Brief Communication
Tumor necrosis factor receptor 1 and 2 proteins are differentially regulated during Wallerian degeneration of mouse sciatic nerve

https://doi.org/10.1016/j.expneurol.2004.11.002Get rights and content

Abstract

The pro-inflammatory cytokine tumor necrosis factor-α (TNF) is involved in injury-induced peripheral nerve pathology and in the generation of neuropathic pain. Here, we investigated local protein levels of the two known TNF receptors, TNF receptor 1 and 2 (TNFR1, TNFR2), on days 0, 1, 3, 7, 14, and 28 after unilateral crush or chronic constriction injury (CCI) of mouse sciatic nerves using enzyme-linked immunoassay. Both receptors were detectable at a low level in nerve homogenates from naive mice. After crush or CCI, TNFR1 increased by 2-fold on days 3 and day 7. Unlike TNFR1, TNFR2 was markedly upregulated already on day 1 after crush or CCI. TNFR2 increased by 7-fold on days 3 and 7, and remained elevated at a lower level until day 28 after both CCI and crush injury. These data indicate that endoneurial TNFR1 and TNFR2 proteins are differentially regulated during Wallerian degeneration.

Introduction

Recent experimental studies revealed that pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF), contribute to injury-induced peripheral nerve pathology and to the development of neuropathic pain (for review, see Sommer, 2001, Stoll et al., 2002). TNF induces axonal damage, macrophage recruitment, and ectopic activity in peripheral nerve fibers (Hall et al., 2000, Liefner et al., 2000, Said and Hontebeyrie-Joskowicz, 1992, Schäfers et al., 2003a, Sorkin et al., 1997, Uncini et al., 1999). TNF is rapidly upregulated at the site of peripheral nerve lesion and is believed to act as an initiator of the cytokine network during Wallerian degeneration (LaFleur et al., 1996, Shamash et al., 2002, Wagner and Myers, 1996). The molecular mechanisms that underlie TNF actions after peripheral nerve injury are incompletely known.

TNF binds and triggers two known receptors (TNFRs): the constitutively expressed TNFR1 and the inducible TNFR2. Both are mainly colocalized on all nucleated cell types (for review, see MacEwan, 2002, Vandenabeele et al., 1995). TNFR1 and TNFR2 have been detected on normal rat or human Schwann cells using immunohistochemistry (Bonetti et al., 2000, Skoff et al., 1998), and along rat sciatic nerve axons and on dorsal root ganglion neurons after sciatic nerve injury (Ohtori et al., 2004, Schäfers et al., 2003b, Shubayev and Myers, 2001). The TNFR1 binds preferably soluble TNF whereas TNFR2 binds preferably to membrane-bound TNF molecules (for review, see Vandenabeele et al., 1995). Binding of TNF to TNFR1 is preferentially followed by internalization, whereas TNF binding to TNFR2 is followed by shedding of the ligand–receptor complex. The signal transduction pathways induced intracellularly by the two receptors differ to a great extent (Ledgerwood et al., 1999). Thus, a TNF signal that is mediated by the TNFR1 can elicit effects that are distinct from those mediated by the TNFR2. Indeed, thermal hyperalgesia after chronic constriction injury (CCI) of mouse sciatic nerve depends on TNFR1 only (Sommer et al., 1998). To further dissect TNF-triggered pathways at the site of peripheral nerve injury, it is essential to know how TNFR1 and TNFR2 are regulated after peripheral nerve injury. Therefore, we asked (a) whether local TNFR levels are altered during different stages of Wallerian degeneration and (b) whether TNFR1 and the TNFR2 are differentially regulated. We used enzyme-linked immunoassays to determine TNFR1 and TNFR2 protein in mouse sciatic nerve homogenates after chronic constriction (CCI) or crush injury.

CCI was chosen as a well-established model of neuropathic pain (Bennett and Xie, 1988), which combines nerve injury with perineurial inflammation due to ligatures tied around the nerve. Nerve crush was used as an additional injury model that allows studying local TNFR changes due to local nerve trauma only. Both nerve injury models induce Wallerian degeneration. We used behavioral testing in the experimental animals to be able to compare the temporal course of pain behavior with that of TNFR1.

Female C57BL/6 mice (16–20 g; Harlan–Winkelmann) were used in procedures approved by the Bavarian State animal experimentation committee. CCI was performed as described by Bennett and Xie (1988) with minor modifications (Sommer et al., 1998). Under deep barbiturate anesthesia, three loosely constrictive ligatures were placed around one sciatic nerve at the mid thigh level with 1-mm spacing. Crush injury was performed by pressing the sciatic nerve at the mid thigh level with a jeweler's forceps no. 5 for three successive periods of 10 s. Sham surgery was performed by exposing the sciatic nerve but without further nerve injury. Animals were monitored for the development of hyperalgesia to heat and of mechanical allodynia to von Frey hairs as previously described (Hargreaves et al., 1988, Sommer et al., 1998). Analysis of variance with Tukey's post hoc test was used to test for significant differences of mean withdrawal thresholds or mean withdrawal latencies after crush or CCI when compared to baseline control values. After CCI, animals were sacrificed on days 1, 3, 7, 14, and 28 (n = 36 per time point; except for day 28 with n = 12). After crush injury, animals were sacrificed on days 0, 1, 3, 7, 14, and 28 (n = 36 per time point; except for day 14 with n = 12). After sham surgery, animals were sacrificed on day 1 (n = 12 per time point). Sciatic nerves were removed by cutting the nerve shortly above the site of the constrictive ligatures or the crush lesion and 1 cm distally. Control nerves from naive animals (day 0) or sham-operated nerves were removed accordingly. Sciatic nerves from 12 animals were pooled per sample and homogenized as described (George et al., 2004). Levels of TNF, TNFR1, and 2 were determined by the Quantikine M TNF-α Elisa and the Mouse sTNFR1 and 2 Elisa (R&D, Wiesbaden, Germany) according to the manufacturer's instructions. These assay systems detect mouse TNF, TNFR1, and TNFR2 protein with a sensitivity of 5, 15, and 12 pg/ml, respectively. TNF and TNFR levels were expressed as pg/mg total protein. Protein content was determined by the bicinchoninic acid protein assay reagent (KMF Laborchemie, St. Augustin, Germany). We used some aliquots from the nerve homogenates obtained in the present study for parallel measurements of interleukin-10 as reported recently (George et al., 2004). Data represent mean values ± standard deviation. Analysis of variance was used to test for differences between groups and subsequent Scheffé test for comparison of individual means throughout the time course. Statistical significance was assumed with P < 0.05.

Thermal hyperalgesia (Fig. 1A) developed rapidly in all CCI- or crush-injured mice with a maximum during weeks 1 and 2. Mechanical allodynia (Fig. 1B) was present at all time points investigated in both models until the last test on day 23.

Protein levels of TNF (Fig. 2A), TNFR1 (Fig. 2B), and TNFR2 (Fig. 2C) were increased after peripheral nerve injury at the site of nerve lesion, but with different kinetics. TNF was increased in the injured mouse sciatic nerve on days 1 and 3 after CCI or crush. TNF was not detectable in nerve homogenates from naive or sham-operated controls or in nerve homogenates from nerve-injured mice on day 7, 14, or 28 after CCI or crush. Thus, endoneurial TNF protein peaked early and transiently in mouse sciatic nerve at the site of lesion. This is in line with former reports in mice (Shamash et al., 2002). Both TNFRs were detectable at a low level in nerve homogenates from naive mice. After crush or CCI, TNFR1 increased 2- to 3-fold between days 1 and 7 (Fig. 2B). Unlike TNFR1, TNFR2 was markedly upregulated already on day 1 after crush or CCI. TNFR2 increased by 7-fold on days 3 and 7, and remained elevated at a lower level until day 28 after both CCI and crush (Fig. 2C). The TNFR1 and TNFR2 levels in nerve homogenates harvested on day 1 after sham surgery were 25.8 and 51.5 pg/mg protein, respectively, and thus only slightly increased over controls.

Our data reveal three major findings. First, an early and transient increase of local TNF was paralleled and followed by an increase of its two receptors. Thus, TNF-initiated effects after peripheral nerve injury may be mediated and maintained by the TNFR1 and TNFR2 at the site of lesion. Furthermore, both TNF receptors are involved in TNF regulation. Thus, modulating the TNF–TNFR1–TNFR2 balance may offer a strategy to antagonize TNF-initiated processes even at time points when TNF has returned to baseline levels again.

Second, the time course of local TNFRs after CCI did not differ markedly from the crush-induced TNFR changes, although there is an additional inflammatory component due to the ligature material in CCI. Thus, local TNFR upregulation during Wallerian degeneration is rather due to the nerve injury with axonal damage than due to an additional local inflammation.

Thirdly, the temporal pattern of nerve injury-induced TNFR1 changes was distinct from that observed for the TNFR2. This indicates that the TNFR1 and the TNFR2 protein are differentially regulated during Wallerian degeneration. The slight and slow increase of TNFR1 after nerve injury in the present study corresponded well with the known constitutive expression of the TNFR1 gene regulated by a non-inducible, housekeeping promoter (Rothe et al., 1993). For TNFR2, which is characterized as the inducible receptor subtype, we observed a rapid, marked and long lasting upregulation after nerve injury. However, higher levels of the TNFR2 may not necessarily be linked to more functional importance. Since TNFR1 is able to self-associate specifically, even low levels of TNFR1 and slight changes might be functionally relevant.

The TNFR1 has been shown to mediate the CCI-induced thermal hyperalgesia (Sommer et al., 1998). In the present study, the maximum of local TNFR1 levels paralleled neuropathic pain associated behavior during the first week after nerve injury. Thus, TNFR1-antagonizing strategies might be helpful during this time window. We also observed that the TNFR2 but not the TNFR1 protein remained elevated by day 28 after CCI or crush injury; thus, at a time when nerve regeneration occurs. This suggests that TNF and the TNFR2 are not only involved in deleterious effects after nerve injury but might also be essential for successful nerve regeneration. The shift of the TNF–TNFR1–TNFR2 balance toward TNFR2 at later timepoints after nerve injury could also indicate that TNFR2 acts as an endogenous TNF antagonist.

Further studies are warranted to further clarify which TNF-iniated effects after peripheral nerve injury are mediated by the TNFR1 or the TNFR2. This may offer a strategy to modulate TNF induced effects after peripheral nerve injury more specifically.

Section snippets

Acknowledgments

The expert technical help of Lydia Biko, Barbara Dekant and Helga Brünner is greatly appreciated. We thank Prof. K. Toyka for critical reading of the manuscript. This work was supported by Volkswagenstiftung.

References (24)

Cited by (82)

  • The Jak/STAT pathway: A focus on pain in rheumatoid arthritis

    2021, Seminars in Arthritis and Rheumatism
  • Death Receptor 6 Promotes Wallerian Degeneration in Peripheral Axons

    2017, Current Biology
    Citation Excerpt :

    This reveals a surprising anti-degenerative role for p75NTR in the context of Wallerian degeneration, which is in contrast to its pro-degenerative role in trophic withdrawal. This, coupled with the known roles for TNFR family members in suppressing axon regeneration, somewhat complicates their role in functional recovery after nerve injury [23, 26, 34, 35]. Sciatic nerves from uninjured DR6−/− mice showed larger axon diameters but normal myelin thicknesses compared to wild-type (Figure S4I).

  • Modulation of Opioid Analgesic Reward by Inflammatory Agents

    2016, Neuropathology of Drug Addictions and Substance Misuse Volume 3: General Processes and Mechanisms, Prescription Medications, Caffeine and Areca, Polydrug Misuse, Emerging Addictions and Non-Drug Addictions
View all citing articles on Scopus
View full text