Abstract Dry eye disease (DED) is the most common ocular surface disorder with meibomian gland dysfunction (MGD) as a leading cause. However, precise diagnosis of MGD based on the intrinsic composition remains unattainable, leading to delayed diagnosis and suboptimal therapeutic interventions. This study proposes a high-precision hyperspectral pathological diagnosis scheme for MGD based on a real-time spectral convolutional neural network (SCNN) chip. We analyze the spectral characteristics of meibum, hemoglobin, and its derivatives of meibomian gland pathological sections. The neural network trained on the SCNN chip for MGD diagnosis achieved a diagnostic accuracy of 96.22%, outperforming models based on RGB images of 84.00%. To our knowledge, this is the first study to develop a spectral pathological diagnostic model for MGD and to apply optical neural networks for the diagnosis of MGD. This finding offers new possibilities for effective and accurate MGD management and promotes the performance of SCNN chip for medical applications.
This study aimed to compare the utility of four different dyes for intracellular lipid detection in immortalized human meibomian gland epithelial cells (IHMGECs). IHMGECs were cultured in a serum-containing medium for 10 days in the presence or absence of Roxadustat (Roxa), a known inducer of IHMGEC differentiation. Cells were then fixed and stained with Oil Red O (ORO), Sudan III (SIII), LipidTOX green (LT), or Nile Red (NR). IHMGECs were evaluated for the number, size, and area of stained intracellular lipid vesicles or the intensity of staining using bright field (ORO, SIII) or fluorescence (LT, NR) microscopy. Data were captured with ImageJ and analyzed with Student's two-tailed t-test. Our findings demonstrate that different staining methods can yield significantly different patterns of intracellular lipid quantity and/or distribution in IHMGECs. ORO and SIII significantly increased the size and area of lipid-containing vesicles in Roxa-treated cells. Neither stain showed a change in the number of vesicles during IHMGEC differentiation. Vesicle size was significantly greater in cells stained with ORO, as compared to SIII. In addtion, LT, but not NR, showed a significant increase in intracellular lipid intensity in IHMGECs following Roxa -induced differentiation. Our results demonstrate significant differences in the distribution patterns and intensities of lipid-containing vesicles in IHMGECs after staining with ORO, SIII, LT, and NR. ORO, SIII, and LT, but not NR staining, are helpful methods to help identify and quantitate the extent of intracellular lipid accumulation during IHMGEC differentiation.
We recently discovered that by changing environmental signals, differentiated immortalized human meibomian gland epithelial cells (IHMGECs) de-differentiate into proliferating cells. We also discovered that following exposure to appropriate stimuli, these proliferative cells re-differentiate into differentiated IHMGECs. We hypothesize that this plasticity of differentiated and proliferative IHMGECs is paralleled by very significant alterations in cellular gene expression. To begin to test this hypothesis, we compared the gene expression patterns of IHMGECs during differentiation and proliferation. IHMGECs were cultured for four days in either differentiating or proliferating media. After four days of culture, cells were processed for the analysis of gene expression by using Illumina BeadChips and bioinformatic software. Our study identified significant differences in the expression of more than 9200 genes in differentiated and proliferative IHMGECs. Differentiation was associated with significant increases in the expression of specific genes (e.g. S100 calcium binding protein P; 7,194,386-fold upregulation) and numerous ontologies (e.g. 83 biological process [bp] ontologies with ≥100 genes were upregulated), such as those related to development, transport and lysosomes. Proliferation also led to a significant rise in specific gene expressions (e.g. cathelicidin antimicrobial peptide; 859,100-fold upregulation) and many ontologies (115 biological process [bp] ontologies with ≥100 genes were upregulated), with most of the highly significant ontologies related to cell cycle (z scores > 13.9). Our findings demonstrate that gene expression in differentiated and proliferative IHMGECs is extremely different. These results may have significant implications for the regeneration of HMGECs and the reversal of MG dropout in MG dysfunction.
Purpose Clinical studies have indicated that the long-term use of topical antiglaucoma drugs, such as carbonic anhydrase inhibitors (CAIs), may lead to meibomian gland dysfunction (MGD). We hypothesize that these adverse effects involve a direct influence on human MG epithelial cells (HMGECs). The purpose our present investigation was to test our hypothesis and determine whether exposure to dorzolamide, a CAI, impacts the proliferation, intracellular signaling and differentiation of HMGECs. Materials and Methods We cultured immortalized (i) HMGECs with vehicle or various concentrations of dorzolamide for 6 days. Cells were enumerated with a hemocytometer, and examined for their morphology, Akt signaling activity, accumulation of neutral lipids, phospholipids and lysosomes, and the expression of protein biomarkers for lipogenesis regulation, lysosomes and autophagosomes. Results Our results show that a high, 500 mu g/ml concentration of dorzolamide causes a significant decrease in Akt signaling and the proliferation of iHMGECs. However, the high dose of dorzolamide also promotes the differentiation of iHMGECs. This response features increases in the number of lysosomes, the accumulation of phospholipids, and the expression of the light chain 3A biomarker for autophagosomes. In contrast, the therapeutic amount (50 mu g/ml) of dorzolamide has no impact on the proliferative or differentiative abilities of iHMGECs. Conclusions Our results support our hypothesis and demonstrate that the CAI dorzolamide does exert a direct influence on the proliferation and differentiation of iHMGECs. However, this effect is elicited only by a high, and not a therapeutic, amount of dorzolamide.