key: cord-0010403-tljeonwm
authors: Caswell, Anthony H.; Brandt, Neil R.
title: Does muscle activation occur by direct mechanical coupling of transverse tubules to sarcoplasmic reticulum?
date: 2003-03-21
journal: Trends Biochem Sci
DOI: 10.1016/0968-0004(89)90265-x
sha: 1c3643c810ba67db37b2b8bc0fe3f06df4d381dd
doc_id: 10403
cord_uid: tljeonwm

Our knowledge of the physiological and biochemical constituents of skeletal muscle excitation has increased greatly during the last few years but this has not led to a consensus of the physiological mode of muscle activation. Three hypotheses of transmission, involving either transmitter-receptor interaction or direct mechanical coupling, are still under active consideration. The hypothesis of direct mechanical coupling currently being evaluated proposes that the dihydropyridine receptor in the transverse tubules serves as a voltage sensor that communicates directly with the junctional foot protein/Ca(2+) channel of sarcoplasmic reticulum to initiate opening of the channel.

In other cellular systems, this receptor has been associated with Ca 2 ~ Anthony 1-t!. Caswell and Nell R. Brandt channel activity. Experimental data in skeletal muscle support the view that drugs which bind to and inhibit this receptor also inhibit the slow Ca 2+

Our knowledge ofthephysiologicaland biochemical constituents ofskeletalmuscle channel. Estimates of dihydropyridine receptor content in T-tubules vary excitation has increased grea,qy during the last few years but this has not led to a from 20 to 200 pmol per mg protein 2. consensus of the physiological mode of muscle activation. Three hypotheses of Assuming each receptor is a channel, transmission, involving either transmitter-receptor interaction or direct mechanical coupling, are still under active consideration. The hypothesis of direct mechanical this content would probably not be coupling currently being evatuated proposes that the dihydropyridine receptor in quite adequate to supply the Ca ~-+ the transverse tubules serves as a voltage sensor that communicates directlv with the necessary for contraction. This number " of channels, however, could supply a

]uncttonalfootprotem/Ca-channelofsarcoplasmic reticulum to initiate opening source of trigger Ca 2+ which, in its of the channel, turn, stimulates opening of the Ca 2+ channel of SR. In a similar context, During the last 20 years three basic occurs between the T-tubule and the experiments withskinnedmuscle fibers (in which the plasma membrane has hypotheses of the mechanism of terminal cisternae of the SR in which been mechanically removed or chemiexcitation-contraction cou?ling in depolarization of the T-tubule causes a cally renderedleaky), aswell as experiskeletal muscle have been proposed, conformational alteration of the span-These have found favor and eisfavor in ning structure, thus opening the Ca 2+ merits in isolated SR demonstrate that a cyclical fashion much as the waves of channel of SR.

Ca ~+ release from SR may be activated in the presence of micromolar concertthe sea. Our understanding of the Atpresent, the first ofthese hypotheses trations of Ca 2+ and inhibited in very physiological and biochemical pro-

cesses of muscle activation has has little experimental support. It is low Ca concentrations " . The rate of increased vastly during this period but, hard to define within the context of the Ca ;+ release from the SR initiated by we cannot as yet delineate unequivo-known ionic gradients across the SR this process although under the nonmembrane and the transverse tubular physiological conditions of low Mg 2+, cally between the theories.

The three basic hypotheses of muscle membrane 1. The other two hypotheses is adequate to account for the rapidity excitation expressed in general terms have been involved in a long running ofmusclecontraction 5.

dogfight for supremacy. This review Why, then, is this hypothesis not questions whether direct mechanical generally accepted as the mode of (1) Direct or capacitive electrical corn-communication can account for muscle muscle activation? The initial evidence munication between the t::ansverse activation. The answer should be against this hypothesis was presented tubule (T-tubule) and the terminal cis-viewed in the context of alternative by Armstrong et al. 6 who showed that ternae causes the passage of a wave of modes of excitation-contraction muscle could continue to contract when potential change that is responsible for coupling, immersed in a solution containing the the opening of the Ca 2+ channel in the The transmitter-receptor hypothesis Ca 2+ chelator EGTA. Subsequent sarcoplasmicreticulum (SR). may be expressed more specifically experiments have supported this obser-(2) Depolarization of the T-tubule in terms of two potential transmitters vation but more sophisticated experi-'9+

causes the release of a transmitter into for muscle activation: (I) Ca -; and mental approaches have demonstrated the junctional space which binds to a (2) inositol 1,4,5-trisphosphate (IP3). a reduction in Ca 2+ release associated receptor and causes Ca 2+ channel These two hypotheses reduced to their with EGTA treatment 7. In addition, opening in the SR. minimal constituents are illustrated in Baylor and Hollingworth s have shown (3) Direct mechanical communication Fig. 1 

muscleisactivatedbydepo-idityofregenerationoftheprecursor larization. It begins and is PIP2. In addition, phospholipase C T.C.

terminated within a few activity has been described as Ca 2+ milliseconds of depolari-dependent in the physiological range ~s. zation I~. The evidence in Therefore, its action may be the result favor of IP3-induced Ca 2+ of stimulation rather than the cause. It release is that injection of is doubtful that the concentration of this compound into skinned PIP2 provides an adequate store to gen-(b) 1,, muscle fibers induces con-crate IP3 at the concentration needed

Phospholipase C traction of the muscle ~2"~3. in the cytoplasm, unless the diffusion of

However, these obser-IP3 is restricted to the triadic region. evidence of the existence of mechanical communication arose hormone receptors which through the observation that a nonexists that the whole apparatus for might use the cascade as occurs in other linear capacitive current was observ-Ca2+-induced Ca 2+ release appears to cellular systems. Phosphatidylinositol able when the membrane potential of exist but, nevertheless, does not appear (PI) kinase and phosphatidylinositol the muscle was altered. The size of this to operate. 4-phosphate (PIP) kinase are localized charge movement was half-maximal at in the T-tubule, although PI kinase a level of depolarization which gave Inositol trisphosphate induced Ca 2+ appears also to be present in sarcoplas-half-maximal activation of muscle conrelease mic reticulum ~v'~8. A phospholipase C traction. The charge movement was The more recently postulated alter-acting on PIP2 is associated with the originally modeled as a plunger which native, IP3, is a superficially attractive T-tubule.

The quantitative re-contained a charge constituent joining candidate for initiating Ca 2+ release quirements for IP3 to be a transmitter the T-tubule to the terminal cisternae. since this is a known activator of Ca 2+ are particularly stringent in view of the The movement of the charge constitrelease from SR or the equivalent rapidity of Ca 2+ release in normalcon-uent caused the opening of the Ca 2+ organelle in non-muscle tissue 9~1°. It traction, and the necessity for regener-channel of the terminal cisternae by should, however, be recognized that ation of lP3 to permit repetitive firing of unblocking the pore 2~. In many ways this crude but simple mode] has with-not translated into direct demonstration, movement is associated with concomistood the test of time. Most particu-tant activation or inhibition of Ca 2+ larly, charge movement is very closely The voltage sensor release from SR and contraction; connected to muscle activation not only A further insight into the mechanism (3) activation or inhibition of Ca 2+ as a function of membrane potential of muscle contraction has arisen movement into the cell is not associated but also in the response to pharmaco-through the development of our under-with concomitant activation or inhilogical manipulation. The limitation of standing of the role of the dihydro-bition of charge movement and the model was that it did not lead to the pyridine receptor, the putative Ca 2+ contraction. defining of a specific moiety through channel of T-tubules. The existence of which the message of transmission this channel was originally invoked to Most of the data currently available occurred. The details of this model, as support the concept of Ca2+-induced support these conclusions although well as the alternative protocols, has Ca 2+ release, but it soon became with some reservation. Two classes of been fleshed out through a growing apparent that the mode of action of this agents which are known to bind to and knowledge of the molecular constit-proteinis more complicated. The early inhibit the dihydropyridine receptor's uentsofthetriadjunction, experiments on drugs which inhibited action as a Ca 2+ channel are also this channel failed to elicit a significant known to inhibit charge movement and The CaZ+-releasesite inhibition of muscle contraction. In contraction. The conditions in which Our ability to obtain molecular addition, the Ca 2+ current associated D600 blocks contraction require that insight into the triad junction has arisen with the action potential was switched the muscle first be depolarized and through fractionationproceduresatthe on by muscle depolarization signifi-cooled and the inhibition can be level of the vesicular components for cantly after the contractile event had reversed by hyperpolarization. A posthe junction and, subsequently, at the begun. Furthermore a long-standing sible explanation for these relevel of the protein constituents of discrepancy existed between the bio-quirementsis that D600 only binds the these vesicles. Sarcoplasmic reticulum chemical data in which dihydropyridine receptor when it is in an inactive state has been physically separated into drugs, such as nitrendipine, gave after depolarization. The blockage of longitudinal reticulum and terminal cis-a Kd for the receptor in the nanomolar charge movement and of contraction ternae. The terminal cisternae fraction range while inhibition of the slow Ca 2+ are both complete. In the presence of has been shown to contain T-tubules current requires much higher con-dihydropyridine drugs blockage of associated in the form ofa triadicjunc-centration, charge movement and Ca 2+ release tion, and the triadic junction has itself

Hui et al. 27 demonstrated that the from the SR appears to be partial and been separated into its component Ca2+-blocking drug, D600, wasindeed critically dependent on the condition of organelles. The T-tubules have been able toinhibit contraction although this the fiber. When the holding potentialof shown to be an extremely rich sourceof inhibition occurred only after the the muscle fiber is hyperpolarized dihydropyridine receptor, the putative muscle had been subjected to K + (-100 mV) the blockage by the dihy-Ca 2+ channel. The identification of the contracture and cooled. Rios and dropyridine is quite limited, but when junctional constituent has been Brum as subsequently reported that the holding potential is at -70 mV a achieved through two complementary inhibition of contraction could be eli-blockage by about 70% is observed. approaches. Cadwell and Caswel122 cited with nitrendipine. These authors RiosandBrum 28 have shown that this is identified a high molecular weight pro-added an interesting corollary, postu-associated with a concomitant inhibitein (confined to the terminal cister-lating that the dihydropyridine recep-tion of the rate and extent of Ca 2+ nae) that exhibited expected properties tor might be acting not in the capacity release from the SR. The concentraof the junctional feet. This protein was of a Ca 2+ channel but in the role of a tions required to produce even a partial subsequently extracted and isolated 23. voltage sensor. The earlier electrophy-blockade are considerably in excess of At thesametime, several investigators siological experiments demonstrated the Kd estimated from isolated Thave followed the pharmacological that the opening of the Ca 2+ channel tubules. Lamb >~ has found that increasapproach of identifying a drug which occurred on depolarization of the fiber, ing dihydropyridine to extremely high binds and reacts specifically with the Rios and Bruin 2s have argued that the concentration does not create more Ca 2+ release channel of sarcoplasmic dihydropyridine receptor can have a than partial blockage. In neonatal reticulum. One drug (ryanodine) has dual role. In skeletal muscle the main muscle there is evidence that the halfnow been radiolabeled as a means to role of this channel is to sense the elec-maximal blocking concentration effecthe isolation of the channel. The recent trical potential of the T-tubule and to ted by the dihydropyridine, PN200isolation of the channel has shown that transmit, via a conformational change, 110, is considerably decreased by deit is identical to the protein previously the information to the SR without polarization of the fiber. If blockade identified as the junctional feet 24-2°. necessarily initiating Ca 2+ influx, by PN200-110 of charge movement is This now allows us to see the model of The physiological evidence germaine similarly voltage dependent this may direct mechanical coupling in a more to this hypothesis rests largely on a explain the discrepancy with the biospecific context: The feet processes requirement for the following correla-chemical Kd. The physiological halfthemselves contain the Ca 2+ channel tions: maximal concentration for the dihydroand this channel is in immediate juxta-pyridine effect on charge movement position to the T-tubule. T;herefore, (1) activation or inhibition of the dihy-may be high because the muscle is the logic of direct mechanical coupling dropyridine receptor is associated with polarized, while in biochemical experibecomes more compeling, but the concomitant activation or inhibition of ments the drug binds with higher afficaveat remains that the understanding charge movement (Q[3) and the two are nity to its receptor because it is in its of muscle contraction has been be-temporally linked;

inactive (depolarized) state. What is deviled by compeling logic which has (2) activation or inhibition of charge not yet clear is whether that portion of the charge movement that is refractory have observed that Ca 2+ release from In each case we have clear evidence for to dihydropyridines represents a scpar-the SR may take place in a stimulated an association between the glycolytic ate component of charge movement or muscle even when the cytoplasmic enzymes, glyceraldehyde phosphate whether it simply represents the fact environment is bathed with a rapidly dehydrogenase and aldolase, with the that under the experimentalconditions binding Ca2+-chelating agent, further junctional foot protein. We have not employed, only a portion of the dihy-suggesting that any Ca 2+ influx which been able to observe any direct associdropyridine receptor has been placed may occur is not directly responsible ation between the junctional foot proin an inactive state capable of binding for Ca 2+ release, tein and the dihydropyridine receptor the drug. In addition, Lamb > has or, for that matter, any intrinsic found, in rabbit muscle fibers, that Abioehemicalmodei? T-tubular protein. Chadwick et a/. 32, although charge movement is consis-If this hypothesis of direct mechani-using a hetcro bifunctional agent, have tently blocked (to approximately 50%) cal coupling expressed in its present described an association between the by nifedipine, the effect on contraction form is to fulfill the role of excitation-junctional foot protein and a protein is quite variable, some fibers being contraction coupling then the physio-of molecular weight 72 000 in the completely refractory to the effects of logical data must be supported by bio-T-tubule. This is not a subunit of the the drug while others are completely chemical evidence that direct mechan-dihydropyridine receptor and is, in any paralysed. This, in its turn, raises the ical communication does occur. In event, an extrinsic protein. Using gel issue as to whether the charge move-many ways the hypothesis is rendered overlay procedures, we have observed ment invariably reflects the activation more attractive by the observation that an association between glyceraldehyde ofthemuscle.

Ca 2+ release is elicited by the junc-phosphate dehydrogenase and the {t~-Evidence of a temporallink between tional foot protein.

For this now subunit of the dihydropyridine rccepactivation or inactivation of the dihy-requires of the mechanical coupling tor; this, however, is not specific in the dropyridine receptor and muscle acti-hypothesis only that there be communi-sense that the glycolytic enzyme binds vation is, as yet, unavailable. The cation between this protein and the to a number of other T-tubular constitdihydropyridine-sensitive Ca 2+ chan-voltage sensor in the T-tubules as illus-uents. Thus, there remains the possinel is switched on with a slow time trated in Fig. lc . This could be fulfilled bility that the junctional foot protein course. It could be argued, however, if the junctional foot protein is physi-binds indirectly through the glycolytic that this represents a multistage pro-cally attached to the dihydropyridine enzymes to the DHP receptor but it is cess in which a rapid conformational receptor. One would further expect doubtful that this could serve for direct response to a voltage change may pre-that the attachment should be specific mechanical coupling. cede the slow gating of the channel, in the sense that all junctional foot It needs to be emphasized that the When the receptor is operating as a particles would have an associated hypothesis that the dihydropyridine voltage sensor for muscle activation, dihydropyridine receptor or receptor receptor is the voltage sensor for only the conformationalchange may be cluster, muscle excitation does not, in itself, required. It is also unclear whether Block et al. 3~ have observed tetrads require that the receptor binds directly blockage of the receptor by drugs gives of large intercalated particles in freeze-to the junctional feet. Communication a time-correlated blockade of acti-fracture replicas of skeletal muscle could take place through indirect vation. It is possible that the inhibition T-tubules. The large size of the inter-means: we are, therefore, currently in of muscle activity may be consequent calated particle is consistent with the the ambiguous situation that our knownot to direct inhibition of the receptor view that this particle may be the dihy-ledge on the processes of muscle excitabut to the modified Ca 2+ homeostasis dropyridine receptor whose total tool-tion has increased vastly but we still are following channel inhibition. T-tubules ecular weight must be in the range of limited by our inability to produce a contain an active Ca2+-extrusion pump 400 000. An unexpected feature of the theory that is fully consistent with the which may deplete cytoplasmic stores observation of these authors is that existing data, although much of our directly and SR stores indirectly if there is one tetrad per two feet indic-current knowledge isin accord with the Ca 2+ entry is blocked, ating that only half the feet are asso-hypothesis of direct mechanical coup-The third requirement for the hy-ciated with these particles, ling. The molecular mechanism of pothesis that the dihydropyridine The biochemical evidence tosupport direct mechanical coupling has not receptor directly mediates the message direct connection between the dihydro-been explained in skeletal muscle but of excitation to the SR is that there pyridine receptor and the junctional could be analogous to the sliding of should be a dissociation between Ca 2+ foot protein is weak. We have era-subunits which has been described in influx and activation. There is no doubt ployed three techniques to determine gap junctions. that under normal conditions a consid-which proteins in the T-tubule bind erable influx of Ca 2+ is elicited by an to the foot protein. These are:

References action potential but this Ca 2+ current is I Somlyo, A. V., Gonzalez-Scrratos, H., Shuman, H., McClellan, E. and Somlyo, A. P. slow in onset. A second component of (1) affinity chromatography, employ-(1981)J. CelIBiol. 911,577 594 much more rapid activity has also been ing the isolated protein and dissolved 2 Fosset, M., Jaimovich, E., Delpont, E. and discerned in neonatal muscle 3°. This T-tubularvesicles; 

Proc. Natl 103

Proc. NatlAcad. Sci. USA 18 Varsanyi

Ce// (1985) Proc. Natl Acad. Sci. USA

Secondary reading frames, 'hidden' under other reading frames, are used ]br The use of an internal AUG codon coordinated expression of prgteins in several eukaryotic viruses. In some genes, for initiation of translation is an alternribosomal frameshifting and initiation or reinitiation of protein synthesis on internal AUG codons are tra.~slational mechanisms allowing access to such 'hid-ative to ribosomal frameshifting, and den' reading flames. In others, secondary reading flames are translated from does not result in production of fusion alternatively spliced or editedmRNAs, proteins ( Fig. 1 , bottom center, black box). Several conditions may allow the use of an internal AUG codon.(1) An Eukaryotic mRNAs are generally codon, only one protein will be internal AUG is sometimes used in monocistronic, but for sev,zral viral expressed (bottom left, stippled box). cases wherc the first AUG occurs in an transcripts ~-s as well as for orLe cellular Ribosomal frameshifting, subsequent unfavourable context for translation transcript 9, it has been shown that pro-to initiation on the first AUG codon, initiation 4.(2) Termination of protein teins encoded in different reading may enable expression of the overlap-synthesis at a stop codon may lead to frames are expressed. Since eukaryotic ping reading frame, creating a fusion reinitiation at a nearby AUG in ribosomes typically start protein syn-protein (bottom right, fused stippled another frame ~12.(3) A 'ribosome thesis at the first AUG codon only 4, it is and black boxes). Since only a fraction landing pad' may direct the ribosome to not immediately evident how second-of the ribosomes change frame at a an internal position in the mRNA, as ary reading frames overlapping with or frameshift signal, fusion proteins are described for the uncapped genomic following the first reading frame can be produced in addition to, rather than RNA of picornaviruses ~3. (4) In a capexpressed, instead of, the 'normal' protein, ped mRNA of a paramyxovirus, some Frameshifting in the -1 frame is ribosomes pass from the cap directly to Ribosomal frameshifting used by most retroviruses to access the an initiation codon far downstream14. A situation present in many viral reverse transcriptasc reading frame 5"~', It is interesting to note that the eukaryotic transcripts is illu~,trated in hidden in the gag mRNA, and by cousins of rctroviruses and the yeast Ty Fig. 1 sequences on which frameshifting hal AUG codon to express reverse