- Email: [email protected]
Tetanus toxin: Retrograde axonal transport of systemically administered toxin
Neuroscience Letters, 4 0 9 7 7 ) 61--65 © Elsevier/North-Holland Scientific Publishers Ltd.
61
TETANUS TOXIN: RETROGRADE AXONkL TRANSPORT OF SYSTEMICALLY ADMINISTERED TOXIN
DONALD L PRICE and JOHN W. GRIFFIN
Departments o f Neuro,ogy and Pathology (Division o f Neuropathology), The Johns Hopkins University School o f Medicine, Baltimore, Md. 21205 (U.S.A.) (Received October 27th, 1976) (Accepted December 10th, 1976)
SUMMARY
After intramuscular injection, [,2s I] tetanus toxin is carried 1;o spinal motor neurons by retrograde axonal transport. In the present study, we asked whether systemically administered toxin can be transported toward the spinal cord via this intraaxonal pathway. Mice were given [,2s I] tetanus toxin intraperitoneally, one sciatic nerve was ligated and, 24 h later, the animals were perfused. Autoradiograms showed labeled toxin accumulated within axons distal to the ligature. We interpret these findings as indicating that circulating toxin can leak out of intramuscular capillaries, be taken up at nerve terminals, and subsequently be carried toward the spinal cord by retrograde axonal transport. This pathway may be an important route by which systemically administered toxin can reach the CNS.
After intramu,~cular injection, [ a2s I] tetanus toxin is taken up at nerve terminals [I4] and carried to the CNS by retrograde axonal transport [3,] 5, 17--19]. In;efference with the retrograde transport system by freezing [$] or colchicine [ 3,19] results in a substantial reduction in the radioactivity in the nerve cells of ven~,ral spinal cord [3] and dorsal root ganglia [19]. However, it has not been demonstrated that blood-borne toxin can be carrS.ed to the CNS by the retrograde transport system. The present investigation was d2signed to evaluate the possibility that circulating tetanus toxin could be taken up at neuromu.:cular junctions and carried centripetally by axonal transport. Our approach was to inject iodinated toxin intraper, toneally into mice, to ligate one sciatic nerve short!y thereafter, and ~,o sacrifice animals 24 h later. The location of labeled toxin was evaluated by scintillation counting and autoradiography. These studies were performed on 20 g female CBA mice. Surgical proce-
62
dures were ca,Tied out under chloral hydrat~ anesthesia. In a group e f four animals, the ~ciatic nerve was ligated at the level of the sciatic notch. Iodinated toxin, prepared as previously described [4,15], was injected intraperitoneally (1.5 × 106 dpm) and, 24 h later, the animals were perfused with 2.5% glutaraldehyd~ and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer. Segments oi sciatic nerve were processed for autoradiography [15]. Thick sections (1 ]Jm) were dipped in Kodak NTB-2 emulsion, e x r ' ~ e d for 103 days, developed in Kodok D-19, stained with toluidine blue: and examined in a Zeiss photomicroscope. In a second group of fi'~'e animals, the nerves were either ligated (three animals ) as above or double ligated (two mdmals). Multiple 3 m m segments of sciatic nerve proximal and distal to the ~gature were counted in a Nuclear Chicago scintillation counter. In the single ligature experiments, autoradiography demonstrated radioactivity only within segments immediately distal to the ligature (Fig. 1). The axons on the distal side of the ligature were swollen with accumulated organelles (pellets) [5] ; the label was associated with ther~e pellets. The swollen axons on the proximal side of the ligature were not labeled. In these respects the autoradiograms resembled those obtained after intramuscular injection of toxin [3,5,15,19]. On the other hand, these autoradiogr~ms differed from those obtained after intramuscular injection in that they showed substantial numbers of silver grains in the perineurium on both sides of the ligature. This distribution of radioactivity was reflected in scintillation count studies which showed no differences between the proximal and distal sides of the ligature. In animals with a second ligature, there was label within nerve seg.
4
Fig. 1. Light autoradiogram showing intraaxonal radioactivity after systemic administration of [1251 ]tetanus toxin. Twenty-four hours before perfusion fixation, labeled toxin was injected intraperitoneally ana the sciatic nerve was ligated. Silver grains are present ,:c.swollen segments of axons (A) distal to the nerve ligation. Magnification u 918.
63
ments distal to the distal ligature, but silver ~ains were not present within axons in any segments proximal to the distal ligature. There was label within the perineurium op both sides of both ligatures and this was presumably the result of local tr,,uma to vessels [13]. Scintillation counting readily detected this radioacti~ ity, but it could not define the exact location of label. Hence, only by au~oradiography were we able to demonstrate the retrograde intraaxonal tran,~port of toxin. The autoradiograms indi~.ate that systemically administered toxin can reach axons and be carried toward the CNS by retrograde axonal transport. This observation suggests one pathway by which tetanus toxin, a protein weighing 130,000 daltons, may circumvent the blood-nerve [ 12] and bloodbrain [16] barriers which normally exclude macromolecules. Our results can be interpreted (Fig. 2) to suggest that circulating toxin leaks out the no, mally permeable vessels within muscle [ 1 0 ] , reaches neuromuscular regions by diffusion, is taken up into nerve terminals [9,14,20], and is subsequently carried to the CNS by retrograde axonal transport [2,3,15,17--19]. This interpretation is suppo:,~-~d by the recent demonstration that systemically administered horseradish peroxidase can pass across the capillaries of muscle and be taken dp at nerve terminals from which it is carried by retrograde transport to the ceU bodies of motor neurons [ 1]. That circulating toxin may enter the CNS by a neural route is consistent with several lines of evidence: (1) circulating toxin is cleared from the blood rapidly but enters the spinal cord slowly [7] ; (2) the toxin does not appear to migrate directly from blood stream to CNS [7,19] ; (3) after intravenous injection, toyin accumulates in neurons having axonal connections with the periphery [8] and (4)
I
t-.,l,..,.
r L_J,
Fig. 2. Schematic drawing of experimental model with interpretation of results. Intraperitoneally injected [~2~I ]tetanus toxin circulated in the bloc~d, leaked out of intramuscular capillaries, and was taken up at nerve terminals. Radioactivity accumulated within axons o n distal side of a ligature (stippled region). The ['25 1 ] toxin was carried o u t toward the CNS by retrograde axonal transp~:t.
64
interference with axonal transport results in a decreased amount of radioactivity m the spinal cord after eitherintramusctflar [3~or ~Tstemic administration [18] of [12s I] toxin. Finally, it has recently been shown that [~2s I] tetems toxin passes transsynaptically from the cell bodies of motor neurons to synaptic terminak in the ventral horn [17]. The time course of synaptic labeling parallels the appearance of clinical tetanus [17]. Thus, experimental studies, as well as c'Anical observations [ 1 I ] , lend ~ p port to the hypothesis that retrograde axonal transport is one pathway by which tetanus toxin may reach the CNS. ACKNOWLEDGEMENTS
This work was s~pported by USPHS Grants NS-10920 and NS-10580. We wish t~ thank Keith Peck and Jim Morris for their expert technical assistance. REFERENCES I Broedweli, R. an~ Brightman, W., Entry of peroxidase into neurons of the central and peripheral nez ~ous systems fr~>m ex'~racerebral and cerebral blood, J. comp. Neurol., 166 (1976) 257--284. 2 Dimpfel, W. and Kabermann, E., Histoautoradiographic localization of J2s I labeled tetanus toxin in rat spinal cord, Naunyn-Sehmiedeberg's Arch. exp. Path. Pharmak., 290 (1973) 177--182. 3 £rdmann, G., Wiegsnd, H. and WellhiSner, H.H., lr.traaxonal and extraaxorml transport of J~s I-tetanus to~in in early local tetanus, Naunyn-Sehmiedebarg's Arch. exp. Path. Pharmak., 290 (19'75) 357--373. 4 Griffin, J.W., Price, D.L., Drachman, D.B. and Engel, W.K., Axonal transport to and from the motor nerve ending, Ann. N.Y. Acad. 8cJ., 274 (1976} 31--45. 5 Griffin, J.W., Price, D.L., Engel.~ W.K. and Drachman, D.B., Pathogenesis of reactive axonal swellings: role of axonal transport, J. Neuropath. exp. Neurol., in press. 6 Habermann, E., Interaction of labeled tetanus toxin and toxoJd with substructures of rat brain and spinal cord in vitro, Naunyn-Schmiedeberg's Arch. exp. Path. Pharmak., 276 (1973) 341--359. 7 Habermann, E. and D;mpfel, W., Distribution of ls, [-tetanus toxin and , s I-toxoid in rats with generalized tetanus, as influenced by antitoxin, Naunyn-Schmiedeberg's Arch. exp. Path. Pharmak., 276 (1973) 327--340. 8 Habermann, E., Dimpfel, W. and Raker, K.D., Interaction of labeled tetanus toxin with substructures of rat spinal cord in v~vo, Naunyn-Schmiedeberg's Arch. exp. Path. Ph~rmak., 276 (1973) 361--373. 9 Heuser, J.E. and Reese, T.S., Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction, J. Cell Biol., 57 (1973) 315--344. 10 Karnovsky, M.J., The ultrastructural basis of capillary permeability studied with peroxidase as a tracer, J. Cell Biol., 35 (1967) 213--236. 11 Laurence, D.R. and Webster, R.A~, Pathologic physiology, pharmacology and therapeutics of tetanus, Clin. Pharmacol. Thvr., 4 (1963) 36--72. 12 Ols.~on, Y. and Reese, T.S., Pe~'meability of vasa nervorum and perine...]um in mouse sciatic nerve studies by fluorescence and electron microscopy, J. Neuropath. exp. Neurol., 30 (1971) i05--J.19. 13 Oissoll: Y. and Kristensson, K., The perineurium as a diffusion barrier to protein tracers following trauma to ner:es, Acts neuropath. (Bed.), 23 (1973) 105--111.
65 14 Price, D., Griffin, J. and Peck, K., Tetanus toxin: evidence for binding at presynaptic r,erve endings, Brain Res., in press. 15 Price, D.L., Griffin, J., Young, A., Peck, K. and ~tucks, A., Tetanus tozin: direct evidence for retrograde intraaxonal transport, Science, 188 (1975) 945--947. 16 Reese, T.S. and Kamovsky, M.J., Fine structur~ loca~ization of a blood-brain barrier for exogenous peroxidase, J. Cell ~iol., 34 (1967) 207--217. 17 Schwab, M.E. and Thoenen, H., Electron microscopic evidence for a t~anssynaptic migration of tetanus toxin in spinal cord motor neurons: an electron microscopic and morphometric study, Brain Res., 105 (1976) 213--227. 18 Seib, U.C., Hensel, B., Wiegand, H. and Weilh~ner, H.H., Supporting evidence for a role of neuronal ascent of toxin in the pathogenesis of general tetanus in cats, NaunynSchmiedeberg's Arch. exp. Path. Pharmak., 276 (1973)403--411. 19 Stoeckei, IC, Schwab, M. and Thoenen, H., Comparison between the retrograde axonal transport of nerve growth factor and tetanus toxin in motor, sensory and adrenergic neurons, Brain Res., 99 (1975) 1--16. 20 Zacks, S.I. and Saito, A., Uptake of exogenous horseradish peroxidase by coated vesicles in mouse neuromuscular junctions, J. Histochem. Cytochem., 17 (i969) 161--170.
61
TETANUS TOXIN: RETROGRADE AXONkL TRANSPORT OF SYSTEMICALLY ADMINISTERED TOXIN
DONALD L PRICE and JOHN W. GRIFFIN
Departments o f Neuro,ogy and Pathology (Division o f Neuropathology), The Johns Hopkins University School o f Medicine, Baltimore, Md. 21205 (U.S.A.) (Received October 27th, 1976) (Accepted December 10th, 1976)
SUMMARY
After intramuscular injection, [,2s I] tetanus toxin is carried 1;o spinal motor neurons by retrograde axonal transport. In the present study, we asked whether systemically administered toxin can be transported toward the spinal cord via this intraaxonal pathway. Mice were given [,2s I] tetanus toxin intraperitoneally, one sciatic nerve was ligated and, 24 h later, the animals were perfused. Autoradiograms showed labeled toxin accumulated within axons distal to the ligature. We interpret these findings as indicating that circulating toxin can leak out of intramuscular capillaries, be taken up at nerve terminals, and subsequently be carried toward the spinal cord by retrograde axonal transport. This pathway may be an important route by which systemically administered toxin can reach the CNS.
After intramu,~cular injection, [ a2s I] tetanus toxin is taken up at nerve terminals [I4] and carried to the CNS by retrograde axonal transport [3,] 5, 17--19]. In;efference with the retrograde transport system by freezing [$] or colchicine [ 3,19] results in a substantial reduction in the radioactivity in the nerve cells of ven~,ral spinal cord [3] and dorsal root ganglia [19]. However, it has not been demonstrated that blood-borne toxin can be carrS.ed to the CNS by the retrograde transport system. The present investigation was d2signed to evaluate the possibility that circulating tetanus toxin could be taken up at neuromu.:cular junctions and carried centripetally by axonal transport. Our approach was to inject iodinated toxin intraper, toneally into mice, to ligate one sciatic nerve short!y thereafter, and ~,o sacrifice animals 24 h later. The location of labeled toxin was evaluated by scintillation counting and autoradiography. These studies were performed on 20 g female CBA mice. Surgical proce-
62
dures were ca,Tied out under chloral hydrat~ anesthesia. In a group e f four animals, the ~ciatic nerve was ligated at the level of the sciatic notch. Iodinated toxin, prepared as previously described [4,15], was injected intraperitoneally (1.5 × 106 dpm) and, 24 h later, the animals were perfused with 2.5% glutaraldehyd~ and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer. Segments oi sciatic nerve were processed for autoradiography [15]. Thick sections (1 ]Jm) were dipped in Kodak NTB-2 emulsion, e x r ' ~ e d for 103 days, developed in Kodok D-19, stained with toluidine blue: and examined in a Zeiss photomicroscope. In a second group of fi'~'e animals, the nerves were either ligated (three animals ) as above or double ligated (two mdmals). Multiple 3 m m segments of sciatic nerve proximal and distal to the ~gature were counted in a Nuclear Chicago scintillation counter. In the single ligature experiments, autoradiography demonstrated radioactivity only within segments immediately distal to the ligature (Fig. 1). The axons on the distal side of the ligature were swollen with accumulated organelles (pellets) [5] ; the label was associated with ther~e pellets. The swollen axons on the proximal side of the ligature were not labeled. In these respects the autoradiograms resembled those obtained after intramuscular injection of toxin [3,5,15,19]. On the other hand, these autoradiogr~ms differed from those obtained after intramuscular injection in that they showed substantial numbers of silver grains in the perineurium on both sides of the ligature. This distribution of radioactivity was reflected in scintillation count studies which showed no differences between the proximal and distal sides of the ligature. In animals with a second ligature, there was label within nerve seg.
4
Fig. 1. Light autoradiogram showing intraaxonal radioactivity after systemic administration of [1251 ]tetanus toxin. Twenty-four hours before perfusion fixation, labeled toxin was injected intraperitoneally ana the sciatic nerve was ligated. Silver grains are present ,:c.swollen segments of axons (A) distal to the nerve ligation. Magnification u 918.
63
ments distal to the distal ligature, but silver ~ains were not present within axons in any segments proximal to the distal ligature. There was label within the perineurium op both sides of both ligatures and this was presumably the result of local tr,,uma to vessels [13]. Scintillation counting readily detected this radioacti~ ity, but it could not define the exact location of label. Hence, only by au~oradiography were we able to demonstrate the retrograde intraaxonal tran,~port of toxin. The autoradiograms indi~.ate that systemically administered toxin can reach axons and be carried toward the CNS by retrograde axonal transport. This observation suggests one pathway by which tetanus toxin, a protein weighing 130,000 daltons, may circumvent the blood-nerve [ 12] and bloodbrain [16] barriers which normally exclude macromolecules. Our results can be interpreted (Fig. 2) to suggest that circulating toxin leaks out the no, mally permeable vessels within muscle [ 1 0 ] , reaches neuromuscular regions by diffusion, is taken up into nerve terminals [9,14,20], and is subsequently carried to the CNS by retrograde axonal transport [2,3,15,17--19]. This interpretation is suppo:,~-~d by the recent demonstration that systemically administered horseradish peroxidase can pass across the capillaries of muscle and be taken dp at nerve terminals from which it is carried by retrograde transport to the ceU bodies of motor neurons [ 1]. That circulating toxin may enter the CNS by a neural route is consistent with several lines of evidence: (1) circulating toxin is cleared from the blood rapidly but enters the spinal cord slowly [7] ; (2) the toxin does not appear to migrate directly from blood stream to CNS [7,19] ; (3) after intravenous injection, toyin accumulates in neurons having axonal connections with the periphery [8] and (4)
I
t-.,l,..,.
r L_J,
Fig. 2. Schematic drawing of experimental model with interpretation of results. Intraperitoneally injected [~2~I ]tetanus toxin circulated in the bloc~d, leaked out of intramuscular capillaries, and was taken up at nerve terminals. Radioactivity accumulated within axons o n distal side of a ligature (stippled region). The ['25 1 ] toxin was carried o u t toward the CNS by retrograde axonal transp~:t.
64
interference with axonal transport results in a decreased amount of radioactivity m the spinal cord after eitherintramusctflar [3~or ~Tstemic administration [18] of [12s I] toxin. Finally, it has recently been shown that [~2s I] tetems toxin passes transsynaptically from the cell bodies of motor neurons to synaptic terminak in the ventral horn [17]. The time course of synaptic labeling parallels the appearance of clinical tetanus [17]. Thus, experimental studies, as well as c'Anical observations [ 1 I ] , lend ~ p port to the hypothesis that retrograde axonal transport is one pathway by which tetanus toxin may reach the CNS. ACKNOWLEDGEMENTS
This work was s~pported by USPHS Grants NS-10920 and NS-10580. We wish t~ thank Keith Peck and Jim Morris for their expert technical assistance. REFERENCES I Broedweli, R. an~ Brightman, W., Entry of peroxidase into neurons of the central and peripheral nez ~ous systems fr~>m ex'~racerebral and cerebral blood, J. comp. Neurol., 166 (1976) 257--284. 2 Dimpfel, W. and Kabermann, E., Histoautoradiographic localization of J2s I labeled tetanus toxin in rat spinal cord, Naunyn-Sehmiedeberg's Arch. exp. Path. Pharmak., 290 (1973) 177--182. 3 £rdmann, G., Wiegsnd, H. and WellhiSner, H.H., lr.traaxonal and extraaxorml transport of J~s I-tetanus to~in in early local tetanus, Naunyn-Sehmiedebarg's Arch. exp. Path. Pharmak., 290 (19'75) 357--373. 4 Griffin, J.W., Price, D.L., Drachman, D.B. and Engel, W.K., Axonal transport to and from the motor nerve ending, Ann. N.Y. Acad. 8cJ., 274 (1976} 31--45. 5 Griffin, J.W., Price, D.L., Engel.~ W.K. and Drachman, D.B., Pathogenesis of reactive axonal swellings: role of axonal transport, J. Neuropath. exp. Neurol., in press. 6 Habermann, E., Interaction of labeled tetanus toxin and toxoJd with substructures of rat brain and spinal cord in vitro, Naunyn-Schmiedeberg's Arch. exp. Path. Pharmak., 276 (1973) 341--359. 7 Habermann, E. and D;mpfel, W., Distribution of ls, [-tetanus toxin and , s I-toxoid in rats with generalized tetanus, as influenced by antitoxin, Naunyn-Schmiedeberg's Arch. exp. Path. Pharmak., 276 (1973) 327--340. 8 Habermann, E., Dimpfel, W. and Raker, K.D., Interaction of labeled tetanus toxin with substructures of rat spinal cord in v~vo, Naunyn-Schmiedeberg's Arch. exp. Path. Ph~rmak., 276 (1973) 361--373. 9 Heuser, J.E. and Reese, T.S., Evidence for recycling of synaptic vesicle membrane during transmitter release at the frog neuromuscular junction, J. Cell Biol., 57 (1973) 315--344. 10 Karnovsky, M.J., The ultrastructural basis of capillary permeability studied with peroxidase as a tracer, J. Cell Biol., 35 (1967) 213--236. 11 Laurence, D.R. and Webster, R.A~, Pathologic physiology, pharmacology and therapeutics of tetanus, Clin. Pharmacol. Thvr., 4 (1963) 36--72. 12 Ols.~on, Y. and Reese, T.S., Pe~'meability of vasa nervorum and perine...]um in mouse sciatic nerve studies by fluorescence and electron microscopy, J. Neuropath. exp. Neurol., 30 (1971) i05--J.19. 13 Oissoll: Y. and Kristensson, K., The perineurium as a diffusion barrier to protein tracers following trauma to ner:es, Acts neuropath. (Bed.), 23 (1973) 105--111.
65 14 Price, D., Griffin, J. and Peck, K., Tetanus toxin: evidence for binding at presynaptic r,erve endings, Brain Res., in press. 15 Price, D.L., Griffin, J., Young, A., Peck, K. and ~tucks, A., Tetanus tozin: direct evidence for retrograde intraaxonal transport, Science, 188 (1975) 945--947. 16 Reese, T.S. and Kamovsky, M.J., Fine structur~ loca~ization of a blood-brain barrier for exogenous peroxidase, J. Cell ~iol., 34 (1967) 207--217. 17 Schwab, M.E. and Thoenen, H., Electron microscopic evidence for a t~anssynaptic migration of tetanus toxin in spinal cord motor neurons: an electron microscopic and morphometric study, Brain Res., 105 (1976) 213--227. 18 Seib, U.C., Hensel, B., Wiegand, H. and Weilh~ner, H.H., Supporting evidence for a role of neuronal ascent of toxin in the pathogenesis of general tetanus in cats, NaunynSchmiedeberg's Arch. exp. Path. Pharmak., 276 (1973)403--411. 19 Stoeckei, IC, Schwab, M. and Thoenen, H., Comparison between the retrograde axonal transport of nerve growth factor and tetanus toxin in motor, sensory and adrenergic neurons, Brain Res., 99 (1975) 1--16. 20 Zacks, S.I. and Saito, A., Uptake of exogenous horseradish peroxidase by coated vesicles in mouse neuromuscular junctions, J. Histochem. Cytochem., 17 (i969) 161--170.