QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online June 23, 2005
doi: 10.1242/10.1242/jcs.02425

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JCS 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 Destaing, O. Articles by Jurdic, P. Search for Related Content PubMed PubMed Citation Articles by Destaing, O. Articles by Jurdic, P. Social Bookmarking                         What's this?

A novel Rho-mDia2-HDAC6 pathway controls podosome patterning through microtubule acetylation in osteoclasts

Olivier Destaing^1,*,, Frédéric Saltel^1,*,, Benoit Gilquin², Anne Chabadel¹, Saadi Khochbin², Stéphane Ory^3,* and Pierre Jurdic^1,*,¶

¹ Laboratoire de Biologie Moléculaire et Cellulaire, UMR 5665 CNRS/ENS, INRA 913, Ecole Normale Supérieure de Lyon, 46, allée d'Italie, 69364 Lyon CEDEX 7, France
² Laboratoire de Biologie Moléculaire et Cellulaire de la Différenciation, INSERM U309, Institut Albert Bonniot, Faculté de Médecine, 38706 La Tronche Cedex, France
³ CRBM, CNRS FRE2593, 1919 route de Mende, 34293 Montpellier CEDEX 5, France

View larger version (89K): [in a new window]   Fig. 1. TAT-C3-mediated Rho inhibition in osteoclasts confers podosome belt resistance to microtubule depolymerisation through stabilisation of a subset of microtubules. Rho-inhibition partially blocks nocodazole-induced microtubule depolymerisation and podosome belt dissociation. Osteoclasts either untreated or treated for 5 hours in the presence of TAT-GFP (0.5 µM) or TAT-C3 (0.5 µM) were incubated in the presence of nocodazole (2 µM) for 50 minutes and then fixed and stained for actin (in red) and ß-tubulin (in green) before observation using a confocal microscope. (A) Kinetics of nocodazole-mediated podosome belts and microtubule disruption. In the presence of the control (TAT-GFP), nocodazole disrupted both microtubules and podosome belts in less than 30 minutes. In the presence of TAT-C3, podosomes belts were resistant to nocodazole treatment for more than 1 hour whereas subsets of microtubules were still observed. (B) TAT-C3 pretreatment had no effect on osteoclast cytoskeletons exhibiting a dense microtubule network and a podosome belt, whereas nocodazole induced complete microtubule dissociation together with podosome belt destabilisation and the subsequent formation of podosome rings (arrowheads) and clusters (open arrowheads) as previously described (Destaing et al., 2003). In contrast, TAT-C3 blocked the action of nocodazole since podosome belts were stabilised at the osteoclast periphery (arrows) and a subset of microtubules was maintained. A close-up of the area within the white insert is presented underneath each image. (C) Rho activation induces microtubule stabilisation in nocodazole-treated NIH3T3 cells. NIH3T3 cells were serum starved for 12 hours in the presence of TAT-GFP or TAT-C3 (0.5 µM) for the last 4 hours. Then cells were stimulated by serum addition for a further 2 hours before a 50-minute nocodazole (2 µM) treatment. Nocodazole-resistant microtubules were barely detectable when Rho was inactivated either in the absence of serum or in the presence of TAT-C3. In contrast, Rho activation by serum induced microtubule stabilisation in TAT-GFP control cells. A close-up of the area within the white insert is presented underneath each image. Bar, 10 µm (A, lower panels in B, C); 20 µm (upper panels in B).

View larger version (63K): [in a new window]   Fig. 2. Rho activity controls the level of tubulin acetylation upstream of HDAC6. (A) Inhibition of Rho by 0.5 µM TAT-C3 for 5 hours induced an accumulation of acetylated microtubules in comparison to 0.5 µM TAT-GFP used as a control. (B) One nucleus per osteoclast was microinjected with either RhoA WT-GFP or a constitutively activated form of Rho, RhoAV14-GFP expression vectors. Cells were fixed 6 hours after microinjection and GFP-expressing cells were detected by GFP fluorescence using a confocal microscope. Acetylated tubulin was detected by indirect immunofluorescence (green) and F-actin by means of phalloidin-RITC (red) and a close-up of each condition is presented. In the presence of RhoA-WT, osteoclasts exhibit the typical podosome belt and dense networks of acetylated microtubules. On the other hand, expression of Rho V14-GFP induced deacetylation of microtubules and disorganisation of podosome belts (arrowhead in close-up area). However, tubulin deacetylation dependent on Rho activation was inhibited after treatment with the HDAC6 inhibitor TSA (3 µM) for 1 hour, showing that this enzyme is downstream of Rho. (C,D) HDAC6 is present and active in osteoclasts. Endogenous HDAC6 was easily detected in osteoclasts by western blotting with a polyclonal anti-HDAC6 antibody (C). The deacetylase activity of HDAC6 was tested with two drugs: TSA, known to inhibit its activity and sodium butyrate, which does not. HDAC6 was indeed active in osteoclasts as confirmed by greatly increased levels of acetylated tubulin in TSA-treated osteoclasts and unchanged levels in the presence of sodium butyrate compared to the control and to the total amount of ß-tubulin (D). Finally, inhibition of Rho by TAT-C3 (for 4 hours) in the presence of TSA had no additional effect on the increase in acetylated tubulin (C). Bar, 20 µm.

View larger version (54K): [in a new window]   Fig. 3. mDIA2 mediates the action of Rho at the level of acetylated microtubules by interacting with and activating HDAC6. (A) Constitutively active mDia2 blocks tubulin acetylation and podosome belt formation. One nucleus per osteoclast was microinjected with either GFP or a constitutively activated form of mDIA2, GFP-mDIA2 GBD expression vectors. 6 hours later, acetylated tubulin was detected by indirect immunofluorescence (green) and F-actin with phalloidin-RITC (red). In GFP-mDIA2 GBD-expressing osteoclasts compared to GFP-expressing osteoclasts, the Ac-MT level was dramatically decreased and the podosome belt disrupted. (B) HDAC6 and mDia2 interact together. COS cells were either transfected with GFP alone, HA-HDAC6, GFP-mDIA2 WT, GFP-mDIA2 GBD vectors or co-transfected together with HDAC6. Transfected HDAC6 or mDia2 were revealed in total cell lysate (TCL) by an anti-HA or an anti-GFP antibody respectively (left panel). HDAC6 was immunoprecipitated with an anti-HA antibody and mDia2-associated proteins were revealed with an anti-GFP antibody (right panel). Both the wild type and activated mDia2 co-precipitated with HA-tagged HDAC6. (C) Mapping of HDAC6 domain interaction with mDia2. To determine the domains of HDAC6 implicated in the interaction with GFP-mDIA2 WT, COS cells were transfected with GFP-mDIA2 WT and with the N-terminal domain, the deacetylase domain 1 (DD1), the deacetylase domain 2 (DD2) or the C-terminal domain of HDAC6, using HA-tagged expression vectors. Cell lysates of transfected cells were immunoprecipitated with a monoclonal anti-HA antibody. (D) mDia2 increases HDAC6 deacetylase activity. In a tubulin deacetylase in vitro assay, COS cells were transfected with mHDAC6-HA WT, lysed at room temperature for 15 minutes and the ratio of acetylated tubulin/tubulin monitored by western blotting. When GFP-mDIA2 WT was transfected with mHDAC6-HA WT, this increased the deacetylase activity of the enzyme whereas GFP-mDIA2 WT alone had no effect on the level of this PTM. Positions of protein standards in kDa are indicated on the left-hand side of blots.

View larger version (60K): [in a new window]   Fig. 4. The level of microtubule acetylation in mature osteoclasts correlates with stabilisation of podosomes into belts. (A,B) Macrophages and immature osteoclasts, differentiated for 6 days in the presence of RANK-L + M-CSF, with podosome rings, presented a low level of acetylated tubulin. (C) Mature osteoclasts differentiated for 8 days in the presence of RANK-L + M-CSF presented specific accumulations of acetylated microtubules just behind the podosome belt (arrowhead). (D) This specific accumulation of acetylated microtubules during the transition period between podosome rings and belt, the last step of podosome patterning, was confirmed by western blot analysis on osteoclast populations differentiated for 6 (D6) or 8 (D8) days. Standard protein markers are indicated in kDa. (E,F) Rho inhibition accelerates the podosome ring to belt transition in immature osteoclasts. Spleen leukocytes were differentiated in the presence of RANKL and M-CSF for 6 days, and then treated or not with TAT-GFP or TAT-C3 for 6 hours before being fixed, and stained for F-actin with Phalloidin-RITC. A general overview of podosome organisation in osteoclasts is presented (E). At D6, osteoclasts in untreated or TAT-GFP treated cultures exhibited mostly podosome clusters or rings (arrowheads) or rare podosome belt (arrows). In contrast, in TAT-C3 treated osteoclasts, podosome belts were seen. This was quantified in osteoclasts containing more than three nuclei by counting 600 osteoclasts per condition (F). At this stage of differentiation, untreated osteoclasts or those treated for 6 hours with TAT-GFP (0.5 µM) exhibited a majority of podosomes arranged in rings or clusters, whereas TAT-C3 (0.5 µM) induced the formation of belts at the cell periphery as usually found in more mature osteoclasts. (G) Osteoclasts at 3, 5 and 7 days of differentiation were lysed to perform an affinity precipitation assay with GST-RBD fusion proteins. The amount of Rho in total cell lysates (Total Rho) and in the precipitated fraction (RhoGTP) was determined by western blotting using the 26C4 monoclonal anti-Rho antibody. The ratio of RhoGTP/total Rho was estimated by densitometry analysis. Bar 20 µm.

View larger version (75K): [in a new window]   Fig. 5. Formation of the sealing zone in bone resorbing osteoclasts is associated with the dynamic regulations of tubulin acetylation. (A,B) Osteoclasts were differentiated in the presence of RANK-L + M-CSF for 8 days, detached, spread on mineralised matrix (ACC, Apatite Collagen Complex) substrate, which mimics dentin slices, fixed and immunostained with phalloidin-RITC and monoclonal anti-acetylated and polyclonal anti-detyrosinated tubulin antibodies. These two osteoclasts are associated with a resorption pit (*). The osteoclast presented in A is still resorbing, as it exhibits a sealing zone, a large band of F-actin, and has a large number of acetylated but undetectable detyrosinated microtubules. The osteoclast in B is a migrating osteoclast without a sealing zone and showed neither acetylated nor detyrosinated microtubules. Bar, 20 µm.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter