A team led by KineMed Inc. researchers has shown that measuring the kinetics of CNS proteins in cerebrospinal fluid could help diagnose Parkinson's disease and amyotrophic lateral sclerosis.1 The company is now exploring the utility of the approach to diagnose multiple neurodegenerative diseases and monitor treatment responses.

Neurons in the brain synthesize and secrete a range of proteins that move along the microtubule structures of axons to the nerve endings-a process known as axonal transport. There, the proteins enter extracellular fluids, including cerebrospinal fluid (CSF).

Multiple studies have suggested links between disruptions in axonal transport and PD, ALS, Alzheimer's disease (AD) and Huntington's disease (HD).2-8 However, the molecular mechanisms underlying those disruptions are poorly understood, largely due to a lack of tools for studying axonal transport in vivo.

To build out the toolbox, the KineMed team turned to a method of deuterium (2H) labeling it had previously developed to investigate the effects of paclitaxel on microtubule structures in cancer cells.9 In that study, the team administered heavy water (2H2O) to cells and animal models to measure the incorporation of deuterium into tubulin dimers and polymers and thus track changes in microtubule structure and function.

In its latest study, the team administered heavy water to disease models of PD and ALS and to patients with PD and collected CSF samples via lumbar puncture for analysis of selected deuterium-labeled proteins.

Mouse models received heavy water orally and/or by intraperitoneal injection and underwent lumbar punctures once daily for up to 10 days. Patients and healthy subjects received heavy water orally and lumbar punctures no more than 4 times over the course of 40 days.

The team analyzed the CSF samples with mass spectrometry to determine the concentrations of four deuterium-labeled cargo proteins known to rely on axonal transport: soluble amyloid precursor protein (APP), chromogranin B (CHGB), neuregulin 1 (NRG1) and α-synuclein (SNCA). After plotting the concentration of each protein as a function of time, the team calculated the rates of appearance and disappearance for each protein in CSF.

Finally, for all proteins the researchers compared the protein appearance and disappearance rates in the disease model or patient with the rates in the corresponding control. Slower rates indicated a disease-related delay in the axonal transport of that protein.

For example, in healthy mice receiving intracerebroventricular infusions of nocodazole-a microtubule-destabilizing agent that disrupts axonal transport-the rates for Chgb and Nrg1 were lower than those in vehicle-treated controls, thus confirming that the method could detect changes in axonal transport.

The team also found that in PD mouse models and patients, the rates for three of the proteins-SNCA, CHGB and soluble APP-were lower than those in healthy controls. In mouse models of ALS, the rates for a different trio of proteins-Nrg1, Chgb and soluble App-were lower than those in healthy controls.

Additional experiments showed that the neuronal synthesis rates of the four proteins in the nocodazole-treated mice and the two disease models were similar to synthesis rates in the respective controls. This provided additional evidence that the method was measuring changes in axonal transport-not neuronal production-of the proteins, the team wrote in its report in The Journal of Clinical Investigation.1

The degree of deuterium enrichment of a protein also could reveal whether or not its axonal transport was disrupted, team coleader Marc Hellerstein told SciBX. By measuring the amount of deuterium incorporation in a protein that appeared in the CSF at a particular time point, the team could determine how long after heavy water administration the protein had been synthesized in neurons-and thus estimate how long it had taken the new protein to reach the CSF, he said.

"We knew the only factor that could explain why it took a week or so between synthesis in neurons and detection in CSF was axonal transport," said Hellerstein.

Hellerstein is cofounder of KineMed and chairman of its scientific advisory board. He also is a professor at the University of California, Berkeley and professor of medicine at the University of California, San Francisco's San Francisco General Hospital.

Patrizia Fanara, team coleader and KineMed's VP of neuroscience, added that deuterium labeling is good for measuring slow processes like axonal transport because heavy water diffuses through living tissues over the course of many days. This means the heavy water remains in tissues and continues to label newly synthesized neuronal proteins with deuterium over the time period required to measure axonal transport rates.

She added, "It is also very easy to administer heavy water in outpatient clinical settings as we did in this study."

The team included researchers from the McLaughlin Research Institute for Biomedical Sciences and the University of Osnabrueck.

"This looks exactly like the kind of good diagnostic tool we need in the difficult space of neurodegenerative disease," said Rémy Luthringer, Mind-NRG board member and venture partner at Index Ventures.

The link between neurodegenerative disease and deficits in axonal transport of cargo proteins was already known, "but now we have a way to systematically assess axonal transport," he said.

Luthringer wanted to know whether the axonal transport kinetics of cargo proteins change during the progression of a neurodegenerative disease. If so, the KineMed approach could enable early diagnosis-for example, before the onset of motor symptoms in PD-and thus guide treatment decisions, he said.

Barbara Tate, VP at Satori Pharmaceuticals Inc., was less sanguine about the approach's utility as a diagnostic because the team tested it only in animals and patients with established disease. "I didn't see anything in the study to suggest the approach could be used to diagnose neurodegenerative disease before current clinical criteria are met" and significant brain damage already has occurred, she said.

Instead, she said, the approach was better suited for monitoring disease progression and treatment responses.

"Validation of this approach for monitoring treatment responses in patients would first require a therapeutic that modifies axonal transport, which in turn would require a valid marker for that process," Tate said. "Thus, the approach is a long way from establishing clinical markers that regulatory agencies would accept for compound registration."

Nevertheless, she said the approach could be useful for monitoring responses in preclinical studies.

Indeed, a problem with neurodegenerative disease models is that behavior is often used to measure cognition, said Jo Ann Dumin, principal research investigator at Satori. "This can result in a lot of false positives, and the measures don't always translate to humans. So the possibility of having a biochemical marker in CSF as a surrogate for cognition in preclinical models is intriguing."

She added, "Down the road, KineMed's approach might have similar utility in patients because the tools for measuring cognition in humans are not always precise either."

For instance, Dumin said, it would be interesting to see whether the approach could measure microtubule-associated protein-t (MAPT; TAU; FTDP-17) and TAU-related proteins in CSF to follow progression and treatment response in patients with AD.

Tate agreed, noting that such biochemical markers-even those not accepted by regulatory agencies-could help determine whether a therapy was on the right track in the clinic. "If a company could use a biochemical marker to get an early read of therapeutic response in a small patient population, that could save a lot of time and money," she said.

This year, Satori hopes to submit an IND for SPI-1865, a g-secretase modulator, to treat AD.

Heavy mettle

Another question is whether the technique can help identify subsets of neurons that are involved in neurodegenerative diseases.

Luthringer wanted to see KineMed's approach combined with MRI or other readouts that could link altered axonal transport to anatomical or functional changes in specific areas of the brain. "That could help us better understand disease progression and assess response to specific therapies," he said.

Fanara agreed and said the team's preclinical findings suggest the axonal transport kinetics of a cargo protein may be specific to differing populations of neurons in each neurodegenerative disease. Thus, she said, it might eventually be possible to diagnose a disease by detecting altered axonal transport kinetics of cargo proteins in specific classes of neurons.

Additionally, "if we can link disruptions in axonal transport of a given target protein to a particular subset of neurons, we might gain insights into disease pathogenesis and progression," Hellerstein said.

"It might be difficult to find patients willing to submit to a lumbar puncture" for diagnostic or monitoring purposes, cautioned Luthringer. "To get clinicians to recommend such an invasive procedure and to get patients to accept it, you would need to show it has some real benefit compared with noninvasive techniques."

Dumin also said there might be diurnal variations in the levels of the proteins in CSF. "One might expect a marker of synaptic activity to show changes according to the time of day," she said. "Also, the results could vary depending on where in the spine you take the sample."

Nevertheless, Luthringer said the technique seemed equally applicable to preclinical and clinical studies, and "we are now thinking about using this method in our own studies at Mind-NRG."

Mind-NRG's NRG-101, the extracellular fragment of the b1 isoform of NRG1, is in preclinical development to treat PD and schizophrenia.

Fanara said KineMed is investigating the potential of the approach to diagnose and monitor treatment responses in PD, ALS, AD and HD.

The company has also begun a cross-sectional study in patients with PD-funded by The Michael J. Fox Foundation for Parkinson's Research (MJFF)-to see "whether there is any meaningful relationship between the degree of axonal transport deficit and severity of disease," she said. The study will also follow the patients over time to look for associations between the degree of axonal transport deficit and progression of disease that could have prognostic value.

Additionally, KineMed is looking for other cargo proteins that rely on axonal transport as potential markers of neurodegenerative disease.

"We hope that tracking additional cargo proteins will reveal whether decreased axonal transport is specific for certain classes of neurons rather than universal" in a given disease, she said.

KineMed has exclusively licensed a portfolio of patents and patent applications covering the technology described in the JCI study from UC Berkeley.

The company is seeking corporate collaborators interested in applying the technology to the preclinical and clinical development of therapies to treat neurological disorders, Fanara said.

In 2008, KineMed received a $700,000 grant from MJFF to fund the PD patient studies reported in the JCI paper. In June, MJFF awarded KineMed an additional $1.2 million to continue developing the technology as a PD diagnostic.

Haas, M.J. SciBX 5(36); doi:10.1038/scibx.2012.944
Published online Sept. 13, 2012


1.   Fanara, P. et al. J. Clin. Invest.; published online Aug. 27, 2012; doi:10.1172/JCI64575
Contact: Patrizia Fanara, KineMed Inc., Emeryville, Calif.
e-mail: pfanara@kinemed.com

2.   Roy, S. et al. Acta Neuropathol. 109, 5-13 (2005)

3.   El-Kadi, A.M. et al. J. Neurosci. Res. 85, 2557-2566 (2007)

4.   Her, L.-S. & Goldstein, L.S.B. J. Neurosci. 28, 13662-13672 (2008)

5.   Williamson, T.L. & Cleveland, D.W. Nat. Neurosci. 2, 50-56 (1999)

6.   Lee, H.-J. et al. Eur. J. Neurosci. 24, 3153-3162 (2006)

7.   Saha, A.R. et al. J. Cell Sci. 117, 1017-1024 (2004)

8.   Barten, D.M. et al. J. Neurosci. 32, 7137-7145 (2012)

9.   Fanara, P. et al. J. Biol. Chem. 279, 49940-49947 (2004)


Index Ventures, Geneva, Switzerland

KineMed Inc., Emeryville, Calif.

McLaughlin Research Institute for Biomedical Sciences, Great Falls, Mont.

The Michael J. Fox Foundation for Parkinson's Research, New York, N.Y.

Mind-NRG, Geneva, Switzerland

San Francisco General Hospital, San Francisco, Calif.

Satori Pharmaceuticals Inc., Cambridge, Mass.

University of California, Berkeley, Calif.

University of California, San Francisco, Calif.

University of Osnabrueck, Osnabrueck, Germany