sign in sign up
<<
Home , Cytosis, Exocytosis

Exo- at mossy fiber terminals: Toward capacitance measurements in cells with arbitrary geometry

Christopher Kushmerick and Henrique von Gersdorff* The Vollum Institute, Oregon Health and Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239

xocytosis and endocytosis are real time as a decrease in ments have been made on secretory ubiquitous cellular phenomena capacitance back to baseline resting lev- cells for which the compact isopotential necessary for diverse functions els (6–9). approximation seems, prima facie,tobe such as , internal sig- Most measurements obtained to date justified, including adrenal chromaffin naling,E traffic, and motility. have relied on one of two general tech- cells (10), mast cells (11), and neuroen- Many different techniques have been niques to relate membrane current to docrine cells (12), which secrete via developed to assay and endo- capacitance (6, 9). Time-domain meth- large dense-core vesicles. In addition, , but to date only electrical mea- ods use the amplitude and time course small clear-core synaptic surements of plasma membrane capaci- of membrane current relaxations after and membrane retrieval have been mea- tance have had the time resolution step changes in electrical potential to sured from retinal bipolar terminals necessary to capture both the fusion and determine parameters. (2, 3, 13), hair cells (4, 5), and photore- reuptake of small clear-core vesicle ceptors (14). However, these sensory membrane during fast neurotransmis- neurons contain nonconventional rib- sion. In this issue of PNAS, Hallermann bon-type active zones (3, 13, 15). et al. (1) present capacitance measure- These are the first Recently, attempts have been made to ments from hippocampal mossy fiber measure exocytosis in cells with complex nerve terminals during stimulated exocy- membrane capacitance geometry and multiple electrical com- tosis. These are the first time-resolved partments by using capacitance mea- membrane capacitance measurements measurements from surements. Hsu and Jackson (16) first from bona fide (albeit relatively large) bona fide bouton-type showed that capacitance measurements bouton-type synaptic terminals, with di- are possible in pituitary slices from ameters of Ϸ3 ␮m and a resting capaci- synaptic terminals. nerve varicosities (with attached ) tance of Ϸ1 pF. The results obtained that secrete both via dense- and clear- not only provide important data on ba- core vesicles (17). Mennerick et al. (18) sic synaptic properties of this nerve ter- used the time-domain method to analyze Frequency-domain methods use the minal, in particular on the vesicle pool the electrotonic structure of the intact phase shift between an applied sinusoi- size and the question of multivesicular goldfish retinal bipolar cell, consisting of dal membrane potential and the induced release, but also provide an example of a soma and terminal connected by a current to calculate the complex imped- how to extend capacitance measure- short . They showed that a two- ance of the circuit. In both techniques ments from cells with simple and com- compartment model adequately de- voltage-dependent conductances are pact geometry to more general classes scribed their recordings and, under the avoided (pharmacologically and by care- of neurons and nerve terminals. assumption of high membrane resis- ful choice of membrane potential) such The accessibility of exocytosis and tance, allowed for unique determination that only passive (capacitive and leak) endocytosis to voltage-clamp measure- of the resistance of the axon and the currents remain. ments stems from the fact that these capacitance of the soma and terminal. A Regardless of the choice of technique, processes, by their nature, change the third example, the calyx of Held, is a interpretation of the data depends on surface area of the plasma membrane. giant nerve terminal in the brainstem the use of an a priori electrical model of Because the plasma membrane acts as ascending auditory pathway. This syn- the preparation. The simplest model, an electrical capacitor, changes in sur- apse, unlike the bipolar cell , appropriate for a compact isopotential face area can be detected as changes in contains conventional active zones. De- cell, consists of the membrane capaci- its total capacitance. Thus by electrical pending on the length of the attached tance in parallel with its conductance measurements, net changes in plasma axon, the calyx appears as one or more membrane (i.e., exocytosis minus endo- (6). The rest of the circuit (recording electrotonic compartments (19). Sun cytosis) can be measured. Coupled with patch pipette, cytoplasm, and extracellu- and Wu (20) used frequency-domain voltage protocols that open voltage- lar path) is combined into the series re- capacitance measurements and the com- gated calcium channels, this technique sistance term for a total of three passive pact isopotential cell approximation to can be used to follow membrane parameters (Cm, Gm, and Rs; see Fig. 1). measure exocytosis in calyces that be- 2ϩ changes during Ca -dependent secre- In this case, theoretical time-domain haved as a single electrical compartment tion of . If exocytosis responses consist of single exponential (i.e., calyces that appeared to have a and endocytosis are temporally distinct relaxations from which the three cell single exponential current relaxation in phenomena, the rate of exocytosis can parameters can be uniquely determined. response to a step hyperpolarization). be determined by measuring capacitance Likewise, for a sinusoidal excitation, the Using the same measurement technique, jumps triggered by step depolarizations components of current in phase and 90° of different durations (see refs. 2–5). In out of phase with the voltage stimulus addition, membrane re- can be analyzed by using standard cir- See companion article on page 8975. trieval (or reinternalization via endocy- cuit theory to determine Cm, Gm, and *To whom correspondence should be addressed. E-mail: tosis) after fusion can be monitored in Rs. To date, most capacitance measure- [email protected].

8618–8620 ͉ PNAS ͉ July 22, 2003 ͉ vol. 100 ͉ no. 15 www.pnas.org͞cgi͞doi͞10.1073͞pnas.1633427100 Downloaded by guest on October 2, 2021 COMMENTARY

the mossy fiber terminal, during which Ca2ϩ influx occurred. They found two components of capacitance increase in response to varying length depolariza- tions: a fast component with a time con- stant near 1 ms and a second, slower component with a time constant near 20 ms. They interpreted these components as distinct pools of synaptic vesicles with differential release kinetics. Recovery of membrane capacitance (i.e., putative endocytosis) after a maximally effective stimulation that depleted the releasable pool of vesicles occurred on a time scale of several seconds. Perhaps the most provocative aspect of the article concerns the quantitative estimate of the rate of vesicle release per and implications for multivesicular release (i.e., release of more than one vesicle per active zone per ) (22). The argu- ments stem from the observation that during a prolonged depolarizing voltage step, the calculated release corre- Fig. 1. Model circuits for different cell geometries. (A) Spherical isopotential cells. These cells can be sponded, on average, to more than one described by three passive circuit parameters: membrane capacitance (Cm), membrane conductance (Gm), vesicle per active zone per millisecond. and series resistance (Rs, a composite parameter usually dominated by the resistance of the attached If release were uniform in time during whole-cell mode recording patch pipette but also including the resistance of the cytoplasmic and the depolarization, this would imply that extracellular paths). (B) Two-compartment model proposed by Mennerick et al. (18) to describe goldfish lateral inhibition of release of vesicles retinal bipolar neurons. This four-component model assumes negligibly low membrane conductance and from the same active zone, if present, axon capacitance. R2 is the axon resistance, C1 and C2 are the synaptic terminal and cell soma capacitance, Ͻ and R1 is series resistance of a patch pipette recording from the synaptic terminal. (C) General n- was very short-lived ( 1 ms, compare compartment model described by 3n parameters. with action potential duration half-width of Ϸ0.6 ms). In fact, during the fast component of release, the rate per ac- Taschenberger et al. (21) demonstrated The authors then used their model tive zone may have been much higher. empirically that capacitance jumps fol- cell to test several time- and frequency- Assuming that all the increase in capaci- lowing exocytosis were similar across domain capacitance measurements tech- tance signal is due to exocytosis of syn- calyces, independent of whether the niques to determine which most faith- aptic vesicles that are docked at active structure consisted of one or two elec- fully reported simulated changes in the zones, this measurement implies that trical compartments. Thus it seems that electrical parameters of the bouton. multivesicular release occurs at this syn- capacitance measurements in the calyx Notwithstanding the complex structure, apse. The list of that experi- of Held are not greatly affected by the they found that time- or frequency-do- ence multivesicular release has been attached axon. However, a rigorous main techniques with analysis based on growing recently and now includes both modeling of such capacitance measure- single-compartment models were ade- inhibitory and excitatory synapses in the ments would be useful to show the theo- quate to detect such changes. However, cerebellum (23, 24), cultured hippocam- retical relationship of measured to true because this was only an approximation, pal and cortical synapses (25, 26), and changes in membrane capacitance for the authors were careful to show the bouton-to-spine synapses in hippocam- different-length axons. extent to which the calculations are valid pal slices (27). During early postnatal and pointed out three artifacts that The approach adopted by Hallermann development, the immature calyx of arose during their simulations. First, Held also experiences multivesicular re- et al. (1) to calculate membrane capaci- large simulated conductance decreases lease, although this probably changes to tance from currents recorded from the in the model terminal were associated univesicular release in more mature ca- mossy fiber terminal in slice started with with small phantom increases in calcu- lyxes because release probability de- an anatomical model obtained by recon- lated terminal capacitance. Second, a creases with age (21). struction of a biocytin-filled terminal. correction (Ͻ10%) was needed to relate The extension of capacitance mea- Although they record directly from the the capacitance change reported by the surements to hippocampal synaptic ter- bouton (where the active zones are lo- calculations to the amplitude of the sim- minals is important also because these cated and thus the putative site of exo- ulated capacitance jump. Finally, there nerve terminals undergo presynaptic cytotic capacitance changes), the pres- was a correlated decrease in calculated long-term potentiation (LTP) and ence of multiple large associated series resistance with increasing simu- long-term depression (LTD) mediated processes makes direct application of lated capacitance increases. by cAMP and protein A (PKA) the spherical cell approximation inade- Having characterized the response of (28, 29). LTP and LTD are ubiquitous quate. Indeed, the authors found that a their model to simulated changes in among cortical synapses and are three-compartment model was necessary membrane parameters, Hallermann et thought to be involved in learning and to provide a fit to their data comparable al. (1) went on to characterize tetanus memory. Accordingly, LTP does not to that provided by the full morphologi- toxin-sensitive capacitance jumps that seem to occur in ribbon-type synapses cal model. occur during a step depolarization of and at the calyx of Held synapse, which

Kushmerick and von Gersdorff PNAS ͉ July 22, 2003 ͉ vol. 100 ͉ no. 15 ͉ 8619 Downloaded by guest on October 2, 2021 probably function as highly specialized interesting, for example, to directly ogy. To date, they have been applied and faithful relay stations of presynap- study presynaptic LTP via the dialysis mostly to certain model cells, chosen tic input under most circumstances in of cAMP into mossy fiber terminals for their simple geometry and accessi- adult animals. Thus capacitance mea- to determine whether presynaptic Ca bility to patch-clamp electrophysiologi- surements, which have played an im- currents and͞or vesicle pool size are cal techniques. As the field moves to- ward more general classes of cells and portant role in studies of short-term augmented and whether release ma- synaptic terminals, correct interpreta- at ribbon-type nerve chinery is affected via phosphorylation tion of electrophysiology data will de- terminals and neuroendocrine cells by PKA. pend on a careful combination of mor- (30), now can be extended to long- Capacitance measurements have phometry and modeling. The article by term plasticity studies in more conven- proved to be a powerful tool in the Hallermann et al. (1) shows the way tional nerve terminals. It will be very study of secretion and synaptic physiol- forward.

1. Hallermann, S., Pawlu, C., Jonas, P. & Heckmann, 11. Alvarez de Toledo, G., Fernandez-Chacon, R. & 22. Korn, H., Sur, C., Charpier, S., Legendre, P. M. (2003) Proc. Natl. Acad. Sci. USA 100, 8975– Fernandez, J. M. (1993) Nature 363, 554–558. & Faber, D. S. (1994) in Molecular and Cellular 8980. 12. Thomas, P., Lee, A. K., Wong, J. G. & Almers, W. Mechanisms of Release, 2. von Gersdorff, H. & Matthews, G. (1994) Nature (1994) J. Cell Biol. 124, 667–675. eds. Stjarne, L., Greengard, P., Grillner, S., 367, 735–739. 13. Llobet, A., Cooke, A. & Lagnado, L. (2003) Hokfelt, T. & Ottoson, D. (Raven, New York), 3. Heidelberger, R., Sterling, P. & Matthews, G. J. Neurosci. 23, 2706–2714. pp. 301–322. (2002) J. Neurophysiol. 88, 98–106. 14. Rieke, F. & Schwartz, E. A. (1994) Neuron 13, 23. Auger, C., Kondo, S. & Marty, A. (1998) J. Neu- 4. Parsons, T. D., Lenzi, D., Almers, W. & Roberts, 863–873. rosci. 18, 4532–4547. W. M. (1994) Neuron 13, 875–883. 15. Lenzi, D. & von Gersdorff, H. (2001) BioEssays 24. Wadiche, J. I. & Jahr, C. E. (2001) Neuron 32, 5. Moser, T. & Beutner, D. (2000) Proc. Natl. Acad. 23, 831–840. 301–313. Sci. USA 97, 883–888. 16. Hsu, S. F. & Jackson, M. B. (1996) J. Physiol. 25. Tong, G. & Jahr, C. E. (1994) Neuron 12, 51–59. 6. Lindau, M. & Neher, E. (1988) Pflugers Arch. 411, (London) 494, 539–553. 26. Prange, O. & Murphy, T. H. (1999) J. Neuro- 137–146. 17. Klyachko, V. A. & Jackson, M. B. (2002) Nature physiol. 81, 1810–1817. 7. Matthews, G. (1996) Curr. Opin. Neurobiol. 6, 418, 89–92. 358–364. 18. Mennerick, S., Zenisek, D. & Matthews, G. (1997) 27. Oertner, T. G., Sabatini, B. L., Nimchinsky, E. A. 8. Henkel, A. W. & Almers, W. (1996) Curr. Opin. J. Neurophysiol. 78, 51–62. & Svoboda, K. (2002) Nat. Neurosci. 5, 657–664. Neurobiol. 6, 350–357. 19. Borst, J. G. & Sakmann, B. (1998) J. Physiol. 28. Castillo, P. E., Weisskopf, M. G. & Nicoll, R. A. 9. Gillis, K. D. (1995) in Single-Channel Recording, (London) 506, 143–157. (1994) Neuron 12, 261–269. eds. Sakmann, B. & Neher, E. (Plenum, New 20. Sun, J. Y. & Wu, L. G. (2001) Neuron 30, 171–182. 29. Tzounopoulos, T., Janz, R., Su¨dhof, T. C., Nicoll, York), pp. 155–198. 21. Taschenberger, H., Lea˜o, R. M., Rowland, K. C., R. A. & Malenka, R. C. (1998) Neuron 21, 837– 10. Neher, E. & Marty, A. (1982) Proc. Natl. Acad. Spirou, G. A. & von Gersdorff, H. (2002) Neuron 845. Sci. USA 79, 6712–6716. 36, 1127–1143. 30. Neher, E. (1998) Neuron 20, 389–399.

8620 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.1633427100 Kushmerick and von Gersdorff Downloaded by guest on October 2, 2021