovine vitreous body was of the order of 8 cm3 in volume (n ) 10, SD ) 1.1), the time required to attain the equilibrium state was three weeks. Each solvent of sufficient volume was, therefore, changed every 2 days. Then the equilibrium-swelling ratio was determined. 2.3. Microscopic Laser Light Scattering Spectroscopy (MLLSS). Since the vitreous body is difficult to be held in a sample cell, we used the technique of the microscope laser light scattering spectroscopy (MLLSS).7 The technique differs from the conventional dynamic laser light scattering technique in that the scattering volume in the MLLSS is some 105 times smaller, making it as low as 2 µm3. It, therefore, becomes possible to analyze the motion of particles inside the vitreous body. The schematic diagram of the MLLSS setup is illustrated in Figure 1. The beam of a 30 mW HeNe laser (632.8 nm, Spectra Physics 124A) was focused onto the equational plane of the vitreous body through an optical fiber connected to the fiber coupler that is equipped with a condensing lens. The beam is sharply focused onto the various positions of the vitreous body placed under an upright

microscope (Nikon Optiphot). The cross-section at the focal point within the vitreous body was approximately 2 × 2 µm2. The focal region was imaged onto a photomultiplier tube (EMI 9863B-350) by a 50 µm-diameter optical fiber embedded in one of the eyepieces of the microscope (Gamma Scientific 700-10-36A). The fluctuations of the number of photons were analyzed using the autocorrelator (Brookhaven Instrument, model BI-2030A). The temporal fluctuations of the scattered light intensity I(t) were analyzed in terms of intensity autocorrelation functions8 〈C(τ)〉 ) 〈I(t) I(t + τ)〉t

(1)

where 〈〉 stands for the time average over t. The rate of the fluctuations of the scattered light intensity that represent the density fluctuations of the vitreous gel is proportional to the rate of local swelling and shrinking of the gel via molecular Brownian motions. There are also permanent and static inhomogeneities within the vitreous that also contribute to light scattering. The light intensity scattered by these inhomogeneities does not fluctuate with time. The scattered light intensity is, thus, the superposition of contributions from scattering elements that are static from those that dynamically fluctuate9 I(t) ) IS + ID(t)

(2)

In dynamic light scattering, the time correlation of the intensity of scattered light is recorded. Assuming the Gaussian nature of the scattered light photons, the correlation function of the intensity of scattered light is rewritten in terms of the autocorrelation function g(τ) of the scattered electric field E(t), which is related to the scattered light intensity, I(t) ) E(t) E*(t):8 g(τ) ≡ Then C(τ) is written as

〈E(t) E*(t + τ)〉t 〈E(t) E*(t)〉t

(3)

1298

Biomacromolecules, Vol. 5, No. 4, 2004

C(τ) ) (IS + ID)2 + A{ID2 g2(τ) + 2IS IDg(τ)}

Matsuura et al.

(4)

where A is the efficiency parameter of the apparatus, which is uniquely determined by the optical configuration of the setup, the value I, and the average intensities scattered by the gel fluctuations and the static inhomogeneities.9 For the present experiments, A was determined to be 0.8. The electric field autocorrelation function g(τ) can be easily extracted from the intensity autocorrelation function C(τ) from knowledge of the initial value C(0), the baseline C(∞), and the coefficient A, using the above relations. As we shall see later and as shown in Figure 6, the autocorrelation functions can have a distinct double-exponential feature. This indicates the presence of two different modes within the gel. Therefore, the correlation functions g(τ) are analyzed using the following relationship: g(τ) )

Figure 2. Swelling ratio of bovine vitreous body in solutions varying pH at 23 °C (n ) 3 for each pH).

As Af exp(-Dfq2τ) + exp(-Dsq2τ) (5) Af + A s Af + As

where Af and As are the amplitudes and Df and Ds the diffusion coefficients of the fast and slow components in the bimodal distribution, respectively. The wavenumber q is defined by the scattering angle θ and the wavelength λ of the laser beam in the vitreous body θ 4π sin q) λ 2

()

(6)

Figure 3. Swelling ratio of bovine vitreous body in different concentrations of CaCl2 aqueous solutions at 23 °C (n ) 3 for each concentration).

Data were obtained using a scattering angle of 135° at a temperature of 23 °C and a focal depth of 1-1.5 mm within the vitreous body. MLLSS measures the intensity correlation functions from the observed time-dependent fluctuations in the intensity of light scattered by the vitreous body. 3. Results and Discussions 3.1. Swelling Behavior. The swelling behavior of a gel is determined by the osmotic pressure of the gel. The osmotic pressure of the gel consists of four contributions, that is, the rubber elasticity of the polymer network, the effects of the counterion of the ionizable group on the polymer network, the interaction free energy between the polymer and the solvent, and the mixing entropy. The balance of these four factors determines the equilibrium-swelling ratio of the gel.1 The vitreous body is a typical biological gel that consists of protein (collagen) and polysaccharide. The collagen is a main protein component of the vitreous body.4 Hyaluronic acid is a typical acidic mucopolysaccharide carrying carboxyl groups as the ionizable side chain. These components form a complex in the vitreous gel and built-up the three-dimensional polymer networks of the gel. The vitreous body is, therefore, one of the multicomponent ionic gels. 3.1.1. pH and Salt Concentration Dependence. Figure 2 shows the swelling ratio of vitreous body as a function of pH. The swelling ratio of vitreous body is almost 2 in the pH region from 4 to 10. It is also found that the vitreous body keeps the original shape and transparency in this pH region. The vitreous body, however, discontinuously collapses into a compact state at pH 4. The vitreous body

Figure 4. Swelling ratio of bovine vitreous body in a mixture of water and methanol plotted as a function of the methanol composition at 23 °C (n ) 3 for each concentration).

becomes opaque below pH 3, but it maintains a spherical shape even in the collapsed state (Figure 5b). The gel that collapsed in the lower pH region does not re-swell to its original volume if soaked in water at a neutral pH. The volume phase transition that is induced by changing pH is, therefore, irreversible. The observed results strongly suggest that a new contracting force is present after the vitreous gel experiences the most collapsed phase that is promoted by lowering pH. Weak-acid polyelectrolytes with carboxyl groups normally become insoluble to water with increasing the degree of protonation. Insolubility of protonate hyaluronic acid may contribute to the observed irreversibility partly. The repeating units of hyaluronic acid may interact with each other through inter- and intramolecular hydrogen bonding between carboxylic acid and hydroxyl group at low pH. Formation of hydrogen bonding may also be one plausible explanation for irreversible phase behavior.



Figure 5. Bovine vitreous body (a) immediately after preparation, (b) collapsed at pH 3, (c) collapsed in 1 mol/L aqueous solution of calcium chloride, and (d) collapsed in the mixture of water/MeOH (5/95 vol %).

In Figure 3, the swelling ratio of vitreous body is plotted as a function of the concentration of calcium ion in the solvent. We observe first, a deswelling at low salt concentration, then a plateau, and then a discontinuous deswelling at calcium concentrations higher than 1 mol/L. The vitreous body lost transparency but the spherical shape remained in the collapsed state as shown in Figure 5c. In this case, it is found that the volume phase transition was reversible. The osmotic pressure, Π, of a charged gel is given by10,11 Π ) Πmix + Πelastic + Πion ) -

∆µ Vs

(7)

where Vs is the molar volume of the solvent and ∆µ is the chemical potential change caused by gel swelling. Πmix, Πelastic, and Πion are the contributions to the osmotic pressure due to polymer-solvent mixing, polymer chain elasticity, and the Donnan potential, respectively. In the context of the conventional theory of swelling,11 Πmix and Πelastic are functions of the polymer volume fraction, φ, and are given by Πmix ) -

RT [φ + ln(1 - φ) + χφ2] V

(8)

and Πelastic ) νeRT

[( ) ( ) ] 1 φ φ 2 φ0 φ0

1/3

(9)

where R is the gas constant, V is the molar volume of the lattice (i.e., the molar volume of the segment), χ is the FloryHuggins interaction parameter, νe is the number of effective chains contributing to the elasticity per unit volume, and φ0 is the polymer volume fraction at the reference state. Due to cross-linking, the gel acts as if it provides its own semipermiable membrane. To allow for changes in swelling caused by altering the concentration of salts outside the gel,

it is necessary to treat the ionic term, Πion, as the effective difference in chemical potential of the solvent due to mobile ions inside the gel. The osmotic pressure generated by the Donnan potential is then given by12,13 Πion ) -

∆µ ≈ RT(ngel - n0) Vs

(10)

where ngel and n0 denote the total molar number of mobile ions per unit volume in the gel and solvent, respectively. The ion concentrations are determined by the Donnan equilibrium. By numerical analysis of the equations which determine the degree of swelling of a gel (eqs 8-10), the ionic term is the significant term contributing to the swelling pressure of the gel, which indicates that the Donnan equilibrium is the main factor determining the swelling behavior of the gel. The vitreous gel swells in 10-7-10-4 mol/L of the salt solution due to the charge repulsion force of the carboxylate group on hyaluronic acid, resulting in expansion of the gel networks. When the salt concentration of the external solution increased from 10-4-10-3 mol/L, the negative charges in hyaluronic acid were neutralized by the cations and the swelling ratio rapidly decreased; that is, the gel show deswelling behavior. Almost all of the negative charges on the polymer chains were neutralized by the external cations in this concentration range. This results from the Donnan effect.10-12 However, the salt concentration further increased over 10-3 mol/L, the vitreous gel became nonionic-type hydrogel, therefore the swelling curve reflected nearly a horizontal line in the range of 10-3-10-1 mol/L. When the salt concentration was raised to 1 mol/L, the salting-out effect was enhanced, due to the high external ionic concentration, and the swelling ratio of the vitreous gel sharply decreased. 3.1.2. Anisotropic Deswelling of Vitreous Body. Figure 4 shows the swelling behavior of bovine vitreous body in

1300


Figure 6. Correlation function as determined by the technique of MLLSS from light scattered by the bovine vitreous body at [CaCl2] ) 1 × 10-4 mol/L. The data were collected using a scattered light collection angle 135° and at a temperature 23 °C. The solid line represents a computer-generated fit to the data using eqs 4-6. The results of the fit yielded the two apparent, collective diffusion motions of the vitreous gel; the “fast” mode Df ) 7.8 ( 1.5 × 10-8cm2/s and the “slow” mode DS ) 3.8 ( 0.60 × 10-9cm2/s, and % As ) 60 ( 5 s, which represents an approximate value for the relative concentration of the fast scatterer present within the vitreous body (mean ( SD for three measurements).

the mixed solvent of methanol and water. The vitreous body continuously collapses into a compact state in the mixed solvent of methanol and water. It was found from these results that the vitreous body swells to about 2-fold when soaked in pure water. The initial state of the gel, which corresponds to the swelling ratio of unity, is regained at a methanol concentration of about 37%. The gel becomes opaque at methanol concentrations above 78%. The vitreous body collapses into a compact state at a methanol concentration above 95%. The gel regains original volume and shape when re-swollen in water. The structure of the gel is, however, totally altered from the intact state. Only the peripheral region of the gel swells in the re-swollen state. The observed irreversible transition may be attributed to denaturation of collagen at high methanol concentration.14 Since methanol is a poor solvent for these polymers, the attractive interaction between polymers becomes dominant when the concentration of methanol is increased in the mixed solvent of methanol and water. Interestingly, we find that the vitreous body collapses nonuniformally as shown in Figure 5d. The vitreous body collapses along an axis parallel to the direction from the lens to the retina. The spherical gel of vitreous body, therefore, collapses to a disk in a mixed solvent of methanol and water. This phenomenon suggests that there are structures which lead to shrinkage along the orbital axis. Also, this may be related to the cisternal structure of the vitreous body that J. G. T. Worst proposed.15 More extensive study is needed, however, to identify the microscopic structure of the vitreous body. 3.2. Microscopic Laser Light Scattering Spectroscopy (MLLSS). Figure 6 exhibits a correlation function obtained by MLLSS from observations of light scattered by the bovine vitreous body immersed in 1 × 10-4 mol/L aqueous CaCl2 solution. The data were collected using a scattering angle of 135° at a temperature 23 °C. The solid line represents a nonlinear least-squares fit to the data using eqs 4-6. The

Matsuura et al.

Figure 7. Diffusion coefficient (fast and slow components) and scattered light intensity as a function of CaCl2 concentration at 23 °C (n ) 3 for each concentration). As evidenced, the observed diffusion coefficients Df and DS both slow appreciably and diminish as the vitreous gel approaches a critical calcium concentration 1 mol/L. Concurrent with the decrease in the observed diffusion coefficients, the total scattered intensity is also observed to increase dramatically. The solid lines are guide for the eyes.

results of the fit yielded the two apparent, collective diffusion motions of the vitreous gel; the “fast” mode Df ) 7.8 ( 1.5 × 10-8 cm2/s and the “slow” mode Ds ) 3.8 ( 0.60 × 10-9 cm2/s. Scattered light was collected at 10 different locations within a focal depth range of 1-1.5 mm, and the diffusion coefficients were found to show no position dependency. Other significant parameters of the fit include the relative contributions of the fast and slow components to overall dynamic scattered light intensity ID, designated as % Af and % As, respectively, where % Af ) 100 × {Af}/{(Af + As)}. The static % IS and dynamic % ID components of scattered light intensities, where % ID ) 100 × {ID}/{(IS + ID)}, to the total scattered light intensity It observed at a particular wave vector, were also determined. For the data in Figure 6, % As ≈ 60 ( 5 and % ID ≈ 30 ( 4. The dynamics of vitreous gels were observed with respect to externally increased calcium concentration. Figure 7 shows the averaged fast and slow diffusion coefficients, 〈Df〉 and 〈Ds〉, and the measured scattered light intensity It in the concentration range between 1.0 × 10-4 and 1.0 mol/L. The collective diffusion coefficients from the intensity correlation function decreased with the concentration of calcium ion, and diminished at 1 mol/L. The vitreous body became opaque as a threshold for calcium concentration was approached in the range of 1 mol/L. Parallel to the diminishment of the diffusion coefficients, the measured intensity It observed at 135° increased and appeared to diverge as the concentration approached the critical threshold, 1 mol/L (Figure 7). The changes in It can be attributed to the increased compressibility of the vitreous gel as it approached the critical point of concentration. The divergent behavior in the observed diffusion coefficients and total scattered light intensities is indicative of the occurrence of a phase transition upon calcium ion concentration. These changes were reversible. Under the condition of increased calcium concentration, the vitreous body approached a critical point where it became opaque. Critical opalescence occurs as a result of large-scale temporal fluctuations in the local densities of the collagen gel, where regions of high and low collagen densities form. Therefore, as a phase transition is approached, the magnitude



the highly flexible hyaluronic acid is interwoven by a semirigid network of collagen threads. It is natural to consider that the collagen motion is the coupled with the dynamics of hyaluronic acid. We are developing the equation for the mode coupling of flexible hyaluronic acid and the semirigid network of collagen. 4. Clinical Implication

Figure 8. Static IS and dynamic ID component contributions to the total scattered intensity It, obtained from the fits to correlation functions using eqs 4-6, as a function of calcium concentration at 23 °C. The corresponding contribution of the “slow” scatterer % AS to the overall increase in ID is also shown (n ) 3 for each concentration). The dramatic increase in ID is indicative of increased amplitudes of local density fluctuations that usually accompany a phase transition. The solid lines are guide for the eyes.

of intensity fluctuations is increased while they become slowed. This behavior was readily evidenced by a divergent behavior in the observed scattered light intensities (It) and diffusion coefficients (Df and Ds). Moreover, the dramatic increase in the observed scattered light intensities It is associated directly with an increase in the dynamic component of the scattered light ID rather than the static component IS (Figure 8). The primary contributor to the increase in ID is the “slow” mode, where an increase in % AS is also observed as the calcium concentration was increased (Figure 8). This behavior is indicative of a gel undergoing a phase transition. The collective diffusion coefficient has been theoretically formulated for gels16,17 D(T) )

kBT 6πη(T)ξ

(11)

where kB is the Boltzmann constant, T is the temperature, η(T) is the temperature-dependent viscosity of the “trapped” fluid, and ξ denotes the correlation length of the concentration fluctuations in the network, which in theory represents conceptually either the average distance between neighboring cross-links in the network (pore size) or the size of segments constituting the network polymers.1,16 In practice, the spatial arrangements of randomly formed gel networks are not well known to offer any direct comparison with the values. Using eq 11, the parameter ξ was calculated for the above-cited experimental values, and we obtained ξf ≈ 30 nm and ξS ≈ 618 nm for the fast and slow diffusion coefficients, respectively. These values provide the “pore” size of the collagen network. It is well known that the average diameter of the collagen fibrils in the vitreous body is approximately 1020 nm.18 Although eq 11 may not predict the actual structure of the vitreous network, the parameter ξ may be useful for determining the swollen state of the vitreous body, and can have practical applications in the clinical determination of the onset of turbidity by metabolic defect or high age. The dynamic light scattering measures the time correlation of density fluctuation of the scattering entities, which are the segments of the hyaluronic acid polymer and the collagen mesh. The vitreous body has a complex structure, in which

As preliminary experiments, we investigate the structural and dynamical properties and phase equilibrium properties of the calf vitreous body by changing the external conditions such as pH, ion concentration, and solvent composition in vitro. Human vitreous body may have basically the same composition and structure as the bovine one (collagen and hyaluronic acid);19,20,21 therefore, the same phenomena are expected to be observed in response to the external conditions applied in this study. Posterior vitreous detachment (PVD) is the most common pathophysiologic condition of the vitreous body. The mechanism of PVD is a phenomenon that is still poorly understood. In this study, we investigate the shrinkage of vitreous body in vitro and consider the mechanisms of the phase transition of vitreous body by changing the external conditions. Regarded from this standpoint, PVD which occurrs clinically may be related to the phase transition of vitreous gel. The existence of critical phenomenon in vitreous body may also explain the abnormal diffusion of macromolecules in human vitreous body observed in the case of various intraocular diseases. Dynamic light scattering experiments in this study show that as the vitreous gel approaches the critical point the spatial correlation among the polymer segments of the network polymers increases and the movements of the segments become correlated and slow. This is closely correlated to the effective pore size of the polymer network, which practically diverges as the gel passes the coexistence curve into the two-phase region. More extensive study, however, is needed to reveal the physicochemical bases of diseases. 5. Concluding Remarks The phase equilibrium property and structural and dynamical properties of bovine vitreous body were studied by macroscopic observation of swelling behavior and dynamic light scattering under various conditions. It was found that the vitreous body collapses isotropically or anisotropically depending on the external conditions. The vitreous body collapses into a compact state while maintaining its shape when the pH and the concentration of calcium ions are changed. On the other hand, the vitreous body collapses along the orbital axis in a mixed solvent of methanol and water. This may be related to the cisternal structure of the vitreous body. From observations of the dynamics of light scattered by the vitreous body, intensity autocorrelation functions that revealed two independent diffusion coefficients were obtained. The collective diffusion coefficients from the intensity correlation function decreased with the concentration of calcium ion and diminished at 1 mol/L. Parallel to the diminishment of the diffusion coefficients, the measured intensity increased and appeared to diverge as the concentra-

1302


tion approached at 1 mol/L. The divergent behavior in the observed diffusion coefficients and total scattered light intensities is indicative of the occurrence of a phase transition upon calcium ion concentration. Acknowledgment. The work was partly supported by part NIH, E.Y.05272-05, and a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture (M. A.). M.A. and T.M. acknowledge the Sumitomo Foundation for financial support. The authors thank Dr. Toyoichi Tanaka and Dr. Mototsugu Saishin for valuable discussions and suggestions. References and Notes (1) (2) (3) (4)

Tanaka, T. Phys. ReV. Lett. 1978, 40, 820. Shibayama, M.; Tanaka, T. AdV. Polym. Sci. 1993, 109, 1. Amiya, T.; Tanaka, T. Macromlecules 1987, 20, 1162. Berman, E. R.; Voaden, M. In Biochemistry of the Eye; Smelser, G. K., Ed.; Academic Press: London, 1970; p 373. (5) Balazs, E. A. In New and ControVersial Aspects of Retinal Detachment; McPherson, A., Ed.; Academic Press: Philadelphia, 1968; Vol. 1, p 3. (6) Worst, J. G. F. Cisternal Anatomy of the Vitreous; Kugler Publications: Amsterdam, 1995; p 1. (7) Matsuura, T.; Gorti, S.; Tanaka, T.; Hara, Y.; Saishin, M. Eur. Biophys. J. 1999, 28, 357.

Matsuura et al. (8) Berne, B. J.; Pecora, R. Dynamic Light Scattering; Plenum Press: New York, 1976. (9) Peeterman, J.; Nishio, I.; Onishi, S.; Tanaka, T. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 352. (10) Tanaka, T.; Fillmore, D.; Nishio, I.; Sun, S.-T.; Shah, A.; Swislow, G. Phys. ReV. Lett. 1980, 45, 1636. (11) Flory, P. J. In Principle of Polymer Chemistry; Cornell University Press: Ithaca, NY, 1953; pp 541-593. (12) Ohmine, I.; Tanaka, T. J. Chem. Phys. 1982, 77, 5725. (13) Ricka, J.; Tanaka, T. Macromolecules 1984, 17, 2916. (14) Creighton, T. E. Proteins, Structure and Molecular Properties; Freeman, New York, 1984. (15) Worst, J. G. F. Cisternal Anatomy of the Vitreous; Kugler Publications: Amsterdam, 1995; p 3. (16) Tanaka, T.; Hocker L. O.; Benedek, G. B. J. Chem. Phys. 1973, 59, 5151. (17) de Gennes, P. G. Scaling Concepts in Polymer Physics; Cornell University Press: Ithaca, NY, 1979; pp 214-216. (18) Kaufman, P. L.; Alm, A. Adler’s Physiology of the eye; Mosby: Washington, DC, 2001; p 296. (19) Laurent, U. B. G.; Granath, K. A. Exp. Eye. Res. 1983, 36, 481. (20) Seery, C. M.; Davison, P. F. InVest. Ophthal., Visual Sci. 1991, 32, 1540. (21) Schepens, C. L.; Neetens, A. The Vitreous and Vitreoretinal Interface; Springer-Verlag: New York, 1987; p 211.

BM049954Y

Volume Phase Transition of Bovine Vitreous Body in Vitro and

Recommend Documents