Mechanical Properties of Eye Lens Cortical and Nuclear Membranes and the Whole Lens

Abstract

Purpose

To elucidate the mechanical properties of the bovine lens cortical membrane (CM), the nuclear membrane (NM) containing cholesterol bilayer domains (CBDs), and whole bovine lenses.

Methods

The total lipids (lipids plus cholesterol) from the cortex and nucleus of a single bovine lens were isolated using the monophasic methanol extraction method. Supported CMs and NMs were prepared from total lipids extracted from the cortex and nucleus, respectively, using a rapid solvent exchange method and probe-tip sonication, followed by the fusion of unilamellar vesicles on a flat, freshly cleaved mica surface. Topographical images and force curves for the CMs and NMs were obtained via atomic force microscopy (AFM) in a fluid cell. Whole bovine lenses were affixed to custom-built glass Petri dishes, and an AFM was used to obtain force curves. Force curves were analyzed to estimate the breakthrough force, membrane stiffness (K_A and E_m), and lens stiffness (E_L).

Results

The NMs containing CBDs exhibited significantly lower breakthrough force, K_A, and E_m than the CMs without CBDs. The E_m values for CMs and NMs were significantly higher than the E_L for the whole lens.

Conclusions

The significantly higher stiffness of the CM and NM compared to the stiffness of the whole lens suggests that slight modulation in CM and NM composition may play a crucial role in altering the overall lens stiffness. Furthermore, the NMs containing CBDs were less stiff than CMs without CBDs, suggesting that CBDs decrease lens membrane stiffness and possibly protect against lens hardening and presbyopia.

Several potential factors have been proposed to explain the cause of presbyopia,¹ the loss of ability to focus on near objects, including lens growth² and the roles of ciliary muscle,³ vitreous,⁴ capsule,⁵ and hardening of the lens.^6–10 Recent experimental evidence supports lens elasticity loss with aging,^8–13 and lens stiffness increases with age,^7,10,14,15 resulting in presbyopia development. The Young's modulus of the nuclear region was ∼10-fold larger than the cortical region in older bovine lenses; however, there was no significant difference between the nuclear and cortical Young's modulus in younger porcine and bovine lenses.¹⁶ The stiffness of human eye lenses, regardless of cataractous or non-cataractous condition, has been found to increase with age; however, cataractous lenses are significantly stiffer than age-matched non-cataractous lenses.^10,17,18 In human lenses, the stiffness in the central nuclear region increases by about 1000 times, whereas the outer cortical region increases by around 50 times with age.¹⁰

The mechanical properties of eye lenses, including stiffness, have been investigated in both ex vivo¹⁹ and in vivo^20,21 experiments, revealing that the center of the aged human lens (51, 64, and 78 years) is stiffer than the periphery.²² In contrast, the periphery of the young human lens (19–49 years) is stiffer than the center, and stiffness increases in both regions with aging.²² It has been reported that the nucleus was softer than the cortex in human lenses younger than age 30 years; in contrast, for human lenses older than 30 years, the nucleus was found to be stiffer than the cortex, and both lens cortex and nucleus stiffness increased with aging, with a massive increase in the stiffness of the nucleus.¹⁰ Several groups have attempted to explain the molecular basis of lens hardening; however, the mechanism of lens hardening is still unclear. It has been reported that protein aggregation and insolubilization increase in concert with lens stiffness,^23,24 where crystallin membrane binding leads to protein insolubilization.²⁵ With aging, the soluble lens proteins in the cortex and nucleus, predominantly crystallins (α-, β-, and γ-crystallins), convert into insoluble high molecular weight (HMW) proteins.^7,26 The HMW protein concentration increases with age and is accompanied by a significant increase in lens stiffness.^7,8 HMW protein is associated with the fiber cell plasma membrane.^27–29 It has been reported that crystallins binding to the membrane deteriorate membrane elasticity,^30,31 which in turn has been suggested to deteriorate the ability of individual fiber cells to change shape, accompanied by lens hardening and presbyopia development,^30,31 suggesting the crucial role of the lens membrane in lens hardening.^32,33

The contribution of the overall cortex and nucleus stiffness to lens stiffness likely results from a synergistic effect involving several factors, including intrinsic membrane proteins,^34,35 cytoskeleton,^34,36 crystallins,^6,9,24 and membrane lipids (phospholipids and sphingolipids) and cholesterol (Chol).^32,33,37,38 As mature lens fiber cells lose their intracellular organelles,³⁹ the plasma membrane, together with the cytoskeleton, accounts for the bulk membrane properties of the cells.³⁸ The eye lens plasma membrane contains extremely high levels of Chol content such that the Chol/lipid molar ratio is greater in the nuclear membrane (NM) than in the cortical membrane (CM) in all groups.^40–42 At elevated concentrations, Chol forms immiscible cholesterol bilayer domains (CBDs) within the membranes.^40,41,43,44 High Chol and CBDs in lens membranes play a positive physiological role by maintaining lens membrane homeostasis.^27,45,46 Previously, we studied the mechanical properties of supported membranes containing very high Chol content in individual lipids and the model of porcine lens lipid membranes^32,33 using atomic force microscopy (AFM). We found that high Chol content and CBD decrease membrane stiffness.^32,33 In this work, we investigated the stiffness of CMs and NMs obtained from total lens lipids (lipids plus Chol) from cortical and nuclear regions of ∼2-year-old bovine eye lenses using AFM. Furthermore, we estimated the stiffness of the whole bovine lens using AFM and linked stiffness of the CM and NM with whole lens stiffness, elucidating the role of high Chol and CBDs in modulating CM and NM stiffness and whole lens stiffness.

Materials and Methods

Materials

Bovine eye globes about 2 years old were collected and placed on wet ice within ∼2 hours of animal sacrifice at a local slaughterhouse (Northwest Premium Meats, Nampa, ID, USA), and two clear lenses were removed and stored in Dulbecco's Modified Eagle Medium (DMEM) at 4°C, followed by stiffness measurements using AFM at ∼24 hours and ∼48 hours post-sacrifice. The transparency of the bovine lens was inspected with a binocular microscope. For the lipid extraction, transparent bovine lenses were stored at –80°C until the lipid extraction was performed. None of the lenses used for measuring whole lens stiffness was frozen. Analytical grade DMEM, methanol, hexane, isopropanol, MgCl₂, HEPES, and NaCl were purchased from Sigma-Aldrich (St. Louis, MO, USA). Buffer A (pH 7.4) consisted of 10-mM HEPES and 150-mM NaCl, with 10-mM MgCl₂ added; buffer B (pH 7.4) was composed of simply 10-mM HEPES and 150-mM NaCl.

Lipid Extraction and Supported Lipid Membrane Preparation

A single lens of an approximately 2-year-old bovine was defrosted at room temperature, and a binocular microscope was used to confirm transparency after defrosting the lens at room temperature. The cortex and nucleus were then separated based on tissue consistency to extract total cortical and nuclear lipids (lipid plus Chol), followed by preparation of the CM and NM as described by us previously.^47,48 As reported in our previous publications,^47,48 total lipids from the cortical and nuclear fractions were isolated separately, with minor modifications to the monophasic extraction process.⁴⁹ Briefly, cortical and nuclear tissues were homogenized separately in a glass tube with methanol, initially using a glass Dounce homogenizer, and finally using a probe-tip sonicator (Model 550; Thermo Fisher Scientific, Waltham, MA, USA). The supernatant from each homogenized solution, obtained by centrifuging (1 hour at ∼3000 relative centrifugal force) using a Beckman Coulter J-26S XP centrifuge (Beckman Coulter, Brea, CA, USA), was collected into a separate glass beaker, and the methanol was evaporated by placing the supernatant solution on a hot plate at 60°C with a gentle stream of industrial-grade N₂ gas flowing across the top of the beaker. Dry lipid films were dissolved in 10 mL of hexane–isopropanol mixed solvent (2:1 v/v) solution using a homogenizer followed by a probe-tip sonicator, then further centrifuged in separate glass tubes. The supernatants were evaporated to 2 mL, then transferred to fresh glass centrifuge tubes, followed by the addition of 3 mL of hexane–isopropanol (2:1 v/v) to each tube and centrifuged again. Finally, the supernatant from each tube was transferred into previously weighed small glass tubes and evaporated until the solution was dry, and the remaining traces were removed by placing them in a vacuum overnight. The weight of each glass tube with lipid films was measured, and the total lipid extracted from the cortex and nucleus of a single bovine lens was estimated. The total cortical and nuclear lipids (lipid plus cholesterol) isolated from a single bovine lens were ∼3.1 mg and ∼1.9 mg, respectively, and they were stored at −20°C in a chloroform solution.

To prepare the CM and NM, multilamellar vesicles of cortical and nuclear total lipids prepared by rapid solvent exchange method were probe tip sonicated to formulate small unilamellar vesicle (SUV) suspension solutions in buffer A with a concentration of ∼0.2 mg/mL using the methods described by us previously.^47,48 Supported lipid membranes (SLMs) of cortical and nuclear lipids (i.e., supported CM and NM) were prepared by dispensing the lipid suspension on a freshly cleaved mica disk in a chamber under the AFM head, enclosed in an O-ring connected to a fluid cell. After ∼40 minutes of incubation, unfused excess SUVs were removed by passing buffer B from an inlet on the fluid chamber and removing the excess from a separate outlet. To prepare the membrane patches of CM and NM, ∼10 to 20 µg/mL of total lipids SUV suspension solution prepared from the cortical and nuclear total lipids was deposited on a freshly cleaved mica disk under an AFM fluid cell, and images were acquired after 5 minutes of incubation without rinsing the SUVs. All of the AFM measurements for CM and NM were performed in a fluid cell at a room temperature of ∼21°C.

Whole Lens Preparation for AFM Experiments

The bovine lenses were carefully removed from the eye globes using a sharp scalpel and scissors. A solution of 5% agarose was heated, and melted agarose was poured into a custom-cut glass beaker, followed by stabilization of the lens by gently immersing it in the gel as the agarose started to solidify (34°–38°C) with the anterior portion of the lens on the top followed by immediately submerging the lens and gels in DMEM to preserve hydration following the approach described previously.⁵⁰ DMEM media were poured to cover the immobilized lenses completely during the AFM experiments.

AFM Measurements

CM and NM were imaged in fluid phase (buffer B) using the PeakForce quantitative nanomechanical mapping (PF-QNM) mode on a Bruker MultiMode 8-HR AFM equipped with a 32-bit NanoScope V controller (Bruker, Billerica, MA, USA), and force curves were acquired to study the mechanical properties of the membranes. Commercially available Bruker DNP-S probes with spring constants of ∼0.5 to 0.7 N/m (estimated utilizing the built-in thermal tune method) and probe end radius of ∼16 nm (estimated by analyzing height images of the Bruker Ti-roughness characterizer, PFQNM-SMPKIT) were used for obtaining images and force curves. Images were collected with a setpoint of 0.6 nN at a 1-Hz scan rate and 384 × 384 lines by samples per line, and force curves were collected with a trigger threshold of 10 nN. Force curves were collected with more than 100 nm spacing between each location on the SLM, and the mechanical properties of the SLM were evaluated.

To measure the mechanical properties of the SLM, we fit the force in the elastic regime of the force curves using the following equations^{30,32,51–53} to obtain K_A and the Young's modulus (E_m):

$$\begin{array}{c}\hfill F=\pi K_{A}R{\left(\frac{D-s}{s}\right)}^2\end{array}$$ (1)

$$\begin{array}{c}\hfill F=\frac{16}{9}{E}_m{R}^{1/2}{\delta }^{3/2}\left(1+0.884\rho +0.781{\rho }^2+0.386{\rho }^3+0.0048{\rho }^4\right)\end{array}$$ (2)

where K_A is the membrane compressibility modulus and is a measure of membrane stiffness, R is the AFM tip-end radius, D is the distance from the mica surface to the initial point of contact between the AFM tip and the membrane, and s is the tip–mica separation distance. Similarly, E_m is Young's modulus and is a measure of membrane stiffness, δ is indentation depth defined by D-s. Here, ρ is a dimensionless parameter defined by $\sqrt{R\delta }/h$, where h = D – t_w, with t_w being the water layer thickness, assumed to be 2 nm.⁵⁴ The complementary approaches to finding the membrane stiffness (K_A and E_m) by fitting the force curve data with Equation 1 and Equation 2 (Hertz model) are included to ensure scientific rigor in our analysis. Furthermore, E_m estimated from Equation 2 enabled us to correlate with the whole bovine lens stiffness (E_L).

Force curves on the whole bovine lenses were obtained using a Bruker Dimension FastScan Bio AFM, equipped with a 32-bit NanoScope V controller, at a room temperature of ∼21°C. Force curves on the lenses were collected under DMEM media at more than 10 different locations on top of the anterior region of each lens, with at least 10 force curves being collected in each location. Force curves were obtained using factory-calibrated Bruker SAA-SPH AFM probes (spring constant, ∼0.2 N/m, as quantified via laser Doppler vibrometry; tip end radius, 10.25 µm, as measured via scanning electron microscope inspection) with a 50-nm trigger threshold limit using PF-QNM mode. Force curves were processed using AtomicJ software,⁵⁵ and the Young's modulus was estimated using a Hertz contact mechanics model:

$$\begin{array}{c}\hfill F=\frac{4}{3\left(1-{\nu }^2\right)}{E}_L{R}^{1/2}{\delta }^{3/2}\end{array}$$ (3)

where E_L is the Young's modulus and is a measure of lens stiffness, R is the tip–end radius, ν is the Poisson's ratio (taken as 0.5 for soft biological materials such as the bovine eye lens), and δ is the indentation depth defined by the system.

Statistics

The measurements from at least three independent experiments were used to estimate the breakthrough (BT) force, K_A, and E_m averages, expressed as mean ± SD throughout this manuscript. The depth of the membrane patches was estimated using the rotating box function in NanoScope Analysis 1.9 software to measure at least four boundary locations of the membrane edge in each AFM topography image, with at least three images from three independent experiments used. For E_L measurements, force curves were obtained in 10 different locations on each sample. About 10 force curves were collected at each location (i.e., ∼100 force curves in total) for each lens to test the statistical significance of differences in lens stiffness at ∼24 hours and ∼48 hours after animal sacrifice. Statistical significance between the measured parameters was tested using Student's t-test, with P ≤ 0.05 considered statistically significant unless otherwise stated.

Results

Topography and Breakthrough Forces

After incubating for ∼40 minutes and flushing the unfused SUVs, the images of CM and NM were collected at several different locations. The surface topography of SLMs prepared from nuclear and cortical total lipids was smooth but included some random aggregation or unfused vesicles, as shown in Figures 1A and 1B.

Figure 1.

(A–E) The surface topography (A, B), collected force curves (C, D), and average BT force (E) of the supported CM (A, C, E) and NM (B, D, and E) isolated from bovine lens. Smooth topography with few bright structures was obtained for the CM and NM (A, B). The force curves displayed in C and D were obtained from the CM and NM, respectively, from one representative experiment. The discontinuity on the force curves, indicated by the black arrows in C and D, represents the BT events, and the corresponding force is the BT force. The BT force is presented as the mean ± SD from at least three independent experiments (E). Scale bar: 1 µm (A, B). Color scale is from –12 nm (dark) to +10 nm (bright). *P < 0.05.

The collection of force curves obtained in the CM and NM is shown in Figures 1C and 1D. The average BT force of the NM was significantly lower than that of CM, measured to be 2.64 ± 0.49 nN and 4.39 ±0.95 nN, respectively. The BT force is defined as a discontinuity in the force curve corresponding to (occurring at) the local maximum in the force along the force–distance curve as the tip approaches the bilayer/substrate. This is in contrast to the local minimum discontinuity observed at shorter separation when the probe pushed through the membrane and came into contact with the underlying mica substrate.

Mechanical Properties of CM and NM

We used the elastic regimes of the force curves from respective membranes to evaluate the mechanical properties of the CM and NM from bovine lenses. We estimated the membrane compressibility modulus (K_A) by assuming a spherical tip end and freestanding membrane and then quadratically fitting the AFM tip force in the elastic regime of the force curves with tip–mica separation distance s using Equation 1. To avoid the substrate effect, we used the force in the elastic regime from the point of contact to 80% of the BT force with the similar approach we described previously,^32,33 where the maximum indentation depth would be ∼20% of the sample thickness. A representative fit is shown in Figure 2A. K_A for the CM and NM were estimated to be 199.58 ± 59.66 mN/m and 72.9 ± 22.7 mN/m, respectively, as shown in Figure 2C.

Figure 2.

Mechanical properties of the CM and NM. (A) An example of a fit (red circles) of the membrane compressibility modulus (K_A) superimposed on a force curve (black line) obtained for a CM. (B) An example of a fit of Young's modulus (E_m) for a force curve obtained on a NM. The black line represents the raw data, and the red circles are the fit. (C, D) K_A and E_m values of the CM and NM as mean ± SD obtained from at least three independent experiments. *P < 0.05. K_A and E_m quantify membrane stiffness.

Similarly, using the approach described in our previous study,^30,31 Young's modulus (E_m) of the SLMs was estimated from the elastic regime using Equation 2 with a similar fitting method as used for K_A. A representative fit is shown in Figure 2B. Based on at least three independent experiments, E_m values for the CM and NM were estimated to be 23.01 ± 6.20 MPa and 12.16 ± 3.40 MPa, respectively, as shown in Figure 2D.

Membrane Thickness

The thickness of the bovine lens CM and NM was determined by measuring the height from the mica to the top of the membrane patches, as shown in Figures 3A and 3B. The thickness of CM was found to be 5.37 ± 0.45 nm, and that of NM was 5.99 ± 0.62 nm, which is not a statistically significant difference.

Figure 3.

Thickness of supported CM and NM. (A) The thickness of the CM was determined from the CM patches by measuring the vertical distance from the mica substrate to the top surface of the membrane, as shown by the red cross-sectional line drawn across the exposed mica region (dark brown) and CM patch (light brown). (B) Similarly, NM thickness was determined from NM patches by measuring the distance from the mica substrate to the top surface of the NM patch, as shown by the red line drawn across the mica substrate (dark brown) and NM patch (light brown). (C) The thicknesses of the CM and NM are displayed as mean ± SD obtained from at least three independent experiments. Scale bar: 0.5 µm (A, B). Color scale is from –12 nm (dark) to +10 nm (bright); ns, not significant.

Bovine Lens Stiffness

Force measurements on the whole lens were performed at more than 10 locations, with at least 10 curves at each location. Force curves were obtained in the central region of the anterior portion of the lens to minimize the effect of the lens curvature. A typical force curve is shown in Figure 4A, where red circles show the fit to obtain E_L using the AtomicJ software package.⁵⁵ The representative distribution of E_L measured on the eye lens within ∼24 hours of animal sacrifice ranged from 12.61 to 69.11 kPa, whereas E_L measured at ∼48 hours of animal sacrifice ranged from 3.63 to 80.63 kPa, as shown in Figures 4C and 4D. The average E_L values measured at ∼24 hours and ∼48 hours of animal sacrifice from three independent experiments were 33 ± 15 kPa and 38 ± 16 kPa, respectively. The bovine lens stiffness was not altered significantly (P = 0.35) between ∼24 hours and ∼48 hours of animal sacrifice when stored at 4°C.

Figure 4.

Stiffness of whole bovine lens measured at different times after animal sacrifice. (A) An example of a Hertz model fit of the force curve obtained on a whole bovine lens. The black line represents the raw data and the red circles the fit. (B) Young's modulus (E_L) of the lenses is displayed as the mean ± SD from three independent experiments at ∼24 hours and three independent experiments at ∼48 hours after animal sacrifice. (C, D) Representative probability distributions (PDs) of the E_L value of the lenses at ∼24 hours (C) and at ∼48 hours (D) after animal sacrifice. The 292 force curves obtained for the bovine lens ∼24 hours after animal sacrifice from three independent experiments and 326 force curves for the bovine lens ∼48 hours after animal sacrifice from three independent experiments were used to estimate E_L. E_L is the measure of lens stiffness; ns, not significant.

Discussion

Whole lens stiffness has been measured using noninvasive and in vitro assessment via Brillouin microscopy,²⁰ ultrasound elastography,²¹ optical coherence tomography,⁵⁶ coverslip squeezing method,¹² dynamical mechanical analysis,¹⁷ and AFM.⁵⁰ The research conducted by the Ziebarth group mainly focused on the elastic properties of whole lenses of primate lenses and capsules using AFM.^50,57–61 In prior work of the Ziebarth group, a custom-built, modified AFM was used to measure the stiffness of whole mouse and monkey lenses using microindentation by an AFM probe.^50,57–61 In this study, we used a Bruker Dimension FastScan Bio AFM to conduct AFM cantilever-based nanoindentation, which enabled us to measure the stiffness at 10 different points in the central anterior region of the bovine lens with ∼1-mm separation and thus minimal or no influence from adjacent measurements. Furthermore, we used AFM to estimate the CM and NM stiffness for the first time, to our knowledge, thus elucidating the positive physiological role of high Chol and CBD in decreasing the lens membrane stiffness and correlating with the stiffness of the whole lens. The K_A and E_m for NM are significantly lower than those for CM (Figs. 2C, 2D), indicating that the NM containing high Chol content and CBDs is less stiff than the CM with low Chol content and absence of CBDs. This suggests that high Chol and CBDs in lens membranes decrease membrane stiffness. The significantly higher membrane stiffness of CM and NM (Fig. 2D) compared to the stiffness of the whole lens (Fig. 4B) suggests that slight modulation in CM and NM stiffness may play a crucial role in altering the overall lens stiffness. We also estimated the stiffness of the bovine whole lens using AFM at ∼24 hours and ∼48 hours after animal sacrifice. Our E_L values (∼35 kPa) for an ∼2-year-old bovine lens (Fig. 4B) obtained with the AFM approach are comparable to E_L values reported earlier for the bovine lens using the microbubble approach.¹⁶ The Young's moduli of the bovine lens cortex and nucleus estimated using the microbubble approach are 2.9 kPa and 23.3 kPa, respectively.¹⁶ The Young's modulus estimated using the volume–strain procedure and creep properties for 2- to 5- to 6-year-old bovine lens capsule was reported to be 1.2 ± 0.08 MPa.⁶² These data suggest that the bovine lens capsule is significantly stiffer than the cortex and nucleus. Our estimation of Young's modulus using AFM for the whole bovine lens includes the intact lens capsule. In this scenario, a stiff covering (lens capsule) encases a soft, squishy tissue (lens cortex), likely leading to the cortex, followed by the nucleus being compressed, resulting in the measured E_L of the whole lens using AFM arising from the synergistic contributions of the cortex, nucleus, and capsule. The thickness of the bovine lens capsule is ∼60 µm^62,63 at the point where the AFM probe comes in direct contact with the lens capsule, and the lens capsule is thick compared to the penetration depth of the ramp/nanoindentation, suggesting the contribution of the lens capsule to measured E_L. The long-term goal of our research is to use the AFM approach to measure stiffness on a single human lens and correlate it with the stiffness of the human lens CM and NM. Experiments on single human lenses are significant because we can include sex, age, race, left versus right eye lens, and health history information (e.g., diabetes, radiation treatment, vitrectomy, smoking) provided by the Eye Bank in our analysis. Based on our experience working with human lenses,^48,64 obtaining human lenses from the Eye Bank within 24 to 48 hours postmortem is feasible. The AFM approach to measuring the stiffness of the single lens and correlating with the measured stiffness of the CM and NM reported in this manuscript forms the basis for estimating the stiffness of a single human lens within 24 to 48 hours postmortem and correlate with CM and NM stiffness, which we plan to pursue in our future research.

Most of the research on the mechanical properties of lenses is focused on whole lenses or tissues. The study of whole lens mechanical properties is paramount; however, the study of the contribution of lens lipid membrane in lens stiffness is equally crucial. In this work, we studied the mechanical properties of CM and NM prepared from a total lipid extract from a bovine lens. Lipid contents in lenses vary with species, age, lens region, and cataractous status of the lenses.^40–42,46 In bovine lenses, the phosphatidylcholine/sphingomyelin ratio ranges from ∼1.8 in the CM to ∼0.2 in the NM,⁶⁵ whereas the Chol/lipid ratio was found to be ∼1 in the CM to ∼2 in the NM.⁶⁶ The difference in the CM and NM lipid composition results in distinct mechanical properties of CM versus NM. The mechanical properties of the lens, predominantly from the nuclear region, are markers of the overall mechanical properties of the lens. A study of the physical properties of bovine lens lipid membranes by electron paramagnetic resonance (EPR) discriminated two domains in high Chol-containing NMs, one similar to the CM, known as the phospholipid cholesterol domain (PCD), and another domain known as the CBD. The profiles of hydrophobicity, order parameter, and oxygen transport parameters are almost identical in PCDs of the NM and CM, whereas the high Chol-containing CBDs exhibited a different pattern.⁶⁷ It has also been found that the thickness of the PCDs in both membranes is similar.⁶⁸ The increase of sphingomyelin in the membrane decreases the membrane mobility near the headgroup regions, while loss of Chol content increases membrane mobility near the headgroup regions.⁶⁹ During the interaction, Chol favors straight, stiffer hydrocarbon chains of saturated lipid types and interaction with neighboring hydrocarbon chains to favor extended conformations, leading to increased thickness.⁷⁰ Favoring the interaction with Chol, sphingolipids display longer, more saturated hydrocarbon chains, and their interaction increases membrane thickness.⁷⁰

In our previous experiments, with an increase in Chol content, we found a sharp rise in BT force until the Chol/phospholipid mixing ratio reached 1, with the BT force decreasing thereafter.^32,33 In the presence of high Chol, depending on the lipid composition, the formation of CBDs is observed at different Chol/lipid mixing ratios.^32,33,71 Based on the force curves, the sharp decrease in the BT force after the initial climb can be attributed to the formation of CBDs. A similar trend has been found in the mechanical property K_A of membranes consisting of variable Chol content. The decreased stiffness of the saturated lipid membrane with high Chol content after a peak was also observed by Al-Rekabi and Contera.⁷² CBDs are obtained only in the NM but not in the CM in bovine lens membrane.⁶⁷ Thus, the lower BT force, K_A, and E_m obtained in the NM can be attributed to the formation of CBDs. Previously, it has been reported that the Chol/lipid ratio is ∼1 for the calf lens membranes (young bovine lens), where the Chol/lipid ratio was estimated from the whole calf lens without separation of the cortex and nucleus.^73,74 For comparison, for the 2-year-old bovine lens with the separation of the cortex and nucleus, the Chol/lipid ratio was found to vary from ∼1 in the CM to ∼2 in the NM.^66,67 Based on these literature data and the findings reported in this manuscript, we can speculate that, for the calf lens, the Chol/lipid ratio is likely smaller in the CM compared to the NM, and a high Chol/lipid ratio in the NM might lead to the formation of CBDs which might lead to a decrease in NM stiffness and possibly protect against lens hardening. Lenses with cataracts contain lower Chol than clear ones, which is also seen in mouse eye lenses with reduced Chol biosynthesis, leading to cataract formation.⁷⁵ This observation indicates the significance of high Chol content in the lens membrane, particularly in the nucleus, to decreasing overall lens stiffness.

The lens membrane proteins, such as AQP0 and Cx46, are suggested to play a role in lens hardening.^34,35 Cytoskeletal structures in lenses also play a part in maintaining the mechanical properties of eye lenses.⁷⁶ The increased binding of crystallin proteins to the fiber cell plasma membrane might increase the lens stiffness.^30,31 The increase in stiffness of the eye lens is also promoted by oxidative damage in the crystalline lens.⁷⁷ It is likely that the oxidation of lipids and Chol decreases the Chol content in the membrane,^78–82 the lower Chol content results in a lower amount and size of CBDs,^40,41,71,78 a decrease in Chol and CBDs content increases binding of crystallins to membranes,^{27,30,47,48,69,83–85} crystallin–membrane binding leads to protein insolubilization,^25,27,86 and protein insolubilization leads to lens stiffness,^23,24 eventually resulting in presbyopia development. Thus, high Chol and CBD content in the lens membrane appear critical to decreasing lens stiffness and likely protecting against presbyopia. Furthermore, a significant difference in lipid and Chol composition in the cortex and nucleus of the lens likely alters membrane binding of crystallin, protein aggregation, and insolubilization into HMW proteins, and membrane/cytoskeleton–crystallin interactions differently in cortical and nuclear regions of the lens, resulting in the synergistic effect of cortex and nucleus stiffness toward the whole lens stiffness.