Serum neurotoxicity was studied by adding whole or fractionated serum (adult human, adult horse, or newborn calf) to neuron-rich cultures prepared from various regions of embryonic (Days 14-15) rat brain, including spinal cord, ventral mesencephalon, cerebellum, septum, and striatum. Effects of serum were also tested on several types of embryonic non-neuronal cells (skeletal muscle myotubes, cardiac muscle myocytes, and fibroblasts from skin and lung). Serum concentrations of 50% or more killed more than 95% of all neurons within 3 days. Serum concentrations as low as 10% also killed some neurons, especially those from cerebellum. Septal, cerebellar, and spinal cord neurons were more sensitive than striatal or mesencephalic neurons. All the tested non-neuronal cells survived much better than neurons at serum concentrations of 20% or more. Neurotoxicity was present in both fresh (human) and previously frozen (human and animal) sera, and affected both young (4 days in vitro) and older (42 days in vitro) cultures. Neurotoxicity was greatly diminished by heating the serum to 56°C for 30 min. Experiments indicated that serum toxicity was not due to lipoprotein, complement, or tumor necrosis factor. All serum neurotoxicity was retained by an ultrafilter with a nominal molecular weight cutoff of 10 kDa. The profile of neurotoxicity following gel filtration at neutral pH was variable, with high toxicity most consistently observed in fractions with apparent molecular weights exceeding 100 kDa, and variable degrees of toxicity at lower molecular weights.
ASJC Scopus subject areas