Elsevier

Experimental Neurology

Volume 237, Issue 1, September 2012, Pages 8-17
Experimental Neurology

Reduced removal of synaptic terminals from axotomized spinal motoneurons in the absence of complement C3

https://doi.org/10.1016/j.expneurol.2012.06.008Get rights and content

Abstract

Complement proteins C1q and C3 play a critical role in synaptic elimination during development. Axotomy of spinal motoneurons triggers removal of synaptic terminals from the cell surface of motoneurons by largely unknown mechanisms. We therefore hypothesized that the complement system is involved also in synaptic stripping of injured motoneurons. In the sciatic motor pool of wild type (WT) mice, the immunoreactivity (IR) for both C1q and C3 was increased after sciatic nerve transection (SNT). Mice deficient in C3 (C3−/−) showed a reduced loss of synaptic terminals from injured motoneurons at one week after SNT, as assessed by immunoreactivity for synaptic markers and electron microscopy. In particular, the removal of putative inhibitory terminals, immunopositive for vesicular inhibitory amino acid transporter (VIAAT) and ultrastructurally identified as type F synapses, was reduced in C3−/− mice. In contrast, lesion-induced removal of nerve terminals in C1q−/− mice appeared similar to WT mice. Growth associated protein (GAP)-43 mRNA expression in lesioned motoneurons increased much more in C3−/− compared to WT mice after SNT. After sciatic nerve crush (SNC), the C3−/− mice showed a faster functional recovery, assessed as grip strength, compared to WT mice. No differences were detected regarding nerve inflammation at the site of injury or pattern of muscle reinnervation. These data indicate that a non-classical pathway of complement activation is involved in axotomy-induced adult synapse removal, and that its inhibition promotes functional recovery.

Highlights

► The absence of complement C3 reduces synaptic stripping of motoneurons after axotomy. ► Dependence of C3 for synaptic stripping is the strongest for inhibitory synapses. ► Normal stripping without C1q implies a non-classical complement activation pathway. ► The absence of C3 improves functional recovery after axotomy.

Introduction

The complement system is a part of the innate immune system and provides a rapid response to infections by opsonizing foreign material, attracting immune cells, and lysis of foreign cell membranes. Activation of the complement system can also contribute to tissue damage. The classical complement pathway is initiated by C1q, eventually leading to activation of the key component of the complement cascade, C3. The complement system can also be activated by non-classical pathways in a manner that is independent of C1q (Gasque, 2004, Ip et al., 2009, Medzhitov and Janeway, 2000, Morgan and Gasque, 1996).

Recent findings suggest that the complement system has also important functions in the development and maintenance of CNS, such as stimulation of adult neurogenesis (Rahpeymai et al., 2006, Shinjyo et al., 2009) and the refinement of developing neuronal circuits. Stevens et al. (2007) reported that complement factors C1q and C3 tag synapses in the visual pathway during development and during neurodegeneration in the retina, thereby helping to mediate their removal(Stevens et al., 2007). Since defects in the synaptic elimination and organization in the retinogeniculate pathway was detected in mice lacking either C3 (C3−/−) or C1q (C1q−/−), the authors concluded that the effects on the synaptic organization were mediated through the classical complement cascade. Loss of synapses marks the onset of several virtually untreatable neurodegenerative diseases, such as amyotrophic lateral sclerosis and Alzheimer's disease (Selkoe, 2002, Zang da et al., 2005). Therefore, it is of great interest to investigate the involvement of complement in synapse elimination in different pathological conditions and experimental models.

One model of CNS synapse elimination is peripheral axotomy of spinal motoneurons. Axotomy results in inflammation both at the lesion site and within the spinal cord. The distal stump of the severed axon is degraded in Wallerian degeneration and within the spinal cord microglia and astrocytes are activated and interact closely with the injured motoneuron. The removal of synaptic terminals from the cell surface of axotomized motoneurons (‘synaptic stripping’) is initiated by still largely unknown mechanisms (Aldskogius et al., 1999, Blinzinger and Kreutzberg, 1968, Chen, 1978, Lindå et al., 2000, Oliveira et al., 2004, Sumner, 1975). However, the major histocompatibility complex (MHC) class I family of immune molecules plays a role in this process (Oliveira et al., 2004).

After peripheral nerve injury, the complement system is activated with increased levels of C1q and C3 within lesioned motor nuclei, and injured motoneurons increase their expression of complement inhibitors such as CD59 (Mattsson et al., 1998). The complement system also has well established roles in the peripheral inflammatory events following peripheral nerve lesions, as reviewed in Ramaglia et al. (2008). Complement expression is increased in the lesioned peripheral nerve (Ohlsson et al., 2003) and complement factor C3 mediates recruitment of macrophages into the damaged nerve and contributes to the clearing of debris in Wallerian degeneration (Bruck and Friede, 1990, Dailey et al., 1998).

We have here investigated whether the complement system is involved in the process of synaptic stripping of axotomized motoneurons. For this purpose we studied the synapse stripping response in mice lacking C1q or C3.

Section snippets

Animal experiments

C1q−/− mice (Botto et al., 1998) were backcrossed to C57Bl/6 genetic background. C3−/− mice (Pekna et al., 1998) were backcrossed to C57Bl/6 genetic background (Charles River, Uppsala, Sweden) for 13 generations. Adult C1q−/− and WT littermates as well as adult C3−/− and age matched WT females were anesthetized with a mixture of midazolam (Dormicum, Roche Diagnostics; 1.25 mg/ml) and Hypnorm (Janssen) i.p. at 0.2 ml per 25 g of body weight. The left sciatic nerve was transected or crushed at the

Expression of complement proteins in the spinal cord after SNT

We first investigated the expression of complements C3 and C1q in the spinal cord after axotomy. Seven days after SNT, immunoreactivity (IR) for both markers was increased in the sciatic motoneuron pool (Figs. 1A–D). C1q displayed a diffuse pattern of small punctae in the neuropil with the strongest density close to the surface of motoneuron somata and dendrites (Figs. 1A, B). This pattern was different from C3, which displayed a more localized distribution in the neuropil (Figs. 1C, D).

Axotomy-induced elimination of synapses from injured motoneurons is mediated by C3 but not C1q

Here we studied one of the hallmarks of the response of spinal motoneurons to axon lesion, i.e. the elimination of synapses from the surface of lesioned neurons, in relation to the presence of complement proteins. The most important finding is that in the absence of complement protein C3, the majority of synaptic terminals that are normally removed are spared. This could be attributed to a very high degree of maintenance of putative inhibitory synaptic terminals, which constitute the large

Acknowledgment

This study was supported by grants by the Swedish Research Council (6815 to S. Cullheim, 20116 to M. Pekna and 11548 to M. Pekny), ALF Göteborg (11267 to M. Pekna and 11392 to M. Pekny), AFA Research Foundation, Sten A. Olsson Foundation for Research and Culture, NanoNet COST Action, the EU FP 7 Programs EduGlia (237956) and TargetBraIn (279017), the Brain Foundation, the Marianne and Marcus Wallenberg Foundation, Friends of Karolinska Institutet, Karolinska Institutet, J. and B. Wennerström's

References (49)

  • B. Stevens et al.

    The classical complement cascade mediates CNS synapse elimination

    Cell

    (2007)
  • B.E. Sumner

    A quantitative analysis of boutons with different types of synapse in normal and injured hypoglossal nuclei

    Exp. Neurol.

    (1975)
  • M. Svensson et al.

    Synaptic density of axotomized hypoglossal motoneurons following pharmacological blockade of the microglial cell proliferation

    Exp. Neurol.

    (1993)
  • M. Svensson et al.

    The response of central glia to peripheral nerve injury

    Brain Res. Bull.

    (1993)
  • B. Ulfhake et al.

    An ultrastructural study of 5-hydroxytryptamine-, thyrotropin-releasing hormone- and substance P-immunoreactive axonal boutons in the motor nucleus of spinal cord segments L7-S1 in the adult cat

    Neuroscience

    (1987)
  • H. Aldskogius et al.

    Glial responses to synaptic damage and plasticity

    J. Neurosci. Res.

    (1999)
  • F.J. Alvarez et al.

    Permanent central synaptic disconnection of proprioceptors after nerve injury and regeneration: I. Loss of VGLUT1/IA synapses on motoneurons

    J. Neurophysiol.

    (2011)
  • K. Barron

    Comparative observations on the cytologic reactions of central and peripheral nerve cells to axotomy

  • K. Blinzinger et al.

    Displacement of synaptic terminals from regenerating motoneurons by microglial cells

    Z. Zellforsch. Mikrosk. Anat.

    (1968)
  • M. Botto et al.

    Homozygous C1q deficiency causes glomerulonephritis associated with multiple apoptotic bodies

    Nat. Genet.

    (1998)
  • W. Bruck et al.

    Anti-macrophage CR3 antibody blocks myelin phagocytosis by macrophages in vitro

    Acta Neuropathol.

    (1990)
  • D.H. Chen

    Qualitative and quantitative study of synaptic displacement in chromatolyzed spinal motoneurons of the cat

    J. Comp. Neurol.

    (1978)
  • Y. Chu et al.

    Enhanced synaptic connectivity and epilepsy in C1q knockout mice

    Proc. Natl. Acad. Sci. U. S. A.

    (2010)
  • S. Conradi

    Ultrastructure and distribution of neuronal and glial elements on the motoneuron surface in the lumbosacral spinal cord of the adult cat

    Acta Physiol. Scand. Suppl.

    (1969)
  • Cited by (0)

    1

    Equal contribution by AB and JZ.

    View full text