Ellen J. Lehning

Neurological evaluation of toxic axonopathies in rats: acrylamide and 2,5-hexanedione.

This research was conducted to determine which neurological test or combination of tests can provide sufficient functional information to compliment biochemical or morphological endpoints in mechanistic studies of toxic axonopathies. Using several neurological indices, we evaluated the effects of two prototypical neurotoxicants that cause distal axonopathy: acrylamide monomer (ACR) and 2,5-hexanedione (HD). For each toxicant, rats were exposed to two daily dosing rates (ACR, 50 mg/kg per day i.p. or 21 mg/kg per day, p.o.; HD, 175 or 400 mg/kg per day, p.o.) and neurological endpoints were determined two to three times per week. Specific tests included observations of spontaneous locomotion in an open field, and measurements of hindlimb landingfoot splay, forelimb and hindlimb grip strength and the hindlimb extensor thrust response. For all neurological parameters, the magnitude of defect induced by either neurotoxicant was not related to daily dose-rate, e.g. both the lower and higher ACR dose-rates produced the same degree of neurological dysfunction. Instead, dose-rate determined onset and progression of neurotoxicity, e.g. the higher ACR dose-rate produced moderate neurotoxicity after approximately 8 days of intoxication, whereas the lower dose-rate caused moderate neurotoxicity after 26 days. Regardless of dose-rate, ACR-exposed rats exhibited gait abnormalities (ataxia, splayed hindlimbs), in conjunction with increased landing hindfoot spread and decreased hindlimb grip strength and extensor thrust HD intoxicated rats exhibited hindlimb muscle weakness as indicated by a gait abnormality (dropped hocks) and decreases in grip strength and the extensor thrust response. However, hindlimb landingfoot spread was not affected by HD exposure. For both neurotoxicants, gait changes preceded or coincided with alterations in other neurologic indices. These results suggest that observations of spontaneous behavior in an open field represent a practical approach to assessing temporal development and extent of neurological dysfunction induced by axonopathic toxicants such as ACR and HD.

Nerve terminals as the primary site of acrylamide action: a hypothesis.

Acrylamide (ACR) is considered to be prototypical among chemicals that cause a central-peripheral distal axonopathy. Multifocal neurofilamentous swellings and eventual degeneration of distal axon regions in the CNS and PNS have been traditionally considered the hallmark morphological features of this axonopathy. However, ACR has also been shown to produce early nerve terminal degeneration of somatosensory, somatomotor and autonomic nerve fibers under a variety of dosing conditions. Recent research from our laboratory has demonstrated that terminal degeneration precedes axonopathy during low-dose subchronic induction of neurotoxicity and occurs in the absence of axonopathy during higher-dose subacute intoxication. This relationship suggests that nerve terminal degeneration, and not axonopathy, is the primary or most important pathophysiologic lesion produced by ACR. In this hypothesis paper, we review evidence suggesting that nerve terminal degeneration is the hallmark lesion of ACR neurotoxicity, and we propose that this effect is mediated by the direct actions of ACR at nerve terminal sites. ACR is an electrophile and, therefore, sulfhydryl groups on presynaptic proteins represent rational molecular targets. Several presynaptic thiol-containing proteins (e.g. SNAP-25, NSF) are critically involved in formation of SNARE (soluble N-ethylmaleimide (NEM)-sensitive fusion protein receptor) complexes that mediate membrane fusion processes such as exocytosis and turnover of plasmalemmal proteins and other constituents. We hypothesize that ACR adduction of SNARE proteins disrupts assembly of fusion core complexes and thereby interferes with neurotransmission and presynaptic membrane turnover. General retardation of membrane turnover and accumulation of unincorporated materials could result in nerve terminal swelling and degeneration. A similar mechanism involving the long-term consequences of defective SNARE-based turnover of Na+/K(+)-ATPase and other axolemmal constituents might explain subchronic induction of axon degeneration. The ACR literature occupies a prominent position in neurotoxicology and has significantly influenced development of mechanistic hypotheses and classification schemes for neurotoxicants. Our proposal suggests a reevaluation of current classification schemes and mechanistic hypotheses that regard ACR axonopathy as a primary lesion.

Mechanisms of Injury‐Induced Calcium Entry into Peripheral Nerve Myelinated Axons: Role of Reverse Sodium‐Calcium Exchange

Abstract: To investigate the route of axonal Ca2+ entry during anoxia, electron probe x‐ray microanalysis was used to measure elemental composition of anoxic tibial nerve myelinated axons after in vitro experimental procedures that modify transaxolemmal Na+ and Ca2+ movements. Perfusion of nerve segments with zero‐Na+/Li+‐substituted medium and Na+ channel blockade by tetrodotoxin (1 µM) prevented anoxia‐induced increases in Na and Ca concentrations of axoplasm and mitochondria. Incubation with a zero‐Ca2+/EGTA perfusate impeded axonal and mitochondrial Ca accumulation during anoxia but did not affect characteristic Na and K responses. Inhibition of Na+‐Ca2+ exchange with bepridil (50 µM) reduced significantly the Ca content of anoxic axons although mitochondrial Ca remained at anoxic levels. Nifedipine (10 µM), an L‐type Ca2+ channel blocker, did not alter anoxia‐induced changes in axonal Na, Ca, and K. Exposure of normoxic control nerves to tetrodotoxin, bepridil, or nifedipine did not affect axonal elemental composition, whereas both zero‐Ca2+ and zero‐Na+ solutions altered normal elemental content characteristically and significantly. The findings of this study suggest that during anoxia, Na+ enters axons via voltage‐gated Na+ channels and that subsequent increases in axoplasmic Na+ are coupled functionally to extraaxonal Ca2+ import. Intracellular Na+‐dependent, extraaxonal Ca2+ entry is consistent with reverse operation of the axolemmal Na+‐Ca2+ exchanger, and we suggest that this mode of Ca2+ influx plays a general role in peripheral nerve axon injury.

Effects of ion channel blockade on the distribution of Na, K, Ca and other elements in oxygen-glucose deprived CA1 hippocampal neurons.

The pathophysiology of brain ischemia and reperfusion injury involves perturbation of intraneuronal ion homeostasis. To identify relevant routes of ion flux, rat hippocampal slices were perfused with selective voltage- or ligand-gated ion channel blockers during experimental oxygen-glucose deprivation and subsequent reperfusion. Electron probe X-ray microanalysis was used to quantitate water content and concentrations of Na, K, Ca and other elements in morphological compartments (cytoplasm, mitochondria and nuclei) of individual CA1 pyramidal cell bodies. Blockade of voltage-gated channel-mediated Na+ entry with tetrodotoxin (1 microM) or lidocaine (200 microM) significantly reduced excess intraneuronal Na and Ca accumulation in all compartments and decreased respective K loss. Voltage-gated Ca2+ channel blockade with the L-type antagonist nitrendipine (10 microM) decreased Ca entry and modestly preserved CA1 cell elemental composition and water content. However, a lower concentration of nitrendipine (1 microM) and the N-, P-subtype Ca2+ channel blocker omega-conotoxin MVIIC (3 microM) were ineffective. Glutamate receptor blockade with the N-methyl-D-aspartate (NMDA) receptor-subtype antagonist 3-(2-carboxypiperazin-4-yl) propyl-1-phosphonic acid (CPP; 100 microM) or the alpha-amino-3-hydroxy-5-methyl-4-isoazole propionic acid (AMPA) receptor subtype blocker 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 microM/100 microM glycine) completely prevented Na and Ca accumulation and partially preserved intraneuronal K concentrations. Finally, the increase in neuronal water content normally associated with oxygen-glucose deprivation/reperfusion was prevented by Na+ channel or glutamate receptor blockade. Results of the present study demonstrate that antagonism of either postsynaptic NMDA or AMPA glutaminergic receptor subtypes provided nearly complete protection against ion and water deregulation in nerve cells subjected to experimental ischemia followed by reperfusion. This suggests activation of ionophoric glutaminergic receptors is involved in loss of neuronal osmoregulation and ion homeostasis. Na+ channel blockade also effectively diminished neuronal ion and water derangement during oxygen-glucose deprivation and reperfusion. Prevention of elevated Nai+ levels is likely to provide neuroprotection by decreasing presynaptic glutamate release and by improving cellular osmoregulation, adenosine triphosphate utilization and Ca2+ clearance. Thus, we suggest that voltage-gated tetrodotoxin-sensitive Na+ channels and glutamate-gated ionotropic NMDA or AMPA receptors are important routes of ion flux during nerve cell injury induced by oxygen-glucose deprivation/reperfusion.

Intracellular concentrations of major ions in rat myelinated axons and glia: calculations based on electron probe X-ray microanalyses.

Abstract: Electron probe x‐ray microanalysis (EPMA) was used to measure water content (percent water) and dry weight elemental concentrations (in millimoles per kilogram) of Na, K, Cl, and Ca in axoplasm and mitochondria of rat optic and tibial nerve myelinated axons. Myelin and cytoplasm of glial cells were also analyzed. Each anatomical compartment exhibited characteristic water contents and distributions of dry weight elements, which were used to calculate respective ionized concentrations. Free axoplasmic [K+] ranged from ≈155 mM in large PNS and CNS axons to ≈120–130 mM in smaller fibers. Free [Na+] was ≈15–17 mM in larger fibers compared with 20–25 mM in smaller axons, whereas free [Cl−] was found to be 30–55 mM in all axons. Because intracellular Ca is largely bound, ionized concentrations were not estimated. However, calculations of total (free plus bound) aqueous concentrations of this element showed that axoplasm of large CNS and PNS axons contained ≈0.7 mM Ca, whereas small fibers contained 0.1–0.2 mM. Calculated ionic equilibrium potentials were as follows (in mV): in large CNS and PNS axons, EK = −105, ENa = 60, and ECl = −28; in Schwann cells, EK = −107, ENa = 33, and ECl = −33; and in CNS glia, EK = −99, ENa = 36, and ECl = −44. Calculated resting membrane potentials were as follows (in mV, including the contribution of the Na+,K+‐ATPase): large axons, about −80; small axons, about −72 to −78; and CNS glia, −91. ECl is more positive than resting membrane potential in PNS and CNS axons and glia, indicating active accumulation. Direct EPMA measurement of elemental concentrations and subsequent calculation of ionized fractions in axons and glia offer fundamental neurophysiological information that has been previously unattainable.

Acrylamide neuropathy. III. Spatiotemporal characteristics of nerve cell damage in forebrain.

Previous studies of acrylamide (ACR) neuropathy in rat PNS [Toxicol. Appl. Pharmacol. (1998) 151:211-221] and in spinal cord, brainstem and cerebellum [NeuroToxicology (2002a) 23:397-414; NeuroToxicology (2002b) 23:415-429] have suggested that axon degeneration was not a primary effect and was, therefore, of unclear neurotoxicological significance. To conclude our studies of neurodegeneration in rat CNS during ACR neurotoxicity, a cupric silver stain method was used to define spatiotemporal characteristics of nerve cell body, dendrite, axon and terminal argyrophilia in forebrain regions and nuclei. Rats were exposed to ACR at a dose-rate of either 50 mg/kg per day (i.p.) or 21 mg/kg per day (p.o.) and at selected times brains were removed and processed for silver staining. Results show that intoxication at either ACR dose-rate produced a terminalopathy, i.e. nerve terminal degeneration and swelling were present in the absence of significant argyrophilic changes in neuronal cell bodies, dendrites or axons. Exposure to the higher ACR dose-rate caused early onset (day 5), widespread nerve terminal degeneration in most of the major forebrain areas, e.g. cerebral cortex, thalamus, hypothalamus and basal ganglia. At the lower dose-rate, nerve terminal degeneration in the forebrain developed early (day 7) but exhibited a relatively limited spatial distribution, i.e. anteroventral thalamic nucleus and the pars reticulata of the substantia nigra. Several hippocampal regions were affected at a later time point (day 28), i.e. CA1 field and subicular complex. At both dose-rates, argyrophilic changes in forebrain nerve terminals developed prior to the onset of significant gait abnormalities. Thus, in forebrain, ACR intoxication produced a pure terminalopathy that developed prior to the onset of significant neurological changes and progressed as a function of exposure. Neither dose-rate used in this study was associated with axon degeneration in any forebrain region. Our findings indicate that nerve terminals were selectively affected in forebrain areas and, therefore, might be primary sites of ACR action.

Acrylamide neuropathy. II. Spatiotemporal characteristics of nerve cell damage in brainstem and spinal cord.

Previous studies of acrylamide (ACR) neuropathy in rat PNS [Toxicol. Appl. Pharmacol. 151 (1998) 211] and cerebellum [Neurotoxicology, 2002a] have suggested that axon degeneration was not a primary effect and was, therefore, of unclear neurotoxicological significance. To continue morphological examination of ACR neurotoxicity in CNS, a cupric silver stain method was used to define spatiotemporal characteristics of nerve cell body, dendrite, axon and terminal degeneration in brainstem and spinal cord. Rats were exposed to ACR at a dose-rate of either 50 mg/kg per day (i.p.) or 21 mg/kg per day (p.o.), and at selected times brains and spinal cord were removed and processed for silver staining. Results show that intoxication at the higher ACR dose-rate produced a nearly pure terminalopathy in brainstem and spinal cord regions, ie. widespread nerve terminal degeneration and swelling were present in the absence of significant argyrophilic changes in neuronal cell bodies, dendrites or axons. Exposure to the lower ACR dose-rate caused initial nerve terminal argyrophilia in selected brainstem and spinal cord regions. As intoxication continued, axon degeneration developed in white matter of these CNS areas. At both dose-rates, argyrophilic changes in brainstem nerve terminals developed prior to the onset of significant gait abnormalities. In contrast, during exposure to the lower ACR dose-rate the appearance of axon degeneration in either brainstem or spinal cord was relatively delayed with respect to changes in gait. Thus, regardless of dose-rate, ACR intoxication produced early, progressive nerve terminal degeneration. Axon degeneration occurred primarily during exposure to the lower ACR dose-rate and developed after the appearance of terminal degeneration and neurotoxicity. Spatiotemporal analysis suggested that degeneration began at the nerve terminal and then moved as a function of time in a somal direction along the corresponding axon. These data suggest that nerve terminals are a primary site of ACR action and that expression of axonopathy is restricted to subchronic dosing-rates.

Mechanisms of injury-induced calcium entry into peripheral nerve myelinated axons : in vitro anoxia and ouabain exposure

In the present investigation, electron probe X-ray microanalysis was used to characterize the effects of in vitro ouabain (2 mM) or anoxia on elemental composition (e.g. Na, K, Ca) and water content of rat peripheral (tibial) nerve myelinated axons and Schwann cells. Results showed that independent of axon size, both ouabain and anoxia markedly increased axoplasmic Na and decreased K concentrations. However, only anoxia was associated with significant elevation of axonal Ca content. Mitochondrial areas from ouabain- or anoxia-exposed fibers exhibited changes in element and water contents that were similar to axoplasmic alterations. Schwann cells and myelin displayed small increases in Na and substantial losses of K in response to ouabain exposure. In contrast, these glial compartments were relatively resistant to anoxia as indicated by the modest and delayed nature of the elemental changes. Nonetheless, neither treatment significantly affected glial Ca concentrations. Our results suggest that Ca2+ accumulation in peripheral nerve axons is complex and involves not only deregulation of Na+ and K+ but other fundamental pathogenic changes as well. In addition to providing baseline information, we have identified an in vitro model (anoxia) which features Ca2+ build-up in PNS myelinated axons. Thus, the present study offers a foundation for investigation into mechanisms of Ca2+ entry following peripheral nerve injury.

Rubidium Uptake and Accumulation in Peripheral Myelinated Internodal Axons and Schwann Cells

Abstract: To study mechanisms of K+ transport in peripheral nerve, uptake of rubidium (Rb+), a K+ tracer, was characterized in rat tibial nerve myelinated axons and glia. Isolated nerve segments were perfused with zero‐K+ Ringers solutions containing Rb+ (1–20 mM) and x‐ray microanalysis was used to measure water content and concentrations of Rb, Na, K, and Cl in internodal axoplasm, mitochondria, and Schwann cell cytoplasm and myelin. Both axons and Schwann cells were capable of removing extracellular Rb+ (Rb+o) and exchanging it for internal K+. Uptake into axoplasm, Schwann cytoplasm, and myelin was a saturable process over the 1–10 mM Rb+o concentration range, although corresponding axoplasmic uptake rates were higher than respective glial velocities. Mitochondrial accumulation was a linear function of axoplasmic Rb+ concentrations, which suggests involvement of a nonenzymatic process. At 20 mM Rb+o, a differential stimulatory response was observed; i.e., axoplasmic Rb+ uptake velocities increased more than fivefold relative to the 10 mM rate, and glial cytoplasmic uptake rose almost threefold. Finally, Rb+o uptake rate into axons and glia was completely inhibited by ouabain (2–4 mM) exposure or incubation at 4°C. These results suggest that Rb+ uptake into peripheral nerve internodal axons and Schwann cells is mediated by Na+,K+‐ATPase activity and implicate the presence of axonal‐ and glial‐specific Na+ pump isozymes.

Reoxygenation of anoxic peripheral nerve myelinated axons promotes re-establishment of normal elemental composition

Previously we have shown that in vitro anoxia of rat peripheral nerve myelinated axons causes sequential deregulation of axoplasmic Na, K and Ca; i.e., an initial influx of Na and loss of K is coupled to subsequent Ca accumulation [7]. In the present study, we examined the ability of PNS axons to recover normal elemental composition following oxygen deprivation. Thus, electron probe X-ray microanalysis was used to determine the effects of post-anoxia reoxygenation on the concentrations of elements (i.e., Na, K, Cl, Ca, Mg, P and S) in rat posterior tibial nerve myelinated axons and Schwann cells. Results indicate that following 180 min of anoxia, peripheral nerve reoxygenation (60 and 120 min) promoted progressive recovery of normal elemental composition in axoplasm and mitochondria of small, medium and large diameter tibial nerve fibers. Our observations also indicate that small axons recovered normal elemental concentrations more rapidly than larger counterparts. Schwann cells and myelin exhibited only modest elemental disruption during anoxia from which reoxygenation promoted full reparation. The ability of peripheral nerve axons to restore normal elemental composition during post-anoxia reoxygenation is in marked contrast to the exacerbation of elemental deregulation which ensued during in vitro reoxygenation of anoxic rat CNS fibers [14]. This differential response to reoxygenation represents a fundamental difference in the pathophysiology of myelinated axons in the CNS and PNS.