Antisera raised in rabbits to three peptides corresponding to amino acid sequences found in human O6-alkylguanine-DNA alkyltransferase were used to study the fate of the alkyltransferase protein in human colon tumor cells after exposure to N-methyl-N'-nitro-N-nitrosoguanidine or to O6-benzylguanidine. Under these conditions, the alkyltransferase protein becomes inactivated, presumably by the conversion of its cysteine acceptor site to S-methylcysteine or S-benzylcysteine respectively. It was found that the protein was rapidly degraded after such inactivation both in intact cells and in cell-free extracts. It is probable that a conformational change in the protein is brought about by conversion of the alkyltransferase to the inactive form by alkylation of the cysteine acceptor site. This change may render the protein very sensitive to proteolytic degradation. The rapid degradation of the inactive form of the protein may serve as a signal for its resynthesis but in the short term ensures that its reactivation by regeneration of the cysteine acceptor site is unlikely to occur to any significant extent. The short half-life of the inactivated alkyltransferase protein makes it probable that measurement of the content of the alkyltransferase protein by immunohistochemistry, which is likely to measure the sum of the active and inactivated forms of the protein, will nevertheless yield an accurate estimation of the cellular capacity to repair O6-methylguanine provided that procedures with sufficient specificity and affinity can be developed.

Antisera were raised in rabbits to three peptides which correspond to sequences of amino acids present in human O6-alkylguanine-DNA alkyltransferase (residues 1-11, 8-20 and 197-207). These antisera recognized the alkyltransferase protein on Western blots of proteins from a number of Mer+ human tumor cell lines but this protein was found to be absent from Mer- tumor cell lines. The alkyltransferase protein was also detected by these antisera in some extracts from human liver samples but other human liver extracts having a high alkyltransferase activity failed to react with antibodies directed towards the carboxyl terminus of the protein, suggesting that part of this region can be removed by protease action without loss of activity or that there is genetic variability in this region. This result indicates that studies with a number of antisera or with an antibody known to be directed towards an essential, invariant region of the alkyltransferase will be needed for unequivocal detection of the alkyltransferase by antibody screening. The immunoreactive human alkyltransferase protein was absent from CHO cell lines that had previously been selected for alkyltransferase expression after transfection with human DNA. It is therefore probable that the hamster gene has been reactivated in these cells.

Gangliosides have been used to treat specific central nervous system lesions and to inhibit proliferation of neuroblastoma cells in vitro. However, the mechanisms by which they (1) cross the blood-brain barrier and (2) inhibit cell proliferation have not been clearly defined. Evidence is presented in support of the hypotheses that (1) serum albumin functions in the transport of gangliosides across the blood-brain barrier, and (2) when gangliosides inhibit cell proliferation, they do so by inhibiting the activity of DNA polymerases alpha and beta.

Addition of gangliosides to neuroblastoma cells maintained in vitro has been shown to enhance neuritogenesis. Although the mechanism by which they exert this effect is unknown, it has been postulated that they may act by adhering to cell surface proteins. In this article, we describe the isolation and identification of an S20Y murine neuroblastoma cell protein recognized by a monoclonal antibody that was prepared against putative GM1‐binding proteins and shown to inhibit GM1‐enhanced neuritogenesis. The protein identified was nonmuscle myosin heavy chain B, which appears to function in neurite formation but may not adhere to gangliosides.