Sage Journals: Discover world-class research

Abstract

Our purpose was to enable an in vivo imaging technology that can assess the anatomy and function of peripheral nerve tissue (neurography). To do this, we designed and tested a fluorescently labeled molecular probe based on the nontoxic C fragment of tetanus toxin (TTc). TTc was purified, labeled, and subjected to immunoassays and cell uptake assays. The compound was then injected into C57BL/6 mice (N = 60) for in vivo imaging and histologic studies. Image analysis and immunohistochemistry were performed. We found that TTc could be labeled with fluorescent moieties without loss of immunoreactivity or biologic potency in cell uptake assays. In vivo fluorescent imaging experiments demonstrated uptake and retrograde transport of the compound along the course of the sciatic nerve and in the spinal cord. Ex vivo imaging and immunohistochemical studies confirmed the presence of TTc in the sciatic nerve and spinal cord, whereas control animals injected with human serum albumin did not exhibit these features. We have demonstrated neurography with a fluorescently labeled molecular imaging contrast agent based on the TTc.

DESPITE A LARGE CLINICAL NEED, no suitable contrast agent is available for visualizing peripheral nerve tissues by imaging. Numerous nerve diseases could be better diagnosed and treated if clinical neurography could be performed effectively and reliably. Current methods for imaging nerves such as ultrasonography and magnetic resonance imaging (MRI) do exist but are limited and could be improved by the use of a suitable contrast agent.¹

A contrast agent for neurography should preferably be based on a robust physiologic mechanism that is present in all nerves and closely tied to the health of nerve tissues. One such mechanism is the two-way cycle of communication between peripheral nerves and their end-organs first proposed by Levi-Montalcini and Hamburger (Figure 1A).² In this cycle, action potentials in the nerve lead to synaptic neurotransmitter release, which gives rise to end-organ activity; over long periods, this activity will lead to hypertrophy of the end-organ. The end-organ, in turn, secretes neurotrophic factors that are absorbed by the nerve, and these factors are retrogradely transported to the nerve cell body, where translation and transcription are then altered to promote nerve health and function. A large body of literature now demonstrates that the nerve and the end-organ mutually support each other's maintenance.³ The failure of retrograde transport has been clearly shown to be a primary cause of peripheral neuropathies such as those caused by diabetes,^4,5 neurotoxic chemotherapy agents,⁶ compressive neuropathy,^7,8 and amyotrophic lateral sclerosis.⁹

The purpose of this exploratory study was to characterize, validate, and provide a first demonstration of an imaging agent that uses the retrograde transport mechanism in nerves. We fluorescently labeled the nontoxic C fragment of tetanus toxin (TTc), which mediates nerve cell-specific uptake and transport of the intact toxin.^10,11 To test our agent, we injected the compound into the calf muscles of mice. Time-lapse fluorescence imaging was used to visualize the process of the compound's retrograde transport in vivo. The specificity of the obtained in vivo data was confirmed by histologic methods. Ex vivo imaging was used to measure the speed of retrograde axonal transport in the sciatic nerve. We found the rate to be compatible with the fast axonal transport mechanism.^12,13

Materials and Methods

Synthesis, Labeling, and Characterization of TTc

Recombinant TTc was isolated from Epicurion coli carrying the expression plasmid pAE-Fc (the kind gift of Dr. Ana L. T. O. Nascimento).¹⁴ This protein is identical to the original published sequences,^15,16 except for six histidine residues appended to the N-terminus to enable affinity chromatography purification. The crystallographic structure has been solved (Figure 1B),¹⁷ and a putative ligand-receptor interaction with nerve terminal membrane gangliosides has been demonstrated with crystallography.¹⁸ The 50 kDa His-tagged protein was purified in its native state using standard procedures¹⁹ and labeled with Alexa Fluor succinimidyl esters 546, 680, and 790 according to the manufacturer's protocol (Invitrogen, Carlsbad, CA). Fluorophores emitting at 546 and 680 nm were used for microscopic studies, whereas the 790 nm probe was used for in vivo studies. The properties of labeled TTc were compared with those of commercially available TTc (Roche Applied Science; Indianapolis, IN) by Western blot and cell uptake studies using PC-12 pheochromocytoma cells (American Type Culture Collection, Manassas, VA) cultured in poly-l-lysine-coated Costar 96-well plates at 2,000 cells/well and differentiated with 50 ng/mL nerve growth factor (Invitrogen). Cellular uptake of TTc-Alexa680 was compared with that of labeled human serum albumin (HSA-Alexa680) by live imaging over a period of 60 minutes. Imaging was performed using an inverted Olympus FV1000 scanning confocal microscope (Olympus America Inc, Center Valley, PA) equipped with the following lasers: multiline Ar laser (457 nm, 488 nm, 515 nm), green HeNe (543 nm), and red HeNe (633 nm) (Melles Griot, Albuquerque, NM).

Figure 1.

A, The nerve-end organ cycle of communication. There is continuous interplay between neurons and their end-organs, with nerve stimulation making up one leg and retrograde transport of nerve growth factors the other leg of the cycle. The health and maintenance of both the nerve and the innervated end-organ rely on unbroken mutual communication. B, The crystallographic structure of the nontoxic tetanus toxin C fragment (TTc) is demonstrated at a resolution of 1.61 angstrom.¹⁷ Additional crystallographic studies have suggested that TTc binds by cross-linking to gangliosides in the cell membranes of nerve terminals.¹⁸ C, Characterization of labeled TTc. Top panel: Protein integrity and immunoreactivity of commercially available TTc and recombinant purified TTc are compared. An SDS-PAGE gel demonstrates commercially available TTc (lane 1), recombinant TTc with a His tag (lane 2), fluorescently labeled TTc with His tag (lane 3), and human serum albumin (HSA) used in this study as a negative control (lane 4). The commercial and purified TTc both run at 50 kDa, and the natively purified protein used in this study did not show any degradation during the purification or fluorescent labeling process. The immunoreactivity of commercially available TTc (lane 5) and TTc with His tag used in this study (lane 6) was identical and was retained after fluorescent labeling (lane 7), whereas HSA (lane 8), the control protein, was not. Fluorescently labeled TTc (lane 9) was also visible when fluorescently scanning the electrophoresis gel, demonstrating the tight covalent association between TTc and the fluorophore. Lower panel: PC-12 cells cultured under differentiating conditions avidly took up both commercial TTc (left image) and recombinant TTc (middle image) used in this study but not in HSA (right image).

In Vivo and Ex Vivo Imaging

All animal experiments were performed in compliance with National Institutes of Health and institutional guidelines. Sixty female C57BL/6 mice purchased from Charles River Laboratories (Wilmington, MA) were used in the experiments. Dose-finding and imaging studies were performed on 24 animals and histologic studies on 36 animals. Animals were anesthetized using isoflurane (Abbott Laboratories, North Chicago, IL), and their dorsal fur was removed with hair removal cream. The mice underwent injections in their calf muscles with 10, 30, or 100 μg of fluorescent compound using Hamilton glass syringes with 26-gauge needles mounted on a stereotactic frame. Animals were placed into the whole-body animal imaging system (Xenogen IVIS 200, Caliper Life Sciences, Hopkinton, MA) equipped with the appropriate bandpass excitation and emission filters (DsRed for Alexa Fluor 546, 500–550/575–650 nm; Cy5.5 for Alexa Fluor 680, 615–665/695–770 nm; and ICG for Alexa Fluor 790, 710–760/810–875 nm). Images of the dorsal aspects of the animals were taken at specific time points under identical conditions: a high fluorescence level, binning level set at medium, and field of view of 13.1 cm. White light photographs were taken with a 0.2 s/f = 8 exposure and fluorescence images with a 0.5 s/f = 4 exposure. After animals were killed, tissues including the sciatic nerves and spines were excised, cryopreserved, placed in the small animal imager and imaged using the same parameters as for in vivo imaging. These same tissues were then cryosectioned and used for histological studies. Images were analyzed using the Living Image 3.0 systems software (Caliper Life Sciences). To allow comparison among fluorescence images obtained during different imaging sessions, fluorescence normalized to photons per second per centimeter squared per steradian (p s⁻¹ cm⁻² sr⁻¹) was displayed on the same scale of intensity.

Histologic Analysis and Fluorescence Confocal Microscopy

Excised spines with cords in situ were cut into approximately 5 mm-thick transverse pieces. Serial frozen sections at a thickness of 20 μm were cut in a cryostat. Thoracic spine sections were fixed in Carnoy fluid (ethanol:acetic acid, 6:1), permeabilized with 0.1% Tween 20 for 10 minutes, and immersed in hematoxylin QS (Vector Laboratories, Burlingame, CA); stained sections were photographed using an Olympus BX51 microscope (Olympus). For immunohistochemical detection of TTc, 4% paraformaldehyde-fixed sections were blocked in a solution of phosphate-buffered saline (PBS), 1% bovine serum albumin, and 0.03% Triton X-100, reacted with a polyclonal rabbit anti-TTc antibody (1 μg/mL; Rockland Immunochemicals, Gilbertsville, PA) in blocking buffer containing 0.01% Triton X-100 for 1 hour at room temperature, and incubated with a secondary goat antirabbit Alexa Fluor 488-labeled antibody (1 μg/mL, Invitrogen) at room temperature for 30 minutes. For detection of fluorescence, sections were mounted in Prolong Gold antifade reagent. Micrographs used were an average of five adjacent slices of a vertical stack captured at 2.5 μm/slice using a 20 × UPLSAPO oil objective with an Olympus FV1000 scanning confocal microscope (Olympus).

Nerves (n = 12) were cut into 3 mm pieces starting from the nerve insertion into the muscle between the two heads of the gastrocnemius muscle. The tissue pieces were either immediately fixed in 4% paraformaldehyde, washed with PBS, and mounted with Prolong Gold antifade reagent with DAPI stain (4′,6-diamidino-2-phenylindole, Molecular Probes, Eugene, OR) or embedded in Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA) and imaged and analyzed using confocal microscopy.

Rate of Transport

Animals underwent injections of 30 μg TTc into the right calf muscle and 30 μg HSA into the left calf muscle. Mice were sacrificed 20 and 60 minutes after the injection (n = 5 per time point), and nerves were excised and imaged. Images were adjusted uniformly using the fluorescence imaging software to achieve a dark background on the HSA-injected side, followed by measurement of the fluorescent front from the amputated nerve on the TTc-injected side. This distance and the time after injection were used to estimate the speed of transport.

Results

Synthesis, Labeling, and Characterization of TTc

His-tag affinity chromatography was used to purify recombinant TTc, yielding protein at a concentration of 5 mg/mL. TTc labeled with Alexa Fluor546, −680, and −790 all retained integrity and immunoreactivity after purification and fluorescent labeling in comparison with commercially available TTc (Figure 1C).

TTc-Alexa680 conjugate was tested on PC-12 cell cultures and demonstrated expected uptake into the cells,²⁰ indicating retained biologic activity compared with commercially available TTc (see Figure 1C). In contrast, Alexa-labeled control HSA (HSA-Alexa680) showed no uptake.

In Vivo Imaging

Initial imaging experiments were directed at establishing an appropriate administration and dosing parameters and resulted in an intramuscular dose of 30 μg of TTc-Alexa790 being selected for further evaluation. Lesser doses yielded insufficient photon flux for imaging, whereas larger doses yielded nonspecific distribution through the body fluids that obscured specific uptake (data not shown).

Subsequent whole-body fluorescence imaging experiments were done using fluorescently labeled TTc and HSA as a control substance at identical doses. These experiments showed that the patterns of fluorescent agent distribution were markedly different for TTc and HSA, compatible with specific and fast nerve uptake for TTc and nonspecific diffusion for HSA. TTc-Alexa790 showed rapid and specific uptake and transport into spinal cords via the sciatic nerves after unilateral intramuscular injection into the calf muscles of mice over a time period of 90 minutes (Figure 2A, right animal). In the same experiment, contralateral injections with HSA-Alexa790 did not result in nerve uptake or transport but in nonspecific diffusion into the body fluids (see Figure 2A, left animal), whereas noninjected controls showed no autofluorescence background (see Figure 2A, central animal). We also demonstrated in a different experiment the time course of uptake from the muscular injection of a single dose of 30 μg TTc-Alexa790, through the nerves and into the spine (Figure 2B). Nerve uptake into the spinal cord was visible within 30 minutes postinjection and became more pronounced as time progressed.

We wanted to assess the possibility that TTc might be reaching the cord through a transvascular route by diffusing into the blood from the muscular injection site rather than via direct nerve transport. To do this, we injected TTc intravenously at the same dose level (30 μg) and imaged in an identical manner. We found only diffuse body uptake and some renal clearance but no specific spinal uptake.

Quantitative analysis was done by drawing regions of interest (ROI) over the lumbar and thoracic areas (Figure 2C). Fluorescence values over the injection site and the adjacent thighs were highly variable and discarded from analysis. TTc after intramuscular injection (n = 5), HSA after intramuscular injection (n = 3), and TTc after intravenous injection (n = 4) all showed markedly different quantitative profiles, paralleling their differing imaging appearance. TTc accumulated over the spine (top two curves in Figure 2C) but only after intramuscular injection, whereas HSA did not distribute to the spine when injected into the calf muscles in the same manner as TTc (middle two curves in Figure 2C). Intravenous injection of TTc resulted in dilute background activity only (bottom two curves in Figure 2C). In the thoracic region, mean fluorescent values showed statistically significant differences between TTc intramuscular and HSA intramuscular injections (p < .05, Student t-test, two-tailed, unequal variance) at all points (from 10 to 90 minutes) excepting baseline (top and middle curves on Figure 2C, right panel). These values tended to statistical significance for the lumbar area but did not reach significance, likely because of increased variability induced by photon leakage from the nearby located injection site in the calf muscles.

Ex Vivo Imaging

Excised tissues from animals after TTc injection showed fluorescent imaging features similar to those observed with in vivo imaging. When the spine was dissected and divided in cross section (Figure 3A), we were able to detect robust fluorescent uptake in the cords of animals injected with TTc (Figure 3A, right panel) but not with HSA (Figure 3B, left panel). Thus, the imaging findings of Figure 2, A and B, were confirmed by the ex vivo findings of Figure 3, A and B (the same animal in each case). We noted a progressive increase in the amount of distal-to-proximal migration of TTc-related fluorescence in the sciatic nerve as retrograde transport carried the protein up the nerve from the muscular injection site.

Histologic Analysis and Fluorescence Confocal Microscopy

Confocal imaging of whole-mounted nerves demonstrated specific nerve transport of TTc-Alexa680. Nerve transport was visible in the inner core or depth of the nerve and appeared to be straight linear, parallel, and nonbranching structures (Figure 4A, left panel) as early as 20 minutes after intramuscular injection; HSA-Alexa680 was not transported (Figure 4A, right panel). Nonspecific lymphatic uptake in both TTc-Alexa680 (see Figure 4A left panel) and HSA-Alexa680 (see Figure 4A, right panel) was observed as curvilinear, nonparallel, and branched structures, located exclusively on the outer surface of the nerves. The cryosectioned nerve and cord tissue showed the presence of TTc-Alexa546 ipsilateral to the site of the injection in the nerves and cords. In the cross-sectioned nerves, we observed TTc-Alexa546 in the nerve (Figure 4B, left panel), whereas the HSA-Alexa546 control showed background levels of fluorescence only (Figure 4B, right panel).

Figure 2.

In vivo imaging of retrograde neuronal transport after intramuscular injection of fluorescently labeled nontoxic tetanus toxin C fragment (TTc) and controls. A, Fused white light-fluorescence images obtained using the IVIS 200 small-animal imager demonstrate the time course (in minutes) after a single intramuscular injection of either 30 μg human serum albumin (HSA)-Alexa790 (left animal injected in left calf, dotted oval) or 30 μg TTc-Alexa790 (right animal injected in right calf, solid circle); a noninjected control animal is shown between the injected mice. TTc was taken up into nerve tissue and progressively transported to the spinal cord (yellow arrows), whereas HSA did not undergo specific transport. B, Fluorescence images of a second animal that underwent injection of a single dose of 30 μg TTc-Alexa790 into the left calf muscle also indicate nerve uptake and transport of TTc into the spinal cord (yellow arrows). At later time points, low-level nonspecific diffused fluorescence can be seen in other body parts, such as the tail and contralateral hindlimb. C, Quantitative analysis of TTc nerve transport, comparing control agents and routes of administration. Regions of interest (ROI) were drawn as shown (left panel) over the injection site, thigh, lumbar (short arrow), and thoracic areas (long arrow) (left panel) and plotted as a function of time (right panel). Three administration paradigms were used: (1) TTc-Alexa790 30 μg intramuscular injection into the calf (n = 5) (top two curves), (2) HSA-Alexa790 30 μg intramuscular injection into the calf (n = 3) (middle two curves), and (3) TTc-Alexa790 30 μg injected intravenously into the tail vein (n = 4) (bottom two curves). There are clear differences between TTc and HSA when injected intramuscularly, with HSA showing a low-level diffuse signal compatible with passive diffusion, whereas TTc shows progressive uptake over the spine compatible with nerve transport. The uptake of TTc over the spine is not the result of transvascular spine localization because intravenous TTc administration gives rise to a diffuse background signal only Thoracic values were statistically significantly different between HSA and TTc for intramuscular injections from 10 minutes onward (p > .05, two-tailed t-test, unequal means). Lumbar values approached but did not reach statistical significance, likely owing to variability induced by photon diffusion from the nearby injection site.

Lymphatic transport was only noted close to the injection site in the popliteal fossa (see Figure 4A, left and right panels) for both TTc and HSA but was present only at the distalmost end of the nerve, close to the site of muscular injection. We did not observe lymphatic transport in the mid- or proximal nerve.

Unilateral TTc calf muscle injections produced ipsilateral TTc-related fluorescence in the gray matter of the cord, whereas no uptake was seen for the negative HSA control. Immunofluorescence images of TTc in the cord show extensive distribution in the cord, but especially ipsilateral to the anterior motoneurons (Figure 4C, middle panel; Figure 4C, left panel, shows cord anatomy). TTc accumulation was intracytoplasmic in the neurons, as demonstrated with nuclear counterstained DAPI images, again supporting transport within the neurons (Figure 4C, right panel).

Rate of Transport

Nerves harvested at 20 and 60 minutes (n = 5 per time point) postinjection were used to estimate the transport rate of 30 mg TTc by ex vivo imaging (Figure 5). There were marked differences between HSA (see Figure 5A, top) and TTc (see Figure 5A, bottom). TTc demonstrated an advancing fluorescent front extending in the nerve from the muscle side toward the spinal cord, whereas HSA showed background fluorescence near the injection site that did not change over time. Plotting the distance of the fluorescent TTc front as a function of time yielded an approximate speed of 12.1 mm/h (see Figure 5B), which is in agreement with published data on the fast axonal transport mechanism.^{21

26}

Figure 3.

Ex vivo imaging of excised tissues showing time- and dose-dependent nerve and cord uptake of nontoxic tetanus toxin C fragment (TTc). A, White-light photographic images of mouse spine and cord. Top panel: Intact dissected spine. Lower panel: Approximately 5 mm cross sections of spine. The cord is clearly visible as a central white structure in higher sections above the level of the conus medullaris in the thoracic spine. B, Ex vivo imaging of thoracic cross sections 3 hours after injection of either 30 μg Alexa790-labeled human serum albumin (HSA) (left, stippled circle) or TTc (right, solid circle); fluorescence emitted from the cord was detected in the TTc-injected, but not the HSA-injected, sections. This is an ex vivo imaging demonstration of the same animals demonstrated in Figure 2A. Note the strong TTc-related signal in the spinal cord, with nonspecific background signal only associated with HSA.

Discussion

We demonstrated that TTc can serve as a molecular imaging agent for neurography, allowing imaging of peripheral and central nerve structures. The mechanism is based on the uptake and retrograde transport of labeled TTc along nerves after intramuscular injections and could be applied to (a) demonstrate nerve anatomy or (b) the functional status of the physiologic mechanisms of retrograde uptake and transport.

This represents the first demonstration of a novel nerve imaging agent, something that is not currently available to imaging science. Nerve imaging agents have great potential for increasing our diagnostic capabilities in a wide variety of nerve diseases and thus to potentially allow a greater understanding of nerve disease and aid in the development of improved therapies.

TTc in the Literature

TTc, which is nontoxic on its own, mediates nerve uptake and retrograde transport of the intact tetanus toxin in nature, as is well known from the literature.^10,11 The fragment can be obtained from the intact toxin by papain digestion; however, the sequencing and subsequent cloning of the fragment have now enabled the purification of recombinant protein.¹⁶ The nerve uptake and transport attributes of TTc have been exploited for years in histologic investigations, leading to its use as a nerve-tracing agent in histologic studies on excised tissues.^{27

30} We have built on this prior work by using modern molecular imaging technologies to make the process of nerve uptake and transport visible to imaging.

Tetanus toxin is known to be transported using the fast retrograde axonal transport mechanism.^12,13,31 Fast axonal transport is mediated by microtubules³² interacting with motor proteins, kinesin for anterograde transport,^{33

35} and dynein^36,37 for retrograde transport. The natural function of the fast transport mechanism is the transport of organelles and growth factors along neuronal processes.

Limitations of the Study

We faced significant challenges in quantifying TTc transport rates in vivo. Specifically, (a) in vivo optical imaging suffers significantly from light attenuation by tissue in a nonlinear, depth-dependent manner and (b) emitted photon flux is in part dependent on the dose of fluorescent agent administered. Both of these effects have a strong impact on measured photon emissions and thus on transport rate calculations. To minimize these effects, we standardized our fluorescent imaging dose to 30 μg and chose to use excised samples, thus eliminating tissue light attenuation as a variable (see Figure 5). This ex vivo methodology is similar to that used by prior investigators, and our measured rates are similar to the previously published rates of fast axonal transport in general.^{21

26} The transport rate of intact tetanus toxin has been studied specifically previously and found to vary from 2.8 mm/h in thin autonomic nerve fibers to 13 mm/h in large sensory nerve fibers³¹; our measured rate of 12.1 mm/h falls in the upper end of this range and likely reflects the high proportion of large-diameter sensory and motor nerve fibers in the sciatic nerve.

We found that successful imaging absolutely required agents such as TTc-Alexa790, which emit in the near-infrared (NIR), similar to the experience of other investigators.^{21

25} Other fluorophors emitting outside the NIR range were useful only as histologic or ex vivo imaging agents.

Figure 4.

Histologic examination of excised tissues confirms nerve transport and distinguishes between nerve and lymphatic transport. A, Confocal longitudinal optical sections of sciatic nerves near the popliteal trifurcation after injecting fluorescently labeled nontoxic tetanus toxin C fragment (TTc)-Alexa680 and human serum albumin (HSA)-Alexa680 (30 μg, 20 minutes after injection) into opposite calf muscles of the same animal. Nerve uptake of TTc (left first panel) is characterized by straight linear, nonbranching, and parallel uptake into nerve fascicles in the depths of the nerve, whereas HSA-injected animals had no specific uptake in their sciatic nerves (right first panel). This contrasts to the appearance of lymphatic uptake, which is branched, nonparallel linear uptake on the outer surface of the nerve in both TTc (left second panel) and HSA (right second panel). B, Cross sections through the sciatic nerve near the popliteal trifurcation 20 minutes after injecting 30 μg TTc-Alexa680 into the calf muscle (left panel). The high-intensity peripheral uptake on the nerve surface (arrow) signifies interstitial and lymphatic fluorophore uptake; this uptake was only noted close to the injection site and was not noted more proximally along the nerve. The fluorescence in the substance of the nerve is due to transported TTc (yellow circle, left panel). Contrast this to HSA-injected controls, where no transport is noted (yellow circle, right panel). C, Cross sections through the spinal cord (within an hour after injection). Left panel: Light micrograph of hematoxylin-stained cryosection demonstrating normal cord anatomy. Middle panel: Confocal micrograph of an immunofluorescent cryosection demonstrating the distribution of TTc in the thoracic cord with anti-TTc fluorescent antibodies after injection of 30 μg TTc into the right calf muscle and 30 μg HSA into the left. The antibody detects TTc in the cord ipsilateral to the TTc injection (red). Right panel: Higher magnification with DAPI counterstain (yellow) demonstrates the intracytoplasmic localization of TTc in the nerve cell bodies as detected with anti-TTc antibodies (red).

Alternative Transport Hypotheses

Although our findings are convincing of nerve transport, alternative hypotheses should also be considered: lymphatic transport and distribution by diffusion through the blood pool.

We assessed the role of lymphatic distribution by using HSA as a control substance, as it would be expected to distribute via lymph and blood but not specifically via nerve transport. Confocal microscopy of excised nerves did show some lymphatic distribution of both TTc and albumin, but only close to the injection site in the calf muscles, never in the mid- to proximal nerves, where nerve uptake and lymphatic uptake might be confused with each other (see Figure 4). The histologic appearance of lymphatic and nerve transport could clearly be distinguished. Lymphatic transport would also not be expected to lead to signal in the spinal cord.

We assessed the role of transvascular distribution through the blood pool by injecting TTc directly intravenously, rather than our regular intramuscular route. We found intravenous TTc signal to be weak and diluted, with no specific spine uptake (see Figure 2C). This led us to discount diffusion through blood and body fluids to the cord as a mechanism for TTc accumulation in the spine.

Figure 5.

The rate of nerve transport. A, Ex vivo images of dissected nerves taken (at 20 and 60 minutes) following calf muscle injections of 30 μg nontoxic tetanus toxin C fragment (TTc) (lower panels) or human serum albumin (HSA) (upper panels). A fluorescent front was identified for TTc (yellow arrows), advancing over time from the muscle to the cord. No such front was present for HSA. B, Scatterplot showing the distance of the front along the sciatic nerve as a function of time. The mean speed of TTc migration was 12.1 mm/h as indicated by the slope of the linear regression.

Having discounted both lymphatic and blood transport, and having observed TTc directly in nerve tissues, we conclude that nerve transport is the most reasonable explanation of our observations. Similar findings were found for intact tetanus toxin by previous investigators using histologic methods.³⁸

Nerve Imaging Applications

An imaging agent for neurography would fill an important gap in our imaging armamentarium as no such agent is currently available. The lack of such an agent has hampered the assessment of nerve anatomy and function. Many authors have worked previously to improve neurographic imaging techniques.^{39

55} There have been calls for a neurographic agent to be developed,¹ and some efforts toward such an agent have previously been undertaken based on the slow axonal transport of magnetic nanoparticles.^{56
^–
59}

We believe that TTc or its derivatives can fill this gap in our armamentarium and be developed to the point where we can (a) demonstrate the anatomy of nerves (eg, during surgery where nerves are at risk) and/or (b) show the functional status of nerve uptake and transport (eg, during pressureinduced neuropathy).

This should complement and augment currently available measures of nerve function (clinical examination and electrophysiology) and anatomy (mainly MRI). Having a molecular imaging agent available could significantly strengthen the contribution of imaging in the assessment of neuropathy.

It is likely that nerve imaging will access new information, not measured by current techniques, once this technology is developed sufficiently. There are indications that retrograde transport is a primary pathologic event in many neuropathies, including diabetes,^4,5 neurotoxic chemotherapy agents,⁶ compressive neuropathy,^7,8 amyotrophic lateral sclerosis,⁹ and others. Having a means to directly assess the nerve transport function could potentially provide information of earlier phases of the pathogenesis of neuropathy and shed new light on the nature of these diseases. This, in turn, will allow (a) earlier and more successful diagnosis, (b) interventions, and (c) more effective monitoring of therapies.

Future Studies

Future studies could focus on TTc in its current form as a fluorescent agent or extend applications to other imaging modalities.

Fluorescent nerve transport imaging in its current form could be applied in experimental animal settings to perform imaging or neuroscience research, but applications in humans would likely be limited to intraoperative settings for surgeries where nerves are at risk.

Many more human applications would be enabled if imaging modalities such as MRI, nuclear scintigraphy, and positron emission tomography were able to interrogate nerve function and anatomy. Efforts to further develop TTc in this manner are under way.

Studies on the metabolism, clearance, and toxicology of the agent will also be needed in the future.

Conclusion

We have demonstrated retrograde axonal transport by noninvasive in vivo fluorescent imaging in animals using a molecular imaging agent based on TTc. This imaging technology has the potential to open up new avenues of research into nerve function and anatomy and could be applied both in the research setting to better our understanding of neuropathies and in improving the clinical care of patients at risk of nerve damage. Neurography, analogous to the early days of angiography, awaits the development of suitable molecular imaging agents to enable its further development; we trust that this study will serve as a contribution to that end.

Footnotes

Acknowledgments

We thank David Bier for the artwork in .

Financial disclosure of authors: The Mike Hogg Award and M. D. Anderson Internal Research Grant (both to Dr. Schellingerhout) provided funding support for this work, along with the New Program Development fund of the Department of Experimental Diagnostic Imaging.

References

Filler

. Axonal transport and MR imaging: prospects for contrast agent development. J Magn Reson Imaging 1994;4:259–67.

Levi-Montalcini

Hamburger

. Selective growth stimulating effects of mouse sarcoma on the sensory and sympathetic nervous system of the chick embryo. J Exp Zool 1951;116:321–61.

Hamburger

. The history of the discovery of the nerve growth factor. J Neurobiol 1993;24:893–7.

Tomlinson

Fernyhough

Diemel

. Neurotrophins and peripheral neuropathy. Philos Trans R Soc Lond B Biol Sci 1996;351:455–62.

Abbate

Atkinson

Breuer

. Amount and speed of fast axonal transport in diabetes. Diabetes 1991;40:111–7.

Russell

Windebank

McNiven

. Effect of cisplatin and ACTH4-9 on neural transport in cisplatin induced neurotoxicity. Brain Res 1995;676:258–67.

Gallant

. The direct effects of graded axonal compression on axoplasm and fast axoplasmic transport. J Neuropathol Exp Neurol 1992;51:220–30.

Kitao

Hirata

Morita

. Transient damage to the axonal transport system without Wallerian degeneration by acute nerve compression. Exp Neurol 1997;147:248–55.

Munoz

Greene

Perl

Selkoe

. Accumulation of phosphorylated neurofilaments in anterior horn motoneurons of amyotrophic lateral sclerosis patients. J Neuropathol Exp Neurol 1988;47:9–18.

10.

Helting

Zwisler

. Structure of tetanus toxin. I. Break down of the toxin molecule and discrimination between polypeptide fragments. J Biol Chem 1977;252:187–93.

11.

Helting

Zwisler

Wiegandt

. Structure of tetanus toxin. II. Toxin binding to ganglioside. J Biol Chem 1977;252:194–8.

12.

Sjostrand

. Fast and slow components of axoplasmic transport in the hypoglossal and vagus nerves of the rabbit. Brain Res 1970;18:461–7.

13.

Willard

Cowan

Vagelos

. The polypeptide composition of intra axonally transported proteins: evidence for four transport velocities. Proc Natl Acad Sci U S A 1974;71:2183–7.

14.

Ramos

Abreu

Nascimento

. A high-copy T7 Escherichia coli expression vector for the production of recombinant proteins with a minimal N-terminal His-tagged fusion peptide. Braz J Med Biol Res 2004;37:1103–9.

15.

Eisel

Jarausch

Goretzki

. Tetanus toxin: primary structure, expression in E. coli, and homology with botulinum toxins. Embo J 1986;5:2495–502.

16.

Fairweather

Lyness

Pickard

. Cloning, nucleotide sequencing, and expression of tetanus toxin fragment C in Escherichia coli. J Bacteriol 1986;165:21–7.

17.

Knapp

Segelke

Rupp

. The 1.61 angstrom structure of the tetanus toxin ganglioside binding region: solved by Mad and Mir phase combination. Proceedings of the Am. Crystallogr. Assoc. Abstr. Papers (Annu. Meeting); Washington DC: July 19–24, 1998.

18.

Fotinou

Emsley

Black

. The crystal structure of tetanus toxin Hc fragment complexed with a synthetic GT1b analogue suggests cross-linking between ganglioside receptors and the toxin. J Biol Chem 2001;276:32274–81.

19.

Copeland

JWR

Nasiadka

Dietrich

Krause

. Patterning of the Drosophila embryo by a homeodomaindeleted Ftz polypeptide. Nature 1996;379:162–5.

20.

Herreros

Lalli

Montecucco

Schiavo

. Tetanus toxin fragment C binds to a protein present in neuronal cell lines and motoneurons. J Neurochem 2000;74:1941–50.

21.

O'Brien

Snyder

. Position-sensitive detector studies of the axonal transport of a pulse of radioisotope. J Neurobiol 1982;13:435–45.

22.

Brown

. Slow axonal transport: stop and go traffic in the axon. Nat Rev Mol Cell Biol 2000;1:153–6.

23.

Lasek

Garner

Brady

. Axonal transport of the cytoplasmic matrix. J Cell Biol 1984;99:212s–21s.

24.

Lubinska

Niemierko

. Velocity and intensity of bidirectional migration of acetylcholinesterase in transected nerves. Brain Res 1971;27:329–42.

25.

Ochs

. Fast transport of materials in mammalian nerve fibers. Science 1972;176:252–60.

26.

Ranish

Ochs

. Fast axoplasmic transport of acetylcholinesterase in mammalian nerve fibres. J Neurochem 1972;19:2641–9.

27.

Fishman

Savitt

Farrand

. Enhanced CNS uptake of systemically administered proteins through conjugation with tetanus C-fragment. J Neurol Sci 1990;98:311–25.

28.

Evinger

Erichsen

. Transsynaptic retrograde transport of fragment C of tetanus toxin demonstrated by immunohistochemical localization. Brain Res 1986;380:383–8.

29.

Fishman

Carrigan

. Retrograde transneuronal transfer of the C-fragment of tetanus toxin. Brain Res 1987;406:275–9.

30.

Fishman

Savitt

. Transsynaptic transfer of retrogradely transported tetanus protein-peroxidase conjugates. Exp Neurol 1989;106:197–203.

31.

Stockel

Schwab

Thoenen

. Comparison between the retrograde axonal transport of nerve growth factor and tetanus toxin in motor, sensory and adrenergic neurons. Brain Res 1975;99:1–16.

32.

Sjostrand

Hansson

. Effect of colchicine on the transport of axonal protein in the retinal ganglion cells of the rat. Exp Eye Res 1971;12:261–9.

33.

Schnapp

Reese

. New developments in understanding rapid axonal transport. Trends Neurosci 1986;9:155–62.

34.

Vale

Reese

Sheetz

. Identification of a novel force-generating protein, kinesin, involved in microtubulebased motility. Cell 1985;42:39–50.

35.

Cyr

Brady

. Molecular motors in axonal transport: cellular and molecular biology of kinesin. Mol Neurobiol 1992;6:137–55.

36.

Schnapp

Reese

. Dynein is the motor for retrograde axonal transport of organelles. Proc Natl Acad Sci U S A 1989;86:1548–52.

37.

Vallee

Bloom

. Mechanisms of fast and slow axonal transport. Annu Rev Neurosci 1991;14:59–92.

38.

Price

Griffin

Young

. Tetanus toxin: direct evidence for retrograde intraaxonal transport. Science 1975;188:945–7.

39.

Chopra

Morris

Friedman

Mendenhall

. Radiotherapy and radiosurgery for benign neurofibromas. Am J Clin Oncol 2005;28:317–20.

40.

Dailey

Tsuruda

Filler

. Magnetic resonance neurography of peripheral nerve degeneration and regeneration. Lancet 1997;350:1221–2.

41.

Dailey

Tsuruda

Goodkin

. Magnetic resonance neurography for cervical radiculopathy: a preliminary report. Neurosurgery 1996;38:488–92, discussion 492.

42.

Filler

. Piriformis and related entrapment syndromes: diagnosis & management. Neurosurg Clin North Am 2008;19:609–22, vii.

43.

Filler

. Diagnosis and treatment of pudendal nerve entrapment syndrome subtypes: imaging, injections, and minimal access surgery. Neurosurg Focus 2009;26:E9.

44.

Filler

Haynes

Jordan

. Sciatica of nondisc origin and piriformis syndrome: diagnosis by magnetic resonance neurography and interventional magnetic resonance imaging with outcome study of resulting treatment. J Neurosurg Spine 2005;2:99–115.

45.

Filler

Howe

Hayes

. Magnetic resonance neurography. Lancet 1993;341:659–61.

46.

Filler

Kliot

Howe

. Application of magnetic resonance neurography in the evaluation of patients with peripheral nerve pathology. J Neurosurg 1996;85:299–309.

47.

Filler

Maravilla

Tsuruda

. MR neurography and muscle MR imaging for image diagnosis of disorders affecting the peripheral nerves and musculature. Neurol Clin 2004;22:643–82, vi–vii.

48.

Friedman

Elias

Midha

. Introduction: peripheral nerve surgery: biology, entrapment, and injuries. Neurosurg Focus 2009;26:2.

49.

Howe

Filler

Bell

Griffiths

. Magnetic resonance neurography. Magn Reson Med 1992;28:328–38.

50.

Howe

Saunders

Filler

. Magnetic resonance neurography of the median nerve. Br J Radiol 1994;67:1169–72.

51.

Koenig

Pedro

Heinen

CPG

. High-resolution ultrasonography in evaluating peripheral nerve intrapment and trauma. Neurosurg Focus 2009;26:6.

52.

Kuntz

Blake

Britz

. Magnetic resonance neurography of peripheral nerve lesions in the lower extremity. Neurosurgery 1996;39:750–6, discussion 756–7.

53.

Moore

Tsuruda

Dailey

. The value of MR neurography for evaluating extraspinal neuropathic leg pain: a pictorial essay. AJNR Am J Neuroradiol 2001;22:786–94.

54.

Spinner

Atkinson

Wenger

Stuart

. Tardy sciatic nerve palsy following apophyseal avulsion fracture of the ischial tuberosity. Case report. J Neurosurg 1998;89:819–21.

55.

Wong

Amrami

Wenger

. Lipomatosis of the sciatic nerve: typical and atypical MRI features. Skel Radiol 2006;35:180–4.

56.

Enochs

Schaffer

Bhide

. MR imaging of slow axonal transport in vivo. Exp Neurol 1993;123:235–42.

57.

Enochs

Weissleder

. MR imaging of the peripheral nervous system. J Magn Reson Imaging 1994;4:251–7.

58.

Petropoulos

Schaffer

Cheney

. MR imaging of neuronal transport in the guinea pig facial nerve: initial findings. Acta Otolaryngol (Stockh) 1995;115:512–6.

59.

van Everdingen

Enochs

Bhide

. Determinants of in vivo MR imaging of slow axonal transport. Radiology 1994;193:485–91.

Fluorescence Imaging of Fast Retrograde Axonal Transport in Living Animals

Abstract

Materials and Methods

Synthesis, Labeling, and Characterization of TTc

In Vivo and Ex Vivo Imaging

Histologic Analysis and Fluorescence Confocal Microscopy

Rate of Transport

Results

Synthesis, Labeling, and Characterization of TTc

In Vivo Imaging

Ex Vivo Imaging

Histologic Analysis and Fluorescence Confocal Microscopy

Rate of Transport

Discussion

TTc in the Literature

Limitations of the Study

Alternative Transport Hypotheses

Nerve Imaging Applications

Future Studies

Conclusion

Footnotes

Acknowledgments

References