Trophic factors such as IGF-1 have shown promise for the treatment of ALS.3–5
In this study, we report that CNS-restricted delivery of IGF-1 is sufficient to modify disease progression in symptomatic ALS mice. Specifically, we showed that injecting a recombinant AAV vector encoding IGF-1 within the DCN of SOD1G93A
mice resulted in axonal transport of vector and/or expressed IGF-1 protein to the brain stem and all segments of the spinal cord. This, in turn, led to improved muscle function and a significant extension of life span. Furthermore, IGF-1 also attenuated astrogliosis, microglial activation, peroxynitrite formation, and glial cell–mediated release of TNF-α and NO.
Results obtained using mouse models of motor neuron disease have demonstrated that trophic factors (e.g.
, IGF-1, BDNF, CNTF, and GDNF) have potent effects on motor neuron survival.3–5
However, systemic administration of some of these recombinant trophic factors into subjects with ALS showed only very modest clinical benefit.33–36
Studies in ALS mice suggested that inadequate delivery of these trophic growth factors to the CNS may have been responsible for the poor response. Only systemic administration of vascular endothelial growth factor has been reported to be effective in treating SOD1G93A
Intrathecal administration of purified IGF-1 to the same mouse model was also efficacious.38
However, in both cases, positive effects were reported only when treatment was initiated in presymptomatic animals. In contrast to the results observed with systemic or intrathecal delivery of purified trophic factors, intramuscular injections of viral vectors encoding these factors demonstrated significant therapeutic benefit in the SOD1G93A
mice, even when administered after the onset of overt disease symptoms.6,39
The results of this study indicate that trophic factor delivery to the CNS is sufficient to modify disease progression in symptomatic ALS mice. An advantage of this delivery strategy over existing approaches is that it permits the targeting of multiple areas that undergo neurodegeneration in ALS with a single injection site and obviates the need for injecting directly into the spinal cord where neurodegeneration is taking place. Comparison of survival benefits achieved with DCN versus intramuscular delivery of AAV-IGF-1 is difficult, given that a “death event” in an ALS mouse is artificially determined (i.e.
, occurs when the mouse can no longer right itself within 30 seconds). In the mouse model, testing a therapeutic is limited to measuring the ability to offer protection to motor neurons predominantly residing in the lumbar division of the spinal cord, which is responsible for the righting reflex. Indeed direct intraspinal injections of AAV-IGF-1 led to significant increases in survival.40
However, given that respiratory failure is the primary cause of death in ALS patients, we believe that delivering AAV-IGF-1 to the DCN may offer an advantage in that it permits targeting regions of the CNS that control respiration. This, in turn, may lead to a level of efficacy in ALS patients beyond what is observed in ALS mice.
Cellular mechanisms that modulate disease progression in ALS have not been known until just recently. While disease onset is initiated by motor neurons in ALS, it appears that glial cells modulate disease progression.8,9
The benefit provided by IGF-1 has mainly been thought to be attributable to activation of anti-apoptotic pathways (e.g.
, AKT and Bcl-2) within motor neurons of the spinal cord.41
However, efficacy is still observed when treatment is initiated during disease onset suggesting that the actions of IGF-1 may also be mediated through additional mechanisms, including muscle enhancement. While IGF-1 has potent effects on muscles, it has been recently demonstrated that muscle is not a direct target for mutant SOD1-mediated toxicity.42
Our in vivo
studies here showed that IGF-1 may also attenuate a number of pathological features that have been linked to motor neuron cell death including increased NO activity, elevated peroxynitrite expression, astrogliosis, and microglial activation. Using a newly developed in vitro
model of ALS, we corroborate our in vivo
findings and also confirm earlier published results demonstrating that motor neurons containing the SOD1 mutation required coculture with astrocytes containing the SOD1 mutation for motor neuron death.29–31
To our knowledge, no study to date has compared the efficacy of a potential therapeutic such as IGF-1 in this in vitro
ALS models system with efficacy results obtained using ALS mice. We show here that IGF-1 is potently neuroprotective when present in the coculture system. We observed a delay in neuritic atrophy, cell death, and caspase-9 activation in motor neurons treated with IGF-1. To decipher whether IGF-1 had effects on non-neuronal cells, we performed experiments using microglial cells that contained the mutant SOD1, because these cells have been directly linked to disease progression. Surprisingly, IGF-1 significantly lowered TNF-α and NO production when these cells were activated with lipopolysaccharide. Furthermore, we show here that the neuroprotective effects of IGF-1 are not exclusively limited to motor neurons in the coculture system, rather IGF-1 exerts strong inhibitory effects on SOD1G93A
-mediated toxicity in astrocytes. These results strongly implicate that pleiotropic effects for IGF-1 in multiple non-neuronal subtypes of ALS suggested by the activation of AKT () could be a potent neuroprotective factor for motor neurons along with the ability to attenuate aberrant glial cell activation and subsequent products produced by astrocytes and microglia, which have been implicated in ALS, such as glutamate, peroxynitrite, TNF-α, and NO. While not all of the signaling pathways of IGF-1 were evaluated in this coculture study, AKT activation appears to play a significant role in protecting motor neurons, in part by its actions for neuroprotection in motor neurons themselves, as well as suppressing the toxicity derived from SOD1G93A
-containing astrocytes. This is of interest because it has recently been demonstrated that activated phosphorylated AKT is absent in motor neurons of both sporadic and familial ALS patients, and that motor neurons from mutant SOD1 mice lose activated AKT early in the disease.43
IGF-1 signaling through other signaling pathways may also be responsible in part for the effects on motor neuron protection.
In summary, these results highlight a novel approach to deliver IGF-1 to multiple regions of the CNS by a single injection site and show for the first time that CNS-restricted delivery of IGF is sufficient to modify disease progression in ALS mice. We also show that in addition to providing motor neuron protection, IGF-1 also modulates pathological events mediated by glial cells in ALS. These findings support the development of therapies that are designed to treat ALS by targeting motor neurons and their cellular environment. Furthermore, the data support the strength of developing therapeutic screens in the ALS coculture system and that potential future therapies may exploit the activation of AKT pathways in the CNS to curtail ALS progression.