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

This Article Full Text (PDF) Free All Versions of this Article: 19/12/1674 04-3281fjev1    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 Nicchia, G. P. Articles by Spray, D. C. Search for Related Content PubMed PubMed Citation Articles by Nicchia, G. P. Articles by Spray, D. C.

New possible roles for aquaporin-4 in astrocytes: cell cytoskeleton and functional relationship with connexin43

Grazia P. Nicchia^*,,1, Miduturu Srinivas, Wei Li, Celia F. Brosnan^,, Antonio Frigeri^, and David C. Spray

^* Department of General and Environmental Physiology and Centre of Excellence in Comparative Genomics (CEGBA), University of Bari, Bari, Italy; and Departments of
Neuroscience and
Pathology, Albert Einstein College of Medicine, Bronx, New York, USA

¹ Correspondence: Dipartimento di Fisiologia Generale ed Ambientale, Università degli Studi di Bari, via Amendola 165/A, Bari 70126 , Italy. E-mail: p.nicchia{at}biologia.uniba.it

SPECIFIC AIMS

The morphology of primary cultures of astrocytes is influenced by many factors, including the presence of neurons, reagents that can increase intracellular cAMP levels, and various hormones. Changes in astrocyte morphology may affect neuronal excitability by changing the volume of the extracellular space. In order to maintain a constant extracellular volume, water needs to be cleared from the neuropil to the vascular side of the glial compartment through gap junctions.

We tested whether aquaporin-4 (AQP4), the main CNS water channel, is involved in cell cytoskeleton modifications. To this purpose, we compared the effect of AQP4 knockdown (KD) by RNA interference (RNAi) in mouse, rat, and human astrocyte primary cultures and compared both cell shape and organization of the F-Actin cytoskeleton. We also tested whether a functional relationship may exist between water channels and gap junctions. To this purpose, we analyzed connexin43 protein levels and the strength of intercellular coupling after AQP4 gene silencing.

PRINCIPAL FINDINGS

1. Primary cultures of rat astrocytes represent a better model for morphological studies of human primary astrocytes compared with mouse cells
In a previous study, we reported that AQP4 KD caused dramatic morphological changes in rat astrocytes. In the present study, we found that this effect is species-specific and that morphological changes are different in mouse vs. rat and human astrocytes. Our results show that AQP4 KD in mouse primary astrocytes was even more efficient and more rapid than in rat but did not cause the same dramatic morphological changes. In contrast, AQP4 KD in human primary cultures induced similar morphological changes as found in rat indicating that, at least for AQP4 related studies, rat astrocytes are more similar to human astrocytes.

The effect of AQP4 KD on human astrocytes is not the only aspect analyzed in this study that makes human astrocytes closer to rat than to mouse cells. A comparative analysis of F-actin organization of untreated astrocytes revealed some substantial differences between mouse vs. rat and human cells. In rat and human astrocytes, F-actin is completely organized in stress fibers (Fig. 1 C and A, respectively) whereas the large majority of F-actin fibers in mouse astrocytes are organized in thin cortical bands (Fig. 1E ). The different organization of the F-actin cytoskeleton could be responsible for the differences in cell shape caused by AQP4 KD among the species.

View larger version (130K): [in this window] [in a new window]   Figure 1. Phalloidin-TRITC was used to stain F-actin in human (A, B), rat (C, D) and mouse (E, F) astrocytes in culture. Note that F-actin is organized in stress fibers in human (A) and rat (C) but in a thick cortical layer in mouse astrocytes (E). AQP4 KD induces rearrangement and depolimerization of the F-Actin cytoskeleton in human and rat (B, D, respectively) and a cytoskeleton rearrangement in mouse astrocytes (F).

These findings demonstrate that RNAi technique can be used to KD AQP4 in rat, mouse, and human primary astrocytes and that specie-specific differences are responsible for different morphological phenotypes associated with AQP4 KD in human and rat vs. mouse astrocytes. Moreover, human primary astrocytes are more comparable to rat than to mouse cells even independently from AQP4 KD experiments.

2. AQP4 is involved in astrocyte cytoskeleton changes
In this study we performed a detailed analysis of the F-actin cytoskeleton of rat, human, and mouse astrocytes after AQP4 KD. In rat and human (Fig. 1D and B , respectively), AQP4 KD is associated with a rearrangement and depolymerization of actin where the change of morphology is more dramatic. A more remarkable F-actin cytoskeleton rearrangement is present in AQP4 KD mouse astrocytes (Fig. 1F ) where the presence of the cortical layer is completely replaced by the presence of fibers with a star-like organization. Thus, even though in a completely different way, both the dramatic morphological changes with actin depolymerization of AQP4 KD rat and human astrocytes and the actin cytoskeleton rearrangement of AQP4 KD mouse astrocytes strongly indicate an involvement of this protein in the astrocyte cytoskeleton organization.

Recent studies indicate that AQP4 in the brain seems to associate with the dystrophin associated protein (DAP) complex, in particular with dystrophin (DP71), ß-dystroglycan, and -syntrophin. The absence of dystrophin (mdx mouse) and -syntrophin (ko mice) has been shown to lead to the loss of perivascular AQP4. Thus, it could be conceivable that alterations in cytoskeleton observed in AQP KD astrocytes might be the result of changes to DAP complex. This hypothesis was tested in this study by measuring the expression level of the DAPs after AQP4 KD. However, our results indicate that the dramatic alterations in astrocytes cytoskeleton do not seem to be related to secondary alterations of DAPs and further support a primary role of AQP4 in the observed changes.

3. AQP4 could be functionally linked to Cx43
In this study we also analyzed the expression level of the main gap-junction protein, Cx43, in AQP4 KD astrocytes. Our results showed that AQP4 KD induced strong down-regulation of Cx43 expression and cell coupling only in mouse astrocyte primary cultures (Fig. 2 ). A possible explanation of this functional relationship might be related to a potential modulation of the ions and water flux through Cx43 when AQP4 mediated water exit is reduced/increased, to avoid accumulation of metabolites at the perivascular level.

View larger version (27K): [in this window] [in a new window]   Figure 2. Top left: double immunofluorescence analysis of AQP4 (red) and Cx43 (green) in mouse astrocytes after AQP4 gene silencing (C, D) compared with CTRLs (A, B). Top right: junctional conductance recorded from pairs of mouse and rat cortical astrocytes after AQP4 inhibition by siRNA and under control conditions (parallel treatment with a CTRL siRNA). Bottom left: scrape loading analysis performed on mouse astrocytes treated with CTRL siRNA (A) and AQP4 siRNA (B). Bottom middle: histogram showing the distance of LY spread from the scrape line. Bottom right: graph showing the correlation between the level of AQP4 reduction and the consequent levels of reduction of Cx43 protein and of junctional coupling for each of 3 experiments.

CONCLUSIONS AND SIGNIFICANCE

In a previous study, we reported that AQP4 KD by RNAi caused dramatic morphological changes in primary culture of rat astrocytes. In the present study, we found that this effect is species-specific and that morphological changes are different in mouse vs. rat and human astrocytes. AQP4 KD in human primary cultures induced similar morphological changes as found in rat indicating that, at least for AQP4-related studies, rat astrocytes are more similar to human astrocytes. In rat and human astrocytes, F-actin is mainly organized in stress fibers whereas the large majority of F-actin fibers in mouse astrocytes is organized in thin cortical bands. The presence of a thick cortical layer of actin, conferring a major mechanical support to the membrane, could make mouse astrocytes more resistant to changes in shape, whereas the organization in stress fibers, which are contractile bundles exerting tension, would make human and rat astrocytes more susceptible to morphological changes. Thus, primary cultures of rat astrocytes represent a better model for morphological studies of human primary astrocytes compared with mouse cells.

However, a detailed analysis of the F-actin revealed profound structural cytoskeleton changes in AQP4 KD astrocytes indicating an involvement of AQP4 in the astrocyte cytoskeleton changes.

Dystrophin and the dystrophin-glycoprotein complex are present not only in muscle but also in brain. In muscle, they link the extracellular matrix to the cytoskeleton but their function in brain is not understood. Recent studies indicate that AQP4 seems to associate with dystrophin (DP71), ß-dystroglycan, and -syntropin in both astrocytes as well as in skeletal muscle. The absence of dystrophin (mdx mouse) and -syntrophin (KO mice) has been shown to lead to the loss of perivascular AQP4. Thus, it could be conceivable that alterations in cytoskeleton observed in AQP KD astrocytes might be the result of changes to DAP complex. This hypothesis was tested in this study by measuring the expression level of the DAPs after AQP4 KD. However, our results indicate that the dramatic alterations in astrocytes cytoskeleton do not seem to be related to secondary alterations of DAPs and further support a primary role of AQP4 in the observed changes. Testing is needed to determine whether AQP4 KD directly affect the cytoskeleton organization because of an interaction with the actin microfilaments or whether the described effect is a consequence of volume changes. It has been shown that astrocyte swelling induces changes in the actin cytoskeleton and cell morphology. Astrocytes react by changing their morphology in different physiological and pathological situations. For example, they change their morphology to interact with other tissues or cells, such as capillaries and neurons, and in response to damage to the CNS. Of particular interest would be to clarify whether modulation of AQP4 activity, and thus cell volume, can be useful under pathological situations in which a cytoskeleton and morphological modification of astrocytes occurs.

We analyzed the expression level of the main gap-junction protein Cx43 in AQP4 KD astrocytes. This analysis was suggested by the observation that gap junctions might be considered "intercellular water channels" through which ions and small molecules move together with water. In the brain, molecules/ions that need to be cleared from the active neuropile toward "sinks," such as blood vessels, move with water through the glial Cx43 gap junction channels to reach perivascular astrocytes, where they find separate channels that regulate their selective extrusion. We can propose that AQP4 and Cx43 work in association to allow a unidirectional water flux from the neuropil to the vasculature side since the water flux associated with ions and small molecules, such as K+, moves through Cx43 and exits through AQP4. Our results in mouse astrocyte primary cultures showed that AQP4 KD induced strong down-regulation of Cx43 expression and cell coupling, suggesting these two proteins might be functionally linked. A possible explanation might be related to a potential modulation of the ions and water flux through Cx43 when AQP4-mediated water exit is reduced/increased, to avoid accumulation of metabolites at the perivascular level. Our results suggest that junctional proteins may also be involved in the inter-species variability and that similar changes may occur in AQP4 KO, implying that the phenotype of these transgenic animals may be complicated by the altered expression of this gap junction protein.

Our study demonstrated that 1) RNAi technique can be used to KD AQP4 in rat, mouse, and human primary astrocytes and that species-specific differences are responsible for different morphological phenotypes associated with AQP4 KD in human and rat vs. mouse astrocytes; 2) human primary astrocytes are more comparable to rat than to mouse cells even independently from AQP4 KD experiments; 3) AQP4 could have a new role in changes of cell cytoskeleton in vivo; and 4) AQP4 could be functionally linked to Cx43 in vivo.

View larger version (33K): [in this window] [in a new window]   Figure 3. A) Organization of F-actin in rat and mouse normal primary cultured astrocytes and after AQP4 KD. AQP4 could directly or indirectly interact with the actin microfilaments (AQP4 in blue) or the observed cytoskeleton depolimerization/rearrangement is a consequence of volume changes. B) Hypothetic functional cooperation between gap-junctions (Cx43) and water channels (AQP4) to clear molecules/ions from the active neuropile toward blood vessels in astrocytes.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-3281fje;

This article has been cited by other articles:

				A. Rossi, F. Pisani, G. P. Nicchia, M. Svelto, and A. Frigeri Evidences for a Leaky Scanning Mechanism for the Synthesis of the Shorter M23 Protein Isoform of Aquaporin-4: IMPLICATION IN ORTHOGONAL ARRAY FORMATION AND NEUROMYELITIS OPTICA ANTIBODY INTERACTION J. Biol. Chem., February 12, 2010; 285(7): 4562 - 4569. [Abstract] [Full Text] [PDF]

				T. Vincent, P. Saikali, R. Cayrol, A. D. Roth, A. Bar-Or, A. Prat, and J. P. Antel Functional Consequences of Neuromyelitis Optica-IgG Astrocyte Interactions on Blood-Brain Barrier Permeability and Granulocyte Recruitment J. Immunol., October 15, 2008; 181(8): 5730 - 5737. [Abstract] [Full Text] [PDF]

				M. D. Sorani, Z. Zador, E. Hurowitz, D. Yan, K. M. Giacomini, and G. T. Manley Novel variants in human Aquaporin-4 reduce cellular water permeability Hum. Mol. Genet., August 1, 2008; 17(15): 2379 - 2389. [Abstract] [Full Text] [PDF]

				M. Ruiz-Meana, A. Rodriguez-Sinovas, A. Cabestrero, K. Boengler, G. Heusch, and D. Garcia-Dorado Mitochondrial connexin43 as a new player in the pathophysiology of myocardial ischaemia-reperfusion injury Cardiovasc Res, January 15, 2008; 77(2): 325 - 333. [Abstract] [Full Text] [PDF]

				A. J. Yool Aquaporins: Multiple Roles in the Central Nervous System Neuroscientist, October 1, 2007; 13(5): 470 - 485. [Abstract] [PDF]

This Article Full Text (PDF) Free All Versions of this Article: 19/12/1674 04-3281fjev1    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 Nicchia, G. P. Articles by Spray, D. C. Search for Related Content PubMed PubMed Citation Articles by Nicchia, G. P. Articles by Spray, D. C.