The therapeutic potential of cAMP was recently assessed in three independent studies of spinal cord injury, each using different injury models and methods of increasing cAMP. In two of these studies, cAMP elevation was accomplished by inhibiting PDE4 activity with rolipram. PDE4 is the major source of phosphodiesterase activity in the CNS (Iona et al., 1998
), making it a logical target for therapeutic intervention, and rolipram is a specific PDE4 inhibitor (Krause and Kuhne, 1988
). It also has the added advantage of being able to cross the blood-brain barrier, which makes subcutaneous and oral administration possible (Krause and Kuhne, 1988
). Rolipram was initially developed as an antidepressant (Horowski and Sastre y Hernandez, 1985
), but has now been shown to enhance neurite outgrowth and axonal regeneration in the presence of myelin inhibitors.
In the first of these studies, rolipram was administered through priming or subcutaneous delivery and neurite outgrowth on MAG-expressing CHO cells or CNS myelin was significantly increased (Nikulina et al., 2004
). To test the efficacy of rolipram in vivo
, spinal cord hemisections were performed in adult rats and this was followed by transplantation of embryonic spinal cord tissue into the lesion site and subcutaneous delivery of rolipram. Axonal regeneration was assessed 6−8 weeks after injury and few axons were observed within the transplant for vehicle treated animals. In animals that received rolipram, however, there was significantly more growth of serotonergic fibers into the transplant (; Nikulina et al., 2004
). Rolipram-treated animals also had greater functional recovery as measured by forelimb paw placement, which suggests that the regenerating axons may have formed synaptic connections (; Nikulina et al., 2004
Figure 2 Rolipram promotes growth of serotonergic axons, reduces astrogliosis, and improves functional recovery after spinal cord injury. Few serotonergic fibers (arrowheads) are present within the embryonic spinal cord tissue grafts of vehicle-treated animals (more ...)
Somewhat unexpectedly, decreased expression of glial fibrillary acidic protein (GFAP) was also observed in animals that received rolipram, and this is indicative of reduced glial scarring (; Nikulina et al., 2004
). The glial scar is another major contributor to regenerative failure, forming both a physical and biochemical barrier to regenerating axons. It is the product of injury-induced astrogliosis, which is characterized by astrocyte proliferation and elevated expression of GFAP and extracellular matrix molecules such as chondroitin sulfate proteoglycans (CSPGs), keratin sulfate proteoglycan, and cytotactin/tenascin (McKeon et al., 1991
; Jones et al., 2002
; Silver and Miller, 2004
). CSPGs are major components of the glial scar and are believed to inhibit axonal regeneration after spinal cord injury (Jones et al., 2002
; Tang et al., 2003
). Our results demonstrated that astrogliosis can be reduced through elevation of cAMP, and this could enhance axonal regeneration by rendering the CNS environment more permissive. It is not known how cAMP signaling mediates this effect, but inhibition of reactive astrocyte proliferation and downregulation of CSPG expression are two possibilities.
Rolipram has also been very effective when used in combination with dbcAMP and cell transplantation. In a study by Pearse and colleagues (2004)
, adult rats received Schwann cell grafts and intraspinal injections of dbcAMP one week after a moderate spinal cord contusion. Some animals were also treated with rolipram beginning at the time of injury (acute). The animals that received acute rolipram, dbcAMP, and Schwann cells displayed enhanced sparing of myelinated axons, greater myelination of spinal cord axons by the engrafted Schwann cells, and an overall increase in the number of axons within the grafts (; Pearse et al., 2004
). These observations indicated that this treatment strategy was highly neuroprotective. Most importantly, there was also significant regeneration of serotonergic axons across the lesion site and recovery of hindlimb locomotor function as measured on the Basso-Beattie-Bresnahan scale (Basso et al., 1995
). These findings demonstrated that elevation of cAMP could increase the regenerative capacity of injured axons following spinal cord contusion.
Figure 3 Rolipram in combination with Schwann cell transplantation and dbcAMP promotes axonal sparing and myelination after spinal cord contusion injury. Transverse sections of control spinal cords (A) showed extensive cavitation at 11 weeks after injury, but (more ...)
The third study combined administration of dbcAMP and NT-3 with transplantation of bone marrow stromal cells (MSCs), which have been shown to physically support axonal regeneration in the spinal cord (Hofstetter et al., 2002
; Lu et al., 2004
). Prior to injury, dbcAMP was injected into the L4 DRG, and NT-3 was injected into the injury site and caudal spinal cord immediately after dorsal column lesion (Lu et al., 2004
). MSCs were also transplanted after injury. The goal of this approach was to replicate the conditioning lesion effect in vivo
by stimulating neuronal cell bodies with dbcAMP, while simultaneously providing trophic support to the transected axons. Like a conditioning lesion, the injections of dbcAMP made prior to injury are prophylactic, and so, this approach should be considered a means of providing proof of principle rather than a clinically applicable treatment strategy. Animals that received a combination of dbcAMP, NT-3, and MSCs were the only treatment group that displayed extensive axonal regeneration beyond the lesion site, more than ever reported before in this lesion paradigm (Lu et al., 2004
). The exact contribution of NT-3 to this response is unknown, but it is undoubtedly acting as a chemoattractant for the regenerating dorsal column axons. NT-3 may also be working synergistically with dbcAMP to overcome inhibition by myelin, but it is important to bear in mind that neurotrophins can only overcome inhibition with priming or administration of PTX (Cai et al., 1999
), neither of which were used in this study. It is therefore possible that myelin signaling may override the effects of NT-3 in this model. Functional recovery was not observed in animals that received dbcAMP, NT-3, and MSCs, but this study still provides another clear example of axonal regeneration mediated by dbcAMP, and also emphasizes the benefits of combinatorial approaches to spinal cord injury repair.