Immuno-histological detection of resistant columnar units and vulnerable networks in the rat retina after asphyxia-induced transient cardiac arrest

Abstract

Background:

Stroke-related loss of vision is one of the residual impairments, restricting the quality of life. However, studies of the ocular manifestations of asphyxia cardiac arrest/resuscitation (ACA/R) have reported very heterogeneous results.

Objective:

We aimed to evaluate the ACA/R-induced degeneration pattern of the different retinal cell populations in rats using different immuno-histological stainings.

Methods:

The staining pattern of toluidine blue and the ganglion cell markers β-III-tubulin and NeuN; the calcium-binding protein parvalbumin, indicating ganglion, amacrine, and horizontal cells; calretinin D28k, indicating ganglion and amacrine cells; calbindin, indicating horizontal cells; Chx 10, indicating cone bipolar cells; PKCα, indicating ON-type rod bipolar cells; arrestin, indicating cones; and rhodopsin, a marker of rods, as well as the glial cell markers GFAP (indicating astroglia and Müller cells) and IBA1 (indicating microglia), were evaluated after survival times of 7 and 21 days in an ACA/R rat model. Moreover, quantitative morphological analysis of the optic nerve was performed. The ACA/R specimens were compared with those from sham-operated and completely naïve rats.

Results:

ACA/R-induced effects were: (i) a significant reduction of retinal thickness after long-term survival; (ii) ganglion cell degeneration, including their fiber network in the inner plexiform layer; (iii) degeneration of amacrine and cone bipolar cells; (iv) degeneration of cone photoreceptors; (v) enhanced resistance to ACA/R by rod photoreceptors, ON-type rod bipolar and horizontal cells, possibly caused by the strong upregulation of the calcium-binding proteins calretinin, parvalbumin, and calbindin, counteracting the detrimental calcium overload; (vi) significant activation of Müller cells as further element of retinal anti-stress self-defense mechanisms; and (vii) morphological alterations of the optic nerve in form of deformed fibers.

Conclusions:

Regardless of the many defects, the surviving neuronal structures seemed to be able to maintain retinal functionality, which can be additionally improved by regenerative processes true to the “use it or lose it” dogma.

Keywords

Keywords: Amacrine cells bipolar cells calcium-binding proteins ganglion cells horizontal cells Müller cells photoreceptors

Abbreviations

ACA/R

asphyxial cardiac arrest/resuscitation

BDNF

brain-derived neurotrophic factor

bFGF

basic fibroblast growth factor

CTCF

corrected total cell fluorescence

DAPI

6-diamidino-2-phenylindole

ECG

electrocardiogram

EIOP

elevating intraocular pressure

FCS

fetal calf serum

FOV

field of vision

GABA

gamma-aminobutyric acid

ganglion cell

GDNF

glial cell line-derived neurotrophic factor

GFAP

glial fibrillary acidic protein

IBA1

ionized calcium-binding adaptor molecule 1

INL

inner nuclear layer

IPL

inner plexiform layer

IPPV

intermittent positive pressure ventilation

Kir

inward-rectifier potassium channel

linear density

MAP

mean arterial pressure

MCAO

middle cerebral artery occlusion

NDS

neurological deficient score

NGF

nerve growth factor

NT-3, 4

neutrophine 3, 4

ONL

outer nuclear layer

OPL

outer plexiform layer

PBS

phosphate-buffered saline

PFA

phosphate-buffered paraformaldehyde

ROSC

return of spontaneous circulation

toluidine blue

1 Introduction

Stroke-related loss of vision is one of the residual impairments of this condition and massively restricts the quality of life of these patients. Transient vision problems are accepted to be a harbinger of stroke (Pula & Yuen, 2017). However, the ocular manifestations of asphyxia cardiac arrest (ACA) with subsequent reoxygenation have been reported rarely and inconsistently. There are reports of a “surprising degree of retinal resistance to damage from circulation arrest” (Miller, 1975) and transient visual impairment after cardiac arrest (Priluck & Eifrig, 1978), as well as descriptions of significant functional and morphological retinal symptoms caused by cardiac arrest (Feher & Antal, 1979; Strosznajder et al., 1998). Moreover, ischemic proliferative retinopathy induced by perinatal asphyxia has been described in rats (Fernandez et al., 2020; 2013; Rey-Funes et al., 2010; 2011) and piglets (Solberg et al., 2013).

The retina requires constant blood flow for secure provision of sufficient oxygen for ATP generation (Caprara & Grimm, 2012). Thus, ischemia/reperfusion has negative consequences on retinal cells and remains a common cause of blindness. Ischemia/reperfusion is a feature of many retinal disorders, including prematurity, diabetic retinopathy, traumatic retinopathy, anterior ischemic optic neuropathies, vein occlusion, and glaucoma (Flammer, 1994; Osborne et al., 2004). Therefore, apoptosis and necrosis of the ganglion cell layer, the inner nuclear layer (INL), and the photoreceptor layer were described (Dvoriantchikova et al., 2014; Fulton et al., 2009; Huang et al., 2012; Nag & Wadhwa, 2012; Osborne et al., 2004; Produit-Zengaffinen et al., 2016). There is evidence that free radicals are important for ischemic retinal injury (Delbarre et al., 1991; Osborne et al., 2004). These radicals are massively produced during the early stage of reperfusion, generating a free radical burst that is able to overcome the endogenous cellular antioxidant defense mechanisms and induces oxidative stress, thus damaging all cellular components, including lipids, proteins, and even DNA (Gilgun-Sherki et al., 2002). This phenomenon also applies to the retina without restrictions (Kuriyama et al., 2001; Neufeld et al., 2002; Shi et al., 2018; Shibuki et al., 2000). Consequently, there is evidence of protective effects of various antioxidants during ischemia−reperfusion in the retina (Chen et al., 2014; Dilsiz et al., 2006); and more than 100 studies of concrete substances). And the role of Müller cells must also be considered here as it is a double-edged sword. On the one hand, induction of Müller cell proliferation resulting in glial scar formation with the consequence of inhibited retinal regeneration and secondary retinal cell death was observed (Reichenbach & Bringmann, 2016). On the other hand, the Müller cells are involved in the widespread neuroprotective system in the retina (reviewed in (Eastlake et al., 2020; Subirada et al., 2018); described in more detail in the discussion).

Because oxygen deficiency and reperfusion are characteristics of cardiac arrest, a corresponding degeneration pattern should be expected. However, studies of cardiac arrest/resuscitation-mediated retinopathies are rare (Feher & Antal, 1979; Miller, 1975; Priluck & Eifrig, 1978; Strosznajder et al., 1998), and sometimes only evaluated short survival intervals (Strosznajder et al., 1998).

Thus, the aim of our study was to determine the degeneration pattern of the different retinal cell populations in our well-established asphyxial cardiac arrest/resuscitation (ACA/R) model in rats (Keilhoff et al., 2017; Keilhoff, 2020; Keilhoff et al., 2011; Keilhoff et al., 2015; Keilhoff, 2020). For this, a broad spectrum of immunohistochemical staining experiments were carried out after ACA/R survival times of 7 and 21 days. Care was taken to ensure that all animals showed massive degeneration of the hippocampal CA1 region, as this is commonly considered to be the gold standard of a cardiac arrest model.

2 Material and methods

2.1 ACA/R rat model

Ethical approval for this study was granted according to the requirements of the German Animal Welfare Act on the Use of Experimental Animals and the Animal Care and Use Committees of Saxony-Anhalt (permit number 42502-2-2-947 Uni MD). The experiments were carried out using male rats (aged between 10 and 12 weeks and weighing 300−400 g, from the breeding population of inbred Wistar rats of our Institute; Harlan-Winkelmann, Borchen, Germany). Animals were housed under controlled laboratory conditions (light cycle of 12 h light/12 h dark, lights on at 6:00 a.m.; temperature, 20°C±2°C; air humidity, 55%–60%) with anytime free access to water and chow. Every effort was made to minimize the amount of suffering and the number of animals used in the experiments.

At the beginning, the study included 3 completely naïve rats, 7 sham-operated rats with a survival time of 7 days, and two groups of 10 ACA/R-treated animals each with survival times of 7 and 21 days, respectively, for a total of 30 animals. Subsequently, 3 animals were excluded from the study (see below).

The ACA/R method has been described in detail previously (Keilhoff et al., 2015). Briefly, the ACA was achieved as follows: (i) anesthesia: 5%sevoflurane (Pfizer GmbH, Berlin, Germany) in an oxygen/nitrous oxide mixture (40:60) via a face mask; (ii) endotracheal intubation with a modified laryngoscope and a 14 G venous catheter; (iii) muscular relaxation with vecuronium (1 mg/kg; Pfizer); (iv) intermittent positive pressure ventilation (IPPV); (v) cannulation of both left femoral vessels with polyethylene catheters for drug administration, blood sampling, and continuous blood pressure monitoring; (vi) complete washout of anesthetic gases by room air ventilation (5 min) and baseline control; and (vii) induction of ACA by end-expiratory interruption of IPPV on paralyzed rats for 6 min. ACA (defined as a non-pulsatile blood pressure of less than 10 mmHg) was reached within approximately 3 min. Subsequently, resuscitation was achieved as follows: (i) administration of epinephrine (i.v.; 1μg/kg; Pfizer) and sodium bicarbonate (1 mEq/kg); (ii) reintroduction of IPPV with 100%oxygen for 1 h; and (iii) manual external chest compression (200 compressions/min). Finally, return of spontaneous circulation (ROSC) was achieved as a pulsatile mean arterial pressure (MAP) above 60 mmHg. Rats with no ROSC within 2 min were excluded from the study (n = 3). The 2-min interval is the laboratory standard to avoid non-homogeneous ACA periods leading to pathophysiological differences and, subsequently, to the need for a larger number of animals. During the first 45 min of the post-resuscitation intensive-care phase, ECG, blood pressure, temperature, and airway pressure were continuously monitored. Moreover, arterial blood samples were collected 5, 15, 30, and 45 min after ROSC for the evaluation of blood gases (pCO₂ and pO₂), pH, and glucose. After establishing of sufficient spontaneous respiration, catheters were removed, vessels were ligated, and incisions were surgical closed. To avoid the neuroprotective effects of hypothermia, we used a heating mat to maintain the body temperature of the animals at 37°C±0.3°C during preparation, insult, and the first 45 min post-ROSC. For this, a heated mat temperature controller TC-813 SN 97106 (CEW, Inc.) was coupled to a tympanic temperature control, thus affording an optimized temperature regime. Moreover, animals were kept in an incubator cage at 34°C for an additional 24 h after the intervention. After successful resuscitation within the 2-min interval, we experienced no further failures because of the use of intensive post-operative care. In sham animals, we carried out the same preparation. Since no ACA/R was induced, anesthesia and paralyze had to be maintained for an additional 45 min.

2.2 Assessment of histology

After survival times of 7 and 21 days, deeply anesthetized rats were sacrificed by perfusion through the left ventricle with PBS (pH 7.4, 3 min) and 300 ml of PBS-buffered paraformaldehyde (PFA, 4%, 0.1 M, pH 7.4; Millipore, Darmstadt, Germany; 30 ml/min). The brains and eyes were quickly removed and post-fixed in the same fixative at 4°C until further processing.

For retinal immunohistochemistry, the whole eyes were cryoprotected in 30%sucrose in PFA (0.4%, pH 7.4) for 1 day and embedded in cryomatrix (Thermo Scientific, Wilmington, USA). Subsequently, the optic nerve was held with forceps and the eyes were aligned with a slight inclination (5°), with the lens facing downward. Then, the samples were rapidly frozen at −20°C and sliced on a cryostat (Jung Frigocut 2800 E, Leica, Bensheim, Germany; coronal cross-sections, between 1300 and 2400μm below the optic nerve head, 20μm thickness, Fig. 1A). To obtain a minimum of 5 slices per animal for each staining, both eyes were processed. Fist, unspecific binding sites were blocked with 10%FCS/0.3%Triton-X-100 (both Sigma-Aldrich, Taufkirchen, Germany) in PBS for 1 h. Then, the slices were incubated with the primary antibodies (alone or in combination; listed in Table 1) at 7°C overnight. It followed the incubation with the secondary antibodies for 3 h (Table 1). Finally, nuclear counterstaining with 4^′,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich, D9542; 15 min, 37 °C) completed the procedure. For each staining experiment, controls of antibody specificity involved the replacement of the primary antibody with buffer or normal serum, thus completely lacking immunoreactivity. Moreover, 5 retinal sections per animal were stained with 0.1%toluidine blue (TB).

Fig. 1

Schematic illustration of the eye and optic nerve preparation procedure. (A) Coronal sections were prepared from 1300 to 2400μm below the optic nerve head. (B) Per each individually stained coronal slice of retina, 4 photomicrograph (447×336μm, red frame) were taken, serving as basis for creating figures 5 –10 and for quantification. Note that it is a not-to-scale illustration. (C) Optic nerve cross-section for the demonstration of the relative position of 24 equal-sized boxes (400μm²) for semi-quantitative analysis of the optic nerves.

Table 1

Used primary and secondary antibodies

Antibody		Dilution	Catalog Nr.	References/wavelength	Company
Primary antibodies
anti-β-III-tubulin	monoclonal mouse	1:1000	ab7751	ganglion cell fiber network (Intartaglia et al., 2020); pigment epithelial cell (Vinores et al., 1995)	Abcam, Cambridge, UK
anti-β-III-tubulin	polyclonal rabbit	1:1000	ab15568
anti-parvalbumin	polyclonal rabbit	1:1000	ab11427	ganglion, amacrine cells (S. A. Kim et al., 2010); (Olivares et al., 2017); horizontal cells (Oguni et al., 1998)
anti-calretinin D28k	polyclonal rabbit	1:100	ab702	ganglion, amacrine cells (S. A. Kim et al., 2010; Sawant et al., 2019)
anti-synaptophysin	polyclonal rabbit	1:500	ab32594	Synaptic vesicle (Brandstatter et al., 1996)
anti-NeuN	monoclonal mouse	1:100	MAB377	ganglion cells (Boretsky et al., 2015)	Millipore, Darmstadt, Germany
anti-arrestin	polyclonal rabbit	1:5000	AB15282	Cones (L. Zhao et al., 2015)
anti-rhodopsin	monoclonal mouse	1:500	MABN15	rods (Bourque et al., 2013)
anti-calbindin	polyclonal rabbit	1:500	AB1778	horizontal cells (S. A. Kim et al., 2010)
anti-vimentin	monoclonal mouse	1:500	MAB 3400	Müller cells (Ishikawa et al., 2014)
anti-Chx 10	monoclonal mouse	1:100	sc-374151	(cone) bipolar cells (Kaufman et al., 2019; Nemitz et al., 2019)	Santa Cruz Biotechnology Inc., Heidelberg, Germany
anti-PKCα	monoclonal mouse	1:100	sc-8393	ON-type rod bipolar cells (Kuny et al., 2010)
anti-IBA1	polyclonal rabbit	1:1000	019-19741	general microglial marker	Wako Chemicals GmbH Neuss, Germany
anti-MAP2	monoclonal mouse	1:1000	SMI-52R	general neuronal marker	Covance; Münster, Germany
anti-GFAP	polyclonal rabbit	1:500	10555	general astroglial marker; Müller cells (Bringmann et al., 2006)	Progen, Heidelberg, Germany
Secondary antibodies
Alexa 488	goat anti-mouse	1:500	A11029	Excitation max.: 490 nm	Invitrogen, Carlsbad, USA
				Emission max: 525 nm
Alexa 546	goat anti-rabbit	1:500	A11010	Excitation max.: 556 nm
				Emission max: 573 nm

All antibodies were diluted in 10 %BSA/0.3 %Triton in PBS.

To evaluate the optic nerve, the two nerves of each animal were fixed in 0.2 M cacodylate buffer containing 2.5%glutaraldehyde, dehydrated, en-bloc stained with 7%uranyl acetate, embedded in Durcupan (ACM, Sigma-Aldrich), and sectioned at semi-thin thickness (500 nm) at a distance of about 4 mm posterior to the emanation from the globe. Sections were stained with 0.1%TB.

2.3 Evaluation and statistical analysis

To evaluate Immu-Mount (Thermo Scientific)-embedded slices (5 alternating slices/staining/animal), a combined light/fluorescence microscope (AxioImager M1, Zeiss, Jena, Germany), equipped with Plan-Neofluar 10×/0.3, 20×/0.50, and 40×/0.75 objectives, and a fluorescein/rhodamine/DAPI filter set and the photo-camera Axiocam 503 color (pixel count: 2.83 Megapixel, 1.936 (H)×1.460 (V); pixel dimension: 4.54μm×4.54μm; FPS @ 1 ms: 38; digitization: 14 bit/pixel) were used. Per each individually stained retinal slice, 4 photomicrograph (447×336μm, red frame in Fig. 1B) were taken. From each photomicrograph, a cutout of either 160×90μm or 120×90μm (field of vision, FOV) was used for figures and quantification. The thickness of the complete retina and of the photoreceptor segment, in both cases without the retinal pigment epithelium, was assessed using the measuring tool of the AxioVision software (SE64, Rel. 4.9; Zeiss). Therefore, 3 measurements were performed per photomicrograph and then averaged. The staining intensities of β-III-tubulin, synaptophysin, and parvalbumin in the photoreceptor layer, as well as GFAP, were semi-quantified as corrected total cell fluorescence (CTCF), which was calculated as: [integrated density –(selected area×mean fluorescence of the background)]×10⁻⁵ in arbitrary units according to the user manual of the ImageJ software (http:/rsb.web.nih.gov/ij/). Thereby, microscope setting (brightness, contrast, ∨_∘) and the exposure times of the fluorescence channels were set based on control slices and kept constant for each antibody. If necessary, background correction was applied by subtracting a background value, respectively. The linear density (LD) of TB-, NeuN-, and calretinin-stained ganglion cells; arrestin-positive cone photoreceptors, and calbindin-positive horizontal cells was calculated and the layer length was measured using the ImageJ software. The total number of clearly immunopositive cells divided by the layer length represented the respective LD (cells/mm). The number of calretinin-positive amacrine cells, Chx 10-positive cone bipolar cells, and PKCα-positive ON-type rod bipolar cells, as well as IBA1-positive microglia cells throughout the retina, was counted. For each staining experiment, a mean value per animal was calculated and used as one value for the statistical analyses. It consisted of 20 individual values: the values from the 4 photomicrographs of each of the 5 retinal sections per staining.

The density and total number of optic nerve axons were determined in five cross-sections from each nerve. Bright-field micrographs were acquired using an oil-immersion objective (100×/1.3), and images of the entire nerve were montaged using the AxioVision software (“Panorama” Zeiss). Twenty-four boxes of 400μm² each were placed as shown in Fig. 1C, and all axons that appeared to be morphologically vital were counted manually using the ImageJ software. To obtain the total number of optic nerve axons, the average axon density was multiplied by the nerve area, which was determined as Ao = minor radius×major radius×π. A mean count per animal was calculated, respectively, and used as one value for the statistical analyses.

The data in Tables 2 3 are presented as the mean±SD per animal. The histological data in Fig. 4 are presented as the median with min-to-max whiskers. Both data sets were analyzed using Graph Pad Prism 6 (GraphPad Software Inc., La Jolla, CA, USA). For histological data, the non-parametric Kruskal−Wallis test and Dunn’s multiple comparison post hoc test were applied. Student’s t-test was chosen to compare the vital parameters of ACA/R and sham animals at each time point. Significance was set at p≤0.05.

Table 2
Quantification of pre- and post-resuscitation physiological parameters

Parameter Post-ROSC time

BL 5 min 15 min 30 min 45 min

Sham animals (n = 7)

Tympanic temperature (°C) 36.5±0.4 36.9±0.4 36.9±0.2 36.5±0.2 36.9±0.3

Rectal temperature (°C) 36.7±0.5 36.8±0.7 36.9±0.4 36.9±0.3 37.1±0.3

MAP (mmHg) 89±12 94±09 91±10 93±09 98±09

Heart rate (min^–1) 328±43 357±29 360±29 371±34 368±43

pCO₂ (mmHg) 40±06 40±04 41±03 41±01 40±01

Glucose (mmol/l) 11.4±3.7 12.1±4.2 12.6±3.2 11.5±3.8 11.9±2.3

pH 7.41±0.04 7.40±0.02 7.39±0.02 7.39±0.01 7.40±0.02

pO₂ (mmHg) 172±30 170±26 174±24 177±14 178±16

ACA/R animals (n = 17)

Tympanic temperature (°C) 36.7±0.6 36.8±0.6 36.8±0.5 37.1±0.4 37.0±0.4

Rectal temperature (°C) 36.4±0.3 36.6±0.5 36.6±0.5 36.8±0.3 36.7±0.6

MAP (mmHg) 77±17 169±14^** 163±21 89±20 84±18

Heart rate (min^–1) 312±19 424±30^* 424±33 324±32^ 301±27^*

pCO₂ (mmHg) 44±04 49±08 49±14 41±16 38±11

Glucose (mmol/l) 12.1±2.3 19.8±2.5^* 17.3±3.1 10.9±2.7 11.0±2.3

pH 7.38±0.03 7.25±0.08^**** 7.31±0.10 7.35±0.09 7.39±0.07

pO₂ (mmHg) 170±37 251±87 251±86 397±97^* 414±104^**

Parameter		Post-ROSC time
Sham animals (n = 7)
Tympanic temperature (°C)	36.5±0.4	36.9±0.4	36.9±0.2	36.5±0.2	36.9±0.3
Rectal temperature (°C)	36.7±0.5	36.8±0.7	36.9±0.4	36.9±0.3	37.1±0.3
MAP (mmHg)	89±12	94±09	91±10	93±09	98±09
Heart rate (min^–1)	328±43	357±29	360±29	371±34	368±43
pCO₂ (mmHg)	40±06	40±04	41±03	41±01	40±01
Glucose (mmol/l)	11.4±3.7	12.1±4.2	12.6±3.2	11.5±3.8	11.9±2.3
pH	7.41±0.04	7.40±0.02	7.39±0.02	7.39±0.01	7.40±0.02
pO₂ (mmHg)	172±30	170±26	174±24	177±14	178±16
ACA/R animals (n = 17)
Tympanic temperature (°C)	36.7±0.6	36.8±0.6	36.8±0.5	37.1±0.4	37.0±0.4
Rectal temperature (°C)	36.4±0.3	36.6±0.5	36.6±0.5	36.8±0.3	36.7±0.6
MAP (mmHg)	77±17	169±14^****	163±21	89±20	84±18
Heart rate (min^–1)	312±19	424±30^***	424±33	324±32^**	301±27^***
pCO₂ (mmHg)	44±04	49±08	49±14	41±16	38±11
Glucose (mmol/l)	12.1±2.3	19.8±2.5^*	17.3±3.1	10.9±2.7	11.0±2.3
pH	7.38±0.03	7.25±0.08^****	7.31±0.10	7.35±0.09	7.39±0.07
pO₂ (mmHg)	170±37	251±87	251±86	397±97^*	414±104^**

BL, base line; MAP, mean arterial pressure; pCO₂, arterial carbon dioxide tension; pO₂, arterial oxygen tension, ACA/R, asphyxia cardiac arrest/resuscitation; Data: mean±SD; Student’s t-test compared the respective parameters of ACA/R and sham animals at each time point, ^****p≤0.0001, ^***p≤0.001, ^**p≤0.005, ^*p≤0.05.

Table 3

Quantification of morphometric data

	Naïve (n = 3)	p; naïve vs. ACA/R 7d	p; naïve vs. ACA/R 21d	Sham (n = 7)	p; sham vs. ACA/R 7d	p; sham vs. ACA/R 21d	ACA/R 7d (n = 9)	ACA/R 21d (n = 8)
Retina
LD of TB-stained GC	132.7±9.7	^{-- -- -- -}	< 0.05	127.1±9.5	< 0.05	< 0.05	83.6±10.3	72.4±14.7
p; LD TB⁺-GC vs. calretinin⁺-GC	< 0.05			< 0.0001			< 0.001	< 0.001
LD of NeuN-stained GC	93.4±11.5	^{-- -- -- -}	< 0.05	90.7±10.9	< 0.05	< 0.05	66.6±18.2	61.9±17.5
p; LD NeuN⁺-GC vs. calretinin⁺-GC	< 0.05			< 0.05			< 0.01	< 0.01
LD of Calretinin-stained GC	45.5±10.9	^{-- -- -- -}	< 0.05	42.2±7.0	< 0.05	< 0.05	27.9±11.8	26.3±9.2
LD of Arrestin-stained cones	146.9±17.5	< 0.001	< 0.001	140.6±16.6	< 0.01	< 0.01	78.5±16.9	79.5±11.9
LD of Calbindin-stained horizontal cells	21.5±10.5	^{-- -- -- -}	^{-- -- -- -}	22.4±7.3	^{-- -- -- -}	^{-- -- -- -}	20.1±6.7	19.4±7.7
Number of Calretinin-stained amacrine cells (per FOV: 160×90μm)	11.2±5.6	< 0.05	< 0.05	10.5±4.2	< 0.01	< 0.05	4.9±2.2	5.3±2.4
Number of PKCα-stained rod bipolar cells (per FOV: 120×90μm)	37.4±7.8	^{-- -- -- -}	^{-- -- -- -}	36.8±5.4	^{-- -- -- -}	^{-- -- -- -}	31.6±6.5	34.8±5.2
Number of Chx 10-stained cone bipolar cells (per FOV: 120×90μm)	30.3±5.5	^{-- -- -- -}	^{-- -- -- -}	33.3±4.7	< 0.01	< 0.01	22.6±4.5	21.7±5.0
Optic nerve
Cross sectional area (μm²;×10⁴)	(20.7±1.5)	^{-- -- -- -}	^{-- -- -- -}	(21.4±1.8)	^{-- -- -- -}	^{-- -- -- -}	(19.8±1.2)	(19.5±1.6)
Axonal density/μm²	0.4±0.0	^{-- -- -- -}	^{-- -- -- -}	0.4±0.1	^{-- -- -- -}	^{-- -- -- -}	0.4±0.1	0.4±0.1
Total axonal number (×10³)	(79.0±5.8)	^{-- -- -- -}	^{-- -- -- -}	(87.1±6.5)	^{-- -- -- -}	^{-- -- -- -}	(76.7±1.4)	(73.2±1.3)
Total number of malformed fibers	72.3±21.5	< 0.05	< 0.05	68.5±16.2	< 0.01	< 0.01	697.4±163.8	651.9±109.8

Linear density in n/mm; all data are mean±SD; FOV, field of vision; GC, ganglion cell; LD, linear density; TB, toluidine blue; d, day; n, animal number.

Fig. 4

Semi-quantification of ACA/R-mediated degenerative patterns in the rat retina. Boxplots show the: (A) significant reduction of retinal thickness after the longer survival time; (B) tendency toward a reduction of the photoreceptor segment after the longer survival time; (C) significant reduction of β-III-tubulin immunofluorescence intensity, indicating degeneration of the ganglion cell network; (D) significant reduction of synaptophysin immunofluorescence intensity, indicating degeneration of the cone photoreceptors and a general reduction of synaptic terminals; (E) significant reduction of parvalbumin-expressing (bipolar and horizontal) cells in the INL; (F) significant induction of parvalbumin immunofluorescence intensity in the inner and outer segment of the photoreceptors; (G) significant induction of GFAP immunofluorescence intensity as an expression of Müller cell activation; and (H) significant increase of IBA1-expressing microglial cells at both survival times, indicating ongoing neurodegenerative processes. Data are the median with min-to-max whiskers. We analyzed 3 naïve, 7 sham-operated, 9 ACA/R (7d), and 8 ACA/R (21d) rats using the non-parametric Kruskal–Wallis test. Dunn’s multiple comparison post hoc test revealed significant differences, with ^*P < 0.05, ^**P < 0.005, and ^***P < 0.001. CTCF, corrected total cell fluorescence.

3 Results

3.1 Effect of ACA/R on vital parameters

The preparation of animals (anesthesia via facemask, trachea intubation, cannulation of the left femoral vessels, and determination of the baseline vital parameters) required 23±5 min. This pre-arrest surgical handling vitiated the vital parameters only marginally. In ACA/R animals, the arterial blood pressure dropped to the null line after 170±29 s of asphyxiation. After 6 min of asphyxiation, resuscitation was initiated. Therefore, ROSC was achieved within 30±24 s.

The 45-min post-ROSC time course of the basic vital parameters is displayed in Table 2. The tympanal and rectal body temperatures and arterial carbon dioxide tension (pCO₂) were unaffected. Significant, though temporary, differences in MAP (increase), heart rate (increase/decrease), blood glucose values (increase), and pH level (decrease) were evident. Because of post-ROSC ventilation with 100%oxygen, the arterial pO₂ of ACA/R animals increased during the 45-min measurement.

3.2 Effect of ACA/R on the histology of the hippocampal CA1 region

The hippocampal CA1 region is recognized as the gold standard for assessing ACA/R-mediated neurodegenerative processes in the brain, as this formation is highly sensitive to ischemia/reperfusion injuries. Figure 2 illustrates the pattern of hippocampal CA1 pyramidal cell degeneration that occurred in our animal model 7 days after ACA/R injury. The comparison of sham-operated (A) and ACA/R-treated (B) animals regarding the neuronal marker MAP2 revealed loss of CA1 pyramidal cells, including their neuronal network. This neurodegenerative process was accompanied by microglial activation (IBA1 positivity).

Fig. 2

The hippocampal CA1 region as gold standard of the rat ACA/R model. (A) sham-operated and (B) an ACA/R animal 7 days after intervention, as assessed using co-immunostaining of the neuronal network marker MAP2 and the microglial marker IBA1 (red). Microglial activation indicates a massive degeneration of CA1 pyramidal cells and their neuronal dendritic network.

3.3 Effect of ACA/R on retinal histology

The measurement of retinal cross-section thickness showed a significant ACA/R-mediated reduction in this parameter after the long-term survival time of 21 days exclusively (Figs. 3A−D 4A). The evaluation of β-III-tubulin immunostaining revealed degeneration of ganglion cells, especially their fiber network in the inner plexiform layer (IPL; Figs. 4C , and 9). The ACA/R-induced degeneration of ganglion cells in the IPL was confirmed by TB staining (Table 3; Fig. 3A−H) and NeuN (Table 3; Fig. 5I−L) and calretinin (Table 3; Figs. 6A−D , and 8A−D) immunostaining. At this point, it was interesting to compare the staining-specific LDs (Table 3), which revealed that calretinin immunostaining was significantly reduced, representing about 33%of the TB-stained ganglion cells regardless of treatment. However, the ACA/R-induced decrease in ganglion cells was identical (approximately 37%at day 7 post-ACA/R and 45%at day 21 post-ACA/R), regardless of the cell (staining) subtype. The anti-NeuN antibody stained less (not significant) ganglion cells compared with TB, but stained significantly more of these cells compared with calretinin. Moreover, the ACA/R-induced decrease in NeuN-positive ganglion cells was less pronounced (approximately 28%at day 7 post-ACA/R and 34%at day 21 post-ACA/R). In addition, calretinin staining indicated an ACA/R-induced degeneration of amacrine cells in the INL [Figs. 6(arrowheads), and 8A−D , respectively; Table 3] and changes in the sub-laminas in the IPL (white asterisk in Fig. 6C, D; C’, D’), representing axonal projections of ON and OFF bipolar cells and dendritic stratifications of amacrine cells. Parvalbumin immunostaining (Figs. 4E , and 8E−H, respectively) confirmed the degeneration pattern of the ganglion and amacrine cells, but also provided evidence of bipolar and horizontal cell damage (Figs. 6 , H). Chx 10 immunostaining (Fig. 7, Table 3) confirmed the degeneration of bipolar cells, especially that of cone bipolar cells. In contrast, the ON-type rod bipolar cell-specific PKCα immunostaining (Fig. 8, Table 3) indicated the ACA/R resistance of these cells as well as their synaptic trees in the inner half of the IPL. Moreover, PKCα is mainly expressed close to the cell membrane. Calbindin immunostaining (Figs. 7 8I−L) seemed to confirm the ACA/R-mediated damage of horizontal cells. However, semi-quantification did not reveal significant differences (Table 3). Furthermore, staining of calretinin (Fig. 6A-D; A’-D’) and parvalbumin (Fig. 6E-H, E’-H’) indicated a massive activation of the inner (empty arrowheads in Fig. 6C, D, G, and H) as well as outer (black asterisk in Fig. 6C, D, G, and H) photoreceptor segment. In turn, arrestin immunostaining (Fig. 5A−H; Table 3) demonstrated an ACA/R-induced loosening of the cones and their synaptic terminals (cone pedicles) in the outer plexiform layer (OPL; Fig. 7M−L). In contrast, rhodopsin immunostaining did not reveal such obvious evidence of an ACA/R effect on rod photoreceptor cells (Fig. 6A−D and I−L); we observed only a tendency toward a reduced thickness of the photoreceptor segment after the long-term survival time of 21 days (Fig. 4B).

Fig. 3

Micrographs of Toluidine-blue-stained retina and optic nerve sections. (A−H) Toluidine-blue-staining demonstrates an ACA/R-mediated loss of ganglion cells (zoomed in E−H), as well as a loosening of the INL and ONL. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; IS, inner segment; OS, outer segment. (E−H) Toluidine-blue-stained semi-thin cross-sections of the optic nerve. Note the degenerated fibers in ACA/R specimens (arrows in K and L). The scale bar is valid for all images of the respective row.

Fig. 5

Fluorescence micrographs of the retina demonstrating ACA/R-induced loss of ganglion cells and cones including their synaptic connections. (A, B vs. C, D) β-III-tubulin immunostaining indicated the damage of ganglion cells, especially their fiber network of the IPL. (A, B vs. C, D; A’, B’ vs. C’, D’; E, F vs. G, H) Arrestin immunostaining (red) showed the ACA/R-induced loosening of the cones and the respective synaptic terminals in the OPL. Note that A’-D’ are not exact details of A-D. They demonstrated the arrestin staining using single-channel analysis. (E-H) Rhodopsin staining confirmed the relative resistance of rods toward ACA/R. (I, J vs. K, L) NeuN staining confirmed the ACA/R-induced ganglion cell loss. (M, N vs. O, P) Synaptophysin immunostaining (red) confirmed the loss of synaptic connections between ganglion, bipolar, and amacrine cells in the IPL, and between horizontal cell dendrites and rod and cone cell bodies in the OPL. Nuclei are stained with DAPI (blue). The scale bar is valid for all images of the respective row.

Fig. 6

Fluorescence micrographs of the retina demonstrating the ACA/R effect on calcium binding proteins related to Rhodopsin-positive photoreceptors. (A, B, A’, B’ vs. C, D, C’, D’) Calretinin D28k immunostaining (red) illustrating the loss of ganglion and amacrine cells (arrowheads), as well as the respective sub-laminas of the IPL (white asterisk). Moreover, staining revealed massive activation of the calcium-binding capacity in the inner (empty arrowheads) and outer (black asterisk) segments of the rhodopsin-immunostained photoreceptor segment. Note that A’-D’ are not exact details of A-D. They demonstrated the calretinin staining using single-channel analysis. (E, F vs. G, H) Parvalbumin immunostaining (red) indicating damage of the ganglion cells, as well as amacrine, bipolar, and horizontal cells in the INL (E’−H’). Note that E’-H’ are not exact details of E-H. They demonstrated the parvalbumin staining using single-channel analysis. The staining confirmed the massive increase in the calcium-binding capacity of the inner (empty arrowheads) and outer (asterisk) segments of the β-III-tubulin-immunostained photoreceptor segment. (I−L) Rhodopsin immunostaining showing a tendency toward a reduction of the photoreceptor segment. Nuclei are stained with DAPI (blue). The scale bar is valid for all images of the respective row.

Fig. 9

Fluorescence micrographs of the retina demonstrating ACA/R-mediated changes in astroglia and Müller cells. (A–D) Vimentin immunostaining (Niwa et al.) did not differ among naïve, sham, and ACA/R animals. (E, F, I, J vs. G, H, K, L) In contrast, GFAP immunostaining (red) was massively intensified in the entire retina, indicating a strong ACA/R-induced activation of Müller cells. Nuclei are stained with DAPI (blue). The scale bar in L is valid for all images.

Fig.7

Fluorescence micrographs of the retina demonstrating the ACA/R effect on calcium binding proteins related to Chx 10-positive bipolar cells. (A, B vs. C, D) The co-immunostaining of calretinin D28k (red) and Chx 10 validated the loss of ganglion and amacrine cells, including the sub-laminas of the IPL. Moreover, Chx 10 staining revealed a loss of (cone) bipolar cells in the INL. (E, F vs. G, H) Parvalbumin immunostaining (red) confirmed the damage of ganglion cells, as well as the neuronal populations of the INL, whereas Chx 10 staining revealed the loss of (cone) bipolar cells. (I−L) The co-immunostaining of Chx 10 and calbindin (red) demonstrated the embedding in the bipolar cell population and horizontal cells in the INL and their synaptic connections in the OPL. (M−N) Co-immunostaining of Chx 10 and arrestin (red) showing a band of cone pedicles in the OPL, which was clearly thinned out as a result of the ACA/R-mediated loss of cones. Nuclei are stained with DAPI (blue). The scale bar is valid for all images of the respective row.

Fig. 8

Fluorescence micrographs of the retina demonstrating the ACA/R effect on calcium binding proteins related to PKCα-positive ON-type rod bipolar cells. (A, B vs. C, D) The co-immunostaining of calretinin D28k (red) and PKCα validated the loss of ganglion and amacrine cells, including the sub-laminas of the IPL. Moreover, PKCα staining revealed no ACA/R-related changes in the ON-type rod bipolar cells of the INL or their synaptic trees in the inner half of the IPL. (E, F vs. G, H) Parvalbumin immunostaining (red) confirmed the damage of ganglion cells and of the neuronal populations of the INL, whereas PKCα staining underscored the stability of ON-type rod bipolar cells. (I−L) The co-immunostaining of PKCα and calbindin (red) demonstrated the embedding in the bipolar cell population and horizontal cells in the INL, as well as their synaptic connections in the inner half of the IPL. Note the expression of the PKCα immunosignal close to cell membrane. Nuclei are stained with DAPI (blue). The scale bar is valid for all pictures.

Immunostaining of synaptophysin, which is the most frequent synaptic vesicle membrane protein, was detected in the IPL and the OPL (Fig. 5M−P). Consistent with the ACA/R-mediated damaged fiber network observed in the IPL, synaptophysin expression was reduced in these animals (Fig. 5O and P). The OPL represents a dense network of synaptic connections between dendrites from bipolar cells (from the INL), horizontal cell axons, and dendrites (from the OPL), as well as the rod and cone cell bodies in the ONL. As we found degenerative signs in most of these cell types, a reduced synaptic density in the OPL was to be expected. The semi-quantification of the overall synaptophysin fluorescence intensity is given in Fig. 4D.

Immunostaining for vimentin, which is constitutively expressed in and, thus, specific for Müller cells (Fig. 9A−D), did not offer clear evidence of Müller cell activation. However, GFAP staining (Figs. 4G 9E−L) indicated a dramatic ACA/R-induced activation of Müller cells. Of note, GFAP staining of activated, hypertrophic Müller cells runs through the entire retina.

In accordance with the degenerative changes described above, we found an increased number of IBA1-positive activated microglial cells, especially in the IPL and OPL (Figs. 4H and Fig. 10). Microglial activation was recognizable also after the long-term survival time of 21 days, albeit at a lower level (Fig. 10 D, H), indicating an ongoing neurodegenerative process. The microglial cells were characterized by compact hypertrophic somata with few processes.

The semi-quantitative exploration of the optic nerve did not reveal any significant ACA/R-mediated effects in terms of axonal density, total axon number, and cross-sectional area (Table 3; Fig. 3E−H). Optic nerve fibers were organized in bundles among septa. However, the number of deformed fibers was significantly higher because of ACA/R (Table 3; arrows in Fig. 3K and L).

Fig. 10

Fluorescence micrographs of the retina demonstrating ACA/R-mediated changes in microglial cells. (A, B, E, F vs. C, D, G, H) The number of IBA1-positive microglial cells (red) was increased, especially in the IPL and OPL. Nuclei are stained with DAPI (blue). The scale bar in H is valid for all images.

4 Discussion

Studies of ischemia-related changes in the retina are numerous. However, most of these studies targeted the retina directly by (i) high intraocular pressure, (ii) ligature of the optic nerve bundle, (iii) ligature of ophthalmic vessels, (iv) photodynamic ablation, or (v) intravitreal injection of glutamate receptor agonists. Moreover, models of occlusion of the common carotid artery (2-vessel-occlusion) or bilateral occlusion of the vertebral and common carotid arteries (4-vessel-occlusion) were described (reviewed by D’Onofrio & Koeberle, 2013; Minhas et al., 2012; Niwa et al., 2016; Osborne et al., 2004). In contrast, studies using the model of cardiac arrest/resuscitation are rare.

Here, we reported the neurodegenerative pattern of the rat retina at 7 and 21 days after ACA/R. We found a significant ACA/R-mediated reduction of retinal thickness only after the long-term survival time. Immunostaining with various antibodies revealed degeneration of the ganglion cells, including their fiber network in the IPL, degeneration of the amacrine and cone bipolar cells of the INL, as well as damage of cone photoreceptor cells. However, rod photoreceptors, their ON-type bipolar cells, and horizontal cells seemed to be more resistant to ACA/R. Moreover, a significant activation of Müller and microglial cells was evident. The optic nerve showed a significant ACA/R-mediated increase of deformed fibers.

Our results confirm the hypothesis of a high but different sensitivity of the different retinal cells to ischemic insults and extend the hypothesis to global ischemia. The loss of ganglion cells has been described repeatedly and independently of the model used [elevating intraocular pressure (EIOP): (Kwon et al., 2005; Lee et al., 2011; Palmhof et al., 2019); middle cerebral artery occlusion (MCAO): (Lee et al., 2011); reviews: (Khalilpour et al., 2017; Sucher et al., 1997)]. The particularly high sensitivity of amacrine cells, cone bipolar cells, and cone photoreceptors has been published, although this was reported in a model of EIOP (Dijk, 2004; Palmhof et al., 2019; Schmid et al., 2014). Cell loss leads to a reduction in retinal thickness, albeit only after longer survival times. This phenomenon was also observed in the model of EIOP (Kim et al., 2013; Palmhof et al., 2019).

Moreover, the resistance of horizontal cells (Chun et al., 1999; Kim et al., 2010), photoreceptor rods (Palmhof et al., 2019), and ON-type rod bipolar cells (Dijk, 2004; Schmid et al., 2014) to ischemia−reperfusion has been demonstrated, although always using a model of EIOP. Thus, this functional unit (i.e., horizontal cells) is an essential guidepost for the interaction of rod photoreceptors with rod bipolar cells (Nemitz et al., 2019) and appears to be particularly resistant to ischemic insults. Ischemia−reperfusion induces an over-excitation of ionotropic glutamate receptors, leading to cell depolarization with excessive Ca^2 + influx into the cells and subsequent activation of cell death mechanisms. Therefore, the resistance of horizontal cells, ON-type bipolar cells, and rods could be attributed to their robust calcium-binding capacity, which counteracts the ischemia-induced detrimental calcium overload. Here, we detected a strong upregulation of calretinin and parvalbumin in the rods. Kim and co-workers (Kim et al., 2010) reported high levels of expression of the calbindin transcript and protein in ischemia−reperfusion-resistant horizontal cells In an EIOP model. Moreover, the functionality of rod ON-type bipolar cells, including their deactivation at the end of the glutamate-driven signal transduction cascade, depends on solid expression of PKCα (Ruether et al., 2010). We (and others) detected PKCα immunosignals close to the cell membrane, indicating PKC activation (Weinreb et al., 2004).

In contrast to this resistant cell network of rods consisting of ON-type bipolar and horizontal cells, we identified a group of significantly more sensitive ganglion, amacrine, and cone bipolar cells, as well as cones. The calcium-buffering capacity of these cells seems to be less effective. In the EIOP model, it was shown that calretinin and parvalbumin expression at both the transcript and protein levels was decreased after ischemia in amacrine and ganglion cells (Dijk & Kamphuis, 2004b; Dijk, 2004; Dijk, 2004; Kwon et al., 2005; Lee et al., 2011).

Ganglion cells, bipolar cells, and photoreceptors are glutamatergic cells. Horizontal and amacrine cells are GABAergic cells (reviewed in (Connaughton, 1995), updated 2007). The latter are also glycinergic, similar to bipolar cells (Menger et al., 1998). Horizontal cells spread the information horizontally, thus generating feedback (presynaptic) inhibition to photoreceptors and feedforward (postsynaptic) inhibition to bipolar cells. Amacrine cells afford reciprocal feedback inhibition of bipolar cells and feedforward inhibition of ganglion cells [for more detail see (Diamond, 2017; Kolb, 1995b)]. To a certain degree, ischemia-induced over-excitation seems to be counteracted by the mentioned inhibitory circuits. However, because of the degeneration of amacrine cells, the protective inhibition of ganglion cells should be less effective.

Regardless of the degeneration of cones in the ONL, there are virtually no activated microglial cells in the outer retina. The group of Li and co-workers (Li et al., 2019) described the same distribution pattern of microglia which was predominately located in the retinal synapse layers. “The adult OPL contains ∼47 %of the microglial population, while 53 %are found in the IPL.” The authors interpreted this phenomenon as result of retinal synapse development and emphasized that the ONL is largely devoid of microglial cells.

In addition, ACA/R also led to the massive activation of astroglia, including Müller cells. This phenomenon has been described in several different ischemia models [reviewed in: (Bringmann et al., 2006; Bringmann et al., 2005); EIOP: (Pannicke et al., 2004; Renner et al., 2017; Wurm et al., 2011)]. This observation should correspond to the expression of a massive self-defense mechanism in the ACA/R-stressed retina. Müller cells afford an additional widespread neuroprotective system for neurons [reviewed in (Eastlake et al., 2020; Subirada et al., 2018)]. Müller cells are a source of neurotrophins, such as BDNF, NGF, NT-3, NT-4, and GDNF (Taylor et al., 2003), as well as bFGF (Berk et al., 2015) and antioxidant factors, such as glutamate transporters and glutathione (Pfeiffer et al., 2016). Furthermore, they are able to modulate neuronal activity by regulating the concentration of neuroactive substances, including K⁺, GABA, and H⁺, in the extracellular space. They also initiate the recycling of the glutamate neurotransmitter, as glutamine synthetase is expressed exclusively in Müller cells (Bringmann et al., 2013). In ischemia, this competence of Müller cells is neuroprotective, while also counteracting neuronal over-excitation by neurotransmitters (Bringmann et al., 2005) and potassium (Szabo et al., 1992). Moreover, Müller cells regulate the retinal blood flow; their endfeet ensheathe retinal blood vessels and release vasoactive agents, thus directly counteracting ischemia (Newman, 2015). Interestingly, the columnar units formed by Müller cells contain the rod photoreceptors and their bipolar cells (Newman & Reichenbach, 1996; Reichenbach et al., 1993). These functional units could provide a further explanation of the heterogeneous sensitivity to ischemia of the different retinal cell types. For example, it has been shown that the spread of excitation within these columns is limited.

On the other hand, strong gliosis is critical for neurodegeneration/neuroprotection (Bringmann et al., 2006). The inward-rectifier potassium channels (Kir) decreases under conditions of ischemia/reperfusion (Pannicke et al., 2004). That diminishes the mentioned above ability of Müller cells to maintain the neuronal potassium homeostasis. Moreover, activated Müller cells are booster of the secondary inflammatory neuronal degeneration (Portillo et al., 2009; Schultz et al., 2018). These ACA/R-induced changes in Müller cells may explain, at least partially, the imperfect neuroprotective potency of Müller cells and the profound retinal neurodegeneration pattern in our ACA/R model.

ACA/R is one of the injuries, which do not include any pathogen invasion but still invoke an inflammatory response (Minhas et al., 2016). ACA/R-mediated inflammation is a response on stress due to reperfusion. It is characterized by endogenous damage-associated molecular patterns (DAMPs), released by necrotic or apoptotic cells (Chen & Nunez, 2010). These components together with released cytokines, chemokines, and other effector molecules stimulate microglia, astroglia and Müller cells. That might be in turn an impulse to further intensify inflammatory processes (Minhas et al., 2016). Interestingly, the retina has long been counted among the immune privileged structures. However, this could not be sustained in view of the manifold immunological processes in responses of diverse cell stressors including ischemia/reperfusion.

The morphometric analysis of the optic nerve did not reveal extensive ACA/R-mediated alterations; however, deformed fibers were obvious. This result agrees with those of other similar studies, which reported in the EIOP model either no ischemia-induced degeneration of the optic nerve (Adachi et al., 1997; Belforte et al., 2011) or degeneration occurring after longer-lasting ischemia (Adachi et al., 1996; Du et al., 2020; Gallyas et al., 1992). One can only speculate about the reasons for this observation. However, there is evidence that endogenous protection via radical scavengers may play a decisive role in this process, as MnSOD production is strongly intensified by ischemia−reperfusion (Mrsic-Pelcic et al., 2004). Interestingly, studies using the model of increased intraocular pressure indicate a more significant and earlier damage of the optic nerve; however, no morphometric parameters were evaluated (Palmhof et al., 2020; Renner et al., 2017; Wong et al., 2015).

As mentioned in the Introduction, loss of vision is one of the most disabling residual impairments after cerebral infarction (Pula & Yuen, 2017). However, vision-related data (experimental as well as clinical) after ischemia−reperfusion, including cardiac arrest, are heterogeneous, ranging from descriptions of persisting complete loss of vision (Stiles et al., 2012), to temporary/reversible blindness (de Souza et al., 2017), and general disregard. We did not perform a vison test in this study, and the vision reflex, which was tested in the context of the neurological deficient score, reached normal values 7 days after ACA/R (Keilhoff, 2020).

Therefore, the regeneration-supporting potency of Müller cells should be considered. There is evidence that reactive Müller cell gliosis exhibits stem cell/neurogenic competence (Beach et al., 2017; Garcia-Garcia et al., 2020; Xia & Ahmad, 2016). The upregulation of GFAP, as demonstrated here and in studies of EIOP (Cho et al., 2011; Wurm et al., 2011) and 4-vessel-occlusion (Osborne et al., 1991), was associated with the dedifferentiation of Müller cells, which represents a module of reprogramming to a neurogenic state (Lahne et al., 2020). Moreover, the ACA/R-induced neuronal cell loss was incomplete. According to the “use it or lose it” dogma in the retina (Fleitas et al., 2020), the surviving neuronal structures seem to be able to maintain retinal functionality, which can be additionally improved by regenerative processes.

Our study contradicts the results of clinical studies who reported no (Miller, 1975), or only transient changes (Nickel & Hoyt, 1982; Priluck & Eifrig, 1978) in retinal morphology after cardiac arrest induced global ischemia−reperfusion or selective impairment in a 4-vessel-occlusion model in rats (Zhao et al., 2013). There is a consensus regarding the contention that the great heterogeneity of ischemia models hampers the comparison of results. In both 2- and 4-vessel-occlusion models, diverse collaterals can diminish the ischemic effect. Moreover, none of the models of increased intraocular pressure considers the impact of whole-body effects, such as those caused by ACA/R. Moreover, the heterogeneity of the markers used for immunostaining will yield different results. As a specific limitation of our study, we would cite our experience of co-staining, in which the staining intensity of an antibody can vary depending on the antibody combination used. Furthermore, it is difficult to decide whether a reduced fluorescence signal describes cell loss or the downregulation of the respective protein.

5 Conclusion

The neurodegenerative pattern of the rat retina 7 and 21 days after ACA/R was characterized by (i) a significant reduction of retinal thickness after long-term survival; (ii) ganglion cell degeneration, including their fiber network of IPL; (iii) degeneration of the amacrine and cone bipolar cells of the INL; and (iv) degeneration of the cone photoreceptor cells. The rod photoreceptors offered a higher resistance to ACA/R, possibly because of the strong upregulation of the calcium-binding proteins calretinin and parvalbumin, which counteracted the detrimental calcium overload. ON-type rod bipolar and horizontal cells were also less vulnerable to ACA/R. Moreover, a significant activation of Müller cells, as a further element of the anti-stress self-defense mechanism of the retina, was evident. The optic nerve showed ACA/R-mediated fiber deformation.

Footnotes

Acknowledgments

The technical assistance of Leona Bück and Andrea Mohrmann is gratefully acknowledged.

References

Adachi,

, Takahashi,

, Nishikawa,

, Miki,

, & Uyama,

(1996). High intraocular pressure-induced ischemia and reperfusion injury in the optic nerve and retina in rats. Graefe’s Archive for Clinical and Experimental Ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie, 234(7), 445–451. doi: 10.1007/BF02539411

Adachi,

, Takahashi,

, Yuge,

, Miki,

, & Uyama,

(1997). [’Ischemic tolerance’ in ischemia-reperfusion injury in the optic nerve in rats]. Nippon Ganka Gakkai zasshi, 101(1), 24–29.

Beach,

K.M.

, Wang,

, & Otteson,

D.C.

(2017). Regulation of Stem Cell Properties of Muller Glia by JAK/STAT and MAPK Signaling in the Mammalian Retina. Stem Cells International, 2017, 1610691. doi: 10.1155/2017/1610691

Belforte,

, Sande,

P.H.

, de Zavalia,

, Fernandez,

D.C.

, Silberman,

D.M.

, Chianelli,

M.S.

, & Rosenstein,

R.E.

(2011). Ischemic tolerance protects the rat retina from glaucomatous damage. PloS One, 6(8), e23763. doi: 10.1371/journal.pone.0023763

Berk,

B.A.

, Vogler,

, Pannicke,

, Kuhrt,

, Garcia,

T.B.

, Wiedemann,

,... Bringmann,

(2015). Brain-derived neurotrophic factor inhibits osmotic swelling of rat retinal glial (Muller) and bipolar cells by activation of basic fibroblast growth factor signaling. Neuroscience, 295, 175–186. doi: 10.1016/j.neuroscience.2015.03.037

Boretsky,

, Gupta,

, Tirgan,

, Liu,

, Godley,

B.F.

, Zhang,

,... Motamedi,

(2015). Nicotine accelerates diabetes-induced retinal changes. Current Eye Research, 40(4), 368–377. doi: 10.3109/02713683.2014.924147

Bourque,

S.L.

, Kuny,

, Reyes,

L.M.

, Davidge,

S.T.

, & Sauve,

(2013). Prenatal hypoxia is associated with long-term retinal dysfunction in rats. PloS One, 8(4), e61861. doi: 10.1371/journal.pone.0061861

Brandstatter,

J.H.

, Lohrke,

, Morgans,

C.W.

, & Wassle,

(1996). Distributions of two homologous synaptic vesicle proteins, synaptoporin and synaptophysin, in the mammalian retina. The Journal of Comparative Neurology, 370(1), 1–10. doi: 10.1002/(SICI)1096-9861(19960617)370:1<1::AID-CNE1>3.0.CO;2-7

Bringmann,

, Grosche,

, Pannicke,

, & Reichenbach,

(2013). GABA and Glutamate Uptake and Metabolism in Retinal Glial (Muller) Cells. Frontiers in Endocrinology, 4, 48. doi: 10.3389/fendo.2013.00048

10.

Bringmann,

, Pannicke,

, Grosche,

, Francke,

, Wiedemann,

, Skatchkov,

S.N.

,... Reichenbach,

(2006). Muller cells in the healthy and diseased retina. Progress in Retinal and Eye Research, 25(4), 397–424. doi: 10.1016/j.preteyeres.2006.05.003

11.

Bringmann,

, Uckermann,

, Pannicke,

, Iandiev,

, Reichenbach,

, & Wiedemann,

(2005). Neuronal versus glial cell swelling in the ischaemic retina. Acta Ophthalmologica Scandinavica, 83(5), 528–538. doi: 10.1111/j.1600-0420.2005.00565.x

12.

Caprara,

, & Grimm,

(2012). From oxygen to erythropoietin: relevance of hypoxia for retinal development, health and disease. Progress in Retinal and Eye Research, 31(1), 89–119. doi: 10.1016/j.preteyeres.2011.11.003

13.

Chen,

G.Y.

, & Nunez,

(2010). Sterile inflammation: sensing and reacting to damage. Nature reviews. Immunology, 10(12), 826–837. doi: 10.1038/nri2873

14.

Chen,

Y.I.

, Lee,

Y.J.

, Wilkie,

D.A.

, & Lin,

C.T.

(2014). Evaluation of potential topical and systemic neuroprotective agents for ocular hypertension-induced retinal ischemia-reperfusion injury. Veterinary Ophthalmology, 17(6), 432–442. doi: 10.1111/vo12105

15.

Cho,

K.J.

, Kim,

J.H.

, Park,

H.Y.

, & Park,

C.K.

(2011). Glial cell response and iNOS expression in the optic nerve head and retina of the rat following acute high IOP ischemia-reperfusion. Brain Research, 1403, 67–77. doi: 10.1016/j.brainres.2011.06.005

16.

Chun,

M.H.

, Kim,

I.B.

, Ju,

W.K.

, Kim,

K.Y.

, Lee,

M.Y.

, Joo,

C.K.

, & Chung,

J.W.

(1999). Horizontal cells of the rat retina are resistant to degenerative processes induced by ischemia-reperfusion. Neuroscience Letters, 260(2), 125–128. doi: 10.1016/s0304-3940(98)00973-2

17.

Connaughton,

(1995). Glutamate and Glutamate Receptors in the Vertebrate Retina. In H. Kolb, E. Fernandez & R. Nelson (Eds.), Webvision: The Organization of the Retina and Visual System. Salt Lake City (UT).

18.

D’Onofrio,

P.M.

, & Koeberle,

P.D.

(2013). What can we learn about stroke from retinal ischemia models? Acta Pharmacologica Sinica, 34(1), 91–103. doi: 10.1038/aps.2012.165

19.

de Souza,

, de Souza,

R.J.

, & Pai Kakode,

V.R.

(2017). Delayed-onset Reversible Cortical Blindness after Resuscitation from Cardiac Arrest. Journal of Neurosciences in Rural Practice, 8(Suppl 1), S133–S135. doi: 10.4103/jnrjnrp_63_17

20.

Delbarre,

, Delbarre,

, Calinon,

, & Ferger,

(1991). Accumulation of amino acids and hydroxyl free radicals in brain and retina of gerbil after transient ischemia. Journal of Ocular Pharmacology, 7(2), 147–155. doi: 10.1089/jo1991.7.147

21.

Diamond,

J.S.

(2017). Inhibitory Interneurons in the Retina: Types, Circuitry, and Function. Annual Review of Vision Science, 3, 1–24. doi: 10.1146/annurev-vision-102016-061345

22.

Dijk,

, & Kamphuis,

(2004a). An immunocytochemical study on specific amacrine cell subpopulations in the rat retina after ischemia. Brain Research, 1026(2), 205–217. doi: 10.1016/j.brainres.2004.08.014

23.

Dijk,

, & Kamphuis,

(2004b). Ischemia-induced alterations of AMPA-type glutamate receptor subunit. Expression patterns in the rat retina–an immunocytochemical study. Brain Research, 997(2), 207–221. doi: 10.1016/j.brainres.2003.08.069

24.

Dijk,

, Kraal-Muller,

, & Kamphuis,

(2004). Ischemia-induced changes of AMPA-type glutamate receptor subunit expression pattern in the rat retina: a real-time quantitative PCR study. Investigative Ophthalmology & Visual Science, 45(1), 330–341. doi: 10.1167/iovs.03-0285

25.

Dijk,

, van Leeuwen,

, & Kamphuis,

(2004). Differential effects of ischemia/reperfusion on amacrine cell subtype-specific transcript levels in the rat retina. Brain Research, 1026(2), 194–204. doi: 10.1016/j.brainres.2004.08.034

26.

Dilsiz,

, Sahaboglu,

, Yildiz,

M.Z.

, & Reichenbach,

(2006). Protective effects of various antioxidants during ischemia-reperfusion in the rat retina. Graefe’s archive for clinical and experimental ophthalmology=Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie, 244(5), 627–633. doi: 10.1007/s00417-005-0084-6

27.

Du,

, Wang,

, & He,

(2020). BDNF improves axon transportation and rescues visual function in a rodent model of acute elevation of intraocular pressure. Science China. Life sciences, 63(9), 1–10. doi: 10.1007/s11427-019-1567-0

28.

Dvoriantchikova,

, Degterev,

, & Ivanov,

(2014). Retinal ganglion cell (RGC) programmed necrosis contributes to ischemia-reperfusion-induced retinal damage. Experimental Eye Research, 123, 1–7. doi: 10.1016/j.exer.2014.04.009

29.

Eastlake,

, Luis,

, & Limb,

G.A.

(2020). Potential of Muller Glia for Retina Neuroprotection. Current Eye Research, 45(3), 339–348. doi: 10.1080/02713683.2019.1648831

30.

Feher,

, & Antal,

(1979). Ischemic retinal alterations in cardiac arrest. Annals of Ophthalmology, 11(6), 909–913.

31.

Fernandez,

J.C.

, Pelaez,

, Rey-Funes,

, Solino,

, Contartese,

D.S.

, Dorfman,

V.B.

,... Martinez,

(2020). Methylene Blue Prevents Retinal Damage Caused by Perinatal Asphyxia in the Rat. Frontiers in Cellular Neuroscience, 14, 157. doi: 10.3389/fncel.2020.00157

32.

Flammer,

(1994). The vascular concept of glaucoma. Survey of ophthalmology, 38(Suppl), S3–6. doi: 10.1016/0039-6257(94)90041-8

33.

Fleitas,

M.F.G.

, Aranda,

M.L.

, Dieguez,

H.H.

, Milne,

, Langellotti,

, Miranda,

,... Rosenstein,

R.E.

(2020). The “Use It or Lose It” Dogma in the Retina: Visual Stimulation Promotes Protection Against Retinal Ischemia. Molecular Neurobiology, 57(1), 435–449. doi: 10.1007/s12035-019-01715-5

34.

Fulton,

A.B.

, Akula,

J.D.

, Mocko,

J.A.

, Hansen,

R.M.

, Benador,

I.Y.

, Beck,

S.C.

,... Harris,

M.E.

(2009). Retinal degenerative and hypoxic ischemic disease. Documenta Ophthalmologica. Advances in Ophthalmology, 118(1), 55–61. doi: 10.1007/s10633-008-9127-8

35.

Gallyas,

, Hsu,

, & Buzsaki,

(1992). Delayed degeneration of the optic tract and neurons in the superior colliculus after forebrain ischemia. Neuroscience Letters, 144(1-2), 177–179. doi: 10.1016/0304-3940(92)90744-r

36.

Garcia-Garcia,

, Locker,

, & Perron,

(2020). Update on Muller glia regenerative potential for retinal repair. Current Opinion in Genetics & Development, 64, 52–59. doi: 10.1016/j.gde.2020.05.025

37.

Gilgun-Sherki,

, Rosenbaum,

, Melamed,

, & Offen,

(2002). Antioxidant therapy in acute central nervous system injury: current state. Pharmacological Reviews, 54(2), 271–284. doi: 10.1124/pr.54.2.271

38.

Huang,

H.M.

, Huang,

C.C.

, Hung,

P.L.

, & Chang,

Y.C.

(2012). Hypoxic-ischemic retinal injury in rat pups. Pediatric Research, 72(3), 224–231. doi: 10.1038/pr.2012.74

39.

Intartaglia,

, Giamundo,

, Marrocco,

, Maffia,

, Salierno,

F.G.

, Nusco,

,... Sorrentino,

N.C.

(2020). Retinal Degeneration in MPS-IIIA Mouse Model. Frontiers in Cell and Developmental Biology, 8, 132. doi: 10.3389/fcell.2020.00132

40.

Ishikawa,

, Yoshitomi,

, Zorumski,

C.F.

, & Izumi,

(2014). Neurosteroids are endogenous neuroprotectants in an ex vivo glaucoma model. Investigative Ophthalmology & Visual Science, 55(12), 8531–8541. doi: 10.1167/iovs.14-15624

41.

Kaufman,

M.L.

, Park,

K.U.

, Goodson,

N.B.

, Chew,

, Bersie,

, Jones,

K.L.

,... Brzezinski,

J.A. t.

(2019). Transcriptional profiling of murine retinas undergoing semi-synchronous cone photoreceptor differentiation. Developmental Biology, 453(2), 155–167. doi: 10.1016/j.ydbio.2019.05.016

42.

Keilhoff,

, Esser,

, Titze,

, Ebmeyer,

, & Schild,

(2017). High-potential defense mechanisms of neocortex in a rat model of transient asphyxia induced cardiac arrest. Brain Research, 1674, 42–54. doi: 10.1016/j.brainres.2017.08.018

43.

Keilhoff,

, Nguyen Thi,

T.M.

, Esser,

, & Ebmeyer,

(2020). Relative Resilience of Cerebellar Purkinje Cells in a Cardiac Arrest/Resuscitation Rat Model. Neurocritical Care, 32(3), 775–789. doi: 10.1007/s12028-019-00799-0

44.

Keilhoff,

, Schweizer,

, John,

, Langnaese,

, & Ebmeyer,

(2011). Minocycline neuroprotection in a rat model of asphyxial cardiac arrest is limited. Resuscitation, 82(3), 341–349. doi: 10.1016/j.resuscitation.2010.11.011

45.

Keilhoff,

, Titze,

, Esser,

, Langnaese,

, & Ebmeyer,

(2015). Constitutive and functional expression of YB- in microglial cells. Neuroscience, 301, 439–453. doi: 10.1016/j.neuroscience.2015.06.023

46.

Keilhoff,

, Titze,

, Rathert,

, Lucas,

, Esser,

, & Ebmeyer,

(2020). Normoxic post-ROSC ventilation delays hippocampal CA neurodegeneration in a rat cardiac arrest model, but does not prevent it. Experimental Brain Research, 238(4), 807–824. doi: 10.1007/s00221-020-05746-6

47.

Khalilpour,

, Latifi,

, Behnammanesh,

, Majid,

A.S. A.

, & Tamayol,

(2017). Ischemic optic neuropathy as a model of neurodegenerative disorder: A review of pathogenic mechanism of axonal degeneration and the role of neuroprotection. Journal of the Neurological Sciences, 375, 430–441. doi: 10.1016/j.jns.2016.12.044

48.

Kim,

B.J.

, Braun,

T.A.

, Wordinger,

R.J.

, & Clark,

A.F.

(2013). Progressive morphological changes and impaired retinal function associated with temporal regulation of gene expression after retinal ischemia/reperfusion injury in mice. Molecular Neurodegeneration, 8, 21. doi: 10.1186/1750-1326-8-21

49.

Kim,

S.A.

, Jeon,

J.H.

, Son,

M.J.

, Cha,

, Chun,

M.H.

, & Kim,

I.B.

(2010). Changes in transcript and protein levels of calbindin Dk, calretinin and parvalbumin, and numbers of neuronal populations expressing these proteins in an ischemia model of rat retina. Anatomy & Cell Biology, 43(3), 218–229. doi: 10.5115/acb.2010.43.3.218

50.

Kolb,

. (1995a). Feedback Loops in the Retina. In H. Kolb, E. Fernandez & R. Nelson (Eds.), Webvision: The Organization of the Retina and Visual System. Salt Lake City (UT).

51.

Kolb,

. (1995b). Neurotransmitters in the Retina. In H. Kolb, E. Fernandez & R. Nelson (Eds.), Webvision: The Organization of the Retina and Visual System. Salt Lake City (UT).

52.

Kuny,

, Gaillard,

, Mema,

S.C.

, Freund,

P.R.

, Zhang,

, Macdonald,

I.M.

,... Sauve,

(2010). Inner retina remodeling in a mouse model of stargardt-like macular dystrophy (STGD3). Investigative Ophthalmology & Visual Science, 51(4), 2248–2262. doi: 10.1167/iovs.09-4718

53.

Kuriyama,

, Waki,

, Nakagawa,

, & Tsuda,

(2001). Involvement of oxygen free radicals in experimental retinal ischemia and the selective vulnerability of retinal damage. Ophthalmic Research, 33(4), 196–202. doi: 10.1159/000055670

54.

Kwon,

O.J.

, Kim,

J.Y.

, Kim,

S.Y.

, & Jeon,

C.J.

(2005). Alterations in the localization of calbindin DK-, calretinin-, and parvalbumin-immunoreactive neurons of rabbit retinal ganglion cell layer from ischemia and reperfusion. Molecules and Cells, 19(3), 382–390.

55.

Lahne,

, Nagashima,

, Hyde,

D.R.

, & Hitchcock,

P.F.

(2020). Reprogramming Muller Glia to Regenerate Retinal Neurons. Annual Review of Vision Science, 6, 171–193. doi: 10.1146/annurev-vision-121219-081808

56.

Lee,

J.H.

, Shin,

J.M.

, Shin,

Y.J.

, Chun,

M.H.

, & Oh,

S.J.

(2011). Immunochemical changes of calbindin, calretinin and SMI in ischemic retinas induced by increase of intraocular pressure and by middle cerebral artery occlusion. Anatomy & Cell Biology, 44(1), 25–34. doi: 10.5115/acb.2011.44.1.25

57.

Li,

, Jiang,

, & Samuel,

M.A.

(2019). Microglia in the developing retina. Neural Development, 14(1), 12. doi: 10.1186/s13064-019-0137-x

58.

Menger,

, Pow,

D.V.

, & Wassle,

(1998). Glycinergic amacrine cells of the rat retina. The Journal of Comparative Neurology, 34–46. doi: 10.1002/(sici)1096-9861(19981109)401:1<34::aid-cne3>3.0.co;2-p

59.

Miller,

J.R.

(1975). Optic nerve and retina after experimental circulatory arrest. Investigative Ophthalmology, 14(2), 146–148.

60.

Minhas,

, Morishita,

, & Anand,

(2012). Preclinical models to investigate retinal ischemia: advances and drawbacks. Frontiers in Neurology, 3, 75. doi: 10.3389/fneur.2012.00075

61.

Minhas,

, Sharma,

, & Khan,

(2016). Cellular Stress Response and Immune Signaling in Retinal Ischemia-Reperfusion Injury. Frontiers in Immunology, 7, 444. doi: 10.3389/fimmu.2016.00444

62.

Mrsic-Pelcic,

, Pelcic,

, Peternel,

, Pilipovic,

, Simonic,

, & Zupan,

(2004). Effects of the hyperbaric oxygen treatment on the Na+,K+-ATPase and superoxide dismutase activities in the optic nerves of global cerebral ischemia-exposed rats. Progress in Neuro-psychopharmacology & Biological Psychiatry, 28(4), 667–676. doi: 10.1016/j.pnpb2004.05.003

63.

Nag,

T.C.

, & Wadhwa,

(2012). Ultrastructure of the human retina in aging and various pathological states. Micron, 43(7), 759–781. doi: 10.1016/j.micron.2012.01.011

64.

Nemitz,

, Dedek,

, & Janssen-Bienhold,

(2019). Rod Bipolar Cells Require Horizontal Cells for Invagination Into the Terminals of Rod Photoreceptors. Frontiers in Cellular Neuroscience, 13, 423. doi: 10.3389/fncel.2019.00423

65.

Neufeld,

A.H.

, Kawai,

, Das,

, Vora,

, Gachie,

, Connor,

J.R.

, & Manning,

P.T.

(2002). Loss of retinal ganglion cells following retinal ischemia: the role of inducible nitric oxide synthase. Experimental Eye Research, 75(5), 521–528. doi: 10.1006/exer.2002.2042

66.

Newman,

(2015). Glial cell regulation of neuronal activity and blood flow in the retina by release of gliotransmitters. Philosophical transactions of the Royal Society of London. Series B, Biological Sciences, 370(1672). doi: 10.1098/rstb.2014.0195

67.

Newman,

, & Reichenbach,

(1996). The Muller cell: a functional element of the retina. Trends in Neurosciences, 19(8), 307–312. doi: 10.1016/0166-2236(96)10040-0

68.

Nickel,

B.L.

, & Hoyt,

C.S.

(1982). The hypoxic retinopathy syndrome. American Journal of Ophthalmology, 93(5), 589–593. doi: 10.1016/s0002-9394(14)77373-8

69.

Niwa,

, Aoki,

, Hirata,

, Tomita,

, Green,

P.G.

, & Hara,

(2016). Retinal Cell Degeneration in Animal Models. International Journal of Molecular Sciences, 17(1). doi: 10.3390/ijms17010110

70.

Oguni,

, Setogawa,

, Shinohara,

, & Kato,

(1998). Calbindin-D kD and parvalbumin in the horizontal cells of rat retina during development. Current Eye Research, 17(6), 617–622.

71.

Olivares,

A.M.

, Han,

, Soto,

, Flattery,

, Marini,

, Mollema,

,... Haider,

N.B.

(2017). The nuclear hormone receptor gene Nrc (Tr) is a critical regulator of early retina cell patterning. Developmental Biology, 429(1), 343–355. doi: 10.1016/j.ydbio.2017.05.021

72.

Osborne,

N.N.

, Block,

, & Sontag,

K.H.

(1991). Reduction of ocular blood flow results in glial fibrillary acidic protein (GFAP) expression in rat retinal Muller cells. Visual Neuroscience, 7(6), 637–639. doi: 10.1017/s0952523800010427

73.

Osborne,

N.N.

, Casson,

R.J.

, Wood,

J.P.

, Chidlow,

, Graham,

, & Melena,

(2004). Retinal ischemia: mechanisms of damage and potential therapeutic strategies. Progress in Retinal and Eye Research, 23(1), 91–147. doi: 10.1016/j.preteyeres.2003.12.001

74.

Palmhof,

, Frank,

, Rappard,

, Kortenhorn,

, Demuth,

, Biert,

,... Joachim,

S.C.

(2019). From Ganglion Cell to Photoreceptor Layer: Timeline of Deterioration in a Rat Ischemia/Reperfusion Model. Frontiers in Cellular Neuroscience, 13, 174. doi: 10.3389/fncel.2019.00174

75.

Palmhof,

, Wagner,

, Nagel,

, Biert,

, Stute,

, Dick,

H.B.

, & Joachim,

S.C.

(2020). Retinal ischemia triggers early microglia activation in the optic nerve followed by neurofilament degeneration. Experimental Eye Research, 198, 108133. doi: 10.1016/j.exer.2020.108133

76.

Pannicke,

, Iandiev,

, Uckermann,

, Biedermann,

, Kutzera,

, Wiedemann,

,... Bringmann,

(2004). A potassium channel-linked mechanism of glial cell swelling in the postischemic retina. Molecular and Cellular Neurosciences, 26(4), 493–502. doi: 10.1016/j.mcn.2004.04.005

77.

Pfeiffer,

R.L.

, Marc,

R.E.

, Kondo,

, Terasaki,

, & Jones,

B.W.

(2016). Muller cell metabolic chaos during retinal degeneration. Experimental Eye Research, 150, 62–70. doi: 10.1016/j.exer.2016.04.022

78.

Portillo,

J.A.

, Okenka,

, Kern,

T.S.

, & Subauste,

C.S.

(2009). Identification of primary retinal cells and ex vivo detection of proinflammatory molecules using flow cytometry. Molecular Vision, 15, 1383–1389.

79.

Priluck,

I.A.

, & Eifrig,

D.E.

(1978). Transient visual impairment following cardiac arrest. Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie. Albrecht von Graefe’s archive for clinical and experimental ophthalmology, 206(1), 67–70. doi: 10.1007/BF00411339

80.

Produit-Zengaffinen,

, Favez,

, Pournaras,

C.J.

, & Schorderet,

D.F.

(2016). JNK Inhibition Reduced Retinal Ganglion Cell Death after Ischemia/Reperfusionand after Hypoxia In Vitro. Advances in Experimental Medicine and Biology, 854, 677–683. doi: 10.1007/978-3-319-17121-0_90

81.

Pula,

J.H.

, & Yuen,

C.A.

(2017). Eyes and stroke: the visual aspects of cerebrovascular disease. Stroke and Vascular Neurology, 2(4), 210–220. doi: 10.1136/svn-2017-000079

82.

Reichenbach,

, & Bringmann,

(2016). Purinergic signaling in retinal degeneration and regeneration. Neuropharmacology, 104, 194–211. doi: 10.1016/j.neuropharm.2015.05.005

83.

Reichenbach,

, Stolzenburg,

J.U.

, Eberhardt,

, Chao,

T.I.

, Dettmer,

, & Hertz,

(1993). What do retinal muller (glial) cells do for their neuronal ‘small siblings’? Journal of Chemical Neuroanatomy, 6(4), 201–213. doi: 10.1016/0891-0618(93)90042-3

84.

Renner,

, Stute,

, Alzureiqi,

, Reinhard,

, Wiemann,

, Schmid,

,... Joachim,

S.C.

(2017). Optic Nerve Degeneration after Retinal Ischemia/Reperfusion in a Rodent Model. Frontiers in Cellular Neuroscience, 11, 254. doi: 10.3389/fncel.2017.00254

85.

Rey-Funes,

, Dorfman,

V.B.

, Ibarra,

M.E.

, Pena,

, Contartese,

D.S.

, Goldstein,

,... Loidl,

C.F.

(2013). Hypothermia prevents gliosis and angiogenesis development in an experimental model of ischemic proliferative retinopathy. Investigative Ophthalmology & Visual science, 54(4), 2836–2846. doi: 10.1167/iovs.12-11198

86.

Rey-Funes,

, Ibarra,

M.E.

, Dorfman,

V.B.

, Lopez,

E.M.

, Lopez-Costa,

J.J.

, Coirini,

, & Loidl,

C.F.

(2010). Hypothermia prevents the development of ischemic proliferative retinopathy induced by severe perinatal asphyxia. Experimental Eye Research, 90(1), 113–120. doi: 10.1016/j.exer.2009.09.019

87.

Rey-Funes,

, Ibarra,

M.E.

, Dorfman,

V.B.

, Serrano,

, Fernandez,

A.P.

, Martinez-Murillo,

,... Loidl,

C.F.

(2011). Hypothermia prevents nitric oxide system changes in retina induced by severe perinatal asphyxia. Journal of Neuroscience Research, 89(5), 729–743. doi: 10.1002/jnr.22556

88.

Ruether,

, Feigenspan,

, Pirngruber,

, Leitges,

, Baehr,

, & Strauss,

(2010). PKCalpha is essential for the proper activation and termination of rod bipolar cell response. Investigative Ophthalmology & Visual Science, 51(11), 6051–6058. doi: 10.1167/iovs.09-4704

89.

Sawant,

O.B.

, Jidigam,

V.K.

, Fuller,

R.D.

, Zucaro,

O.F.

, Kpegba,

, Yu,

,... Rao,

(2019). The circadian clock gene Bmal is required to control the timing of retinal neurogenesis and lamination of Muller glia in the mouse retina. FASEB journal: official publication of the Federation of American Societies for Experimental Biology, 33(8), 8745–8758. doi: 10.1096/fj.201801832RR

90.

Schmid,

, Renner,

, Dick,

H.B.

, & v Joachim,

S.C.

(2014). Loss of inner retinal neurons after retinal ischemia in rats. Investigative Ophthalmology & Visual Science, 55(4), 2777–2787. doi: 10.1167/iovs.13-13372

91.

Schultz,

, Krug,

, Precht,

, Wohl,

S.G.

, Witte,

O.W.

, & Schmeer,

(2018). Frataxin overexpression in Muller cells protects retinal ganglion cells in a mouse model of ischemia/reperfusion injury in vivo. Scientific Reports, 8(1), 4846. doi: 10.1038/s41598-018-22887-5

92.

Shi,

, Ebrahim,

A.S.

, & Berger,

E.A.

(2018). A Contrast in Pathogenic Responses between CBL/J and BALB/cJ Mice Using a Model of Retinal Injury. The American Journal of Pathology, 188(12), 2717–2728. doi: 10.1016/j.ajpath.2018.08.010

93.

Shibuki,

, Katai,

, Yodoi,

, Uchida,

, & Yoshimura,

(2000). Lipid peroxidation and peroxynitrite in retinal ischemia-reperfusion injury. Investigative Ophthalmology &Visual Science, 41(11), 3607–3614.

94.

Solberg,

, Escobar,

, Arduini,

, Torres-Cuevas,

, Lahoz,

, Sastre,

,... Quintas,

(2013). Metabolomic Analysis of the Effect of Postnatal Hypoxia on the Retina in a Newly Born Piglet Model. PloS One, 8(6), e66540. doi: 10.1371/journal.pone.0066540

95.

Stiles,

, Weil,

A.B.

, Packer,

R.A.

, & Lantz,

G.C.

(2012). Post-anesthetic cortical blindness in cats: twenty cases. Veterinary Journal, 193(2), 367–373. doi: 10.1016/j.tvjl.2012.01.028

96.

Strosznajder,

, Chalimoniuk,

, Strosznajder,

R.P.

, Walski,

, Lupo,

, Anfuso,

C.D.

,... Alberghina,

(1998). Arachidonate transport through the blood-retina and blood-brain barrier of the rat after reperfusion of varying duration following complete cerebral ischemia. International journal of developmental neuroscience: the official journal of the International Society for Developmental Neuroscience, 16(2), 103–113. doi: 10.1016/s0736-5748(98)00005-7

97.

Subirada,

P.V.

, Paz,

M.C.

, Ridano,

M.E.

, Lorenc,

V.E.

, Vaglienti,

M.V.

, Barcelona,

P.F.

,... Sanchez,

M.C.

(2018). A journey into the retina: Muller glia commanding survival and death. The European Journal of Neuroscience, 47(12), 1429–1443. doi: 10.1111/ejn.13965

98.

Sucher,

N.J.

, Lipton,

S.A.

, & Dreyer,

E.B.

(1997). Molecular basis of glutamate toxicity in retinal ganglion cells. Vision Research, 37(24), 3483–3493. doi: 10.1016/S0042-6989(97)00047-3

99.

Szabo,

M.E.

, Droy-Lefaix,

M.T.

, Doly,

, & Braquet,

(1992). Ischaemia- and reperfusion-induced Na+,+, Ca+and Mg+shifts in rat retina: effects of two free radical scavengers, SOD and EGB. Experimental Eye Research, 55(1), 39–45. doi: 10.1016/0014-4835(92)90089-b

100.

Taylor,

, Srinivasan,

, Wordinger,

R.J.

, & Roque,

R.S.

(2003). Glutamate stimulates neurotrophin expression in cultured Muller cells. Brain Research. Molecular Brain Research, 111(1-2), 189–197. doi: 10.1016/s0169-328x(03)00030-5

101.

Vinores,

S.A.

, Derevjanik,

N.L.

, Mahlow,

, Hackett,

S.F.

, Haller,

J.A.

, deJuan,

,... Campochiaro,

P.A.

(1995). Class III beta-tubulin in human retinal pigment epithelial cells in culture and in epiretinal membranes. Experimental Eye Research, 60(4), 385–400. doi: 10.1016/s0014-4835(05)80095-8

102.

Weinreb,

, Bar-Am,

, Amit,

, Chillag-Talmor,

, & Youdim,

M.B.

(2004). Neuroprotection via pro-survival protein kinase C isoforms associated with Bcl- family members. FASEB journal: official publication of the Federation of American Societies for Experimental Biology, 18(12), 1471–1473. doi: 10.1096/fj.04-1916fje

103.

Wong,

, Li,

, Zhang,

, Li,

, Zhang,

,... Huang,

(2015). Therapeutic Retrobulbar Inhibition of STAT Protects Ischemic Retina Ganglion Cells. Molecular Neurobiology, 52(3), 1364–1377. doi: 10.1007/s12035-014-8945-9

104.

Wurm,

, Iandiev,

, Uhlmann,

, Wiedemann,

, Reichenbach,

, Bringmann,

, & Pannicke,

(2011). Effects of ischemia-reperfusion on physiological properties of Muller glial cells in the porcine retina. Investigative Ophthalmology & Visual science, 52(6), 3360–3367. doi: 10.1167/iovs.10-6901

105.

Xia,

, & Ahmad,

(2016). Unlocking the Neurogenic Potential of Mammalian Muller Glia. International Journal of Stem Cells, 9(2), 169–175. doi: 10.15283/ijsc16020

106.

Zhao,

, Zabel,

M.K.

, Wang,

, Ma,

, Shah,

, Fariss,

R.N.

,... Wong,

W.T.

(2015). Microglial phagocytosis of living photoreceptors contributes to inherited retinal degeneration. EMBO Molecular Medicine, 7(9), 1179–1197. doi: 10.15252/emmm.201505298

107.

Zhao,

, Yu,

, Xiang,

Y.H.

, Han,

X.J.

, Xu,

, So,

K.F.

,... Ruan,

Y.W.

(2013). Changes in retinal morphology, electroretinogram and visual behavior after transient global ischemia in adult rats. PloS One, 8(6), e65555. doi: 10.1371/journal.pone.0065555

Parameter		Post-ROSC time
	BL	5 min	15 min	30 min	45 min
Sham animals (n = 7)
Tympanic temperature (°C)	36.5±0.4	36.9±0.4	36.9±0.2	36.5±0.2	36.9±0.3
Rectal temperature (°C)	36.7±0.5	36.8±0.7	36.9±0.4	36.9±0.3	37.1±0.3
MAP (mmHg)	89±12	94±09	91±10	93±09	98±09
Heart rate (min^–1)	328±43	357±29	360±29	371±34	368±43
pCO₂ (mmHg)	40±06	40±04	41±03	41±01	40±01
Glucose (mmol/l)	11.4±3.7	12.1±4.2	12.6±3.2	11.5±3.8	11.9±2.3
pH	7.41±0.04	7.40±0.02	7.39±0.02	7.39±0.01	7.40±0.02
pO₂ (mmHg)	172±30	170±26	174±24	177±14	178±16
ACA/R animals (n = 17)
Tympanic temperature (°C)	36.7±0.6	36.8±0.6	36.8±0.5	37.1±0.4	37.0±0.4
Rectal temperature (°C)	36.4±0.3	36.6±0.5	36.6±0.5	36.8±0.3	36.7±0.6
MAP (mmHg)	77±17	169±14^****	163±21	89±20	84±18
Heart rate (min^–1)	312±19	424±30^***	424±33	324±32^**	301±27^***
pCO₂ (mmHg)	44±04	49±08	49±14	41±16	38±11
Glucose (mmol/l)	12.1±2.3	19.8±2.5^*	17.3±3.1	10.9±2.7	11.0±2.3
pH	7.38±0.03	7.25±0.08^****	7.31±0.10	7.35±0.09	7.39±0.07
pO₂ (mmHg)	170±37	251±87	251±86	397±97^*	414±104^**