HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 17/13/1886 02-0870fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by PROCINO, G. Articles by VALENTI, G. Search for Related Content PubMed PubMed Citation Articles by PROCINO, G. Articles by VALENTI, G.

Ser-256 phosphorylation dynamics of Aquaporin 2 during maturation from the ER to the vesicular compartment in renal cells¹

GIUSEPPE PROCINO, MONICA CARMOSINO, ORIANO MARIN^*, ANNA M. BRUNATI^*, ANTONELLA CONTRI^*, LORENZO A. PINNA^*, ROBERTA MANNUCCI, SOREN NIELSEN, TAE-HWAN KWON, MARIA SVELTO^*,¶ and GIOVANNA VALENTI^*,¶,2

Dipartimento di Fisiologia Generale ed Ambientale, University of Bari, 70126 Bari, Italy;
^* Dipartimento di Chimica Biologica, University of Padova, 35121 Padova, Italy;
Dipartimento di Medicina Clinica e Sperimentale, Universitá di Perugia, 06100 Perugia, Italy;
Department of Cell Biology, Institute of Anatomy, University of Aarhus, 8000 Aarhus, Denmark;
Department of Physiology, School of Medicine, Dongguk University, Kyungju 780-714, Korea; and
^¶ Centro di Eccellenza di Genomica in Campo Biomedico ed Agrario, University of Bari

²Correspondence: Dipartimento di Fisiologia Generale ed Ambientale, Via Amendola 165/A, 70126 Bari, Italy. E-mail: g.valenti{at}biologia.uniba.it

SPECIFIC AIMS

Vasopressin-induced phosphorylation of Aquaporin 2 (AQP2) water channel at Ser-256 by protein kinase A (PKA) is considered a key signal for its translocation to the apical plasma membrane of renal principal cells. This process renders the kidney collecting duct water permeable leading to water reabsorption and urine concentration. The aim of this study is to analyze phosphorylation dynamics of AQP2 during its maturation from the endoplasmic reticulum (ER) to the vesicular compartment and analyze whether phosphorylation-defective AQP2 are impaired in their routing in renal cells. Detailed analysis of AQP2 phosphorylation dynamics helps in understanding the molecular basis of an autosomal dominant form of nephrogenic diabetes insipidus (NDI), a severe disease characterized by the kidney inability to concentrate urine in response to vasopressin, caused by a mutation in the AQP2 water channel gene.

PRINCIPAL FINDINGS

1. AQP2 transit in the Golgi area is associated to an increase in phosphorylation
Phosphorylation dynamics of AQP2 were analyzed in AQP2-transfected renal CD8 cells in which newly synthesized AQP2 was accumulated in the ER by incubation with Brefeldin A (BFA). BFA is a fungal metabolite that blocks anterograde vesicular transport from the ER.

In control cells (Fig. 1 A, CTR) AQP2 was localized to small intracellular vesicles. After 16 h incubation with BFA, newly synthesized AQP2 was localized in the ER (Fig. 1A , BFA). After 2 h BFA washout AQP2 was localized in typical large perinuclear structures resembling the Golgi complex (Fig. 1A , 2h wo). AQP2 localization was similar to that found in control cells after 4 h BFA washout (Fig. 1A , 4h wo). In all experimental conditions, the effects of BFA treatment or washout on the organization of the Golgi complex was analyzed by staining with the Golgi marker NBDC6 (Fig. 1A , insets).

View larger version (71K): [in this window] [in a new window]   Figure 1. Immunolocalization and phosphorylation dynamics of newly synthesized AQP2 during maturation from the ER to the vesicular compartment in CD8 cells. *P <0.05; **P <0.01; ***P <0.005; ****P <0.001; *****P <0.0001 (Student’s t test for paired data).

Phosphorylation experiments performed in parallel revealed that both the 29 kDa and the high mannose 32 kDa AQP2 were weakly phosphorylated in the ER (Fig. 1B , BFA). After 2 h of BFA washout, when newly synthesized AQP2 is mainly located in the Golgi area, the amount of phosphorylated AQP2 increased significantly (Fig. 1B , B’, 2h wo), even in the presence of the PKA inhibitor H89 (Fig. 1B , B’, 2h wo+H89). This indicates that AQP2 transition in the Golgi apparatus is associated with a PKA-independent increase in AQP2 phosphorylation. Prolongation of BFA washout to 4 h resulted in AQP2 transport from the Golgi to the vasopressin-regulated vesicular compartment with a concomitant decrease in AQP2 phosphorylation (Fig. 1B , B’, 4h wo).

Western blot analysis showed that the high mannose 32 kDa form was substituted by the mature 35–48 kDa form after a longer BFA washout, (Fig. 1C , 8 h wo), demonstrating that the BFA effect was fully reversible.

2. A serine-kinase distinct from PKA is responsible for AQP2 phosphorylation at Ser-256 during Golgi transit
Immunolocalization experiments using an antibody specifically recognizing Ser-256-phosphorylated AQP2 (p-AQP2) revealed that in control CD8 cells, p-AQP2 was present in intracellular vesicles as described in native renal principal cells. In cells pretreated for 16 h with BFA, a very low staining was visible, indicating that in the ER AQP2 is weakly phosphorylated at Ser-256. After 2 h of BFA washout, when newly synthesized AQP2 is mainly located in large perinuclear structures likely corresponding to the Golgi complex, p-AQP2 staining strongly increased, even if the PKA inhibitor H89 was present during the 2 h BFA washout. This confirms that AQP2 transition from the ER to the post-ER compartment is associated with a PKA-independent increase in AQP2 phosphorylation at Ser-256. Prolongation of BFA washout to 4 h resulted in AQP2 transport from the Golgi to the vesicular compartment with a concomitant decrease in p-AQP2 staining.

We therefore searched for the known consensus sequences of protein kinases active on this serine. The first choice appeared to be a kinase termed Golgi apparatus casein kinase (G-CK) expressed also in mammalian kidney. Accordingly, a synthetic peptide reproducing the AQP2 sequence around Ser-256 was readily in vitro phosphorylated by purified G-CK.

3. E258K-AQP2 mutant, lacking the G-CK consensus site and responsible for a dominant form of Nephrogenic Diabetes Insipidus (NDI), is routed to lysosomes
A dominant form of NDI, caused by AQP2 mutation E258K, generates a protein impaired in its routing to plasma membrane. This mutation affects the crucial glutamic acid sitting in the putative G-CK consensus sequence. RC.SV3 rabbit cortical collecting duct cells (wild-type CD8 cells) were stably transfected with the cDNA encoding for E258K-AQP2 and S256A-AQP2. Both mutations, the former affecting the hallmark residue for G-CK activity and the latter implying the target amino acid for both G-CK and PKA activity, are predicted to impair AQP2 phosphorylation by G-CK. Immunolocalization experiments demonstrated that E258K-AQP2 and S256A-AQP2 did not redistribute to the apical membrane upon forskolin stimulation. Laser scanning confocal microscopy of cells double labeled with anti-AQP2 antibody and a monoclonal antibody specific for the lysosomal resident protein AC17 demonstrated that both mutated forms of AQP2 were mainly expressed in lysosomes.

4. Impaired constitutive phosphorylation of AQP2 at Ser-256 in the Golgi complex may cause AQP2 routing to lysosomes
Analysis of phosphorylation dynamics of E258K-AQP2 and S256A-AQP2 before reaching the lysosomal compartment in cells pretreated with BFA showed that E258K-AQP2 and S256A-AQP2 were localized to the ER as for wt AQP2. After 2 h BFA washout, E258K-AQP2 and S256A-AQP2 staining appeared to be localized in the Golgi area, as observed for wt AQP2. After 4 h BFA washout, the pattern of E258K-AQP2 and S256A-AQP2 localization was indistinguishable from the CTR condition, displaying most of the immunoreactivity in intracellular large vesicles previously identified as lysosomes. Parallel studies of the kinetics of E258K-AQP2 and S256A-AQP2 phosphorylation under the same experimental conditions described above revealed that under control conditions, the phosphorylation level of AQP2 immunoprecipitated from E258K-AQP2 and S256A-AQP2 clonal cells was dramatically reduced compared with wild-type AQP2 immunoprecipitated from CD8 cells. Most important, no increase in phosphorylation for any mutants after the release from the ER following 2 h BFA washout was found even if H89 was present during the washout. Together, these results demonstrate that, in contrast to what was observed for wt AQP2, intracellular trafficking of E258K-AQP2 and S256A-AQP2 from the ER to Golgi is not associated with an increase in phosphorylation. This raises the possibility that the lack of phosphorylation at Ser-256 might cause AQP2 routing to lysosomes.

CONCLUSIONS AND SIGNIFICANCE

This work represents a significant advance in the cell biology research on how protein phosphorylation regulates their constitutive and regulated trafficking within cell compartments. We show that AQP2 transition from the ER into the Golgi apparatus is associated with a PKA-independent increase in AQP2 phosphorylation at Ser-256. A kinase distinct from PKA, which we have tentatively identified as Golgi casein kinase (G-CK), is likely responsible for this constitutive phosphorylation. In fact, the AQP2 amino acidic sequence around Ser-256 also fulfills the consensus sequence of the G-CK found in several mammalian tissues, including rat kidney and rat kidney cell lines. G-CK phosphorylates seryl residues specified by an acidic amino acid at position +2. This consensus sequence occurs around AQP2 Ser-256 (...RRQSVEL...), making it a potential substrate for G-CK; the same site is also a good target for PKA by virtue of the two arginyl residues at positions -2 and -3.

To investigate whether AQP2 phosphorylation in the Golgi is a physiologically relevant phenomenon, the routing and the phosphorylation of intact wild-type AQP2, E258K-AQP2 and S256A-AQP2 transfected renal cells were compared. The mutation E258K is regarded as the hallmark residue for G-CK activity and causes a dominant form of NDI and the S256A involves the target amino acid for both G-CK and PKA activity.

These studies led to two key results: first, when compared with wt AQP2, transit of mutated E258K-AQP2 and S256A-AQP2 through the Golgi was not associated with an increase in phosphorylation. Second, double labeling experiments revealed that most E258K-AQP2 and S256A-AQP2 labeling colocalized with the lysosomal protein marker AC17.

A possible explanation for the first effect is that in E258K-AQP2 phosphorylation is impaired by the lack of Glu-258 and in S256A-AQP2 because of the absence of the Ser-256 itself. The second observation might be a physiological consequence of the first: AQP2 phosphorylation in the Golgi may be required for subsequent AQP2 sorting to the intracellular vesicles and, consequently, phosphorylation-defective AQP2 is routed to lysosomes.

These findings are fully supported by a naturally occurring mutation of dominant inherited NDI caused by a mutation of the Glu-258 for a lysine in the water channel AQP2. This mutation affects the crucial glutamic acid of the G-CK consensus sequence (S-x-E) leading to a functional AQP2 protein that is impaired in its routing to the plasma membrane upon vasopressin stimulation. The prediction is that the lack of Glu-258 will prevent phosphorylation of AQP2 by G-CK, a signal that might play an essential role in sorting AQP2 to the regulated vesicular compartment.

Conversely, one would expect that mutation in the PKA consensus sequence not affecting the G-CK consensus sequence would generate a protein that is constitutively phosphorylated by the G-CK but unable to translocate in response to elevation of intracellular cAMP leading to an NDI phenotype. Indeed, a family with dominant NDI was recently identified in which an AQP2 gene mutation was found encoding R254L-AQP2, a mutation that destroys the PKA consensus sequence but not the putative G-CK consensus sequence. When expressed in MDCK cells, R254L-AQP2 was retained within the cell and was phosphorylated at Ser-256, supporting the hypothesis that Ser-256 in AQP2 is phosphorylated by kinases other than PKA.

In conclusion, the present work demonstrates for the first time that intracellular trafficking of AQP2 from the ER to the regulated vesicular compartment is parallel to multistep phosphorylation dynamics (Fig. 2 ). We propose a double functional role of Ser-256 phosphorylation in the AQP2 water channel trafficking: it regulates both the exit from the Golgi complex and targeting to the apical plasma membrane. The findings presented provide clues to understanding the molecular basis of the dominant form of NDI characterized by the E258K-AQP2 mutation. We propose that the lack of Glu-258 will prevent phosphorylation of AQP2 by G-CK, impairing E258K-AQP2 entry in regulated vesicular compartment. Thus, E258K-AQP2 does not translocate to plasma membrane in response to vasopressin stimulation leading to an NDI phenotype. The physiological importance of AQP2 phosphorylation during transition in the Golgi complex needs to be elucidated. We might speculate that this phosphorylation represents a routing determinant to trigger the binding of adaptor proteins probably required for the exit of AQP2 from this organelle.

View larger version (132K): [in this window] [in a new window]   Figure 2. Schematic diagram of AQP2 maturation in renal cells. A tentative model summarizing our current view. In the Golgi compartment, AQP2 is constitutively phosphorylated by the G-CK. Phosphorylated AQP2 is selected probably by the help of adaptor proteins recognizing phosphoserines sitting in a specific amino acidic motif around Ser-256. AQP2-containing vesicles are probably sorted to the regulated vesicular compartment. This step would be associated with AQP2 dephosphorylation, probably accomplished by phosphatases associated with AQP2-bearing vesicles. Vasopressin stimulation controls the acute AQP2 phosphorylation by activating PKA, a signal required for AQP2 targeting to the apical plasma membrane. If AQP2 is not properly phosphorylated during its transition in the Golgi (as in AQP2-E258K or in AQP2-S256A expressing cells), this causes traffic of defective AQP2 to lysosomes. An essential feature in this model is the prediction that phosphorylation of AQP2 at Ser-256 is a mechanism for coupling exit from the Golgi complex and targeting to the plasma membrane.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0870fje; doi: 10.1096/fj.02-0870fje

This article has been cited by other articles:

				S. Prak, S. Hem, J. Boudet, G. Viennois, N. Sommerer, M. Rossignol, C. Maurel, and V. Santoni Multiple Phosphorylations in the C-terminal Tail of Plant Plasma Membrane Aquaporins: Role in Subcellular Trafficking of AtPIP2;1 in Response to Salt Stress Mol. Cell. Proteomics, June 1, 2008; 7(6): 1019 - 1030. [Abstract] [Full Text] [PDF]

				H. A. J. Lu, T.-X. Sun, T. Matsuzaki, X.-H. Yi, J. Eswara, R. Bouley, M. McKee, and D. Brown Heat Shock Protein 70 Interacts with Aquaporin-2 and Regulates Its Trafficking J. Biol. Chem., September 28, 2007; 282(39): 28721 - 28732. [Abstract] [Full Text] [PDF]

				K.-P. Yip Epac-mediated Ca2+ mobilization and exocytosis in inner medullary collecting duct Am J Physiol Renal Physiol, October 1, 2006; 291(4): F882 - F890. [Abstract] [Full Text] [PDF]

				J. L. Gooch, R. L. Guler, J. L. Barnes, and J. J. Toro Loss of calcineurin A{alpha} results in altered trafficking of AQP2 and in nephrogenic diabetes insipidus J. Cell Sci., June 15, 2006; 119(12): 2468 - 2476. [Abstract] [Full Text] [PDF]

				G. Procino, D. B. Caces, G. Valenti, and J. E. Pessin Adipocytes support cAMP-dependent translocation of aquaporin-2 from intracellular sites distinct from the insulin-responsive GLUT4 storage compartment Am J Physiol Renal Physiol, May 1, 2006; 290(5): F985 - F994. [Abstract] [Full Text] [PDF]

				G. Valenti, G. Procino, G. Tamma, M. Carmosino, and M. Svelto Minireview: Aquaporin 2 Trafficking Endocrinology, December 1, 2005; 146(12): 5063 - 5070. [Abstract] [Full Text] [PDF]

				L. Deak, J. Zheng, A. Orem, G.-G. Du, S. Aguinaga, K. Matsuda, and P. Dallos Effects of cyclic nucleotides on the function of prestin J. Physiol., March 1, 2005; 563(2): 483 - 496. [Abstract] [Full Text] [PDF]

				N. Wajih, T. Borras, W. Xue, S. M. Hutson, and R. Wallin Processing and Transport of Matrix {gamma}-Carboxyglutamic Acid Protein and Bone Morphogenetic Protein-2 in Cultured Human Vascular Smooth Muscle Cells: EVIDENCE FOR AN UPTAKE MECHANISM FOR SERUM FETUIN J. Biol. Chem., October 8, 2004; 279(41): 43052 - 43060. [Abstract] [Full Text] [PDF]

				V. Henn, B. Edemir, E. Stefan, B. Wiesner, D. Lorenz, F. Theilig, R. Schmitt, L. Vossebein, G. Tamma, M. Beyermann, et al. Identification of a Novel A-kinase Anchoring Protein 18 Isoform and Evidence for Its Role in the Vasopressin-induced Aquaporin-2 Shuttle in Renal Principal Cells J. Biol. Chem., June 18, 2004; 279(25): 26654 - 26665. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 17/13/1886 02-0870fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by PROCINO, G. Articles by VALENTI, G. Search for Related Content PubMed PubMed Citation Articles by PROCINO, G. Articles by VALENTI, G.