Sphingomyelinases in retinas and optic nerve heads: Effects of ocular hypertension and ischemia

Abstract

Sphingomyelinases (SMase), enzymes that catalyze the hydrolysis of sphingomyelin to ceramide, are important sensors for inflammatory cytokines and apoptotic signaling. Studies have provided evidence that increased SMase activity can contribute to retinal injury. In most tissues, two major SMases are responsible for stress-induced increases in ceramide: acid sphingomyelinase (ASMase) and Mg²⁺-dependent neutral sphingomyelinase (NSMase). The purposes of the current study were to determine the localization of SMases and their substrates in the retina and optic nerve head and to investigate the effects of ocular hypertension and ischemia on ASMase and NSMase activities.

Tissue and cellular localization of ASMase and NSMase were determined by immunofluorescence imaging. Tissue localization of sphingomyelin in retinas was further determined by Matrix-Assisted Laser Desorption/Ionization mass spectrometry imaging. Tissue levels of sphingomyelins and ceramide were determined by liquid chromatography with tandem mass spectrometry. Sphingomyelinase activities under basal conditions and following acute ischemic and ocular hypotensive stress were measured using the Amplex Red Sphingomyelinase Assay Kit. Our data show that ASMase is in the optic nerve head and the retinal ganglion cell layer. NSMase is in the optic nerve head, photoreceptor and retinal ganglion cell layers. Both ASMase and NSMase were identified in human induced pluripotent stem cell-derived retinal ganglion cells and optic nerve head astrocytes. The retina and optic nerve head each exhibited unique distribution of sphingomyelins with the abundance of very long chain species being higher in the optic nerve head than in the retina. Basal activities for ASMase in retinas and optic nerve heads were 54.98 ± 2.5 and 95.6 ± 19.5 mU/mg protein, respectively. Ocular ischemia significantly increased ASMase activity to 86.2 ± 15.3 mU/mg protein in retinas (P = 0.03) but not in optic nerve heads (81.1 ± 15.3 mU/mg protein). Ocular hypertension significantly increased ASMase activity to 121.6 ± 7.3 mU/mg protein in retinas (P < 0.001) and 267.0 ± 66.3 mU/mg protein in optic nerve heads (P = 0.03). Basal activities for NSMase in retinas and optic nerve heads were 12.3 ± 2.1 and 37.9 ± 8.7 mU/mg protein, respectively. No significant change in NSMase activity was measured following ocular ischemia or hypertension.

Our results provide evidence that both ASMase and NSMase are expressed in retinas and optic nerve heads; however, basal ASMase activity is significantly higher than NSMase activity in retinas and optic nerve heads. In addition, only ASMase activity was significantly increased in ocular ischemia or hypertension. These data support a role for ASMase-mediated sphingolipid metabolism in the development of retinal ischemic and hypertensive injuries.

1. Introduction

Glaucoma is an optic neuropathy associated with characteristic changes in the optic nerve and a loss of retinal ganglion cells (RGCs) and their axons (Davis et al., 2016; Nuschke et al., 2015). Although there are many different types of glaucoma, it is often associated with elevated intraocular pressure (IOP) (Burgoyne et al., 2005) due to the blockage or dysfunction of the conventional outflow pathway. Current treatments for glaucoma are primarily centered on pharmacological, surgical or laser approaches to lower IOP. Retinal ischemia is considered a common pathophysiological event associated with several ocular disorders such as age-related macular degeneration, diabetic retinopathy, glaucoma or retinal vascular occlusion (Osborne et al., 2004). Although retinal ischemia and glaucoma differ in their clinical presentation, they share many pathophysiological features. These include elevated extracellular glutamate, cytokine and matrix metalloproteinase secretion, glial activation and inner retinal degeneration (Renner et al., 2017). Understanding the mechanisms involved in neuronal cell death is essential for preventing or slowing the development of glaucoma and ischemia-associated vision loss.

Ceramide, a bioactive sphingolipid, regulates many physiological and pathological processes (Chen et al., 2014; Mondal and Mandal, 2019). In the retina, studies have demonstrated that ceramide levels are elevated in several retinal injury models, and suppressing ceramide elevation can provide significant neuroprotection from light damage, ischemic and ocular hypertensive injuries (Aslan et al., 2014; Chen et al., 2013; Fan et al., 2016, 2021). Although ceramides can be generated by three independent routes, the sphingomyelinase pathway, the de novo synthesis pathway, and the salvage pathway (Canals et al., 2018), the sphingomyelinase pathway is the route most often associated with stress-induced ceramide production. Sphingomyelinases catalyze the hydrolysis of sphingomyelin to ceramide (Kanfer et al., 1966). In non-ocular tissues, multiple extracellular stress signals, such as heat shock, ischemia/reperfusion, ionizing radiation and oxidative stress, induce sphingomyelinase activity (Andrieu-Abadie et al., 2001; Chang et al., 1995; Fan et al., 2016; Kolesnick and Fuks, 2003; Novgorodov and Gudz, 2011). Multiple sphingomyelinases have been identified; however, the two major enzymes mediating stress-induced ceramide production are acid sphingomyelinase (ASMase) and Mg²⁺-dependent neutral sphingomyelinase (NSMase) (Goni and Alonso, 2002; Samet and Barenholz, 1999).

Activation of ASMase is often an early response to various cellular stresses and precedes ceramide production, which accelerates signaling pathways involved in cell death (Marchesini and Hannun, 2004). Previous studies have shown that elevation of ceramide levels occurs in several in vivo ischemia models, including heart (Bielawska et al., 1997; Cordis et al., 1998; Cui et al., 2004; Der et al., 2006; Zhang et al., 2001), liver (Bradham et al., 1997; Kukimoto et al., 1995), kidney (Zager et al., 1998), brain (Herr et al., 1999; Kubota et al., 1989, 1996; Nakane et al., 2000; Takahashi et al., 2004; Yu et al., 2000) and retinas (Fan et al., 2016). In the brain, the upregulation of ASMase activity is responsible for the ischemia-induced rise in astroglial ceramide levels (Ohtani et al., 2004). The extent of tissue damage in the ischemic brain, heart, liver and retina can be attenuated by the genetic deletion of ASMase or in vivo administration of an AMSase inhibitor (e.g., desipramine) or siRNA (Der et al., 2006; Fan et al., 2016; Llacuna et al., 2006; Opreanu et al., 2010, 2011; Yu et al., 2000). Our initial studies demonstrated that ASMase is necessary for the maintenance of retinal structure and function, and the genetic deletion of ASMase results in age-dependent photoreceptor degeneration and increased RPE lipofuscin accumulation (Wu et al., 2015). More recently, we have shown that reduction of ASMase expression protects against ocular hypertension-induced RGC degeneration (Fan et al., 2021).

The NSMases are a group of Mg²⁺-dependent enzymes with optimal activity observed at neutral pH ranges. The most active and widely studied isoform is NSMase2 (Smpd3). NSMase2 activity and associated rise in ceramide levels in the brain have been linked to the development of age-related cognitive deficits such as Alzheimer and Parkinson’s diseases (Dinkins et al., 2016; Tan et al., 2018). Mechanistic studies have provided evidence that oxidative stress, TNF-α and amyloid-β can all increase the activity and/or expression of NSMase2 (Clarke et al., 2011; Hernandez et al., 2000; Ichi et al., 2009; Jana and Pahan, 2010; Lee et al., 2004). Activation of NSMase2 contributes to the elevation of ceramide levels in astrocytes in rat hippocampi (Gu et al., 2013). Suppressing ceramide production using the NSMase inhibitor GW4869 reduces ocular hypertensive injury (Aslan et al., 2014). However, the specific contributions of ASMase or NSMase activation to ocular hypertensive and retinal ischemic injury remain unclear.

In this study, we investigate how sphingomyelinases, the enzymes primarily responsible for stress-induced ceramide increases that lead to cell death, respond to ischemic and ocular hypertensive injury. The goals of the present study were to determine the expression patterns of ASMase and NSMase in the retina and optic nerve head and to compare their basal activities and early responses of these enzymes to ischemic and ocular hypertensive stresses. We present evidence that both ASMase and NSMase are expressed in the retina and optic nerve head; however, basal ASMase activity was significantly higher than NSMase activity. In the retina, only ASMase activity increased following ocular ischemia or hypertension stress. Along with our previous reports showing reduction of ASMase expression protects against retinal ischemia (Fan et al., 2016) and ocular hypertension-induced RGC degeneration (Fan et al., 2021), the data presented here support the idea that ASMase activation plays an early role in RGC degeneration following ischemic and ocular hypertensive stress.

2. Materials and methods

2.1. Animals

Adult male and female brown Norway rats (3–5 months of age, 125–150 g; Charles River Laboratories, Inc., Wilmington, MA) were used in this study. Rats were maintained in an environmental cycle of 12-h light/12-h dark with ambient light intensity (150 ± 20 lux). All experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research; the study protocol (IACUC-2021–01267) was approved by the MUSC Animal Care and Use Committee.

2.2. Cell culture

Human induced pluripotent stem cells (iPSC) were purchased from WiCell Research Institute (Madison, WI) and were cultured following the manufacturer’s instructions. Human iPSC-derived RGCs were generated from human iPSCs following published procedures (Ohlemacher et al., 2015). Retinal neurospheres differentiated for 50–55 days were collected and dissociated by accutase (Sigma-Aldrich, MO) and incubated with CD90.2 (THY1.2) microbeads (Miltenyi Biotec, Germany) for 20 min at 4 °C. The microbead-containing cells were enriched on a MACS MS column (Miltenyi Biotec, Germany) following the manufacturer’s instructions. Purified cells were cultured overnight in BrainPhys^™ Neuronal medium (Stem Cell Technologies, Canada) including B27 supplement (Thermo Fisher Scientific, MA) on matrigel (Corning, AZ) coated slides. Primary human optic nerve head astrocytes were isolated, purified and cultured as previously described (Yang and Hernandez, 2003).

2.3. Immunofluorescence microscopy

Rat eyes were fixed in 4% paraformaldehyde overnight at 4 °C, cryoprotected by immersion in 30% sucrose for at least 2 h, and then embedded in optimal cutting temperature compound. Frozen retinal cross-sections were cut at a thickness of 12 μm iPSC-derived RGC cells and human optic nerve head astrocytes were fixed for 15 min using 4% paraformaldehyde. Sections or cells were incubated with blocking buffer (3% bovine serum albumin, 4% normal goat serum, 0.1% Triton 100-X in 1x phosphate buffered saline) for 1 h at room temperature and incubated with primary polyclonal Brn3a (Santa Cruz, TX, 1:200), GFAP (Sigma-Aldrich, MO, 1:500), ASMase (ASM H181, Santa Cruz, 1:200) or NSMase (nSMase 2, ECM bioscience, 1:100) antibody overnight at 4 °C. Goat anti-rabbit Alexa 488 or 594 secondary antibody (Molecular Probes) was used at a dilution of 1:500. Propidium iodide at a dilution of 1:3000 was used to visualize nuclei. Retinal sections were imaged using a Leica confocal fluorescence microscope with a 20x objective lens. iPSC-derived RGCs and human optic nerve head astrocytes were imaged using a Nikon A1Rsi confocal fluorescence microscope with 20x and 60x objective lenses.

2.4. LC-MS/MS sphingolipid analyses

Retinas and optic nerve heads (n = 4) were lysed in lysis buffer containing 50 mM Tris-base, 1.0 mM EDTA, 0.2% Triton X-100, pH 7.4 with protease inhibitor mini tablet (Roche). Tissue debris and unlysed cells were pelleted and removed by centrifugation at 300g for 5 min at 4 °C. Cellular homogenates were used for the sphingolipid analyses, which were performed by the Lipidomics Core Facility (Medical University of South Carolina, Charleston, SC, USA) using high-performance liquid chromatography coupled to electrospray ionization followed by tandem mass spectrometry (HPLC-MS/MS) using a triple-stage quadrupole mass spectrometer (TSQ 7000; Thermo Finnigan, San Jose, CA, USA) operating in a multiple reaction monitoring positive ionization mode as described previously (Bielawski et al., 2006).

2.5. MALDI imaging of sphingolipids

Freshly isolated rat eyes were embedded in 2.7% carboxymethyl-cellulose (Sigma Aldrich) and flash frozen with liquid nitrogen vapor. Retinal cross-sections (10 μm) were sprayed with the matrix 2,5-diaminonaphthalene in 90% acetonitrile/water using a TM Sprayer (HTX Imaging) with 10 passes at a 1300 mm/min velocity at 60 °C. Images were obtained in the positive ion mode collected by MALDI-FTICR (7.0 solariX Legacy, Bruker Scientific, LLC) equipped with a Smartbeam2 laser using 25 laser shots per pixel at a 25 μm stepsize. Transients of 1 Mega word were acquired in broadband mode over m/z 200–1600 resulting in an estimated 160,000 resolving power on tissue at m/z 400 calculated at full width half max (FWHM). Data were visualized in flexImaging 4.1 build 116 (Bruker Daltonics) and analyzed for regional distribution in SCiLS Lab software version 2016b. Images showed the distribution of the signals of sphingomyelins 16:0, 18:0 and 24:1. Adjacent sections were post fixed with 4% of paraformaldehyde and stained with H&E.

2.6. Retinal ischemia

Retinal ischemia was induced using techniques described previously (Husain et al., 2009) with minor modification. Briefly, rats (n = 6) were anesthetized with ketamine (75 mg/kg) and xylazine (8 mg/kg) (Henry Schein Medical, Columbus, OH). Proparacaine (5 μL, 0.5%, Akorn, Inc., Buffalo Grove, IL) was applied for corneal analgesia. Body temperature was maintained with a temperature-controlled heating pad (Harvard Apparatus; Holliston, MA) at 37 °C during the experiment. The anterior chamber was cannulated with a 27-gauge needle (World Precision Instruments, Inc., Sarasota, FL) which was connected to a reservoir of sterile PBS, pH 7.4. The container was elevated to raise the intraocular pressure (IOP) to 160 mmHg for 45 min. The IOP was monitored by a calibrated pressure transducer (Argon Medical Devices, Athens, TX), and the absence of retinal blood flow was confirmed by direct ophthalmoscopy. The contralateral eye was untreated and served as a control. Following ischemic injury and recovery from anesthesia, the pupillary light reflex was verified as present and not different from the contralateral control eye. The eyes were then allowed to reperfuse for 90 min, animals were sacrificed, and the tissues were collected for evaluating changes in ASMase and NSMase activity.

2.7. Ocular hypertension

Rats (n =6) were allowed to acclimate to their surroundings for 7–10 days prior to baseline IOP measurements and study initiation. On study day 0, rats were anesthetized with ketamine (75 mg/kg) and xylazine (8 mg/kg) (Henry Schein Medical, Columbus, OH), and corneal analgesia achieved with the application of proparacaine (0.5%; 5 μL; Akorn, Inc., Buffalo Grove, IL). Body temperature was maintained at 37 °C with a heating pad (Harvard Apparatus; Holliston, MA). A glass micropipette was used to inject microbeads into the anterior chamber as described previously (Bunker et al., 2015; Ito et al., 2016). A total of 25 μL of sterile 8 μm magnetic microbeads (Bangs Laboratories, Inc) in a 9× 10⁷ microbeads/mL solution was injected. Contralateral eyes were untreated and served as controls. Prophylactic neomycin antibacterial ointment was applied to the site of injection to prevent infection. Intraocular pressure was recorded as discrete readings using a calibrated Tonolab tonometer (Colonial Medical Supply Co., Inc., Franconia, NH). Tissues were collected 14 days after microbead injection.

2.8. Sphingomyelinase activities

Retinas and optic nerve heads were isolated 90 min following acute retinal ischemia and 14 days following the chronic elevation in IOP. Retinas and optic nerve heads were dissected into halves, with one half used to measure ASMase activity and the other half NSMase. To determine ASMase activity, tissues were placed in lysis buffer (50 mM sodium acetate, 1 mM EDTA, 1% Triton 100-X, protease inhibitor cocktail [Roche], pH 5.0). To determine NSMase activity, tissues were placed in a different lysis buffer (20 mM Tris HCl, 1 mM EDTA, 1% Triton 100X, protease inhibitor cocktail, pH 7.5). All samples were homogenized by sonication with three 5 s pulses at 4 °C. The homogenates were centrifuged at 12,000 g for 4 min, and the supernatants were used for determining protein levels and measuring ASMase or NSMase activity by means of the commercial Amplex Red Sphingomyelinase Assay Kit following the manufacturer’s instructions (Molecular Probes, Eugene, OR). Fluorescence was measured in a fluorescence microplate reader with excitation at 530 nm and emission detection at 590 nm. The readings were corrected for background fluorescence by subtracting the values derived from the negative control and then normalizing to protein levels. Protein levels were determined from the same supernatants used for the corresponding SMase activity assay with the Pierce BCA assay (ThermoFisher Scientific).

2.9. Statistics

For all experiments, data were expressed as mean ± SEM. Data were analyzed using a two-tailed Student t-test or ANOVA. A P < 0.05 was considered significant.

3. Results

3.1. Localization of ASMase and NSMase

To begin to understand how ASMase or NSMase might contribute to stress-induced retinal degeneration, we evaluated the expression profiles of these enzymes in the retina and optic nerve head. As shown in Fig. 1A, expression of ASMase and NSMase in the retina displays a distinct pattern. Retinal ASMase is mainly localized in the inner retinal layers, with the highest immunolocalization evident in the retinal ganglion cell layer. Little or no ASMase expression was observed in the outer retina. On the other hand, retinal NSMase is primarily expressed in the photoreceptor outer segments although it is still present in the retinal ganglion cell layer at lower levels. In the optic nerve head, both ASMase and NSMase are expressed in a diffuse pattern (Fig. 1B). The specificity of the ASMase and NSMase antibodies was verified by the absence of autofluorescence and nonspecific binding of the secondary antibody (left panels, Fig. 1). To confirm the cellular expression of ASMase and NSMase in the retinal ganglion cells and optic nerve head astrocytes and to provide information regarding human expression, we evaluated the expression of these enzymes in iPSC-derived human RGCs and purified human optic nerve head astrocytes. Brn3a and GFAP polyclonal antibodies were used for labeling iPSC-derived RGCs and astrocytes, respectively (Fan et al., 2021). As shown in Fig. 2A and B, both ASMase and NSMase are expressed in each cell type.

Fig. 1.

Localization of ASMase and NSMase in rat (A) retina and (B) optic nerve head. Immunofluorescence of tissue probed with the secondary antibody alone as a negative control (green if non-specific binding or autofluorescence is present, left panels), ASMase (green, middle panels) and NSMase (green, right panels), overlaid with the nuclei imaged with propidium iodide (red). PRL, photoreceptor layer; ONL, outer nuclear layer; INL, inner nuclear layer; IPL, inner plexiform layer; RGC, ganglion cell layer. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2.

Localization of ASMase and NSMase in (A) induced pluripotent cell derived RGCs and (B) purified human optic nerve head astrocytes. Brn3a and GFAP polyclonal antibodies were used for labeling iPSC-derived RGCs and astrocytes, respectively.

3.2. Distribution of sphingomyelins and ceramides

Sphingomyelins are substrates for both ASMase and NSMase. We utilized LC-MS/MS to determine the ocular distribution and levels of sphingomyelin and its enzymatic product ceramide in the retina and optic nerve head. Ceramide and sphingomyelin profiles in rat retinas and optic nerve heads are presented in Fig. 3. C16:0-, C18:0-, and C20:0-ceramides are the three dominant species in the retina. The overall retinal ceramide profiles are similar to those detected in mouse retina (Bruggen et al., 2016; Fan et al., 2016; Garanto et al., 2013; Yu et al., 2019). However, in the optic nerve head, longer chain ceramides are more dominant with the C18:0, C20:0, C24:1-ceramides exhibiting the highest levels (Fig. 3B). In order to determine whether these different sphingolipid species localize in different retinal layers like ASMase and NSMase, we used mass spectrometry MALDI imaging of frozen cross-sections to determine the ocular distribution and relative levels of sphingomyelin in the retina and optic nerve head. As shown in Fig. 4, sphingomyelins are not uniformly distributed across the retina. The sphingomyelin species 16:0 and 24:1 are mainly localized in the retinal ganglion cell layer, optic nerve and choroidal regions. The sphingomyelin species 18:0 is present throughout all retinal layers, optic nerve and choroidal regions (Fig. 4). Normal retinal morphology was verified with H&E staining of adjacent sections.

Fig. 3.

(A) Ceramide and (B) sphingomyelin LC-MS profiles of rat retinas and optic nerve heads. Data are represented as the mean ± SEM; n = 4.

Fig. 4.

Localization of sphingomyelin species in the retina and optic nerve head. Unfixed rat eyes were embedded in 2.7% carboxymethyl-cellulose and flash frozen. MALDI-IMS was performed on retinal cross-sections. MALDI-IMS images display the relative abundances of sphingomyelins 16:0, 18:0 and 24:1. Adjacent sections were stained with H&E, and shown in the left panel for this set of data.

3.3. ASMase and NSMase activities

To investigate whether acute ischemia/reperfusion or chronic ocular hypertensive stress can alter ASMase and NSMase activities in the retina and optic nerve head, tissues were isolated 90 min following unilateral retinal ischemia or 14 days following unilateral IOP elevation. Contralateral baseline IOP was 9.9 ± 0.2 mmHg. On days 7, 10 and 14 following a single microbead injection, IOPs from all the injected eyes remained elevated significantly (P < 0.001), compared to the untreated contralateral control eyes (Fig. 5).

Fig. 5.

Intraocular pressure were measured on day 0 prior to microbead injection, and days 3, 7 and 14 after microbead injection. Significant differences (P < 0.001, n = 6) in IOP were observed between the contralateral eyes and microbead injected hypertensive eyes on days 7, 11 and 14.

To determine basal ASMase and NSMase retinal activities, data from the untreated contralateral eyes from both ischemia and ocular hypertension groups were combined, as subgroup analysis did not reveal any difference in ASMase or NSMase activities between contralateral eyes from either model. In the retina, the basal ASMase activity was 54.9 ± 2.5 mU/mg protein. The basal NSMase activity in the retina was 12.31 ± 2.1 mU/mg protein, significantly (P < 0.001) lower than the basal ASMase activity (Fig. 6A and B). Ischemia/reperfusion produced a significant increase in retinal ASMase activity of 56.8% (P = 0.03). Ocular hypertensive stress produced a dramatic increase of 121.1% (P < 0.001) in retinal ASMase activity. No significant changes in NSMase retinal activity were measured following ischemia/reperfusion or ocular hypertensive stress.

Fig. 6.

(A) ASMase activities and (B) NSMase activities from contralateral untreated controls (n = 12, white bar), ischemic (n = 6, grey bars) and ocular hypertensive (n = 6, black) retinas and optic nerve heads (ONH). Data are expressed as mean ± SE. *Indicates significant difference (P < 0.05).

In the optic nerve head, the basal ASMase and NSMase activities were 95.6 ± 19.5 and 37.9 ± 8.7 mU/mg protein, respectively (Fig. 6A and B), both significantly higher than the corresponding activities measured in the retina (P = 0.04, P = 0.01). While ischemic stress did not significantly alter ASMase activity in the optic nerve, ocular hypertensive stress induced a substantial increase (P = 0.03) in optic nerve head ASMase activity of 179.2%. Neither Ischemia nor ocular hypertensive stress significantly altered optic nerve head NSMase activity.

4. Discussion

In many systems, different sphingomyelinases respond to different physiological and pathophysiological stressors. Both ASMase and NSMase activation have been considered to be early events of the response to various injuries. For example, Gu et al. (2013) reported that ceramide production in astrocytes during early cerebral ischemia was accompanied by induction of NSMase rather than ASMase activity. In the eye, extensive studies evaluating the physiological and pathophysiological roles of sphingolipids have focused on photoreceptors and the RPE. Chen et al. (2013) have shown that light damage to the retina slightly increased NSMase activity but dramatically decreased ASMase activity. Others have provided evidence that sphingolipids also play a central role in the development of inner retinal injury resulting from ischemic and ocular-hypertensive stress (Aslan et al., 2014; Fan et al., 2016, 2021; You et al., 2014). Our previous study provided evidence showing ceremide is elevated in ischemic retinas 2 h after ischemic injury (Fan et al., 2016). Several studies have highlighted the neuroprotection associated with reduced ceramide production (Argaud et al., 2004; Aslan et al., 2014; Chen et al., 2013; Fan et al., 2016). In particular, suppression of ASMase expression reduces the retinal ischemia- and ocular hypertension-induced retinal degeneration (Fan et al., 2016, 2021).

Extensive studies with animal models have shown that in several tissues, overall ceramide levels increase in response to stress (Argaud et al., 2004; Chen et al., 2013; Kubota et al., 1989; Zager et al., 1998; Zhang et al., 2001). In the retina, we have demonstrated that acute retinal ischemic stress significantly elevated ceramide levels resulting in RGC degeneration (Fan et al., 2016). A previous study by Bhattacharya’s group found significant alterations of some selected ceramide species in late-stage glaucomatous optic nerves; however, total ceramides were not significantly altered (Chauhan et al., 2019). It is important to note that ceramide is not only generated by the sphingomyelinase pathway, which requires activation of ASMase or/and NSMase, but also by the de novo and salvage pathways. Ceramide produced by these pathways is rapidly metabolized and converted to sphingosine and sphingosine1-phosphate by the actions of multiple enzymes and processes. Therefore, further investigations will be needed to understand how catabolic and metabolic enzymes influence ceramide levels over time in both ischemic and ocular hypertensive models.

In this study, we wished to investigate the potential roles of sphingomyelinase activity in the eye being subjected to retinal ischemia and ocular hypertension. First, we evaluated the expression patterns of NSMase and ASMase in the inner retina and found that they were similarly distributed diffusely in the optic nerve head (Fig. 1B). However in the retina, NSMase was predominantly in the outer retina, while ASMase was predominantly in the inner retina (Fig. 1A). In addition, basal ASMase activity was substantially higher than that of NSMase in both the optic nerve head and the retina (Fig. 6). This higher basal activity is consistent with a previous finding showing ASMase mRNA expression level in the retina being 20–50-fold higher than NSMase mRNA expression (Chen et al., 2013). Early changes of ASMase and NSMase activities following ischemic and ocular hypertensive stress were then determined. We found that the retinal ASMase activity was significantly elevated following ischemic or ocular hypertensive stress; however, in the optic nerve head, ASMase activity was only elevated by ocular hypertensive stress (Fig. 6A). NSMase activity was not significantly altered by either stressor in the retina or optic nerve head (Fig. 6B). We note that Aslan et al. (2014) found no significant changes in overall sphingomyelinase activity in the retinas from ocular hypertensive eyes after 6 weeks of elevated IOP. The difference between Aslan et al. (2014) and our results could be due to when the samples were collected; our current study collected tissues 2 weeks rather than 6 weeks after the induction of the ocular hypertensive injury. Normally, sphingomyelinase activation is an early event of a pathophysiological cascade, with the activity subsequently returning to baseline levels (Gu et al., 2013; Hernandez et al., 2000; Llacuna et al., 2006). Our results demonstrated that ASMase activity in both retinas and optic nerve heads is elevated 14 days after the induction of ocular hypertensive stress and only in the retina 90 min following ischemia. This is consistent with studies in other tissues that concluded ASMase activation is one of the early events in response to stress stimuli (Zeidan and Hannun, 2010). We have also shown that ASMase activation is involved in the general development of ocular hypertensive and ischemic injury to the retina (Fan et al., 2016, 2021).

Our previous studies have demonstrated that both retinal ischemia and ocular hypertension eventually lead to structural and functional damage to RGCs at late time points (Alsarraf et al., 2014a, 2014b; Fan et al., 2016, 2021). Early changes are more subtle. Although overall retinal morphology appeared normal 2 h after ischemic injury, some RGC damage has been detected (Palmhof et al., 2019). In the microbead injected ocular hypertension model, RGC numbers and axon density decrease were detected as early as 2 weeks after injury (Chen et al., 2011). These damages are relatively minor at the early stages after injury but become more evident over time (Chen et al., 2011; Palmhof et al., 2019; Zhu et al., 2006). We have shown that upregulation of inflammatory cytokine was detected early in both ischemia and ocular hypertension models (Fan et al., 2016; Husain et al., 2011, 2012) and wanted to identify the early sphingolipid changes preceding significant RGC degeneration. We now have identified early sphingomyelinase activity changes in retinal ischemia and ocular hypertension models that have not been previously reported. These findings coupled with our previous work that showed that decreased levels of ASMase were protective against retinal ischemic injury (Fan et al., 2016) and ocular hypertensive injury (Fan et al., 2021) suggest a potential role for ASMase in the pathogenesis of these disorders.

Abbreviations

RGC

retinal ganglion cell

ASMase

acid sphingomyelinase

NSMase

neutral sphingomyelinase

IOP

intraocular pressure

ONH

optic nerve head

pERG

pattern electroretinogram

Data availability

Data will be made available on request.