Akt Substrate of 160 Kd Regulates Na ,K -Atpase Trafficking in Response

BASIC RESEARCH www.jasn.org

Akt Substrate of 160 kD Regulates Na+,K+-ATPase Trafﬁcking in Response to Energy Depletion and Renal Ischemia

† ‡ † Daiane S. Alves,* Gunilla Thulin, Johannes Lofﬁng, Michael Kashgarian, and Michael J. Caplan*

Departments of *Cellular and Molecular Physiology and †Pathology, Yale University School of Medicine, New Haven, Connecticut; and ‡Institute of Anatomy, University of Zurich, Zurich, Switzerland

ABSTRACT Renal ischemia and reperfusion injury causes loss of renal epithelial cell polarity and perturbations in tubular solute and fluid transport. Na+,K+-ATPase, which is normally found at the basolateral plasma membrane of renal epithelial cells, is internalized and accumulates in intracellular compartments after renal ischemic injury. We previously reported that the subcellular distribution of Na+,K+-ATPase is modulated by direct binding to Akt substrate of 160 kD (AS160), a Rab GTPase-activating protein that regulates the trafficking of glucose transporter 4 in response to insulin and muscle contraction. Here, we investigated the effect of AS160 on Na+, K+-ATPase trafficking in response to energy depletion. We found that AS160 is required for the intracellular accumulation of Na+,K+-ATPase that occurs in response to energy depletion in cultured epithelial cells. Energy depletion led to dephosphorylation of AS160 at S588, which was required for the energy depletion–induced accumulation of Na,K-ATPase in intracellular compartments. In AS160-knockout mice, the effects of renal ischemia on the distribution of Na+,K+-ATPase were substantially reduced in the epithelial cells of distal segments of the renal tubules. These data demonstrate that AS160 has a direct role in linking the trafficking of Na+,K+-ATPase to the energy state of renal epithelial cells.

J Am Soc Nephrol 26: 2765–2776, 2015. doi: 10.1681/ASN.2013101040

The Na,K-ATPase, or sodium pump, creates the transport protein whose activity is governed through driving force for solute and ﬂuid transport in most stimulus-induced trafﬁcking between the plasma tissues. The energy released through the hydrolysis of membrane and intracellular compartments.8 GLUT4 onemoleculeofATPisusedbytheenzymetodrive is delivered to the cell surface from intracellular stor- exportof three Na+ ions and the import of two K+ ions, age vesicles in response to stimuli that favor increased and this transport is essential for the maintenance of glucose uptake, including insulin and muscle contrac- cellular electrochemical gradients. The Na,K-ATPase tion. AS160, or TBC1D4, is a Rab GTPase-activating is restricted in its distribution to the basolateral domain protein (GAP) that participates in regulating the trans- of the plasma membrane in most polarized epithelial location of GLUT4 to the plasma membrane.9,10 cells.1 Many cell types appear to contain two function- ally separable pools of Na,K-ATPase. In addition to the principal pool at the plasma membrane, a population Received October 3, 2013. Accepted January 6, 2015. of Na,K-ATPase is also frequently associated with the Published online ahead of print. Publication date available at 2–4 membranes of intracellular compartments. Physio- www.jasn.org. logic stimuli can promote Na,K-ATPase endocytosis Correspondence: Dr. Michael J. Caplan, Department of Cellular or translocation from the intracellular pool to the and Molecular Physiology, Yale University School of Medicine, plasma membrane.2,5–7 PO Box 208026, New Haven, CT 06520-8026. Email: michael. The glucose transporter 4 (GLUT4) of muscle and [email protected] fat cells is one of the best-studied examples of a Copyright © 2015 by the American Society of Nephrology

J Am Soc Nephrol 26: 2765–2776, 2015 ISSN : 1046-6673/2611-2765 2765 BASIC RESEARCH www.jasn.org

AS160 is phosphorylated by Akt on at least six amino acid res- Na,K-ATPase in a stable clonal cell line in which AS160 is ro- idues after insulin stimulation9 and this phosphorylation inhibits bustly knocked down by small hairpin RNA (shRNA) (AS160 the Rab-GAP activity of AS160. Because the cell surface accumu- KD).23 Wild-type (WT) Madin-Darby Canine Kidney (MDCK) lation of GLUT4 is dependent upon the GTP-bound state of Rab cells or AS160 KD MDCK cells were subjected or not to energy 8, Rab 10, and Rab 14,11 inhibition of AS160’s Rab-GAP activity depletion. Energy depletion was achieved by incubating cells in a leads to a redistribution of GLUT4 from intracellular storage glucose-free medium containing antimycin A and 2-deoxy-glucose compartments to the plasma membrane. Consistent with this (AA/DG). A mAb directed against the Na,K-ATPase a-subunit was model, the small interfering RNA–induced knockdown (KD) of used to detect the distribution of the sodium pump. Figure 1 AS160 expression increases the basal levels of GLUT4 at the cell depicts the Na,K-ATPase localization in untreated WT and surface and concomitantly reduces the size of the intracellular AS160 KD cells (2). After energy depletion treatment (AA/DG), pool.12,13 AS160 may also play a similar role in regulating the the Na,K-ATPase was dramatically redistributed. Abundant distribution and hence the activity of the epithelial sodium chan- Na,K-ATPase was detected in association with intracellular struc- nel in response to aldosterone stimulation.14 tures in the cytoplasm in the WT cells. The intracellular accumu- Energy depletion in renal epithelial cells results in a re- lation of the Na,K-ATPase in response to energy depletion was distribution of the cell surface Na,K-ATPase, resulting in its substantially reduced in the AS160 KD cell line, suggesting that accumulation in intracellular compartments.15 Energy depriva- AS160 plays an obligate role in sodium pump accumulation in tion induced by renal ischemia can lead to AKI, which is a com- intracellular compartments after energy depletion. mon condition associated with very significant morbidity and In order to confirm the results obtained by immunofluores- mortality.16–18 AKI is associated with detachment of the sodium cence andto quantify the extent ofNa,K-ATPaseaccumulationin pump from the plasma membrane’s subcortical cytoskeleton and cytoplasmiccompartmentsinresponsetoenergydepletioninthe with the loss of cell polarity and resultant impairment of renal MDCK WTand AS160 KD cell lines, a surface biotinylation assay function.7,19–21 Recovery from ischemic renal injury involves was performed. The basolateral surfaces of MDCK cells were restitution of cellular polarity and return of the sodium pump labeled with Sulfo-NHS-SS biotin, after which cells were sub- from intracellular compartments to the plasma membrane.22 jected to energy depletion (AA/DG). As a control, biotinylated Recently, we reported that AS160 interacts directly with the cells were maintained in regular media during the treatment (2). Na,K-ATPase a-subunit and modulates the sodium pump’ssub- Todetermine the total amount of plasma membrane Na,K-ATPase cellular distribution.23 In this study, we investigated the role of that was labeled, aliquots of cells were lysed immediately after AS160 in mediating the accumulation of the Na,K-ATPase in the biotinylation process. After the energy depletion treatment intracellular compartments after energy depletion in cultured or control incubation interval, the biotin that remained exposed at renal epithelial cells. Our data indicate that AS160 is essential the plasma membrane was stripped through an incubation with for sodium pump accumulation in cytoplasmic compartments the membrane-impermeable reducing agent 2-mercaptoethane in response to ATP depletion and that the redistribution of the sulfonate sodium (MesNa), allowing us to detect only the pool of Na,K-ATPase in response to energy depletion is mediated by internalized Na,K-ATPase that was protected from exposure to changes in the phosphorylation state of AS160. Our studies uti- the MesNa reagent. Cell lysates were incubated with streptavidin lizing AS160 knockout (KO) mice further suggest that AS160 beads and samples were subjected to SDS-PAGE and Western plays a role in governing the redistribution of the Na,K-ATPase blotting. Figure 1B shows the resulting Western blot, and Figure in distal segments of the renal tubule that is induced by renal 1C shows the quantification of the relative levels of biotinylated ischemia in vivo. Na,K-ATPase detected under each condition. Similar levels of energy depletion were achieved in both the WTand AS160 KD cells, as evidenced by the comparable increase in the level of the RESULTS phosphorylated form of the energy-sensing kinase adenosine monophosphate-stimulated protein kinase (AMPK) that is AS160 Mediates the Intracellular Accumulation of Na, detected in both cell types. As expected, upon ATP depletion, K-ATPase That Is Induced by Energy Depletion the levels of intracellular Na,K-ATPase were significantly AS160 plays an important role in regulating the GLUT4 protein’s higher in the MDCK WT cells than in the AS160 KD cell intracellular retention and its translocation to the plasma mem- line. These results are consistent with the results observed brane in response to insulin or muscle contraction.24,25 Previ- in the immunofluorescence experiments and suggest that ously, we characterized the interaction between AS160 and Na, AS160 is an essential component of the machinery that mediates K-ATPase and found that AS160 interacts with the cytoplasmic the intracellular accumulation of Na,K-ATPase in response to NP domain of the a-subunit of the sodium pump.23 Na, energy depletion. K-ATPase is internalized and retained in intracellular compartments in response to energy depletion and renal ischemia.7,15 The AS160 and Na,K-ATPase Interaction Is Not To investigate whether AS160 plays a role in this Na,K-ATPase Affected by Energy Depletion redistribution after energy depletion, we examined the effects We previously showed that AS160 interacts directly with the of energy depletion on the subcellular distribution of the Na,K-ATPase in MDCK cells.23 To analyze whether the

2766 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015 www.jasn.org BASIC RESEARCH

Figure 1. shRNA-mediated KD of AS160 inhibits the intracellular accumulation of Na,K-ATPase that is induced by energy depletion. (A) Immunofluorescence analysis is performed to detect the distribution of the endogenous Na,K-ATPase in MDCK renal epithelial cells in culture. MDCK cells are stained with an antibody directed against the sodium pump a-subunit (a5). WT MDCK cells and MDCK cells knocked down for AS160 expression (AS160 KD) are treated with (+) or without (2) AA/DG. Typical results from one of five experiments are shown. (B) Cell surface biotinylation. WT or AS160 KD cells are biotinylated at the basolateral surface, then subjected or not to energy depletion. The biotin that remained exposed at the cell surface after the treatments is stripped by treating the cells with MesNa. Biotinylated proteins are recovered through incubation with streptavidin agarose beads. The biotinylated Na,K-ATPase a-subunit is detected with the mAb a5. Lysates are blotted with anti-AS160, anti–phospho-AMPK antibody, and with anti–b-actin to assess total protein loading. The extent of energy depletion is assessed by probing for phospho-AMPK. (C) Quantification of the biotinylated Na,K- ATPase band intensity normalized to the b-actin levels. The results indicate that the fraction of the pool of sodium pump that was initially biotinylated at the cell surface that accumulated within the cell following energy depletion is lower in the AS160 KD cells compared with the WT line. *P,0.01 (n=4). CT, total biotinylated Na,K-ATPase in untreated cells; 2, biotinylated Na,K-ATPase in untreated cells subjected to MesNa strip; AA/DG, biotinylated Na,K-ATPase in energy-depleted cells treated with MesNa; p-AMPK, phospho-AMPK. Bar, 5 mm. interaction between these two proteins was affected by ATP results indicated that, in MDCK cell lysates treated with alkaline depletion, a coimmunoprecipitation assay was performed. phosphatase, the apparent molecular weight of AS160 was re- MDCK cells were treated or not with media lacking glucose in ducedtoasimilarextentasisobservedinmaterialpreparedfrom the presence of AA/DG. Total cell lysates were then incubated cells that were subjected to energy depletion alone (Figure 3A). with protein G beads and a5 antibody to immunoprecipitate This result supports the interpretation that the reduction of the endogenous Na,K-ATPase a-subunit. The associated AS160 that apparent molecular weight of AS160 that follows energy deple- coprecipitated with the sodium pump was detected by Western tion could be attributable to the effects of dephosphorylation. blotting with anti-AS160 antibody (Figure 2). The results indi- To further explore the effects of energy depletion on AS160 cated that the energy depletion treatment does not affect the phosphorylation, lysates were prepared from MDCK cells that extent of the interaction between AS160 and Na,K-ATPase. In- had been subjected or not to the energy depletion protocol and terestingly, we observed a slight decrease in the apparent molec- analyzed by Western blotting (Figure 3, B and C). As observed ular weight of the coprecipitated AS160 as a result of the energy previously, the apparent molecular weight of AS160 was reduced depletion treatment. in lysates prepared from cells subjected to energy depletion. We tested the effects of energy depletion on the phosphorylation AS160 Phosphorylation at S588 Is Essential for Na,K- status of two specific AS160 phosphorylation sites: S588 and ATPase Intracellular Accumulation in Response to T642. The results show that the phosphorylation of AS160 Energy Depletion T642 does not change in response to energy depletion. However, AS160 phosphorylation plays a critical role in determining S588 phosphorylation was abolished after the treatment, sug- whether GLUT4 is retained intracellularly or translocated to the gesting that the change in the apparent molecular weight of plasma membrane.9 AS160 dephosphorylation is essential for AS160 induced by ATP depletion is due to the dephosphoryla- GLUT4 intracellular storage under basal conditions.12,13 tion of AS160 at S588. As noted previously, the dephosphory- Because a decrease in the apparent molecular weight of AS160 lated form of AS160 promotes the intracellular accumulation of was observed after ATP depletion, we explored whether this GLUT4.9,12–14,26 It is tempting to suggest, by analogy, that the change was attributable to alterations in the phosphorylation dephosphorylation of AS160 at S588 is responsible for the in- state of AS160. We began this analysis by performing an alkaline tracellular accumulation of Na,K-ATPase that accompanies en- phosphatase assay. Total cell lysates from MDCK cells subjected ergy depletion. or not to energy depletion were incubated in the presence or Finally, to test directly whether the intracellular accumulation absence of calf intestinal alkaline phosphatase, after which they ofNa,K-ATPase inresponse toenergydepletionisdependentupon were analyzed by Western blotting with anti-AS160 antibody. The dephosphorylation of AS160 at S588, we prepared constructs

J Am Soc Nephrol 26: 2765–2776, 2015 AS160 and Sodium Pump Trafﬁcking 2767 BASIC RESEARCH www.jasn.org

KD by the shRNA that was expressed in these cells. The resultant clonal cell lines do not express detectable levels of endogenous AS160 (Supplemental Figure 1) compared with control MDCK cells. Figure 4A depicts the immunofluorescence images obtained with these cell lines. Antibodies directed against the FLAG epitope and the a5 mAb were used to detect exogenous AS160 and endogenous Na,K-ATPase, respectively. AS160 WT- FLAG and AS160 S588D-FLAG cells were subjected (4A, II and IV) or not (4A, I and Figure 2. The interaction between AS160 and Na,K-ATPase is not affected by energy III) to energy depletion. As can be seen in depletion treatment. Endogenous Na,K-ATPase is immunoprecipitated using a mAb di- Figure 4AII, Na,K-ATPase was detected in in- rected against the a-subunit (a5). Immunoprecipitates are probed with AS160 polyclonal tracellular compartments after energy deple- antibody to detect endogenous AS160. MDCK cells are treated (+) or not (2) with energy tion in cells expressing AS160 WT-FLAG, depletion AA/DG, for 1 hour. (A) Immunoblot of the coimmunoprecipitation. Lysates are demonstrating that the AS160 WT-FLAG –b blotted with anti-AS160 to detect the endogenous levels of AS160 and with anti -actin as construct can effectively substitute for the ac- fi a control for the total protein loaded. (B) Coimmunoprecipitation quanti cation. The results tivity of the absent endogenous AS160 in the indicate that the interaction between AS160 and sodium pump is not affected by energy AS160 KD cells. By contrast, sodium pump depletion. n=3. IP, immunoprecipitation. was found predominantly at the plasma membrane after energy depletion in AS160 encoding wild-type human AS160 carrying a FLAG epitope tag KD cells expressing the S588D mutant form of AS160 (Figure (AS160 WT-FLAG) or a mutated version of human AS160 in 4AIV). To measure the relative quantities of Na,K-ATPase present which the serine at position 588 has been replaced with a at the basolateral surface versus in intracellular compartments in phosphomimetic aspartic acid residue (AS160 S588D-FLAG). response to energy depletion in cells expressing the S588D mutant These constructs were expressed in MDCK cells in which form of AS160, a surface biotinylation/MesNa stripping assay was expression of the endogenous AS160 had been reduced by shRNA performed as described above. Figure 4B depicts the resultant KD (AS160 KD). The human sequence was used to generate the Western blot. Consistent with the immunofluorescence results transfected constructs because this sequence is not susceptible to shown in Figure 4A, intracellular accumulation of Na,K-ATPase

Figure 3. Energy depletion leads to AS160 dephosphorylation at S588. (A) The apparent molecular weight of AS160 decreases as a result of its dephosphorylation. MDCK cells are subjected (AA/DG) or not (2) to energy depletion for 1 hour. Total cell lysates are incubated in the presence (+) or absence (2) of 10 units of CIP for 30 minutes at 37°C. Lysates are blotted with anti-AS160 antibody to visualize the shift of the apparent molecular weight of AS160. Blots are also probed with anti–phospho-AMPK and p-AS160 S588 antibodies to document the efficacy of both the ATP depletion and the CIP treatments. Anti–b-actin is also used to assess the total amount of protein loaded. (B) Total MDCK cells lysates are blotted with rabbit anti-AS160 to detect the endogenous AS160 and with phospho-specific antibodies directed against the AS160 phosphorylation sites S588 and T642. Lysates are also probed with antibodies directed against p-AMPK. Blots are probed with anti–b-actin to assess the total amount of protein loaded. The data indicate that energy depletion is associated with a decrease in the apparent molecular weight of AS160 that is induced by the dephosphorylation of AS160 at S588. (C) Quantification of AS160 phosphorylation at S588 and T642 is based on four independent experiments. *P,0.01. Typical results of one of four experiments are shown. CIP, calf intestinal alkaline phosphatase; p-AMPK, phospho-AMPK.

2768 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015 www.jasn.org BASIC RESEARCH

Figure 4. Intracellular accumulation of Na,K-ATPase in response to energy depletion does not occur in cells expressing the S588D mutant of AS160. (A) Immunoﬂuorescence analysis of the distribution of endogenous Na,K-ATPase. MDCK cells untreated (2) (I and III) or treated with energy depletion (AA/DG) (II and IV) are stained with an antibody directed against the Na,K-ATPase a-subunit (a5) and with anti-FLAG to detect exogenous AS160. Na,K-ATPase accumulates in intracellular compartments in response to AA/DG in cells stably transfected with AS160 WT-FLAG (II). However, Na,K-ATPase is not localized in intracellular structures in the AS160 S588D cell line treated with AA/DG (IV). (B) Cell surface biotinylation. MDCK WT or AS160 S588D cells are biotinylated at the basolateral surface, then exposed or not to energy depletion (AA/DG). The biotin that remained at the cell surface after the energy depletion is stripped by treating the cells with MesNa. The biotinylated sodium pump is detected with the antibody a5 directed against the Na,K-ATPase a-subunit. Lysates are blotted with anti- AS160, anti-FLAG, and with anti–b-actin antibodies to assess total protein loading. (C) Quantiﬁcation of the biotinylated Na,K-ATPase band intensity normalized to the b-actin levels. The amount of sodium pump that was initially biotinylated at the cell surface and that accumulated within the cell after energy depletion is lower in the AS160 S588D cells compared with the MDCK WT cell line. *P,0.05 (n=3). CT, total biotinylated Na,K-ATPase in untreated cells; 2, biotinylated Na,K-ATPase in untreated cells subjected to MesNa strip; AA/DG, biotinylated Na,K-ATPase in energy-depleted cells treated with MesNa. Bar, 5 mm.

in MDCK WT cells treated with AA/DG was significantly higher assess the effectiveness of dynamin K44A expression in blocking than that measured in the AS160 S588D cell line. Quantification endocytosis in the transfected cells, MDCK cells were incubated of the Na,K-ATPase band intensity detected under each condition with wheat germ agglutinin linked to Alexa-647 probe (WGA- is depicted in Figure 4C. Together, these findings strongly suggest 647) for 30 minutes at 37°C (Figure 5A). In the absence of the that AS160 is necessary for the redistribution of Na,K-ATPase to dominant negative form of dynamin, the surface-bound WGA- intracellular structures in response to energy depletion and that 647 was extensively internalized into the cells. As expected, in the dephosphorylation of AS160 at S588 plays a crucial role in this cells expressing dynamin K44A-green fluorescent protein process. (GFP), the endocytosis of WGA-647 was inhibited. MDCK cells transfected with dynamin K44A-GFP were treated (+AA/DG) or Na,K-ATPase–Induced Endocytosis in Response to not with energy depletion (2) (Figure 5B). Interestingly, the Na, Energy Depletion Is a Unique Pathway That Is Not K-ATPase internalization induced by energy depletion contin- Regulated by Dynamin and Does Not Involve Caveolae ued to be detected in cells that express the dominant negative To investigate the mechanism that controls Na,K-ATPase endo- from of dynamin. These findings strongly suggest that the in- cytosis in response to energy depletion, MDCK cells were tracellular accumulation of Na,K-ATPase that follows energy de- transfected with a dominant negative form of dynamin contain- pletion is not dependent upon the dynamin/clathrin pathway. ing the mutation K44A. The cells were then treated or not with It has been shown that caveolin-1 interacts with the Na,K- energy depletion for 30 minutes at 37°C according to our stan- ATPase27 and is involved in the ouabain-induced endocytosis dard protocol, after which they were subjected to immunofluo- of the sodium pump.28 To assess whether the Na,K-ATPase rescence analysis, the results of which are shown in Figure 5. To becomes associated with caveolae in response to energy

J Am Soc Nephrol 26: 2765–2776, 2015 AS160 and Sodium Pump Trafﬁcking 2769 BASIC RESEARCH www.jasn.org

Figure 5. Na,K-ATPase endocytosis after energy depletion does not involve dynamin or caveolin-1. (A) Immunoﬂuorescence analysis of MDCK cells transfected with dynamin K44A-GFP and incubated with WGA-647. Cells positive for the expression of the dominant negative form of dynamin do not display WGA-647 endocytosis. (B) MDCK cells transfected with dynamin K44A-GFP are subjected or not (2) to energy depletion (AA/DG). Na,K-ATPase localization is determined with an antibody directed against the Na,K-ATPase a-subunit (a5). (C) Immunoﬂuorescence analysis of endogenous Na,K-ATPase and caveolin-1 in MDCK cells untreated (2)ortreated with energy depletion, (AA/DG). The results indicate that sodium pump endocytosis induced by energy depletion does not depend upon dynamin or involve transit through caveolin-1–containing compartments. Typical results of one of three experiments are presented. Bar, 5 mm.

depletion, MDCK cells were subjected or not to energy de- AS160 Mediates the Intracellular Accumulation of the pletion and examined by immunofluorescence analysis, which Na,K-ATPase in Epithelial Cells of Distal Segments is depicted in Figure 5C. No colocalization between the intra- after Renal Ischemia And Reperfusion Injury In Vivo cellular pool of Na,K-ATPase and caveolin-1 was detected in The Na,K-ATPase is internalized and mislocalized as a result of cells subjected to energy depletion, suggesting that the redis- renal ischemia in vivo.7,30 To determine whether AS160 is re- tribution of the Na,K-ATPase that follows energy depletion quired for the intracellular accumulation of the Na,K-ATPase does not involve its transit through caveolae. as a consequence of renal ischemia, WT and AS160 null mice Like the Na,K-ATPase, E-cadherin is also endocytosed in were subjected to renal ischemia and reperfusion. Both WT response to energy depletion.15 To investigate the generality of and AS160 KO animals underwent 30 minutes of bilateral re- the AS160-regulated mechanism that drives endocytosis and renal ischemia that was imposed by clamping the renal pedicles, tention of the sodium pump, an immunofluorescence assay was followed by 24 hours of reperfusion. Control animals of both performed to detect E-cadherin endocytosis after energy deple- genotypes were subjected to sham surgery, in which renal tion. Figure 6A shows immunofluorescence images from blood flow was not compromised. Immunofluorescence lo- MDCK (WT) and AS160 KD cells subjected or not to energy calization of the Na,K-ATPase in kidney sections of two dif- depletion. The results indicate that AS160 does not play an ob- ferent animals subjected to ischemia followed by 24 hours of ligate role in the energy depletion–induced endocytosis and cy- reperfusion is depicted at high magnification in Figure 7 toplasmic retention of E-cadherin. Finally, we assessed whether (lower-magnification images are presented in Supplemental the energy depletion–induced AS160-dependent internalization Figure 2). In total, eight mice from both groups were analyzed. mechanism is specific for the Na,K-ATP or instead affects the AS160 is expressed predominantly in the distal convoluted entire basolateral plasma membrane. Toward this end, WT or tubule (DCT).31 Kidney sections that exhibited calbindin- AS160 KD MDCK cells were transfected to express the plasma positive DCT profiles were selected for imaging of Na,K- 29 membrane marker GFP-PH-PLCd1 and then treated or not ATPase distribution. In control and sham-treated animals, with energy depletion. The results indicated that GFP-PH- the Na,K-ATPase is localized to the basolateral infoldings of PLCd1 is not internalized or retained in the cytoplasm (Figure DCTepithelial cells (Figure 7, A and B). WT mice subjected to 6B) in response to energy depletion. These observations strongly ischemia followed by 24 hours of reperfusion exhibited Na,K- support the conclusion that AS160 regulates Na,K-ATPase traf- ATPase mislocalization and accumulation of sodium pump in ficking in response to energy depletion via auniquepathway. cytoplasmic structures in DCT cells (Figure 7, C and D). By

2770 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015 www.jasn.org BASIC RESEARCH

Figure 6. The mechanism that regulates Na,K-ATPase trafficking after energy depletion is specific and AS160 dependent. (A) Im- munofluorescence analysis of WT MDCK cells and AS160 KD MDCK cell lines stained with an antibody that detects endogenous levels of E-cadherin. Cells are treated (AA/DG) or not (2) with energy depletion for 30 minutes. The images indicate that E-cadherin endocytosis induced by energy depletion is AS160 independent. (B) Immunofluorescence images of WT and AS160 KD MDCK cells expressing a membrane marker GFP-PH-PLCd1 that are treated (AA/DG) or not (2) with energy depletion. Endogenous Na,K-ATPase is detected with an antibody directed against the a-subunit (a5). The resultant immunofluorescence images show that energy depletion does not produce a generalized internalization of plasma membrane components, indicating that the AS160-dependent modulation of Na,K- ATPase distribution is specific. Bar, 5 mm. contrast, the plasma membrane localization of the Na,K-ATPase internalization in response to ischemic injury was not apprecia- was preserved and sodium pump accumulation in intracellular bly different in PTs of WT versus AS160 KO animals (Supple- compartments was dramatically reduced in the DCTsegments of mental Figure 3). No significant differences in the levels of serum the AS160 KO mice subjected to the ischemia and reperfusion creatinine measured after 24 hours of reperfusion were detected protocol (Figure 7, E and F). The Na,K-ATPase distribution in between AS160 null mice and WT mice, nor were substantial each condition was scored as being predominantly basolateral or differences noted in the histologic appearance of renal tissue predominantly cytoplasmic by an observer who was blinded to obtained from WTand AS160 KO mice subjected to the ischemia the experimental conditions. Sections from three animals for protocol (data not shown). To better quantify the extent of tissue each of the four experimental groups were analyzed, and damage, a terminal deoxynucleotidyl transferase–mediated .100 tubule cross-sections per animal were scored. As expected, digoxigenin-deoxyuridine nick-end labeling assay was performed in sections from WTand AS160 KO mice subjected to the sham (Supplemental Figure 4). The results indicate that the levels of cell procedure, a very small fraction of the tubules were scored as death after injury are similar in the DCTs of WT and AS160 KO exhibiting extensive cytoplasmic labeling (4% and 5%, respec- mice. Thus, as expected, the absence of AS160 does not alter the tively). In animals subjected to ischemia followed by 24 hours of effects of ischemia and reperfusion on the distribution of the Na, reperfusion, the fraction of tubules with predominantly cyto- K-ATPase in PT cells or on the severity of renal injury. plasmic staining was significantly higher in the WT compared with the AS160 KO renal tissue (62% versus 43%, respectively; P,0.02). These results are consistent with the conclusion that DISCUSSION AS160 plays an obligate role in the intracellular accumulation of Na,K-ATPase in DCT that accompanies renal ischemia and ARF is a common and clinically important problem whose reperfusion. incidence and effect is increasing.32 Ischemic renal injury in- Because AS160 is not detectably expressed in proximal tubule duces defects in tubular reabsorption of solute and fluid. The (PT) epithelial cells,31 we anticipated that ischemia-induced in- Na,K-ATPase creates the driving forces responsible for most ternalization of the Na,K-ATPase in PT cells would occur to the ion transport in kidney tubules. Renal ischemia induces the same extent in WT and AS160 KO mice. Consistent with this loss of epithelial cell polarity, and the normally basolateral Na, hypothesis, we found that the extent of Na,K-ATPase K-ATPase relocalizes to intracellular compartments and to the

J Am Soc Nephrol 26: 2765–2776, 2015 AS160 and Sodium Pump Trafﬁcking 2771 BASIC RESEARCH www.jasn.org

Figure 7. Renal ischemia–induced redistribution of Na,K-ATPase is reduced in the DCTs of AS160 KO mice. Immunofluorescence images of calbindin-positive tubules of kidney cortex stained with an antibody directed against the Na,K-ATPase a-subunit (a5). (A and B) WT (A) and AS160 KO (B) mice exposed to sham surgery. Tissue from two separate WT and AS160 KO animals exposed to 30 minutes of bilateral renal pedicle clamping followed by 24 hours of reperfusion. (C–F) Na,K-ATPase mislocalization after ischemia and reperfusion is less dramatic in the DCT cells of the AS160 KO mice (E and F) compared with those of the WT mice (C and D). Images are taken at a uniform contrast and brightness and are representative of at least eight animals per condition. Bar, 10 mmand5mmforlow- magnification and enlarged images, respectively. apical surface.33,34 The mechanism that controls this Na,K- between intracellular compartments and the cell surface.12–14 ATPase redistribution in response to energy deprivation is Recently, we described that AS160 interacts with the Na,K- poorly understood. In this study, we characterized the role ATPase through the cytosol-facing NP domain of the sodium of AS160 in this process in vitro and in vivo and found that pump’s a-subunit.23 AS160 mediates the intracellular accumulation of the Na,K- Previous studies have shown that the reduction in the size of ATPase after ischemic injury. the pool of the Na,K-ATPase at the plasma membrane after Intracellular compartments within renal epithelial cells can hypoxia is specific and not a consequence of general membrane exchange Na,K-ATPase molecules with the pool of sodium internalization, because levels of both the transferrin receptor pump present at the plasma membrane. Up to 30% of a renal and GLUT-1 remained unchanged.37 In this work, we find that epithelial cell’s complement of Na,K-ATPase may be contained sodium pump endocytosis induced by energy depletion is me- in intracellular compartments that have the potential capacity diated by a pathway that is not dependent upon dynamin and to be translocated to the plasma membrane.5 The Na,K-ATPase does not involve passage through caveolae. The Na,K-ATPase can also be internalized by endocytosis and retained in intracel- endocytosis that occurs in response to energy depletion was lular structures in response to stimuli such as hormones and not affected by the overexpression of a dominant negative hypoxia. These trafficking processes appear to be controlled form of dynamin and the Na,K-ATPase did not colocalize by a variety of protein kinases.5,35–38 with caveolin-1 at any point during or after its internalization. We have found that AS160 mediates the intracellular It would appear, therefore, that the AS160-dependent mech- accumulation of the Na,K-ATPase that is induced by energy anism that controls sodium pump endocytosis in response to depletion. We find that shRNA-mediated KD of AS160 energy depletion does not involve these two major endocytic expression is sufficient to prevent the relocalization of the pathways. Na,K-ATPase to intracellular structures in MDCK renal We find that the mechanism throughwhich AS160 mediates epithelial cells that have been subjected to ATP depletion. Na,K-ATPase trafficking in response to energy depletion This result is consistent with the finding that small interfering involves regulation of the phosphorylation of AS160 at S588. RNA–mediated KD of AS160 increases the cell surface expres- The phosphorylation state of AS160 determines its capacity to sion of both GLUT4 and the epithelial sodium channel in control GLUT4 translocation to the plasma membrane.39,40 En- muscle and renal epithelial cells, respectively. AS160 appears ergy depletion treatment in MDCK cells results in a substantial to regulate the trafficking of both of these transport proteins decrease in the phosphorylation of AS160 at S588. We

2772 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015 www.jasn.org BASIC RESEARCH generated a stable cell line expressing a phosphomimetic form of Alargebodyofliteraturedocumentsthatrenalischemiaresults AS160 by substituting S588 with an aspartic acid residue. When in partial redistribution of the Na,K-ATPase from the basolateral this construct is expressed in MDCK cells in which expression of domain of the plasma membrane to the apical surface in PT the endogenous AS160 has been reduced by shRNA-mediated cells.20,43 Although no such loss of Na,K-ATPase polarity is ob- KD, the intracellular accumulation of the sodium pump that is served in distal nephron segments, biochemical studies suggest induced by ATP depletion is substantially reduced compared with that ischemic injury disrupts the attachment of the Na,K-ATPase that observed in AS160 KD cells transfected to express WTAS160. to the subcortical cytoskeleton throughout the nephron, thus Together, these results support the conclusion that the dephos- enhancing its susceptibility to endocytosic internalization.30,43 phorylation of AS160 S588 is required to induce the intracellular The ischemia-induced internalization of the Na,K-ATPase may accumulation of Na,K-ATPase. Furthermore, our data indicate be of both physiologic and pathophysiologic significance, be- that loss of phosphorylation at S588 is sufficient to induce the cause an analysis of cadaveric transplanted kidneys demonstrated intracellular accumulation of the Na,K-ATPase and persistence of that those kidneys that manifest delayed graft function exhibited this phosphorylation is sufficient to prevent this relocalization. significantly larger pools of intracellular Na,K-ATPase than did Further experiments will be required to determine whether and those that demonstrated prompt graft function.44 In addition, how the regulation of kinase and phosphatase activities contrib- the redistribution of the sodium pump from the basolateral utes to determining the phosphorylation state of AS160 at S588. membrane to intracellular compartments could account at least Overexpression of a dominant active mutant form of AS160 in part for the decrease in tubular Na absorption that occurs as a (AS160-4P), which has four phosphorylation sites mutated to consequence of renal ischemia. Our data suggest that, at least in alanines, inhibits insulin-stimulated GLUT4 translocation to the the DCT, AS160 plays a central role in this clinically important plasma membrane in adipocytes, whereas the overexpression of process. the WTAS160 protein has no effect on this process.9 Addition of another mutation to AS160-4P, in the GAP domain at AS160 R973K, overcomes the inhibitory effect of AS160-4P.These find- CONCISE METHODS ings indicate that the GAP activity of AS160 is required for it to manifest its capacity to induce the intracellular accumulation of Antibodies and Reagents GLUT4.9 Stöckli et al. suggested that S588 phosphorylation A polyclonal antibody directed against amino acids 1178–1189 of rat confers a stronger inactivation of the AS160 GAP activity than AS160 was obtained from EMD Millipore (Billerica, MA). Monoclonal does phosphorylation of AS160 at T642 alone.26 Interpreted in anti–Na,K-ATPase antibody a5 is directed against the amino terminus of light of these observations, our findings suggest that energy de- the rat Na,K-ATPase a1 subunit.45 Polyclonal phospho-specificantibodies pletion results in the dephosphorylation of AS160 at S588, which recognizing AS160 phosphorylated at S588 and T642 were obtained from triggers the activation of the AS160 GAP activity and results in Symansis (New Zealand). Monoclonal anti–b-actin antibody and anti– the accumulation of the GDP-bound form of a critical Rab pro- calbindin polyclonal antibody were purchased from Abcam, Inc. tein that participates in determining the subcellular distribution (Cambridge, MA), and anti–caveolin-1 polyclonal and anti–E-cadherin of the Na,K-ATPase. AS160 displays in vitro GAP activity toward mAbs were obtained from BD Transduction Laboratories (San Jose, CA). several Rab proteins, including Rab 2A, Rab 8A, Rab 10, Rab 11, Rabbit anti–phospho-AMPK (T172) antibody was purchased from Cell Rab 13, and Rab 14.11,13,41 Recently, Comellas et al. suggested Signaling Technology (Danvers, MA). The cell death quantification ex- that Rab 10 is implicated in sodium pump trafficking in response periment was done using the terminal deoxynucleotidyl transferase– toinsulininpulmonarycells.42 It will be important to identify mediated digoxigenin-deoxyuridine nick-end labeling assay following the Rab protein involved in the AS160-mediated Na,K-ATPase the manufacturer’s protocol (Roche, Indianapolis, IN). trafficking in response to energy depletion, as well as the role of the GAP domain of AS160 in this process. Cell Culture and Energy Depletion Protocol Our studies utilizing AS160 KO mice indicate that at least MDCK cells were grown in a 5% CO2 and 95% air humidified incubator some component of the intracellular accumulation of Na,K- at 37°C in a-MEM (Invitrogen, Carlsbad, CA) supplemented with 10% ATPase that is induced by renal ischemia in distal segments of the FBS,2mML-glutamine,50U/mlpenicillin,and50g/mlstreptomycin. nephron in vivo is dependent upon the participation of AS160. The cells were subjected to energy depletion via incubation in prewarmed AS160 is expressed at high levels in the renal DCT.31 Our data glucose-free DMEM containing 0.1 mM AA and 25 mM DG (Sigma- indicate that, in the absence of AS160 expression, the relocaliza- Aldrich, St. Louis, MO) to inhibit aerobic and substrate-dependent ATP tion of sodium pump that occurs in response to renal ischemia generation, respectively. Cells were subjected to 30 minutes to 1 hour of followed by 24 hours of reperfusion in cells of the DCT is sub- energy depletion as previously described.46 stantially reduced. The absence of AS160 does not reduce the ischemia-induced internalization of the Na,K-ATPase in PTs, Constructs and DNA Cloning consistent with the lack of detectable AS160 expression in prox- The sequence chosen for the shRNA construct targeting expression of 9 9 imal segments.31 Clearly other mechanisms must be involved in canine AS160 (5 GCAAGGGAGCATGGTATTA3 ) was subcloned into driving the substantial internalization of Na,K-ATPase that oc- pSUPER plasmid (Oligoengine, Seattle, WA). The detailed description curs in response to ischemic injury in PTs. of the production and characterization of the stable KD AS160 MDCK

J Am Soc Nephrol 26: 2765–2776, 2015 AS160 and Sodium Pump Trafficking 2773 BASIC RESEARCH www.jasn.org cell line are presented elsewhere.23 The CMV-10 plasmid encoding the Cells were first biotinylated at the basolateral surface with Sulfo-NHS-SS WT human AS160 was obtained from Thermo Scientific (Thermo biotin (Thermo Fisher Scientific) as described previously,47 and then Fisher Scientific, Rockford, IL) and the construct with a triple FLAG- subjected to energy depletion. Biotin that remained exposed at the baso- tag inserted at the amino terminus (AS160 WT-FLAG) was generated lateral cell surface was stripped through incubation with 100 mM according to previously described protocols.9,24 To create the MesNa for 40 minutes at 4°C, and cells were washed with PBS++ (PBS pcDNA3.1-Hygro (+) plasmid encoding the human AS160 WT- supplemented with 10 mM MgCl2 and 1 mM CaCl2). Cells were then FLAG, the AS160-FLAG sequence was inserted at NotIandBamHI re- lysed in 1 ml of lysis buffer (1% Triton, 150 mM NaCl, 50 mM Tris, pH striction sites. The construct encoding the AS160-S588D mutation was 7.4, and 1 mM EDTA, containing protease and phosphatase inhibitors), generated in this plasmid with the QuikChange XL site-directed muta- and incubated for 8 hours at 4°C with streptavidin-conjugated agarose genesis kit from Stratagene, and the mutations were verified by DNA beads (Pierce). Precipitated proteins were eluted from the beads through sequencing. The AS160 WT-FLAG and S588D-FLAG constructs were incubation in SDS-PAGE sample buffer supplemented with 100 mM transfected into the AS160 KD MDCK cell line with Lipofectamine 2000 dithiothreitol and analyzed by standard SDS-PAGE and Western immu- (Invitrogen). The selection and maintenance of stable MDCK cell clones noblotting. To assess the level of Na,K-ATPase expression, equal were performed in a-MEM containing 5 mg/ml G418 and Hygromycin amounts of total lysates were subjected to SDS-PAGE and Western im- B 0.4 mg/ml (Invitrogen). Clones were screened both for the reduced munoblotting using a mAb (a5) directed against an epitope of the expression levels of endogenous AS160 and for the expression of the a-subunit of Na,K-ATPase. Band density was quantified using ImageJ exogenous human AS160-FLAG by Western blotting. Constructs software (National Institutes of Health). encoding a dominant negative form of dynamin (dynamin K44A- GFP) and a plasma membrane marker corresponding to a PI (4,5) P2 Alkaline Phosphatase Treatment binding PH domain with a GFP tag (GFP-PH-PLCd1) were kindly pro- MDCK cells were plated on six-well tissue culture plates and allowed to vided by Dr. Pietro De Camilli (Yale University, New Haven, CT). polarize for 3–4daysafterreachingconfluence. Cells were subjected to energy depletion as described above and then lysed in 1 ml of lysis buffer: Immunofluorescence 1% Triton X-100, 150 mM NaCl, 50 mM Tris, and 1 mM EDTA, con- MDCK cells were plated on 12-mm Transwell filter inserts (Corning Life taining protease inhibitors. Cells were incubated with lysis buffer for 30 Sciences, Lowell, MA) and allowed to polarize for 3–4 days after reaching minutes at 4°C, and the soluble fraction was recovered after a centrifuga- confluence. Cells were fixed with 4% paraformaldehyde and subse- tion at 10,000g for 30 minutes at 4°C. The lysates were incubated with or quently permeabilized with PBS containing 1 mg/ml BSA and 0.3% without 10 units of calf intestinal alkaline phosphatase (Sigma-Aldrich) for Triton X-100. Nonspecific binding was blocked using goat serum di- 30minutesat37°C.Samplesthatdidnotreceive the calf intestinal alkaline lution buffer (33% goat serum, 40 mM NaPi, pH 7.4, 450 mM NaCl, phosphatase treatment were lysed in a buffer containing phosphatase in- and 0.6% Triton X-100). Primary and Alexa Fluor–conjugated second- hibitors. Lysates were dissolved in SDS-PAGE sample buffer, separated by ary (Invitrogen) antibodies were diluted in goat serum dilution buffer. SDS-PAGE electrophoresis and examined by Western blotting. Cells were visualized with a confocal laser scanning microscope (model LSM-780; Carl Zeiss Microimaging, Thornwood, NY). Contrast and Acute Renal Ischemia brightness settings were chosen so that all pixels were in the linear range. Animal studies were conducted in accordance with the Yale School of Images are the product of 8-fold line averaging. Medicine Institutional Animal Care and Use Committee. The AS160 KO mice had been generated by targeted deletion of part of exon 1 of Immunoprecipitation the AS160 gene as described by Lier et al.31 The 129/SvEv-C57BL/6 MDCK cells treated or untreated with the energy depletion protocol were TBC1D4-deficient mice were crossed with C57/BL6 WT mice to incubated with 1 ml of lysis buffer containing 150 mM NaCl, 50 mM Tris- generate a mixed 129/SvEv-C57BL/6 background. Eight-week-old HCl,pH7.4,1%Lubrol,5mMEDTA,proteaseinhibitors(SantaCruz male AS160 KO mice or WT littermates weighing 20–25 g were anes- Biotechnology, Santa Cruz, CA) and phosphatase inhibitors (Sigma- thetized with ketamine (100 mg/kg) by intraperitoneal injection and, Aldrich) for 30 minutes at 4°C. The insoluble fraction was removed using aseptic techniques, subjected to a laparotomy only (sham surgery) through centrifugation at 10,000g for 30 minutes at 4°C. After the cen- or a laparotomy with bilateral renal pedicle clamping (ischemia) for 30 trifugation, the supernatants were incubated with a5 antibody directed minutes. Ischemia was confirmed by color change observed in kidneys against the endogenous Na,K-ATPase and with protein G conjugated to after clamping. For analgesia, buprenorphine (0.05 mg/kg) was injected Sepharose (Pierce, Rockford, IL) for 8 hours at 4°C. Beads were washed subcutaneously before surgical procedures and then postoperatively four times with lysis buffer. Proteins were eluted in SDS-PAGE sample every 12 hours. Supportive fluids were given throughout the operative buffer and separated by SDS-PAGE electrophoresis and analyzed by West- period, and hypothermia was prevented by use of an isothermal heating ern blotting. Blots were then probed with peroxidase-conjugated mouse pad and warming lights. After 24 hours of reperfusion, the animals were secondary antibody and visualized with the enhanced chemilumines- once again anesthetized with ketamine (100 mg/kg) and subjected to cence reagent (GE Healthcare, Little Chalfont, Buckinghamshire, UK). laparotomy. Kidneys were collected after being snap frozen in situ.Fro- zen kidneys were stored at 280°C or immediately homogenized in 1 ml Cell Surface Biotinylation of cold homogenization buffer containing phosphatase and protease MDCK cells were plated on Transwell filter inserts (Corning, Corning, inhibitors (150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 1% Triton X-100, NY), and wereallowedtopolarizefor3–4 days after reaching confluence. and 5 mM EDTA). Homogenates were then prepared as previously

2774 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015 www.jasn.org BASIC RESEARCH published.48 Specimens were rotated for 1 hour at 4°C, centrifuged at washed with 1% SDS and then blocked with 0.1% BSA in 10% goat 16,000g for 5 minutes at 4°C to separate nuclei and tissue debris, and serum buffer. Primary antibody a5, directed against the sodium then ultracentrifuged at 100,000g for 60 minutes at 4°C. Protein con- pump, was applied at a dilution of 1:400, followed by Alexa Fluor centrations of supernatants were determined by the Bradford assay 488–conjugated goat anti-mouse secondary antibody at a 1:200 di- (Bio-Rad, Hercules, CA) and subjected to Western blot analysis. lution (Molecular Probes, Carlsbad, CA).

Fixation and Preparation of Kidney Tissue Statistical Analyses Kidney tissue was fixed according to our laboratory’s standard pro- All experiments were carried out in at least three independent tocol.49 Briefly, mice were fixed via whole-body ventricular perfusion replicates. Values are expressed as means6SEM. Comparisons be- with PBS, pH 7.4, for 15 seconds followed with freshly prepared 4% tween experimental groups were made using the t test. formaldehyde in PBS for 2 minutes. Kidneys were removed, dissected into transverse slices 1 mm in thickness, and postfixed for 1 hour at room temperature, followed by three 15-minute washes in PBS. The tissue was then prepared for cryostat sectioning or further postfixed ACKNOWLEDGMENTS in PBS with 1% glutaraldehyde followed by osmication and embed- ding in EPOX-812 resin. Epoxy-embedded tissue was sectioned at We thank Vanathy Rajendran, SueAnn Mentone, Tom Ardito, and 2 mm with a Reichert Ultracut-E ultramicrotome and stained Deren Shao for their invaluable technical support. We also thank all of with a mixture of 1% azure I and II in water and 0.5% sodium borate. the members of the Caplan Laboratory for their helpful discussions and suggestions. Analysis of Images of Renal Tissue This work was supported by the Leducq Foundation Transatlantic Random images of cortex sampling both proximal and distal tubules Network for Hypertension and by the National Institutes of Health were obtained from three experiments and labeled only with image (Grants DK-072612 and DK-17433). accession number, date, and time. Images were analyzed at a later time period in a randomly blinded fashion. All distal tubules in each image were counted. Images were scored based upon whether the subcellular DISCLOSURES distribution of the Na,K-ATPase was predominantly restricted to the None. basolateral plasma membrane or whether the Na,K-ATPase staining was predominantly distributed throughout the cytoplasm. Those with a predominantly cytoplasmic distribution were counted as a separate set REFERENCES and the ratio of tubules with a cytoplasmic distribution to the total number of distal tubules was obtained for each image. Over 100 tubule 1. Jørgensen PL: Sodium and potassium ion pump in kidney tubules. – cross-sections were scored for each condition. The experimental groups Physiol Rev 60: 864 917, 1980 2. Barlet-Bas C, Khadouri C, Marsy S, Doucet A: Enhanced intracellular were then correlated with date and time of accession and the results of all sodium concentration in kidney cells recruits a latent pool of Na-K- of the images of each group were combined to give totals of all tubules in ATPase whose size is modulated by corticosteroids. JBiolChem265: each group and the ratio and percentage was calculated. 7799–7803, 1990 3. Horisberger JD, Rossier BC: Aldosterone regulation of gene transcription leading to control of ion transport. Hypertension 19: 221–227, 1992 Wheat Germ Agglutinin Endocytosis Assay 4. Liang M, Tian J, Liu L, Pierre S, Liu J, Shapiro J, Xie ZJ: Identification of a MDCK cells were transfected or not with a plasmid encoding dynamin pool of non-pumping Na/K-ATPase. J Biol Chem 282: 10585–10593, 2007 fi K44A-GFP,platedonTranswell lterinserts(Corning),andwereallowed 5. Chibalin AV, Ogimoto G, Pedemonte CH, Pressley TA, Katz AI, Féraille E, to polarize for 3–4 days. Cells were treated at the apical surface with Berggren PO, Bertorello AM: Dopamine-induced endocytosis of Na+, 25 mg/ml WGA-647 conjugate (Invitrogen) for 30 minutes at 37°C. The K+-ATPase is initiated by phosphorylation of Ser-18 in the rat alpha sub- apical WGA-647 that remained bound to the apical surface after this unit and Is responsible for the decreased activity in epithelial cells. JBiol Chem 274: 1920–1927, 1999 D treatment was stripped with 0.1 M of N-acetyl- -glucosamine (Sigma- 6. Sweeney G, Klip A: Regulation of the Na+/K+-ATPase by insulin: Why Aldrich) for 10 minutes at 18°C with three different washes in a CO2 and how? Mol Cell Biochem 182: 121–133, 1998 independent media (Invitrogen). After a wash step in in PBS++, cells 7. Molitoris BA, Geerdes A, McIntosh JR: Dissociation and redistribution were then fixed with 4% paraformaldehyde and analyzed by confocal of Na+,K(+)-ATPase from its surface membrane actin cytoskeletal – microscopy, as described above. complex during cellular ATP depletion. JClinInvest88: 462 469, 1991 8. Coster AC, Govers R, James DE: Insulin stimulates the entry of GLUT4 into the endosomal recycling pathway by a quantal mechanism. Traffic Immunofluorescence Analyses of Mouse Kidneys and 5: 763–771, 2004 Measurement of Biochemical Parameters 9. Sano H, Kane S, Sano E, Mîinea CP, Asara JM, Lane WS, Garner CW, Serum creatinine measurements were performed by the Mouse Lienhard GE: Insulin-stimulated phosphorylation of a Rab GTPase- Metabolic Phenotyping Center at Yale University. Cryostat-prepared activating protein regulates GLUT4 translocation. J Biol Chem 278: 14599–14602, 2003 kidney sections were subjected to 15 minutes of antigen retrieval 10. Bruss MD, Arias EB, Lienhard GE, Cartee GD: Increased phosphoryla- through incubation in 10 mM citrate buffer, pH 6.0. After multiple tion of Akt substrate of 160 kDa (AS160) in rat skeletal muscle in re- – washes with PBS and quenching with 0.5 M NH4Cl, sections were sponse to insulin or contractile activity. Diabetes 54: 41 50, 2005

J Am Soc Nephrol 26: 2765–2776, 2015 AS160 and Sodium Pump Trafﬁcking 2775 BASIC RESEARCH www.jasn.org

11. Mîinea CP, Sano H, Kane S, Sano E, Fukuda M, Peränen J, Lane WS, 31. Lier N, Gresko N, Di Chiara M, Loffing-Cueni D, Loffing J: Immunoflu- Lienhard GE: AS160, the Akt substrate regulating GLUT4 translocation, orescent localization of the Rab-GAP protein TBC1D4 (AS160) in mouse has a functional Rab GTPase-activating protein domain. Biochem J 391: kidney. Histochem Cell Biol 138: 101–112, 2012 87–93, 2005 32. Slocum JL, Heung M, Pennathur S: Marking renal injury: Can we move 12. Eguez L, Lee A, Chavez JA, Miinea CP, Kane S, Lienhard GE, McGraw beyond serum creatinine? Transl Res 159: 277–289, 2012 TE: Full intracellular retention of GLUT4 requires AS160 Rab GTPase 33. Molitoris BA, Falk SA, Dahl RH: Ischemia-induced loss of epithelial activating protein. Cell Metab 2: 263–272, 2005 polarity. Role of the tight junction. JClinInvest84: 1334–1339, 1989 13. Larance M, Ramm G, Stöckli J, van Dam EM, Winata S, Wasinger V, 34. Molitoris BA, Hoilien CA, Dahl R, Ahnen DJ, Wilson PD, Kim J: Char- Simpson F, Graham M, Junutula JR, Guilhaus M, James DE: Character- acterization of ischemia-induced loss of epithelial polarity. JMembr ization of the role of the Rab GTPase-activating protein AS160 in insulin- Biol 106: 233–242, 1988 regulated GLUT4 trafficking. JBiolChem280: 37803–37813, 2005 35. Tian J, Li X, Liang M, Liu L, Xie JX, Ye Q, Kometiani P, Tillekeratne M, Jin 14. Liang X, Butterworth MB, Peters KW, Frizzell RA: AS160 modulates R, Xie Z: Changes in sodium pump expression dictate the effects of aldosterone-stimulated epithelial sodium channel forward trafficking. ouabainoncellgrowth.J Biol Chem 284: 14921–14929, 2009 MolBiolCell21: 2024–2033, 2010 36. Dada LA, Sznajder JI: Hypoxic inhibition of alveolar fluid reabsorption. 15. Mandel LJ, Doctor RB, Bacallao R: ATP depletion: A novel method to Adv Exp Med Biol 618: 159–168, 2007 study junctional properties in epithelial tissues. II. Internalization of Na 37. Dada LA, Chandel NS, Ridge KM, Pedemonte C, Bertorello AM, +,K(+)-ATPase and E-cadherin. J Cell Sci 107: 3315–3324, 1994 Sznajder JI: Hypoxia-induced endocytosis of Na,K-ATPase in alveolar 16. Brodie JC, Humes HD: Stem cell approaches for the treatment of renal epithelial cells is mediated by mitochondrial reactive oxygen species failure. Pharmacol Rev 57: 299–313, 2005 and PKC-zeta. JClinInvest111: 1057–1064, 2003 17. Thadhani R, Pascual M, Bonventre JV: Acute renal failure. NEnglJMed 38. Yudowski GA, Efendiev R, Pedemonte CH, Katz AI, Berggren PO, 334: 1448–1460, 1996 Bertorello AM: Phosphoinositide-3 kinase binds to a proline-rich motif 18. Schrier RW, Wang W, Poole B, Mitra A: Acute renal failure: Definitions, in the Na+, K+-ATPase alpha subunit and regulates its trafficking. Proc diagnosis, pathogenesis, and therapy. JClinInvest114: 5–14, 2004 Natl Acad Sci U S A 97: 6556–6561, 2000 19. Gaudio KM, Thulin G, Ardito T, Kashgarian M, Siegel NJ: Metabolic 39. Tzatsos A, Tsichlis PN: Energy depletion inhibits phosphatidylinositol alterations in proximal tubule suspensions obtained from ischemic 3-kinase/Akt signaling and induces apoptosis via AMP-activated pro- kidneys. Am J Physiol 257: F383–F389, 1989 tein kinase-dependent phosphorylation of IRS-1 at Ser-794. JBiol 20. Molitoris BA, Chan LK, Shapiro JI, Conger JD, Falk SA: Loss of epithelial Chem 282: 18069–18082, 2007 polarity: A novel hypothesis for reduced proximal tubule Na+ transport 40. Towler MC, Hardie DG: AMP-activated protein kinase in metabolic following ischemic injury. J Membr Biol 107: 119–127, 1989 control and insulin signaling. Circ Res 100: 328–341, 2007 21. Riordan M, Sreedharan R, Wang S, Thulin G, Mann A, Stankewich M, 41. Sun Y, Bilan PJ, Liu Z, Klip A: Rab8A and Rab13 are activated by insulin Van Why S, Kashgarian M, Siegel NJ: HSP70 binding modulates and regulate GLUT4 translocation in muscle cells. Proc Natl Acad Sci U detachment of Na-K-ATPase following energy deprivation in renal SA107: 19909–19914, 2010 epithelial cells. Am J Physiol Renal Physiol 288: F1236–F1242, 2005 42. Comellas AP, Kelly AM, Trejo HE, Briva A, Lee J, Sznajder JI, Dada LA: 22. van Why SK, Kim S, Geibel J, Seebach FA, Kashgarian M, Siegel NJ: Insulin regulates alveolar epithelial function by inducing Na+/K+-ATPase Thresholds for cellular disruption and activation of the stress response translocation to the plasma membrane in a process mediated by the in renal epithelia. Am J Physiol 277: F227–F234, 1999 action of Akt. JCellSci123: 1343–1351, 2010 23. Alves DS, Farr GA, Seo-Mayer P, Caplan MJ: AS160 associates with 43. Molitoris BA, Dahl R, Geerdes A: Cytoskeleton disruption and apical the Na+,K+-ATPase and mediates the adenosine monophosphate- redistribution of proximal tubule Na(+)-K(+)-ATPase during ischemia. stimulated protein kinase-dependent regulation of sodium pump surface Am J Physiol 263: F488–F495, 1992 expression. Mol Biol Cell 21: 4400–4408, 2010 44. Alejandro VS, Nelson WJ, Huie P, Sibley RK, Dafoe D, Kuo P, Scandling JD 24. Kane S, Sano H, Liu SC, Asara JM, Lane WS, Garner CC, Lienhard GE: A Jr, Myers BD: Postischemic injury, delayed function and Na+/K(+)-ATPase method to identify serine kinase substrates. Akt phosphorylates a novel distribution in the transplanted kidney. Kidney Int 48: 1308–1315, 1995 adipocyte protein with a Rab GTPase-activating protein (GAP) domain. 45. Tamkun MM, Fambrough DM: The (Na+ + K+)-ATPase of chick sensory JBiolChem277: 22115–22118, 2002 neurons. Studies on biosynthesis and intracellular transport. JBiol 25. Cartee GD, Wojtaszewski JF: Role of Akt substrate of 160 kDa in insulin- Chem 261: 1009–1019, 1986 stimulated and contraction-stimulated glucose transport. Appl Physiol 46. Seo-Mayer PW, Thulin G, Zhang L, Alves DS, Ardito T, Kashgarian M, Nutr Metab 32: 557–566, 2007 Caplan MJ: Preactivation of AMPK by metformin may ameliorate the 26. Stöckli J, Davey JR, Hohnen-Behrens C, Xu A, James DE, Ramm G: epithelial cell damage caused by renal ischemia. Am J Physiol Renal Regulation of glucose transporter 4 translocation by the Rab guanosine Physiol 301: F1346–F1357, 2011 triphosphatase-activating protein AS160/TBC1D4: Role of phosphor- 47. Gottardi CJ, Dunbar LA, Caplan MJ: Biotinylation and assessment of ylation and membrane association. Mol Endocrinol 22: 2703–2715, membrane polarity: Caveats and methodological concerns. Am J 2008 Physiol 268: F285–F295, 1995 27. Cai T, Wang H, Chen Y, Liu L, Gunning WT, Quintas LE, Xie ZJ: Regu- 48. Mount PF, Hill RE, Fraser SA, Levidiotis V, Katsis F, Kemp BE, Power DA: lation of caveolin-1 membrane trafficking by the Na/K-ATPase. JCell Acute renal ischemia rapidly activates the energy sensor AMPK but Biol 182: 1153–1169, 2008 does not increase phosphorylation of eNOS-Ser1177. Am J Physiol 28. Liu J, Liang M, Liu L, Malhotra D, Xie Z, Shapiro JI: Ouabain-induced Renal Physiol 289: F1103–F1115, 2005 endocytosis of the plasmalemmal Na/K-ATPase in LLC-PK1 cells re- 49. Pluznick JL, Zou DJ, Zhang X, Yan Q, Rodriguez-Gil DJ, Eisner C, Wells quires caveolin-1. Kidney Int 67: 1844–1854, 2005 E, Greer CA, Wang T, Firestein S, Schnermann J, Caplan MJ: Functional 29. Stauffer TP, Ahn S, Meyer T: Receptor-induced transient reduction in expression of the olfactory signaling system in the kidney. Proc Natl plasma membrane PtdIns(4,5)P2 concentration monitored in living Acad Sci U S A 106: 2059–2064, 2009 cells. Curr Biol 8: 343–346, 1998 30. Aufricht C, Bidmon B, Ruffingshofer D, Regele H, Herkner K, Siegel NJ, Kashgarian M, Van Why SK: Ischemic conditioning prevents Na,K-ATPase dissociation from the cytoskeletal cellular fraction after repeat renal This article contains supplemental material online at http://jasn.asnjournals. ischemia in rats. Pediatr Res 51: 722–727, 2002 org/lookup/suppl/doi:10.1681/ASN.2013101040/-/DCSupplemental.

2776 Journal of the American Society of Nephrology J Am Soc Nephrol 26: 2765–2776, 2015

Supplemental Data 2

Supplemental Data 2. Renal ischemia-induced redistribution of Na,K-ATPase is reduced in the DCTs of AS160 knockout mice, low magnification images. Images depicting the distribution of the Na,K-ATPase in renal tissue derived from wild type (WT) and AS160 knockout (AS160 KO) mice subjected to sham surgery (Sham) or to ischemia followed by 24 hours of reperfusion (24h). The scale bars in lower left hand corners of each of the four central panels correspond to 50 µ. The insets correspond to higher magnifications of the indicated portions of the fields in each of the four central panels. Na,K-ATPase labeling associated with basolateral plasma membrane infoldings is present under sham conditions in tissue derived from both genotypes, as well as in the tissue derived from AS160 knockout mice subjected to ischemia and 24 hours of reperfusion. This pattern is lost in tissue derived from wild type mice subjected to ischemia and 24 hours of reperfusion, and is replaced by a largely intracellular pattern of staining.

Supplemental Data 3

Supplemental Data 3. Na,K-ATPase mis-localization after renal injury is similar in the proximal tubules of WT and AS160 knockout mice. Immunofluorescence images of kidney cortex stained with an antibody directed against Na,K-ATPase -subunit (5). Wild type (WT) and AS160 KO (KO) mice were subjected to 30 min of bilateral renal pedicle clamping followed by 24 h of reperfusion (24h R). Na,K-ATPase appears to be redistributed Supplemental Data 4

100 WT KO NS

20 NS DCT Tunel staining/nucleus (a.u.) staining/nucleus DCT Tunel

0 Shamsham 24h24h R R

Supplemental Data 4. TUNEL analysis indicates that WT and AS160 KO mice exhibit similar levels of post-ischemic apoptosis. Cryostat sections from WT and AS160 KO animals subjected to sham or ischemia surgery followed by 24hs of reperfusion were analyzed by Tunel staining and the prevalence of positive-staining DCT cell nuclei was quantified in confocal images. Wild type and AS160 KO mice show similar levels of apoptosis in DCTs. NS: not significant.