Abstract
We explored the effect of astaxanthin on vascular endothelial growth factor in the aqueous humor, by measuring vascular endothelial growth factor levels and oxidation-related parameters, including O2•− scavenging activity, H2O2 level, and total hydroperoxide level in the aqueous humor, obtained from 35 patients before and after astaxanthin administration. We evaluated the relationship between vascular endothelial growth factor and the oxidation-related parameters as well as the patient’s diabetic status, age, and sex. Vascular endothelial growth factor levels did not change significantly but O2•− scavenging activity and total hydroperoxide level significantly (p<0.05) increased and decreased, respectively. Both pre- and post- astaxanthin intake, vascular endothelial growth factor and total hydroperoxide levels were positively correlated (Pearson: r = 0.42, p<0.05; r = 0.55, p<0.01, respectively). Analysis of vascular endothelial growth factor levels and O2•− scavenging activities gave a negative correlation but only pre-astaxanthin intake (r = −0.37, p<0.05). Differences in levels pre- and post-astaxanthin only showed association between vascular endothelial growth factor and total hydroperoxide (r = 0.49, p<0.01) analyzed by multiple linear regression. Using multivariate analysis, pre-astaxanthin vascular endothelial growth factor level was associated with two factors of total hydroperoxide and O2•− scavenging activity (r = 0.49, p<0.05), and post-astaxanthin vascular endothelial growth factor level with two factors of total hydroperoxide and sex (r = 0.60, p<0.01). Astaxanthin intake may have affected vascular endothelial growth factor level through its antioxidant effects by increasing O2•− scavenging activity and suppressing peroxide production.
Keywords: astaxanthin, aqueous humor, vascular endothelial growth factor, oxidation, superoxide
Introduction
Oxidation reactions are involved in various pathologies including brain and heart ischemia, reperfusion injury after such ischemic events and tumor growth.(1) This has led to increased leading research on antioxidant agents. In the field of ophthalmology, oxidation has been implicated(2,3) in such pathologies as cataracts, diabetic retinopathy, uveitis and age-related macular degeneration (AMD) and the benefits of antioxidant treatments have been considered.
Lutein, a carotenoid, has been recommended as a supplement based on its reported effects in prevention of AMD.(4) Similarly, astaxanthin (AX) (Fig. 1), another kind of carotenoid, has recently attracted attention and a number of studies have focused on its potent antioxidant activity and its safety.(5,6)
Fig. 1.
Structural formula of astaxanthin.
Our laboratory has conducted research on use of an AX supplement in ophthalmology, and has reported the following findings: (i) AX intake suppressed inflammation after cataract surgery;(7) (ii) AX intake increased superoxide (O2•−) scavenging activity in the aqueous humor of diabetic patients;(8) (iii) AX intake changed hydrogen peroxide (H2O2) levels in the aqueous humor;(9) and (iv) AX intake decreased total hydroperoxide (TH) levels in the aqueous humor, indicating overall suppression of peroxidation reactions.(10) More recently, we reported on the relationship between AX administration and three parameters relevant to oxidative reactions, O2•−, H2O2, and TH.(11)
Intravitreal injections of anti-vascular endothelial growth factor (VEGF) formulations are increasingly used to treat AMD, with the goal of suppressing macular angiogenesis.(12,13) In human retinal pigment epithelial cells, VEGF mRNA levels increased in response to superoxide.(14) We hypothesized that, if AX, through its antioxidant effects, would decrease VEGF levels, it might attenuate pathologies associated with intraocular angiogenesis.
In this study, we evaluated effects of AX treatment on VEGF levels in the aqueous humor of patients, also analyzing levels of substances related to oxidation (O2•−, H2O2, and TH). Our analysis also included consideration of sex, age and diabetic status of the subjects.
Materials and Methods
The study subjects were 35 patients who underwent bilateral cataract surgery (intraocular lens implantation) at Tsukuba Hashimoto Optical Clinic after giving an informed consent based on a detailed explanation of the purpose of the study (Table 1). Patients with inflammatory diseases such as uveitis, with a high degree of refractive error of 8 diopters or above and who had been taking other supplements were excluded. The study was approved by the Bioethics Committee, Dokkyo Medical University Koshigaya Hospital (approval number: 22025).
Table 1.
Sex and diabetes status of the subjects. Error bars represent standard deviations
| Diabetic | Non diabetic | Male | Female | |
|---|---|---|---|---|
| Patients (n) | 16 | 19 | 16 | 19 |
| Average age (years) | 70.3 ± 6.2 | 71.5 ± 7.6 | 71.3 ± 6.4 | 70.6 ± 7.4 |
Patients began AX (6 mg/day) intake immediately after receiving surgery in one eye, then underwent surgery in the other eye after 2 weeks. The AX supplement used in this study was Astavita® (Fuji Chemical Industry, Toyama, Japan) derived from algae. Aqueous humor was taken from each of the eyes during surgery for analysis of O2•− scavenging activity and levels of H2O2, TH and VEGF.(15–19)
We measured VEGF levels before and after AX intake and calculated the difference between these levels, known as “change” or ΔVEGF. Relationships between these VEGF-related parameters and factors including O2•− scavenging capacity, levels of H2O2 and TH, diabetic status, age, and sex were analyzed by multivariate analysis. In the analysis, presence and absence of diabetes were scored as 1 and 0, respectively, and male and female were scored as 1 and 0, respectively.
Wilcoxon’s signed rank sum test was used for quantitative comparisons of values before and after AX treatment, stepwise multiple linear regression was used for multivariate analysis and Pearson’s correlation coefficient was used for analysis of the correlations among the factors. For each statistical test, values of p<0.05 were considered significant.
Nitro blue tetrazolium (NBT) reduction(16) was used to measure O2•− scavenging activity and a titanium colorimetric method(17) was used to measure H2O2. The NBT assay was performed with a SOD test kit, “SOD Test Wako R” (Wako Pure Chem. Ind., Ltd., Osaka, Japan). This method measures O2•− scavenging activities by various O2•− scavengers such as reduced glutathione (GSH) or l-ascorbic acid (l-AsA) and is not limited to only detecting SOD.
TH was measured by a microassay using the Free d-ROMs reagent (Diacron Srl, Grosseto, Italy).(18) N,N-diethylparaphenylenediamine, the chromogen pigment in the Free d-ROMs reagent, reacts with H2O2, lipid peroxides, peroxidized nucleic acids and nucleotides, as well as peroxides of proteins, peptides and amino acids. Thus measured TH levels indicate the total amount of these peroxidized (-OOH modified) substances.(19)
VEGF levels were measured with an enzyme-linked immunosorbent assay (ELISA) using Quantikine (R&D Systems, Minneapolis, MN) and detecting VEGF165 at a limit of detection of 5.0 pg/ml.(15)
For each parameter measured, the “change” caused by AX administration was expressed by subtracting the value measured in the aqueous humor sample collected after AX administration (from the second eye surgery) from that in the sample collected before AX administration (from the first eye surgery).
Results
Levels before and after AX intake (Fig. 2 and Table 2)
Fig. 2.
Relationship between VEGF levels in the aqueous humor before and after AX intake. VEGF levels were measured by ELISA in aqueous humor samples collected from patients before and after AX intake, plotted as shown. Each symbol represents data from one patient.
Table 2.
Measurements of each parameter in aqueous humor samples from patients before and after AX intake
| Parameters | Before AX intake | After AX intake |
|---|---|---|
| O2•− scavenging activity (U/ml) | 18.2 ± 4.1 | 19.9 ± 3.6* |
| H2O2 level (nmol/ml) | 109.8 ± 81.4 | 134.1 ± 81.1 |
| Total hydroperoxides (U CARR) | 1.16 ± 0.18 | 1.04 ± 0.31* |
| VEGF (pg/ml) | 108.7 ± 61.4 | 110.2 ± 71.2 |
Values are means ± SD. *p<0.05.
There was no statistical difference in VEGF levels in samples from subjects before and after AX intake; however, a significant positive correlation was seen between values before AX intake (r = 0.68, p<0.01). As compared with samples taken before AX intake, those taken after AX intake showed significantly higher O2•− scavenging activities (p<0.05) and lower TH levels (p<0.05).
Factors associated with VEGF levels determined by multiple linear regression analysis (Tables 3A, B and C)
Table 3.
Results of multivariate analyses of relationships between VEGF measurements (levels before and after AX intake and change) and various other parameters, as shown (multiple correlations)
| A: Factors affecting VEGF level in subjects before AX intake | ||
|---|---|---|
| Rank | Item | Standardized partial regression coefficient (β) |
| 1 | TH level before AX intake | 0.3339 |
| 2 | O2•− scavenging activity before AX intake | −0.271 |
| Multiple regression equation: (VEGF level before AX intake) = 115.3557 × (TH level before AX intake) − 4.0149 × (O2•− scavenging activity before AX intake) + 48.5403 | ||
| Multiple correlation coefficient: r = 0.49 (p = 0.0123) | ||
| B: Factors affecting VEGF level in subjects after AX intake | ||
|---|---|---|
| Rank | Item | Standardized partial regression coefficient (β) |
| 1 | TH level after AX intake | 0.6026 |
| 2 | sex | 0.2381 |
| Multiple regression equation: (VEGF level after AX intake) = 136.5143 × (TH level after AX intake) + 33.5714 × (sex) + 47.1114 | ||
| Multiple correlation coefficient: r = 0.60 (p = 0.0008) | ||
| C: Factors affecting change in VEGF levels (difference between values after and before AX intake in the same subject) | ||
|---|---|---|
| Rank | Item | Standardized partial regression coefficient (β) |
| 1 | Change in TH levels | 0.4923 |
| Multiple regression equation: (Change in VEGF level) = 96.5404 × (Change in TH level) + 12.7185 | ||
| Multiple correlation coefficient: r = 0.49 (p = 0.0027) | ||
Note: All parameters were entered into the stepwise analysis, but change in VEGF level was associated with only one factor, change in TH level, thus ended up with a simple correlation.
- Before AX intake: Two factors of 1) TH level and 2) O2•− scavenging activity (r = 0.49, p<0.05)
- After AX intake: Two factors of 1) TH level and 2) sex (r = 0.60, p<0.01)
- ΔVEGF level: One factor of 1) ΔTH (r = 0.49, p<0.01)
Correlation between VEGF levels and each factor, before and after AX intake (Fig. 3, 4 and 5, and Table 4, 5 and 6)
Fig. 3.
Relationships between VEGF levels and O2•− scavenging activities in the aqueous humor, before and after AX intake. VEGF levels and O2•− scavenging activity were measured in aqueous humor samples collected from patients before (left panel) and after (right panel) AX intake. Each symbol represents data from one patient.
Fig. 4.
Relationships between VEGF and TH levels in the aqueous humor, before and after AX intake. VEGF and TH levels were measured in aqueous humor samples collected from patients before (left panel) and after (right panel) AX intake. Each symbol represents data from one patient.
Fig. 5.
Relationship between changes (with respect to AX intake) in VEGF and TH levels in the aqueous humor. Values of each parameter after AX intake were subtracted from corresponding values before AX intake for each patient. Each resulting difference is expressed as ΔVEGF or ΔTH, respectively.
Table 4.
Relationship between VEGF level and various factors (simple correlation)
| VEGF level |
||
|---|---|---|
| Before AX intake | After AX intake | |
| O2•− scavenging activity before AX intake | r = −0.374*, p = 0.027 | r = −0.0282, p = 0.100 |
| H2O2 level before AX intake | r = 0.094, p = 0.592 | r = −0.007, p = 0.968 |
| TH level before AX intake | r = 0.417*, p = 0.013 | r = 0.554**, p = 0.0005 |
| Diabetic status | r = 0.247, p = 0.152 | r = 0.105, p = 0.547 |
| Age | r = 0.041, p = 0.814 | r = 0.228, p = 0.187 |
| Sex | r = 0.127, p = 0.469 | r = 0.116, p = 0.506 |
*p<0.05, **p<0.01
Table 5.
Relationship between ΔVEGF and various factors (simple correlation)
| Change in VEGF level | |
|---|---|
| ΔO2•− scavenging activity | r = −0.038, p = 0.830 |
| ΔH2O2 level | r = −0.026, p = 0.883 |
| ΔTH level | r = 0.492**, p = 0.0027 |
| Diabetic status | r = −0.143, p = 0.412 |
| Age | r = 0.256, p = 0.137 |
| Sex | r = 0.001, p = 0.957 |
**p<0.01
Table 6.
The number of samples that showed following changes in VEGF and TH levels after AX intake
| Change | TH level decreased | TH level increased |
|---|---|---|
| VEGF level increased | 8 | 6 |
| VEGF level decreased | 17 | 4 |
Before AX intake, the VEGF level was negatively correlated with O2•− scavenging capacity (r = −0.37, p<0.05). TH level was the only factor showing a positive correlation both before and after AX intake (before AX intake: r = 0.42, p<0.05, after AX intake: r = 0.55, p<0.01). Regarding the “change”, ΔVEGF showed a significant positive correlation only with ΔTH (r = 0.492, p<0.01). In nearly half the cases (48.6%), both VEGF and TH levels had decreased in samples taken after AX intake (Table 6).
Discussion
In diabetic retinopathy and AMD, VEGF levels increased with aggravation of symptoms(20) and many studies in ophthalmology have shown efficacy of anti-VEGF drugs in the treatment of these diseases.(12,13) These findings give an impression that VEGF is a “villain” in the field of ophthalmology, but it is also a cytokine indispensable in the human body and is produced under normal circumstances. In contrast to anti-VEGF therapy, treatments administering VEGF to patients with lower limb ischemia or ischemic heart diseases, aiming to increase circulation, have been studied.(21,22) Since VEGF is an essential cytokine, a substantial decrease would be of concern, potentially inducing infarction of the myocardium or brain. Also, the intravitreal administration of anti-VEGF agent in an attempt to suppress the development of oxygen-induced retinopathy of prematurity has been studied, but there are some concerns of systemic effect of anti-VEGF agent which escaped from the eye into the systemic circulation.(23) In this study, VEGF levels in the aqueous humor did not change after 2-week administration of AX, at 6 mg/day, as a supplement. This dose corresponds the amount of AX in, for example, a 300 g salmon fillet, within the expected range obtained through a normal daily diet.
In this study, AX intake did not cause significant quantitative changes in VEGF levels in the aqueous humor. However, VEGF levels before and after AX intake as well as ΔVEGF showed significant positive correlations with those of TH. In nearly half of the cases (48.6%), levels of both VEGF and TH decreased after AX intake, indicating that levels of TH and VEGF in the aqueous humor are likely be linked.
VEGF levels in the aqueous humor or vitreous body fluctuate according to the severity of the pathology in diabetic retinopathy,(15,20) suggesting that VEGF is secreted locally in the eye irrespective of its serum levels. VEGF, with a molecular weight of about 20,000 Da, is unlikely to cross the blood–ocular barrier readily. A significant rise in O2•− scavenging capacity and decrease in TH levels in the aqueous humor after AX intake would indicate an overall suppression of oxidative stress. In this study, VEGF levels were significantly correlated with those of TH. This was true in samples from subjects before and after AX intake and the changes of TH and VEGF were also correlated. These findings indicate that VEGF production is affected by oxidative stress. In addition, because only levels of TH, among various factors, showed a correlation with those of VEGF both before and after AX intake, we infer that overall oxidation levels played a role in VEGF production by surrounding tissues and, therefore, its levels in the aqueous humor.
TH includes lipid peroxides. In rabbit corneal parenchymal cell model, administration of lipid peroxides (peroxidized linoleic acid) induced VEGF release.(24) Similar effects are possible in the aqueous humor.
In an experimental mouse macular degeneration model induced by laser irradiation, intraperitoneal injection of AX suppressed VEGF levels in the retinal pigment epithelium (RPE).(6) In contrast, we found that VEGF levels did not decrease significantly after AX intake. This discrepancy might be explained by our use of aqueous humor, which is behind the blood-aqueous barrier, as the specimen. It is also possible that the total AX concentration, on a per tissue weight basis, was far lower in our study than in the mouse model, where up to 100 mg/kg body weight was reported. Administering such large doses of VEGF and harvesting RPE specimens for analysis would be ethically unfeasible in humans. Any studies on the effects of supplement intake in humans should use the range of recommended supplement doses. Using HPLC, we could detect a small peak corresponding to AX in the aqueous humor of patients after oral AX administration (data not shown), indicating that this dosing did result in AX being present in the aqueous humor.
Relevant to the relationship between peroxidation processes and VEGF levels, one study reported dose dependent NO release in cultured vascular endothelial cells treated with VEGF.(25) NO is believed to immediately react with O2•− to form peroxynitrite (ONOO−), resulting in loss of the vasodilating effects of NO, also known as endothelium-derived relaxing factor (EDRF). This would lead to poor blood perfusion.
In our study, before AX intake, O2•− scavenging capacity showed a negative correlation with VEGF levels, indicating that O2•− affects VEGF levels in the aqueous humor normally, even without AX intake. A relationship between O2•− and VEGF levels may exist not only in the RPE cells but also in the aqueous humor. Our multivariate analysis showed that, while TH levels were consistently and most strongly associated with VEGF levels in the aqueous humor, both before and after AX intake, O2•− scavenging activity showed the next strongest association with VEGF levels only before the AX intake. No such association was observed after AX intake. We propose that AX affects not only O2•− levels but also those of other unknown agents influencing oxidation reactions.
We showed that VEGF levels in the aqueous humor did not change significantly with AX intake, on average, in our patient population. However, VEGF levels did show a tendency to change in association with those of TH and TH levels were significantly lower with AX intake.
Based on our findings, we believe that AX led to suppression of oxidative reactions in the aqueous humor and this could have had some influence on VEGF production in surrounding tissues and the level in the aqueous humor. This implies that AX may be of value in the prevention and treatment of AMD and diabetic retinopathy through suppression of neoangiogenesis.
Abbreviations
- AMD
age-related macular degeneration
- AX
astaxanthin
- EDRF
endothelium-derived relaxing factor
- ELISA
enzyme-linked immunosorbent assay
- Gpx
glutathione peroxidase
- GSH
glutathione
- H2O2
hydrogen peroxide
- l-AsA
l-ascorbic acid
- NBT
nitro blue tetrazolium
- NO
nitric oxide
- O2•−
superoxide
- -OOH
peroxidated substances
- ONOO−
peroxynitrite
- ROMs
reactive oxygen metabolites
- RPE
retinal pigment epithelium
- SOD
superoxide dismutase
- TH
total hydroperoxide
- VEGF
vascular endothelial growth factor
Conflict of Interest
No potential conflicts of interest were disclosed.
References
- 1.Chakraborti T, Das S, Mondal M, Roychoudhury S, Chakraborti S. Oxidant, mitochondria and calcium: an overview. Cell Signal. 1999;11:77–85. doi: 10.1016/s0898-6568(98)00025-4. [DOI] [PubMed] [Google Scholar]
- 2.Ho L, van Leeuwen R, Witteman JC, et al. Reducing the genetic risk of age-related macular degeneration with dietary antioxidants, zinc, and ω-3 fatty acids: the Rotterdam study. Arch Ophthalmol. 2011;129:758–766. doi: 10.1001/archophthalmol.2011.141. [DOI] [PubMed] [Google Scholar]
- 3.Ohira A, Ueda T, Ohishi K, Hiramitsu T, Akeo K, Obara Y. Oxidative stress in ocular disease. Nippon Ganka Gakkai Zasshi. 2008;112:22–29. (in Japanese) [PubMed] [Google Scholar]
- 4.Moeller SM, Parekh N, Tinker L, et al. Associations between intermediate age-related macular degeneration and lutein and zeaxanthin in the Carotenoids in Age-Related Eye Disease Study (CAREDS): ancillary study of the Women’s Health Initiative. Arch Ophthalmol. 2006;124:1151–1162. doi: 10.1001/archopht.124.8.1151. [DOI] [PubMed] [Google Scholar]
- 5.Suzuki Y, Ohgami K, Shiratori K, et al. Suppressive effects of astaxanthin against rat endotoxin-induced uveitis by inhibiting the NF-kappaB signaling pathway. Exp Eye Res. 2006;82:275–281. doi: 10.1016/j.exer.2005.06.023. [DOI] [PubMed] [Google Scholar]
- 6.Izumi-Nagai K, Nagai N, Ohgami K, et al. Inhibition of choroidal neovasucularization with an anti-inflammatory carotenoid astaxanthin. Invest Ophthalmol Vis Sci. 2008;49:1679–1685. doi: 10.1167/iovs.07-1426. [DOI] [PubMed] [Google Scholar]
- 7.Hashimoto H, Takahashi J, Chikuda M, Obara Y. Anti-inflammatory effects of astaxanthin after cataract surgery. Atarashii Ganka. 2007;24:1357–1360. (in Japanese) [Google Scholar]
- 8.Hashimoto H, Arai K, Takahashi J, Chikuda M, Obara Y. Effect of astaxanthin consumption on superoxide scavenging activity in aqueous humor. Atarashii Ganka. 2009;26:229–234. (in Japanese) [Google Scholar]
- 9.Hashimoto H, Arai K, Hayashi S, Takahashi J, Chikuda M, Obara Y. Effect of astaxanthin consumption on superoxide scavenging activity and hydrogen peroxide in the human aqueous humor. Folia Japonica de Ophthalmologica Clinica. 2012;5:119–126. (in Japanese) [Google Scholar]
- 10.Hashimoto H, Arai K, Okamoto H, Takahashi J, Chikuda M, Obara Y. Effect of astaxanthin consumption on hydroperoxides in the aqueous. Jpn J Ophthalmol. 2011;65:465–470. (in Japanese) [Google Scholar]
- 11.Hashimoto H, Arai K, Hayashi S, et al. Effects of astaxanthin on antioxidation in human aqueous humor. J Clin Biochem Nutr. 2013;53:1–7. doi: 10.3164/jcbn.13-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med. 2006;355:1419–1431. doi: 10.1056/NEJMoa054481. [DOI] [PubMed] [Google Scholar]
- 13.Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versus verteporfin for neovascular. Age-related macular degeneration. N Engl J Med. 2006;355:1432–1444. doi: 10.1056/NEJMoa062655. [DOI] [PubMed] [Google Scholar]
- 14.Kuroki M, Voest EE, Amano S, et al. Reactive oxygen intermediates increase vascular endothelial growth factor expression in vitro and in vivo. J Clin Invest. 1996;98:1667–1675. doi: 10.1172/JCI118962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hashimoto H, Arai K, Chikuda M, Obara Y. Relationship between vascular endothelial growth factors and advanced glycation end products in the human vitreous. Nippon Ganka Gakkai Zasshi. 1998;102:442–446. [PubMed] [Google Scholar]
- 16.McCord JM, Fridovich I. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein) J Biol Chem. 1969;244:6049–6055. [PubMed] [Google Scholar]
- 17.Winterbourn CC, Garcia RC, Segal AW. Production of the superoxide adduct of myeloperoxidase (compound III) by stimulated human neutrophils and its reactivity with hydrogen peroxide and chloride. Biochem J. 1985;228:583–592. doi: 10.1042/bj2280583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cesarone MR, Belcaro G, Carratelli M, et al. A simple test to monitor oxidative stress. Int Angiol. 1999;18:127–130. [PubMed] [Google Scholar]
- 19.Alberti A, Bolognini L, Macciantelli D, Caratelli M. The radical cation of N,N-diethyl-para-phenylendiamine: a possible indicator of oxidative stress in biological samples. Res Chem Intermed. 2000;26:253–267. [Google Scholar]
- 20.Aiello LP, Avery RL, Arrigg PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med. 1994;331:1480–1487. doi: 10.1056/NEJM199412013312203. [DOI] [PubMed] [Google Scholar]
- 21.Baumgartner I, Pieczek A, Manor O, et al. Constitutive expression of phVEGF 165 after intramuscular gene transfer promotes collateral vessel development in patients with critical limb ischemia. Circulation. 1998;97:1114–1123. doi: 10.1161/01.cir.97.12.1114. [DOI] [PubMed] [Google Scholar]
- 22.Rosengart TK, Lee LY, Patel SR, et al. Angiogenesis gene therapy: phase I assessment of direct intramyocardial administration of an adenovirus vector expressing VEGF 121 cDNA to individuals with clinically significant severe coronary artery disease. Circulation. 1999;100:468–474. doi: 10.1161/01.cir.100.5.468. [DOI] [PubMed] [Google Scholar]
- 23.Matsubara M, Saito Y, Nakanishi-Ueda T, et al. Influence of the difference of breastfeeding volume on a rat model of oxygen-induced retinopathy. J Clin Biochem Nutr. 2014;55:129–134. doi: 10.3164/jcbn.14-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ueda T, Nakanishi-Ueda T, Fukuda S, et al. Lipid hydroperoxide-induced tumor necrosis factor (TNF)-α, vascular endothelial growth factor and neovascularization in the rabbit cornea: effect of TNF inhibition. Angiogenesis. 1997;1:174–184. doi: 10.1023/A:1018377621102. [DOI] [PubMed] [Google Scholar]
- 25.van der Zee R, Murohara T, Luo Z, et al. Vascular endothelial growth factor/vascular permeability factor augments nitric oxide release from quiescent rabbit and human vascular endothelium. Circulation. 1997;95:1030–1037. doi: 10.1161/01.cir.95.4.1030. [DOI] [PubMed] [Google Scholar]