Stevens and Beutel studied the activity of several carbamate anticholinesterases. Among other things, they found that the (4-trimethylammonio)phenyl dimethylcarbamate iodide (The para-analog of neostigmine) is significantly less active than neostigmine (Subcutaneous LD50 in mice of the para-analog was 120 mg/kg compared to 0.45 mg/kg for neostigmine). However, ((4-trimethylammonio)-3-isopropyl)phenyl dimethylcarbamate iodide, also known as TL-599, was more toxic than neostigmine, having an LD50 of 0.075 mg/kg. How exactly do the differences in the structures of TL-599, neostigmine, and the para-analog of neostigmine contribute to their differences in toxicity?
Stevens, J. R., & Beutel, R. H. (1941). Physostigmine Substitutes. Journal of the American Chemical Society, 63(1), 308–311. doi:10.1021/ja01846a078