The tubulin molecule is a heterodimer of á and â subunits, each of which consists of multiple isotypes differing in amino acid sequence and encoded by different genes. The specific differences among the isotypes have been widely conserved in vertebrate evolution, suggesting that the differences are physiologically significant. The vertebrate â isotypes have fairly distinct tissue distributions. âI is expressed in most tissues. âIV is very similar with a particularly high concentration in cilia and flagella. âII is particularly abundant in brain, nerves and muscles and occurs in low amounts in several tissues and also occurs in the nuclei of a variety of cancer cells. âIII occurs largely in neurons and testes and in low amounts in a very small number of tissues. âVI is restricted to hematopoietic tissues and the distribution of V is still largely unknown. Among the â isotypes, âIII is particularly intriguing for the following reasons. 1. âIII lacks the easily oxidized and very widely conserved cys239 (present in âI, âII and âIV. 2. âIII contains an unusual cys124 which is virtually unknown in non-animal â-tubulins; together with the widely conserved cys127 and cys129 this constitutes a cysteine cluster. 3. âIII is phosphorylated at a serine in the C-terminal domain. 4. Despite having a very narrow distribution in normal tissues, âIII is widely expressed among cancers as the Katsetos laboratory has shown.
We have used a variety of approaches to explore the possible functions of âIII. We have observed that, in breast cancer excisions, the concentration of âIII is positively correlated with the grade of the tumor. We have also observed that the few normal tissues that express âIII also are rich in free radicals and reactive oxygen species. We hypothesize that âIII may have the functions of a) protecting microtubules from free radicals and b) possibly protecting other cell structures as well. The latter hypothesis is consistent with the report from the Kavallaris group that silencing âIII expression in cancer cells increases the cells susceptibility not only to drugs that target tubulin but to other drugs as well. Both hypotheses are consistent with our observation that treatment of breast cancer cells with the oxidizing agent diamide leads to a five-fold over-expression of âIII.
In a purely biochemical approach, we have purified the áâII, áâIII and áâIV dimers from bovine brain and examined their ability to form microtubules in vitro in the presence of a superoxide generating system. We find that the extent of microtubule assembly of áâII and áâIV is inhibited by 25% and 10%, respectively in the presence of the free radical superoxide. In contrast, assembly of áâIII is not inhibited at all. Interestingly, the rates of assembly of áâII and áâIV are dramatically inhibited by superoxide while that of áâIII actually increases. This is consistent with the hypothesis that lack of cys239 may allow áâIII to assemble in the presence of free radicals. Interestingly, assembly of unfractionated tubulin is not inhibited by superoxide indicating that the presence of âIII, which accounts for 25% of brain â-tubulin, can protect the other isotypes as well. Perhaps these observations account for the presence of âIII in tumors, which are very rich in free radicals.
In another approach, we have examined the roles of the âI, âII and âIII isotypes in SK-N-SH neuroblastoma cells, which are a model system for differentiating neurons. We have prepared siRNA's for the individual isotypes and silenced them one at a time, meanwhile inducing the cells to differentiate and form neurons. Silencing âI makes little difference to cell morphology but causes about a 50% decrease in cell viability. Silencing âII causes a 20% decrease in cell viability but also strongly blocks neurite outgrowth. Silencing âIII also causes a 20% decrease in cell viability and causes some inhibition of neurite outgrowth.
Our results suggest that âI is best adapted to carry out normal microtubule functions, including mitosis and intracellular transport, that âII is specifically required for neuronal differentiation and that âIII has multiple roles, protecting cells from free radicals and also promoting neuronal differentiation.
Speaker: Richard F. Luduena
Department of Biochemistry, University of Texas Health Science Center at San Antonio, San Antonio, TX 78229-3900, USA
Time: Wednesday, 15 October 2008, 13:00