Embed Size (px)
Citation preview
B R I E F COM M U N I C AT I O N S
Osmometry in osmosensoryneuronsZizhen Zhang & Charles W Bourque
Mechanosensitive channels may act as osmosensorytransducers because osmotically evoked changes in cellvolume alter forces applied to the channel and affect openingprobability1–4. However, osmotically stimulated cells regulatevolume5 normally, whereas systemic osmoregulatory responsesare sustained6,7. Here we report that osmosensory neurons of rat supraoptic nucleus expressed significant membranereserves, and could not regulate volume or total membranesurface area during osmotic perturbations, which may allowmechanosensitive channels to encode the amplitude and timecourse of an osmotic stimulus.
To transduce dynamic aspects of osmotic stimuli, changes in cell vol-ume should be sustained and cause proportional forces onmechanosensitive channels1. However, most animal cells, includingneurons8 and glia5, undergo regulatory volume decrease (RVD) orincrease (RVI) upon swelling or shrinking, respectively9. We there-fore examined whether osmosensory neurons (ONs) and non-osmosensory (hippocampal) neurons (HNs) differ in these respects.ONs were from supraoptic nucleus, an area that contains intrinsi-cally osmosensitive neurons and that releases antidiuretic and natri-uretic hormones from the neurohypophysis in proportion to bloodosmolality10.
Physiological changes in osmolality proceed over minutes orhours7. We therefore examined the effects of slow osmotic ramps.During hypotonic ramps, ONs progressively increased their volume
(mean maximum change was +10.8 ± 1.5%, P < 0.005, n = 6, Fig. 1a),whereas hypertonic ramps caused ONs to shrink by 9.5 ± 1.5% (P < 0.01, n = 5, Fig. 1b). Volume changes evoked in ONs duringosmotic ramps followed osmometric predictions (Fig. 1c). By con-trast, HNs did not show significant changes in volume in response tohypotonic (+0.2 ± 1.1%, P = 0.814, n = 5, Fig. 1a) or hypertonicramps (−1.2 ± 3.2%, P = 0.69, n = 5, Fig. 1b), indicating that RVD andRVI operate in these cells. In HNs, washout of hypotonic rampscaused rebound shrinking (−7.6 ± 1.1%, P < 0.05), whereas reboundswelling was not observed following hypertonic ramps (+0.28 ±3.69%, P = 0.416), suggesting that RVI is slower than RVD. Volumechanges evoked by osmotic ramps in ONs and HNs were similar inisolated cells and in brain slices (Fig. 1d).
Volume regulation during slow and fast perturbations may reflectdistinct mechanisms11. We therefore examined the effects of osmoticsteps. Hypotonic steps evoked gradual swelling in ONs (Fig. 2a,n = 7). Volume peaked after ∼ 4 min (τ = 61 ± 11 s), was maintainedthroughout, and no rebound was observed upon washout. In con-trast, HNs responded to hypotonic steps by initial swelling (< 100 s)followed by RVD (n = 6). Upon washout, HNs showed reboundshrinking (–9.3 ± 2.1%, P < 0.05), indicating relatively slow RVI.Hypertonic steps induced gradual shrinking in ONs (τ = 60 ± 9s,n = 4, Fig. 2b). Minimum volume was reached within 4 min, wasmaintained throughout, and no overshoot was observed uponwashout. Hypertonic steps applied to HNs also caused shrinking thatwas gradual (over ∼ 40 s) and sustained (n = 4). Steady-state volumedecrease, however, was smaller in HNs (–6.4 ± 0.9%) than ONs (–11.5 ± 3%, P < 0.05), suggesting partial RVI in HNs. Steady-statevolume changes evoked by osmotic steps in ONs matched changesobserved during ramps (Fig. 2c). Thus, unlike HNs, ONs showedsteady-state osmometric behavior.
Different cell types express varying amounts of reserve membraneas surface irregularities (e.g., 0.3–1% in fibroblasts12 and up to 70%in lymphocytes13). Additionally, some cells undergoing osmoticallyevoked volume changes regulate their total membrane surface area(TMSA) through mechanisms that include exocytosis and endocyto-sis, and this regulation can blunt osmotically evoked changes inmembrane tension14. We therefore compared membrane reserve andTMSA regulation in ONs and HNs. TMSA, which was determined bymeasuring input capacitance in ONs, exceeded values predictedmorphometrically by 54 ± 8% (P < 0.05, n = 12, Fig. 3a). In HNs, bycontrast, TMSA determined electronically did not differ
Centre for Research in Neuroscience, Montreal General Hospital & McGill University, 1650 Cedar Avenue, Montreal, Quebec H3G 1A4, Canada. Correspondenceshould be addressed to C.W.B. ([email protected]).
Published online 14 September 2003; doi:10.1038/nn1124
NATURE NEUROSCIENCE VOLUME 6 | NUMBER 10 | OCTOBER 2003 1021
Figure 1 Effects of slow osmotic stimulation. (a) Neurons acutely isolatedfrom adult rats (according to a protocol approved by the Animal CareCommittee of McGill University) were exposed to a 60-min hypotonic ramp(–40 mosm/kg [mosm]) and mean (± s.e.m.) normalized volume (volumerelative to control) was determined at each time point. Osmosensoryneurons (ONs) showed reversible swelling, whereas hippocampal neurons(HNs) did not swell but showed rebound shrinking during washout (*P < 0.05 vs. control). (b) In response to a hypertonic ramp, ONs showedreversible shrinking, but HNs did not. (c) Normalized volume in ONs andHNs as a function of osmolality. Solid line is a Boyle-van’t Hoff osmometryfit assuming a non-osmotic compartment of 22.4%. (d) Bars depict mean(± s.e.m.) changes in volume evoked at the end of osmotic ramps in acutelyisolated ONs and HNs, and in cells in acute brain slices (n is indicated oneach bar). See Supplementary Methods online for details.
©20
03 N
atu
re P
ub
lish
ing
Gro
up
h
ttp
://w
ww
.nat
ure
.co
m/n
atu
ren
euro
scie
nce
B R I E F COM M U N I C AT I O N S
1022 VOLUME 6 | NUMBER 10 | OCTOBER 2003 NATURE NEUROSCIENCE
significantly from that estimated morphometrically (P > 0.05, Fig.3a), indicating an absence of membrane reserve. Measurements ofinput capacitance further revealed that when challenged osmotically,HNs regulated TMSA during volume changes (slope = 1.26, r2 = 0.8),whereas TMSA in ONs did not vary during swelling or shrinking(slope = 0.09, r2 = 0.26, Fig. 3b).
Our results show that in contrast to HNs, ONs showed changesin cell volume that were sustained and inversely proportional tostimulus amplitude because of lack of volume regulation. The basisfor the absence of volume regulation remains to be established. Thepresence of a large membrane reserve in ONs, and the absence ofmembrane addition or retrieval during swelling or shrinking,implies that changes in tension could occur within elastic compo-nents of the membrane-associated subcortical cytoskeleton as theplasma membrane expands or contracts to adjust to changes in cellsize. Such features may enable mechanosensitive channels toencode the time course and amplitude of an osmotic stimulus.
Note: Supplementary information is available on the Nature Neuroscience website.
ACKNOWLEDGMENTS
We thank E. Trudel for technical assistance during experiments with brain slices,and R. Sharif and M. Ghamari-Langroudi for helpful comments on themanuscript. This work was funded by Canadian Institutes of Health ResearchOperating and Senior Investigator Awards to C.W.B. and a McGill UniversityHealth Center Research Institute Studentship Award to Z.Z.
COMPETING INTERESTS STATEMENTThe authors declare that they have no competing financial interests.
Received 24 June; accepted 21 July 2003Published online at http://www.nature.com/natureneuroscience/
1. Oliet, S.H. & Bourque, C.W. Nature 364, 341–343 (1993).2. Colbert, H.A., Smith, T.L. & Bargmann, C.I. J. Neurosci. 17, 8259–8269 (1997).3. Suzuki, M. et al. J.Biol.Chem. 274, 6330–6335 (1999).4. Liedtke, W. et al. Cell 103, 525–535 (2000).5. Strange, K. & Morrison, R. Am. J. Physiol. 263, 412-419 (1992).6. Verbalis, J.G., Baldwin, E.F. & Robinson, A.G. Am. J. Physiol. 250, 444–451
(1986).7. Wakerley, J.B., Poulain, D.A. & Brown, D. Brain Res. 148, 425–440 (1978).8. Pasantes-Morales, H., Maar, T.E. & Moran, J. J. Neurosci. Res. 34, 219–224
(1993).9. Lang, F. et al. Physiol. Rev. 78, 247–306 (1998).10. Bourque, C.W., Oliet, S.H.R. & Richard, D. Front. Neuroendocrinol. 15, 231–274
(1994).11. Pasantes-Morales, H. Franco, R., Ordaz, B. & Ochoa, L.D. Arch. Med. Res. 33,
237–244 (2002).12. Raucher, D. & Sheetz, M.P. Biophys. J. 77, 1992–2002 (1999).13. Ross, P.E. et. al. Biophys. J. 66, 169–178 (1994).14. Morris, C.E. & Homann, U. J. Membr. Biol. 179, 79–102 (2001).
1022 VOLUME 6 | NUMBER 10 | OCTOBER 2003 NATURE NEUROSCIENCE
Figure 3 Membrane reserve and surface area in ONs and HNs. (a) Graphscompare mean (± s.e.m.) surface area in ONs (n = 12) and HNs (n = 9)estimated morphometrically (assuming a smooth surface; filled bars) anddetermined electronically via whole-cell recording as Ci/Cm (open bars; *P < 0.001), where Ci is input capacitance and Cm is specific membranecapacitance (1 µF/cm2). (b) Total membrane surface area (TMSA)expressed as normalized capacitance (Ci/control) versus normalized volumeat different times during fast osmotic steps in 12 ONs and 9 HNs. Solidline plots predicted normalized capacitance if changes in volume areaccompanied by proportional changes in TMSA (see SupplementaryMethods online for further details).
Figure 2 Effects of fast osmotic stimulation. (a) ONs and HNs weresubjected to a fast hypotonic step. Although both cell types showed initialswelling, ONs displayed a sustained response whereas HNs displayedRVD(*). Upon washout, ONs returned to control, whereas HNs showedrebound shrinking. (b) In response to a fast hypertonic step, HNs shrankless than ONs. (c) Mean (± s.e.m.) normalized steady-state volume duringfast osmotic steps in ONs and HNs as a function of osmolality. Solid linerepeats Boyle-van’t Hoff osmometry plot in Fig. 1c.
©20
03 N
atu
re P
ub
lish
ing
Gro
up
h
ttp
://w
ww
.nat
ure
.co
m/n
atu
ren
euro
scie
nce
