Schwann cells regulate sensory neuron gene expression before and after peripheral nerve injury

Abstract

Sensory neurons in the PNS demonstrate substantial capacity for regeneration following injury. Recent studies have identified changes in the transcriptome of sensory neurons, which are instrumental for axon regeneration. The role of Schwann cells (SCs) in mediating these changes remains undefined. We tested the hypothesis that SCs regulate expression of genes in sensory neurons before and after PNS injury by comparing mice in which LDL Receptor-related Protein-1 (LRP1) is deleted in SCs (scLRP1^-/- mice) with wild-type (scLRP1^+/+ ) littermates. LRP1 is an endocytic and cell-signaling receptor that is necessary for normal SC function and the SC response to nerve injury. scLRP1^-/- mice represent a characterized model in which the SC response to nerve injury is abnormal. Adult DRG neurons, isolated from scLRP1^-/- mice, with or without a conditioning nerve lesion, demonstrated increased neurite outgrowth when cultured ex vivo, compared with neurons from wild-type mice. Following sciatic nerve crush injury, nerve regeneration was accelerated in vivo in scLRP1^-/- mice. These results were explained by transcriptional activation of RAGs in DRG neurons in scLRP1^-/- mice prior to nerve injury. Although the presence of abnormal SCs in scLRP1^-/- mice primed DRG neurons for repair, nerve regeneration in scLRP1^-/- mice resulted in abnormalities in ultrastructure, principally in Remak bundles, and with the onset of neuropathic pain. These results demonstrate the importance of SCs in controlling RAG expression by neurons and the potential for this process to cause chronic pain when abnormal. The SC may represent an important target for preventing pain following PNS injury.

Keywords: DRG; LRP1; Schwann cell; axonal growth; pain; peripheral nerve; regeneration associated genes (RAGs).

Figure 1. Genetic deletion of LRP1 in Schwann cells accelerates functional sensory recovery

(A) Cutaneous innervation was assessed by pinprick scores of right hind paw post sciatic nerve crush from scLRP1^+/+ and scLRP1^−/− mice. (B) No difference in toe spreading reflex was evident between scLRP1^+/+ and scLRP1^−/− mice. Scores were measured up to 23 days post crush. Data are mean±SEM (n=6–11 mice per genotype, scLRP1^+/+ and scLRP1^−/−). *p<0.05, **p<0.01 repeated measures ANOVA and Schefe post hoc test.

Figure 2. Conditional deletion of LRP1 in Schwann cells enhances early axonal growth after crush injury in vivo

(A) STMN2 (SCG10) immunoreactivity in longitudinal sections of sciatic nerve 72 h after crush injury in scLRP1^+/+ and scLRP1^−/− mice. Distance from the nerve injury site (arrow) at 5 mm (*) and 10 mm (**). Scale bar: 1 mm. (B) Immunoblot of growth cone markers, GAP43 and STMN2 and loading control, (tERK1/2) in nerve sections 5–10mm from the crush sites after 72 h. (C) Quantification of growth cone markers. Data represent mean±SEM (n=3 individual mice; *p<0.05; **p≤0.01).

Figure 3. Neuronal cytoskeletal components, Neurofilament 200 (NF200) and βIII neuronal tubulin (Tuji), are increased, yet p75NTR, a SC nerve repair marker, is decreased after crush injury in scLRP1^−/− mice

(A) NF200 immunoreactivity in transverse sections of distal sciatic nerve (5 and10 mm) one week after crush. Scale bar: 50 μM. (B) Quantitation of NF200-immunoreactive axons in 5-μm-thick transverse sciatic nerve sections (n=8/group; **p<0.01). (C) Representative immunoblot analysis of neuronal βIII-tubulin, NFM, p75^NTR and GFAP one week post injury. GAPDH is used as a loading control. (D) Densitometric measurement of immunoblot in (C) βIII-tubulin, NFM, p75^NTR and GFAP normalized to GAPDH one week post injury. Data represent mean±SEM (n=4 mice per genotype; *p<0.05; *** p<0.001).

Figure 4. scLRP1^−/− DRG neurons demonstrate increased neurite outgrowth in vitro

Adult DRG neurons were cultured for 24 hours from either scLRP1^+/+ or scLRP1^−/− mice that were uninjured or injured (sciatic nerve crush) 3 days prior to plating. (A) Representative images of DRG neurons from uninjured scLRP1^+/+ mice (left, upper panel) and scLRP1^−/− mice (left, lower panel) and after preconditioning injury, scLRP1^+/+ mice (right, upper panel) and scLRP1^−/− mice (right, lower panel). Scale bar 200 μm. Quantification of βIII-tubulin immunopositivity shows total neurite length per cell in (B) uninjured and (C) fold increase in injured DRG neurons from scLRP1^+/+ and scLRP1^−/− mice after 24 hours in vitro. Data represent mean±SEM (n=2–8 mice per genotype and 3 wells per condition; *p<0.05 two tailed T-test).

Figure 5. qPCR analysis of dorsal root ganglions (DRGs) from scLRP1^−/− and scLRP1^+/+ mice show upregulation of regeneration associated genes (RAGs) before and after injury

Taqman qPCR analysis of (A) uninjured DRGs showing CREB3, ATF3, STAT3 and SPRR1A in scLRP1^−/− mice (n=3–5 mice per genotype; *p<0.05; **p<0.01, two tailed t-test). (B) Immunoblot showing increased βIII neuronal tubulin in scLRP1^−/− mice. GAPDH is used as a loading control. Quantification of increased βIII neuronal tubulin in scLRP1^−/− mice (n=2–3 mice per genotype; **p<0.01, two tailed t-test). (C) Fold change of CREB3 mRNA in DRGs before and 3 days post sciatic nerve crush injury. mRNA expression in scLRP1^−/− and scLRP1^+/+ DRGs is normalized to uninjured scLRP1^+/+ expression. Data represent mean±SEM. (n=3–5 mice per genotype; *p<0.05, ***p<0.001 one-way ANOVA with a Neuman Keuls post hoc analysis).

Figure 6. Aberrant Remak bundle formation in scLRP1^−/− distal nerve after sciatic nerve crush injury

Transverse section of nerve showing unmyelinated axons (Remak bundle) three weeks after crush injury. Each of the individual axons in this bundle is compartmentalized by a mesaxonal process (mes) provided by the central Schwann cells. Segments of myelinated axons (my) are visible in the corners. 1900 X. Bundles of non-myelinated axons including those manifesting a neuroma phenotype are identified by * (axon that appose each other; axon sprouts) are shown in scLRP1^−/− mice. Although the Schwann cell has produced some mesaxonal cytoplasm, many axons are in close opposition (asterisks). 4800 X. Quantification of (B) the axon numbers in Remak bundles (n=3–4 mice per genotype; *p<0.05, **p<0.01 one-way ANOVA with Tukeys post hoc); (C) total number of unmyelinated axons; (D) total number of myelinated axons; and (E) total number of DRG neurons. Data represent mean±SEM (n=3–4 mice per genotype; *p<0.05, **p<0.01, two tailed T-test).

Figure 7

Thermal and tactile allodynia develop in scLRP1^−/− mice. (A) Thermal withdrawal latencies and (B) tactile withdrawal thresholds were evaluated in female scLRP1^+/+ and scLRP1^−/− mice over 4 weeks. Data represent mean±SEM (n=4–7 mice group *p<0.05, **p<0.01 compared to sham control and naïve by two way repeated ANOVA followed by a Scheffe’s test).