J Neurophysiol 105: 1963–1965, 2011.
First published January 27, 2011; doi:10.1152/jn.01104.2010.
Neuro Forum
Developmental plasticity of descending motor pathways
Hsiu-Ling Li1 and Curtis O. Asante2
1
Department of Neuroscience, Columbia University, and 2Department of Physiology, Pharmacology and Neuroscience, The
City College of The City University of New York, New York City, New York
Submitted 17 December 2010; accepted in final form 25 January 2011
functional recovery; juvenile animal brains
JUVENILE ANIMAL BRAINS ARE highly plastic because refinement
and maturation of the CNS continues after birth. Thus, juvenile
brain injury often triggers remodeling of the CNS, which may
lead to functional recovery if the newly formed circuits are
properly wired. For instance, it has been shown that juvenile
rats receiving unilateral lesions of the sensorimotor cortex
(SMC) can develop fairly normal grasping and reaching movements later in adulthood (Barth and Stanfield 1990; Hicks and
D’Amato 1970). This functional recovery is attributed to the
remodeling of the corticospinal system from the undamaged
hemisphere (Barth and Stanfield 1990). While the mature
corticospinal terminations predominantly innervate the contralateral side, unilateral brain damage in neonatal rats may
induce aberrant sprouting of undamaged corticofugal projections to the ipsilateral side. Moreover, these atypical circuits
are functional since intracortical microstimulation of the undamaged SMC elicits responses of ipsilateral limb muscles
(Kartje-Tillotson et al. 1985). Undoubtedly, large-scale reorganization of the corticospinal system occurs to compensate for
the functional loss after early brain injury. However, the
organization of circuits that originate from the undamaged
SMC and comprise connections with the motoneurons (MNs)
of ipsilateral forelimbs, is still unclear.
In a recent issue of Journal of Neurophysiology, Umeda et
al. (Umeda et al. 2010) performed a series of electrophysiological recordings to identify the descending pathways that
relay motor commands to the affected forelimb muscles in
hemidecorticated rats. They first determined that cervical segments C5–C7 compared with other cervical segments received
the densest innervation from corticospinal axons since the
pyramidal stimulation evoked the strongest field potentials at
these cervical levels. In control rats, pyramidal stimulation
induced a significant field potential on the contralateral, but not
Address for reprint requests and other correspondence: H. Li, 1051 Riverside Drive, PI annex 645, New York, NY 10032 (e-mail: [email protected]
or [email protected]).
www.jn.org
the ipsilateral, side. In contrast, stimulation of the undamaged
pyramid of the hemidecorticated rats evoked comparable responses bilaterally, suggesting the emergence of compensatory
corticospinal circuits that might mediate the functional recovery of injured rats.
The authors next asked whether these atypical projections
are able to excite the ipsilateral MNs in hemidecorticated rats.
Following single or repetitive pulses of pyramidal stimulation,
the intracellular responses of the MNs located in spinal segments C5–C7 were recorded. In control rats, a single pulse of
stimulation was not sufficient to elicit bilateral MN responses.
However, repetitive stimulation evoked significant EPSPs in a
high percentage of contralateral MNs (92.5%). It is estimated
that 50 – 60% of the recorded MNs are deep radial (DR) MNs,
which control the contraction of forelimb muscles via DR
nerves. In contrast, the same protocol only induced modest
responses from a smaller population of ipsilateral MNs
(67.2%). In hemidecorticated rats, repetitive stimulation on the
undamaged pyramid evoked significant EPSPs of MNs bilaterally. Moreover, the proportions of MNs responsive to the
stimulation were comparably high on both contralateral
(95.2%) and ipsilateral (89.4%) sides. Together, these results
suggest that compensatory circuits from the undamaged pyramid act to activate ipsilateral MNs on the affected side.
Interestingly, the average segmental latencies of contralateral and ipsilateral MNs in hemidecorticated rats were comparable to each other, but significantly shorter than those of their
counterparts in control rats. Moreover, a high percentage of
MNs from both sides of hemidecorticated rats exhibited very
rapid responses with latencies less than 3.0 ms, whereas the
MNs in control rats displayed a wide range of latencies. Based
on the segmental latencies, several pathways are hypothesized
to mediate the activation of ipsilateral MNs in injured rats. The
monosynaptic pathway between the dorsal, lateral, and ventral
components of the corticospinal tracts (CSTs) and segmental
spinal neurons is known to have latencies from 0.3 to 0.9 ms
(Alstermark et al. 2004). Furthermore, the authors speculate
that the reticulospinal pathway has faster conduction velocity
compared with CSTs, which could possibly evoke the responses of MNs within 0.3 ms. In contrast, the delayed responses with latencies longer than 0.9 ms could be mediated
via the oligosynaptic connections from CSTs or extra-pyramidal tracts to spinal interneurons. These data suggest that both
corticospinal and cortico-reticulo-spinal pathways might play a
role in activating the ipsilateral MNs in hemidecorticated rats.
The authors next investigated to what extent the pyramidal
excitation of ipsilateral MNs is relayed via the reticulospinal
neurons in the brain stem by transecting the dorsal column at
C2 to interrupt CST transmission in hemidecorticated rats. As
expected, this treatment abolished both the field potential and
0022-3077/11 Copyright © 2011 the American Physiological Society
1963
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 14, 2017
Li HL, Asante CO. Developmental plasticity of descending motor
pathways. J Neurophysiol 105: 1963–1965, 2011. First published
January 27, 2011; doi:10.1152/jn.01104.2010.—Juvenile animal
brains are highly plastic and thus often achieve better functional
recovery after injury compared with adult brains. Recently, Umeda et
al. (Umeda T, Takahashi M, Isa K, Isa T. J Neurophysiol 104:
1707–1716, 2010) have shown that the remodeling of both corticospinal and extra-pyramidal pathways can contribute to the recovery of
grasping and reaching ability in hemidecorticated juvenile rats. They
have further unveiled the strengthening of the cortico-reticulo-spinal
pathway after injury, that mediates the fast excitation of ipsilateral
motoneurons for functional recovery.
Neuro Forum
1964
PLASTICITY OF MOTOR PATHWAYS
therefore likely that although the effects obtained may be
largely attributed to cortico-reticular-spinal pathways, one cannot discount the contribution of these other pathways. In fact,
the authors did explore the possible involvement of rubrospinal
tracts and propriospinal circuits in the functional recovery of
hemidecorticated rats. However, their results by retrograde
labeling have excluded these possibilities (unpublished data).
Thus, future experiments will still be needed to test the role of
other aforementioned pathways in mediating the motor recovery of hemidecorticated rats. However, compared with the high
conduction velocity and efficiency of the direct cortico-reticulo-spinal pathway, it is likely that these pathways may only
mediate delayed, rather than fast, excitation of ipsilateral MNs.
Nonetheless, this series of experiments has clearly demonstrated the importance of direct cortico-reticulo-spinal pathways in evoking fast responses of ipsilateral motor units and
thereby promoting motor recovery after juvenile brain injury.
In contrast, although less characterized in this study, the
indirect cortico-reticulo-spinal pathways might also contribute to slow responses of ipsilateral MNs via spinal interneurons, including commissural interneurons. In fact, these
ipsilateral cortico-reticulo-spinal pathways, although minor,
are normally present in control rats (Fig. 1B). However,
ipsilateral pyramidal stimulation alone rarely evokes any
responses of ipsilateral MNs under physiological conditions.
Fig. 1. Hemidecortication-induced plasticity of corticospinal (A) and reticulospinal (B) pathways. A: in
control rats (left), mature corticospinal tract (CST)
projections predominantly innervate spinal interneurons (IN) on the contralateral side. However, in juvenile hemidecorticated rats (right), undamaged CST
projects bilaterally (blue) and therefore can mediate
the activation of both ipsilateral (red) and contralateral (blue) motor neurons (MN) following pyramidal
stimulation. B: under physiological conditions (left),
the cortico-reticulo-spinal pathways are not effective
enough to activate ipsilateral MNs (dashed lines).
However, the effectiveness of these pathways is strengthened (thick line) after hemidecortication (right) and
thereby sufficient to mediate both fast and slow excitation
of ipsilateral MNs via monosynaptic (red) and oligosynaptic (black) connections, respectively.
J Neurophysiol • VOL
105 • MAY 2011 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 14, 2017
the excitation of MNs on the contralateral side following
pyramidal stimulation. In contrast, the same protocol was still
able to evoke EPSPs on ipsilateral MNs, despite a smaller
magnitude. The latencies to evoke these residual responses
(less than 3.0 ms) are similar to those before transection.
Together, these data further confirm that both corticospinal and
reticulospinal pathways could account for the excitation of
ipsilateral MNs in hemidecorticated rats (Fig. 1). More importantly, disruption of CST transmission further suggested that
the reticulospinal pathway might activate ipsilateral MNs efficiently within 0.3 ms via monosynaptic connections or with a
longer delay via oligosynaptic connections (Fig. 1B). These
findings are further supported by their unpublished data that
direct stimulation on the medullary reticular formation could
activate ipsilateral MNs efficiently.
Here, it is noteworthy that dorsal column transection at
cervical level C2 is an effective way to eliminate the main
component of the CST (dorsal CST), but this treatment does
not disrupt the smaller components of the CST that travel down
more lateral and ventral parts of the spinal cord. In addition,
apart from CSTs, other prominent descending tracts such as
rubrospinal and vestibulospinal pathways are also implicated in
motor function. Importantly, these pathways would not have
been affected by the dorsal column transection since their route
to spinal cord gray matter is not via the dorsal column. It is
Neuro Forum
PLASTICITY OF MOTOR PATHWAYS
J Neurophysiol • VOL
over, clinical evidence has further suggested that these aberrant
projections might aggravate the motor deficits associated with
the hemiplegic CP patients. Loss of responses from the injured
cortex and maintenance of fast ipsilateral projections from the
unaffected cortex is strongly associated with a poor outcome in
hemiplegic CP patients. A possible explanation for this discrepancy is that these aberrant projections, at the time of brain
injury, fail to establish functional linkages with the cortical and
subcortical network required for effective arm and hand control. Furthermore, these ipsilateral projections could even displace the surviving projections from the injured hemisphere
and cause a “dominant negative” effect that leads to a gradual
loss of residual motor skills. Thus, injury-induced plasticity
may be reparative only if the newly formed circuits are functionally integrated into the CNS; otherwise, it may only aggravate the progression of motor impairments.
In summary, Umeda et al. (2010) demonstrate the importance of corticospinal and reticulospinal pathways in mediating
the functional recovery of reaching and grasping ability in
hemidecorticated rats (Fig. 1). Of particular interest is that the
authors have underscored an alternative pathway, the corticoreticulo-spinal pathway, in mediating the fast excitation of
ipsilateral MNs for functional recovery.
ACKNOWLEDGMENTS
The authors thank Drs. Kathleen Friel and Ben S. Huang for comments.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
REFERENCES
Alstermark B, Ogawa J, Isa T. Lack of monosynaptic corticomotoneuronal
EPSPs in rats: disynaptic EPSPs mediated via reticulospinal neurons and
polysynaptic EPSPs via segmental interneurons. J Neurophysiol 91: 1832–
1839, 2004.
Barth TM, Stanfield BB. The recovery of forelimb-placing behavior in rats
with neonatal unilateral cortical damage involves the remaining hemisphere.
J Neurosci 10: 3449 –3459, 1990.
Eyre JA. Corticospinal tract development and its plasticity after perinatal
injury. Neurosci Biobehav Rev 31: 1136 –1149, 2007.
Friel KM, Martin JH. Bilateral activity-dependent interactions in the developing corticospinal system. J Neurosci 27: 11083–11090, 2007.
Hicks SP, D’Amato CJ. Motor-sensory and visual behavior after hemispherectomy in newborn and mature rats. Exp Neurol 29: 416 – 438, 1970.
Jankowska E, Edgley SA. How can corticospinal tract neurons contribute to
ipsilateral movements? A question with implications for recovery of motor
functions. Neuroscientist 12: 67–79, 2006.
Kartje-Tillotson G, Neafsey EJ, Castro AJ. Electrophysiological analysis of
motor cortical plasticity after cortical lesions in newborn rats. Brain Res
332: 103–111, 1985.
King VR, Michael GJ, Joshi RK, Priestley JV. trkA, trkB, and trkC
messenger RNA expression by bulbospinal cells of the rat. Neuroscience 92:
935–944, 1999.
Martin JH, Friel KM, Salimi I, Chakrabarty S. Activity- and use-dependent
plasticity of the developing corticospinal system. Neurosci Biobehav Rev 31:
1125–1135, 2007.
Nashmi R, Fehlings MG. Mechanisms of axonal dysfunction after spinal cord
injury: with an emphasis on the role of voltage-gated potassium channels.
Brain Res Brain Res Rev 38: 165–191, 2001.
Umeda T, Takahashi M, Isa K, Isa T. Formation of descending pathways
mediating cortical command to forelimb motoneurons in neonatally hemidecorticated rats. J Neurophysiol 104: 1707–1716, 2010.
105 • MAY 2011 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 14, 2017
It is therefore likely that the synaptic efficacy of these
pathways is augmented in developing motor circuits after
hemidecortication (Fig. 1B). One possible mechanism is that
these ipsilateral corticoreticular fibers may expand their
synaptic territories in the absence of their contralateral
counterparts during development and thereby enhance the
efficacy of transmission. Alternatively, hemidecortication
may cause changes in the microenvironment of both brain
and spinal cord, such as the expression levels of neurotransmitters and growth factors. Neurotransmitters, such as serotonin and nor-adrenaline, have been shown to modulate
the activation of commissural interneurons and may thereby
strengthen the indirect cortico-reticulo-spinal pathways
(Jankowska and Edgley 2006). Moreover, it has been shown
that reticulospinal tracts express high levels of TrkB and
TrkC receptors and thus the outgrowth and the synaptic
strength of reticulospinal tracts may be influenced by
changes in the levels of BNDF and neurotrophins (King et
al. 1999). Together, ipsilateral cortico-reticulo-spinal pathways may provide an alternate pathway that replaces the
functions of damaged motor circuits after juvenile brain
injury. In fact, treatment with the potassium channel
blocker, 4-AP, has been shown to enhance the effectiveness
of these cortico-reticulo-spinal pathways to activate the
ipsilateral MNs (Jankowska and Edgley 2006). Clinical use
of 4-AP has also been proven beneficial for the motor
deficits associated with multiple sclerosis and after various
spinal cord injuries (Nashmi and Fehlings 2001). For future
studies, it will be of clinical importance to better understand
the mechanisms underlying the enhancement of ipsilateral
cortico-reticulo-spinal pathways for the treatment of motor
deficits associated with early brain injury, such as cerebral
palsy.
While the reticulospinal pathways are functionally important
in hemidecorticated rats, CSTs appear to be the major locus of
remodeling following juvenile brain damage across animal
species. Developing CST projections reach the mature pattern
after birth (Martin et al. 2007). This maturation is thought to be
driven by primary motor (M1) cortical activity. Thus, early
brain damage often leads to the rewiring of developing CSTs.
Intuitively, this injury-induced plasticity would be expected to
be beneficial. However, studies in both cats and humans have
raised the question as to whether the anatomical or synaptic
plasticity associated with early brain injury are always reparative (Eyre 2007; Martin et al. 2007). Unilateral inactivation of
M1 activity in kittens leads to hemiparesis that correlates with
a paucity of CST projections from the damaged hemisphere,
whereas CST projections from the undamaged side send exuberant projections bilaterally (Friel and Martin 2007). Accordingly, the CSTs in hemiplegic cerebral palsy (CP) patients no
longer respond to transcranial magnetic stimulation (TMS) on
the plegic side, yet TMS of the intact side results in bilateral
activation (Eyre 2007). The presence of bilateral CST projections in hemiplegic cats and CP patients is similar to the
undamaged CST pathways, which activate bilateral MNs in
hemidecorticated rats (Fig. 1A). However, the hemiplegic cats
are still unable to use their forelimbs even in the presence of
atypical ipsilateral CST projections in the plegic side. More-
1965