There are approximately 130 million contact lens (CL) wearers worldwide and 38.5 million in the US . Contact lenses are regulated as medical devices that need to be biocompatible with the human tissues with which they come in contact. According to the United States Food and Drug Administration (FDA), a device is considered biocompatible if its component materials do not either directly or through the release of their material constituents: (i) produce adverse local or systemic effects; (ii) act as carcinogens; or (iii) produce adverse reproductive and developmental effects . Therefore, the FDA requires data from systemic testing of the device to confim that the benefits provided by the final product exceed any potential risks produced by device materials before clearing the device for marketing. Even after becoming commercially available, devices may exhibit biocompatibility issues. For example, CL can develop protein and lipid deposition on their surfaces and/or matrix, which can be a concern for biocompatibility with the ocular surface, as deposits have been associated with adverse reactions including symptomatology, inflammation and infection [3-9].
Silicone hydrogels (SiHys) are the most widely prescribed soft contact lens materials in most parts of the world . Silicone hydrogels, although highly permeable to oxygen, still exhibit deposition and there is debate as to whether SiHy materials deposit more lipids than hydrogel materials [11, 12][13-16]. Some researchers have reported that SiHy materials attract more lipids from the tears, which is proposed to be due to their inherent hydrophobicity, and can lead to subsequent dewetting of the CL surface and tear film instability [14, 16, 17]. Begley et al in 2001 reported frequencies of 77% and 73% of CL wearers with discomfort and dryness, respectively. Nichols et al in 2005 reported that 52% of CL wearers experienced increased dryness and discomfort when compared to spectacle wearers (23.9%) and emmetropes (7.1%) . Dryness and discomfort have been reported to be a major cause for permanent discontinuation of CL wear. Giannoni and Nichols reported that 40% of permanent discontinuations were due to dryness and discomfort. Similarly, Pritchard et al also reported 40% of CLs wearer discontinuations to be due to discomfort. Therefore, with such a large proportion of CL wearing population affected with the problem of discomfort, there is a great need to understand the etiology of CL discomfort, including the role of deposition.
Lipid deposition on CLs has also been associated with CL discomfort . The lipids that deposit on CLs come from the meibum, which makes its way into the tear film from the meibomian gland orifices. Meibum is produced in the meibomian glands located in the upper and lower eyelids. The major lipids present in the meibum include wax (25 – 68%) and cholesterol esters (0 – 65%), although there are many other types of lipids present [23-25]. These and other ocular surface lipids can oxidize and degrade to products like malondialdehyde, which may be toxic to ocular surface tissues . Together, this may lead to dryness and discomfort [27, 28] as such products have been found to be increased in intolerant CL wearers .
There are various methods of studying lipid deposits on the surfaces and in the matrix of CLs . They can mainly be categorized into observational techniques and assays. Observational techniques include imaging through microscopy (light, confocal, scanning electron, transmission electron, atomic force etc.) and techniques like chromatography. Assays include specific CL lipid analyses like a total lipid assay, and cholesterol and phospholipid assays . Any one of the above-mentioned techniques or assays cannot provide a full description of lipid deposition alone. Observational methods generally show the morphology of deposits while assays provide more quantitative insights, although many assays require extraction from the material using organic solvents, which may lead to contamination of the sample with polymer from the CL material. Imaging techniques capable of quantifying lipids without extraction (and potential contamination) avoid this issue, and concurrently provide the morphology and distribution pattern of the deposition.
Previously, imaging studies on CL materials have generally used Oil Red O and Nile Red to stain lipids [32, 33]. LipidTOX is a neutral lipid stain that can be detected by fluorescence microscopy or a high-content screening reader. LipidTOX has been used previously to stain lipids in rat cortical neurons , human hepatoma cells, adipocytes macrophages  and in immortalized human meibomian gland epithelial cells . To our knowledge, this dye has not been used to stain lipid deposits on CL materials. Thus, the purpose of this research was to evaluate whether LipidTOX could stain lipids on CL materials using contemporary contact lens materials and compare lipid deposition patterns qualitatively and quantitatively using this dye and an imaging method.
2.1 In vitro Arm
2.1.1 CL Materials
Six commercially available unworn SiHy materials and one hydrogel material (10 total for each type of material) were included in the study (Table 1). All lenses had an optical power of -2.00.
Insert table 1 here.
A cholesteryl oleate stock solution was prepared at a concentration of 8.60 mM (5.60 mg/ml) as directed by the manufacturer . In brief, 56.00 mg of cholesteryl oleate (C9253, Sigma Aldrich, St. Louis, MO) was dissolved in 1.00 mL of nonaethylene glycol monododecyl ether (P-9641, Sigma-Aldrich) followed by the addition of 9.00 mL of hot 0.10 M, pH 7.0 phosphate buffer. The solution was then stored at room temperature. For a positive control, one microliter of olive oil was added to 99.00 µL of stock phosphate buffered saline (PBS) solution (0.10 M, pH 7.0) to prepare 1% olive oil in PBS. Eighty-five µL of propylene glycol (100%) (P4347, Sigma-Aldrich) was added to 15 µL distilled water giving 85% solution of propylene glycol in distilled water.
To make sure that cholesteryl oleate dissolved completely in the solvent and did not suspend as small droplets, the cholesteryl oleate solution was mixed with both the dyes separately and examined under the microscope at 10x magnification. Cholesteryl oleate was observed to have dissolved completely and no suspended droplets were visible. Ten unworn CLs of each of the seven materials were then placed in 10 wells of closed CL cases with 1.00 mL of 5.60 mg/ml cholesteryl oleate solution. Similarly, two unworn CL of each material were doped in 1.00 mL of 1% olive oil in PBS as positive controls and another two unworn CL of each material were doped in 1.00 mL of PBS as negative controls. The CL cases were then placed on a rocker table with gentle agitation for 12 hours at room temperature. Contact lenses were then removed from the incubation solution with clean metal forceps and were lightly blotted with a Kim Wipe. The CLs were then stained with Oil Red O and LipidTOX as described below.
2.1.3 Oil Red O and LipidTOX Staining
Oil Red O was prepared just prior to staining the CLs. In brief, 0.5 gram (g) of powdered Oil Red O (P O0625, Sigma Aldrich) was added to 100 mL propylene glycol (100%) and heated at 100°C for 10 minutes. It was then allowed to stand at room temperature and was filtered with a 0.20 µm syringe filter. Contact lenses were rinsed with PBS, stained with Oil Red O for two minutes in a contact lens case, placed on a rocker and were then differentiated with 85% propylene glycol in distilled water for one minute. Lastly, the CLs were rinsed with PBS for 20 minutes and lightly blotted with a Kim Wipe.
LipidTOX was also prepared just prior to staining the CLs. In brief, 10 µL LipidTOX (H34475, Thermo Fisher Scientific) was added to 990 µL of PBS to prepare a 1% solution in PBS per ml in room temperature. Contact lenses were first rinsed with PBS and then incubated in 1 mL of LipidTOX solution for 30 minutes in a CL case for optimal staining of the lipids. They were then rinsed with PBS for 20 minutes and lightly blotted with Kim Wipes.
In order to make sure that both the dyes were staining the same lipids on the contact lens surfaces, an unworn senofilcon A (Acuvue Oasys) CL was stained with both the dyes simultaneously and then imaged through a tetramethylrhodamine (TRITC) filter for Oil Red O, through a fluorescein isothiocynate (FITC) filter for LipidTOX, and finally through a combined filter (Figure 1) for comparison of fluorescence at 100x. It was clearly seen that both dyes stained the same lipid deposit on the contact lens surface.
Insert figure 1 here. Please use color images.
2.1.4 Fluorescence Microscopy
All CLs were cut radially on one side taking care not to cut the central 2 mm diameter region to allow for flattening of the CLs over glass slides. The CLs were then covered with a cover slip and a central circular zone of 2 mm diameter was marked with a permanent board marker. A fluorescence microscope (either a Deltavision Spectris Core microscope (GE Healthcare, Pittsburgh, PA) with SoftWoRx (1.0) software or a Zeiss Axioplan 2 imaging microscope (Carl Zeiss Microscopy, LLC, Thornwood, NY) with AxioVision 4.0 software was used to visualize the deposits. A tetramethylrhodamine (TRITC) filter was used for Oil Red O imaging and a fluorescein isothiocynate (FITC) filter was used for LipidTOX, both at 100x objective magnification. An area of 4402.32 µm2 that could be visualized within one frame of the microscope was imaged of the central 2 mm diameter zone of the CL.
2.1.5 Image Analysis & Quantification
Using the aforementioned images, qualitative descriptions of deposition shape, size and fluorescence are described in the results section along with comparison among the various materials using each stain. For quantification, all images were converted into grayscale (8 bit) and local thresholding was performed using the Otsu method in Fiji software [40, 41]. Thresholding is done to control for background staining (staining of CL material itself and not the lipid deposits). Only the pixels above the intensity of the background are considered as being the deposits while those below are discarded from the analysis. The “Analyze Particles” command was then used to calculate the surface area (in µm2) of the thresholded image (Figure 2).
Insert Fig 2a here (in color)
Insert Fig 2c here
Insert Fig 2b here
Color image 8-bit image Thresholded image
2.2 Ex Vivo Arm
This work was approved by an Institutional Review Board in accordance with the Declaration of Helsinki. Two worn CLs, one each for staining with the two dyes, were collected from ten wearers of senofilcon A CLs (Acuvue Oasys; Vistakon Inc., Jacksonville, FL), to compare the deposition patterns and examine the method ex-vivo. Eligible subjects included those that were at least 18 years of age, who were current senofilcon A CL wearers. Subjects were required to wear their CLs at least eight hours per day (on average) and to have worn the same brand of CLs for at least one month. Subjects were excluded from the study if they had any ocular surface disorders, were using any ocular medications, or were currently pregnant or breast-feeding. Eligible subjects were asked to wear their contact lenses for 14 days, and on the 14th day, their contact lenses were removed from the eye with sterile metal forceps. The worn CLs were placed in glass vials and stored at -80°C until analysis. The CL removed from the right eye was stained with Oil Red O and the CL removed from the left eye was stained with LipidTOX as described above.
In order to ensure that both the dyes were staining same lipids on the worn contact lens surfaces, a worn senofilcon A (Acuvue Oasys) CL was stained with both the dyes and imaged similarly as described above under TRITC and FITC filters (Figure 3) through a fluorescence microscope at 100x. The figures clearly showed that both the dyes stained the same lipid deposit on the worn contact lens surface.
Insert figure 3 here. Please use color images.
2.3 Statistical Analysis
The surface area of staining was described using medians and ranges for all the CL materials. Comparison of surface areas of deposition between various materials using an individual dye was done with the Kruskal-Wallis test for the in vitro portion of the study. The Mann-Whitney U test was done to compare staining areas for the two dyes for the same CL material in both the in vitro and ex vivo arms of the study.
3.1 In vitro arm
Staining of cholesterol oleate by both dyes was confirmed with positive and negative controls, where positive controls showed similar patterns of deposition as the cholesterol doped CLs, while negative controls showed no deposition (Figures 4 & 5).
Insert figure 4 here. Please use color images.
Insert figure 5 here. Please use color images.
3.1.1 Between Materials Comparison with Oil Red O
Balafilcon A appeared to have the largest sizes of individual lipid deposits (Figure 6a). The deposits ranged from round to elliptical in shape and were highly fluorescent, while background staining was the least with this material. Delefilcon A had highly fluorescent circular deposits with a darker areas surrounding them (Figure 6b), with the greatest background staining. Enfilcon A showed more flattened deposits that were circular to oval in shape (Figure 6c). Lotrafilcon A showed more elliptical deposits aggregated towards one area of the imaged section (Figure 6d). Narafilcon A had diffuse circular and oval deposits that covered most of the parts of the imaged section (Figure 6e). Senofilcon A showed irregular and oval shaped deposits with some stained more strongly than others (Figure 6f). Etafilcon A showed elongated oval shaped deposits (Figure 6g).
Insert figure 6 here. Please use color images.
Although deposition was observed on all material types, not all CLs sampled within a certain material type showed deposition using Oil Red O. Thus, the quantitative values and the deposit size seen on the representative image may not correspond to each other. The quantitative values represent data from 10 CLs within each material type. Table 2 shows that enfilcon A showed the largest median area and range of staining with Oil Red O (5.08 µm) among the materials compared while etafilcon A showed the smallest median area (0.04 µm). However, there was no statistical difference between the median ranked areas of lipid deposition across all materials (Kruskal-Wallis test: H (6) = 2.66, p = 0.75).
3.1.2 Between Materials Comparison with LipidTOX
LipidTOX staining showed mostly large, round to irregular-shaped deposits on most of the CL materials. One or a few large irregular deposits were observed on four balafilcon A CLs while rest of the balafilcon A CLs showed no deposition. The representative image shows the largest deposit observed on this CL material (Figure 7a). One or more irregular deposits were also observed on lotrafilcon A (Figure 7b) and narafilcon A (Figure 7c) CLs. Multiple smaller sized deposits were observed on narafilcon A CLs. Brightly fluorescent irregular deposits were seen on most of the delefilcon A CLs (Figure 7d). A large deposit was seen on one of the ten enfilcon A CLs while the remaining nine CLs showed smaller sized deposits. The representative image shows the largest deposit (Figure 7e). A clumping of multiple irregular deposits was observed on senofilcon A CLs (Figure 7f) while round to oval shaped deposits were observed on etafilcon A CLs (Figure 7g). The deposits observed on lotrafilcon A, narafilcon A and etafilcon A materials were generally smaller than on balafilcon A, enfilcon A, delefilcon A and senofilcon A CLs.
Insert figure 7 here. Please use color images.
With LipidTOX, narafilcon A showed the largest median area (268.76 µm), while balafilcon A and enfilcon A showed the smallest median area of deposition (Table 3). However, there was no statistical difference across the CL materials in median ranked deposition areas when using LipidTOX (Kruskal-Wallis test: H (6) =6.28, p=0.28.)
3.1.3 Comparison of LipidTOX and Oil Red O
Lipid deposits stained with LipidTOX were found to be larger when compared to those stained with Oil Red O with balafilcon A (Figures 6 & 7). Although there was no statistical difference in the median ranked areas of lipid deposits with the two dyes in balafilcon A, multiple small oval deposits were observed with Oil Red O while few but bigger irregular deposits were observed with LipidTOX with this material (Figures 6 & 7). Delefilcon A showed background staining quite low to non-existent with LipidTOX stain compared to Oil Red O, and LipidTOX also gave a higher signal to noise ratio than Oil Red O (Figures 6 & 7). Enfilcon A had similar type of deposition pattern with both dyes although patterns that are more circular were seen with Oil Red O (Figures 6 & 7). In lotrafilcon A and narafilcon A, large aggregated deposits were observed with LipidTOX while small separate droplets were seen with Oil Red O (Figures 6 & 7). Senofilcon A had more irregular and polygonal shaped deposits seen with both Oil Red O and LipidTOX but there appeared to be more aggregation of deposits at one area with LipidTOX staining when compared to Oil Red O (Figures 6 & 7).
Although lipid deposition areas stained by the two dyes were different among the compared CLs, the differences were not statistically significant (balafilcon A, p = 0.97; delefilcon A, p = 0.74, enfilcon A, p = 0.08, lotrafilcon A, p = 0.18; narafilcon A, p = 0.19; senofilcon A, p = 0.06; etafilcon A, p = 0.09).
Insert table 2 here.
Insert table 3 here.
3.2 Ex vivo arm
Staining of deposits from worn CLs was confirmed with positive and negative controls with both dyes, where positive controls showed similar patterns of staining as the doped CLs, while negative controls showed no staining (Figure 8 & 9).
Insert figure 8 here. Please use color images.
Insert figure 9 here. Please use color images.
Images of lipid deposits with Oil Red O in worn senofilcon A CLs showed large irregular, aggregated lipid deposits along with other smaller deposits scattered around within the imaged central area (Figure 8). Images of lipid deposits with LipidTOX in worn senofilcon A CLs also showed larger aggregated deposit along with more diffuse deposits spread within the imaged area (Figure 9).
The median ranked area of lipid deposits in the ex-vivo lenses was not significantly different using either of the dyes (73.13 (Oil Red O) & 168.47 (LipidTOX), p = 0.06).
To our knowledge, this is the first report on the use of LipidTOX in staining lipid deposits on CL materials and several overall conclusions were established. Bright staining of cholesteryl oleate deposits was observed on both unworn and worn CL materials using LipidTOX, which was similar to that of Oil Red O staining. Using positive and negative controls, the stained material was confirmed to be cholesteryl oleate and not debris on the CL surfaces. The median area of deposition on CL surfaces within the central 2 mm diameter zone was not significantly different among the various materials using either dyes. Lipid deposition area on the CL surface was quantified by the simple yet effective imaging procedure described above using fluorescence microscopy. Lastly, LipidTOX was found to be more convenient to use than Oil Red O during the course of the study.
Deposition patterns observed using both Oil Red O and LipidTOX in in vitro and ex vivo experiments on the same senofilcon A CL material were quite similar (Figures 6 & 7, and Figures 8 & 9). Peng et al used 25-NBD cholesterol to image fouled SiHy CLs and found similar pattern of deposition observed in the present study using senofilcon A and LipidTOX dye. Although Peng et al used a different method of fouling CLs with cholesterol, the deposit images obtained using a fluorescence microscope were similar in morphology to that seen in this study.
The lipid deposition morphology observed with Oil Red O in the in vitro portion of this study was similar to the in vitro portion of a study by Pucker and Nichols. Lotrafilcon A showed robust lipid deposits when incubated under 1 mL of 5.6 mg/mL of cholesteryl oleate solution using Oil Red O in both the studies. Similar staining was also observed with the use of LipidTOX in the present study. Oil Red O was also used by Mirejovsky et al  to stain lipid deposits on CL surfaces. They observed more granular type of deposits unlike the present study that found bigger round or oval shaped deposits on the CL surfaces. This difference may be due to the use of a mixture of different types of lipids along with proteins and electrolytes in the artificial tear solution used by Mirejovsky and colleagues. Similarly, the use of phase contrast microscopy (unlike fluorescence microscopy used in the present study) might have also given different morphology of lipid deposits. The in vitro findings of this study showed that Oil Red O stains cholesteryl oleate deposits on both SiHy and hydrogel CLs in a similar manner. This differed from Mirejovsky et al study that showed Oil Red O to be a poor stain for hydrogel CLs. The difference may be due to the use of various lipids, proteins, and electrolytes in the artificial tear solution by Mirejovsky and colleagues and a different hydrogel material than that used in this study.
To our knowledge, quantification of lipid deposits on CL surfaces using a similar imaging method as in the present study has not previously been performed. Ho and Hlady evaluated CL deposits using a fluorescence assay. They used Nile Red dye and compared the fluorescence intensity of lipid deposits on CL surfaces, whereas we assessed the area of the deposits on CL surfaces. Therefore, direct comparison to the present imaging method may not be relevant.
Oil Red O is a fat-soluble diazo dye that stains neutral lipids and cholesteryl esters, which has been used for the detection of accumulated lipid droplets in various animal as well as human cells and tissues [44-46]. It needs to be prepared fresh before use and differentiation is necessary with other solvents. LipidTOX is a lipophilic dye with high specificity to neutral lipids and has been previously used to image intracellular lipid accumulation in various animal as well as human cells and tissues and in examination of lipid production by human meibocytes [34-38, 47-49]. It has a ready-to-use formulation and requires no wash or filtering steps, and does not require differentiation to remove excess amounts after staining. These conveniences of using LipidTOX for staining lipids and the similar staining patterns of lipids in in vitro and ex vivo experiments in this study suggests that LipidTOX can be a more efficient dye in similar applications as those presented here. LipidTOX also showed low background staining upon gross examination of the images when compared to Oil Red O.
There were a few limitations in this study. Deposition of lipids during CL wear occurs when the CL is in the presence of the tear film composed of not just lipids but also proteins, mucins and other electrolytes. The present study used only one of the more abundant lipids found in the tear film, cholesteryl oleate, for simplicity and not a mixture of other compounds found in tears. Further, lipids may also penetrate the surface of the CLs and enter the matrix of the material. Thus, quantification of lipids absorbed by the CL matrix would be a further step in lipid deposition analysis in CLs.
LipidTOX can be used to stain lipid deposits on CL surfaces both in vitro and ex vivo and maybe a dye of choice due to its convenience. The imaging method is suitable for quantification of lipid deposits on CL surfaces and may be used to quantify other types of deposits too on CL surfaces. Lipid deposition on the central 2 mm region of various commercial contact lens materials was not significantly different.
1. Nichols JJ. Contact Lenses 2013. Contact Lens Spectrum. 2014;29:22-8
2. Use of International Standard ISO-10993, ‘Biological Evaluation of Medical Devices Part 1: Evaluation and Testing.’
3. Fonn D, Pritchard N, Brazeau D, Michaud L. Discontinuation of contact lens wear: The numbers, reasons and patient profiles. Invest Ophthalmol Vis Sci. 1995;36(4):S312. Abstract 1455.
4. Weed K, Fonn D, Potvin R. Discontinuation of contact lens wear. Optom Vis Sci 1993;70(12s):140.
5. Hart DE, Schkolnick JA, Bernstein S, Wallach D, Gross DF. Contact lens induced giant papillary conjunctivitis: a retrospective study. Journal of the American Optometric Association. 1989;60(3):195-204.
6. Versura P, Maltarello MC, Caramazza R, Laschi R. Immunocytochemical analysis of contact lens surface deposits in transmission electron microscopy. Current eye research. 1988;7(3):277-86.
7. Abdelkader A. Cosmetic soft contact lens associated ulcerative keratitis in southern Saudi Arabia. Middle East African journal of ophthalmology. 2014;21(3):232-5.
8. Dhiman R, Singh A, Tandon R, Vanathi M. Contact lens induced Pseudomonas keratitis following descemet stripping automated endothelial keratoplasty. Contact lens & anterior eye : the journal of the British Contact Lens Association. 2015.
9. Nichols JJ, Sinnott LT. Tear film, contact lens, and patient-related factors associated with contact lens-related dry eye. Invest Ophthalmol Vis Sci. 2006;47(4):1319-28.
10. Morgan PB, Woods CA, Tranoudis IG, Helland M, Efron N, Jones L, et al. International Contact Lens Prescribing in 2014. Contact Lens Spectrum. 2015;30:28-33.
11. Nichols JJ. Deposition on silicone hydrogel lenses. Eye & contact lens. 2013;39(1):20-3.
12. Luensmann D, Jones L. Protein deposition on contact lenses: The past, the present, and the future. Contact Lens and Anterior Eye. 2012;35(2):53-64.
13. Pucker AD, Thangavelu M, Nichols JJ. In vitro lipid deposition on hydrogel and silicone hydrogel contact lenses. Invest Ophthalmol Vis Sci. 2010;51(12):6334-40.
14. Jones L, Senchyna M, Glasier MA, Schickler J, Forbes I, Louie D, et al. Lysozyme and lipid deposition on silicone hydrogel contact lens materials. Eye & contact lens. 2003;29(1 Suppl):S75-9; discussion S83-4, S192-4.
15. Maissa C, Guillon M, Cockshott N, Garofalo RJ, Lemp JM, Boclair JW. Contact lens lipid spoliation of hydrogel and silicone hydrogel lenses. Optom Vis Sci. 2014;91(9):1071-83.
16. Lorentz H, Jones L. Lipid deposition on hydrogel contact lenses: how history can help us today. Optom Vis Sci. 2007;84(4):286-95.
17. Tighe BJ. A decade of silicone hydrogel development: surface properties, mechanical properties, and ocular compatibility. Eye & contact lens. 2013;39(1):4-12.
18. Begley CG, Chalmers RL, Mitchell GL, Nichols KK, Caffery B, Simpson T, et al. Characterization of ocular surface symptoms from optometric practices in North America. Cornea. 2001;20(6):610-8.
19. Nichols JJ, Ziegler C, Mitchell GL, Nichols KK. Self-reported dry eye disease across refractive modalities. Invest Ophthalmol Vis Sci. 2005;46(6):1911-4.
20. Giannoni AG, Nichols JJ. 2012 Annual Report on Dry Eye Diseases. Contact Lens Spectrum. 2012;27(July):26 – 30.
21. Pritchard N, Fonn D, Brazeau D. Discontinuation of contact lens wear: a survey. International contact lens clinic (New York, NY). 1999;26(6):157-62.
22. Jones L, Brennan NA, Gonzalez-Meijome J, Lally J, Maldonado-Codina C, Schmidt TA, et al. The TFOS International Workshop on Contact Lens Discomfort: report of the contact lens materials, design, and care subcommittee. Invest Ophthalmol Vis Sci. 2013;54(11):Tfos37-70.
23. Pucker AD, Nichols JJ. Analysis of meibum and tear lipids. The ocular surface. 2012;10(4):230-50.
24. Butovich IA, Uchiyama E, McCulley JP. Lipids of human meibum: mass-spectrometric analysis and structural elucidation. Journal of Lipid Research. 2007;48(10):2220-35.
25. Chen J, Green-Church KB, Nichols KK. Shotgun Lipidomic Analysis of Human Meibomian Gland Secretions with Electrospray Ionization Tandem Mass Spectrometry. Invest Ophthalmol Vis Sci. 2010;51(12):6220-31.
26. Chen Y, Mehta G, Vasiliou V. Antioxidant Defenses in the Ocular Surface. The ocular surface. 2009;7(4):176-85.
27. Campbell D, Griffiths G, Tighe BJ. Tear analysis and lens-tear interactions: part II. Ocular lipids-nature and fate of meibomian gland phospholipids. Cornea. 2011;30(3):325-32.
28. Griffiths HR, Moller L, Bartosz G, Bast A, Bertoni-Freddari C, Collins A, et al. Biomarkers. Molecular aspects of medicine. 2002;23(1-3):101-208.
29. Glasson M, Stapleton F, Willcox M. Lipid, lipase and lipocalin differences between tolerant and intolerant contact lens wearers. Current eye research. 2002;25(4):227-35.
30. Panthi S, Nichols JJ. Imaging Approaches for Contact Lens Deposition. Eye & contact lens. 9000;Publish Ahead of Print.
31. Brennan NA, Coles MLC. Deposits and symptomatology with soft contact lens wear. ICLC. 2000;27(3):75-100.
32. Mirejovsky D, Patel AS, Rodriguez DD, Hunt TJ. Lipid adsorption onto hydrogel contact lens materials. Advantages of Nile red over oil red O in visualization of lipids. Optom Vis Sci. 1991;68(11):858-64.
33. Pucker AD, Nichols JJ. A method of imaging lipids on silicone hydrogel contact lenses. Optom Vis Sci. 2012;89(5):E777-87.
34. Du L, Hickey RW, Bayir H, Watkins SC, Tyurin VA, Guo F, et al. Starving neurons show sex difference in autophagy. J Biol Chem. 2009;284(4):2383-96.
35. Ma Y, Yates J, Liang Y, Lemon SM, Yi M. NS3 helicase domains involved in infectious intracellular hepatitis C virus particle assembly. J Virol. 2008;82(15):7624-39.
36. Majka SM, Miller HL, Helm KM, Acosta AS, Childs CR, Kong R, et al. Analysis and Isolation of Adipocytes by Flow Cytometry. Methods in enzymology. 2014;537:281-96.
37. Grandl M, Schmitz G. Fluorescent high-content imaging allows the discrimination and quantitation of E-LDL-induced lipid droplets and Ox-LDL-generated phospholipidosis in human macrophages. Cytometry Part A : the journal of the International Society for Analytical Cytology. 2010;77(3):231-42.
38. Liu Y, Kam WR, Ding J, Sullivan DA. Effect of azithromycin on lipid accumulation in immortalized human meibomian gland epithelial cells. JAMA Ophthalmol. 2014;132(2):226-8.
39. Pucker AD, Thangavelu M, Nichols JJ. Enzymatic quantification of cholesterol and cholesterol esters from silicone hydrogel contact lenses. Invest Ophthalmol Vis Sci. 2010;51(6):2949-54.
40. Otsu N. A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man and Cybernetics. 1979;9(1):62–6.
41. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature methods. 2012;9(7):676-82.
42. Peng CC, Fajardo NP, Razunguzwa T, Radke CJ. In Vitro Spoilation of Silicone-Hydrogel Soft Contact Lenses in a Model-Blink Cell. Optom Vis Sci. 2015;92(7):768-80.
43. Ho CH, Hlady V. Fluorescence assay for measuring lipid deposits on contact lens surfaces. Biomaterials. 1995;16(6):479-82.
44. Koenig U, Fobker M, Lengauer B, Brandstetter M, Resch GP, Groger M, et al. Autophagy facilitates secretion and protects against degeneration of the Harderian gland. Autophagy. 2015;11(2):298-313.
45. Razolli DS, Moraes JC, Morari J, Moura RF, Vinolo MA, Velloso LA. TLR4 expression in bone marrow-derived cells is both necessary and sufficient to produce the insulin resistance phenotype in diet-induced obesity. Endocrinology. 2015;156(1):103-13.
46. Abramczyk H, Surmacki J, Kopec M, Olejnik AK, Lubecka-Pietruszewska K, Fabianowska-Majewska K. The role of lipid droplets and adipocytes in cancer. Raman imaging of cell cultures: MCF10A, MCF7, and MDA-MB-231 compared to adipocytes in cancerous human breast tissue. Analyst. 2015;140(7):2224-35.
47. Dionne K, McDermott AM, Nichols JJ, Zhu H, Nichols KK. The Impact of Treatment on 13-Cis Retinoic Acid-Challenged Human Meibomian Gland Epithelial Cells. Investigative Ophthalmology & Visual Science. 2014;55(13):20-.
48. Jester JV, Potma E, Brown DJ. PPARgamma Regulates Mouse Meibocyte Differentiation and Lipid Synthesis. The ocular surface. 2016;14(4):484-94.
49. Mauris J, Dieckow J, Schob S, Pulli B, Hatton MP, Jeong S, et al. Loss of CD147 results in impaired epithelial cell differentiation and malformation of the meibomian gland. Cell Death Dis. 2015;6:e1726.