Soluble adenylyl cyclase is required for netrin-1 signaling in nerve growth cones.

Axonal outgrowth is essential for the proper development of the nervous system as well as for axonal regeneration following injury. Axonal growth cones detect outgrowth factors in their environment and transduce these signals into rearrangements in the axonal cytoskeleton, ultimately resulting in axonal elongation1. Several families of extracellular molecules have been discovered that promote axon outgrowth, such as netrins2,3 and neurotrophins4. Members of the netrin family can trigger axonal outgrowth2 and can also elicit either attractive or repulsive turning responses that are fundamental for axonal pathfinding5-7. Netrin-1, the founding member of the netrin family2, has outgrowth and pathfinding roles in the development of many types of neurons, including cortical8, spinal commissural3,9 and peripheral neurons10. Netrin-1 binds to the deleted in colorectal cancer (DCC) family of receptors, which include neogenin and DCC in vertebrates11, resulting in the reorganization of the growth cone cytoskeleton and subsequent axonal elongation12. This process involves the regulation of the activity of Rho family monomeric GTPases, including Cdc42 and Rac1 (ref. 13). Bath application of netrin-1 induces ‘growth cone elaboration’, characterized by significant increases in growth cone surface area and in the number of filopodia12,13—the F-actin–rich, finger-like protrusions at the leading edge of the growth cone. cAMP has critical roles in mediating the responses of axons to netrin-1 and also has direct effects on axonal elongation. Application of netrin-1 increases cAMP levels in growth cones6, and blockade of cAMP signaling by inhibition of protein kinase A (PKA) or by growing axons on laminin-1, a substratum that reduces basal cAMP levels, blocks netrin-1–mediated attractive responses6,14. cAMP seems to signal outgrowth and attractive turning, as Xenopus laevis embryonic spinal axons turn and extend toward gradients of forskolin or dibutyryl cAMP (ref. 15), pharmacological agents that activate cAMP signaling pathways, and elevation of intracellular cAMP promotes outgrowth of injured spinal cord axons in the rat16. Although cAMP is a key mediator of netrin-1 signaling, the mechanisms by which the netrin-1 activation of DCC leads to increased cAMP levels in growth cones are still unclear. DCC does not seem to be linked to heterotrimeric G protein–responsive transmembrane adenylyl cyclases (tmACs), and attempts to link netrin-1 to G protein–coupled receptors (GPCRs) remain controversial. Additionally, although several proteins interact with DCC (ref. 13), none of these are known to be coupled to tmAC activation. Mammalian cells possess a second source of cAMP, the evolutionarily conserved, bicarbonate- and calcium-responsive sAC. sAC was originally identified as a soluble activity in the testis that was not activated by forskolin17. Molecular cloning of the enzyme revealed that sAC does not contain transmembrane domains, is evolutionarily more related to bacterial cyclases than to tmAC (ref. 18), and is ubiquitously expressed19 with several alternatively spliced isoforms that exhibit tissue-specific expression patterns20. Isoforms enriched in the testis are involved in sperm motility21,22 and maturation22. In the testis, sAC activity is predominantly cytosolic, but in other cell types, sAC protein is particulate, with isoforms distributed to the nucleus, mitochondria and cytoskeletal structures23. sAC is not regulated by G protein pathways17,24; instead, it is modulated by calcium25,26. Here we report that sAC is expressed in the axons and growth cones of developing neurons. The effects of sAC overexpression—axonal outgrowth and elaboration of growth cones—resemble morphological changes elicited by the treatment of axons with netrin-1. Using pharmacological and siRNA approaches, we found that sAC activity is required for netrin-1–induced cAMP generation and for netrin-1–mediated growth cone elaboration and axon outgrowth. Blockade of G protein–responsive tmACs had no effect on netrin-1 signaling, indicating that GPCR activation is not involved in these effects of netrin-1. Our data reveal a new role for sAC as a downstream effector of netrin-1 and demonstrate that sAC-mediated cAMP production accounts for the morphologic effects of netrin-1 on growth cones.