Lipid conformational order and the etiology of cataract and dry eye

Abstract

Lens and tear film lipids are as unique as the systems they reside in. The major lipid of the human lens is dihydrosphingomylein, found in quantity only in the lens. The lens contains a cholesterol to phospholipid molar ratio as high as 10:1, more than anywhere else in the body. Lens lipids contribute to maintaining lens clarity, and alterations in lens lipid composition due to age are likely to contribute to cataract. Lens lipid composition reflects adaptations to the unique characteristics of the lens: no turnover of lens lipids or proteins; the lowest amount of oxygen of any tissue; and contains almost no intracellular organelles. The tear film lipid layer (TFLL) is also unique. The TFLL is a thin (100 nm) layer of lipid on the surface of tears covering the cornea that contributes to tear film stability. The major lipids of the TFLL are wax esters and cholesterol esters that are not found in the lens. The hydrocarbon chains associated with the esters are longer than those found anywhere else in the body (as long as 32 carbons), and many are branched. Changes in the composition and structure of the 30,000 different moieties of TFLL contribute to the instability of tears. The focus of the current review is how spectroscopy has been used to elucidate the relationships between lipid composition, conformational order and function, and the etiology of cataract and dry eye.

Keywords: lens; meibum; membrane; spectroscopy.

Fig. 1

Top: Cross-section through a human eye. Bottom: Schematic of the human lens. Used with permission from LifeMap Sciences, Inc. (https://discovery.lifemapsc.com).

Fig. 2

Schematic of the TFLL on the surface of the cornea adapted from slideshare.net (https://www.pharmaccutical-journal.com).

Fig. 3

Cross-section through the eyelid. Sebum (red) is shown mixing with meibum (yellow), which forms a continuous TFLL film over the ocular surface and eyelid. From (32).

Fig. 4

Left: A typical membrane. Right: Human lens membrane. Typical membranes contain fluid lipids with relatively few cholesterol molecules (red cylinders). Human lens membranes are unique. Most of the lipid is associated with proteins such as α-crystallin (α-crystallin assembly shown as gray balls, one large ball and one small ball for each α-crystallin) and aquaporin, which limits their mobility. Human lens membranes are some of the most saturated ordered (stiff) membranes in the human body. The major lipid of the human lens is dihydrosphingomyelin (green shaded balls). Found in quantity only in the human lens. From (3).

Fig. 5

Schematic of phospholipids and the conformation of hydrocarbon chains that define lipid order. The more trans rotamers, the tighter the packing, the greater the van der Waal’s interactions between lipids, and the greater the lipid order (stiffness). The opposite is true for gauche rotamers. From (3).

Fig. 6

The relationship between lens sphingolipid content and hydrocarbon chain order. Hydrocarbon chain order reflects the structural stiffness of the hydrocarbon chain region of lipids in membranes. Clear human lens cortex and nucleus (closed square); cataractous human lenses (closed triangle). This figure was adapted from Figure 5 in (43). All the data except those related to cataractous lens lipid are from Borchman, Yappert, and Afzal (42). Cataractous lipid order information is extracted from Paterson et al. (45).

Fig. 7

Schematic of WE and CE packing from X-ray crystallography. A: Molecular size of CE and WE with 22 carbon hydrocarbon chains. B:) Potential lamellar packing of WE. B (top): Shows rhombic packing of the hydrocarbon chains. B (right): The trans orientation for ordered hydrocarbons, gauche rotamer orientations for disordered hydrocarbon chains. C: Smectic phase packing of CE. D: Speculative packing of a WE, CE, and phospholipid mixture on an aqueous surface. From (68).

Fig. 8

Lifespan versus lipid phase transition parameters from (74) (black stars). Data from (43, 74) (circles).

Fig. 9

Correlation between lens sphingolipid content and hydrocarbon chain order (stiffness) from (75). Black stars, pinniped lipid [data from (43, 74, 111, 112)]; filled squares, human lens nuclear lipid; open squares, human lens cortical lipids [from (1, 95, 96)]; open circles, lenses from various species [from (75)].

Fig. 10

Relationship between the molar amounts of lens sphingolipid and cholesterol. Pinnipeds from (75) and bowhead whale (100% SL) from (74) (black stars). Calf lens cortex and nucleus and 2- to 6-year-old cow from (104) and 1-year-old cow from (120) (open squares). Cow, sheep, human, rat, mouse, pig, and chicken from (17) (open circles). human lens from references (37, 106) (open triangles). Mice (10 and 45 days old) from (121) (open inverted triangles). Figure from (75).

Fig. 11

Tear breakup and BR are a measure of tear stability. Human meibum structural order was assessed by quantifying hydrocarbon chain order measured in vitro using infrared spectroscopy and (C_s⁻¹)_max, reciprocal compressibility modulus, measured using Langmuir trough technology. A: Changes in TBUT with age. The first bar is from (126); the last three bars are from (127). B: Changes in BR with age. The first bar is from (48, 128); the middle two bars are from (48); the last bar is from (129). C: Relationship between tear film breakup time and BR from a cohort of 28 year olds (129). D: The first and third bars are from (64); the second and last bars are from (47, 50, 61). E: A larger reciprocal compressibility indicates a stiffer more elastic lipid layer. Filled circles and squares are from (130); filled triangles are from (67). Data are the average ± the standard error of the mean. The number of subjects are in parentheses.

Fig. 12

Structural functional relationships and dry eye. Tear breakup and BR are a measure of tear stability. Human meibum structural order was assessed by quantifying hydrocarbon chain order measured in vitro using infrared spectroscopy and (C_s⁻¹)_max, the reciprocal compressibility modulus, measured using Langmuir trough technology. A: Tear film stability measured by the BR. The first and last bars are from (131); the middle bar is from (132) and is for Parkinson’s patients at stages 1 and 2 receiving dopamine agonist therapy. B: Tear film stability measured by noninvasive TBUT (NTBUT). The first bar is from (126); the second bar is from (127); the last two bars are from (133). C: Hydrocarbon chain order, a measure of lipid stiffness using infrared spectroscopy. The first bar is from (47, 50, 56) for donors 68 ± 8 years old; the second bar is from (63, 128) for donors 66 ± 6 years old; the third bar is from (41, 58) for donors 54 ± 2 years old; the last bar is from (55) for donors 66 ± 10 years old. D: Reciprocal compressibility modulus, a measure of TFLL elasticity or stiffness (130). E: A pilot study showing how when dry eye symptoms are ameliorated with treatment, lipid order is restored (51). Data are the average ± the standard error of the mean. The number of subjects is in parentheses. MGD, Meibomian gland dysfunction; HSCT, dry eye associated with hematopoietic stem cell transplantation.

Fig. 13

A: A training set was used to discriminate between the spectra of meibum from normal donors and the spectra of meibum from donors with meibomian gland dysfunction. This shows that the infrared spectra must contain compositional and structural information about the changes that occur with meibomian gland dysfunction. A score above 50 (vertical line) is considered a sample with meibomian gland dysfunction. B: The same training set used for A was used to predict the age of the donors with meibomian gland dysfunction within ±20 years at a 95% confidence limits. M_n group (filled circles); Md group (open circles); linear regression fit (solid line); and 95% confidence limits of M_n group (dashed lines). From (53).

Fig. 14

Relationship between phase transition temperature and hydrocarbon chain saturation. Samples measured in this study: oo, oleyloleate; po, palmityloleate; pp, palmitylpalmitate; sp, sterylpalmitate; pppo, equal molar mixture of pp and po; spoo, equal molar mixture of sp and oo. HL, human lens lipid; ROS P, bovine rod outer segment plasma membrane,;SR F, fast twitch rabbit muscle sarcoplasmic reticulum membrane; SR S, slow twitch rabbit muscle sarcoplasmic reticulum membrane. Least squares linear regression fit to all of the data is denoted by the diagonal line. From (58).

Fig. 15

Correlation between the lipid phase transition temperature and lipid order at 33.4°C for human meibum. Meibum from donors without dry eye (filled circles); meibum from donors with dry eye and hematopoietic stem cell transplantation (open circles); meibum from donors with meibomian gland dysfunction (open stars). From (47).

Fig. 16

CE/WE molar ratios calculated from the NMR spectra of meibum. Solid bars: Molar ratios calculated from the intensity of the CE resonance at 4.6 ppm and the WE resonance at 4.0 ppm. Correcting for (O)-acylated ω-hydroxy fatty acids the solid bars would be lower. For instance, the value for Meibomian gland disfunction corrected would be 31, lower than the reported value of 34. Open bars: Molar ratios calculated from the intensity of the CE resonances from cholesteryl carbon numbers 18 and 19 and the WE resonance at 4.0 ppm. From (68).

Fig. 17

Examples of straight chain hydrocarbons and branched chain hydrocarbons. From (194).