Schwann cell GFAP expression increases in axonal neuropathies

Neuromuscular Junction Changes in a Mouse Model of Charcot-Marie-Tooth Disease Type 4C

The neuromuscular junction (NMJ) appears to be a site of pathology in a number of peripheral nerve diseases. Charcot-Marie-Tooth (CMT) 4C is an autosomal recessive, early onset, demyelinating neuropathy. Numerous mutations in the SH3TC2 gene have been shown to underlie the condition often associated with scoliosis, foot deformities, and reduced nerve conduction velocities. Mice with exon 1 of the Sh3tc2 gene knocked out demonstrate many of the features seen in patients. To determine if NMJ pathology is contributory to the pathomechanisms of CMT4C we examined NMJs in the gastrocnemius muscle of SH3TC2-deficient mice. In addition, we performed proteomic assessment of the sciatic nerve to identify protein factors contributing to the NMJ alterations and the survival of demyelinated axons. Morphological and gene expression analysis of NMJs revealed a lack of continuity between the pre- and post-synaptic apparatus, increases in post-synaptic fragmentation and dispersal, and an increase in expression of the gamma subunit of the acetylcholine receptor. There were no changes in axonal width or the number of axonal inputs to the NMJ. Proteome investigations of the sciatic nerve revealed altered expression of extracellular matrix proteins important for NMJ integrity. Together these observations suggest that CMT4C pathology includes a compromised NMJ even in the absence of changes to the innervating axon.

Serum Metabolites Associated with Computed Tomography Findings after Traumatic Brain Injury

There is a need to rapidly detect patients with traumatic brain injury (TBI) who require head computed tomography (CT). Given the energy crisis in the brain following TBI, we hypothesized that serum metabolomics would be a useful tool for developing a set of biomarkers to determine the need for CT and to distinguish among different types of injuries observed. Logistical regression models using metabolite data from the discovery cohort (n = 144, Turku, Finland) were used to distinguish between patients with traumatic intracranial findings and those with negative findings on head CT. The resultant models were then tested in the validation cohort (n = 66, Cambridge, United Kingdom). The levels of glial fibrillary acidic protein and ubiquitin C-terminal hydrolase-L1 were also quantified in the serum from the same patients. Despite there being significant differences in the protein biomarkers in patients with TBI, the model that determined the need for a CT scan validated poorly (area under the curve [AUC] = 0.64: Cambridge patients). However, using a combination of six metabolites (two amino acids, three sugar derivatives, and one ketoacid) it was possible to discriminate patients with intracranial abnormalities on CT and patients with a normal CT (AUC = 0.77 in Turku patients and AUC = 0.73 in Cambridge patients). Further, a combination of three metabolites could distinguish between diffuse brain injuries and mass lesions (AUC = 0.87 in Turku patients and AUC = 0.68 in Cambridge patients). This study identifies a set of validated serum polar metabolites, which associate with the need for a CT scan. Additionally, serum metabolites can also predict the nature of the brain injury. These metabolite markers may prevent unnecessary CT scans, thus reducing the cost of diagnostics and radiation load.

Neurofilament light protein levels in cerebrospinal fluid predict long-term disability of Guillain-Barré syndrome: A pilot study

OBJECTIVES

Although the recovery from Guillain-Barré syndrome (GBS) is good in most patients, some develop permanent severe disability or even die. Early predictors would increase the likelihood to identify patients at risk for poor outcome at the acute stage, allowing them intensified therapeutic intervention.

MATERIALS AND METHOD

Eighteen patients with a history of GBS 9-17 years ago were reassessed with scoring of neurological disability and quality of life assessment (QoL). Their previous diagnostic work-up included clinical examination with scoring of disability, neurophysiological investigation, a battery of serology tests for infections, and cerebrospinal fluid (CSF) examination. Aliquots of CSF were frozen, stored for 20-28 years, and analyzed by ELISA for determination of neurofilament light protein (NFL) and glial fibrillary acidic protein (GFAP).

RESULTS

Patients with poor outcome (n = 3) had significantly higher NFL and GFAP levels at GBS nadir than those with good outcome (n = 15, P < .01 and P < .05, respectively). High NFL correlated with more prominent disability and worse QoL at long-term follow-up (r = .694, P < .001, and SF 36 dimension physical component summary (PCS) (r =-.65, P < .05), respectively, whereas GFAP did not correlate with clinical outcome or QoL.

CONCLUSION

High NFL in CSF at the acute stage of GBS seems to predict long-term outcome and might, together with neurophysiological and clinical measures, be useful in treatment decisions and clinical care of GBS.

Addressing the needs of traumatic brain injury with clinical proteomics

Background

Neurotrauma or injuries to the central nervous system (CNS) are a serious public health problem worldwide. Approximately 75% of all traumatic brain injuries (TBIs) are concussions or other mild TBI (mTBI) forms. Evaluation of concussion injury today is limited to an assessment of behavioral symptoms, often with delay and subject to motivation. Hence, there is an urgent need for an accurate chemical measure in biofluids to serve as a diagnostic tool for invisible brain wounds, to monitor severe patient trajectories, and to predict survival chances. Although a number of neurotrauma marker candidates have been reported, the broad spectrum of TBI limits the significance of small cohort studies. Specificity and sensitivity issues compound the development of a conclusive diagnostic assay, especially for concussion patients. Thus, the neurotrauma field currently has no diagnostic biofluid test in clinical use.

Content

We discuss the challenges of discovering new and validating identified neurotrauma marker candidates using proteomics-based strategies, including targeting, selection strategies and the application of mass spectrometry (MS) technologies and their potential impact to the neurotrauma field.

Summary

Many studies use TBI marker candidates based on literature reports, yet progress in genomics and proteomics have started to provide neurotrauma protein profiles. Choosing meaningful marker candidates from such ‘long lists’ is still pending, as only few can be taken through the process of preclinical verification and large scale translational validation. Quantitative mass spectrometry targeting specific molecules rather than random sampling of the whole proteome, e.g., multiple reaction monitoring (MRM), offers an efficient and effective means to multiplex the measurement of several candidates in patient samples, thereby omitting the need for antibodies prior to clinical assay design. Sample preparation challenges specific to TBI are addressed. A tailored selection strategy combined with a multiplex screening approach is helping to arrive at diagnostically suitable candidates for clinical assay development. A surrogate marker test will be instrumental for critical decisions of TBI patient care and protection of concussion victims from repeated exposures that could result in lasting neurological deficits.

Impaired expression of ciliary neurotrophic factor in Charcot-Marie-Tooth type 1A neuropathy

We investigated the contribution of Schwann cell-derived ciliary neurotrophic factor (CNTF) to the pathogenesis of Charcot-Marie-Tooth disease type 1A (CMT1A) and addressed the question as to whether it plays a role in the development of axonal damage observed in the disease, with aging. Ciliary neurotrophic factor was underexpressed in experimental CMT1A but not in other models of hereditary neuropathies. Sciatic nerve crush experiments and dosage of CNTF at different time points showed that expression of this trophic factor remained significantly lower in CMT1A rats than in normal controls; moreover, in uninjured CMT1A sciatic nerves CNTF levels further decreased with ageing, thus paralleling the molecular signs of axonal impairment, that is increased expression of non-phosphorylated neurofilaments and amyloid precursor protein. Administration of CNTF to dorsal root ganglia cultures reduced dephosphorylation of neurofilaments in CMT1A cultures, without improving demyelination. Taken together, these results provide further evidence that the production of CNTF by Schwann cells is markedly reduced in CMT1A. Moreover, the observations suggest that trophic support to the axon is impaired in CMT1A and that further studies on the therapeutic use of trophic factors or their derivatives in experimental and human CMT1A are warranted.

Glial fibrillary acidic protein: a marker of axonal Guillain-Barrè syndrome and outcome

Glial fibrillary acid protein (GFAP) is increased in serum and cerebrospinal fluid of patients with dementia, traumatic brain injury, stroke, and multiple sclerosis. To determine whether GFAP is increased in Guillain-Barré syndrome (GBS) we evaluated serum GFAP in 30 controls, 20 patients with acute inflammatory demyelinating neuropathy (AIDP), and 17 with primary axonal GBS. Serum GFAP levels were increased in axonal GBS (median, 0.74) compared with controls (median, 0.41; P < 0.0001) and AIDP (median, 0.58; P = 0.0015). GFAP levels correlated with Hughes grades (serum r = 0.74; P < 0.0001) 6 months after neuropathy onset. Applying the cutoff value in serum of 0.63 to the diagnosis of axonal GBS, we obtained a sensitivity of 76.5% and a specificity of 86%. Thus, serum GFAP levels may be used in GBS as a diagnostic marker of the axonal variant and to predict outcome.

Dorsal root rupture injury induces extension of astrocytic processes into the peripheral nervous system and expression of GDNF in astrocytes

Preganglionic brachial plexus injuries fall into two categories according to the lesion site, root avulsion injury and root rupture injury. The latter type of injury involves part of the peripheral nervous system (PNS) component at the injured spinal cord surface. Previous investigators have used rhizotomy of experimental animals as a model for dorsal root rupture injury. However, the effect on the central nervous system (CNS)-PNS junction accompanied by the mechanical stress from traction force is hard to estimate in this model. The current study aimed to demonstrate temporal molecular alterations from the CNS-PNS junction to the ruptured dorsal root after traction injury by immunohistochemical procedures. At 28 days after dorsal rupture injury, GFAP-positive structures could be clearly identified showing rather straight lines from the centro-peripheral junction toward the peripheral stump in the ruptured dorsal root. Immunoelectron microscopy for GFAP verified GFAP IR within the astrocytic processes at the injured dorsal root at 28 days after dorsal rupture injury. Glial cell line-derived neurotrophic factor immunoreactivity (GDNF IR) was slightly upregulated within the Schwann cell bodies on the injured dorsal root at 24-48 h after rupture injury. However, GDNF IR had appeared showing a process-like profile on the ruptured dorsal root by 28 days, and it was closely related with GFAP-positive structures. In contrast, a small increase in GFAP IR was only detected on the proximal side on the rhizotomized dorsal root at 28 days after rhizotomy. A marked decrease in NF IR and S-100 IR was observed at the ruptured dorsal root from 7 days. On the other hand, laminin IR was strongly upregulated on the ruptured dorsal root from 48 h to 7 days, and was still evident at 28 days. We therefore conclude that the astrocytes show a unique ability to extend their processes toward the stump. This ability may provide a new medium for the study of axonal regeneration in future clinical experiments.

The p75 neurotrophin receptor, relative to other Schwann cell and melanoma markers, is abundantly Expressed in spindled melanomas

Seventeen cases of spindled melanomas and eleven cases of epithelioid melanomas were immunolabeled with various melanoma and Schwann cell markers. Standard melanoma markers included S100, HMB45, HMB50, tyrosinase, and Melan A. Schwann cell markers included the p75 neurotrophin receptor (p75NTR), glial fibrillary acidic protein (GFAP), and the L1 adhesion protein. The degree of immunocytochemical labeling was scored by levels of both intensity and pervasiveness. The results confirmed a distinct difference in labeling between epithelioid and spindled melanomas. The p75NTR was strongly expressed in spindled melanomas and weakly expressed in the epithelioid melanomas. The usual melanoma markers, including HMB45, HMB50, MelanA, and tyrosinase had the reverse pattern, being strongly expressed in virtually all epithelioid melanomas, but rarely expressed in the spindled variants. S100 was unique among the markers in being expressed by both epithelioid and spindled melanomas. Glial fibrillary acidic protein and L1 adhesion protein were expressed moderately, with preferential labeling of the spindled melanomas. The greatest immunophenotypic difference between spindled and epithelioid melanomas was the high abundance of p75NTR expression in spindled melanomas. The functional significance of the high level of p75 neurotrophin receptor expression may contribute to the high predisposition of perineural extension in the desmoplastic subset of spindled melanomas.

Fibroblasts can express glial fibrillary acidic protein (GFAP) in vivo

Neuropathologists use anti-glial fibrillary acidic protein (GFAP) antibodies as specific markers for glial cells, and neurobiologists use GFAP for targeting transgenes to glial cells. Since GFAP has also been detected in non-glial cells, we systematically analyzed GFAP expression in human and murine non-CNS tissues using a panel of anti-GFAP antibodies. In human tissues we confirm previously observed GFAP expression in Schwann cells, myoepithelial cells, and chondrocytes, and show for the first time GFAP expression in fibroblasts of epiglottic and auricular perichondrium, ligamentum flavum, and cardiac valves. In mice we show GFAP expression in Schwann cells, bone marrow stromal cells, chondrocytes, and in fibroblasts of dura mater, skull and spinal perichondrium, and periosteum, connective stroma of oral cavity, dental pulp, and cardiac valves. Anti-GFAP immunoblotting of human non-CNS tissues reveals protein bands with a molecular mass ranging between approximately 35 and approximately 42 kDa. In GFAP-v-src transgenic mice, whose oncogenic v-src transgene transforms GFAP expressing cells, non-CNS tumors originate from fibroblasts. We conclude that human and murine fibroblasts can express GFAP in vivo. The somatic distribution of GFAP expressing fibroblasts indicates origin from the neural crest. Development of non-CNS tumors from fibroblasts in GFAP-v-src mice functionally confirms GFAP expression in these cells.

Insulin treatment enhances expression of IGF-I in sural nerves of diabetic patients

We studied the expression of insulin-like growth factor I (IGF-I) and its receptor in sural nerves from 8 diabetic patients divided into insulin-treated (IT) and non-insulin-treated (NIT) groups, compared with 5 patients with axonal neuropathies and 4 control patients (undergoing biopsies for diagnostic purposes). Insulin-like growth factor I mRNA levels did not differ in diabetic cases compared with control subjects. In sural nerves from IT patients and axonal neuropathies, IGF-I expression was higher than in NIT subjects and diagnostic controls. Changes in IGF-I receptor mRNA levels paralleled those of the ligand. Insulin-like growth factor I immunoreactivity was higher in nerves undergoing axonal degeneration and higher in IT than NIT diabetic patients and diagnostic controls. These findings suggest that insulin treatment increases IGF-I expression in diabetic nerves. Our data do not support the hypothesis of an absolute IGF-I deficiency in human diabetic neuropathy. A Schwann cell's incapacity to increase IGF-I expression after severe nerve damage, as happens in axonal neuropathies, may be a cofactor in the pathogenesis of diabetic neuropathy.

Differential regulation of the zinc finger genes Krox-20 and Krox-24 (Egr-1) suggests antagonistic roles in Schwann cells

Krox-20 and Krox-24 (Egr-1) encode closely related zinc finger transcription factors, which interact with the same DNA target sequences. Krox-20 is required for myelination in the peripheral nervous system. Using lacZ knock-in mutant mouse lines as well as immunohistochemical analyses, we have studied the expression of Krox-20 and Krox-24 in the Schwann cell lineage during normal development and following nerve lesion in the mouse and in human neuropathies. During embryogenesis, the two genes are expressed in a successive and mutually exclusive manner, Krox-24 being restricted to Schwann cell precursors and Krox-20 to mature Schwann cells. At birth, Krox-24 is reactivated and the two genes are coexpressed. In the adult, Krox-20 is expressed in myelinating cells, while Krox-24 is restricted to nonmyelinating cells. Following nerve lesion, Krox-24 is strongly induced in Schwann cells, reinforcing the link between its expression and the nonmyelinating and/or proliferative state, whereas Krox-20 is downregulated. These data are consistent with Krox-20 and Krox-24 playing antagonistic roles during the development of the Schwann cell lineage. In particular, their balance of expression might participate in the choice between myelinating and nonmyelinating pathways.

Perineurium talin immunoreactivity decreases in diabetic neuropathy

We studied the immunolocalization of Dp116 (a 116 kDa protein product of the dystrophin gene), vinculin, talin, vimentin, desmin, spectrin and titin in the sural nerve biopsies of 25 patients with peripheral neuropathies of different origin. 4 patients presented with HMSN type 1, 4 with HMSN type 2, 2 with HNPP, 4 with CIDP, 5 with chronic axonal neuropathy of unknown origin, 3 with vasculitic neuropathy, 3 with diabetic neuropathy. Expression and localization of Dp116, vinculin, vimentin, desmin, spectrin and titin did not differ from normal control cases. Spectrin and titin immunoreactivities were absent and desmin was occasionally found in few epineurial vessels. A thin rim of Dp116 binding surrounded the outermost layer of myelin sheaths. Perineurium and epineurial vessels stained deeply for vinculin. Vimentin immunoreactivity was seen in all endoneurial, perineurial and epineurial cells. Immunoreactivity for talin was normally found at endoneurial and epineurial vessel walls, perineurial cells and epineurial fibroblasts in all the sural nerves except diabetic nerves. In the latter, whereas talin binding was normal in the vessel walls and epineurial fibroblasts, it was markedly reduced in the perineurium. On immunoblot, two bands at 235 and 190 kDa were found in the sural nerves with the antibody anti-talin, and both were reduced only in the patients with diabetic neuropathy. We postulate that decreased perineurium talin in diabetic polyneuropathy may be related to the known alterations of the tight junctions of the perineurial cells, which have been proposed to be a contributory factor to impaired permeability barrier properties.

Beta 4 integrin and other Schwann cell markers in axonal neuropathy

Schwann cell gene expression is dynamically regulated after peripheral nerve injury and during regeneration. We hypothesized that the changes in protein expression described after rat peripheral nerve injury could be used to identify single Schwann cell-axon units in human axonal neuropathy. Therefore, we performed immuno-fluorescence staining on sections of injured rat sciatic nerves compared with sections of neuropathic human sural nerves. We chose the markers beta 4 integrin, P0 glycoprotein, and glial fibrillary acidic protein (GFAP) to characterize Schwann cells, and neurofilament-heavy (NF-H) to recognize axons. Normal rat or human myelin-forming units demonstrated a sharp ring of beta 4 staining at their outer surface, P0 staining in the myelin sheath, and NF-H staining in the axon. Acutely denervated rat units transited from broken rings of beta 4 and P0 staining, to diffuse beta 4 and absent P0 and NF-H staining. Chronically denervated rat Schwann cells re-expressed beta 4 more highly, but in a diffuse, non-polarized pattern. In contrast, regenerating units re-expressed beta 4, P0, and NF-H; beta 4 staining was polarized to the outer surface of Schwann cells. Finally, GFAP staining increased progressively after injury and decreased during regeneration in the distal nerve stump. In neuropathic human sural nerves, we identified units exhibiting each of these beta 4, P0, and NF-H staining patterns; the proportion of each pattern correlated best with the extent and chronicity of axonal injury. Thus, synchronous injury of rat sciatic nerve predicts patterns of Schwann cell marker expression in human axonal neuropathy. In addition, the unique changes in the polarity of beta 4 integrin expression, in combination with changes in P0 and NF-H expression, may distinguish normal from denervated or reinnervated myelin-forming Schwann cells in human sural nerve biopsies.

Expression of common acute lymphoblastic leukemia antigen (CD 10) by myelinated fibers of the peripheral nervous system

The common acute lymphoblastic leukemia antigen (CALLA), CD10, is a 100-kDa surface glycoprotein endowed with neutral endopeptidase activity, shared by a number of hemopoietic and non-hemopoietic cells. In this report, immunohistochemical and Western blot techniques, using different anti-CD10 monoclonal antibodies, were utilized to demonstrate that CD10 is also expressed by myelin sheaths of the human peripheral nervous system (PNS), but not of the central nervous system. CD10-positive immunoreactivity appeared to be localized in the outer and inner borders of myelinated fibers, in nodes of Ranvier and in the Schmidt-Lantermann clefts, thus showing a distribution pattern very similar to that of myelin-associated glycoprotein (MAG). The above findings suggest that CD10 antigen, through its enzymatic activity, may have a functional role in the assembly and maintenance of PNS myelin. In addition, it is not known whether CD10, similarly to MAG, may be a target antigen in some PNS immune-mediated disorders.

GFAP expression of human Schwann cells in tissue culture

We have studied the expression of the intermediate filament (IF) proteins, vimentin and glial fibrillary acidic protein (GFAP), in cultured human Schwann cells (SC) from patients with different neuropathies and normal control cases. SC cultures from sural nerve biopsies of 8 subjects with axonal neuropathies, 8 with demyelinating neuropathies and 3 normal controls were included in this study and processed with double immunofluorescence technique, using anti-vimentin and anti-GFAP antibodies, during the 2nd, 4th and 6th week of culture. Five cultures incubated with anti-GFAP antibodies were also processed for immunoelectron microscopy. Specificity tests of the used antibodies were performed. We have found that: (1) cultured human SC constantly express vimentin; (2) SC from normal controls are GFAP-negative in the first period of culture; (3) SC from pathologic nerves can contain GFAP-immunoreactive IF and the percentage of GFAP-positive SC is higher in axonal than in demyelinating neuropathies; (4) during the permanence in culture human SC from both normal and pathologic cases acquire the ability to synthesize GFAP. The obtained data suggest that the removal from axonal contact and the resulting loss of myelinating function induce a cytoskeletal cellular response in human SC characterized by the cytoplasmic accumulation of GFAP-immunoreactive IF.