Molecular and Histopathological Study on the Ameliorative Impacts of Petroselinum Crispum and Apium Graveolens against Experimental Hyperuricemia

Hyperuricemia is an abnormal metabolic condition characterized by an increase in uric acid levels in the blood. It is the cause of gout, manifested by inflammatory arthritis, pain and disability. This study examined the possible ameliorative impacts of parsley (PAR) and celery (CEL) as hypouricemic agents at biochemical, molecular and cellular levels. PAR and CEL alone or in combination were orally administered to hyperuricemic (HU) mice and control mice for 10 consecutive days. Serum levels of uric acid and blood urea nitrogen (BUN), xanthine oxidase activity, antioxidants, inflammatory (IL-1β and TNF-α) and anti-inflammatory cytokines (IL-10) were measured. mRNA expression of urate transporters and uric acid excretion genes in renal tissues were examined using qRT-PCR (quantitative real time PCR). Normal histology and immunoreactivity of transforming growth factor-beta 1 (TGF-β1) in kidneys was examined. Administration of PAR and CEL significantly reduced serum BUN and uric acids in HU mice, ameliorated changes in malondialdehyde, catalase, and reduced glutathione, glutathione peroxidase (GPX), IL-1β, TNF-α and IL-10 in hyperuricemic mice. Both effectively normalized the alterations in mURAT-1, mGLUT-9, mOAT-1 and mOAT-3 expression, as well as changes in TGF-β1 immunoreactivity. Interestingly, combined administration of PAR and CEL mitigated all examined measurements synergistically, and improved renal dysfunction in the hyperuricemic mice. The study concluded that PAR and CEL can potentially reduce damaging cellular, molecular and biochemical effects of hyperuricemia both individually and in combination.

Xanthine Oxidase activity. The kit used depends on the catalysis of hypoxanthine to form xanthine and superoxide anion free radicals. In the presence of chromogenic agent and electronic receptors, it will form a purplish-red substance that can be measured at the OD value of 530 nm. For liver tissues, homogenate in 1:9 normal saline was placed on ice, centrifuged for 10 minutes and the supernatant used for XO assay. The measurement unit for serum is U/l and for liver is U/g protein tissue. The protocol used for XO is a partially modified version of the method used by Haidari et al. 31 Histological and immunohistochemistry analyses of kidney tissue. Kidney slices were dehydrated, embedded in paraffin and sectioned at 4 µm. Slides were stained with hematoxylin and eosin (H&E). Morphological changes were examined using a microscope (Nikon Eclipse 80i, Japan) and images were captured with a digital camera (Canon, SX620 HS -20 MP, Japan). For immunohistochemistry, paraffin-embedded renal sections were deparaffinized, rehydrated and immersed in 2% H 2 O 2 for 15 minutes (to inhibit peroxidase activity). Sections were then washed in phosphate buffer saline. Bovine serum albumin (5%) was used to block nonspecific binding sites. TGF-β1 polyclonal antibody was added for kidney slides in a dilution of 1:350 overnight at 4 °C. Slides were then incubated with secondary antibody after washing in PBS, developed with 3,3-diaminobezidine tetrahydrochloride then counterstained with hematoxylin. The percentage of positive immunoreactive cells for TGF-β1 was shown as the ratio of positively stained cells to the total cell number in the same field. Statistical analysis. Data are expressed as means ± standard error for values collected from 7 mice per group. One-way ANOVA was used to analyze the data. The probability level P < 0.05 and the individual comparisons were obtained using Duncan's multiple range tests for SPSS software version 12.5 for Windows (SPSS, IBM, Chicago, IL, USA). P < 0.05 was considered statistically significant.
Ethical statement. All procedures used in this study were carried out based on the NIH Guide for the care and use of laboratory animals. All precautions were followed to minimize animal suffering throughout the experiments.

Results
Administration of parsley and celery ameliorated liver and kidney dysfunction in hyperuricemic mice. Administration of PO increased serum levels of GOT, GPT, uric acid and BUN in hyperuricemic (HU) mice compared to normal control mice. ALP, PAR and CEL treated HU mice exhibited decreased altered parameters (Table 2). Co-administration of PAR and CEL showed a greater ameliorative effect (P < 0.05) on normalization of assayed parameters (Table 2) compared with PAR or CEL alone in HU mice.
Administration of parsley and celery decreased serum and hepatic Xanthine Oxidase (XO) activity in hyperuricemic mice. As shown in Table 3, serum and liver activities of XO were increased in HU mice and were normalized significantly (P < 0.05) in the ALP, PAR and CEL administered group compared to Administration of parsley and celery ameliorated disorders in cytokine levels in hyperuricemic mice. Table 4 shows the changes in serum levels of IL-1β and TNF-α inflammatory cytokines, and IL-10 anti-inflammatory cytokine. Hyperuricemia induced a state of inflammation and significantly increased IL-1β and TNF-α levels (P < 0.05), while decreasing serum levels of IL-10. PAR and CEL administration ameliorated these effects. Co-administration of PAR and CEL induced a greater (P < 0.05) inhibitory effect on IL-1β and TNF-α, and a stimulatory effect on secretion of IL-10 (Table 4).
Antioxidant activities of parsley and celery against oxidative stress associated with hyperuricemia in mice. Hyperuricemia increased tissue degradation by increasing MDA levels in the HU group    www.nature.com/scientificreports www.nature.com/scientificreports/ (Table 5). These increases in MDA were normalized by PAR and CEL treatment. Hyperuricemia decreased catalase, and GSH and GPX levels but these returned to nearly control levels after PAR and CEL administration ( Table 5). Co-administration of PAR and CEL to HU mice induced an additive ameliorative effect on the changes induced in measured antioxidants ( Table 5).
Impacts of PAR and CEL on mRNA expression of genes associated with renal hyperuricemia. We examined mRNA expression of mOAT-1, mOAT-3, mURTA-1 and mGlut9 genes responsible for urate excretion and reabsorption in the kidney. Figure 1 shows significant oxonate down-regulation in mRNA expression of mOAT-1 and mOAT-3, and significant (p < 0.05) up-regulation in mURAT-1 and mGlut-9 mRNA expression in HU mice kidneys compared with the control group. The alteration in mRNA expression of urate transporter-related genes was consistent with the elevation of serum uric acid and BUN levels reported in Table 2. PAR and CEL administration alone showed significant down-regulation in mURAT-1 and mGlut-9 and up-regulation in mOAT-1 and mOAT-3 mRNA expression (Figs. 1 and 2). The additive synergistic effect on altered genes was clear when both PAR and CEL were co-administered for the HU group.
Impacts of PAR and CEL on mRNA expression of liver genes associated with uric acid metabolism. We examined mRNA expression of mice PNP and mice guanine Gda genes responsible for uric acid metabolism in the liver. As shown in Fig. 3, oxonate administration induced significant up-regulation in mPNP and mGda mRNA expression in HU mice (p < 0.05) compared to the control group. PAR and CEL regulated the alteration reported in HU groups. There was an additive synergistic effect for PAR and CEL when administered together to HU mice (Fig. 3).   www.nature.com/scientificreports www.nature.com/scientificreports/ Impacts of parsley and celery on renal histology and TGF-β1 immunoreactivity in hyperuricemic mice. Kidney histology. Control mice kidneys showed normal glomerular and tubular structure (Fig. 4A), whereas HU group kidneys showed a dense eosinophilic mass occluding the tubular lumina as well as leukocytic infiltration (Fig. 4B). Shrunken glomerular tufts, and periglomerular and interstitial (*) round cells infiltration were also observed. HU group kidneys treated with allopurinol showed normal glomerular architecture with normal tubular histology (Fig. 4C). Kidneys of parsley administered mice showed normal renal tissue with a normal tubular and glomerular picture (Fig. 4D). Kidneys of celery administered mice showed the normal histological picture of both glomerular and tubular sections (Fig. 4E). Kidneys of the HU group treated with parsley alone showed restoration of the normal picture with mild perivascular round cells infiltration (Fig. 4F). Kidneys of the HU group treated with celery showed restoration of glomerular and tubular tissue histology (Fig. 4G). Kidneys of the HU group treated with celery and parsley showed a normal histological picture of both glomerular and tubular tissue (Fig. 4H).  www.nature.com/scientificreports www.nature.com/scientificreports/ Immunoreactivity of renal TGF-β1. Kidneys of the control group showed an absence of TGF-β1 expression in renal tissue (Fig. 5A). Kidneys of the HU group showed high intensity and immunoreactivity for TGF-β1 in renal tubular tissue (Fig. 5B). Kidneys of the HU group treated with allopurinol showed no marked expression of TGF-β1 in renal tissue (Fig. 5C). Kidneys of parsley administered HU mice showed an absence of expression of TGF-β1 in renal tubular tissue (Fig. 5D). Kidneys of the celery group showed an absence of TGF-β1 expression in tubular tissue (Fig. 5E). Kidneys of the HU group treated with parsley alone showed no observed reactivity for TGF-β1 in renal tissue (Fig. 5F). Kidneys of the HU group treated with celery showed glomerular and tubular tissue with no TGF-β1 expression (Fig. 5G). Kidneys of the HU group treated with celery and parsley together showed more restoration in renal cells without expression of TGF-β1 (Fig. 5H). Table 6, shows particularly high intensity scores for TGF-β1 expression in the PAR and CEL administered HU groups. www.nature.com/scientificreports www.nature.com/scientificreports/

Discussion
The results suggested that parsley and celery are safe herbal remedies that can be used either alone or in combination 30,31 to lower the effects on serum levels of uric acid and xanthine oxidase activity in hyperuricemic mice. Hyperuricemia increases the production of oxygen free radicals, induces lipid peroxidation, and up-regulates inflammatory and down-regulates anti-inflammatory cytokine expression and secretion 32,34,35 . Herbal plants can increase antioxidant content in experimental animals and rodents 24,36 . The major functions of most flavonoids present in medicinal plants are their ability to scavenge free radicals and increase antioxidant activities 37,38 . As shown in previous reports, medicinal plants increased total antioxidant capacity, suppressed reactive oxygen species (ROS) and prevented damage induced by oxidative stress 39,40 . Here, parsley and celery showed the potency to improve and increase antioxidant activities, eliminate tissue destruction and reduce inflammatory effects of hyperuricemia.
A negative correlation between the levels of antioxidants and XO activity has been confirmed in patients with acute herbicide poisoning 41 . XO is the key enzyme responsible for catalytic synthesis of uric acid from xanthine and hypoxanthine 42 , and is responsible for ROS generation 43 . Consequently, higher amounts of ROS are generated alongside uric acid production 43 .
Therefore, the suppressive effects of parsley and celery on experimental hyperuricemia may be attributed to inhibition of oxidative stress. Deposition of urate crystals in the kidney and joints stimulates inflamed cells to produce IL-1β, which promotes the release of a series of inflammatory cytokines (TNF-α and IL-6) 44 , causing a state of general inflammation 45 . Furthermore, patients with hyperuricemia exhibit decreased levels of the anti-inflammatory cytokine, IL-10 46 . Clinical trials have also shown that gout is associated with elevated IL-1β 47 . These alterations in cytokine levels were ameliorated by PAR and CEL administration either alone or in combination; the combination effect was more effective.
In this study, parsley and celery reduced inflammatory cytokines (IL-1β and TNFα), enhanced serum antioxidant activities and eliminated pathological changes in the kidney. The results suggest that the effect of parsley and celery on IL-1β and TNF-α may be through the modulation of oxidative stress and the enhancement of antioxidant activities. Celery contains furocoumarins, flavonoids (apigenin), phenolic compounds and tannins 48 . The hyperuricemic and xanthine oxidase inhibitory activity of celery was investigated to a lesser extent. Lin et al. 49 reported in vitro studies that apigenin interacts with XO in its active site.
Several transporter genes play critical roles in urate secretion and excretion during hyperuricemia. URAT1, a renal urate anion exchanger and an integral membrane protein found primarily in kidney, transports urate across the proximal convoluted tubules 50,51 . Its expression depends on the uric acid levels in the blood. mGlut-9 is another urate transporter that regulates urate transport through the proximal tubules 52 . OAT-1 and OAT-3 are localized in the proximal convoluted tubules (in the basolateral membrane) 53 . OAT-1 plays a role in the uptake and secretion of urate 53 . OAT-3 participates in the cellular uptake of urate and in urate secretion. URAT1, OAT-1 and OAT-3 have recently been considered the ideal targets for hyperuricemia treatment 54 . This study is the first to show that parsley and celery have the potential to regulate urate excretion associated genes (URAT1, GLUT-9, OAT-1 and OAT-3), either alone or in combination. PO administration significantly up-regulated mURAT1 and mGlut-9 expression, and down-regulated mOAT-1 and mOAT-3 expressions in mouse kidneys. Oxonate-induced urate reabsorption and reduced urate secretion is counteracted by parsley and celery, which, when co-administered, reduced the effect of disorders associated with hyperuricemia. Both PAR and CEL effectively cured hyperuricemia through: control of xanthine oxidase activity, control of inflammatory cytokines, increase in antioxidant activities and decrease in oxidative stress. Further, the genes responsible for urate transporter expression were controlled. These effects are illustrated in Fig. 6.