Is brain gliosis a characteristic of chronic methamphetamine use in the human?

Highlights

• Brain of human methamphetamine users unexpectedly showed little evidence of gliosis.

• Glial markers CR3/43, TAL.1B5, GLUT-5, GFAP, vimentin, and Hsp27 were examined.

• A positive control of brain of patients with multiple system atrophy was employed.

• Increased proteolysis of vimentin and Hsp27 were observed.

Abstract

Animal data show that high doses of the stimulant drug methamphetamine can damage brain dopamine neurones; however, it is still uncertain whether methamphetamine, at any dose, is neurotoxic to human brain.

Since gliosis is typically associated with brain damage and is observed in animal models of methamphetamine exposure, we measured protein levels (intact protein and fragments, if any) of markers of microgliosis (glucose transporter-5, human leukocyte antigens HLA-DRα [TAL.1B5] and HLA-DR/DQ/DPβ [CR3/43]) and astrogliosis (glial fibrillary acidic protein, vimentin, and heat shock protein-27) in homogenates of autopsied brain of chronic methamphetamine users (n = 20) and matched controls (n = 23). Intact protein levels of all markers were, as expected, elevated (+ 28%–1270%, P < 0.05) in putamen of patients with the neurodegenerative disorder multiple system atrophy (as a positive control) as were concentrations of fragments of glial fibrillary acidic protein, vimentin and heat shock protein-27 (+ 170%–4700%, P < 0.005). In contrast, intact protein concentrations of the markers were normal in dopamine-rich striatum (caudate, putamen) and in the frontal cortex of the drug users. However, striatal levels of cleaved vimentin and heat shock protein-27 were increased (by 98%–211%, P < 0.05), with positive correlations (r = 0.41–0.60) observed between concentrations of truncated heat shock protein-27 and extent of dopamine loss (P = 0.006) and levels of lipid peroxidation products 4-hydroxynonenal (P = 0.046) and malondialdehyde (P = 0.11).

Our failure to detect increased intact protein levels of commonly used markers of microgliosis and astrogliosis could be explained by exposure to methamphetamine insufficient to cause a toxic process associated with overt gliosis; however, about half of the subjects had died of drug intoxication suggesting that “high” drug doses might have been used. Alternatively, drug tolerance to toxic effects might have occurred in the subjects, who were all chronic methamphetamine users. Nevertheless, the finding of above-normal levels of striatal vimentin and heat shock protein-27 fragments (which constituted 10–28% of the intact protein), for which changes in the latter correlated with those of several markers possibly suggestive of damage, does suggest that some astrocytic “disturbance” had occurred, which might in principle be related to methamphetamine neurotoxicity or to a neuroplastic remodeling process. Taken together, our neurochemical findings do not provide strong evidence for either marked microgliosis or astrogliosis in at least a subgroup of human recreational methamphetamine users who used the drug chronically and shortly before death. However, a logistically more difficult quantitative histopathological study is needed to confirm whether glial changes occur or do not occur in brain of human methamphetamine (and amphetamine) users.

Introduction

High doses of the psychostimulant drug methamphetamine (MA) can damage brain dopamine neurones in experimental animal studies (see Cadet et al., 2003, O'Callaghan et al., 2008, Yamamoto et al., 2010 for reviews), and it has been speculated that even “low” doses of amphetamine used clinically in psychiatry might cause brain damage (Ricaurte et al., 2005). Further, results of an epidemiological study revealed increased risk of development of Parkinson's disease in hospitalized patients with MA-use disorders (Callaghan et al., 2012). However, dopamine neurone loss has been difficult to prove in the human, primarily because of uncertainty whether reported low levels of brain dopamine nerve terminal markers (e.g., dopamine, dopamine transporter; for extensive review see (Kish, 2014)) in human MA users equal actual “physical” damage to the neurone.

Some attention therefore has been focused on whether brain of MA users displays signs of either microgliosis or astrogliosis, as these cellular changes, although not equaling brain damage, are typically observed following brain injury in general (Kreutzberg, 1996, Ridet et al., 1997), and in brain, primarily in the dopamine-rich striatum, of experimental animals exposed to high doses of MA (or amphetamine) (Escubedo et al., 1998, Hess et al., 1990, Krasnova et al., 2010, O'Callaghan and Miller, 1994, O'Callaghan et al., 2008, Thomas et al., 2004). Gliosis is a natural reactive process of glial cells (astrocytes and microglia) to brain injury, damage, infection, or disturbed homeostasis and is characterized in part by increased expression of glial-specific proteins and morphologically hypertrophied cell body and processes, and in some cases, also cell proliferation and migration. The functions of gliosis following brain injury and whether the response is helpful or detrimental continue to be debated (Sofroniew and Vinters, 2010, Streit, 2010, Zhang et al., 2010), but most likely include removal of cellular debris, wound repair, and possibly, some involvement in neuroregeneration.

In our postmortem (forensic) neuropathological examinations of brains of human MA users we did not observe any obvious signs of above-normal gliosis (Moszczynska et al., 2004); however, these findings are uncertain as no quantitative assessment could be performed on the material. The literature on this question is scanty and contradictory, limited to a brain imaging finding of massively increased binding of a putative marker of activated microglial cells (¹¹C-PK11195) throughout the brain of abstinent MA users (Sekine et al., 2008) and a neuropathological investigation showing increased number of glucose transporter-5 (GLUT-5)-positive microglia in striatum of users who died of MA intoxication, but no evidence of reactive microgliosis (GLUT-5 or CR3/43) or astrocytosis [glial fibrillary acidic protein (GFAP) or S100B] (Kitamura et al., 2010).

The aim of our study was to establish, using measurement of protein levels of generally accepted markers of microglia and astroglia in postmortem brain homogenates, whether experimental animal findings showing brain gliosis following MA exposure might be relevant to chronic human (recreational) use of the drug. We hypothesized that concentrations of all gliosis markers would be above-normal in the dopamine-rich striatum, but less so, if at all, in dopamine-poor brain areas.