Citrulline protects human retinal pigment epithelium from hydrogen peroxide and iron/ascorbate induced damages

Abstract Oxidative stress plays an important role in the ageing of the retina and in the pathogenesis of retinal diseases such as age‐related macular degeneration (ARMD). Hydrogen peroxide is a reactive oxygen species generated by the photo‐excited lipofuscin that accumulates during ageing in the retinal pigment epithelium (RPE), and the age‐related accumulation of lipofuscin is associated with ARMD. Iron also accumulates with age in the RPE that may contribute to ARMD as an important source of oxidative stress. The aim of this work was to investigate the effects of L‐Citrulline (CIT), a naturally occurring amino acid with known antioxidant properties, on oxidative stressed cultured RPE cells. Human RPE (ARPE‐19) cells were exposed to hydrogen peroxide (H2O2) or iron/ascorbate (I/A) for 4 h, either in the presence of CIT or after 24 h of pretreatment. Here, we show that supplementation with CIT protects ARPE‐19 cells against H2O2 and I/A. CIT improves cell metabolic activity, decreases ROS production, limits lipid peroxidation, reduces cell death and attenuates IL‐8 secretion. Our study evidences that CIT is able to protect human RPE cells from oxidative damage and suggests potential protective effect for the treatment of retinal diseases associated with oxidative stress.

Moreover, intracellular iron can interact with bisretinoid lipofuscin in RPE to promote cell damage. 14 Evidence for the involvement of oxidative stress and free radical damage in RPE degeneration during ageing and ARMD are also reported from studies showing that oral intake of antioxidants could reduce the risk of developing ARMD. 15 In addition, inflammation is implicated in the molecular mechanisms of ARMD pathogenesis, leading to RPE damage. The systemic and ocular levels of some pro-inflammatory and pro-angiogenic cytokines, such as Interleukin 8 (IL- 8), have been correlated with the incidence of ARMD. 16 Increased expression of IL-8 induced by oxidative stress is one of the earliest events of inflammation which could explain, at least in part, the inflammatory events involved in ARMD. 17 L-Citrulline (CIT), a naturally occurring amino acid, could be a good candidate for the prevention or treatment of retinal pathologies associated with oxidative stress. CIT has already won its spurs as antioxidant since it is a powerful hydroxyl radical scavenger. 18 CIT has also been reported to protect against lipid peroxidation and circulating lipoprotein oxidation, as well as to decrease protein carbonylation in muscle and brain. [19][20][21][22] Moreover, studies evidenced that CIT is beneficial in neurological pathologies associated with oxidative stress [23][24][25] and that this amino acid could be protective in the neurodegenerative process associated with ageing. 26 Finally, this amino acid is a precursor of arginine and nitric oxide, and therefore plays a key role at the cardiovascular and cerebral levels. 27 CIT is naturally synthesized by enterocytes from arginine or glutamine, and once released into the bloodstream escapes splanchnic sequestration and reaches the kidney where it is converted to arginine. 25 This amino acid is almost absent from the diet, with the exception of watermelon (citrullus vulgaris) where it is present in high concentrations.
It is also present in smaller amounts in cucumbers, pumpkins, melons and squashes. Furthermore, this amino acid is safe, well tolerated and has excellent bioavailability (80% of ingested CIT is found in the systemic blood circulation), as it has been largely demonstrated in both young adults and elderly subjects. [28][29][30][31] For these reasons, it seems that CIT could be a therapeutic strategy for the prevention/ treatment of retinal pathologies. CIT could easily spread in the retina, due to its very good bioavailability, and its involvement in other retinal function (vasodilation of retinal arterioles) has recently been shown after oral administration in rats. 32 The aim of the present study was to investigate the effects of CIT on oxidative stressed RPE cells. We have shown that CIT can protect human RPE cells from damage induced by H 2 O 2 or iron/ ascorbate. To our knowledge, this is the first study describing the effects of CIT against oxidative stress in RPE cells.

| Chemicals and reagents
Citrulline (CIT) was kindly provided by CITRAGE ® Company.

| Cell treatments
Pre-and co-treatments were carried out with CIT as follows. In cotreatment, cell cultures received a medium containing the oxidant in the presence of CIT for 4 h. In pretreatment, cultures first received a medium containing CIT for 24 h; the medium was then removed and replaced with a fresh culture medium containing the oxidant for 4 h.

| Cell metabolic activity
Cell metabolic activity was determined by the 3-(4,5-dimethylthiaz ol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. After treatment with the oxidant and/or CIT, RPE cells were rinsed in phosphate buffer saline (PBS) and incubated for 2 h with fresh culture medium containing 0.5 mg/ml MTT. During this incubation time, mitochondrial dehydrogenases of living cells reduced MTT to purple formazan.
Cells were then rinsed in PBS and the insoluble purple formazan product was dissolved with dimethyl sulfoxide, forming a coloured solution. After centrifugation at 2000 g for 5 min, the absorbance of the supernatants, proportional to the number of living cells, was read at 570 nm with a microplate reader. The results are expressed as the percentage of control condition representing 100% of viability (cells incubated in normal medium only = 100% of absorbance).

| LDH release
Lactate dehydrogenase (LDH) released from injured RPE cells into the culture medium was quantified by a coupled enzymatic reaction in which LDH catalyses the conversion of lactate to pyruvate via NAD+ reduction to NADH. Diaphorase then uses NADH to reduce a tetrazolium salt to a red formazan product. After treatment with the oxidant and/or CIT, the supernatants were collected and mixed with reaction mixture. Following incubation in the dark for 30 min at room temperature, the absorbance, proportional to the quantity of LDH released into the culture medium, was determined at 490 nm using a microplate reader. The results are expressed in units of absorbance (LDH levels).

| Interleukin-8 production
Interleukin-8 (IL-8) released in the medium was determined by enzyme-linked immunosorbent assay (ELISA). After treatment of RPE cells with the oxidant and/or CIT, the media were removed and replaced with a fresh culture medium. Following 24 h incubation, the supernatants were harvested for measuring IL-8 by ELISA. Briefly, capture antibody was diluted in coating buffer and applied to a 96well plate overnight. Next, cell culture medium samples were added to each well and incubated for 2 h at room temperature after which the detection antibody was added for 1 h. After washing, avidinhorseradish peroxidase was added to each well and left to incubate for 30 min at room temperature. The substrate solution was then added to each well for 30 min in the dark. Finally, a stop solution was added to inhibit the reaction, and the absorbance was read at 450 nm. The results are expressed as pg of IL-8 per ml of medium.

| Statistical analysis
The results correspond to the means ± SEM of n independent experiments. In each experiment, all conditions were done at least in triplicate. Statistical analysis was performed using Student's t-test: *p < 0.05, **p < 0.01, ***p < 0.001.

| Cytotoxic effects of CIT in RPE cells
We first assessed the toxicity of CIT on human RPE cells. For this purpose, cell cultures were incubated with several concentrations of CIT for 24 h, and cell viability was measured using MTT. As shown in Figure 1A, CIT did not affect metabolic activity of ARPE-19 cells from 1 mM to 100 mM, but exhibited significant decreases from 200 mM. Therefore, the results show that CIT is relatively safe for RPE cells at concentrations up to 100 mM.

| CIT improves cell metabolic activity in oxidative stressed RPE cells
To determine whether CIT can protect RPE cells from oxidative damage, we examined the effect of CIT against oxidative stress induced by H 2 O 2 . In a first set of experiments, cell cultures were incubated with H 2 O 2 in the presence of CIT at different concentrations for 4 h, and cell metabolic activity was measured using MTT. As shown in Figure 1B, treatment of RPE cells with H 2 O 2 0.6 mM caused a significant decrease in cell viability (52 ± 11%), whereas co-treatment with CIT 50, 100 and 200 mM significantly reduced this decrease (75 ± 9%, 79 ± 9% and 74 ± 12% of cell viability, respectively). In another set of experiments, cell cultures were pretreated with increasing concentrations of CIT for 24 h, washed and exposed to H 2 O 2 0.6 mM for 4 h.

Pretreatment of the cells had no protective effect against H 2 O 2 , at
any of the CIT concentrations tested (data not shown).
We also examined the effect of CIT against damage induced by iron/ascorbate (I/A). The combination of iron and ascorbate triggers a Fenton reaction with formation of hydroxyl radicals, which causes lipid peroxidation, membrane damage and cell death. As shown in Figure 1C, exposure of RPE cells to I/A 7.5 mM/0.3 M for 4 h led to a significant decrease in cell viability (21 ± 9%), whereas pretreatment of the cells with CIT 50, 100, 200 and 300 mM for 24 h significantly reduced this decrease (37 ± 14%, 44 ± 14%, 42 ± 10% and 38 ± 9%, respectively). Conversely, co-treatment of the cells with CIT and I/A did not improve metabolic activity, at any of the CIT concentrations tested (data not shown).
Thus, CIT is able to reduce the toxicity of H 2 O 2 and I/A in RPE cells, as shown by MTT assay. CIT is effective against H 2 O 2 in cotreatment, while it is effective against I/A in pretreatment. As we did not observe any improvement against H 2 O 2 with CIT in pretreatment, only co-treatments were carried out in the following experiments. Likewise, as no protection was observed against I/A with CIT in co-treatment, only pretreatments were performed in the following experiments.

| CIT decreases H 2 O 2 -induced ROS production in RPE cells
We investigated whether CIT could counteract intracellular produc-  We also examined the effect of CIT against lipid peroxidation induced by I/A. Cell cultures were pretreated with CIT for 24 h and then exposed to I/A for 4 h, and lipid peroxidation was quantified by flow cytometry. As shown in Figure 3B, treatment of RPE cells with I/A 7.5 mM/0.3 M resulted in 46 ± 11% of green-stained cells.

| CIT limits lipid peroxidation in oxidative stressed RPE cells
A pretreatment of the cells with CIT 100 mM before exposure to I/A significantly reduced the proportion of stained cells (19 ± 7%).

| CIT reduces cell death in oxidative stressed RPE cells
To To examine the effect of CIT on I/A-induced cell death, cell cultures were pretreated with CIT for 24 h and then exposed to I/A for 4 h, and cell death was quantified by flow cytometry. As shown in Figure 5, exposure of RPE cells to I/A 7.5 mM/0.3 M led to significant cell death (44%), mainly by late apoptosis/necrosis (38 ± 9%). A pretreatment of the cells with CIT 50 and 100 mM, before exposure to I/A, significantly reduced the percentage of total dead cells (35% and 27%, respectively) and specifically late apoptotic/necrotic cells (30 ± 10% and 20 ± 5%, respectively).

| CIT attenuates H 2 O 2 -induced IL-8 secretion in RPE cells
We measured the level of the pro-inflammatory cytokine IL-8 in RPE cells treated with H 2 O 2 in the presence of CIT. As shown in Figure 6, is also generated by the photo-excited pigment lipofuscin accumulating during ageing in the RPE, and the accumulation of lipofuscin is strongly associated with ARMD. 8 The iron/ascorbate system is commonly used in many studies to generate free radicals and lipid peroxidation. [33][34][35][36][37] Iron levels increase in RPE during ageing, which may contribute to ARMD as an important source of oxidative stress. [9][10][11] The increase of intracellular ferrous iron produces hydroxyl and lipid alkoxyl radicals through the Fenton reaction, which causes lipid peroxidation, membrane damage and cell death. 12,13 First, before testing the potential efficiency of a CIT treatment, we assessed its toxicity towards RPE cells. We observed that CIT alone did not affect the RPE viability at concentrations up to 100 mM, as determined by MTT assay. Thus, CIT safety was confirmed in cultured RPE cells, as it had already been evidenced in humans. 25,28 Then, we evaluated the protective effects of CIT against oxidative damage on RPE cell metabolic activity. We showed that CIT was effective against H 2 O 2 in co-treatment and against iron/ascorbate in pretreatment. Furthermore, the antioxidant effect of CIT has proved its efficiency by decreasing ROS production in RPE cells exposed    (%)

Living cells Early apoptoƟc cells Late apoptoƟc / necroƟc cells
Lipid peroxidation, a consequence of oxidative stress, plays an important role in the degeneration of RPE. As described above, the yellow-brown fluorescent pigment lipofuscin accumulates in RPE with age and this aged accumulation has been associated with ARMD. Moreover, lipofuscin has been shown to produce ROS (singlet oxygen, superoxide anion and hydrogen peroxide) and to increase lipid peroxidation. 40 In our study, exposure of RPE cells to H 2 O 2 increased lipid peroxidation, while co-treatment with CIT had a significant beneficial effect. On the contrary, incubation of RPE cells with iron/ascorbate also increased lipid peroxidation, whereas pretreatment with CIT limited this oxidation. This is in accordance with a work of Fu et al., 19 who evaluated the protective effects of CIT against renal ischaemia-reperfusion injury in rats.
The authors showed that CIT administered by gavage was able to decrease renal oxidative stress and to inhibit lipid peroxidation.
Our findings are also in accordance with the study of Moinard et al. 21 exploring the impact of CIT-enriched diet in healthy aged rats. The authors found that CIT supplementation was able to lower the susceptibility to oxidation of lipoproteins (lag phase significantly higher and maximal concentration of conjugated diene significantly lower).
We also examined whether the protective effects of CIT on   45 also reported that high concentration of H 2 O 2 was able to cause RPE cell death with typical features of necrosis such as cell swelling, loss of plasma membrane integrity and nuclear condensation. They also reported that H 2 O 2 -induced necrosis was a regulated process with cellular calcium overload as a critical step in the cell death program. In addition, Hanus et al. 41 showed that features of apoptosis were not observed in RPE cells induced RPE cell death. 12,46 Retinal iron levels increase with age, 47 and excessive iron accumulation is a source of free radical production in RPE. 11 Moreover, iron levels in RPE have been found to be higher in ARMD patients, 10,11 suggesting that it may be implicated in the pathogenesis of the disease. Recently, a study has shown that intracellular iron can interact with bisretinoid lipofuscin in RPE to promote cell damage. 14 In our study, exposure of RPE cultures to iron/ascorbate led to a significant cell death, mainly by late apoptosis/necrosis, and pretreatment with CIT significantly reduced this cell death. Our results are in agreement with the work of Fu et al. 19 who examined the effects of CIT on renal ischaemia-reperfusion injury in rats. Kidneys of ischaemic rats showed glomerular lesions and massive tubular epithelial cells necrosis or collapse, whereas pretreatment with CIT preserved the normal morphology of the kidneys.
Inflammation is also implicated in the molecular mechanisms of ARMD pathogenesis, leading to RPE damage. IL-8, a proinflammatory and pro-angiogenic cytokine, is an important mediator of inflammation, and the increased expression of IL-8 could explain, at least in part, the inflammatory events involved in ARMD. 17 In the present work, we observed that treatment with H 2 O 2 caused a significant production of IL-8 by RPE cells. Our results are in agreement with the work of Fernandes et al. 17 48 Thus, through its hydroxyl radical scavenging activity, CIT could have a direct action on IL-8 expression. Furthermore, it has been reported, in intestinal ischaemia and reperfusion rat model, that oral CIT supplementation can act on the activity of transcription factor NF-kB by decreasing the ratio of the phosphorylated to the total NF-kB. 49 Preclinical and clinical studies have also reported anti-inflammatory effects of CIT. For instance, Breuillard et al. 50 have evidenced anti-inflammatory properties of CIT, which normalizes nitric oxide production variability by peritoneal macrophages, both in vitro and in vivo, in aged rats with endotoxin challenge. Van Vliet et al. 51 have also reported in patients with chemotherapy-induced mucosal barrier injury that plasma CIT was negatively correlated to plasma IL-8 levels. Also, Luiking et al. 52 have shown in patients with sepsis that C-reactive protein was negatively correlated to plasma CIT concentration.
Finally, an important question is how CIT could have such effects? This could be related to the direct antioxidant potential of CIT, 18,21 to the activity of CIT on nuclear factors involved in the IL-8 regulation, 48,49 and also to its capacity to generate nitric oxide as already observed. 23 The last hypothesis could be related to the thermodynamic properties of CIT. We recently demonstrated that CIT was able to reallocate ATP consumption to muscle protein synthesis. 53 To summarize, in stress situations (like in our conditions), there is a decrease in ATP/ADP ratio that leads to a decrease in Gibbs free energy of ATP hydrolysis. In such conditions, many reactions (requiring high levels energy in cells) are no longer possible and it may lead to cell death. By its thermodynamic action, CIT may decrease activation energies of one or several ATP (and GTP)-consuming reactions involved in cell and, in fine, preserve cell from death. 54 Alterations of the cellular energy dynamics with reduced ATP have been reported in H 2 O 2 -treated ARPE-19 cells 55 and are classically observed during oxidative stress. Thus, we assume that the thermodynamic properties of CIT could also explain in part its protective effect.
In summary, our results evidence that CIT is capable to protect human RPE cells against H 2 O 2 -and iron/ascorbate-induced damages: CIT improves cell metabolic activity, decreases ROS production, limits lipid peroxidation, reduces cell death and attenuates IL-8 secretion. This suggests potential effects of CIT in the prevention or treatment of retinal diseases associated with oxidative stress, such as ARMD. Further studies will be necessary to examine in more details the mechanisms of action of the effective CIT against oxidative damage in RPE cells.

ACK N OWLED G EM ENTS
This work was supported in part by a grant from the CITRAGE ® Company and was presented at the XXIII Biennial Meeting of the International Society for Eye Research (ISER), Belfast, Northern Ireland, September 9-13, 2018. We would like to thank Florence Caldefie-Chezet (INRAE, UNH, ECREIN, UCA, Clermont-Ferrand), for allowing us to use their microplate reader. We also thank the Cellular Health Imaging Center (UCA, Clermont-Ferrand) for help with flow cytometry, and Inserm and UCA for their support. The authors are grateful to Dr. Eric Wersinger for its assistance in English language editing.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.