Combined Proteomics and Pathways Analysis of Collecting Duct Reveals a Protein Regulatory Network Activated in Vasopressin Escape

Basic Science Articles

Ewout J. Hoorn, Jason D. Hoffert, and Mark A. Knepper Laboratory of Kidney and Electrolyte Metabolism; National Heart, Lung, and Blood Institute; National Institutes of Health, Bethesda, Maryland

Low sensitivity is characteristic of many proteomics methods. Presented here is an approach that combines proteomics based on difference gel electrophoresis (DIGE) with bioinformatic pathways analysis to identify both abundant and relatively nonabundant proteins in inner medullary collecting duct (IMCD) altered in abundance during escape from vasopressin- induced antidiuresis. Rats received the vasopressin analog dDAVP by osmotic minipump plus either a daily water load (vasopressin escape) or only enough water to replace losses (control). Immunoblotting confirmed the hallmark of vasopressin escape, a decrease in aquaporin-2, and demonstrated a decrease in the abundance of the urea transporter UT-A3. DIGE identified 22 mostly high-abundance proteins regulated during vasopressin escape. These proteins were analyzed using pathways analysis software to reveal protein clusters incorporating the proteins identified by DIGE. A single dominant cluster emerged that included many relatively low-abundance proteins (abundances too low for DIGE identification), including several transcription factors. Immunoblotting confirmed a decrease in total and phosphorylated c-myc, a decrease in c-fos, and increases in c-jun and p53. Furthermore, immunoblotting confirmed hypothesized changes in other proteins in the proposed network: Increases in c-src, receptor for activated C kinase 1, calreticulin, and caspase 3 and decreases in steroid receptor co-activator 1, Grp78/BiP, and annexin A4. This combined approach proved capable of uncovering regulatory proteins that are altered in response to a specific physiologic perturbation without being detected directly by DIGE. The results demonstrate a dominant protein regulatory network in IMCD cells that is altered in association with vasopressin escape, providing a new framework for further studies of signaling in IMCD. J Am Soc Nephrol 16: 2852–2863, 2005. doi: 10.1681/ASN.2005030322

scape from the antidiuretic action of vasopressin (“va- inner medullary collecting ducts (IMCD) to produce cAMP sopressin escape”) is an important physiologic process in response to vasopressin (7) in association with a fall in E that limits the severity of the syndrome of inappropri- vasopressin-binding capacity of the V2R receptor (8). In con- ate antidiuresis (SIADH) and other hyponatremic disorders trast, there is upregulation of AQP-3 (2) and the ␣-subunit of (1,2). During vasopressin escape, humans and experimental the epithelial Na channel (ENaC) (9), suggesting that the animals undergo a brisk water diuresis, despite high circulating process that is responsible for suppression of AQP2 expres- levels of vasopressin (2–4). Vasopressin escape is of consider- sion is selective. With regard to possible mediators of escape, able clinical importance with regard to SIADH and other vaso- roles have been suggested for nitric oxide and prostaglan- pressin-dependent dilutional hyponatremic states, because dins (10) and aldosterone (11). However, little is known without escape, further water retention and hyponatremia about which intracellular signaling processes orchestrate the could be fatal (4–6). escape and how the vasopressin escape phenomenon is trig- Studies in a rat model of vasopressin escape have demon- gered and maintained. strated that the central feature of the vasopressin-indepen- To address which proteins are co-regulated with AQP2 in dent increase in water excretion is a marked suppression of IMCD during vasopressin escape, we used a relatively new the expression of the water channel aquaporin-2 (AQP2) (2). differential proteomics method called difference gel electro- The fall in AQP2 protein abundance is due in part to de- phoresis (DIGE) coupled with matrix-assisted laser desorp- creased levels of AQP2 mRNA in collecting duct (2). This tion/ionization time-of-flight (MALDI-TOF) mass spectrome- response is associated with a decrease in the capacity of try to identify differentially expressed proteins. Recently, we demonstrated the feasibility of such an approach using DIGE- Received March 26, 2005. Accepted June 25, 2005. based proteomics to identify vasopressin-responsive proteins

Published online ahead of print. Publication date available at www.jasn.org. in the collecting duct of Brattleboro rats that were treated with the selective V2R agonist dDAVP (12). Address correspondence to: Dr. Mark A. Knepper, National Institutes of Health, 10 Center Drive, Building 10, Room 6N260, Bethesda, MD 20892. Phone: 301-496- To address further the signaling pathways that are activated 3064; Fax: 301-402-1443; E-mail: [email protected] during vasopressin escape, we analyzed the DIGE results using

Table 1. IMCD proteins regulated at early (day 1) or later (day 4) stages of vasopressin escapea

Protein Abundance Ratiob MW GenBank (Escape/Control) Protein Identification (kD) pI No.

Decreased early in vasopressin escape (day 1) 0.39 ATP synthase ␤ 50 5.0 gi͉54792127 0.40 keratin complex 2, gene 8 51 5.7 gi͉40786432 0.42 mitochondrial aconitase 86 7.9 gi͉10637996 0.43 NADH dehydrogenase 81 5.5 gi͉51858651 0.51 ␣ B-crystallin 20 6.3 gi͉16905067 0.58 GRP 75, stress 70 protein 75 5.9 gi͉55584140 0.62 ␤-tubulin-1 51 4.9 gi͉56754676 0.63 annexin A2 39 9.2 gi͉584760 0.66 malate dehydrogenase 1 37 6.2 gi͉15100179 0.71 F-actin capping protein ␤ 31 5.7 gi͉53734561 0.73 ␣ enolase 52 6.7 gi͉56757324 0.74 pyruvate kinase 58 7.2 gi͉125601 0.79 transketolase 68 7.2 gi͉12018252 0.82 eukaryotic translation elongation factor 2 96 6.4 gi͉8393296 0.82 ␣-1-antiproteinase 46 5.7 gi͉1703024 0.83 prohibitin 29 5.6 gi͉34873234 0.88 Sec23B 87 6.5 gi͉34858849 0.90 T-complex protein 1 58 6.6 gi͉6981642 Decreased later in vasopressin escape (day 4) 0.71 pyruvate kinase 58 7.2 gi͉125601 0.76 heat-shock protein 70 70 5.5 gi͉450932 0.81 transketolase 68 7.2 gi͉12018252 Increased early in vasopressin escape (day 1) 1.28 calreticulin 48 4.3 gi͉11693172 1.93 pyruvate dehydrogenase ␤ 35 5.6 gi͉1352624 Spot shifted early in vasopressin escape (day 1) 0.19/25.15 protein disulfide-isomerase 57 5.9 gi͉1352384

aIMCD, inner medullary collecting duct; MW, molecular weight; pI, isoelectric point. bCalculated from average protein abundance ratios for four pairs of samples. Only protein spots that showed statistically significant changes by difference gel electrophoresis analysis were picked for matrix-assisted laser desorption/ionization time- of-flight/time-of-flight identification. All of the identified proteins were in the expected ranges of pI and MW in the two- dimensional gel in Figure 3. bioinformatic pathways analysis (13,14). The pathways anal- Materials and Methods ysis software generates hypothetical protein networks, based Animals and Sample Preparation on large databases of protein interactions culled from the Osmotic minipumps (model 2001; Alzet, Palo Alto, CA) that deliver biologic literature, including physical binding reactions, cis- 5 ng/h dDAVP (Peninsula Laboratories, San Carlos, CA; ACUC Pro- trans interactions in transcriptional regulation, and enzyme– tocol 2-KE-3) were implanted subcutaneously in Male Sprague-Dawley substrate relationships. Such networks can be used to predict rats (Taconic Farms, Germantown, NY). After 3 d, rats were divided signaling pathways that are activated during vasopressin into two groups: “escape” and “control.” Escape rats were given excess escape. The rationale for using the pathways analysis ap- daily water (0.4 ml/g body wt) via a gelled-agar diet (71% water, 28% finely ground rat food, 1% agar; BACTO-AGAR; Difco Laboratories, proach was not only to facilitate the interpretation of the Detroit, MI). This diet forced the rats to take the water load to consume relationships between the identified proteins but also to the food ration. Control rats were given the same amount of food but identify relatively low-abundance proteins (abundances too with only enough water (0.075 ml/g body wt) to compensate for low for DIGE identification) that may be involved in vaso- insensible losses plus 0.015 ml/g body wt per d urine. Rats did not pressin escape (15). The hypothetical changes in these low- receive ad libitum water. This represents a small modification from our abundance proteins then can be tested by immunoblotting. previous studies in which control rats received no water mixed with the This integrated approach identified a single dominant net- food but were allowed to receive ad libitum water (2,7,9). The rats were work of proteins that includes several proteins that may play maintained in metabolic cages in a temperature- and humidity-con- key regulatory roles in vasopressin escape. trolled room with a 12:12-h light-dark cycle, and urine was collected 2854 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2852–2863, 2005 daily for measurement of volume and osmolality. Because the onset of escape is known to occur between 1 and 2 d after the start of water loading (2), three time points were analyzed (four control and four escape for each time point): After 1 and2dofwater loading (early stages of escape) and after4dofwater loading (late stages of escape). Thus, a total of 24 rats were separated into 12 control rats and 12 escape rats, and four versus four rats were selected arbitrarily for IMCD analysis at each of the three time points. The day 1 and day 4 time points were selected for DIGE analysis, whereas all three time points were analyzed by immunoblotting. IMCD suspensions were prepared using the method of Stokes et al. (16) with some modifications (17) (Supple- mental Materials available online).

Semiquantitative Immunoblotting Immunoblotting was carried out as described previously (18) (see Supplemental Materials). The IMCD pellet was solubilized in 5ϫ Lae- mmli sample buffer (1 vol per 4 vol of sample) followed by heating to 60°C for 15 min before electrophoresis. Equal loading was confirmed by staining gels loaded for all three time points (24 samples) with Coo- massie blue (18). This loading gel was scanned with a linear fluorescence scanner (Odyssey; Li-Cor Biosciences, Lincoln, NE) at an excita- tion wavelength of 700 nm (Supplemental Figure 1 available online).

DIGE DIGE analysis was carried out (for day 1 and day 4 time points, each four versus four samples) as described previously (12,19) (see Supple- mental Materials). Briefly, before two-dimensional (2-D) gel electrophoresis, IMCD proteins were solubilized in 2-D sample buffer (7 M urea, 2 M thiourea, 30 mM Tris Cl, and 4% CHAPS [pH 8.5]). The samples were labeled on lysine side chains with Cy3- (control), Cy5- (escape), or Cy2- (mixture of control ϩ escape samples, internal stan- dard) fluorophores using N-hydroxysuccinimide chemistry (Amer- sham). Isoelectric focusing was performed using an IPGphor apparatus (Amersham, Piscataway, NJ). Isoelectric focusing strips were loaded onto Ettan DALT-6 electrophoresis unit (Amersham) and further separated on a 10% SDS-PAGE gel (5 W/gel). Fluorescence analytical gel images were obtained (Typhoon scanner; 100 ␮m resolution; Amersham) using the following emission filters: Cy2 (520 BP 40), Cy3 (580 BP 30), and Cy5 (670 BP 30). Spot matching, quantification, and statistical analyses were performed using DeCyder software (Version 5.0; Amersham). The corresponding Cy3 (control) Figure 1. Urine excretion rate, urine osmolality, plasma sodium and Cy5 (escape) images were normalized to the pooled internal stan- concentration, and plasma urea concentration during vasopres- dard (Cy2) for that gel using a Least-Means-Squared-Gradient-Descent sin escape. (A) Urine excretion rate over the course of the algorithm. One gel was chosen as the “master,” and all remaining experiment with water loading commencing on day 0. Urine analytical gels were matched and normalized to the Cy2 master spot excretion rate was increased significantly from day 2 in the map. The resulting protein abundance ratios, now represented as stan- escape group relative to control (four versus four rats per time dardized log abundance values, were compared using an unpaired t point). Negative time points represent the equilibration period. test. The inverse log of these values is presented in Table 1 as protein (B) Urine osmolality over the course of the experiment. Osmo- abundance ratio. A protein abundance ratio Ͼ1 corresponds to an lality was decreased significantly from day 1 to the end of the increase in escape compared with control samples, whereas a ratio Ͻ1 experiment in the escape group relative to control (four versus corresponds to a decrease in escape (significance criterion, P Յ 0.05). four rats per time point). (C) Plasma sodium concentrations at For picking, gels were fixed in 30% ethanol/7.5% acetic acid for 2 h days 1, 2, and 4. Plasma sodium was significantly lower in the followed by SYPRO Ruby (610 BP 30) staining overnight for total escape group relative to control on all 3 d (four versus four rats protein visualization. per time point). (D) Plasma urea concentrations at days 1 and 4. A robotic workstation (Ettan; Amersham) was used to excise protein Plasma urea was significantly lower in the escape group rela- spots, perform in-gel tryptic digestion, extract peptides from the gel, tive to control on both days (four versus four rats per time and transfer the extracts onto a MALDI substrate. Spectra were ac- point). *P Ͻ 0.05 for all. quired with an ABI 4700 MALDI-TOF/TOF mass spectrometer, and proteins were identified by database matching using Mascot. J Am Soc Nephrol 16: 2852–2863, 2005 Proteomics of Vasopressin Escape 2855

Figure 2. Immunoblots showing changes in abundances of aquaporin 2 (AQP2), ␣ subunit of epithelial Na channel (␣-ENaC), and collecting duct urea transporters in inner medullary collecting ducts (IMCD) from rats that underwent vasopressin escape versus control rats. Immunoblots are of IMCD cells purified from rat renal medullas at late stage of vasopressin escape (day 4 time point). Each lane is loaded with a sample from a different rat (n ϭ 4 rats per treatment). Thirty micrograms of total protein was loaded in each lane, and the resulting immunoblots were probed with anti-AQP2 antibody (L127), anti–␣-ENaC (Q3560-2), anti–UT-A3 (Q2695-2), or anti–UT-A1 (L403). To the right of each immunoblot is a bar graph showing densitometry values for all three time points studied. *P Ͻ 0.05.

Bioinformatic Pathways Analysis differentially expressed proteins identified in the DIGE experiment Regulated proteins identified by DIGE were analyzed further by is uploaded into IPA. IPA then builds hypothetical networks from bioinformatic pathways analysis (Ingenuity Pathway Analysis [IPA]; these proteins, and other non–DIGE-identified proteins from the Ingenuity Systems, Mountain View, CA; www.ingenuity.com). IPA database that are needed fill out a protein cluster. Network gener- constructs hypothetical protein interaction clusters on the basis of a ation is optimized for inclusion of as many proteins from the input- regularly updated “Ingenuity Pathways Knowledge Base.” The In- ted expression profile as possible and aims for highly connected genuity Pathways Knowledge Base is a very large curated database networks. that consists of millions of individual relationships between pro- IPA computes a score for each possible network according to the fit teins, culled from the biologic literature. These relationships involve of that network to the inputted proteins. The score is calculated as the direct protein interactions, including physical binding interactions, negative base-10 logarithm of the P value that indicates the likelihood enzyme substrate relationships, and cis-trans relationships in tran- of the inputted proteins in a given network being found together as a scriptional control. The networks are displayed graphically as nodes result of random chance. Therefore, scores of 2 or higher have at least (individual proteins) and edges (the biologic relationships between a 99% confidence of not being generated by random chance alone. For the nodes). previous studies using IPA, see Siripurapu et al. (13) and Raponi et al. In practice, a data set that contains the GenBank identifiers of (14). 2856 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2852–2863, 2005

Results Changes in Rat IMCD Proteome in Vasopressin Escape Verifying Vasopressin Escape Figure 3 shows an example of a DIGE gel with superim- In this model of vasopressin escape, both control and exper- position of Cy3 (control, red) and Cy5 (escape, green) images imental rats received a continuous dDAVP infusion starting on of the gel. Spots corresponding to proteins expressed at day Ϫ3, but only the experimental rats received a daily water nearly equal levels in the two samples appear yellow, those load, mixed with the food, starting on day 0. As previously upregulated in response to vasopressin escape appear green, noted (2,7,9), rats began to “escape” from dDAVP-induced and those downregulated in response to vasopressin escape antidiuresis on the second day, i.e., 24 to 48 h after initiation of appear red. Flanking the 2-D gel image are 3-D pixel density water loading, as evidenced by a marked increase in urine plots for nine selected proteins identified by MALDI-TOF/ excretion rate (Figure 1A). Urine osmolality (Figure 1B) was re- TOF mass spectrometry, including heat-shock protein 70 duced significantly in escape on the second day of water loading. (HSP70), ATP synthase, calreticulin, prohibitin, mitochon- Plasma sodium levels (Figure 1C) showed an acute decrease be- drial aconitase, Sec23B, annexin A2, malate dehydrogenase, Ϯ Ϯ tween days 1 and 2 (from 137 2to105 3 mmol/L) and, and protein disulfide isomerase (PDI). PDI had an apparent Ϯ subsequently, a partial recovery on day 4 (115 3 mmol/L) in shift in isoelectric point, suggestive of a posttranslational response to vasopressin escape. Plasma urea levels were signifi- modification. Only protein spots with statistically significant Ϯ cantly lower in escape animals both at early (day 1: 7.0 0.7 versus abundance ratios (P Յ 0.05 for four pairs of samples) were 5.0 Ϯ 0.4 mmol/L) and late (day 4: 7.0 Ϯ 0.5 versus 5.0 Ϯ 0.4 selected for MALDI-TOF/TOF identification. Moreover, only mmol/L) stages of vasopressin escape (Figure 1D). those identifications with expectation values (i.e., an indica- tor of the degree of certainty of an identification) larger than Changes in Abundances of Transport Proteins in IMCD 95% were accepted. A total of 22 protein spots were identi- Figure 2 shows immunoblots for AQP2, ␣-ENaC, and collect- fied by MALDI-TOF/TOF mass spectrometry (Table 1). More ing duct urea transporters in IMCD on day 4 of vasopressin proteins with altered abundance levels were identified at escape and densitometry values for all three time points. day 1 than at day 4. Those identified included proteins that Downregulation of AQP2 and upregulation of ␣-ENaC is con- were downregulated and upregulated and one protein that sistent with previous studies (7,9). A novel finding was that shifted position in the gel (Table 1; Figure 3) in response to UT-A3 showed a 50% decrease in band density on day 2 of vasopressin escape. Also listed in Table 1 are the theoretical vasopressin escape and was decreased further on day 4. Con- molecular weights and theoretical isoelectric points. The mo- versely, UT-A1 did not show a significant decrease. lecular weights and isoelectric points for all of these proteins

Figure 3. Two-dimensional (2-D) gel showing changes in the IMCD proteome in vasopressin escape. Superimposed images from samples labeled with Cy3 (control, red pseudocolor), and Cy5 (escape, green pseudocolor) and 3-D representation of spot intensities. Spots that appear red or green represent proteins that are respectively down- or upregulated in vasopressin escape, whereas proteins that are equally abundant in both samples appear yellow. Full range of horizontal axis is from 3 pH units (left) to 10 pH units (right). Full range of vertical axis is 15 kD (bottom) to 120 kD (top). pI, isoelectric point; MW, molecular weight. J Am Soc Nephrol 16: 2852–2863, 2005 Proteomics of Vasopressin Escape 2857

Figure 4. Protein regulatory network associated with vasopressin escape. Protein regulatory network was generated by bioinformatic pathways analysis through the use of the Ingenuity Pathways Analysis (IPA) software. Proteins that are listed in Table 1 were analyzed. Individual proteins are displayed as nodes, using different shades of gray to represent how the protein was identified and studied. In addition, different shapes are used to represent the functional class of the gene product (see figure insert). The edges describe the nature of the relationship between the nodes: An edge with arrowhead means that protein A acts on protein B, whereas an edge without an arrowhead represents binding only between two proteins. P indicates phosphorylation as a special case of the former. The gene names associated with the proteins are shown; see Table 2 for glossary. See Supple- mental Materials for detailed descriptions of individual protein–protein interactions, including literature references. The overall score for the depicted network was 34, indicating that the probability of matching the indicated proteins by a purely random event Ϫ was 10 34. 2858 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2852–2863, 2005 matched those derived from the spot position on the gel, by DIGE only are represented as light gray nodes, whereas pro- providing additional verification of the identifications. teins that were identified by DIGE and studied further by immunoblotting are represented as black nodes. Proteins that were Pathways Analysis of Vasopressin Escape identified by IPA only are represented as white nodes, whereas Figure 4 shows the largest protein cluster that was generated proteins that were identified by IPA and studied further by im- by the pathways analysis of the proteins listed in Table 1. This munoblotting are represented as dark gray nodes. Finally, the “vasopressin escape cluster” consists of a network of 33 pro- index protein AQP2 is shown. Because we were chiefly interested teins, including eight of the 22 proteins that were identified by in identifying candidate proteins for follow-up by immunoblot- DIGE-based proteomics and 25 additional proteins that were ting, we did not discriminate between the early and late time recognized as being related because of their reported interac- points for the pathways analysis. Annotation of the interactions in tions with the proteins identified by DIGE. The nodes represent Figure 4 is provided as Supplemental Materials. individual proteins listed by gene name (see Table 2 for glossary), whereas the edges represent the interactions, which include direct Confirmation by Semiquantitative Immunoblotting physical binding, substrate–enzyme interactions, and/or cis-/ Four of the 22 proteins that were identified by DIGE were trans relationship in transcriptional regulation. In Figure 4, nodes selected for semiquantitative immunoblotting on the basis of are displayed using different gray levels that represent how the the availability of high-quality antibodies. These were PDI, protein was identified and studied. Proteins that were identified calreticulin, ␤-tubulin, and HSP70 (Figure 5). Although DIGE

Table 2. Glossary: Protein regulatory network associated with vasopressin escape

Gene Name Protein Name Origin Protein Identification

ANXA2 Annexin A2 DIGE ANXA4 Annexin A4 DIGE AQP2 Aquaporin-2 IPA ATF6 Activating transcription factor 6 IPA CALR Calreticulin DIGE CANX Calnexin IPA CAPZB F-actin capping protein beta DIGE CASP3 Caspase 3 IPA CREB1 cAMP responsive element binding protein 1 IPA CRYAB Alpha B crystallin IPA EIF2AK3 Eukaryotic translation initiation factor 2 IPA EEF2 Eukaryotic translation elongation factor 2 DIGE FOS c-Fos IPA GNB2L1 Receptor for activated kinase C (RACK1) IPA GPI Glucose phosphate isomerase IPA HAS2 Hyaluronan synthase 2 IPA HSPA1 Heat shock protein 70 (HSP70) DIGE HSPA5 Glucose regulated protein 78 (GRP78) IPA IGK Immunoglobin kappa locus IPA JUN c-Jun IPA KRT8 Keratin 8 DIGE LAMP2 Lysosomal associated membrane protein 2 IPA MYC Myelocytomatosis oncogen (c-myc) IPA NCOA1 Steroid receptor co-activator 1 (SRC-1) IPA NR3C1 Glucocorticoid receptor IPA PRKACA Protein kinase A IPA RARA Retinoic acid receptor alpha IPA SEC23B Transport protein Sec-23 B DIGE SMT3B Ubiquitin-like protein SMT3B IPA SRC Sarcoma-oncogen IPA TAGLN2 Transgelin 2 IPA TP53 Tumor protein 53 IPA XBP1 X-box binding protein 1 IPA

IPA, ingenuity pathways analysis. J Am Soc Nephrol 16: 2852–2863, 2005 Proteomics of Vasopressin Escape 2859

c-src, and c-jun; Figure 7B). The abundances of the proteins shown in Figures 5, 6, and 7 were changed only at the reported time point and were unchanged at the other time points (data not shown). To address whether the protein abundance changes are specific for the IMCD or occur in other tissues, we immunoblotted renal cortex and brain samples from the same experiments and probed with selected antibodies (Figure 8). Among the responding proteins in IMCD, only c-src showed a similar response in renal cortex, and only HSP70 showed a similar response in brain.

Discussion Proteomics analysis is seeing increasing use as a means of identifying new mechanistic hypotheses in physiology (20). An important disadvantage of all 2-D gel-based proteomics approaches is that low-abundance proteins, including many cell- Figure 5. Immunoblots confirming selected protein abundance signaling proteins, are unlikely to be identified (15,21). Hence, changes identified by DIGE-based proteomics. IMCD suspen- alternative approaches are needed to discover transcription sions, prepared at the day 1 and day 4 time points. Each lane is factors and other regulatory proteins that are involved in the loaded with a sample from a different rat (n ϭ 4 rats/treat- responses to physiologic perturbations. In this study, we used ment). Thirty micrograms of total protein were loaded in each a combination of DIGE-based proteomics and pathways anal- lane, and the resulting immunoblots were probed with (for day ysis to identify a protein regulatory network whose state is 1) anti–protein disulfide isomerase (PDI), anti-calreticulin, anti– altered in association with the vasopressin escape process and ␤ -tubulin, and (for day 4) anti–heat shock protein 70 (HSP70). the associated downregulation of AQP2 expression. Proteins ␤ The apparent change in mobility of -tubulin may represent an that were identified by DIGE were used as input data for unidentified posttranslational modification. *P Ͻ 0.05. bioinformatic pathways analysis, which pointed to several additional proteins that hypothetically could be involved in the and immunoblotting gave percentage changes that differed escape process. These additional proteins corresponded indi- somewhat for these four proteins, in general, immunoblotting vidually to “hypotheses” that could be tested by immunoblot- confirmed the qualitative responses detected with the DIGE ting asking whether each protein was up- or downregulated as technique, as found previously (12,19). Of the 25 proteins iden- part of the escape process. Indeed, 10 of 10 proteins that were tified by pathways analysis as potentially involved in vasopres- tested in this way displayed the hypothesized regulation (Fig- sin escape (Figure 4), 10 were selected for semiquantitative ures 6 and 7). The outcome of this integrated approach was a immunoblotting on the basis of the central positions in the cluster of proteins, discussed below, that could participate in protein cluster and of antibody availability. One of these is the onset and maintenance of escape or alternatively in second- c-myc, a transcription factor that is regulated in part by phos- ary responses after vasopressin escape. Most of the demon- phorylation. At early stages of escape, total c-myc abundance strated protein abundance changes occurred only in IMCD, not was decreased, whereas at later stages, the abundance of the in renal cortex or brain, indicative of a selective response in the phosphorylated form of c-myc was decreased (Figure 6). The IMCD. abundances of four additional proteins that were identified by pathways analysis were decreased (annexin A4, GRP78/BiP, UT-A3 but Not UT-A1 Is Downregulated during steroid receptor co-activator 1 [SRC-1], and c-fos; Figure 7A), Vasopressin Escape whereas the abundances of five other proteins were increased As a preliminary step in these experiments, we carried out (caspase 3, p53, receptor for activated C kinase 1 [RACK1], immunoblotting to verify the downregulation of AQP2 seen in

Figure 6. Immunoblots of IMCD proteins using antibodies to total and phosphorylated c-myc at early and late stages of vasopressin escape. Each lane is loaded with a sample from a different rat (n ϭ 4 rats/treatment). Thirty micrograms of total protein were loaded in each lane, and the resulting immunoblots were probed with anti–c-myc and anti-phosphorylated c-myc. *P Ͻ 0.05. 2860 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2852–2863, 2005 previous studies (2,9). This study confirmed the decrease in AQP2 expression and showed that the ␣-subunit of ENaC is upregulated in IMCD in the same manner as in more proximal parts of the collecting duct (9). In addition, we extended our studies to collecting duct urea transporter abundances and demonstrated that UT-A3 but not UT-A1 is markedly decreased in abundance during vasopressin escape (Figure 2). The decrease occurred in parallel with the reduction in AQP2. Because both UT-A1 (unchanged) and UT-A3 (downregulated) share the same transcription start site and upstream regulatory re- gions (22), it is unlikely that transcriptional regulation is the basis of the decrease in UT-A3 expression. Instead, recognizing that the 3Ј end of UT-A1 and UT-A3 transcripts differ (22), it seems possible that the differential regulation of these two proteins is based on the different 3Ј ends of the mRNA. Both mRNA stability regulation and translational regulation are based on specialized processes involving the 3Ј end of mRNA molecules (23), raising the hypothesis that either of these mech- anisms may be involved in UT-A3 regulation.

Figure 8. Immunoblots for selected proteins in renal cortex and Figure 7. Immunoblots for selected proteins that were identified brain tissue during vasopressin escape (day 4). Each lane is by pathways analysis. IMCD proteins that were identified by loaded with a sample from a different rat (n ϭ 4 rats/treat- pathways analysis were immunoblotted (30 ␮g of total protein ment), using renal cortex (A) and whole brain (B) homogenates. per lane). Immunoblots show the regulation of proteins at late Thirty micrograms of total protein were loaded in each lane, stages of vasopressin escape (day 4 time point); all proteins and the resulting immunoblots were probed with antibodies to were unchanged at early stages of vasopressin escape (day 1; proteins indicated at left. *P Ͻ 0.05. data not shown). Each lane is loaded with a sample from a different rat (n ϭ 4 rats/treatment). Immunoblots were probed with antibodies to proteins indicated at left. *P Ͻ 0.05. J Am Soc Nephrol 16: 2852–2863, 2005 Proteomics of Vasopressin Escape 2861

The observed downregulation of UT-A3 could contribute to “activated protein 1” (AP-1), for which there is an enhancer site the increase in water excretion seen during the escape process in the 5Ј-flanking region of the human AQP2 gene (31–33). by a mechanism similar to that demonstrated in knockout mice Previously, we demonstrated that c-fos and c-jun are upregu- that lack both UT-A1 and UT-A3 (24). The knockout mice lated in the rat renal inner medulla in response to long-term developed a urinary concentrating defect because urea failed to vasopressin infusion (34). Previous studies have demonstrated accumulate in the inner medulla, resulting in a urea-dependent that AP-1 and a cAMP response element are necessary for osmotic diuresis. Vasopressin escape is associated with modest maximal transcriptional activation of the AQP2 gene in re- extracellular fluid volume (ECF) expansion and hypertension sponse to increased intracellular cAMP (31). Conceivably, the (10). This ECF volume expansion could play a role in the demonstrated decrease in c-fos expression contributes to the decreased expression of collecting duct urea transporters, as fall in AQP2 expression in vasopressin escape. c-Fos is itself suggested in a previous study that implicated aldosterone/salt- regulated in part via a cAMP binding element in its 5Ј-flanking induced extracellular volume expansion in regulation of col- region (33). SRC-1 is a transcriptional co-regulator that interacts lecting duct urea transporter expression (25). In our study, with AP-1 and other transcription factors to mediate transcrip- serum urea levels were markedly decreased as seen previously tional regulation (35,36). (11), a presumed consequence of UT-A3 downregulation. Abundances of Several Other Regulatory Proteins Are Abundances of Several Transcription Factors Are Altered Altered during Vasopressin Escape during Vasopressin Escape Process Aside from transcription factors, the protein regulatory net- Amid the protein regulatory network identified in this work that was identified by pathways analysis included several study are several transcription factors, whose cellular abun- other regulatory proteins that potentially could play a role in dances are presumably too low to be detected via the DIGE vasopressin escape. One of these was c-src, a nonreceptor ty- technique. Several of these were demonstrated to undergo rosine kinase whose abundance was increased three-fold dur- changes in abundance by immunoblotting. The earliest tran- ing late stages of vasopressin escape (Figure 7). c-Src is a critical scription factor to exhibit abundance changes was c-myc, protein in the coupling between G-protein–coupled receptors whose abundance was found to be decreased just before the and mitogen-activated protein kinase pathways (37). c-Src as increase in water excretion (day 1). c-Myc is a basic helix- well as some of its substrates binds to ␤-arrestin 2 (38), a key loop-helix leucine-zipper protein that binds as a heterodimer protein in the internalization of the V2R. Note that c-src was to so-called E-box cis-elements. The presence of three such also increased in renal cortical samples (Figure 8), suggesting E-box elements in the 5Ј-flanking region of the AQP2 gene that c-src upregulation may have been due to a systemic factor. (26) raises the possibility that c-myc abundance changes Similarly, HSP70 was decreased in abundance not only in could contribute to the fall in AQP2 expression during the IMCD but also in brain, perhaps related to the fall in systemic escape response. Although c-myc is best known as a tumor tonicity (39). Another protein whose abundance was upregu- promotor oncogene, it also has physiologic functions in all lated was RACK1. Its function seems to be broader than its cells that seem to be related to the state of differentiation and name suggests, because it constitutes a component of the ribo- proliferation through general regulation of transcription and some and thus may play a role in translational regulation (40). translation (27). Its role in translational regulation was re- RACK1 is also antiapoptotic and a binding partner for c-src vealed in DNA array studies, which showed that changes in (41). Finally, many of the identified regulated proteins also play c-myc levels are associated with parallel changes in the a role in the endoplasmic reticulum (ER) stress response, in- expression of a host of ribosomal proteins (28). The demon- cluding calreticulin, GRP78/BiP, PDI, and caspase 3. ER stress strated fall in c-myc expression therefore may be expected to results from situations in which the protein folding capacity of be associated with a decrease in total cellular transcription the ER is exceeded such as generalized acceleration of transla- and translation, at least at early stages of escape. tion (42). At later stages of vasopressin escape, we found a decrease in In conclusion, combined proteomics and pathways analysis phosphorylated c-myc. The antibody recognizes c-myc phos- served to identify a protein network that is associated with phorylated at threonine-58 and serine-62. These modifications vasopressin escape and contains both high- and low-abundance result in a decrease in half-life of the c-myc protein, thereby proteins. The network included several transcription factors increasing the abundance of total c-myc. The phosphorylation that may be involved in vasopressin escape as well as other is mediated largely by two kinases: c-jun N-terminal kinase (29) relatively low-abundance regulatory proteins. These findings and glycogen synthase kinase-3 (30). The decrease in phosphor- provide a new framework for the study of AQP2 regulation in ylation may play a role in the restoration of total c-myc toward the collecting duct, which is critical to the understanding of control levels on day 4 of the vasopressin escape protocol. SIADH and other forms of hyponatremia. In addition to c-myc, several other transcription factors were identified whose abundances were altered at later stages of Acknowledgments vasopressin escape, viz. c-fos (decreased), c-jun (increased), and This work was supported by the intramural budget of the National p53 (increased). In addition, the abundance of a transcriptional Heart, Lung, and Blood Institute (Z01-HL-01282-KE to M.A.K). E.J.H. co-factor, the SRC-1, was decreased. was supported by the Dutch Kidney Foundation. c-Fos and c-jun together form the transcription factor called We thank Dr. M. Michalak (University of Alberta, Canada) for kindly 2862 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 2852–2863, 2005 providing a calreticulin antibody, Angel Aponte for expert help with collecting duct cells: functional and metabolic assessment. 2-D electrophoresis, Ellis Johns for assistance with immunoblotting, Am J Physiol 253: F251–F262, 1987 David Caden for expert help with blood chemistry, and Dr. R. Zietse for 17. Chou CL, DiGiovanni SR, Luther A, Lolait SJ, Knepper helpful discussions. MA: Oxytocin as an antidiuretic hormone. II. Role of V2 vasopressin receptor. Am J Physiol 269: F78–F85, 1995 18. Kim GH, Masilamani S, Turner R, Mitchell C, Wade JB, Knepper MA: The thiazide-sensitive Na-Cl cotransporter is References an aldosterone-induced protein. Proc Natl Acad Sci U S A 1. Adrogue HJ, Madias NE: Hyponatremia. N Engl J Med 342: 95: 14552–14557, 1998 1581–1589, 2000 19. Hoffert JD, van Balkom BW, Chou CL, Knepper MA: Ap- 2. Ecelbarger CA, Nielsen S, Olson BR, Murase T, Baker EA, plication of difference gel electrophoresis to the identifica- Knepper MA, Verbalis JG: Role of renal aquaporins in tion of inner medullary collecting duct proteins. Am J escape from vasopressin-induced antidiuresis in rat. J Clin Physiol Renal Physiol 286: F170–F179, 2004 Invest 99: 1852–1863, 1997 20. Thongboonkerd V, Barati MT, McLeish KR, Benarafa C, 3. Jaenike JR, Waterhouse C: The renal response to sustained Remold-O’Donnell E, Zheng S, Rovin BH, Pierce WM, administration of vasopressin and water in man. J Clin Epstein PN, Klein JB: Alterations in the renal elastin-elas- Endocrinol Metab 21: 231–242, 1961 tase system in type 1 diabetic nephropathy identified by 4. Schwartz WB, Bennett W, Curelop S, Bartter FC: A syn- proteomic analysis. J Am Soc Nephrol 15: 650–662, 2004 drome of renal sodium loss and hyponatremia probably 21. Thongboonkerd V: Proteomics in nephrology: Current sta- resulting from inappropriate secretion of antidiuretic hor- tus and future directions. Am J Nephrol 24: 360–378, 2004 mone. 1957. J Am Soc Nephrol 12: 2860–2870, 2001 22. Fenton RA, Cottingham CA, Stewart GS, Howorth A, He- 5. Murase T, Ecelbarger CA, Baker EA, Tian Y, Knepper MA, witt JA, Smith CP: Structure and characterization of the Verbalis JG: Kidney aquaporin-2 expression during escape mouse UT-A gene (Slc14a2). Am J Physiol Renal Physiol 282: from antidiuresis is not related to plasma or tissue osmo- F630–F638, 2002 lality. J Am Soc Nephrol 10: 2067–2075, 1999 23. Audic Y, Hartley RS: Post-transcriptional regulation in 6. Verbalis JG: Escape from antidiuresis: A good story. Kidney cancer. Biol Cell 96: 479–498, 2004 Int 60: 1608–1610, 2001 24. Fenton RA, Chou CL, Stewart GS, Smith CP, Knepper MA: 7. Ecelbarger CA, Chou CL, Lee AJ, DiGiovanni SR, Verbalis Urinary concentrating defect in mice with selective dele- JG, Knepper MA: Escape from vasopressin-induced antidi- tion of phloretin-sensitive urea transporters in the renal uresis: Role of vasopressin resistance of the collecting duct. collecting duct. Proc Natl Acad Sci U S A 101: 7469–7474, Am J Physiol 274: F1161–F1166, 1998 2004 8. Tian Y, Sandberg K, Murase T, Baker EA, Speth RC, Ver- 25. Wang XY, Beutler K, Nielsen J, Nielsen S, Knepper MA, balis JG: Vasopressin V2 receptor binding is down-regu- Masilamani S: Decreased abundance of collecting duct lated during renal escape from vasopressin-induced urea transporters UT-A1 and UT-A3 with ECF volume antidiuresis. Endocrinology 141: 307–314, 2000 expansion. Am J Physiol Renal Physiol 282: F577–F584, 2002 9. Ecelbarger CA, Knepper MA, Verbalis JG: Increased abun- 26. Ma T, Yang B, Kuo WL, Verkman AS: cDNA cloning and dance of distal sodium transporters in rat kidney during gene structure of a novel water channel expressed exclu- vasopressin escape. J Am Soc Nephrol 12: 207–217, 2001 sively in human kidney: Evidence for a gene cluster of 10. Song J, Hu X, Khan O, Tian Y, Verbalis JG, Ecelbarger CA: aquaporins at chromosome locus 12q13. Genomics 35: 543– Increased blood pressure, aldosterone activity, and re- 550, 1996 gional differences in renal ENaC protein during vasopres- 27. Schmidt EV: The role of c-myc in regulation of translation sin escape. Am J Physiol Renal Physiol 287: F1076–F1083, initiation. Oncogene 23: 3217–3221, 2004 2004 28. Louro ID, Bailey EC, Li X, South LS, McKie-Bell PR, Yoder 11. Murase T, Tian Y, Fang XY, Verbalis JG: Synergistic effects BK, Huang CC, Johnson MR, Hill AE, Johnson RL, Ruppert of nitric oxide and prostaglandins on renal escape from JM: Comparative gene expression profile analysis of GLI vasopressin-induced antidiuresis. Am J Physiol Regul Integr and c-MYC in an epithelial model of malignant transfor- Comp Physiol 284: R354–R362, 2003 mation. Cancer Res 62: 5867–5873, 2002 12. van Balkom BW, Hoffert JD, Chou CL, Knepper MA: Pro- 29. Noguchi K, Kitanaka C, Yamana H, Kokubu A, Mochizuki teomic analysis of long-term vasopressin action in the in- T, Kuchino Y: Regulation of c-Myc through phosphoryla- ner medullary collecting duct of the Brattleboro rat. Am J tion at Ser-62 and Ser-71 by c-Jun N-terminal kinase. J Biol Physiol Renal Physiol 286: F216–F224, 2004 Chem 274: 32580–32587, 1999 13. Siripurapu V, Meth J, Kobayashi N, Hamaguchi M: DBC2 30. Gregory MA, Qi Y, Hann SR: Phosphorylation by glycogen significantly influences cell-cycle, apoptosis, cytoskeleton synthase kinase-3 controls c-myc proteolysis and sub- and membrane-trafficking pathways. J Mol Biol 346: 83–89, nuclear localization. J Biol Chem 278: 51606–51612, 2003 2005 31. Yasui M, Zelenin SM, Celsi G, Aperia A: Adenylate cycla- 14. Raponi M, Belly RT, Karp JE, Lancet JE, Atkins D, Wang Y: se-coupled vasopressin receptor activates AQP2 promoter Microarray analysis reveals genetic pathways modulated via a dual effect on CRE and AP1 elements. Am J Physiol by tipifarnib in acute myeloid leukemia. BMC Cancer 4: 56, 272: F443–F450, 1997 2004 32. Hozawa S, Holtzman EJ, Ausiello DA: cAMP motifs regu- 15. Knepper MA: Proteomics and the kidney. J Am Soc Nephrol lating transcription in the aquaporin 2 gene. Am J Physiol 13: 1398–1408, 2002 270: C1695–C1702, 1996 16. Stokes JB, Grupp C, Kinne RK: Purification of rat papillary 33. Uchida S, Sasaki S, Fushimi K, Marumo F: Isolation of J Am Soc Nephrol 16: 2852–2863, 2005 Proteomics of Vasopressin Escape 2863

human aquaporin-CD gene. J Biol Chem 269: 23451–23455, 38. Miller WE, Maudsley S, Ahn S, Khan KD, Luttrell LM, 1994 Lefkowitz RJ: Beta-arrestin1 interacts with the catalytic 34. Brooks HL, Ageloff S, Kwon TH, Brandt W, Terris JM, Seth domain of the tyrosine kinase c-SRC. Role of beta-arres- A, Michea L, Nielsen S, Fenton R, Knepper MA: cDNA tin1-dependent targeting of c-SRC in receptor endocytosis. array identification of genes regulated in rat renal medulla J Biol Chem 275: 11312–11319, 2000 in response to vasopressin infusion. Am J Physiol Renal 39. Zhang Z, Ferraris JD, Brooks HL, Brisc I, Burg MB: Expression of Physiol 284: F218–F228, 2003 osmotic stress-related genes in tissues of normal and hypos- 35. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Miz- motic rats. Am J Physiol Renal Physiol 285: F688–F693, 2003 zen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, 40. Sengupta J, Nilsson J, Gursky R, Spahn CM, Nissen P, O’Malley BW: Steroid receptor coactivator-1 is a histone Frank J: Identification of the versatile scaffold protein acetyltransferase. Nature 389: 194–198, 1997 RACK1 on the eukaryotic ribosome by cryo-EM. Nat Struct 36. Lee SK, Kim HJ, Na SY, Kim TS, Choi HS, Im SY, Lee JW: Mol Biol 11: 957–962, 2004 Steroid receptor coactivator-1 coactivates activating protein- 41. Chang BY, Harte RA, Cartwright CA: RACK1: A novel 1-mediated transactivations through interaction with the c- Oncogene Jun and c-Fos subunits. J Biol Chem 273: 16651–16654, 1998 substrate for the Src protein-tyrosine kinase. 21: 37. Luttrell LM: ‘Location, location, location:’ Activation and 7619–7629, 2002 targeting of MAP kinases by G protein-coupled receptors. 42. Schroder M, Kaufman RJ: ER stress and the unfolded pro- J Mol Endocrinol 30: 117–126, 2003 tein response. Mutat Res 569: 29–63, 2005

Supplemental information for this article is available online at http://www.jasn.org.