ORIGINAL ARTICLE

Oxidative Stress and Protective Effect of Erythropoietin

on Methotrexate-induced Esophageal Damage
Mesut Garipardic, MD,* Vedat Bakan, MD,w Mehmet Davutoğlu, MD,z
Hamide Sayar, MD,y and Ergül Belge Kurutaş, MDJ

Esophageal mucositis is a common toxic side effect of MTX

Introduction: An experimental study was conducted to investigate in childhood period and can be dose limiting in therapeutic
the role of oxidative stress and effects of erythropoietin (EPO) on regimens, and treatment reduction or withdrawal.3,4
methotrexate-induced esophageal damage. Free oxygen radicals or reactive oxygen species (ROS)
Materials and Methods: Twenty-four female Sprague–Dawley rats play an important role in the acute stage of esophageal burn
were equally divided into 3 groups: Sham operation animals (group and reflux esophagitis.5–7 ROS, in particular superoxide
S) were administered subcutaneous injections of 0.2 mL of 0.9% anion (O2 ), play an important role in the pathogenesis
NaCl; control animals (group MTX) were administered subcuta- of esophagitis induced by acid and pepsin.7 The role of
neous injections of methotrexate (5 mg/kg) and EPO-treated oxidative stress in MTX-induced small intestinal toxicity
animals (group EPO) were administered subcutaneous injections has been researched and found that animals that received
of methotrexate (5 mg/kg) and EPO (2000 IU/kg) once daily for
N-acetylcysteine or tungsten showed a decreased produc-
4 consecutive days. At the fifth day, the distal 1.5-cm esophageal
segments were harvested for biochemical and histologic investiga- tion of ROS and decreased myeloperoxidase activity.8
tions. Oxidative damage was assessed by measuring the levels Oxidative stress in the esophageal mucosa caused a
of malondialdehyde and nitric oxide, and activities of superoxide reduction in tissue concentrations of antioxidant such as
dismutase and catalase in homogenized samples of esophageal glutathione in a rat esophagitis model.9
tissue. Histologic damage to esophageal tissue was scored and total Erythropoietin (EPO) is a hypoxia-inducible hemato-
tissue damage scores were calculated. poietic growth factor, which has antiapoptotic, antioxidant,
Results: Malondialdehyde levels in the S and EPO groups were antiinflammatory, angiogenic, and neuroprotective effects.10–13
significantly lower than those in the MTX group (P<0.05). In addition to its traditional hematopoietic role, EPO’s
Catalase and superoxide dismutase activities, and nitric oxide levels cytoprotective effects have been highlighted in recent
in the S and EPO groups were significantly higher than those in the studies.11–13 The role of ROS and the effect of EPO on
MTX group (P<0.05). Esophageal tissue damage was significantly MTX-induced esophageal damage has not been investi-
less in the EPO group than that in the MTX group (P<0.05). gated earlierand further research into new therapeutic
Conclusions: Free radicals elevate in methotrexate given rats’ avenues for treatment or prevention of these undesirable
esophagus and EPO has significant preventive effects on metho- side affects is required. Therefore this study, for the first
trexate-induced oxidative damage of esophagus in a rat model. time to our knowledge, was designed to evaluate the role of
free radicals and the protective role of EPO on MTX
Key Words: methotrexate, esophagus, oxidative damage, erythro- induced-esophageal damage.
poietin
(J Pediatr Hematol Oncol 2010;32:108–112)
MATERIALS AND METHODS
Animals
T he gastrointestinal system is highly sensitive to antic-
ancer drugs owing to a major cellular renewal and fast
protein turnover in the mucosa.1–3 Alimentary mucositis is
This study was approved by the Ethics Committee on
Animal Research at our Faculty of Medicine. Twenty-four
female Sprague-Dawley rats weighing 220 to 240 g each
a significant complication of high-dose chemotherapy in were selected for the study and acclimatized for 10 days in
cancer patients, which may cause treatment reduction or the animal laboratory of our university research center,
withdrawal. Methotrexate (MTX), a folic acid antagonist, receiving a standard diet ad lib. The death of an animal
is widely used as a cytotoxic chemotherapeutic agent for was accepted as an exclusion criterion and new animals
several malignancies and various inflammatory diseases. receiving the same protocol were added instead of dieing
animals. The rats were divided into 3 groups: Group S
Received for publication September 12, 2009; accepted November 25,
(sham operation, n = 8) animals were administered sub-
2009. cutaneous injections of 0.2 mL of 0.9% NaCl once daily
From the Departments of *Pediatric Hematology and Oncology; for 4 consecutive days. Group MTX (Control, n = 8)
wPediatric Surgery; zPediatrics; yPathology; and JBiochemistry, animals were administered subcutaneous injections of
Sütcü İmam University, Kahramanmaraş, Turkey.
This study was presented at the 27th Annual Meeting of the Turkish
MTX (5 mg/kg) once daily for 4 consecutive days and Group
Association of Pediatric Surgeons, Malatya, Turkey, September EPO (EPO-treated, n = 8) animals were administered sub-
30–October 4, 2009. cutaneous injections of MTX (5 mg/kg) and EPO (2000 IU/
Reprints: Mesut Garipardic, MD, Department of Pediatrics, Faculty of kg, Recormon, Roche Diagnostics GmbH, Mannheim,
Medicine, Kahramanmaras Sutcu Imam University, Kahramanmaras
(TR-46050), Turkey (e-mail: mgaripardic@hotmail.com; mgaripardic@
Germany) once daily for 4 consecutive days. At the fifth
mynet.com). day, animals in all groups were sacrificed by decapitation
Copyright r 2010 by Lippincott Williams & Wilkins and the distal 1.5-cm esophageal segments were harvested

108 | www.jpho-online.com J Pediatr Hematol Oncol  Volume 32, Number 2, March 2010
J Pediatr Hematol Oncol  Volume 32, Number 2, March 2010 Oxidative Stress and Protective Effect of Erythropoietin

for biochemical and histologic investigations. The two-third were significantly different between Groups S, MTX, and
portion of the harvested esophagus was stored at  801C EPO (P<0.001). MDA levels were significantly higher in
until biochemical assays and the remaining one-third was group MTX (1.37 ± 0.41 nmol/mg protein) compared with
fixed in formalin for histopathologic evaluation. the Sham (0.65 ± 0.15 nmol/mg protein) and EPO groups
(0.73 ± 0.08 nmol/mg protein) (P<0.05). CAT activities in
Biochemical Evaluation the group MTX were significantly lower than those sham
Oxidative stress was evaluated by the measurement of and EPO groups (P<0.05). NO levels (1.10 ± 0.73 mmol/mg
the levels of malondialdehyde (MDA), nitric oxide (NO), protein) were significantly lower in the MTX group
and activities of superoxide dismutase (SOD) and catalase compared with the sham (2.88 ± 0.36 mmol/mg protein)
(CAT). Tissues were weighed, placed in 1.15% KCl and EPO (2.03 ± 0.77 mmol/mg protein) groups (P<0.05).
solution, and homogenized for 30 minutes of 14,000 speeds. NO levels also were significantly lower in the EPO group
Aliquots of the homogenate were then centrifugated at compared with the sham group. Tissue MDA levels and
10,000 rpm for 30 minutes and the supernatants were CAT and SOD activities in the S and EPO groups were not
analyzed for MDA and protein levels, CAT, and SOD significantly different from each other (P>0.05, Fig. 1).
activities. The concentration of plasma lipid peroxidation Sections of the esophagus from the sham operation
(total MDA, expressed in nanomoles per mg protein) was group were normal in histopathologic appearance (Fig. 2).
determined by the Ohkawa method with slight modifica- In all rats of group MTX, histopathologic examination of
tions.14 Protein was measured according to the Lowry the esophagus showed edema, epithelial thinning, and
method15 SOD activity was determined as described by hemorrhage, containing 1 or more of them (Fig. 3). Only
Fridovich.16 This method uses xanthine and xanthine oxidase 1 animal in EPO group showed epithelial thinning (Fig. 4).
to generate superoxide radicals that react with 2-(4-iodo- Esophageal tissue damage scores, of the EPO group
phenyl)-3-(4-nitro phonol-s-phenyl tetrazolium chloride) to (0.12 ± 0.35) were lower than those of the MTX group
form a red formazon dye. SOD activity is then measured by (1.12 ± 0.83) (P<0.05).
the degree of inhibition of this reaction. CAT activity was
spectrophotometrically measured by the disappearance of
H2O2 at 230 nm.17 The determination of nitrite, the stable DISCUSSION
end product of NO radicals, is most often used as a measure In this study, we showed the participation of free
of NO production.18 NO measurement was carried out radicals in the esophageal injury secondary to subcuta-
using the Griess method for detection of nitrite levels.19 neous MTX administration. We also established the
beneficial effects of exogenous EPO in the prevention of
Histopathologic Evaluation tissue damage, which was observed by a reduction in
Esophageal tissues fixed in 10% neutral buffered oxidative stress, an increase in antioxidant enzymes, and
formalin solution were embedded in paraffin. Serial sections less histopathologic damage. This result is in agreement
were cut at a thickness of 5 mm, stained with hematoxylin- with earlier experimental models showing that oxidative
eosin, and examined by light microscopy for the presence of stress plays an important role in MTX-induced small
tissue damage. Examination and scoring of the esophageal intestinal damage8 and antioxidants such as vitamin E,
sections was carried out in a blinded fashion by the same vitamin A, garlic extract, N-acetyl cysteine, and sodium
pathologist. Five microscopy fields were evaluated for the tungstate attenuated side effects of MTX.20
presence of congestion, hemorrhage, elongation of the lamina EPO is a strong antioxidant11–13 and increases the
propriety papillae, basal hyperplasia, inflammation, erosion activity of antioxidant enzymes, such as SOD, CAT, and
or ulcer, necrosis, and epithelial thinning. Each characteristic glutathione peroxidase, and decreases lipid peroxidation
was scored as 1 point, the total tissue damage scores were levels in hypoxic-ischemic brain injury21 and in cultured
calculated by adding the scores for each characteristic. mouse astrocytes.22 In this study, subcutaneous adminis-
Statistical Analysis tration of EPO to MTX received rats increased the level of
esophageal antioxidant enzymes including SOD and CAT,
Individual group biochemical parameters were as-
and decreased the level of MDA, 1 of the lipid peroxidation
sessed with 1-sample Kolmogorov-Smirnov Z test and
products, when compared with control group. We also
found normal (P>0.05). Analysis of variance (ANOVA)
found that histopathologic changes in esophagus of MTX
was carried out on the biochemical data to examine
given rats were more serious than those MTX and EPO
differences among groups. When a significant group effect
given rats. We believed that these findings were related to
was found, a Tukey HSD test was used to identify the
its antioxidant and antiinflammatory effects.
location of differences between the groups. Statistical NO is an important signalling molecule, generated
significance was defined as P<0.05. Tissue damage scores in most cells by nitric oxide synthase enzyme. It has
were compared by nonparametric analysis, and statistical vasodilator, antioxidant, antiplatelet, and antineutrophil
significance was determined by Kruskal-Wallis followed by actions and is essential for normal cellular function.23–26
Bonferroni corrected Mann-Whitney U test. The data were NO has both damaging and protective effects,25,26 so it
expressed as means ± SD. can be considered as a 2-edged sword. NO reacts rapidly
with O2 to form peroxynitrite when both are produced
RESULTS simultaneously and excess NO could be detrimental
All but 6 animals survived till the end of the if it combines with superoxide anion (O2 ).26,27 Peroxy-
experiment. Four animals in Group C and 2 in Group nitrite is thought to be a key mediator of NO-mediated
EPO died owing to congestion and intraalveolar hemor- tissue injury.26,27 SOD eliminates O2 physiologically, but
rhage of the lung on the third day of the experiment. excess production of O2 in pathologic conditions such
Instead of dead rats, 6 rats received the same protocol and as ischemia-reperfusion syndromes causes the formation of
were added to the study. All biochemical parameters peroxynitrite.6,24,26–28 Inhibition of SOD activity leads to

r 2010 Lippincott Williams & Wilkins www.jpho-online.com | 109

Garipardic et al J Pediatr Hematol Oncol  Volume 32, Number 2, March 2010

FIGURE 1. Levels of malondialdehyde (MDA) and nitric oxide (NO), and superoxide dismutase (SOD) and catalase (CAT) activities in all
groups. MDA, NO, levels, and SOD, CAT activities in methotrexate (MTX) group were significant different than those of the S and EPO
groups. *P < 0.05 versus sham and EPO. EPO indicates MTX given and erythropoietin-treated group; MTX, methotrexate given group;
S, shame operated group.

elevation in hypoxia-induced EPO gene expression.29 Single-

dose EPO pretreatment inhibits NO mediated free-radical
formation in rat liver,30 and reduces oxidative stress after
ischemic injury.31 In our study, esophageal NO levels were
found to be lower in control groups, which might have
occurred as a result of reaction between NO and superoxide
anion, as elevated levels of superoxide anion would be
predicted to decrease NO concentrations.32 The high levels of
SOD and NO in the EPO treatment group than the control
might be linked to the antioxidant effect of EPO.
The pathogenesis of mucositis can be attributed to
the direct mucosal toxicity of high-dose chemotherapy and
ionizing radiation and to indirect mucosal damage caused
by concomitant local bacterial, viral, and mycotic infec-
tions.33,34 Owing to mucosal desquamations, esophageal
mucosal barriers are exposed to injury by proinflammatory
factors, such as inflammatory cytokines, leukocytes, and
oxidative stress. The evidence specifically indicates that
FIGURE 2. Normal histologic appearance of rat esophagus in the inflammatory response to radiation and/or chemo-
sham operated (S) group (H&E, original magnification  40). therapy plays an important role in the pathogenesis of

110 | www.jpho-online.com r 2010 Lippincott Williams & Wilkins

J Pediatr Hematol Oncol  Volume 32, Number 2, March 2010 Oxidative Stress and Protective Effect of Erythropoietin

development of true esophagitis and further studies are

needed to create an animal esophagitis model.
In conclusion, we found that ROS elevated in
esophageal tissue of MTX received rats and EPO had
significant preventive effects on oxidative damage, histo-
pathologically, and biochemically, in the early stage of
MTX-induced esophageal damage in a rat model. We
speculate that it may be possible to reduce MTX-induced
esophageal damage by adding EPO to the treatment proto-
col as an antioxidant agent. Further additional investiga-
tions using EPO and other antioxidants are needed to
confirm our results. Future studies should evaluate the dose
ranges, intervals and administration duration of EPO, and
other easily available antioxidants.

ACKNOWLEDGMENTS
The authors thank Prof Dr Yakup Gümüşalan, MD,
FIGURE 3. A, Edema (white arrow) and epithelial thinning Sutcu Imam University, Faculty of Medicine, Department of
(between two black arrows) (H&E, original magnification Anatomy and Assoc. Prof Dr Gökhan Özdemir, MD, Sutcu
 40). B, Mild congestion (short arrows) and hemorrhage (long Imam University, Faculty of Medicine, Department of
arrow) (H&E, original magnification  100) in methotrexate Ophthalmology for English editing and grammar control.
given group.

mucositis.3,33,34 In animal models of radiation35 and MTX- REFERENCES

induced mucositis,36 the antiinflammatory cytokine inter- 1. Radtke F, Clevers H. Self-renewal and cancer of the gut: two
leukin-11 showed a significant reduction in mucosal injury. sides of a coin. Science. 2005;307:1904–1909.
In our study, too little damage such as edema, epithelial 2. Coeffier M, Claeyssens S, Hecketsweiler B, et al. Enteral
thinning, and luminal narrowing was observed in the glutamine stimulates protein synthesis and decreases ubiquitin
mucosal epithelium in microscopic examination, however, mRNA level in human gut mucosa. Am J Physiol Gastrointest
there was no leukocyte infiltration. These minimal changes Liver Physiol. 2003;285:G266–G273.
might be owing to direct mucosal toxicity of chemotherapy 3. Lalla RV, Schubert MM, Bensadoun RJ, et al. Anti-inflamma-
tory agents in the management of alimentary mucositis. Support
or ROS in the early phase of experiment during initial 4
Care Cancer. 2006;14:558–565.
days. A longer experiment period may be necessary for the 4. Leblond J, Le Pessot F, Hubert-Buron A, et al. Chemotherapy-
induced mucositis is associated with changes in proteolytic
pathways. Exp Biol Med (Maywood). 2008;233:219–228.
5. Günel E, Cağlayan F, Cağlayan O, et al. Reactive oxygen
radical levels in caustic esophageal burns. J Pediatr Surg. 1999;
34:405–407.
6. Ishiyama F, Iijima K, Asanuma K, et al. Exogenous luminal
nitric oxide exacerbates esophagus tissue damage in a reflux
esophagitis model of rats. Scand J Gastroenterol. 2009;44:
527–537.
7. Lanas A, Soteras F, Jimenez P, et al. Superoxide anion and
nitric oxide in high-grade esophagitis induced by acid and
pepsin in rabbits. Dig Dis Sci. 2001;46:2733–2743.
8. Miyazono Y, Gao F, Horie T. Oxidative stress contributes to
methotrexate-induced small intestinal toxicity in rats. Scand
J Gastroenterol. 2004;39:1119–1127.
9. Wetscher GJ, Perdikis G, Kretchmar DH, et al. Esophagitis in
Sprague-Dawley rats is mediated by free radicals. Dig Dis Sci.
1995;40:1297–1305.
10. Santhanam AV, Katusic ZS. Erythropoietin and cerebral
vascular protection: role of nitric oxide. Acta Pharmacol Sin.
2006;27:1389–1394.
11. Jelkmann W. Erythropoietin after a century of research:
younger than ever. Eur J Haematol. 2007;78:183–205.
12. Brines M, Cerami A. Discovering erythropoietin’s extra-
hematopoietic functions: biology and clinical promise. Kidney
Int. 2006;70:246–250.
13. Mori S, Sawada T, Okada T, et al. Erythropoietin and its
derivative protect the intestine from severe ischemia/reper-
fusion injury in the rat. Surgery. 2008;143:556–565.
14. Ohkawa H, Ohishi N, Yagi K. Assay for lipid peroxides in
animal tissues by thiobarbituric acid reaction. Anal Biochem.
1979;95:351–358.
FIGURE 4. Epithelial thinning in erythropoietin-treated group 15. Lowry OH, Rosebrough NJ, Farr A. Protein measurement
(H&E, original magnification  100). with the folin phenol reagent. J Biol Chem. 1951;193:265–275.

r 2010 Lippincott Williams & Wilkins www.jpho-online.com | 111

Garipardic et al J Pediatr Hematol Oncol  Volume 32, Number 2, March 2010

16. Fridovich I. Superoxide dismutase. Adv Enzymol. 1974;41: 27. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and
35–97. peroxynitrite: the good, the bad, and ugly. Am J Physiol.
17. Beutler E. Red Cell Metabolism. 2nd ed. New York: Grune and 1996;271:C1424–C1437.
Stratton Company; 1975:261–265. 28. Demchenko IT, Oury TD, Crapo JD, et al. Regulation of
18. Moshage H, Kok B, Huizenga JR, et al. Nitrite and nitrate the brain’s vascular responses to oxygen. Circ Res. 2002;91:
determinations in plasma: a critical evaluation. Clin Chem. 1031–1037.
1995;41:892–896. 29. Zelko IN, Folz RJ. Extracellular superoxide dismutase functions
19. Cortas NK, Wakid NW. Determination of inorganic nitratein as a major repressor of hypoxia-induced erythropoietin gene
serum and urine by akinetic cadmium reduction method. Clin expression. Endocrinology. 2005;146:332–340.
Chem. 1990;36:1440–1443. 30. Joyeux-Faure M, Ramond A, Béguin PC, et al. Early
20. Yüncü M, Eralp A, Celik A. Effect of aged garlic extract pharmacological preconditioning by erythropoietin mediated
against methotrexate-induced damage to the small intestine in by inducible NOS and mitochondrial ATP-dependent potas-
rats. Phytother Res. 2006;20:504–510. sium channels in the rat heart. Fundam Clin Pharmacol. 2006;
21. Kumral A, Gonenc S, Acikgoz O, et al. Erythropoietin 20:51–56.
increases glutathione peroxidase enzyme activity and decreases 31. Sakanaka M, Wen TC, Matsuda S, et al. In vivo evidence that
lipid peroxidation levels in hypoxicischemic brain injury in erythropoietin protects neurons from ischemic damage. Proc
neonatal rats. Biol Neonate. 2005;87:15–18. Natl Acad Sci USA. 1998;95:4635–4640.
22. Nagai A, Nakagawa E, Choi HB, et al. Erythropoietin and 32. Marklund SL. Extracellular superoxide dismutase and other
erythropoietin receptors in human CNS neurons, astrocytes, superoxide dismutase isoenzymes in tissues from nine mam-
microglia, and oligodendrocytes grown in culture. J Neuro- malian species. Biochem J. 1984;222:649–655.
pathol Exp Neurol. 2001;60:386–392. 33. Demarosi F, Bez C, Carrassi A. Prevention and treatment of
23. van der Vliet A, Eiserich JP, Halliwell B, et al. Formation of chemo- and radiotherapy-induced oral mucositis. Minerva
reactive nitrogen species during peroxidase-catalyzed oxidation Stomatol. 2002;51:173–186.
of nitrite. A potential additional mechanism of nitric oxide- 34. Köstler WJ, Hejna M, Wenzel C, et al. Oral mucositis
dependent toxicity. J Biol Chem. 1997;272:7617–7625. complicating chemotherapy and/or radiotherapy: options
24. Ferdinandy P. Peroxynitrite: just an oxidative/nitrosative for prevention and treatment. CA Cancer J Clin. 2001;51:
stressor or a physiological regulator as well? Br J Pharmacol. 290–315.
2006;148:1–3. 35. Sonis ST, Peterson RL, Edwards LJ, et al. Defining mechanisms
25. Wink DA, Feelisch M, Vodovotz Y. The chemical biology of of action of interleukin-11 on the progression of radiation-
nitric oxide. In: Colton CA, Gilbert DL, eds. Reactive Oxygen induced oral mucositis in hamsters. Oral Oncol. 2000;36:
Species in Biological Systems. New York: Kluwer Academic/ 373–381.
Plenum Publishing; 1999:245–291. 36. Gibson RJ, Keefe DM, Thompson FM, et al. Effect of
26. Ortega Mateo A, Amaya Aleixandre de Artiñano AA. Nitric interleukin-11 on ameliorating intestinal damage after metho-
oxide reactivity and mechanisms involved in its biological trexate treatment of breast cancer in rats. Dig Dis Sci.
effects. Pharmacol Res. 2000;42:421–427. 2002;47:2751–2757.

112 | www.jpho-online.com r 2010 Lippincott Williams & Wilkins