NEUROLOGICAL PROGRESS Cellular and Molecular Mechanisms of Neural Repair after Stroke: Making Waves S. Thomas Carmichael, MD, PhD Stroke is associated with a limited degree of functional recovery. Imaging studies in humans have shown that reorganization in periinfarct and connected cortical areas most closely correlates with functional recovery after stroke. On a cellular level, two major regenerative events occur in periinfarct cortex: axons sprout new connections and establish novel projection patterns, and newly born immature neurons migrate into periinfarct cortex. Stroke induces a unique microenvironment for axonal sprouting in periinfarct cortex, in which growth-inhibitory molecules are reduced for 1 month after the infarct. During this period, neurons activate growth-promoting genes in successive waves. Neurogenesis also occurs through waves of migration of immature neurons from their origin in the subventricular zone into periinfarct cortex. This migration is mediated, in part, by the cytokine erythropoietin. These data indicate that the cellular environment after stroke is far from one of just death and destruction, but rather involves a longer evolving process of neuronal regeneration. Poststroke neuronal regeneration is characterized by waves of specific cellular and molecular events. Manipulating these waves of regeneration may provide for novel therapies that will improve recovery after stroke. Ann Neurol 2006;59:735–742 Stroke is the leading cause of adult disability. Sixty percent of survivors have disabilities in arm or leg use, and up to one-third need placement in a nursing home or assisted living environment. This translates to more than $30 billion in annual care for stroke survivors— well more than half of the annual cost of stroke.1,2 Recently, studies have shown that after stroke existing neurons can sprout new connections and new neurons can migrate to areas of injury. Axonal sprouting and neurogenesis after stroke occur in periinfarct tissue, a region of cortical remapping that closely correlates with recovery after stroke in humans. This review focuses on the cellular and molecular mechanisms of axonal sprouting and neurogenesis in periinfarct cortex, and how an understanding of these two processes might lead to novel therapies for neural repair after stroke. Patterns of Recovery in the Human Brain after Stroke: Periinfarct Cortex Stroke reorganizes cognitive maps in cortex in patterns that correlate with functional recovery. Subcortical stroke first induces a bilateral cortical activation to sensory or motor stimulation of the affected limb. This bilateral pattern then becomes more restricted as patients recover, to a reorganized and expanded sensorimotor activation in the cortex contralateral to the affected limb (the side of the stroke) and an increase in From the Department of Neurology, David Geffen School of Medicine at the University of California Los Angeles, Los Angeles, CA. Received Feb 21, 2006. Accepted for publication Feb 25, 2006. Published online Apr 24, 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20845 activity in the connected regions of supplementary motor and premotor areas on this same side.3–5 In crosssectional imaging studies of patients with good functional recovery, cortical activation occurs in an expanded or altered representation in sensorimotor areas contralateral to the affected limb, and this “ipsilesional” activation is most closely correlated with good recovery. Cortical activity in the cortex opposite to the stroke correlates with reduced functional recovery.6 – 8 These studies are limited by the types of sensory or motor activation paradigms that can be performed during scanning; other studies suggest that alternative recovery patterns may be present.9 However, similar data have been obtained after functional recovery from hemispatial neglect10 and aphasia,11,12 in which recovery is associated with a reorganization of cortical activity over time in a network of areas ipsilateral to the lesion. Similarly, transcranial magnetic stimulation mapping studies indicate that stroke induces new movement representations in periinfarct motor cortex, which correlate with functional recovery.13,14 Focal inactivation with transcranial magnetic stimulation in recovered patients in cortex adjacent or ipsilateral to the stroke produces deficits in movement of the affected limb, but inactivation of the cortex contralateral to the stroke does not alter limb movement.15,16 These findings from multiple brain activation paradigms indicate Address correspondence to Dr Carmichael, Department of Neurology, David Geffen School of Medicine at UCLA, Neuroscience Research Building, 635 Charles Young Drive South, Los Angeles, CA 90095. E-mail: firstname.lastname@example.org © 2006 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 735 that functional recovery after stroke occurs largely through an evolving reorganization of cortical activity within cortex adjacent to, or connected with, the stroke site, and not through the development of novel contralateral cortical activation patterns. Axonal Sprouting in Periinfarct Cortex Adult cortex responds to injury or deafferentation with axonal sprouting. Peripheral nerve injury or altered limb use produce substantial axonal sprouting in sensory cortex in nonhuman primates.17 Retinal lesions induce neurons in cortex that borders the deafferented area to sprout new projections into this area.18 Cerebellar lesions induce sprouting from primary sensory cortex into motor cortex.19 These studies indicate that axonal sprouting in cortex after peripheral deafferentation occurs in different species, over long distances, and between functionally different cortical areas. Axonal sprouting occurs in periinfarct and connected cortical areas after stroke. Stroke induces GAP43, a growth cone phosphoprotein that is highly linked to axonal sprouting,20 in human periinfarct cortex21 and in periinfarct cortex in experimental stroke models.22,23 GAP43 messenger RNA and protein are increased from 3 days to 1 month after stroke,21–23 followed by increases in a marker for synapses within the same periinfarct region at later time points.22 This progression from an increase in a growth cone molecule to increases in a marker for synapses suggested a progression toward axonal sprouting in periinfarct cortex. Axonal sprouting was directly demonstrated after stroke using detailed neuroanatomical mapping of cortical connections in control and poststroke brains. Stroke induces neurons in cortex both near the infarct and several millimeters away to form new connections. In the rodent, this poststroke axonal sprouting is robust and remaps the primary sensory cortex.24 In nonhuman primates, poststroke axonal sprouting extends new connections from premotor cortex to primary sensory cortex, in some cases, over distances of up to 1cm.25 The pattern of poststroke axonal sprouting in experimental animals resembles the pattern of cortical remapping in human stroke imaging studies: Axonal sprouting produces an altered and expanded primary sensory map, and new connections are formed between periinfarct cortex and premotor, motor, and somatosensory cortical areas. Stroke Induces a Permissive Environment for Axonal Sprouting The adult brain is normally inhibitory to axonal sprouting. Axonal growth inhibition is mediated through three general classes of proteins: myelinassociated proteins (NogoA, myelin-associated glycoprotein, oligodendrocyte myelin glycoprotein), extracellular matrix proteins (tenascin and chondroitin sulfate proteoglycans), and developmentally associated growth 736 Annals of Neurology Vol 59 No 5 May 2006 cone inhibitors (such as molecules of the ephrin and semaphorin classes).26 After injury to the central nervous system (CNS), many of these molecules are induced near the injury site. This reexpression of axonal growth inhibitors has been particularly well studied in spinal cord injury, where trauma induces chondroitin sulfate proteoglycans, NogoA, ephrin B/EphB class members, and semaphorin IIIa in the region of glial scar.27–31 The pattern of growth-inhibitory protein expression in the brain also has been studied extensively after brain trauma. Brain injury from stab wounds or aspiration of tissue induces chondroitin sulfate proteoglycans, NogoA, and semaphorin IIIa in the region of glial scar and its vicinity.27,30,32 These data have suggested that injury to the brain resembles injury to the spinal cord in broadly inducing growth-inhibitory molecules in a gradient that extends from the injury site into surrounding tissue. However, brain trauma is not stroke. The mechanisms of injury in brain trauma differ substantially from those in stroke, especially regarding the dynamics and distribution of inflammation, apoptotic cell death, and reperfusion injury/free radical damage.33–36 It is not surprising, therefore, that the degree of axonal sprouting differs markedly between brain trauma and stroke. Stab wounds or aspiration injury to the brain does not induce a significant axonal sprouting response,37– 40 whereas ischemic injury induces a substantial local and long-distance axonal sprouting response.24,25,37,40 The tissue environment that is induced by stroke differs from that seen after brain trauma in containing a region that is permissive for axonal sprouting. Axonal sprouting in periinfarct cortex takes place within a favorable environment that is adjacent to, and larger than, the glial scar. Stroke induces gliosis in a region of apoptotic cell death that closely borders the infarct, extending several hundred microns into periinfarct cortex in the rat.41,42 Within this region of glial scar, stroke induces the growth-inhibitory chondroitin sulfate proteoglycans,23,43 the semaphorin IIIa receptor neuropilin 1,44 and myelin-associated glycoprotein.45 Importantly, stroke also induces growth-promoting proteins in this region of glial scar, including GAP43, CAP23, MARCKS, and small proline repeat rich protein 1 (SPRR1)22,23,46,47 (Fig 1). Thus, the glial scar after stroke is a narrow region that closely borders the infarct and contains increased expression levels of both growth-promoting and growth-inhibitory molecules.23 In contrast, a large region of periinfarct cortex that lies just outside of the glial scar has reduced levels of growth-inhibitory molecules. After stroke, chondroitin sulfate proteoglycans, such as aggrecan, phosphacan, and versican, are actually reduced in this region of periinfarct cortex23 (see Fig 1). These chondroitin sulfate proteoglycans normally form perineuronal nets, which encase both pyramidal and nonpyramidal neurons in Fig 1. Growth-associated gene expression in periinfarct cortex after stroke. The schematic (top) shows the location of photomicrographs in A through J. (A–D) Aggrecan and versican are expressed in perineuronal nets in control cortex (A, C). Seven days after stroke (B, D), aggrecan and versican expression in perineuronal nets is markedly reduced, and these two proteins are expressed densely within the glial scar that borders the infarct (asterisks). (E–J) In situ hybridization for the growth-promoting genes GAP43, CAP23, and small proline repeat rich protein 1 (SPRR1). The bright signal corresponds to messenger RNA expression level within cells in cortex. (E, G, I) Control, nonstroke cortex. (F, H, J) Periinfarct cortex at 7 days after stroke. Note that there is intense upregulation of the expression for these three growth-promoting genes in the same region and time points as for the reduction in the growth-inhibitory aggrecan and versican proteins. Data are modified from Carmichael and colleagues.23 Scale bars ⫽ 100M. adult cortex.48 After stroke, the chondroitin sulfate proteoglycan perineuronal nests are reduced in periinfarct cortex.23,43 The reduction in a major class of growth-inhibitory proteins is time-locked to an increase in growth-promoting molecules.23 Thus, stroke induces a glial scar, in which both growth-promoting and growth-inhibitory molecules are upregulated, and a growth-permissive zone, in which growth-promoting molecules are induced and growth-inhibitory molecules are reduced. This growth-permissive zone corresponds to the region of poststroke axonal sprouting.23,24 Molecular Growth-Promoting Programs for Axonal Sprouting Poststroke axonal sprouting means that fully differentiated adult cortical neurons must engage a neuronal growth program: elaborate a growth cone, extend an axon, and form new connections. The genes involved in axonal growth have been studied in the successful neuronal regeneration that occurs in the peripheral nervous system after injury and in the initial phases of axonal sprouting during brain development.49 –52 Axonal sprouting requires a network of molecules that link membrane-signaling events to growth cone phosphorylation cascades, cytoskeletal rearrangement, and new gene transcription. These genes are induced at specific time points in the axonal sprouting process in peripheral nerve regeneration or brain development, and they allow growing axons to sense the local environment and modify their cytoskeleton for directed extension.49 –54 Axonal sprouting after stroke also progresses through specific biological time points. A trigger phase is present 1 to 3 days after stroke, in which rhythmic and synchronized neuronal discharges Carmichael: Neural Repair after Stroke 737 induce axonal sprouting.40 Seven and 14 days after the lesion are initiation and maintenance phases of the sprouting response in stroke and other CNS lesions.22,56,57 Twenty-eight days after stroke is a time period in which axonal sprouting has formed new patterns of connections that can be detected anatomically.24 These time points establish trigger, initiation, maintenance, and maturation phases in poststroke axonal sprouting. Stroke induces neuronal growth-promoting genes in sequential waves after the infarct that correspond to the biological time points in the sprouting response. These waves can be categorized as an early response, an early/ sustained response, a middle response, and a late response.23 In the early response, SPRR1 is induced 3 to 7 days after stroke. Genes for the growth cone lipid raft proteins GAP43, CAP23, and MARCKS and the transcription factor c-Jun are induced at 3 days and persist for the duration of the sprouting response. The cell adhesion molecule L1, cyclin-dependent kinase inhibitor p21/waf1, and embryonic tubulin isoform T␣1 tubulin are induced at middle phases of the sprouting response. The cytoskeletal reorganizing genes SCG10 and SCLIP are induced late, during the maturation phase of the sprouting response. Overall, this pattern of neuronal growth-promoting gene expression in cortex after stroke is unique compared with neuronal development, peripheral nervous system regeneration, and nonstroke CNS lesions. Genes such as SCG10, SCLIP, stathmin, and Rb3, which are important in growth cone motility and responsiveness to guidance cues53–55 and are induced early in neuronal growth phases in other model systems after injury,58 – 60 are induced late in the sprouting response or are repressed after stroke. L1, a member of the immunoglobulin superfamily of cell adhesion molecules that is expressed throughout periods of axonal sprouting during development and regeneration,61– 63 is only transiently expressed in middle phases of poststroke axonal sprouting. These differences between the molecular profile of axonal sprouting after stroke and the molecular profiles of axonal sprouting in the peripheral nervous system and developing nervous system have begun to identify specific molecular targets that may account for the relative failure of regeneration in the adult brain after ischemic injury. Axonal sprouting after stroke requires not only the induction of a growth-promoting program within periinfarct neurons, but also a reduction in the growthinhibitory environment. This principle has clearly emerged from studies in other forms of CNS injury. In spinal cord injury, a reduction in the myelin inhibitor Nogo alone is insufficient to promote axonal regeneration.64,65 Injury to the optic nerve induces a significant axonal sprouting only with both a reduction in Nogo levels and an induction in a growth-promoting 738 Annals of Neurology Vol 59 No 5 May 2006 program.66 In stroke, gene expression profiling shows that the induction of the unique poststroke neuronal growth-promoting program occurs during a window of time in which many growth-inhibitory molecules are not yet activated. Messenger RNA levels for the chondroitin sulfate proteoglycans aggrecan, phosphacan, and versican are induced late after stroke, that is, after the early, early/sustained, and middle phases of growthpromoting genes. This late induction of chondroitin sulfate proteoglycan genes occurs over the same interval in which chondroitin sulfate proteoglycan proteins are removed in the region of axonal sprouting in periinfarct cortex (see earlier). Thus, both the molecular and cellular data support the idea that this 2- to 3-week poststroke interval is a window for axonal sprouting: neuronal growth-promoting genes are induced, growthinhibitory proteins are removed, and many growthinhibitory genes are not yet induced. A small number of growth-inhibitory molecules are induced during this axonal sprouting window. Stroke induces a wave of expression of ephrin A5, semaphorin IIIa and its receptor neuropilin 1, and the chondroitin sulfate proteoglycan neurocan during the initial expression peaks of the neuronal growth-promoting genes.23,67 Ephrin A5 is a classic axonal growth inhibitor in the developing sensorimotor cortex.68 Neurons express neuropilin 1, and elements of the CNS scar express its ligand, semaphorin IIIa.27 With both the inhibitory ligand and the receptor induced in periinfarct cortex, the semaphorin system is in a position to mediate direct signaling between sprouting neurons and the local extracellular environment. As the periinfarct cortex begins activating sequential waves of growth-promoting genes during the initial 2- to 3-week window after stroke, ephrin A5, neurocan, and semaphorin IIIa/neuropilin 1 are induced and in a position to meet growing axons. Poststroke Neurogenesis Stroke signals to the sites of neural stem cells within the brain, to induce the migration of newly born immature neurons (neuroblasts) into areas of ischemic damage.69,70 In the mouse, thousands of neuroblasts migrate long distances from their origin near the ventricles to periinfarct cortex.71 Labeling a cohort of these cells with the thymidine analog bromodeoxyuridine allows a “tagging experiment”—neuroblasts that are born at the same time can be followed in sequential periods after the stroke. Tagged neuroblasts migrate from the subventricular zone (SVZ) in waves that extend over several days. They occupy positions around the infarct, but the successive waves do not increase the total number of neuroblasts that are present in this region (Fig 2). This appears to be because some of the newly born neuroblasts die within days of their arrival.71,72 Thus, poststroke neurogenesis is a dynamic process, with Fig 2. Erythropoietin (EPO) system in poststroke neurogenesis. (A) Schematic diagram of coronal section through mouse brain after stroke. Boxes indicate position of photomicrographs of corresponding panels. (B) Stereological quantification of doublecortin (DCX)positive cells in cortex after stroke. The interior columns within the day 5 and 7 bars indicate the percentage of doublecortin cells that are labeled by bromodeoxyuridine (BrdU) administration on days 1 to 3 after stroke. These doublecortin/BrdU double-labeled cells represent one cohort of neuroblasts that are all born on days 1 to 3 after stroke. These double-labeled cells progressively accumulate in control animals after stroke. In animals with a knock-down in the erythropoietin receptor (EPOR), this cohort of neuroblast initially migrates to periinfarct cortex normally (yellow bar in day 5), but then fails to migrate to cortex by day 7 compared with control (yellow bar vs red bar in day 7). #p ⬍ 0.0001 versus day 3 and control Mut/Con; ˆp ⬍ 0.02 versus Mut day 7; *p ⬍ 0.0002 versus Mut day 7; @p ⬍ 0.02 versus day 3 Mut/Con and nonstroke. Modified from Tsai and colleagues.71 (C) Doublecortin (green) and EPOR (red) staining in subventricular zone (SVZ) after 7 days after stroke. Doublecortin staining recognized migrating neuroblasts after stroke.71 Colocalization of both signals is seen in overlay images in yellow. (D–F) DCX and EPOR staining in subcortical white matter 7 days after stroke. Note that all commercially available EPOR antibodies stain EPOR, as well as other antigens80; thus, this staining must be interpreted as showing the presence of EPOR and possibly other antigens. (G, H) EPO expression in control (G) and 7 days after stroke (H). Note that EPO is induced in white matter (wm) after stroke in close proximity to migrating neuroblasts. (I) Summary of axonal sprouting and neuroblast migration after stroke. Bar ⫽ 100M. Carmichael: Neural Repair after Stroke 739 waves of migration of newly born neuroblasts, and waves of cell death in these recently arrived neuroblasts. Stroke must initiate molecular signals that recruit migrating neuroblasts to areas of injury. One of the signals that promotes neuroblast migration after stroke is the cytokine erythropoietin (EPO). EPO is induced near the infarct after stroke73,74 (see Fig 2) and is under the control of the hypoxia-inducible factor 1 transcription factor, an early molecular signal after stroke.73,74 In vitro evidence suggests that EPO promotes neuronal differentiation of neural stem cells.75,76 Pharmacological doses of EPO promote cell division and increased neuronal progenitor cells in the SVZ after stroke.77 Using a conditional knock-out of the EPO receptor (EPOR), researchers recently showed that EPO functions in the normal brain to promote neural stem cell proliferation in the SVZ and to promote neuroblast migration to periinfarct cortex after stroke.71 In these studies, tagging experiments of migrating neuroblasts (noted earlier) were combined with knock-down of the EPOR in neurons and astrocytes, to show that EPO promotes the continued migration of neuroblasts to areas of injury after stroke. When EPOR is knocked down, only the initial wave of neuroblast migration normally takes place. Subsequent waves of migration are significantly smaller (see Fig 2). A conceptual model utilizing these data and previous studies suggests that stroke activates hypoxia-inducible factor 1 in periinfarct cortex,78 which induces endogenous EPO production around the infarct. Neuroblasts respond to this ligand with their EPOR, and this interaction promotes neuroblast migration to periinfarct cortex. EPO applied pharmacologically also induces angiogenesis near the infarct, with subsequent vascular production of growth factors that stimulate neurogenesis.77 There are many gaps in this framework of poststroke neurogenesis, such as the identity of the signals that initiate poststroke neurogenesis and the cellular microenvironment near the stroke that fosters neuroblast survival or death. Conclusions Functional recovery after stroke in humans correlates with remapping of cognitive operations in periinfarct cortex and connected cortical areas. Within these regions, axons sprout new connections and newly born neuroblasts migrate into areas of damage. Molecular and cellular studies of poststroke axonal sprouting and neurogenesis identify unique regeneration programs within periinfarct cortex. These programs occur as waves of growth-promoting genes and migrating neuroblasts in unique tissue microenvironments of reduced inhibitory molecules and induced promigratory cytokines. In one of the first California metaphors of surfing as Life, Eugene Burdick79 describes a group of surfers 740 Annals of Neurology Vol 59 No 5 May 2006 who waited only to catch the largest wave of the day, the ninth wave of the ninth set. Smaller waves were mere distractions. On the beach and later in life, riding the ninth wave made the effort complete.79 There are many gaps in the data on regeneration after stroke, such as the exact relationship of axonal sprouting and neurogenesis to behavioral recovery and the identification of the key control points or master switches that influence these processes. However, with continued study these gaps will be filled, and the function of the genes that ultimately complete neuronal regeneration after stroke will be determined. We will in short order identify the ninth wave of neural repair. However, the endogenous mechanisms of neural repair lead to only a limited functional recovery. To truly improve recovery after stroke, we will have to apply the molecular principles of neural repair toward novel regenerative therapies after stroke. We will have to make waves of our own. This work was supported by the NIH (National Institute of Neurological Disorders and Stroke, 1R01 NS45729), American Heart Association (0555013Y), and a Distinguished Scholar Award from the Larry L. Hillblom Foundation. References 1. American Heart Association. Heart Disease and Stroke Statistics Update 2006. Available at: http://www.americanheart.org/ presenter.jhtml?identifier⫽1200026. Accessed Feb 17, 2006. 2. Dobkin BH. The clinical science of neurorehabilitation. Oxford, United Kingdom: Oxford University Press, 2003. 3. Calautti C, Leroy F, Guincestre J, et al. Sequential activation brain mapping after subcortical stroke: changes in hemispheric balance and recovery. Neuroreport 2001;12:3883–3886. 4. Tombari D, Loubinoux I, Pariente J, et al. A longitudinal fMRI study: in recovering and then in clinically stable sub-cortical stroke patients. Neuroimage 2004;23:827– 839. 5. Ward NS. Functional reorganization of the cerebral motor system after stroke. Curr Opin Neurol 2004;17:725–730. 6. Cramer SC. Functional imaging in stroke recovery. Stroke 2004;35(suppl 1):2695–2698. 7. Binkofski F, Seitz RJ. Modulation of the BOLD-response in early recovery from sensorimotor stroke. Neurology 2004;63: 1223–1229. 8. Cramer SC, Crafton KR. Somatotopy and movement representation sites following cortical stroke. Exp Brain Res 2006;168: 25–32. 9. Butefisch CM, Kleiser R, Korber B, et al. Recruitment of contralesional motor cortex in stroke patients with recovery of hand function. Neurology 2005;64:1067–1069. 10. Corbetta M, Kincade MJ, Lewis C, et al. Neural basis and recovery of spatial attention deficits in spatial neglect. Nat Neurosci 2005;8:1603–1610. 11. Karbe H, Thiel A, Weber-Luxenburger G, et al. Brain plasticity in poststroke aphasia: what is the contribution of the right hemisphere? Brain Lang 1998;64:215–230. 12. Fernandez B, Cardebat D, Demonet JF, et al. Functional MRI follow-up study of language processes in healthy subjects and during recovery in a case of aphasia. Stroke 2004;35: 2171–2176. 13. Traversa R, Cicinelli P, Bassi A, et al. Mapping of motor cortical reorganization after stroke. A brain stimulation study with focal magnetic pulses. Stroke 199728:110 –117. 14. Thickbroom GW, Byrnes ML, Archer SA et al. Motor outcome after subcortical stroke correlates with the degree of cortical reorganization. Clin Neurophysiol 2004;115:2144 –2150. 15. Werhahn KJ, Conforto AB, Kadom N, et al. Contribution of the ipsilateral motor cortex to recovery after chronic stroke. Ann Neurol 2003;54:464 – 472. 16. Fridman EA, Hanakawa T, Chung M, et al. Reorganization of the human ipsilesional premotor cortex after stroke. Brain 2004;127:747–758. 17. Florence SL, Taub HB, Kaas JH. Large-scale sprouting of cortical connections after peripheral injury in adult macaque monkeys. Science 1998;282:1117–1121. 18. Darian-Smith C, Gilbert CD. Axonal sprouting accompanies functional reorganization in adult cat striate cortex. Nature 1994;368:737–740. 19. Keller A, Arissian K, Asanuma H. Formation of new synapses in the cat motor cortex following lesions of the deep cerebellar nuclei. Exp Brain Res 1990;80:23–33. 20. Benowitz LI, Routtenberg A. GAP-43: an intrinsic determinant of neuronal development and plasticity. Trends Neurosci 1997; 20:84 –91. 21. Ng SC, de la Monte SM, Conboy GL, et al. Cloning of human GAP-43: growth association and ischemic resurgence. Neuron 1988;1:133–139. 22. Stroemer RP, Kent TA, Hulsebosch CE. Neocortical neural sprouting, synaptogenesis, and behavioral recovery after neocortical infarction in rats. Stroke 1995;26:2135–2144. 23. Carmichael ST, Archibeque I, Luke L, et al. Growth-associated gene expression after stroke: evidence for a growth-promoting region in peri-infarct cortex. Exp Neurol 2005;193:291–311. 24. Carmichael ST, Wei L, Rovainen CM, Woolsey TA. New patterns of intracortical projections after focal cortical stroke. Neurobiol Dis 2001;8:910 –922. 25. Dancause N, Barbay S, Frost SB, et al. Extensive cortical rewiring after brain injury. J Neurosci 2005;25:10167–10179. 26. Silver J, Miller JH. Regeneration beyond the glial scar. Nat Rev Neurosci 2004;5:146 –156. 27. De Winter F, Oudega M, Lankhorst AJ, et al. Injury-induced class 3 semaphorin expression in the rat spinal cord. Exp Neurol 2002;175:61–75. 28. Bundesen LQ, Scheel TA, Bregman BS, Kromer LF. Ephrin-B2 and EphB2 regulation of astrocyte-meningeal fibroblast interactions in response to spinal cord lesions in adult rats. J Neurosci 2003;23:7789 –7800. 29. Hunt D, Coffin RS, Prinjha RK, et al. Nogo-A expression in the intact and injured nervous system. Mol Cell Neurosci 2003; 24:1083–1102. 30. Jones LL, Margolis RU, Tuszynski MH. The chondroitin sulfate proteoglycans neurocan, brevican, phosphacan, and versican are differentially regulated following spinal cord injury. Exp Neurol 2003;182:399 – 411. 31. Benson MD, Romero MI, Lush ME, et al. Ephrin-B3 is a myelin-based inhibitor of neurite outgrowth. Proc Natl Acad Sci U S A 2005;102:10694 –10699. 32. Morgenstern DA, Asher RA, Fawcett JW. Chondroitin sulphate proteoglycans in the CNS injury response. Prog Brain Res 2002;137:313–332. 33. Arvin B, Neville LF, Barone FC, Feuerstein GZ. The role of inflammation and cytokines in brain injury. Neurosci Biobehav Rev 1996;20:445– 452. 34. Lipton P. Ischemic cell death in brain neurons. Physiol Rev 1999;79:1431–1568. 35. Raghupathi R, Graham DI, McIntosh TK. Apoptosis after traumatic brain injury. J Neurotrauma 2000;17:927–938. 36. Lewen A, Sugawara T, Gasche Y, et al. Oxidative cellular damage and the reduction of APE/Ref-1 expression after experimental traumatic brain injury. Neurobiol Dis 2001;8:380 –390. 37. Napieralski JA, Butler AK, Chesselet MF. Anatomical and functional evidence for lesion-specific sprouting of corticostriatal input in the adult rat. J Comp Neurol 1996;373:484 – 497. 38. Bush TG, Puvanachandra N, Horner CH, et al. Leukocyte infiltration, neuronal degeneration, and neurite outgrowth after ablation of scar-forming, reactive astrocytes in adult transgenic mice. Neuron 1999;23:297–308. 39. Kartje GL, Schulz MK, Lopez-Yunez A, et al. Corticostriatal plasticity is restricted by myelin-associated neurite growth inhibitors in the adult rat. Ann Neurol 1999;45:778 –786. 40. Carmichael ST, Chesselet MF. Synchronous neuronal activity is a signal for axonal sprouting after cortical lesions in the adult. J Neurosci 2002;22:6062– 6070. 41. Katsman D, Zheng J, Spinelli K, Carmichael ST. Tissue microenvironments within functional cortical subdivisions adjacent to focal stroke. J Cereb Blood Flow Metab 2003;23:997–1009. 42. Yasuda Y, Tateishi N, Shimoda T, et al. Relationship between S100beta and GFAP expression in astrocytes during infarction and glial scar formation after mild transient ischemia. Brain Res 2004;1021:20 –31. 43. Hobohm C, Gunther A, Grosche J, et al. Decomposition and long-lasting downregulation of extracellular matrix in perineuronal nets induced by focal cerebral ischemia in rats. J Neurosci Res 2005;80:539 –548. 44. Beck H, Acker T, Puschel AW, et al. Cell type-specific expression of neuropilins in an MCA-occlusion model in mice suggests a potential role in post-ischemic brain remodeling. J Neuropathol Exp Neurol 2002;61:339 –350. 45. Ishiguro H, Inuzuka T, Fujita N, et al. Expression of the large myelin-associated glycoprotein isoform in rat oligodendrocytes around cerebral infarcts. Mol Chem Neuropathol 1993;20: 173–179. 46. Li Y, Jiang N, Powers C, Chopp M. Neuronal damage and plasticity identified by microtubule-associated protein 2, growth-associated protein 43, and cyclin D1 immunoreactivity after focal cerebral ischemia in rats. Stroke 1998;29:1972–1980. 47. Lu A, Tang Y, Ran R, et al. Genomics of the periinfarction cortex after focal cerebral ischemia. J Cereb Blood Flow Metab 2003;23:786 – 810. 48. Wegner F, Hartig W, Bringmann A, et al. Diffuse perineuronal nets and modified pyramidal cells immunoreactive for glutamate and the GABA(A) receptor alpha1 subunit form a unique entity in rat cerebral cortex. Exp Neurol 2003;184:705–714. 49. Mody M, Cao Y, Cui Z, et al. Genome-wide gene expression profiles of the developing mouse hippocampus. Proc Natl Acad Sci U S A 2001;98:8862– 8867. 50. Bonilla IE, Tanabe K, Strittmatter SM. Small proline-rich repeat protein 1A is expressed by axotomized neurons and promotes axonal outgrowth. J Neurosci 2002;22:1303–1315. 51. Tanabe K, Bonilla I, Winkles JA, Strittmatter SM. Fibroblast growth factor-inducible-14 is induced in axotomized neurons and promotes neurite outgrowth. J Neurosci 2003;23: 9675–9686. 52. Stead JD, Neal C, Meng F, et al. Transcriptional profiling of the developing rat brain reveals that the most dramatic regional differentiation in gene expression occurs postpartum. J Neurosci 2006;26:345–353. 53. Mori N, Morii H. SCG10-related neuronal growth-associated proteins in neural development, plasticity, degeneration, and aging. J Neurosci Res 2002;70:264 –273. 54. Suh LH, Oster SF, Soehrman SS, et al. L1/Laminin modulation of growth cone response to EphB triggers growth pauses and regulates the microtubule destabilizing protein SCG10. J Neurosci 2004;24:1976 –1986. 55. Golub T, Pico C. Spatial control of actin-based motility through plasmalemmal PtdIns(4,5)P2-rich raft assemblies. Biochem Soc Symp 2005;72:119 –127. Carmichael: Neural Repair after Stroke 741 56. Steward O. The process of reinnervation in the dentate gyrus of adult rats: gene expression by neurons during the period of lesion-induced growth. J Comp Neurol 1995;359:391– 411. 57. Leon S, Yin Y, Nguyen J, et al. Lens injury stimulates axon regeneration in the mature rat optic nerve. J Neurosci 2000;20: 4615– 4626. 58. Iwata T, Namikawa K, Honma M, et al. Increased expression of mRNAs for microtubule disassembly molecules during nerve regeneration. Brain Res Mol Brain Res 2002;102:105–109. 59. Mason MR, Lieberman AR, Anderson PN. Corticospinal neurons up-regulate a range of growth-associated genes following intracortical, but not spinal, axotomy. Eur J Neurosci 2003;18: 789 – 802. 60. Voria I, Hauser J, Axis A, et al. Improved sciatic nerve regeneration by local thyroid hormone treatment in adult rat is accompanied by increased expression of SCG10. Exp Neurol 2006;197:258 –267. 61. Zhang Y, Roslan R, Lang D, et al. Expression of CHL1 and L1 by neurons and glia following sciatic nerve and dorsal root injury. Mol Cell Neurosci 2000;16:71– 86. 62. Mintz CD, Dickson TC, Gripp ML, et al. ERMs colocalize transiently with L1 during neocortical axon outgrowth. J Comp Neurol 2003;464:438 – 448. 63. Becker CG, Becker T, Meyer RL. Increased NCAM-180 immunoreactivity and maintenance of L1 immunoreactivity in injured optic fibers of adult mice. Exp Neurol 2001;169: 438 – 448. 64. Kim JE, Li S, GrandPre T, et al. Axon regeneration in young adult mice lacking Nogo-A/B. Neuron 2003;38:187–199. 65. Zheng B, Ho C, Li S, et al. Lack of enhanced spinal regeneration in Nogo-deficient mice. Neuron 2003;38:213–224. 66. Fischer D, He Z, Benowitz LI. Counteracting the Nogo receptor enhances optic nerve regeneration if retinal ganglion cells are in an active growth state. J Neurosci 2004;24:1646 –1651. 67. Deguchi K, Takaishi M, Hayashi T, et al. Expression of neurocan after transient middle cerebral artery occlusion in adult rat brain. Brain Res 2005;1037:194 –199. 742 Annals of Neurology Vol 59 No 5 May 2006 68. Bolz J, Uziel D, Muhlfriedel S, et al. Multiple roles of ephrins during the formation of thalamocortical projections: maps and more. J Neurobiol 2004;59:82–94. 69. Zhang RL, Zhang ZG, Chopp M. Neurogenesis in the adult ischemic brain: generation, migration, survival, and restorative therapy. Neuroscientist 2005;11:408 – 416. 70. Lichtenwalner RJ, Parent JM. Adult neurogenesis and the ischemic forebrain. J Cereb Blood Flow Metab 2006;26:1–20. 71. Tsai PT, Ohab JJ, Kertesz N, et al. A critical role of erythropoietin receptor in neurogenesis and post-stroke recovery. J Neurosci 2006;26:1269 –1274. 72. Nadareishvili Z, Hallenbeck J. Neuronal regeneration after stroke. N Engl J Med 2003;348:2355–2356. 73. Dirnagl U, Simon RP, Hallenbeck JM. Ischemic tolerance and endogenous neuroprotection. Trends Neurosci 2003;26: 248 –254. 74. Bernaudin M, Marti HH, Roussel S, et al. A potential role for erythropoietin in focal permanent cerebral ischemia in mice. J Cereb Blood Flow Metab 1999;19:643– 651. 75. Studer L, Csete M, Lee SH, et al. Enhanced proliferation, survival, and dopaminergic differentiation of CNS precursors in lowered oxygen. J Neurosci 2000;20:7377–7383. 76. Shingo T, Sorokan ST, Shimazaki T, Weiss S. Erythropoietin regulates the in vitro and in vivo production of neuronal progenitors by mammalian forebrain neural stem cells. J Neurosci 2001;21:9733–9743. 77. Wang L, Zhang Z, Wang Y, et al. Treatment of stroke with erythropoietin enhances neurogenesis and angiogenesis and improves neurological function in rats. Stroke 2004;35: 1732–1737. 78. Sharp FR, Lu A, Tang Y, Millhorn DE. Multiple molecular penumbras after focal cerebral ischemia. J Cereb Blood Flow Metab 2000;20:1011–1032. 79. Burdick E. The ninth wave. New York: Dell Publishing, 1956. 80. Elliott S, Busse L, Bass MB, et al. Anti-Epo receptor antibodies do not predict Epo receptor expression. Blood 2006;107: 1892–1895.