TAxI to the Spinal Cord: A Peripherally Injected Peptide Carries Proteins to Motor Neurons

A peripherally injected peptide can ferry enzymes and other proteins into spinal motor neurons, according to a new study published March 1 in Proceedings of the National Academy of Sciences. The technique avoids the risks of a more invasive central administration route, but is likely to require an intact axonal connection, raising the question of how long following diagnosis it could be applicable in ALS.

The blood-brain-barrier, composed of specialized endothelial cells that line the blood vessels surrounding the brain, provides a security system for the brain, but also poses a nearly insurmountable barrier for entry of therapeutics that could help treat disease (see Jun 2011 news). A similar barrier exists between the blood and spinal cord, preventing delivery of protein biologics to neurons of the spinal cord. In order to overcome the challenge of biologics delivery to spinal cord motor neurons, senior co-authors Suzie Pun of the University of Washington in Seattle and Philip Horner of Houston Methodist Research Institute in Texas adapted the technique of a phage display screen to identify peptides that could be taken up by motor neurons in the spinal cord by retrograde transport from the innervated muscles.

In the traditional phage display screen, a library of bacteriophages is genetically engineered to each express a different randomly generated peptide fused to a coat protein expressed on its outer surface. A known nucleic acid or protein target, such as a membrane receptor or channel, is exposed to the full library of phages, and protein-protein or protein-DNA binding interactions can be identified and traced back to individual clones (Pande, J. et. al., 2010).

First author Drew Sellers and colleagues modified that system by injecting phage into the gastrocnemius muscle of mice, to hone in on those that reached the spinal cord. “We tried a higher-risk approach,” said Horner, “using the entire animal as a filter. We didn’t have a high degree of confidence it would work, but it did.”

After iterative rounds of injection, isolation of spinal cord tissue, amplification of recovered phage, and reinjection, the team identified one peptide that was present at the highest titers in the lumbar spinal cord 24 hours following intramuscular injection. The phage associated with this 14-amino acid peptide, which they dubbed TAxI (targeted axonal import), colocalized with neuronal axons and glial processes at the ventral root entry zone, and co-stained with motor neuron markers.

Next, the team set out to determine whether the phage was necessary for peptide transport, and additionally, whether the peptide could deliver protein cargos. Immunohistochemical analysis revealed that a conjugate of TAxI and NeutrAvidin colocalized with spinal motor neurons and their axons, which were retrogradely labeled from the gastrocnemius muscle with Fluorogold (FG). The peptide was taken up by about 50% of FG-labeled motor neurons within a targeted motor unit, Horner said, suggesting that TAxI “is an efficient means to facilitate delivery of biologics to the central nervous system.” Rhizotomy before injection prevented ipsilateral transport, indicating that an intact axonal path is required for efficient transport of protein to the spinal cord.

TAxI was also capable of delivering a functional enzyme cargo to the spinal cord: Cre recombinase. High Cre-recombinase activity was detected using mTmG mice, whose cells convert from expressing membrane-tomato to membrane-GFP (mGFP) upon Cre activation in the nucleus. Mice injected intramuscularly with TAxI-Cre exhibited a higher rate of mGFP conversion than those injected with control fusion proteins. In a second reporter line, Cre activity was detected not only in the motor neurons of the lumbar spinal cord but also in the cervical cord. There, the majority of Cre-recombinase activity was associated with glial, rather than neuronal, markers, suggesting some type of transcytosis between motor neurons and glia, a result that Horner said would need to be investigated further.

Philip Horner

Philip Horner. 

In order to determine whether human motor neurons would also be able to take up TAxI, the team exposed fresh frozen sections of human spinal cord to TAxI complexed with quantum dots to track the proteins, and found that the complex localized to nuclei of ChAT-positive motor neurons in the human spinal cord.

Horner suggested that the system might offer a strategy for targeting motor neurons that is complementary to current therapeutic approaches, such as gene delivery or antisense, which largely rely on central administration, and which are more invasive. The drawback may be the need for repeated or sustained delivery with TAxI, if the cargo were individual proteins rather than genes; timing of readministration would likely depend on the half-life of the administered protein. Delivery of DNA cargoes is also likely to be possible as well. The peptides themselves are probably too small to be immunogenic, he said, though this remains to be tested rigorously.

In addition to delivery of therapeutic proteins, the TAxI system may offer a way to visualize the remaining motor neurons in ALS patients or animal models, by attaching motor neuron-specific neuroimaging ligands to the peptide. “This may allow you to quantify functional axons in the patient,” Horner said.

The biggest challenge for therapy, he said, may be the loss of those same functional axons, which are thought to begin withdrawal from the muscle very early in the ALS disease process.

“This is quite an attractive approach using a unique peptide for retrograde delivery,” said Masatoshi Suzuki of the University of Wisconsin in Madison, who was not involved in the study and whose work has explored trophic factor therapy targeting the skeletal muscle using stem cells and viral vectors. “The viral vector would still be more challenging from a safety perspective, so this approach may offer an alternative,” one that may avoid the possible side effects and immunogenicity of viral methods.

Masatoshi Suzuki. Image used with permission

Masatoshi Suzuki. 

“On the other hand, he said, the early occurrence of axonal degeneration in ALS may limit its utility as a therapy after diagnosis, although clearly there are many intact axons that could potentially be targeted even then.”

“The fact that this can transport proteins or small molecules certainly is appealing,” said Kevin Foust of Ohio State University in Columbus, who has studied diverse gene therapy approaches for protecting motor neurons in ALS and other neurodegenerative diseases (see Sept 2013 news), and who did not participate in the study. “While delivering a high enough dose of a growth factor may be a problem, there may be other proteins, such as Crispr/Cas9 for inactivation of mutant SOD1, for instance, that could be effective at much lower concentrations.”

He suggested that it may also be useful to think about delivering a TAxI-like peptide intrathecally, thereby leveraging its specificity for motor neurons (or developing a different one for glia), while taking advantage of circulation of the cerebrospinal fluid to reach many more cells up and down the cord than would be possible with intramuscular injection. “The more ways we can deliver something, the better,” said Foust.


Sellers DL, Bergen JM, Johnson RN, Back H, Ravits JM, Horner PJ, Pun SH. Targeted axonal import (TAxI) peptide delivers functional proteins into spinal cord motor neurons after peripheral administration. Proc Natl Acad Sci U S A. 2016 Mar 1;113(9):2514-9. [PubMed].

topic-newmethods topic-preclinical
Share this: