Serum matrix metalloproteinase-9 activity is dysregulated with disease progression in the mutant SOD1 transgenic mice

Abstract

Amyotrophic lateral sclerosis (ALS) is an adult-onset fatal neurodegenerative disorder characterized by progressive deterioration of motor neurons in the spinal cord, brainstem, and cerebral cortex. Matrix metalloproteinase-9 (MMP-9) is proposed to be a biomarker for ALS due to a potential pathological role in the disease. However, despite numerous studies, it is still unclear whether there is a direct correlation between MMP-9 expression in serum and progression of disease. Therefore, we used a TgSOD1^G93A mouse with a low transgene copy number. This model shows slow disease progression analogous to human ALS and provides a useful model to study biomarker expression at different stages of disease. Using zymography, we found that serum MMP-9 activity was significantly elevated in animals showing early signs of disease when compared to the younger, pre-symptomatic animals. This was followed by a decrease in MMP-9 activity in TgSOD1^G93A mice with end-stage disease. These results were confirmed in serum of a high copy number strain of TgSOD1^G93A mice with rapid progression. MMP-9 expression was changed accordingly in spinal motor neurons, glia and neuropil, suggesting a spinal cord contribution to blood MMP-9 activity. Serum MMP-2 activity followed a similar profile as the MMP-9 in these two models. These data indicate that circulating MMP-9 is altered throughout the course of disease progression in mice. Further studies in human ALS may validate the suitability of serum MMP-9 activity as a biomarker for early stage disease.

Introduction

Amyotrophic lateral sclerosis (ALS) is the most common form of the adult-onset motor neuron disease (MND), and is a fatal neurodegenerative disorder that is distinguished by selective loss of cortical and spinal motor neurons leading to the clinical characteristics of muscle weakness, wasting, and spasticity [1]. These symptoms lead to paralysis and eventually death through respiratory failure. Death usually occurs within 2–5 years of disease onset depending on the severity and the stage of disease when diagnosed. More than 90% of ALS cases are sporadic (sALS), but 5–10% of ALS are inherited [2]. The aetiology of this debilitating disease, for which no pharmacotherapies and only minimal treatments are available, remains unclear. Mutations (more than 100) in the superoxide dismutase 1 (SOD1) gene are responsible for 10–15% of familial ALS (fALS) cases [3], implicating a role for SOD1 in ALS pathogenesis. Since most of the mutations retain normal superoxide dismutase activity and SOD1 knock-out mice do not fatally develop ALS-like symptoms [4], this suggests that SOD1 mutations lead to a toxic gain of function. Several hypotheses concerning the mechanisms triggering motor neuron degeneration in ALS have been proposed, including abnormal intracellular protein misfolding and aggregation, oxidative stress, mitochondrial dysfunction, glutamatergic excitotoxicity, axonal transport malfunction caused by defective neurofilament networks, metal dyshomeostasis, and neuroinflammation [5], [6], [7], [8]. Identification of disease biomarkers will help diagnosis and early intervention to obtain better outcomes for the disease.

Matrix metalloproteinases (MMPs) are a family of proteolytic enzymes that degrade and remodel extracellular matrix components. MMP activity is calcium (Ca²⁺) dependent and the proteins contain a zinc ion (Zn²⁺) at their catalytic sites [9]. Both MMP-9 and MMP-2 are secreted as proenzymes and can be cleaved into active enzymes. These metalloproteinases have beneficial roles in addition to detrimental effects in the central nervous system [10]. MMP-9 has been linked to ALS pathogenesis [11] and increased MMP-9 immunostaining has been observed from biopsies of atrophic muscles of ALS patients compared to normal muscle [12]. High levels of MMP-9 activity were observed in motor cortex and spinal cord (lumbar and thoracic cord) of ALS patients compared to control specimens [13]. An increase in the expression levels of pro and active forms of MMP-9 [14] as well as an increase in MMP-9 activity (free active and free proMMP-9) measured by ELISA [15] was observed in serum samples of ALS patients. Further to this, Lorenzl et al. reported a slight but insignificant elevation in proMMP-9 levels by ELISA in ALS plasma [16]. MMP-9 levels in ALS cerebrospinal fluid (CSF) is apparently 10- to 1000-fold less than in serum and below the limit of the ELISA [14] and was not detected by zymography [17], thus studies have focused on circulating MMP-9. Oral administration of 100 mg/kg broad spectrum MMP inhibitor Ro 26–2853 during the pre-symptomatic stage in transgenic SOD1^G93A mice with a high gene copy number improved motor function and significantly extended their survival time [18]. A recent study revealed a significant increase in the life span of MMP-9 “knock-out” transgenic SOD1^G93A mice on a B6SJL background as well as a reduction of neuronal death in the spinal cord [19]. These studies suggest a pathogenic role of MMP-9 in the disease, and that inhibition of MMP-9 could be a potential therapeutic approach for treating ALS. However, in an independent study, knock-out of the MMP-9 gene in TgSOD1^G93A mice (on a C57BL6 background with a life span of ∼138 days) accelerated disease progression and reduced survival [20], indicating that MMP-9 has a protective effect. The discrepancy between all these studies may lie in the influence of the mouse genetic background on the phenotype, the construct used in creation of the knock-out, as well as the rate of disease progression. The role of MMP-2 is less clear as no MMP-2 knock-out study in the mutant SOD mouse model is available. MMP-2 activity was found to be decreased in ALS motor cortex [13], and in biopsies, MMP-2 immunostaining was found to be increased in atrophic muscle of ALS patients compared to normal muscle [12]. However no significant change was detected in ALS plasma [16].

Due to the disconcordant results in previous studies, we examined MMP-9 activity using a murine model of ALS with low SOD1^G93A transgene copy number. These mice over-express human mutant SOD1^G93A on a C57Bl6 background, have slow disease progression and produce an adult-onset phenotype that closely replicates both clinical and pathological hallmarks of human ALS. This allowed us to follow serum MMP-9 activity through clearly defined pre-symptomatic (100 days), early stage (200 days) and end-stage disease (270 days). We also examined MMP-9 activity in serum from TgSOD1^G93A mice with a high-transgene copy number which develop terminal disease in a shorter time frame (∼120 days).