Role of HTLV Tax proteins post-Translational modifications in the induction of signaling pathways

The two retroviruses HTLV-1 and HTLV-2 present a very close genomic structure but differ significantly in pathogenicity. In particular, HTLV-1 is associated to Adult T-cell Leukemia, a rare and aggressive T-cell malignancy and is also the causative agent of the HTLV associated myelophaty/tropical spastic paraperesis (HAM/TSP), a progressive and disabling demyelinating neurological disease. In contrast to the case for HTLV-1, HTLV-2 has not been linked to lymphoproliferative pathologies though it appears to be correlated to rare neurological disorders similar to HAM/TSP. This difference is generally attributed to the properties of their transactivating Tax proteins, Tax1 and Tax2, both of which activate gene expression via the ATF/CREB and NF-κB pathways. Since Tax cellular role is strictly governed by post-translational modifications, this study was aimed at comparing the role of specific lysine residues in the processes of ubiquitination and sumoylation of Tax1 and Tax2. Five of ten Tax1 lysine residues, namely from K1 to K10, were found to be the major targets of Tax-1 modifications. More precisely, lysines from K4 to K8 are known to be essential for Tax1 ubiquitination, while lysine K7 and K8 are more directly involved in Tax1 sumoylation. Moreover, both ubiquitination and sumoylation of Tax1 were found to be required for NF-κB activation and localization in nuclear bodies and in Golgi-apparatus related structures. To determine whether ubiquitination and sumoylation control Tax2B transcription activity and to outline possible differences between Tax1 and Tax2B, a series of Tax lysine-to-arginine mutants were constructed and their ubiquitination and sumoylation status, their capacity to activate gene expression via the NF-κB and ATF/CREB pathways and their intracellular distribution with the corresponding Tax1 mutants were compared. The results obtained indicated a notable difference between the lysine sites involved in Tax2B sumoylation. More precisely, other Tax2B lysines appear to be involved in Tax2B sumoylation, since the mutation in Tax2B K7-8R, contrary to Tax1 K7-8R, had little effects on Tax2B sumoylation level and pattern. More importantly, when all six Tax2B central lysines were mutated to arginines, a detectable level of sumoylation was maintained. These results also explain the significant differences between Tax1 and Tax2B showing that Tax1 K7-8R and K4-8R were defective for NF-κB activation whereas the homologous Tax2B K7-8R and K3ii-8R maintained 75% and 25% of NF-κB activation. As expected , Tax1 and Tax2B mutants with substitution of all lysine residues had undetectable level of sumoylation. The non-exclusive role of lysines K7 and K8 in Tax2B sumoylation was confirmed by confocal analysis of intracellular distribution of Tax1 and Tax2B mutants showing that Tax1 K7-8R does not accumulate in nuclear bodies, whereas the homologous Tax2 K7-8R clearly shows its presence in nuclear bodies. Taken together these results indicate that though Tax2B lysine sumoylation pattern is different from that of Tax1, the NF-κB activation mechanism by the two Tax proteins is linked to the same sumoylation process.