Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Dec 15;31(24):1955-66.
doi: 10.1089/neu.2014.3355. Epub 2014 Oct 16.

Human NgR-Fc Decoy Protein via Lumbar Intrathecal Bolus Administration Enhances Recovery From Rat Spinal Cord Contusion

Affiliations
Free PMC article

Human NgR-Fc Decoy Protein via Lumbar Intrathecal Bolus Administration Enhances Recovery From Rat Spinal Cord Contusion

Xingxing Wang et al. J Neurotrauma. .
Free PMC article

Abstract

Axonal growth and neurological recovery after traumatic spinal cord injury (SCI) is limited by the presence of inhibitory proteins in myelin, several of which act via the NgR1 protein in neurons. A truncated soluble ligand-binding fragment of NgR1 serves as a decoy and promotes recovery in acute and chronic rodent SCI models. To develop the translational potential of these observations, we created a human sequence-derived NgR1(310)-Fc protein. This protein is active in vitro. When the human NgR1 decoy is administered by continuous intracerebroventricular infusion to rats with a spinal contusion injury at doses of 0.09-0.53 mg/kg/d, neurological recovery is improved. Effective doses double the percentage of rats able to bear weight on their hindlimbs. Next, we considered the half-life and distribution of NgR1(310)-Fc after bolus delivery to the lumbar intrathecal space. The protein is found throughout the neuraxis and has a tissue half-life of approximately 2 days in the rat, and 5 days in the nonhuman primate. At an intermittent, once every 4 day, lumbar bolus dosing schedule of 0.14 mg/kg/d, NgR1(310)-Fc promoted locomotor rat recovery from spinal cord contusion at least as effectively as continuous infusion in open field and grid walking tasks. Moreover, the intermittent lumbar NgR1(310)-Fc treatment increased the growth of raphespinal axons into the lumbar spinal cord after injury. Thus, human NgR1(310)-Fc provides effective treatment for recovery from traumatic SCI in this preclinical model with a simplified administration regimen that facilitates clinical testing.

Keywords: Nogo; Nogo receptor; contusion; myelin; pharmacokinetics; translational science; traumatic spinal cord injury.

Figures

<b>FIG. 1.</b>
FIG. 1.
Production and purification of human NgR1(310)-Fc. (A) Size exclusion chromatography of NgR1(310)-Fc after purification (lot 15). The absorbance at 280 nm is plotted as a function of elution time from an ACQUITY UPLC BEH200 SEC column using a sodium phosphate/sodium chloride buffer system. The average peak retention times for thyroglobulin (670 kDa), γ-globulin (158 kDa), bovine serum albumin (67 kDa), myoglobin (16.5 kDa), and uracil (0.112 kDa) are indicated. (B) Purified NgR1(310)-Fc protein was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and stained with Coomassie Blue. The boxed regions were used to calculate a purity for the major band at 66 kDa of 97%. The migration of molecular weight standards is at right. (C) Purified NgR1(310)-Fc protein was analyzed by SDS-PAGE under non-reducing conditions.
<b>FIG. 2.</b>
FIG. 2.
Human NgR1(310)-Fc binds myelin ligands and blocks outgrowth inhibition. (A–C) Streptavidin or Strep-Tactin plates coated with biotin-Nogo-66 (A) or myelin-associated glycoprotein (MAG) (B) or myelin glycoprotein (OMgp) (C) were incubated with the indicated concentrations of NgR1(310)-Fc, washed, and bound human Fc was detected with secondary peroxidase-labeled antibody. The absorbance of the peroxidase reaction is plotted as a function of NgR1(310)-Fc added. (D) Mouse E18 cortical neurons were cultured until a mature lawn of neurites formed. Then the cultures were scrape-injured, and regeneration over 5 days was measured in the presence of Nogo-22 kDa protein, NgR1(310)-Fc protein, or both. The inhibitory effect of Nogo is blocked by an excess of NgR1(310)-Fc. Data are mean±standard error of the mean from three separate experiments. Color image is available online at www.liebertpub.com/neu
<b>FIG. 3.</b>
FIG. 3.
Human NgR1(310)-Fc dose response for locomotor recovery after rat spinal cord contusion. (A) Three days after the contusion and intracerebroventricular (i.c.v.) cannula implantation, rats were assigned randomly to one of three treatment groups. The osmotic pumps filled with 0.66 mg human hNgR1(310)-Fc (0.09 mg/kg/d, medium dose), 4.0 mg human hNgR1(310)-Fc (0.53 mg/kg/d, high dose) or equal volume of 2 mL phosphate buffered saline (PBS) (vehicle). The duration of treatment was 4 weeks, and follow-up was through 9–10 weeks. The Basso, Beattie, and Bresnahan (BBB) scores are plotted as a function of time after injury. Data are mean±standard error of the mean (SEM). ** p=0.007 for the effect of medium dose NgR-Fc versus vehicle treatment, and * p=0.046 for the effect of high dose NgR-Fc versus vehicle treatment by repeated measures analysis of variance (RM-ANOVA) from 14–63 days for post hoc pairwise comparisons with least significant difference (LSD) correction. At specific time points, one-way ANOVA for post hoc comparison with LSD correction of medium dose NgR-Fc versus vehicle, * p<0.05; ** p<0.005. (B) Three days after the contusion and i.c.v. cannula implantation, rats were assigned randomly to one of two treatment groups. The osmotic pumps filled with 0.10 mg hNgR1(310)-Fc (0.014 mg/kg/d, low dose), or equal volume of 2 mL PBS as vehicle group. The duration of treatment was 4 weeks and the follow-up was 8 weeks. Data are mean±SEM; n=5 for vehicle and n=11 for low dose NgR1-Fc group, p=0.29, not significant (n.s.) by RM-ANOVA for the effect of low dose NgR1 decoy versus vehicle treatment. (C) The BBB scores at day 56 from A and B are replotted as a function of hNgR1(310)-Fc dose. Data are mean±SEM. By one-way ANOVA with pairwise post hoc comparisons and LSD correction, the high dose (* p=0.025) and medium dose (** p=0.0076) NgR1-Fc treatment groups had better behavioral outcome compared with the vehicle group. Color image is available online at www.liebertpub.com/neu
<b>FIG. 4.</b>
FIG. 4.
Recovery of weight bearing after hNgR-Fc treatment of spinal cord contused rats. The percentage of rats capable of body weight support with at least one hindlimb is reported for the cohort described in the figure at 35 or 63 days after contusion injury. * p<0.05 for the indicated comparisons; analysis of variance with post hoc pairwise comparisons and least significant difference correction. Color image is available online at www.liebertpub.com/neu
<b>FIG. 5.</b>
FIG. 5.
Recovery of grid walking after hNgR-Fc treatment of spinal cord contused rats. Spinal cord contused rats from the cohort described in Fig. 4 were assessed for ability to walk a 50 cm distance over a horizontal wire grid at 63 days after injury. (A) Successful hindlimb (HL) steps (steps without the foot slipping below the grid) were counted from walking over a 50 cm distance. Data are mean±standard error of the mean (SEM), for vehicle (n=11), 0.09 mg/kg/d (n=15), and 0.53 mg/kg/d (n=14). One-way analysis of variance (ANOVA) with pairwise post hoc comparisons with least significant difference (LSD) correction, *** p<0.0001. (B) Grid walking HL versus forelimb (FL) success ratio was evaluated as successful HL step number divided by FL number. Data are mean±SEM, for vehicle (n=11), 0.09 mg/kg/d (n=15), and 0.53 mg/kg/d (n=14). One-way ANOVA with pairwise post hoc comparisons with LSD correction, *** p<0.0001. (C) Grid walking HL foot fault ratio was scored as the HL foot faults divided by the number of HL steps. Data are mean±SEM, for vehicle (n=9), 0.09 mg/kg/d (n=14), and 0.53 mg/kg/d (n=14). Two vehicle rats and one 0.09 mg/kg/d rat were excluded from this analysis because the HL was dragging rather than mis-stepping. One-way ANOVA with pairwise post hoc comparisons with LSD correction, * p<0.05; ** p<0.01. Color image is available online at www.liebertpub.com/neu
<b>FIG. 6.</b>
FIG. 6.
Detection of human NgR-Fc protein in central nervous system tissue after intermittent lumbar intrathecal administration. (A) The standard curve for sandwich enzyme-linked immunosorbent assay detection for known concentrations of hNgR1(310)-Fc diluted into brain extract demonstrates detection by europium-conjugated reagent. (B–D) Pharmacokinetics of brain and spinal cord tissue hNgR1(310)-Fc level after a single dose of 100 μg hNgR1(310)-Fc in 20 μL administered via an intrathecal lumbar (IT-L) catheter to the rat. Values for cervical, thoracic, and lumbar spinal cord tissue were similar and averaged together for each rat. Data are mean±standard error of the mean (SEM) for n=3 rats for each time. The time course from 1–168 h is plotted in B, the early time period in 1–12 h in C, and 12–168 h in D. (E) Brain and spinal cord levels of tissue hNgR1(310)-Fc after five separate 15 μL IT-L doses of 150 μg hNgR1(310)-Fc or vehicle alone spaced 4 days apart, over a 16-day course. The rats were sacrificed and tissue analyzed 24–28 h after the last dose on the 17th day. Data are mean±SEM for n=5 rats. (F) Pharmacokinetics of brain and spinal cord tissue hNgR1(310)-Fc level after administration via an IT-L catheter to nonhuman primate. Individual animals were dosed with 2.0 mg hNgR1(310)-Fc via a catheter placed in the IT-L space, and then the animals were sacrificed and tissue collected at the indicated time points after bolus administration. Values for cerebral cortex and three regions for spinal cord were measured for each animal and averaged together. N=1 monkey at each time point. Color image is available online at www.liebertpub.com/neu
<b>FIG. 7.</b>
FIG. 7.
Intermittent intrathecal lumbar (IT-L) administration of hNgR1(310)-Fc improves locomotor recovery after rat spinal cord contusion. Three days after spinal cord contusion and IT-L catheter insertion, rats were randomly assigned to one of two treatment groups and received seven bolus doses spaced 4 days apart (arrows). One group (n=14) received hNgR1(310)-Fc (15 μL containing 0.15 mg protein), and the other group (n=13) received an equal volume of phosphate buffered saline vehicle. The locomotor Basso, Beattie, and Bresnahan (BBB) scores are reported as a function of time. Data are mean±standard error of the mean. The vehicle and hNgR1(310)-Fc-treated groups are significantly different; repeated measures analysis of variance (RM-ANOVA) from 35–63 days, * p=0.031. At specific time points, one-way ANOVA for NgR-Fc versus vehicle, * p≤0.05. Color image is available online at www.liebertpub.com/neu
<b>FIG. 8.</b>
FIG. 8.
Intermittent lumbar hNgR1(310)-Fc improves weight bearing and grid walking after rat spinal cord contusion. (A) The percentage of rats capable of body weight support with at least one hindlimb (HL) is reported for the cohort described in Fig. 7 at 28 or 56 days after contusion injury. Data are mean±standard error of the mean (SEM). * p=0.020 for the indicated comparison, analysis of variance. (B) Spinal cord contused rats from the cohort described in Fig. 7 were assessed for ability to walk a 50 cm distance over a horizontal wire grid at 56 days after injury. Successful HL steps (steps without the foot slipping below the grid) were counted. Data are mean±SEM. The hNgR1(310)-Fc treated group (n=9) had significantly more successful steps than the vehicle group (n=11). * p=0.01, Student two-tailed t test. Color image is available online at www.liebertpub.com/neu
<b>FIG. 9.</b>
FIG. 9.
Intermittent lumbar hNgR1(310)-Fc IT-L increases raphespinal serotonergic innervation of the caudal spinal cord after contusion injury. (A–D) Transverse sections of spinal cord for rats treated with vehicle (A, C) or hNgR1(310)-Fc (B, D) as in Fig. 7 were stained with antiserotonin antibodies. Sections containing the ventral horn gray matter are shown at a level 11–16 mm caudal to the contusion site. The images in A and B are standard epifluorescence, and the images in C and D are higher magnification Z-stack confocal images. Note the greater fiber density of B and D versus A and C. Scale bars, 100 μm. (E) Rostral (11–16 mm) 5-hydroxytryptamine (5HT) immunopositive fiber length per unit area (μm/mm2) in the ventral horn was measured from images similar to the caudal images in C and D for the cohort of rats treated as in Fig. 7. Data are mean±standard error of the mean (SEM). (F) Caudal (11–16 mm) 5HT immunopositive fiber length per unit area (μm/mm2) in the ventral horn was measured from images as in C and D for the rats treated as in Fig. 7. Data are mean±SEM. The hNgR1(310)-Fc treated group (n=11) had significantly more serotonergic fibers than the vehicle group (n=11). * p=0.002, Student two-tailed t test. Color image is available online at www.liebertpub.com/neu

Similar articles

See all similar articles

Cited by 12 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback