Regenerating the Brain

David A. Greenberg; Kunlin Jin    Buck Institute for Age Research, Novato, California 94945, USA

Recent discoveries related to adult neurogenesis suggest that the long‐sought goal of regenerating injured adult brain tissue may be achievable in principle. A variety of human neurological diseases and rodent models of these diseases— including stroke, Alzheimer’s disease, Huntington’s disease, Parkinson’s disease, and amyotrophic lateral sclerosis—are associated with an increase in the brain’s neuroproliferative capacity. In some cases, this capacity can be enhanced further by growth factors or drugs. Therefore, therapeutic manipulation of endogenous neurogenesis may be feasible. In addition to its potential clinical utility, the study of injury‐induced adult neurogenesis is helping to reveal mechanisms of neuronal proliferation, migration, and differentiation that may also operate during normal development.

I Introduction

The regeneration of human tissue has been a theme since antiquity. Osiris, the Egyptian god of earth and vegetation, was murdered by his brother Seth, and Osiris’ body was thrown into the Nile. When Osiris’ sister Isis recovered the body, Seth chopped it into 14 parts, which he scattered. However, Isis searched for and found these parts, which she reassembled to resurrect Osiris. In Greek mythology, the titan Prometheus stole fire from the gods on Mount Olympus and gave it to man. For this and other offenses, Zeus ordered that Prometheus be chained to a rock, where the eagle Ethon would eat out his liver; the liver grew back each day, only to be eaten again.

Tissue regeneration is a well‐known phenomenon to the zoologist—as exemplified by the ability of creatures like sponges, hydra, flatworms, annelids, sea stars, newts, and salamanders to regrow body parts. It is also recognized in clinical medicine, where it occurs in the setting of wound healing. In some organ systems of vertebrates, however, the conventional wisdom has been that little or no regeneration can take place, and this is perhaps nowhere more commonly held to be the case than in the central nervous system (CNS).

In the 1960s and 1970s, Joseph Altman and colleagues at Purdue University mapped the development of the rodent brain, using [3H]thymidine to birthdate neurons. They found that in some brain regions, the birth of new neurons could be observed well into adulthood. However, this insight seems to have receded from attention until relatively recently, when it provided the basis for a new field of investigation—that of adult neurogenesis.

What makes adult neurogenesis more than just a biological oddity akin to the regrowth of a sea star’s arms is the possibility it suggests for therapeutic application. The CNS is notoriously difficult to repair, and injuries or diseases that affect it often dramatically change a patient’s quality of life. One has only to see a quadriplegic victim of spinal cord injury, an individual with profound amnesia due to alcoholic Korsakoff’s syndrome, or a survivor of stroke who can no longer comprehend or produce language, to be impressed with the extent to which brain damage can depreciate life.

A substantial body of evidence indicates that cerebral injury of several types can stimulate neurogenesis in the adult brain. The most compelling outstanding issue regarding the potential clinical importance of this phenomenon is to determine if it can generate functional neurons that contribute to enhanced recovery.

II Neurodegenerative Diseases: Therapeutic Targets for Neurogenesis

Acute and chronic neurodegenerative diseases are common, disabling, and poorly responsive to current treatment. Stroke, the most frequent cause of acute neurodegeneration, has a prevalence of ∼4.8 million and an incidence of ∼700,000 individuals per year in the United States, where it is the third leading cause of death (American Heart Association, 2004). Even among those who survive stroke, disability due to hemiparesis, gait disorders, aphasia, and other syndromes is common, and ∼20% of these patients require institutional care at 6‐months poststroke. This long‐term disability contributes to the average lifetime cost for stroke care of ∼$140,000 and an annual national cost of ∼$54 billion. The only major recent advance in treatment, the use of thrombolytic agents to dissolve clots in the acute aftermath of stroke, has had limited impact because it appears to be effective only within about the first 3 hours after onset of symptoms (Brott and Bogousslavsky, 2000).

Chronic neurodegenerations include Alzheimer’s disease (AD), Parkinson’s disease (PD), hereditary polyglutamine disorders like Huntington’s disease (HD), and amyotrophic lateral sclerosis (ALS). These diseases affect different (albeit overlapping) regions of the CNS and vary in prevalence, from ∼4.5 million cases in AD to ∼1.5 million cases in PD, and ∼30,000 cases in HD and ALS in the United States. However, all typically culminate in an extended period of functional disability preceding death. Except for PD, in which drugs and surgery are available to reduce symptoms, at least temporarily, even symptomatic treatment for chronic neurodegenerations is extremely limited at present. In AD, acetylcholinesterase inhibitors and the N‐methyl‐D‐aspartate (NMDA)‐type glutamate receptor antagonist memantine exert modest behavioral effects in some patients (Cummings, 2004). In ALS, another NMDA antagonist, riluzole, has provided minimal benefit (Rowland and Shneider, 2001). Perhaps most notably, no treatment exists for any of these diseases that can restore lost function.

Impaired brain function in acute and chronic neurodegenerative diseases results primarily from cell (especially neuronal) loss. One reason for the limited responsiveness of neurodegenerative diseases to treatment may be that it is more difficult to overcome the loss of cells than the impairment of selected cellular functions. As an example, among neurological disorders, the greatest therapeutic successes have come in conditions where cell loss is not a major feature, such as epilepsy and migraine. Even in PD, where cell loss is relatively circumscribed, pharmacological restoration of a key cellular function like dopaminergic neurotransmission, without the temporal, spatial, and stimulus‐coupled regulation that a cellular context provides, has been an imperfect stratagem.

Based on this experience, it is reasonable to conclude that cell‐replacement therapy, technically challenging though it may be, is worth pursuing. In addition to the prospect of more completely restoring brain function, cell‐replacement therapy has the further advantage that it might be effective at later stages of a disease. This is an important consideration not only in disorders like stroke, which often evolve too quickly for acute treatment to be instituted, but also in chronic neurodegenerations, where cell loss is already extensive before the onset of symptoms.

III Building Brains: Evolution and Development

Evidence for the feasibility of cell replacement in the brain and principles to guide cell‐replacement research come from several sources, including evolution and development. The challenge of cell replacement for neurodegenerative diseases is, in simple terms, to (re)build the brain. This is a task that is faced in one form or another: (1) in evolution, as brain size increases and (2) in ontogeny, as the brain develops from the neural tube.

As larger brains evolved, they appear to have done so primarily through an increase in neuron number, rather than, for example, neuron size or proportional connectivity (Streidter, 2005). This suggests that supplying new cells might also be the principal requirement for brain rebuilding. The evolutionary principle of epigenetic population matching suggests that trophic influences of surviving brain cells may help direct new neurons to reestablish appropriate connections. A related concept, the parcellation hypothesis, predicts a mechanism for pruning of exuberant axonal connections to help restore normal patterns of circuitry. Finally, the phenomenon of connectional invasion presages a capacity for restoring connections over an altered neuronal landscape and, perhaps, forming...