Syneretic Response of Aging Normal Human Lens to Pressure

Abstract

purpose. The study was designed to observe whether a reversible syneretic response to pressure is operative in normal human lenses and whether such a response demonstrates a uniform age dependence.

methods. Seven sections (from the anterior outer cortex to the posterior outer cortex) of 10 human lenses were imaged at 2 atmospheres (atm) pressure and the T₁ (spin-lattice) and T₂ (spin-spin) relaxation data on each section were collected. The pressure was then released and NMR relaxographic data were collected under 1 atm.

results. Both T₁ and T₂ relaxation times were at their minimum in the nuclear region and at their maximum at the two outer cortexes. With increasing pressure, T₂ relaxation times decreased. The pressure-dependent change in T₂ relaxation times decreased with age. Changes in T₁ relaxation times showed no consistent pressure or age dependence. The population index of T₂ relaxation, M₂, had a maximum in the nucleus and a minimum in the two cortexes. The population index of T₁ relaxation, M_1, was minimal in the nucleus and maximal at the two cortexes. M₂ increased with increasing pressure, whereas M₁ did not show consistent pressure dependence. The percentage of change in M₂ (ΔM₂) showed a statistically significant increase with increasing age, whereas the %ΔM₁ showed no significant age-dependent trend.

conclusions. The positional dependence of relaxation times and the population indexes indicated that spin-spin relaxation represents the behavior of the bound water and the spin-lattice relaxation that of total water. As pressure increases, the strength of hydrogen bonding as well as the amount of bound water increases. The pressure-induced change in the total water is minimal. Thus, the free water-to-bound water ratio decreases with increasing pressure, demonstrating a significant syneretic response. The extent of reversible syneretic response decreases with age and is actually reversed in older lenses. The implication is that the ability of the human lens to respond reversibly to pressure decreases with the decrease in accommodation, and, when the ability is lost altogether, an increase in free water, a possible source of cataract formation, may ensue.

Asyneretic process or syneretic response refers to the movement of water in or out of the hydration layer. These terms designate both syneresis, in which bound water is released from the hydration layer of biopolymers and becomes free water and inverse syneresis. It has recently been predicted on theoretical grounds that in the ocular lens a reversible syneretic process could be an operative response to hydrostatic pressure changes. ¹ Explicit application of pressure occurs in hyperbaric medicine. The effect of hyperbaric conditions on the lens in vitro has been explored by exposure to 100 atmospheres (atm) of nitrogen or oxygen pressures for different durations and repeatedly up to 100 times. ^{2 3 4 5} Such exposures create nuclear light-scattering and oxidative damage. However, the effect of elevated pressures on the hydration of the lens has not yet been studied. 

Furthermore, during accommodation repeated hydrostatic pressure changes are experienced by the lens. The zonules, which insert into the lens capsule ⁶ apply stress that has both parallel (stretching) and perpendicular (compressive) components. ⁷ These discrete stresses are transformed by the capsule into a uniform stress approximately perpendicular to the lens surface. ⁸ The transition from the unaccommodated to the accommodated state, then, would include a reduction of stresses perpendicular to the lens surface. Hydrostatic pressure on the lens surface has an equilibrium counterpart in osmotic pressure. As a consequence of accommodation, there would be a tendency for water movement from the lens to the surroundings, which would increase the osmolarity of the lens. One way to counteract this outflow of water is to convert bound water to free water, a syneretic process. ¹ Experiments using differential scanning calorimetry and thermogravimetric analysis (DSC/TGA) have demonstrated that such a reversible syneretic response to pressure change occurs in bovine ¹ and rhesus monkey lenses ⁹ in vitro. Similar results were obtained in nuclear magnetic resonance (NMR) relaxographic studies on bovine lenses. ¹⁰ The ratio of free water to bound water decreased with increasing pressure and increased when the pressure was reduced. 

In contrast, irreversible syneresis has a number of potential consequences. It accounts for the liquid pocket formation in the aging vitreous. ¹¹ Similar irreversible syneresis is involved in the formation of cataract. ¹² In aging and cataractous lenses, irreversible syneresis contributes to the turbidity of the lens by increasing the amplitude of refractive index fluctuations. ^{13 14} Besides light-scattering studies, the role of syneresis in cataractogenesis has been shown by NMR ^{15 16 17 18} and DSC/TGA, in human lenses ¹⁹ and in different animal models. ^{20 21}  

Because presbyopia and cataractogenesis are both age-related afflictions and because both processes have syneretic components, albeit the first is a reversible and the second an irreversible process, it occurred to us that the two may have a common origin. ²²  

The aging human lens shows definite syneretic involvement. ^{15 16 23 24 25 26 27 28} Whether the change from a reversible inverse syneresis to an irreversible process of syneresis is indicative of disease is the question we tried to answer in this study. 

Materials and Methods

Ten human eyes were obtained from the National Disease Research Interchange (NDRI; Philadelphia, PA). The research was conducted according to the provisions of the Declaration of Helsinki for research involving human tissue. Eyes were enucleated and maintained in a cool, moist chamber until arrival at the laboratory 36 to 48 hours after death. The lenses were dissected as previously described ²⁹ and were immersed in prepared medium (Leibowitz L-15 with glutamine but no phenol red, adjusted to pH 7.5). Antibiotics (30 U/mL penicillin and 30 μg/mL streptomycin) were added, and the osmolarity of the medium was adjusted to 300 ± 2 mOsmol. 

The lenses were placed in special holders containing the medium. The glass holder had an indentation to accommodate the lens snugly while the tube was in a horizontal position. Two-part sample holders were used. The lower part contained the sample and the upper part had a stopcock that could maintain vacuum or pressure up to 3 atm. The stopcock and the joints of the two parts were lubricated by high-vacuum grease. The sample holder fit into the bore of the NMR magnet so that the lens could be reproducibly localized for imaging. The lenses were exposed to 2 atm pressure that was obtained by connecting the sample holder to a nitrogen cylinder and maintaining the pressure with control valves. All incubations were kept at 37°C by placing the vessels inside a tissue culture incubator. After 12 hours’ exposure, the vessels containing the lenses, but still under 2 atm pressure, were removed and allowed to equilibrate with room temperature for 2 hours. They were transferred to the cavity of the NMR magnet. After the NMR data were collected on a lens at room temperature while under 2 atm pressure, the stopcock was opened and the pressure released to normal 1 atm. MMR data collection was started again and continued repeatedly over an 8- to 12-hour period, yielding four to six sets of data on the normal 1-atm condition of the lens under investigation. 

All NMR measurements were performed at 4.7 Tesla. A 3-mm slice was determined that bisected the imaged lens along the optic axis. ²⁴ T₁ (spin-lattice) data were acquired for this slice, by the method of inversion recovery. ³⁰ T₁ relaxation images were obtained by using an imaging sequence consisting of an initial-slice, selective-inversion pulse followed by a delay and then a standard spin-echo imaging sequence. T₂ (spin-spin) relaxation images were acquired using a standard spin-echo imaging sequence with varying echo time. ³⁰ Relaxation times were calculated from numerical fits to regions of interest in the lens images. Using a mouse-driven graphic analysis program (Research Systems Inc., Boulder, CO), the perimeter of the lens was traced. The perimeter was then scaled down by factors of 0.8, 0.6, and 0.4 and split into anterior and posterior portions. The region from 1.0 to 0.8 was considered the outer cortex (anterior, AOC; posterior, POC), from 0.8 to 0.6 the middle cortex (anterior, AMC; posterior, PMC) and from 0.6 to 0.4 of the inner cortex (anterior, AIC; posterior, PIC). The region under 0.4 was considered the nucleus (N). The average pixel intensity was measured in each region, and the values were fit to a one- or two-term exponential expression of relaxation, by using a least-squares fitting algorithm. ³¹  

Results

Both the T₁ and T₂ relaxation could be fitted with a one-term exponential having a goodness of fit (r ² =1.000).  

\[\mathrm{Intensity}\ {=}\ \mathit{M}\mathrm{exp}({-}\mathit{t}/\mathit{T})\ {+}\ \mathit{C}\]

where M, the preexponential term, is a population index, T is the relaxation time, t is the time in seconds, and C is the constant. 

The T₂ relaxation times of two human lenses, 39 and 77 years old, are presented as a function of pressure and position in Figure 1 , top and bottom, respectively. In both cases the T₂ relaxation times decreased from cortex to nucleus. The T₂ relaxation times decreased with increasing pressure, which is a syneretic response. ^{1 10} This is more evident in the younger lens than in the older, and the cortexes show this effect more dominantly than the inner regions. The age-dependent syneretic response for all 10 lenses investigated is presented in Figure 2 for the anterior outer cortex region. With increase in age, there was a steady decrease in ΔT₂, which is the change in T₂ relaxation times with pressures changing from 1 to 2 atm. For all the regions within the lens, the slope of such a decrease, the goodness of the least-square fit, and the statistical significance of the difference from zero are presented in Table 1 . 

The T₁ relaxation times of the same two human lenses are given in Figure 3 , top and bottom, respectively, as a function of pressure and position. The T₁ relaxation times decreased, moving from cortex to nucleus as well as with increasing pressure from 1 to 2 atm. As with the T₂ relaxation times the change in the T₁ relaxation times with pressure is more evident in the young lens than in the old one and more in the cortex than in the inner regions. The age-dependent changes in the ΔT₁—that is, the change with pressure increasing from 1 to 2 atm—are presented in Figure 4 for the outer anterior cortex. Similar changes for all other segments are provided in Table 1 . These changes have no statistical significance. 

The preexponential terms of the T₂ relaxation, M₂, are presented in Figure 5 , top and bottom, for two human lenses, age 39 and 77 years, respectively. M₂ increased from cortex to nucleus in both lenses. Increasing the pressure from 1 to 2 atm increased M₂ in all regions of the young lens, but this effect was mostly the opposite in the 77-year-old lens. The percentage of change in M₂ with pressure increasing from 1 to 2 atm is given in Figure 6 for the outer anterior cortex in all 10 lenses investigated. The statistical parameters for the other segments are presented in Table 2 . 

The values of the preexponential terms of the T₁ relaxation, M₁, are graphed in Figure 7 , top and bottom, for two human lenses, 39 and 77 years old, respectively, as a function of pressure and position. M₁ decreased moving from cortex to the inner regions of both lenses and increase with increasing pressure in the young lens, but the reverse was true in the old lens. The percentage change in M₁ with pressure changing from 1 to 2 atm is provided in Figure 8 for the outer anterior cortex, as a function of age. Basically, there was no change with age. Similar data for the other the regions of the lens are given in Table 2 . 

Discussion

A syneretic response to pressure that was predicted and observed in bovine and monkey lenses ^{1 9} is evident in normal human lenses. The gist of the response is the conversion of free water to bound water as the pressure across the fiber cell membranes increases. This effect is observable by an increase in the strength of hydrogen bonding and by the increase in the amount of bound water with increasing pressure. 

Water molecules close to the biopolymer (protein) surface interact by H-bonding, their diffusion is slowed down and their rotation is hindered. Bound water may have different degrees of freedom: immobilized partially or fully. Measurements, such as differential scanning calorimetry combined with thermogravimetric analysis, usually deal with only two forms of water, freezable (free water) and nonfreezable (bound water). ^{1 9 20 23} NMR measurements, depending on the fitting of the relaxation curves, may perceive one or up to three different forms of water. ^{25 26 27} Because in our measurements there was no significant improvement in the two-term exponential fitting over the one-term exponential, we could not assume a two-component system. This is what we also observed in relaxographic studies of bovine lenses. ¹⁰  

The most likely explanation for the differences between our data and earlier measurements is that we looked at segments of the lens, whereas Stankiewicz et al. ²⁶ and Racz et al. ^{15 25 27} measured the relaxation of whole lenses. Thus, our seven lens segments—from anterior cortex, to nucleus, to posterior cortex—each represented a more homogeneous sample than the whole lens of previous studies. The T₂ relaxation times for two water components in normal human lenses found by Racz et al. ^{15 27} were 62 and 22 ms. These two water components correspond broadly with the T₂ times we found in the cortex and nucleus, respectively. ²⁴ T₂ relaxation times attributed to bound and free water by Racz et al. ^{25 27} may represent the behavior of the bound water we found in the cortex and nucleus, respectively. 

On the basis of positional dependence of the T₁ and T₂ relaxation parameters within the lens we could determine that T₂ relaxation describes the behavior of bound water, whereas T₁ relaxation pertains to the behavior of total water. ^{10 24} This was also evident from the present studies, in which we found maxima of M₂ in the nucleus and minima in the cortexes (Fig. 5) , whereas M₁ possessed a positional dependence with minima in the nucleus and maxima in the cortexes (Fig. 7) . Total water had minimal concentration in the nucleus, whereas bound water had its highest concentration in the nucleus. From the relaxation parameters we can state the following. 

Strength of Hydrogen Bonds in Bound Water

The effect on the strength of hydrogen bonding can be seen from the changes in the T₂ relaxation times with pressure. A shorter T₂ relaxation time is indicative of a more restrictive motion of the protons—hence, a stronger hydrogen bonding. As can be seen in Figure 1 the T₂ relaxation times decreased with increasing pressure in every segment of the lens, the change being greatest in the outer cortical segments and diminishing from cortex to nucleus. This decrease in the T₂ relaxation times was more prominent in the 39-year-old lens (Fig. 1 , top) than in the 77-year-old lens (Fig. 1 , bottom). The age dependence of change in T₂ relaxation is displayed dramatically in Figure 2 for the anterior outer cortical segments. The negative slope of the diagram indicates that the syneretic response, as far as the strength of hydrogen bonding is concerned, decreased with age. The younger lens converted free water to bound water to a greater degree than the older lens, such that, in the old lenses, the increase in the strength of hydrogen bonding with increasing pressure was minimal or absent. Similar data are available for the other segments of the normal human lens, as tabulated in Table 1 . The age-dependent slope of the ΔT₂ parameter is negative in the cortical segments and shows no age dependence in the nucleus. This latter observation indicates that the diminishing of ability to respond to pressure is not only relevant to the chronological aging of the lens but in the developmental aging (cortex to nucleus) as well. 

Amount of Bound Water

The preexponential term of the T₂ relaxation, M₂, is a measure of the amount of protons experiencing the magnetic field. On the basis of the positional dependence of M₂ we concluded that this represents the amount of bound water. ^{10 24} The preexponential terms, M₂, increased with pressure, moving from cortex to nucleus in every lens (Fig. 5) . Such an increase was characteristic of the young lens (Fig. 5 , top) and was reversed in the older lens (Fig. 5 , bottom). The age dependence of change in the amount of bound water can be seen in Figure 6 for the anterior outer cortical segments. We elected to express the amount of change in bound water as a percentage change in M₂ as pressure changed from 1 to 2 atm. Because in each lens the pressure change was measured within 24 hours, there was no need to normalize the preexponential term as we did in the 2-year study on the aging of normal human lenses. ²⁴ Expressing the syneretic response as the percentage of change allowed the comparison of different lenses that were studied over a period of 2 years. The positive slope of the linear least-square fit (Fig. 4) means that the negative values of %ΔM₂ observable in the young lens become smaller and smaller, eventually becoming zero or positive in old age. Similar data are represented in Table 2 for the different segments of the lens. These negative percentages of change with increasing pressure can be compared with those obtained in other species, albeit under different experimental conditions. The average change in the bound water in 2-year-old rhesus monkey lenses was 12% ⁹ and that in 2-year-old bovine lenses 30%. ¹ These values are in the same range that we can see in Figure 6 in the younger (39–53 years) normal human lenses. All the slopes of percentage changes are positive and, in general, mirror the positional dependence of the absolute values of M₂ within each lens. This means that younger lenses convert free water to bound water with increasing pressure, but in older lenses, this ability is diminished and in some cases reversed. The implication of the latter is that with aging the ability of the lens to compensate for the hydrostatic pressures decreases to the point that the lens at old age starts to behave, not as a membrane-bound body, but more like a gel. This development has significance, not only in reference to decrease in accommodation with age, but also with regard to cataractogenesis. Under such conditions more and more free water may accumulate, potentially contributing to the cataract by formation of a lake. 

Strength of Hydrogen Bonds in Total Water

On the basis of the data collected for T₁ relaxation times we can state the following: Although in the 39-year-old lens a uniform decrease in relaxation time was observed with increasing pressure (Fig. 3 , top), such a uniformity was not evident in all lenses. In some older lenses this relationship was reversed in only one segment (Fig. 3 , bottom), whereas in others many or all segments showed an increase in relaxation times with increasing pressure. This lack of uniformity is obvious in Figure 4 , which shows positive and negative changes in ΔT₁. The age dependence of ΔT₁ in the anterior outer cortex segments is not much different from zero. Tabulation of the age dependence of ΔT₁ in the other segments of the lenses also indicated (Table 1) that not only can ΔT₁ be positive or negative but the age-dependent slopes can become positive and negative as well. None of this age dependence is statistically significant. It must therefore be concluded that there is no particular trend in the strengthening of hydrogen bonding in total water in contrast to what we observed for the bound water. A few young lenses demonstrated a slight increase in the strength of hydrogen bonding of total water (Fig. 3 , top). Such an increase may indicate the effect of the bound water on the total water. However, in older lenses where the syneretic response was diminished and therefore the strengthening of the hydrogen bonds of water was lessened on pressure change, this effect on the total water was minimal. This suggests that in older lenses if there is strengthening of hydrogen bonds in the bound water, there is an equivalent weakening in the free water. 

The Amount of Total Water

The preexponential term of T₁ relaxation, Μ_1, showed a similar trend. In young lenses the response was uniformly an increase in M₁ with increasing pressure (Fig. 7 , top), indicating an increase in total water as well. However, in older lenses more often than not there was a decrease in M₁ with increasing pressure. Figure 8 shows the data for the anterior outer cortex segments of all lenses investigated, revealing a mix of positive and negative M₁. There is no statistically significant age dependence in this parameter, as can be seen in Figure 8 and Table 2 . This indicates that there is no significant change in the status of the total water when pressure changes, nor is there any significant change in its age dependence. The latter also has been observed in aging lenses by other techniques. ^{32 33 34}  

The combined effects of pressure on the strength of hydrogen bonding and on the amount of bound water illustrate the decreasing ability of the aging lens to compensate for changes in hydrostatic pressure. This, combined with the absence of detectable change in the behavior and amount of total water, substantiates the syneretic theory. Furthermore, if the diminishing ability of the aging lens to provide a syneretic response to hydrostatic pressure becomes an irreversible process, turbidity ensues. In the oldest normal human lenses, we have seen that the increase in pressure actually causes the release of bound water to become free water. When such a response to pressure is irreversible, the released free water may accumulate in lakes. Formation of lakes is a well-known cause of turbidity, inasmuch as it increases the size and the amplitude of the density fluctuations. ^{12 13 14} Thus, our findings indicate that the diminishing accommodation and even cataract formation in aging lenses may have a common causative element. 

 Supported in part by National Eye Institute Grant EY12496.

 Submitted for publication April 29, 2002; revised July 10, 2002; accepted July 24, 2002.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

T₂ relaxation times in (top) a 39-year-old and (bottom) a 77 year-old normal human lens as a function of pressure and position.

Figure 2.

 

Age dependence of the pressure-induced change in T₂ relaxation times in the imaged anterior outer cortex.

Table 1.

 

Age Dependence of Parameters, ΔT₁ and ΔT₂, in Different Segments of Normal Human Lenses

Position	Slope	Goodness of Fit (r 2)	F	P
ΔT1
AOC	0.0033	0.0068	0.055	0.8198
AMC	−0.0023	0.0169	0.138	0.7197
AIC	−0.0006	0.0041	0.032	0.8607
N	−0.0003	0.0015	0.011	0.9156
PIC	−0.0010	0.01366	0.111	0.7478
PMC	−0.0001	0.0000	0.000	0.9869
POC	0.0006	0.0009	0.007	0.9330
ΔT2
AOC	−0.3773	0.4955	7.858	0.0231*
AMC	−0.0633	0.0874	0.766	0.4069
AIC	0.0056	0.0010	0.008	0.9306
N	0.0006	0.00001	0.000	0.9916
PIC	−0.0443	0.04633	0.388	0.5504
PMC	−0.0167	0.00437	0.035	0.8561
POC	−0.1737	0.0330	0.275	0.6138

Figure 3.

 

T₁ relaxation times in (top) a 39-year-old and (bottom) a 77-year-old normal human lens as a function of pressure and position.

Figure 4.

 

Age dependence of the pressure-induced change in T₁ relaxation times in the imaged anterior outer cortex.

Figure 5.

 

Preexponential terms of T₂ relaxation, M_2, in (top) a 39-year-old and (bottom) a 77 year-old normal human lens as a function of pressure and position.

Figure 6.

 

Age dependence of the pressure-induced percentage change in M₂ in the imaged anterior outer cortex.

Table 2.

 

Age Dependence of the Parameters, %ΔM₁ and %ΔM₂ in Different Segments of Normal Human Lenses

Position	Slope	Goodness of Fit (r 2)	F	P
%ΔM1
AOC	0.0502	0.0007	0.006	0.9409
AMC	0.3649	0.0867	0.759	0.4089
AIC	0.3489	0.1508	1.420	0.2675
N	0.3940	0.2102	2.129	0.1827
PIC	0.4289	0.2150	2.191	0.1771
PMC	0.4664	0.2286	2.370	0.1622
POC	0.4870	0.1584	1.506	0.2546
%ΔM2
AOC	0.4105	0.3141	3.663	0.0920*
AMC	0.4343	0.2239	2.308	0.1672
AIC	0.4256	0.2101	2.127	0.1828
N	0.5893	0.3503	4.314	0.0715*
PIC	1.027	0.4274	5.972	0.0403, †
PMC	0.738	0.4685	7.051	0.0290, †
POC	1.005	0.4920	7.749	0.0238, †

Figure 7.

 

Preexponential terms of T₁ relaxation, M_1, in (top) a 39-year-old and (bottom) a 77 year-old normal human lens as a function of pressure and position.

Figure 8.

 

Age dependence of the pressure-induced percentage change in M₁ in the imaged anterior outer cortex.