Search

Purpose To determine the effects of prolonged cryopreservation at subzero-degree temperatures on corneal transparency and histology after treatment with preservation medium containing the phosphodiester glycerylphosphorylcholine (GPC). Methods Rabbit corneas (n = 30) were immersed for 3 hours in K-Sol preservation medium containing 30 mM GPC. Three groups with 6 corneas each were refrigerated at -8°C for 2 weeks and liquid nitrogen temperature for 2 and 6 weeks, respectively. Two groups with 6 corneas each immersed in K-Sol preservation medium only were refrigerated at -8°C for 2 weeks and liquid nitrogen temperature for 6 weeks, respectively. Postthawing corneal transparency was measured on a grading scale after which corneas were prepared for and analyzed by light and transmission electron microscopy. Results All 3 groups of corneas preserved with GPC maintained a greater degree of corneal transparency compared with corneas preserved without GPC. The number of corneas retaining epithelial and endothelial layers increased in all groups where corneas were preserved in medium containing GPC, in contrast to corneas preserved in medium without GPC. Cytoplasmic vacuolization or nuclear damage was greater in corneas preserved without GPC. Similar findings were found in corneas stored at -8°C and liquid nitrogen temperatures. Conclusions This study demonstrates a cryoprotective effect of corneas preserved in K-Sol containing the phosphodiester GPC at subzero-degree temperatures. In corneas immersed in preservation medium containing GPC, a higher degree of transparency is maintained and a lesser degree of histopathologic changes is observed with storage at both -8°C and in liquid nitrogen.

This study documents the absorption of glycerylphosphorylcholine (GPC) into corneas ex vivo. Corneas in quadruplicate were incubated in preservation medium containing 30 mM GPC, which is used as a reference marker. The GPC reference marker is used to calibrate 31P nuclear magnetic resonance (NMR) spectral chemical-shift positions for identification of phosphatic metabolites and to calculate intracorneal pH in intact tissues ex vivo. Following baseline NMR ex vivo analysis, corneas were stored in eye bank chambers in preservation medium containing 30 mM GPC at 4 °C overnight for 8 h. After returning to room temperature, NMR analysis was repeated on the same corneas in fresh GPC-free preservation medium. NMR analysis also was performed on the 30 mM GPC preservation medium alone from the eye bank chambers for detection of the GPC signal. The elevated GPC signal unexpectedly persisted in corneas incubated at 4 °C overnight even though GPC was not present in the fresh GPC-free preservation medium. In fact, the concentration of GPC in the intact cornea was many times higher than that found in the cornea endogenously. The levels of phosphatic metabolites and the energy modulus, after subtracting the spectral contribution of the 30 mM exogenous GPC, as well as the intracorneal pH remained unchanged from pre-refrigeration analyses. Corneas also retained transparency through the time-course of this study irrespective of temperature or change in temperature. The GPC signal in the NMR analysis of the preservation medium from the eye bank chambers was nearly undetectable. GPC was unexpectedly absorbed into the corneal tissue without detectable metabolic or physical toxicity. The intracorneal uptake of GPC at reduced temperatures parallels the increase in GPC that occurs naturally in muscle tissue in animals during wintering periods and the very high concentration of GPC in sperm, a cryogenically compatible cell, suggestive of a potential role for GPC in cryopreservation.

The hypothesis proposed herein is presented to explain the unexpectedly high concentration of ATP and provide evidence to support its hydrotropic function in the crystalline lens determined using 31P NMR. The lens, historically considered to be a metabolically quiescent organ, has the requisite machinery to synthesize ATP, such that the homeostatic level is maintained at about 3 mM. This relatively high concentration of ATP has been found to be consistent among multiple mammalian species including humans. This millimolar quantity is many times greater than the micromolar amounts required for the other known functions of ATP. The recent postulation that ATP at millimolar concentrations functions as a hydrotrope in various cell/tissue homogenates preventing protein aggregation coupled with observations presented herein, provide support for extending the hypothesis that ATP functions as a hydrotrope not only in homogenates but in an intact functioning organ, the crystalline lens. Concentrations of ATP of this magnitude are hypothesized to be required to maintain protein solubility and effectively prevent protein aggregation. This concept is important considering protein aggregation is the etiology for age-related cataractogenesis. ATP is a common ubiquitous intracellular molecule possessing the requisite hydrotropic properties for maintaining intracellular proteins in a fluid, non-aggregated state. It is proposed that the amphiphilic ATP molecule shields the hydrophobic regions on intralenticular fiber cell protein molecules and provides a hydrophilic interfacial surface comprised of the ATP negatively charged triphosphate side chain. Evidence is presented that this side chain is exposed to and has been reported to organize intracellular interstitial water to form an interfacial rheologically dynamic water layer. Such organization of water is substantiated with the effect of deuterium oxide (heavy water) on ATP line widths of the side chain phosphates measured ex vivo by 31P NMR. A novel model is presented to propose how this water layer separates adjacent lens fiber cell proteins, keeping them from aggregating. This hypothesis proposes that ATP can prevent protein aggregation in normal intact lenses, and with declining concentrations can be related to the disease process in age-related cataractogenesis, an affliction that affects every older human being.

The inner border of the eyelid margin is critically important for ocular surface integrity because it guarantees the thin spread of the tear film. Its exact morphology in the human is still insufficiently known. The histology in serial sections of upper and lower lid margins in whole-mount specimens from 10 human body donors was compared to in vivo confocal microscopy of eight eyes with a Heidelberg retina-tomograph (HRT II) and attached Rostock cornea module. Behind the posterior margin of the Meibomian orifices, the cornified epidermis stopped abruptly and was replaced by a continuous layer of para-keratinized (pk) cells followed by discontinuous pk cells. The pk cells covered the muco-cutaneous junction (MCJ), the surface of which corresponded to the line of Marx (0.2-0.3 mm wide). Then a stratified epithelium with a conjunctival structure of cuboidal cells, some pk cells, and goblet cells formed an epithelial elevation of typically about 100 μm initial thickness (lid wiper). This continued for 0.3-1.5 mm and formed a slope. The MCJ and lid wiper extended all along the lid margin from nasal to temporal positions in the upper and lower lids. Details of the epithelium and connective tissue were also detectable using the Rostock cornea module. The human inner lid border has distinct zones. Due to its location and morphology, the epithelial lip of the lid wiper appears a suitable structure to spread the tear film and is distinct from the MCJ/line of Marx. Better knowledge of the lid margin appears important for understanding dry eye disease and its morphology can be analysed clinically by in vivo confocal microscopy.

The cat, albeit an expensive animal to breed and maintain, affords a relatively large eye ideal for studies on the embryonic retina. This study reports on the efficacy of serial removal of feline foetuses for studies on the histogenesis of the retina. Single feline foetuses were serially removed from the uterus of pregnant cats to determine if there were histologic changes in the foetal retina of the remaining foetuses during subsequent development. The retinas of neonates from mothers that had undergone hysterotomies were compared by light and transmission electron microscopy to those from animals whose mother's had not had hysterotomies. No histologic differences in the retinas were detected between the 2 groups. The duration of gestation was unaffected. This study indicates that it is possible to obtain embryonic retinal tissue at more than one stage of development from the same pregnant female without inducing histologically observable anomalies.

Objectives. The lid wiper is defined as that portion of the marginal conjunctiva of the upper eyelid that wipes the ocular surface during blinking. The purpose of this study was to investigate whether lid wiper epitheliopathy occurred with patients who reported dry eye symptoms, yet had normal fluorescein breakup time (FBUT) and Schirmer test values and an absence of fluorescein corneal staining. Methods. One hundred patients were divided into two groups based on the presence or absence of dry eye symptoms, as determined with the Standard Patient Evaluation of Eye Dryness questionnaire. Other criteria for admission to both groups were FBUT of 10 seconds or more, Schirmer test value of 10 mm or more, and absence of fluorescein corneal staining. After instillation of fluorescein and rose bengal dyes, the lid wipers of 50 asymptomatic and 50 symptomatic patients were graded for staining from grade 0 (absent) to grade 3 (severe). Results. Of the symptomatic patients, 76% had staining of the lid wiper: 44%, grade 1; 22%, grade 2; and 10%, grade 3. Of the asymptomatic patients, 12% had staining; 8%, grade 1; 4%, grade 2; and 0%, grade 3. The difference in prevalence of lid wiper staining between the symptomatic and asymptomatic groups was significant (P0.0001). Conclusions. Lid wiper epitheliopathy, diagnosed by staining with fluorescein and rose bengal dyes, is a frequent finding when symptoms of dry eye are experienced in the absence of routine clinical dry eye findings.

The phospholipids of the skin are difficult to quantify because they represent only a small fraction of the skin tissue. In this study, 31 P nuclear magnetic resonance, which permits precise profiling of these phospholipids, was used to compare the phospholipids of upper eyelid epidermal and dermal lipid extracts (n = 13 profiles). Phospholipid profiles included alkylacylphosphatidylcholine (AAPC), dihydrosphingomyelin (DHSM), diphosphatidylglycerol (cardiolipin), ethanolamine plasmalogen (EPLAS), lysophosphatidylcholine, phosphatidic acid, phosphatidylcholine (PC), phosphatidylethanolamine, phosphatidylinositol, phosphatidylserine, sphingomyelin, and uncharacterized phospholipids (U1 and U2, particularly enriched in the epidermis). The computed phospholipid metabolic index (n = 86 indexes) findings can be summarized as follows: a lower content of the enol and ether phospholipids in the epidermis relative to the dermis, internal compensation among the component phospholipids so as to maintain the choline functional group ratio, and a greater concentration of hydroxyl-containing functional groups in the epidermis. A membrane index (f mem ) value of -0.37 for the epidermis deviated considerably from the value of -0.06 characteristic of living membranes and the dermis. The production of the reduced phosphatides, EPLAS and AAPC, indicates the use of alternative pathways between the two tissues. Relative to the dermis, increased PC in the epidermis coupled with decreased DHSM, EPLAS, and AAPC are factors enabling the epidermis of eyelid tissue to be an effective water barrier.

The purpose of this study was to determine the relationship between the tear film and humidity by examining whether alterations in periocular humidity influence the thickness of the tear film lipid layer. Thirteen dry eye subjects presenting with a baseline lipid layer thickness of < or = 60 nm were fitted with modified swim goggles in which the right eye (OD) was exposed to conditions of high humidity and the left eye (OS) remained exposed to ambient room conditions. The lipid layer was monitored over a 60-min time course with goggles on and for an additional 60 min following goggle removal. The OD lipid layer increased significantly in thickness within 5 min of exposure to conditions of high humidity (p < 0.0001), reaching a maximum increase of 66.4 nm after 15 min of goggle wear (p < 0.0001). This maximum increase to a lipid layer thickness of 120.5 nm was maintained at the 30- and 60-min goggle time points. No significant change was detected OS. Following goggle removal, OD values declined but remained significantly elevated over the OS lipid layer thickness throughout the 60-min postgoggle period. Moderate to total relief of dry eye symptoms was reported during goggle wear and generally persisted at a reduced level for 1-3 h following goggle removal. Increased periocular humidity results in an increase in tear film lipid layer thickness, possibly by providing an environment that is more conducive to the spreading of meibomian lipid and its incorporation into the tear film.

The content of the meibomian gland lipid exprimate is known, but little is known about the phospholipids that comprise the glandular cells. The purpose of the present study is to identify and quantitate the phospholipid complement of the meibomian gland cells that produce the lipid secretion of meibomian oil and which is vital to tear film stability. Eyelids (n = 50) were excised from rabbits, and after surgical removal of surrounding tissues, the tarsal plates with and without expressing meibomian oil were extracted and phospholipids of the plates quantified by 31P nuclear magnetic resonance (NMR). Seventeen phospholipids were quantified from tarsal plates expressed of oil and tarsal plates containing meibomian oil: alkylacylphosphatidylcholine (AAPC), dihydrosphingomyelin (DHSM), dimethylphosphatidylethanolamine, diphosphatidylglycerol (cardiolipin), ethanolamine plasmalogen (EPLAS), lysoethanolamine plasmalogen, lysophosphatidylcholine, lysophosphatidylethanolamine, lysophosphatidylserine, phosphatidic acid, phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylglycerol, phosphatidylinositol, phosphatidylserine, sphingomyelin (SM), sphingosylphosphorylcholine. The six zwitterionic and neutral phospholipids, DHSM, EPLAS, PE, SM, AAPC, and PC together comprise 79.5% of the total meibomian gland phospholipid profile (in meibomian oil this value is 84.2%). The zwitterionic and neutral phospholipids dominate meibomian gland phospholipid profiles. Since the meibomian gland cells undergo holocrine secretion and form the meibomian glad secretion, such a composition is consistent with the hypothesis that a chemically stable lamellar surfactant layer phospholipids bind non-polar meibomian oil to the aqueous layer of the tear film.

Phospholipids of optic nerve (n = 30) from 5.6-kg rabbits were analyzed by 31P nuclear magnetic resonance (NMR) spectroscopy. Phospholipid metabolites detected were as follows (mol %): phos

This study establishes the metabolic similarities and differences of intact abattoir corneas and the human cornea ex vivo, using 31P nuclear magnetic resonance spectroscopy. Phosphorylated sugars (SP), inorganic orthophosphate (Pi), phosphodiesters, ATP, ADP, and the nucleoside diphosphosugars and dinucleotides (NS&DN) were quantitated. The intracorneal pH also was determined. In addition, metabolic indices ATP/Pi, ATP/ADP, ATP/ (ADP + Pi), SP/Pi, the phosphorylation potential, and the 31P energy modulus were calculated. Significant differences were observed between the abattoir and human corneas in phosphorylated sugars, NS&DN, and intracorneal pH, as well as in the indices ATP/ADP and SP/Pi. The overall energy status as measured by the 31P energy modulus was significantly higher in the bovine cornea when compared to the porcine and human corneas, suggestive of a very high energy reserve in the bovine cornea.

From three groups of patients with corneal edema, the degree of edema was clinically quantified prior to penetrating keratoplasty. Corneas from patients with failed grafts (n = 16) had the greatest edema (mean score 3.8 +/- 0.6 SE), corneas with Fuchs' endothelial dystrophy (n = 12) had the least edema (1.4 +/- 0.4), and corneas with aphakic and pseudophakic bullous keratopathy (n = 40) had intermediate edema (2.7 +/- 0.5). Phosphorus-31 nuclear magnetic resonance spectroscopy was used to quantitate five high-energy phosphates and nine low-energy phosphates and calculate 12 metabolic indices from the specimens of these three edematous groups. Corneas with aphakic and pseudophakic bullous keratopathy had lower alpha-glycerophosphate (alpha-GP) and glycerol 3-phosphorylethanolamine and glycerol 3-phosphorylcholine (GPE & GPC), and higher phosphocreatine (PCr), adenosine diphosphate (ADP), monoesters/diesters, PCr/Pi (phosphocreatine/inorganic orthophosphate), and PCr/ATP (phosphocreatine/adenosine triphosphate) than corneas with Fuchs' endothelial dystrophy (p less than 0.05). Corneas from failed grafts had lower alpha-GP, ethanolamine phosphate (EP), ethanolamine phosphate and choline phosphate (EP&CP), monoesters/Pi, ATP/Pi, and energy modulus, no detectable GPE and GPC, and higher Pi, phosphorylglycans (PG), PCr, ADP, diesters, PCr/Pi, and PCr/ATP than those with Fuchs' endothelial dystrophy (p less than 0.05). Corneas from failed grafts had lower ATP, monoesters/Pi, ATP/Pi, PCr/Pi, and energy modulus and higher Pi, PG, ADP, diesters, and PCr/ATP than corneas with aphakic and pseudophakic bullous keratopathy (p less than 0.05). In summary, there was increasing decline in the high-energy metabolites and energy status with increasing edema.

This study compares porcine scleral and corneal phosphorylated intermediate metabolites determined using 31P NMR. These tissues comprise the tunica fibrosa (outer coat) of the eye. Since the sclera does not possess epithelia as does the cornea, comparative analysis of these tissues included examination of the cornea with and without its epithelia. The phosphorylated intermediates detected include: dihydroxyacetone phosphate, hexose 6-phosphates, alpha-glycerol phosphate, beta-glycerol phosphate, ethanolamine and choline phosphates, nucleoside mono-, di-, and tri-phosphates, inorganic orthophosphate, glycerol 3-phosphoryl-ethanolamine and -choline, phosphoglycans, phosphocreatine, nucleoside diphosphosugars, dinucleotides, and four uncharacterized (unknown) signals. Metabolic indices, comprised of individual or grouped metabolites, were calculated to further compare and contrast metabolites and to provide more pathway specific metabolic interrelations for phosphorylated metabolites. Significant differences exist between the corneal stroma and the sclera in 16 of the 22 phosphorylated metabolites determined, whereas differences exist between the whole cornea and the sclera in six of the 22 phosphorylated metabolites. Considering all metabolite levels and metabolic indices in aggregate, the sclera is most similar to the whole cornea and least similar to the corneal stroma.

Ocular humors (aqueous humor and vitreous body) were obtained from porcine eyes. Perchloric acid extracts were prepared from aqueous and vitreous aspirates, and phosphorus-31 nuclear magnetic resonance spectroscopy 31P-NMR performed. Resonance signals from four high-energy phosphate groups (phosphocreatine, adenosine tri- and diphosphate and the dinucleotides) and nine low-energy phosphate groups [dihydroxyacetone phosphate, hexose-6-phosphate, alpha-glycerophosphate, fructose-1,6-diphosphate, ethanolamine phosphate, choline phosphate, inorganic orthophosphate (Pi), glycerol-3-phosphorylethanolamine and glycerol-3-phosphorylcholine] were present in the aqueous humor and/or vitreous body. Vitreous body phosphates significantly elevated relative to that of the aqueous humor were: DHAP, Hex 6-P, alpha-GP, Fru 1,6-diP, CP, GPE, PCr, ATP, ADP, and the dinucleotides; only Pi was significantly lower. Six metabolic indices (monoesters, diesters, monoesters/diesters, monoesters/Pi, ATP/Pi, energy modulus) were determined; all vitreous body indices were significantly higher than those of the aqueous humor. In particular, the energy modulus, a measure of overall metabolic energy status, was higher in the vitreous body (0.121 +/- 0.047) than in the aqueous humor (0.05 +/- 0.021).

Purpose This study correlates measurement of lipid layer thickness (LLT) with two frequently used dry eye tests, fluorescein break-up time (FBUT) and Schirmer's test with anaesthesia (STA). Methods Subjects (n=44 eyes) with symptoms of dry eye and positive results for dry eye with either FBUT or STA or both were selected. Quantification of LLT was performed by the observation of colour interference patterns in zones of specular reflection using a custom-designed instrument. Results All correlations among pairs of tests were strong and exhibited a significance of P

Purpose To compare the quantity of fluorescein delivered to the eye via fluorescein-impregnated paper strips of various sizes and surface areas and via various microliter volumes of fluorescein sodium using an in vitro assay. Methods A commercially available fluorescein-impregnated strip (75 mm2) and three modified strips of reduced fluorescein-impregnated surface areas (10, 7.5, and 5.0 mm2) were used. The amount of fluorescein delivered to the eye for each of the four strips was approximated by applying each strip to a Whatman No. 1 filter paper under conditions simulating application of the strip to the eye, extracting the fluorescein from the filter paper in an aqueous solution, and performing spectrophotometric analysis at 484 nm. Similarly, this filter paper analytical system was calibrated using 1, 2, and 3 microl volumes of 2% w/v fluorescein delivered to the filter paper. Results Using calibration curves, linearity was observed between absorbance and concentration of fluorescein sodium with an R2 value > or = 0.99. Using these calibration curves, the amount of fluorescein delivered to the eye for the four strips and the three fluorescein solution samples was determined. Fluorescein-impregnated strips with surface areas of 75, 10, and 5 mm2 delivered approximately the same quantity of fluorescein to the ocular surface as 3 microl, 1 microl, and 0.5 microl of fluorescein 2% solution, respectively. Conclusions The surface area of the fluorescein-impregnated portion of the strip can be designed to control the amount of fluorescein delivered to the eye.

Tear film stability has been a common denominator among conditions causing irritation of the ocular surface since Holly’s report in 1973.1 Pflugfelder et al.2 noted “An unstable tear film is the hallmark of various dry eye states.” The NEI/Industry Workshop on Clinical Trials in Dry Eyes reported breakup time (BUT) to be the only clinical test available to measure tear film stability.3 The fluorescein BUT (FBUT) test was introduced as corneal wetting time by Norn4 in 1969 and subsequently refined into the BUT test by Lemp5 in 1973. Although a reduced BUT is considered a global criterion of dry eye,3 it is criticized as not accurate or reproducible, and is not universally accepted.6–9 Inherently, the FBUT test is invasive, thus compromising its reproducibility.7,9

The lipid layer of the tear film is composed of meibomian gland secretions. The current physical-chemical model for the lipid layer consists of a thick nonpolar hydrophobic outer layer, the bulk of which is composed of sterol esters, monoester waxes, long-chain alkane hydrocarbons and triglycerides, with a surfactant interfacial layer formed by phospholipids having polar hydrophilic head groups and hydrophobic tails.

To compare the repeatability of fluorescein break-up time (FBUT) measurements determined with either a standard fluorescein strip or the Dry Eye Test (DET) modified fluorescein strip methods.This was a prospective, randomized contralateral study of 100 patients, in which FBUT measurements were determined with a standard FUL-GLO fluorescein strip (Akorn, Inc., Buffalo Grove, IL, U.S.A.) in one eye and a DET strip (Akorn, Inc., Buffalo Grove, IL, U.S.A.) for the contralateral eye. Three consecutive measurements were made immediately after fluorescein instillation. The second eye was evaluated 1 minute after completion of the first eye. Data from patients with FBUT values less than 20 seconds were included in the data analysis, because measurements greater than 20 seconds are not diagnostically significant.Seventy-five patients met enrollment and FBUT measurement criteria. For three consecutive FBUT measurements, the DET values were within 3 seconds for 72 of the 75 patients (96%). Eighty percent of patients reported no sensation with the DET strip, 20% reported mild sensation, and no patient reported moderate sensation. With FUL-GLO strips, measurements were within 3 seconds for 53 of the 75 patients (71%) (p0.005). The standard fluorescein strip method elicited reports of no sensation from 31% of patients, mild sensation from 60%, and moderate sensation from 9% (p0.001).The DET strip provides a significant reduction in sensation upon application, improved single measurement reliability, and enhanced measurement precision, compared with a conventional fluorescein strip.

•Background: Since a potential exists for untoward effects on the cornea from the high magnetic fields and radio-frequency energies, and the further manipulation required for phosphorus-31 magnetic resonance spectroscopy (31P-MRS), we determined the effects of this technology on tissues using paired human corneas (n=4) meeting criteria acceptable for transplantation. • Methods: Slit-lamp biomicroscopy, pachometry, specular microscopy, and redux fluorophotometry were performed on all corneas. One cornea of each pair was examined (

Meibomian gland secretion (meibom) in human1–5 and rabbit6,7 comprises triglycerides, neutral lipids, and polar lipids. Phospholipids have been further identified and quantified.8 Collectively, these lipids are believed to form the lipid layer of the tear film, which can be observed in the interpalpebral fissure by interferometry.9–11

The lipid layer is extremely important to tear film stability1 as it is believed to prevent excess evaporation from the ocular surface. The meibomian glands of the upper and lower eyelids are considered to be the principal sites of source lipids destined for the tear film.2,3 These meibomian secretions in human4–8 and rabbit9–11 are composed of triglycerides, neutral lipids, and polar lipids. The phospholipid component of the latter group has also been analyzed.11

A large variation in the degree of fluorescence is observed clinically when using fluorescein dye for the evaluation of the ocular surface. Although this variation may not be so important in the case of diagnosing an epithelial defect, such as in a corneal or conjunctival abrasion, it becomes very important when attempting to evaluate the ocular surface in the presence of various degrees of corneal defects that occur in dry eye syndromes. For example, clinically there may be a high degree of variation in fluorescence when using paper impregnated with fluorescein (fluorescein strips) to introduce dye onto the ocular surface, in contrast to the more consistent observations using fluorescein liquid dye, though the appearance is also dependent upon the concentration and volume instilled onto the eye. The degree of fluorescence observed on the ocular surface appears to be concentration dependent, and light of a longer wavelength is emitted with increasing fluorescein concentration.1 We have observed that increasing the concentration of fluorescein dye to 10% results in nearly 100% quenching of fluorescence for the first few minutes after instillation. Quenching is related to the fluorescein concentration on the ocular surface.2 The purpose of this study was to evaluate this fluorescein quenching phenomenon in normal eyes and its relationship to the tear film.

Eye rubbing results in ocular surface changes, such as irregular epithelial cell morphology, alterations in microprojections, and increased cellular exfoliation. It is unknown whether such changes are related entirely to the mechanical trauma of eye rubbing or manifest secondary to the effects of histamine released from degranulated mast cells associated with eye rubbing. To distinguish between these effects, the present study examines the effects of topical histamine application in the absence of applied mechanical trauma. The bulbar and tarsal conjunctiva of rabbits (n=10) exposed to 25 μl of histamine (25 mg/ml) for 10 minutes were examined using scanning electron microscopy. Contralateral untreated eyes served as controls. Regions studied included the bulbar and tarsal conjunctivae. The bulbar conjunctiva was divided into two regions, the interpalpebral bulbar conjunctiva and that portion unexposed to the environment. In contrast to control eyes, the interpalpebral bulbar conjunctiva of treated eyes had surface cells with irregular morphology and elevated apices. The goblet cell intercellular crypt openings in treated eyes had minimal mucus, while in control conjunctiva most intercellular openings contained mucus. The upper tarsal conjunctiva of treated eyes had an irregular surface and empty intercellular crypt orifices when compared to the smooth tarsal conjunctival surface and mucus-con- taining crypts of control eyes. The lower tarsal conjunctiva had minimal differences in surface cell morphology and did not appear notably altered, in contrast to the upper tarsal conjunctiva. In summary, topical histamine resulted in alteration of the exposed bulbar conjunctiva and upper tarsal conjunctiva, whereas the unexposed bulbar conjunctiva and lower tarsal conjunctiva surfaces exhibited only subtle changes. Perhaps the changes observed in the upper tarsal conjunctiva resulted from the blinking (wiping) action of the upper lid over the chemotic exposed bulbar conjunctiva. The paucity of mucus in the goblet cell orifices after histamine exposure was unexpected.

The purpose of this study was to compare and contrast quantitative crystalline lens phospholipid profiles among human, pig, rabbit, rat mouse, dog, lamb, guinea pig, beef, calf, chinook salmon, and golden roach. Lenses were extracted using chloroform-methanol. The extracts were prepared for phospholipid 31P NMR quantitative analysis using an NMR analytical reagent specifically designed for this purpose. Lens phospholipid profiles vary among vertebrate species. Thirteen different phospholipids [phosphatidylglycerol, lysophosphatidylethanolamine, phosphatidic acid, diphosphatidylglycerol, the ethanolamine plasmalogen (EPLAS), phosphatidylethanolamine (PE), phosphatidylserine (PS), sphingomyelin (SM), lysophosphatidylcholine, phosphatidylcholine (PC), including four uncharacterized (unknown) phospholipids at 1.31 ppm (human), 1.20 ppm (human, rabbit, dog, lamb, beef, calf), 0.13 ppm (all except rat), and -0.17 ppm (rat, beef) were detected. EPLAS, PE, PS, SM, and PC are the major lens phospholipids in all species except the human, where the major phospholipid is the unknown at 0.13 ppm. The lens content of this major unknown in mole percentage of the total detected phospholipid profile is: human 43.7; pig 6.7; rabbit, 6.1; rat (not detected); mouse, 3.2; dog, 5.0; guinea pig, 2.0; lamb, 7.0; beef, 7.7; calf, 5.6; chinook salmon, 6.7; and golden roach, 1.6. Large qualitative and quantitative differences were observed among lens species, indicating the necessity for prudent selection of appropriate animal models. The most striking finding is that no other species except the human species exhibits such a profound amount of the unknown phospholipid at 0.13 ppm. In the human lens, this phospholipid is the major phospholipid.

Since diseases of the neural retina and optic nerve can result in alteration of biological membranes, this study determines similarities and differences in the membrane phospholipid content of the neural retina, optic nerve head, and optic nerve to serve as baseline data. Neural retina, optic nerve head, and optic nerve were dissected, isolated as 5 sets from 20 rabbits and frozen in liquid N2. Separate pooled-tissue extracts were prepared for each set of tissues and phosphorus-31 nuclear magnetic resonance (31P NMR) analyses performed. Ten phospholipids were quantified (respective neural retina, optic nerve head, and optic nerve mole % are given for the 5 major phospholipids detected): phosphatidylcholine (PC), 44.61, 27.67, 26.40; PC plasmalogen or alkylacyl PC (CPLIP); phosphatidylinositol (PI); sphingomyelin (SM); phosphatidylserine (PS), 12.63, 14.77, 15.09; phosphatidylethanolamine (PE), 21.21, 9.59, 8.69; PE plasmalogen (EPLAS), 11.07, 30.96, 33.93; an unidentified (unknown) phospholipid (U) at the chemical-shift value of 0.13 ppm; diphosphatidylglycerol (DPG); and phosphatidic acid (PA), 0.46, 2.92, 1.57. Significant differences between the various tissues were determined by the one-way analysis of variance, using a Scheffe range value of P The neural retina in all phospholipids detected except for the uncharacterized (unknown) phospholipid was significantly different from the optic nerve head tissue. The optic nerve head was significantly different from the optic nerve in PC, CPLIP, PE, EPLAS, U, DPG, and PA. The data provide a baseline for studies on pathologically changed neural retina, optic nerve head, and optic nerve.

Alterations in the tear film lipid layer as a function of blinking were investigated using a custom-designed specular reflection monitoring system. The tear film lipid layer of 104 subjects under conditions of normal ("baseline") blinking and "forceful" blinking was quantitated on the basis of specific interference colors. Deliberate, forceful blinking was found to significantly increase the lipid layer thickness (LLT) of the tear film. The magnitude of increase was found to be correlated with the baseline LLT values; individuals with baseline LLT values of 75-150 nm demonstrated a mean increase in LLT of 33 nm following forceful blinking, whereas subjects with baseline LLT values < or = 60 nm experienced a mean increase of 19 nm. The difference in the magnitude of increase between the groups was highly significant (p = 0.0001). The data suggest that, in addition to playing a role in the spreading of lipid across the tear film, the blinking mechanism may be important in the maintenance of the lipid layer by augmenting the expression of lipids from the meibomian glands.

We examined the effect of the growth factors, human epidermal growth factor (hEGF) and insulin, on corneal metabolism during storage in Optisol, a chondroitin-sulfate-(CS)-based storage medium. Paired cat corneas, in either Optisol only or Optisol with growth factor(s), were analyzed using ex vivo 31P nuclear magnetic resonance, after storage for 1 week at 4 degrees C. Lysosomal enzyme release into the media at the end of the storage period also was measured fluorometrically. Both epithelial-intact and epithelial-denuded corneal pairs were examined for all conditions. Considering corneas having either intact epithelia or epithelium-denuded corneas, the addition of either growth factor alone to Optisol did not alter the relative corneal concentrations of five of the six phosphatic metabolite spectral bands measured or two metabolic ratios calculated from these bands. Phosphodiesters, however, were significantly lower in corneas stored in Optisol containing both hEGF and insulin (23%) than in corneas stored in Optisol alone (30%). Intracorneal pH was unaffected by the addition of growth factor(s). A significantly higher release of alpha-glucosidase and alpha-mannosidase was noted in those corneas stored in Optisol containing both hEGF and insulin. Optisol maintains high-energy phosphate corneal metabolism similar to other CS-based media, K-Sol and Chondroitin Sulfate Corneal Storage Medium (CSM). The addition of the growth factors hEGF and insulin to Optisol alters corneal metabolic activity during storage in a manner indicative of conserving corneal phospholipids.

Paired human lenses ranging from 2 to 96 yr old (n = 17) were harvested within 2 hr post-mortem and immersed in liquid nitrogen. Phosphorus containing metabolites from perchloric acid extracts were quantitated using phosphorus-31 nuclear magnetic resonance spectroscopy. Low-energy metabolites detected include the phosphomonoesters (PME) hexose-6 phosphate, alpha-glycerol phosphate, fructose 1,6-diphosphate, beta-glycerol phosphate (beta-GP), uncharacterized phosphomonoesters at 4.63 delta, 4.34 delta and 3.05 delta, phosphorylethanolamine (PE), inosine and adenosine monophosphates, and phosphorylcholine (PC). Inorganic orthophosphate (Pi), glucose 1-phosphate and glycerol 3-phosphorylcholine were the other low-energy metabolites detected. High-energy metabolites phosphocreatine (PCr), adenosine tri- and di-phosphates (ATP, ADP), nucleoside diphosphorylsugars (NS) and the dinucleotides were also detected. The following metabolic indices were calculated: PCr/Pi, PME/Pi; ATP/Pi, PCr/ATP, ATP/ADP, PE + PC, PC/PE, energy charge, phosphorylation potential, and the energy modulus. A significant linear increase with age (P less than 0.05) occurred only in the uncharacterized resonance at 3.05 delta (r = 0.4593). Significant linear decreases with age (P less than 0.05) occurred in the uncharacterized resonance at 4.63 delta (r = -0.5950), beta-GP (r = -0.5018), PCr (r = -0.4495), N.S. (r = -0.4882). There was a significant (P less than 0.05) linear decrease in the energy modulus (ratio of high-energy to low-energy metabolites).

Crystalline lenses excised from 5–7-month-old rabbits and pigs were extracted for lipids with chloroform/methanol, 2 1 using the Folch method. The extracted crude lipids were analyzed at 202 MHz by 31P-NMR spectroscopy. Twelve membrane phospholipids were detected. Both rabbit and pig phospholipid profiles contained phosphatidylcholine (PC), lysophosphatidylcholine, phosphatidylcholine plasmalogen, phosphatidylethanolamine (PE), phosphatidylethanolamine plasmalogen (PE plas), phosphatidylserine, sphingomyelin (SPH), and an uncharacterized phospholipid. In addition, pig lens profiles contained lysophosphatidylethanolamine, phosphatidylinositol, cardiolipin and phosphatidylglycerol. The data indicate that these two animal models have significant differences in membrane phospholipid profiles. In each species, however, the bulk phospholipid component resides in the neutral phospholipids PC, PE, PE plas, and SPH.

Since the ATP molecule is associated with the intracellular milieu, which consists mainly of water, this study suggests an intimate intracellular association with water molecules. Intact canine crystalline lenses (n = 6) were incubated at 37 degrees C in a buffer containing a pure D2O solvent and examined by [31P]-nuclear magnetic resonance (NMR) spectroscopy. Control lenses (n = 6) from contralateral eyes were incubated in a pure H2O-based buffer. During the first 4 h of a 24-hour incubation, the D2O-incubated lens maintained its phosphorus metabolic profile, indicating that the tissue maintains its metabolic health. During this period, however, the spectral signal widths of ATP narrowed progressively. Spectral-simulation analysis of ATP alpha-, beta- and gamma-phosphate signals revealed that the gamma-phosphate signal was narrowed significantly in contrast to the alpha-phosphate signal. Such differential signal narrowing within a single NMR spectrum indicates that the principal intracellular site of water hydrogen binding to the ATP molecule is at the gamma-group phosphate.

THE OCULAR SURFACE / JANUARY, 2005, VOL. 3, NO. 1 / SUPPLEMENT S67 THE BIOCHEMISTRY OF FILM-FORMING MOLECULES PRODUCING A LIPID MEMBRANE THAT FACILITATES THE RESTORATION OF THE PREOCULAR TEAR FILM. Thomas Glonek, Jack V. Greiner, Donald R. Korb. Midwestern University, Chicago, IL; Department of Ophthalmology, Harvard Medical School, Schepens Eye Research Institute, Boston, MA; 2 Korb & Associates, Boston, MA. Purpose. The majority of the millions of dry eye patients in the USA suffer from lipid layer deficiency in the preocular tear film. The purpose is to describe the detailed biochemistry and evolution of a lipid-membraneforming solution that restores the preocular tear film. Conceptual Development. Foundation of concept based on the following: (1) the understanding that, of all the dry eye syndromes, those with lipid layer deficiency from meibomian gland dysfunction comprise the greatest percentage of dry eye sufferers, (2) the question of how to produce an artificial tear that restores the lipid layer began with lecithin, (3) the understanding that lecithin from natural sources is comprised of a class of phospholipids and not just glycerol 3-phosphorylcholine, and (4) the discovery that it is interfacial molecules that hold the key to building a thick neutral superficial oily layer. Results. (1) The biochemical construction by interfacial molecules produces an interfacial layer that has the chemical characteristics of a strong membrane. (2) The production of this biochemical membrane supports an overlying neutral lipid layer, whose thickness is dependent upon the strength of the interfacial layer. (3) Using a tear film apparatus and interference patterns, we demonstrate the ability to produce a thickened lipid layer with a lipid-membrane-forming solution on individuals with meibomian gland dysfunction and resultant lipid layer deficiency. This membrane film can be maintained as part of the preocular tear film for up to 24 hours. Conclusions. An eye drop using interfacial molecule technology, allows for: (1) the production of a strong tear-film lipid-layer (that does not produce visual blur beyond seconds), and (2) products including: tear supplements, contact lens wetting and storage solutions, and drug delivery systems with a vastly enhanced pharmaceutical residence time and potency.

Tear film stability has been a common denominator among conditions causing irritation of the ocular surface since Holly’s report in 1973.1 Pflugfelder et al.2 noted “An unstable tear film is the hallmark of various dry eye states.” The NEI/Industry Workshop on Clinical Trials in Dry Eyes reported breakup time (BUT) to be the only clinical test available to measure tear film stability.3 The fluorescein BUT (FBUT) test was introduced as corneal wetting time by Norn4 in 1969 and subsequently refined into the BUT test by Lemp5 in 1973. Although a reduced BUT is considered a global criterion of dry eye,3 it is criticized as not accurate or reproducible, and is not universally accepted.6–9 Inherently, the FBUT test is invasive, thus compromising its reproducibility.7,9

Phosphatic metabolites from human corneas, pooled into 7 decades ranging from ages < 1 year through 79 years, were quantitated using phosphorus-31 magnetic resonance (31P MR) spectroscopy. Relative concentrations of phosphorus-containing compounds measured included the low-energy metabolites [phosphomonoesters (PME), inorganic orthophosphate (Pi), phosphodiesters (glycerol 3-phosphorylethanolamine and glycerol 3-phosphorylcholine)] and the high-energy metabolites [phosphocreatine (PCr), adenosine triphosphate (ATP), adenosine diphosphate (ADP), nucleosidediphosphosugars and the dinucleotides]. Significant linear changes attributable to age occur in the relative mole percentage decrease of phosphate concentrations of human corneal PME, PCr and ATP, and in the increase of Pi. Age-attributable rates of decrease in PME at -0.162 MPP/YR (mole percent phosphorus per year), PCr at -0.015 MPP/YR and ATP at -0.487 MPP/YR combined, approximate the rate of increase in Pi determined to be +0.729 MPP/YR. Of the indices computed from the human corneal spectral data, the ratios of ATP/Pi and PME/Pi and the tissue energy modulus were all found to decrease significantly with age. These changes in corneal phosphatic metabolites are indicative of an overall decline in high-energy metabolism with age.

Using phosphorus-31 nuclear magnetic resonance (P-31 NMR) spectroscopy the rat crystalline lens was evaluated as a potential animal model for the study of lens metabolic processes relevant to the human. The P-31 NMR spectra of intact rat lenses and rat lens PCA extracts provided comparable qualitative and quantitative results concerning the tissue levels of the phosphorus-containing metabolites of intermediate metabolism. The study also establishes the feasibility of examining intact rat lens metabolism by P-31 NMR. Comparisons with reported human and rabbit lens metabolite levels revealed significant differences; consequently, metabolic information derived from rat lens studies should be extrapolated to the human with caution.

The concentrations of the major and minor organophosphate metabolites of the rabbit and human crystalline lens were quantitated from perchloric acid extracts analyzed by phosphorus-31 nuclear magnetic resonance spectroscopy. The concentrations of adenosine diphosphate, phosphocreatine, inorganic orthophosphate and the nucleoside diphosphosugars were greater in the human lens samples when compared with the rabbit lens profile. A 10-fold greater phosphocretine content of the human lens was the most remarkable species difference. The adenosine triphosphate and the glycerol derivative, glycerol 3-phosphorylcholine, were enriched in the rabbit lens relative to the human. A prominent resonance signal at 6 p.p.m. corresonding to an as yet, unidentified phosphorus-containing molecule and two other less prominent phosphorus resonances of unidentified origin at 17·9 and 10·7 p.p.m., detected in the rabbit lens were absent from the human lens profile. In addition, the α-glycerophosphate content of the human lens was considerably less than that of the rabbit lens. These profile data document similarities and differences in the human versus the rabbit lens metabolites. The magnitude and diversity of these differences in lens metabolite concentrations emphasize the need for cautious rabbit to human extrapolation of experimental findings.

This study describes the phosphatic metabolite contribution of the porcine corneal stroma [by phosphorus-31 nuclear magnetic resonance (P-31 NMR)] and the corneal epithelium (by difference calculation) to the total phosphatic metabolic profile of the intact (i.e. whole, undissected) cornea. In order to provide such a regional analysis, the epithelium and endothelium were surgically removed from three sets of 12 corneas. Three additional sets of 12 intact corneas were used as controls. Histological examination of two randomly selected stromata from each set of corneas were analyzed to verify that epithelium and endothelium were removed. Perchloric acid extracts were prepared from the 10 remaining stromata for P-31 NMR analysis. Twenty phosphatic metabolites were measured including six unidentified (unknown) signals. Components with diminished relative concentrations in the stromal profile compared to that of the intact cornea profile are: the triose and hexose phosphates associated with glycolysis (including alpha-glycerophosphate), choline phosphate, ATP, uridine diphosphohexoses and the unknown compound resonating at 0.93 delta. The relative level of inorganic phosphate in the stroma is nearly doubled. The remaining measured phosphates are not significantly changed relative to the total extractable phosphorus content. Consistent with previous biochemical findings, phosphates associated with energy metabolism are found primarily in the epithelial fraction. The metabolic status of the endothelial monolayer cannot be assessed because of low tissue mass and the relatively low sensitivity of P-31 NMR spectroscopy. This study serves as baseline data for the interpretation of experimental and clinical phosphorus NMR data from intact cornea by providing the contribution to the whole corneal spectrum of metabolites of the individual corneal layers.

The rate of change in concentration of corneal phosphatic metabolites of cat corneas stored in moist chamber and McCarey-Kaufman (M-K) medium was determined in order to provide a basis for prediction of the corneal metabolic status at a given storage time. Perchloric acid corneal extracts were examined by phosphorus-31 magnetic resonance after storage at 4 degrees C of whole globes under moist-chamber conditions up to 48 h and of excised corneas in M-K medium up to 168 h. A significant decline in the corneal concentrations of ATP and a significant increase in inorganic phosphate occurred for both storage methods; however, depending on the metabolite, the rate of decline or increase was significantly greater for the moist-chamber-stored corneas. The phosphorylated sugars significantly increased and the glycerophosphodiesters significantly decreased in the moist-chamber-stored corneas, whereas both metabolites remained unchanged in the M-K-medium-stored corneas. There was no significant change in the dinucleotides and nucleoside diphospho-sugars during the time course for both storage methods. A threefold greater rate of decline was noted in the tissue energy modulus for the moist-chamber-stored corneas than for the M-K-medium-stored corneas (-0.0465 vs. -0.0121 modulus values/h). M-K medium was significantly more effective in the maintenance of high-energy phosphatic metabolites. The mathematical model for these rate determinations provides a basis for prediction of the corneal metabolic status at a given time in moist-chamber or M-K medium storage.

Sixteen corneal buttons were obtained from 16 patients with keratoconus at the time of penetrating keratoplasty (mean age, 34 +/- 3.2 yr) and prepared for phosphorus-31 nuclear magnetic resonance spectroscopy following perchloric acid extraction. The corneas were compared with age-matched corneas from fresh eye-bank eyes with normal slit-lamp biomicroscopic appearance. Higher levels (P less than 0.05) of nucleoside monophosphates and choline phosphate and lower levels (P less than 0.05) of adenosine diphosphate were detected in keratoconus than in eye-bank corneas. The level of an unidentified metabolite at 3.96 delta, present in eye-bank corneas, was ninefold higher in keratoconus corneas. Another unidentified metabolite at 3.31 delta, accounting for 1.5% of the total detected phosphatic metabolites, appeared in keratoconus but not in eye-bank cornea spectra. These findings of altered phosphatic metabolites are consistent with altered corneal metabolism. The unidentified, unknown metabolites in the spectral profile uniquely distinguish keratoconus from eye-bank corneas.

Successful corneal transplantation was accomplished following metabolic phosphorus magnetic resonance analysis. Four cat corneas were analyzed using phosphorus-31 magnetic resonance following storage in modified McCarey-Kaufman (M-K) medium for 24 h. Corneas were re-stored in M-K medium and transplanted 24 h after MR analysis. Four control corneas (contralateral eye, no magnetic resonance analysis performed) were also transplanted following storage in M-K medium under identical conditions. Successful corneal transplantation was accomplished with minimal ATP tissue levels. Corneas stored for 48 h maintained a pH of 7.3. The phosphorus-31 spectral modulus, which is the ratio of the high-energy phosphates to the low-energy phosphates, was calculated using the spectral integral (range, 0.49-0.77). No difference in endothelial cell density or morphology was detected between corneas following magnetic resonance analysis and control corneas when evaluated by specular microscopy.

Spectra of intact bovine and canine crystalline lenses in vitro at 37°C were generated using phosphorus-31 nuclear magnetic resonance (NMR). Lens incubation in a pure D2O buffer results in a progressive narrowing of the tissue P-31 NMR signals.

• Application of nondestructive and noninvasive nuclear magnetic resonance (NMR) techniques to the evaluation of corneal donor tissue metabolism may provide information that would allow improved selection of donor material for keratoplasty. A phosphorus 31 NMR spectrum was generated from a single intact human cornea, evaluating phosphatic metabolites qualitatively and quantitatively. The NMR analyses were conducted on four corneas (from patients aged from 73 to 75 years). Intracorneal pH (7.2) was monitored from the resonance shift position of inorganic orthophosphate (Pi). A numerical index (1.15) of tissue energy status was determined by comparing the relative tissue content of high-field, high-energy phosphate signals (adenosine triphosphate, adenosine diphosphate, nicotinamide-adenine dinucleotide, and the uridine diphospho-hexoses) to that of the low-field, low-energy phosphate signals (Pi, sugar phosphates, and phosphodiesters). This is the first demonstration of the feasibility of using nondestructive 31P NMR to monitor the metabolic status of a single intact eye bank cornea as would be required in the evaluation of eye bank tissue before keratoplastic procedures.

Changes in intracorneal and storage-medium pH values of organ-culture-stored cat corneas were monitored over a 4-week period. The intracorneal pH was determined using the phosphorus-31 magnetic resonance spectroscopy (31P MRS) chemical shift of inorganic orthophosphate in conjunction with a standard pH titration curve. We incubated 32 adult cat corneas using two similar standard organ-culture methods, one with chondroitin sulfate (method 1) and the other without (method 2). Time-course data at 0, 1, 3 and 4 weeks of storage were used to calculate the rate of pH change. The intracorneal pH was not changed significantly for either organ-culture method; however, the storage-medium pH rate of change declined significantly for both methods (method 1, 0.15 pH units/week; method 2, 0.12 pH units/week). The difference between intracorneal and storage-medium pH values over time increased at a rate of 0.12 and 0.11 pH units/week for method 1 and method 2, respectively. The declining storage-medium pH in conjunction with the maintenance of intracorneal pH contributes to an increased metabolic demand on the cornea.

Main and accessory lacrimal tissues from autopsy and biopsy specimens were compared histologically and immunohistologically. Formaldehyde-fixed, paraffin-embedded specimens were studied by light microscopy with hematoxylinand-eosin and PAS staining. Glutaraldehyde-fixed, Epon-embedded specimens were sectioned at 1 micron, stained with alkaline Giemsa, and studied by light microscopy. Specimens fixed in a solution of alcohol and acetic acid were stained by immunofluorescence techniques for lactoferrin, lysozyme, secretory component, and the immunoglobulins IgG, IgA, IgM, IgD, and IgE. The main and the accessory lacrimal tissues were identical histologically and had identical distributions of secretory products and immunoglobulin-containing plasma cells. The finding of myoepithelial cells in 1-micron sections of accessory lacrimal tissue indicates autonomic innervation in that tissue. This finding, in conjunction with the identical immunohistology, indicates a common source for unstimulated and stimulated tears.

Phosphatic metabolites of the intact rabbit lens were quantitated as a function of time by phosphorus-31 nuclear magnetic resonance (P-31 NMR) spectroscopy during in vitro incubations at 37°C in calcium-sufficient and calcium-deficient modified Earle's buffer with and without the osmotic agent, Dextran-40. Intralenticular pH was determined from the resonance shift position of inorganic orthophosphate (Pi). Incubation of lenses in calcium-deficient buffer resulted in a pronounced, time-dependent decrease in lenticular adenosine triphosphate (ATP) levels. The half-life of ATP within the lens was 11 hr under these experimental conditions. A concomitant, essentially stoichiometric increase in adenosine diphosphate and Pi levels was observed also. The other phosphatic metabolites were unaffected by exogenous calcium deprivation except for adenosine and inosine monophosphate which accumulated with time. Dextran-40 (6%), which has been shown to prevent lens swelling under these same experimental conditions, did not influence the metabolic responses of the lens to external calcium deprivation and did not facilitate subsequent restoration of lens phosphatic metabolites following restoration of a physiologic calcium concentration to the supporting medium. The Dextran-40 did, however, promote the retention of intralenticular pH environment during the experimental period. These findings suggest that the previously reported Dextran-40-dependent recovery of intralenticular sodium and potassium concentrations to control levels following 10 hr of incubation in calcium-deficient media cannot be attributed to a direct energy-sparing action of Dextran-40 on lenticular energy metabolism. Instead, the mechanistic basis for the action of Dextran-40 would appear to be related to its colloid osmotic properties and its ability to prevent lenticular swelling, which otherwise occurs in the absence of Dextran under these experimental conditions.

• Possible local sources of human tear lysozyme were investigated using an indirect immunofluorescence technique. Lysozyme was identified in 20% to 50% of acinar and ductular epithelial cells of both main and accessory lacrimal glands. The staining was granular in character and confined to the apices of the cells. Cells that stained positive tended to be grouped. Interstitial tissues of main and accessory lacrimal tissues did not stain. Conjunctiva and all other ocular tissues examined were unstained by antilysozyme antisera. Our findings are compatible with lysozyme either being produced in lacrimal tissue or being concentrated from plasma. The absence of any other lysozyme-specific fluorescence in the interstitial elements of the lacrimal tissues supports the notion of local synthesis by acinar lacrimal tissue.