Effects of sulfatide on peripheral nerves in metachromatic leukodystrophy

Abstract

Objective: To evaluate the longitudinal correlations between sulfatide/lysosulfatide levels and central and peripheral nervous system function in children with metachromatic leukodystrophy (MLD) and to explore the impact of intravenous recombinant human arylsulfatase A (rhASA) treatment on myelin turnover.

Methods: A Phase 1/2 study of intravenous rhASA investigated cerebrospinal fluid (CSF) and sural nerve sulfatide levels, 88-item Gross Motor Function Measure (GMFM-88) total score, sensory and motor nerve conduction, brain N-acetylaspartate (NAA) levels, and sural nerve histology in 13 children with MLD. Myelinated and unmyelinated nerves from an untreated MLD mouse model were also analyzed.

Results: CSF sulfatide levels correlated with neither Z-scores for GMFM-88 nor brain NAA levels; however, CSF sulfatide levels correlated negatively with Z-scores of nerve conduction parameters, number of large (≥7 μm) myelinated fibers, and myelin/fiber diameter slope, and positively with nerve g-ratios and cortical latencies of somatosensory-evoked potentials. Quantity of endoneural litter positively correlated with sural nerve sulfatide/lysosulfatide levels. CSF sulfatide levels decreased with continuous high-dose treatment; this change correlated with improved nerve conduction. At 26 weeks after treatment, nerve g-ratio decreased by 2%, and inclusion bodies per Schwann cell unit increased by 55%. In mice, abnormal sulfatide storage was observed in non-myelinating Schwann cells in Remak bundles of sciatic nerves but not in unmyelinated urethral nerves.

Interpretation: Lower sulfatide levels in the CSF and peripheral nerves correlate with better peripheral nerve function in children with MLD; intravenous rhASA treatment may reduce CSF sulfatide levels and enhance sulfatide/lysosulfatide processing and remyelination in peripheral nerves.

Figure 1

Intravenous rhASA study (A) design and (B) assessments. Doses were adjusted monthly in the Phase 1/2 study and every 6 weeks in the extension to account for body weight changes. ^aChildren initially received a single dose of 25 U/kg, then 50 U/kg for all doses thereafter. ^bChildren who received 50 U/kg in the Phase 1/2 study were assigned 1:1 to treatment with either 100 or 200 U/kg in the extension. ^cThe extension was terminated because of lack of efficacy after 104 weeks (cumulative exposure of 156 weeks, including the Phase 1/2 trial). CSF, cerebrospinal fluid; EOW, every other week; GMFM‐88, 88‐item Gross Motor Function Measure; NAA, N‐acetylaspartate; rhASA, recombinant human arylsulfatase A; U, units.

Figure 2

Z‐scores from baseline to 78 weeks for (A) GMFM‐88 total score and (B) NAA levels in the centrum semiovale. (C) Relationship between Z‐scores for GMFM‐88 and NAA. (D) Relationship between CSF sulfatide levels and Z‐scores for GMFM‐88. (E) Relationship between CSF sulfatide levels and Z‐scores for NAA. CSF, cerebrospinal fluid; GMFM‐88, 88‐item Gross Motor Function Measure; NAA, N‐acetylaspartate.

Figure 3

Relationships between CSF sulfatide levels and (A) combined CMAP and SNAP amplitude Z‐scores in the abductor pollicis brevis, (B) combined MNCV and SNCV Z‐scores in the abductor pollicis brevis, (C) cortical latencies of SSEPs in the median nerve, and (D) cortical latencies of SSEPs in the tibial nerve. Relationship between the change in CSF sulfatide and (E) the change in combined CMAP and SNAP amplitudes and (F) the change in combined MNCV and SNCV. CMAP, compound muscle action potential; CSF, cerebrospinal fluid; MNCV, motor nerve conduction velocity; SNAP, sensory nerve action potential; SNCV, sensory nerve conduction velocity; SSEP, somatosensory‐evoked potential.

Figure 4

Relationship between the number of large (≥7 μm) myelinated sural nerve fibers and (A) CSF sulfatide levels and (B) sural nerve sulfatide levels. CSF, cerebrospinal fluid.

Figure 5

Relationships between sulfatide levels and peripheral nerve morphometry. (A) Relationship between CSF sulfatide levels and sural nerve g‐ratios. (B) Relationship between CSF sulfatide levels and myelin thickness versus fiber diameter slope. (C) Relationship between sural nerve sulfatide levels and endoneural litter. (D) Relationship between sural nerve lysosulfatide levels and endoneural litter. CSF, cerebrospinal fluid.

Figure 6

(A) Change in nerve g‐ratios for individual patients from baseline to 26 weeks of treatment. (B) Change in inclusion bodies per non‐myelinating Schwann cell unit from baseline to 26 weeks of treatment.

Figure 7

Example sural nerve electron micrographs from an individual patient in (A) the right sural nerve at baseline and (B–D) the left sural nerve at 26 weeks. The thick arrows in panels C and D indicate inclusion bodies in Remak bundles; the thin arrows indicate endoneural litter. Scale bar indicates 2 μm.

Figure 8

Example electron micrographs from (A) the sciatic nerve of the tg/ASA(+/−) (control) mouse, (B) the sciatic nerve of the tg/ASA(−/−) mouse, (C) the unmyelinated fibers within the urethra of the tg/ASA(+/−) (control) mouse, and (D) the unmyelinated fibers within the urethra of the tg/ASA(−/−) mouse. The black arrows on panel B indicate Remak‐bundle Schwann cells containing abnormal sulfatide storage material. The white arrow on panel B indicates myelinating Schwann cells containing abnormal sulfatide storage material. The black arrows on panels C and D indicate Remak bundles. Scale bar in panels A and B indicates 2 μm. Scale bar in panels C and D indicates 500 nm. ASA, arylsulfatase A; tg, transgenic.