This Week in The Journal

Cellular/Molecular

Amyloid Precursor Protein Fragments Are Trafficked Separately

Virgil Muresan, Nicholas H. Varvel, Bruce T. Lamb, and Zoia Muresan

Amyloid β is notorious for its hypothesized role in Alzheimer's disease, but the normal function of its parent protein—amyloid precursor protein (APP)—is unknown. APP contains multiple functional domains that, together with effects of APP knock-out, suggest roles in vesicle transport, transcriptional regulation, neurogenesis, neurite outgrowth, and synaptogenesis. But each APP molecule is cleaved by two secretases to create three fragments, and whether the fragments, the full-length protein, or both have important roles is not known. To gain insight into this question, Muresan et al. labeled neurons with antibodies that recognized N-terminal, central (amyloid β), or C-terminal domains of APP. Labeling in neurites was largely nonoverlapping, suggesting the fragments are trafficked along distinct pathways and the full-length molecule is rarely transported out of the soma. Interestingly, phosphorylated C-terminal fragments were the only fragments present in the growth cone periphery, and they were concentrated along the protruding edge in turning growth cones.

Phosphorylated C-terminal fragments of APP (red), localize to the peripheral regions of growth cones, whereas N-terminal fragments (green) are confined to the central portion. See the article by Muresan et al. for details.

Development/Plasticity/Repair

BDNF Is Required for Taste Bud Innervation

Liqun Ma, Grace F. Lopez, and Robin F. Krimm

(see pages 3354–3364)

Gustatory neurons innervate fungiform papillae on the tongue. These taste regions express brain-derived neurotrophic factor (BDNF) during development, suggesting that the precise growth of gustatory axons to their targets depends on BDNF. In support of this hypothesis, Ma et al. report that taste buds were not innervated in BDNF-null mice. Instead, branching of the gustatory nerve increased near the surface of the tongue and axon terminals appeared randomly distributed. After a few days, the number of axon terminals in mutant mice was reduced, but remaining terminals innervated papillae. This suggests that excessive gustatory nerve branching resulted in chance contacts between gustatory nerves and papillae, and that these contacts were maintained while unconnected terminals were pruned. Together with previous studies in which overexpression of BDNF caused inappropriate innervation of the tongue, the results reported here suggest that BDNF is necessary and sufficient for the final steps in guiding gustatory axons to their targets.

Behavioral/Systems/Cognitive

Exogenous TRPM8 Expression Allows Selective Neuronal Activation

Nathan C. Peabody, Jascha B. Pohl, Fengqiu Diao, Andrew P. Vreede, David J. Sandstrom, Howard Wang, Paul K. Zelensky, and Benjamin H. White

(see pages 3343–3353)

A new genetic technique that allows selective activation of specific neurons has been developed by Peabody et al. to investigate the neuronal control of wing expansion in Drosophila. The authors expressed TRPM8, a cold-sensitive nonselective cation channel, specifically in neurons that endogenously express crustacean cardioactive peptide (CCAP), which are thought to regulate wing expansion. Subsequent cooling shortened the time spent perching by flies that had recently emerged from the pupal case, and it accelerated the start of abdominal-flexing and air-swallowing behaviors that pump hemolymph into the unfolding wings. This effect was most dramatic in flies that were confined to a small space. Such confinement increased by approximately 20-fold the amount of time normal flies walked around before perching. But in flies expressing TRPM8, this perch-selection phase was reduced to the duration seen in unconfined wild-type flies. These results establish CCAP-expressing neurons as key control points in the production of wing-expansion behaviors.

Neurobiology of Disease

Synaptic Degeneration Occurs Early in EAE

Diego Centonze, Luca Muzio, Silvia Rossi, Francesca Cavasinni, Valentina De Chiara, Alessandra Bergami, Alessandra Musella, Marcello D'Amelio, Virve Cavallucci, Alessandro Martorana, Andrea Bergamaschi, Maria Teresa Cencioni, Adamo Diamantini, Erica Butti, Giancarlo Comi, Giorgio Bernardi, Francesco Cecconi, Luca Battistini, Roberto Furlan, and Gianvito Martino

(see pages 3442–3452)

Multiple sclerosis results from chronic inflammation of the CNS that causes demyelination and axon loss. A similar pathology can be produced in mice by injecting myelin protein fragments into the blood, which induces experimental autoimmune encephalomyelitis. Using this model system, Centonze et al. found that synaptic loss in the striatum is an early event in disease progression. In the presymptomatic stage of the disease, before demyelination or motor deficits were detected, AMPA receptor expression and phosphorylation were increased in striatal neurons, which prolonged the decay phase of spontaneous EPSCs and increased their frequency. This was accompanied by an increase in markers for synaptic degeneration and spine loss. These changes were likely induced by release of tumor necrosis factor by activated microglia and subsequent downregulation of the immediate early gene Arc/Arg3.1. Blocking AMPA receptors reduced clinical signs and increased spine density, and therefore may be an effective treatment in multiple sclerosis.