Morphine Induces Albuminuria by Compromising Podocyte Integrity

Morphine has been reported to accelerate the progression of chronic kidney disease. However, whether morphine affects slit diaphragm (SD), the major constituent of glomerular filtration barrier, is still unclear. In the present study, we examined the effect of morphine on glomerular filtration barrier in general and podocyte integrity in particular. Mice were administered either normal saline or morphine for 72 h, then urine samples were collected and kidneys were subsequently isolated for immunohistochemical studies and Western blot. For in vitro studies, human podocytes were treated with morphine and then probed for the molecular markers of slit diaphragm. Morphine-receiving mice displayed a significant increase in albuminuria and showed effacement of podocyte foot processes. In both in vivo and in vitro studies, the expression of synaptopodin, a molecular marker for podocyte integrity, and the slit diaphragm constituting molecules (SDCM), such as nephrin, podocin, and CD2-associated protein (CD2AP), were decreased in morphine-treated podocytes. In vitro studies indicated that morphine modulated podocyte expression of SDCM through opiate mu (MOR) and kappa (KOR) receptors. Since morphine also enhanced podocyte oxidative stress, the latter seems to contribute to decreased SDCM expression. In addition, AKT, p38, and JNK pathways were involved in morphine-induced down regulation of SDCM in human podocytes. These findings demonstrate that morphine has the potential to alter the glomerular filtration barrier by compromising the integrity of podocytes.


Introduction
Morphine, a metabolite of heroin, is the main stay of pain management after surgery, angina, myocardial infarction, and trauma [1]. However, heroin has also been one of the main drugs abused for addiction, and has significantly contributed to morbidity and mortality since the early 1990s in the United States [2,3,4,5].
Morphine has been reported to exert a bimodal effect on the growth of kidney fibroblasts and glomerular mesangial cells [6,7,8,9,10]. Intravenous opiate addiction has also been considered a risk factor for the development of human immunodeficiency (HIV)-associated nephropathy [11,12]. Johnson et al reported that morphine administration altered the microprojections on podocytes [13]. We also reported earlier that morphine promoted glomerular epithelial cell proliferation at lower concentration while triggered apoptosis at higher concentration [8]. We demonstrated that morphine-induced alteration in epithelial cell phenoytype was mediated by alteration of hemoxygenase (HO)-1 activity; however, in these studies, we did not study any loss of visceral epithelial cell (podocyte) markers in response to morphine [8]. Thus, the present study not only provides data on the effect of morphine on glomerular visceral epithelial cells (podocytes) in vivo, but also examines the effect of morphine on the markers of visceral epithelial cell integrity.
As the major constituent of glomerular filtration barrier (GFB), slit diaphragm (SD) plays an important role in the prevention of glomerular protein leakage both in physiological and in pathological states. SD consists of proteins specifically expressed by podocytes, such as nephrin, podocin, CD2-associated protein (CD2AP), and synaptopodin [14]. On that account, the podocyte is considered the key cell contributing to the development of albuminuric kidney diseases. Decrease of nephrin and podocin has been shown to be linked with albuminuria and progressive renal disease [15,16]. Whether morphine impairs the SD constituting molecules (SDCM) has yet to be studied.
In the present study, we evaluated the effect of morphine on glomerular filtration barrier in general and podocyte integrity in particular. We showed that morphine-receiving mice not only developed albuminuria but also displayed attenuated expression of SDCM in human podocytes. We also demonstrated that morphine-induced loss of podocyte integrity through loss of slit diaphragm associated molecules is mediated by reactive oxygen species (ROS) generation. These findings provide a basis to support the hypothesis that morphine-induced direct impairment of podocytes contributes to kidney disease.

Reagents
Morphine pellets were obtained from National Institute on Drug Abuse (Bethesda, MD, USA). For in vitro study, morphine was dissolved in normal saline, stored at a concentration of

Animal experiments
All animal experiments complied with Institutional Animal Care and Use Committee (IACUC)-approved protocols. FVB/N mice (purchased from Jackson Laboratory, eight in each group) were administered either normal saline or morphine (by subcutaneous implantation of a 75-mg slow-release morphine pellet). After 72 h, spot urine samples were collected by giving gentle pressure over the urinary bladder. The mice were subsequently sacrificed and the kidneys were isolated for renal immunohistochemical and Western blotting studies. Albumin and creatinine in the urine were measured with Albuwell M Test Kit (Cat# 1011, Exocell, Philadelphia, PA) and Creatinine Companion (Cat# 1012, Exocell), respectively, following the instruction of the manufacturer.

Culture of human podocytes
Human podocytes were cultured as previously reported [17]. Briefly, immortalized human podocytes proliferated in the growth medium containing RPMI 1640 supplemented with 10% fetal bovine serum, 1 X pecicillin-streptomycin, 1 mM Lglutamine, and 1 X insulin, transferrin, and selenium (ITS) (Invitrogen, Grand Island, NY) at permissive temperature (33uC). When the cells reached about 80% confluence, they were transferred to 37uC for differentiation in a medium without ITS for 7 days.

Intracellular ROS measurement
Human podocytes were differentiated in 96-well plates for 7 days as mentioned above, and were then cultured in serum free medium for 12 h. Subsequently, 10 210 to 10 26 M morphine was added. After incubation for another 12 h, intracellular ROS was determined by analyzing the fluorescence intensity of the intracellular fluoroprobe 2, 7-dichlorofluorescin (DCFH) (Molecular probe, Carlsbad, CA, USA), following the manufacturer's instructions.

Electron Microscopic (EM) studies
Mice in groups of five were administered either normal saline or morphine. Kidneys were harvested and fixed in glutaraldehyde and embedded in EPON. Thin sections were cut and reviewed and examined under a transmission microscope. All the EM sections were reviewed in a blinded fashion. Pictures from each kidney were acquired in a double blind manner using sample numbers without the identity of the mouse.

Statistical analyses
Data were presented as means 6 standard deviation (SD) unless otherwise noted. All experiments were repeated at least three times with duplicate or triplicate samples in each assay. All data were evaluated statistically by the analysis of variance (ANOVA), followed by Nweman-Keuls multiple comparison tests using software (Prism 4.0, GraphPad Software). In the case of single mean comparison, data were analyzed by t test. P values ,0.05 were regarded as statistically significant.

Classic opiate receptors are present on podocytes
Opiate receptors (ORs) have been reported to express in kidney messangial cells [6], but their expression in podocytes has not been studied. Therefore, we examined the expression of ORs in podocytes. In in vitro study, we conducted RT-PCR analysis by using human podocytes as the RNA source. The results revealed high level expression of KOR and MOR in human podocytes, but the expression of DOR was barely detectable (Fig. 1). To confirm this observation in in vivo studies, we performed immunofluorescence staining of mouse kidney sections. Both KOR and MOR were highly expressed in kidney, and they were colocalized with podocyte marker nephrin; while no DOR expression was found in podocyte (Fig. 2). These results demonstrate that podocytes display expression of KOR and MOR only.

Morphine treatment leads to acute albuminuria
It has been reported that chronic morphine treatment leads to glomerulophathy particularly proteinuria [5]. To examine the effect of morphine on podocytes, we treated the mice with high dose morphine as described in Materials and Methods. Three days after the treatment, mouse urine samples were collected, and total albumin and albumin/creatinine ratio were determined. The histology in Periodic acid-Schiff (PAS)-stained mouse kidney sections was also examined. Although we didn't find obvious differences between the control and treatment groups in the histology (data not shown), we did observe that high dose morphine treatment significantly increased both total albumin and albumin/creatinine ratio in the urine (Fig. 3). These results suggest that morphine treatment directly impairs the blood urine barrier.

Morphine treatment compromises podocyte integrity
To determine the effect of morphine on podocytes, we performed immunofluorescent staining of mouse kidney sections and immunoblotting studies of the kidney lysates. Morphine attenuated the expression of podocyte specific molecules such as synaptopodin and nephrin (Fig. 4, A-D). In vitro study also showed that morphine treatments decreased the protein expression of SDCMs in human podocytes (Fig. 5). Transmission electron microscopy of the mouse kidney sections showed that morphine administration increased the podocyte foot process effacement ( Figure 4E). These results indicate that morphine treatment may compromise podocyte integrity.

Morphine mediates alterations in podocyte SDCMs via MOR and KOR
Weber et al reported that morphine affects mesangial cells mainly via KOR but not MOR [6]. To determine the receptors involved in podocytes, we stimulated human podocytes with ORspecific agonists. Western blot results revealed that both KORspecific agonist LKD26 and MOR-specific agonist DAMGO decreased the expression of SDCMs, while DOR-specific agonist DPDPE did not show this function (Fig. 6). These results indicate that morphine-induced alterations in podocyte SDCM occurred via both KOR and MOR.

Morphine induces podocyte DNA damage through ROS generation
Our group and others have demonstrated that ROS is a significant contributing factor for podocyte injury and for the progression of chronic kidney disease [8,19,20]. ROS has also been shown to damage SD [21,22,23]. To examine the effect of morphine on intracellular ROS production, the fluorescence intensity of the intracellular fluoroprobe (DCFH) was evaluated. Results showed that ROS generation was significantly increased in morphine-treated cells as compared with unstimulated cells (Figure 7). Morphine increased ROS generation in a dosedependent manner ( Figure 7A).
To determine the effect of morphine-induced ROS generation on oxidative DNA damage in the podocyte, we carried out immunofluorescent staining of podocytes for 8-hydroxyguanine (8-OHdG), a molecular marker of oxidative damage to DNA, in morphine-receiving mice. Morphine-receiving mice displayed increased positive staining of 8-OHdG in podocytes ( Figure 7C).
To test the effect of oxidative stress on SDCMs, aliquots of variable concentrations of hydrogen peroxide (H 2 O 2 ) were added to the media of human podocytes for 24 h. Western blotting studies revealed that H 2 O 2 decreased SDCMs expression in a dose-dependent manner (Figure 8). To further confirm these observations, podocytes were pre-treated with either superoxide dismutase or catalase and then incubated in media containing H 2 O 2 . As shown in Figure 9, pretreating the human podocytes with SOD or catalase significantly attenuated morphine-induced compromise of SDCMs. Taken together, these data suggest that morphine-induced ROS generation may be a contributor to the decrease of SDCMs.
AKT, JNK, and p38 pathways are involved in opioidinduced podocyte injury Several kinases, including AKT, JNK, Erk1/2 and p38 have been implicated in podocyte injury and the progression of chronic kidney diseases (CKD) [6,15,24,25,26]. A transcription factor, signal transducer and activator of transcription 3 (STAT3), has also been reported to regulate HIV-induced podocyte proliferation [24]. To examine the involvement of these kinases and factors in morphine-induced down-regulation of SDCMs, we first evaluated the phosphorylation of these proteins. We treated the human podocytes with morphine, and collected the cell lysates at different time points for Western blot studies. Results showed that morphine stimulation significantly activated AKT, JNK, and p38 at early time points (Fig. 10), while neither Erk1/2 nor STAT3 could be activated by morphine at any time point (data not shown). We also used KOR-and MOR-specific agonists, LKD26 and DAMGO, to stimulate the human podocytes, and the same activation results with morphine were observed (data not shown). These results suggest that morphine down-regulates SDCMs via AKT, JNK, and p38, but not Erk1/2 or STAT3.
To further examine the role of activation of AKT, JNK, and p38 in morphine-induced down-regulation of SDCMs, the expression of SDCMs was measured in podocytes treated with morphine in the presence or absence of LY294002, an inhibitor of PI13/AKT, SP600125, an inhibitor of JNK, or SB203580, an   inhibitor of p38. As presented in Figure 11, all these inhibitors partially attenuated morphine-induced down-regulation of SDCMs. These results indicate that AKT, JNK, and p38 pathways are involved in the regulation of morphine-induced impairment of SDCMs.

Discussion
Morphine has been reported to cause kidney cell injury in both in vitro and in vivo studies [7,8,9,10,11,12]. In the current study, we observed that administration of morphine contributed to albuminuria which appeared to be glomerular in origin and a consequence of the loss of slit diaphragm integrity [27]. Both in vivo and in vitro data confirmed that morphine inflicted podocyte injury in the form of attenuated expression of SDCMs (Figure 4 and 5); these findings further confirmed that loss of integrity of SD contributed to albuminuria in morphine-receiving mice. Since these SDCMs are actin-associated molecules, the reduced level of expression of these molecules by morphine may change the actin cytoskeleton organization. Transmission electron microscopy of the mouse kidney sections showed effacement of podocyte foot process in morphine-receiving mice ( Figure 4E), which might be due to the change of the actin cytoskeleton organization. However, further studies are needed to explore this aspect.
In previously reported studies, morphine has been demonstrated to exert bimodal effects (both apoptosis and proliferation) on glomerular epithelial cells [8]; however the role of opiate receptors was not evaluated in these studies. In the current study, we found that KOR and MOR, but not DOR, were expressed in both human and mouse podocytes (Figures 1 and 2). We also found that stimulating either KOR or MOR in podocytes could activate related kinases and down-regulate SDCMs. To our knowledge, this is the first report on the opiate receptors in podocytes. Weber  et al reported that although both KOR and MOR were expressed in glomerular mesangial cells, only KOR played the function to activate STAT3 and led to the proliferation and glomerulopathy [6]. These findings indicate that podocytes behave differently from mesangial cells in morphine milieu.
Oxidative stress is a common cause of cellular injury. Several investigators have demonstrated that enhanced ROS generation leads to albuminuria by damaging SD components [21,28,29]. Morphine has been demonstrated to stimulate the production of superoxide by macrophage and mesangial cells [22,23]. As we previously reported, generation of ROS induces apoptosis of rat glomerular epithelial cells [8]. Here we also found that morphine increased ROS generation in human podocytes in a dosedependent manner ( Figure 7A). Addition of H 2 O 2 to the media decreased the expression of SDCMs (Figure 8), while free radical scavengers prevented this damaging effect of morphine ( Figure 9). All these findings strongly suggest the role of ROS in morphineinduced down-regulation of SDCMs. One potential possibility is that ROS induced injury hampers the expression of SDCMs. Further investigation into the detailed underlying mechanisms need to be carried out in future studies.
Investigating the kinase or transcription factor pathways involved in morphine-induced kidney injury may provide insight into new potential targets for therapy. STAT3, AKT and mitogenactivated protein (MAP) kinases, including ERK1/2, JNK, and p38, have been implicated in podocyte injury and the progression of chronic kidney diseases (CKD) [6,14,24,25,26]. All these kinases or transcription factors may also be activated by morphine in various cells [6,30,31]. We examined the effect of morphine on the activation of these kinases and factors in podocytes. Our results revealed that morphine stimulated the phosphorylation of AKT, JNK and p38 ( Figure 10), but could not activate Erk1/2 or STAT3 (data not shown). These observations were further confirmed by the use of KOR and MOR specific agonists on human podocytes in morphine milieu (data not shown).
Takano et al reported that activation of AKT suppressed the expression of nephrin [32]; Ikezumi et al claimed that activation of JNK or p38 decreased the expression of nephrin and podocin, while inhibiting these kinases restored their expressions [33]. Consistent with these reports, we also found that inhibiting AKT, JNK, and p38 could partially prevent the morphine-induced decrease of SDCMs (Figure 11), indicating the regulation of these three kinases in morphine-induced SD damage.
In conclusion, we have demonstrated that morphine has the potential to directly impair the SDCMs in podocytes, which will contribute to acute kidney injury. The effects of morphine on podocytes are mediated through both MOR and KOR. These impairments are through generation of ROS, and are regulated by AKT, JNK, and p38 pathways. Our study provides insight into new mechanisms involved in morphine-induced podocyte damage, and highlights some new therapeutic targets for morphine induced kidney injury.