Human retinal endothelial cells

HRECs (Cat. No. ACBRI 181, Cell Systems Corp., Kirkland, WA, USA) were cultured in Dulbecco’s modified Eagle Medium (DMEM) supplemented with fetal bovine serum, endothelial cell growth supplement, glutamax, insulin-transferrin-selenium and antibiotic/antimycotic. Cells (~ 80% confluent) from the 6th to 8th passage were incubated in 5 mM d-glucose (NG group), or in 20 mM d-glucose without (HG group) or with 100 µM l-homocysteine thiolactone hydrochloride (Cat. No. S784036; Sigma-Aldrich, St. Louis, MO, USA; Hcy group) for 24 to 96 h. Based on Hcy levels in diabetic patients ranging from 16 to over 100 µM (moderate to severe hyperhomocysteinemia), 100 µM homocysteine was used in our study; this concentration was shown to result in mitochondrial damage without having any effect on the cell phenotype [21, 24, 26]. To analyze the effect of H2S regulation, 150 µM GYY4137 (Cat. No. SML0100; Sigma-Aldrich) was added to a group of cells incubated with high glucose + homocysteine (GY group) as described previously [24, 27]. As an osmotic control, each experiment included cells incubated with 20 mM l-glucose (L-Gl group), instead of 20 mM d-glucose. Incubation medium was replaced every 48 h, and fresh homocysteine or GYY4137 was supplemented during each media change. Duplicate measurements were taken and each experiment was repeated with three to five different HREC preparations.

Animals

Mice deficient in Cbs, a model of hyperhomocysteinemia, were generated by breeding Cbs heterozygous (Cbs+/−) mice (B6.129P2-Cbstm1Unc/J; Jackson Laboratories, Bar Harbor, ME, USA), and their genotyping was performed following the method reported by others [28, 29]. Since Cbs−/− mice have a very short life span, studies were performed using Cbs+/− mice, and wildtype (Cbs+/+) mice were used as controls. Diabetes was induced in 8 to 10 weeks old Cbs+/− and Cbs+/+ mice (male and female) by streptozotocin administration (55 mg/kg) for four consecutive days. Mice were maintained diabetic (blood glucose > 350 mg/dL) for eight weeks to six months. Age-matched nondiabetic mice served as the controls, and each group had 6–8 mice. The retina was isolated immediately and was used for measurement of biochemical/molecular parameters. For mice diabetic for over 16 weeks, the retina was isolated from one eye and the other eye was fixed in 10% buffered formalin for histopathology [30, 31]. Animal care and maintenance followed Wayne State University’s institutional guidelines for treating animals humanely, and Association for Research in Vision and Ophthalmology’s declaration for using animals in ophthalmic and vision research (protocol #21-03-3331). This study is reported in accordance with the Animal research: reporting of in vivo experiments (ARRIVE) guidelines.

Retinal microvessels

Microvessels were prepared from the mouse retina following the hypotonic shock method. Briefly, the retina was placed in distilled water for an hour at 37 °C, and the vasculature was prepared as reported previously [27, 30].

Homocysteine

Total homocysteine was measured in the retina (25 µg protein) using an ELISA kit (Cat. No. STA-670; Cell Bio Labs Inc., San Diego, CA, USA) using homocysteine-BSA as a standard, as reported previously [22].

Activity of cystathione β synthase (Cbs)

Cbs activity was assessed utilizing 40 to 60 µg of retinal protein in accordance with the cystathionine-β-synthase activity assay kit (Cat. No. K998; Bio Vision, Milpitas, CA, USA) using cysteine and homocysteine as substrates [24]. To ascertain the specificity of Cbs activity, 7-amino-4-methylcoumarin was employed as a standard.

Hydrogen sulfide

H2S was quantified using the methylene blue assay [22, 24]. Briefly, cell culture medium (250 µL) or retinal homogenate (50 µg protein) was incubated with 1% zinc acetate at room temperature for 20 min. This was followed by the addition of 20 mM N-dimethyl-p-phenylenediamine sulfate and 30 mM FeCl3, and the mixture was incubated at 37 °C for one hour in the dark. After precipitating the protein with 10% trichloroacetic acid, followed by centrifugation at 10,000xg for 5 min, the absorbance of the supernatant was measured at 670 nm. Sodium hydrosulfide (Cat. No. 161527; Sigma-Aldrich) was employed as a standard.

Mitochondrial ROS

Mitochondrial ROS were quantified fluorometrically in retinal microvessels using MitoSox red (Cat. No. M36008; Thermo Fisher Scientific, Waltham, MA, USA). Cell homogenate (10 µg protein) was incubated in a microplate with 5 µM MitoSox red in the dark for 10 min at 37 °C, and the fluorescence was measured at excitation/emission wavelengths of 500 nm/580 nm. As a positive control, 5 mM d-glucose incubated cells were exposed to 2 μM antimycin A for one-hour before being analyzed.

RNA extraction and mtDNA damage

Total RNA was extracted using TRIzol reagent as per manufacturer’s instructions (Cat. No. 15596026, Invitrogen, Waltham, MA, USA). Purity of the extracted RNA was assessed by quantifying the ratio of absorbance at 260 nm and 280 nm; a ratio of ~ 2.0 was considered pure. cDNA synthesis was processed after treating RNA with DNase I. Mitochondrial DNA (mtDNA) damage was evaluated in HRECs or retinal microvessels by quantifying gene transcripts of mtDNA-encoded NADH-ubiquinone oxidoreductase chain 6 (ND6) and cytochrome B (CytB) by SYBR-green based real-time polymerase chain reaction (PCR), as reported previously [32], using gene- and species-specific primers (Table 1). Relative expression of the gene was calculated by the Delta-Delta-Ct method using β-actin (HRECs) or 18S rRNA (mouse) as the housekeeping gene [32].

DNA isolation and mitochondrial copy numbers

Genomic DNA was isolated from HRECs or retinal microvessels using the DNeasy kit (Cat. No. 69504, Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol. Mitochondrial copy numbers were quantified in the genomic DNA by SYBR-green based amplification of the mitochondrial marker, CytB, and nuclear marker (β-actin or 18S rRNA), using DNA primers provided in Table 1 [32].

Oxygen consumption rate (OCR)

A Seahorse XF analyzer (Agilent Technologies, Santa Clara, CA, USA) was employed to measure OCR. HRECs (~ 90% confluent) were trypsinized, and after washing the cell pellet with DMEM, it was suspended in the growth medium. The suspension (10 μL) was then loaded into a Neubauer chamber (Microyn Technologies, Baltimore, MD, USA) to count the cells, and was diluted to achieve a concentration of 50,000 cells/mL. Cell suspension (100 μL) was seeded in each well of the 96 well Seahorse culture plate. At the end of experimental treatments, HRECs were washed with 100 µL assay medium (Seahorse XF medium supplemented with 1 mM pyruvate, 2 mM glutamine and 10 mM glucose), and were then incubated at 37 °C for 30 to 45 min with 180 µL assay medium. OCR was measured using Seahorse XF Cell Mito Stress Test kit (Cat. No. 103015-100, Agilent Technologies) by injecting 1.5 µM oligomycin, 2.0 µM FCCP and 0.5 µM rotenone/antimycin A in ports A, B and C, respectively. The data was analyzed using the Wave software (Agilent Technologies), as reported recently [13].

Mitophagosome formation

Mitophagosome formation was quantified by an Autophagy Detection kit (Cat. No. ab139484; Abcam, Boston, MA, USA). Following experimental incubations, cells were washed with the assay buffer (supplied by the manufacturer) containing 5% fetal serum and incubated with the green detection reagent in the dark for 30 min at room temperature. MitoTracker red and Hoechst were used to counterstain the mitochondria and nuclei, respectively. Cells were imaged under a Zeiss Apotome at 20 × objective, and arithmetic mean intensity (AMI) of green fluorescence was calculated using the Zeiss software module. Each experiment included rapamycin (1 µM) as an autophagy inducer control.

Mitophagy flux

Mitophagy flux was assessed in live cells using the Mitophagy detection kit (Cat. No. MD01-10, Dojindo Molecular Technologies, Rockville, MD, USA). Briefly, cells were washed with DMEM and incubated with 100 nM Mtphagy dye for 30 min at 37 °C in dark. After washing the cells two times with FACS buffer (0.5% BSA in PBS w/v) to remove trypsin, scattering of Mtphagy Dye was quantified by flow cytometry under PerCP Cy5.5 channel at an excitation/emission wavelength of 486 nm/679 nm. Flow cytometry standard files were analyzed by FlowJo v.10.8.1 software (BD Biosciences, San Jose, USA), and relative Mtphagy scattering for each group was plotted, as recently reported [13]. As a positive control, 1 μM rapamycin was used.

Mitophagy flux in the mouse retina was quantified by flow cytometry [13]; briefly, tissue disaggregation was carried out by incubating small pieces of retina with 50 µL Accumax™ (Sigma-Aldrich, USA) for 10 min at 37 °C. The resultant suspension was filtered through a 40 µm cell strainer and incubated for 30 min at 37 °C with staining solution containing 50 nM MitoTracker deep red (MTDR, Cat No. M22426, Thermo Fisher Scientific). Cells were washed with FACS buffer (2X) and scanned under FL3 640 nm wavelength using BD Accuri C6 plus flow cytometer. Raw Flow Cytometry Standard files were analyzed employing FlowJo software, and relative MTDR scattering in each group was plotted, as reported recently [13].

Retinal histopathology

Retina isolated from the formalin-fixed eyes was rinsed overnight in running water and incubated at 37 °C with 3% crude trypsin (Invitrogen-Gibco, Grand Island, NY, USA) containing 200 mM sodium fluoride for 45 to 70 min. After cleaning the retinal vasculature under a dissecting microscope, it was stained with periodic acid-Schiff-hematoxylin to count acellular capillaries [24, 31].

Retinal vascular permeability

Permeability of retinal vasculature was determined by fluorescein angiography using a Micron IV retinal imaging microscope (Phoenix Research Laboratories, Pleasanton, CA, USA). After anesthetizing the animals with Ketamine-Xylazine (67 mg and 10 mg per kg, respectively; i.p.) and dilating pupils with 0.1% tropicamide ophthalmic solution and lubricating their corneas with Goniovisc (hypromellose 2.5%). The fundus photographs were taken by fundus camera for small animals. AK-FLUOR (0.5% solution, 0.01 mL/g BW) was then injected intraperitoneally, and the fundus was photographed using a barrier filter for fluorescein angiography, as reported previously [24, 31].

Statistical analysis

GraphPad Prism was used to perform statistical analysis. Data are presented as mean ± standard deviation of three or more experiments. Inter-group comparisons were made using one-way ANOVA followed by post hoc Tukey test. P values were calculated using the Student’s t-test and P values less than 0.05 were considered statistically significant.