CIRI seminar: Alessia Zamborlini

Loading Map....

Date / Heure
Date(s) - 05/02/2020
11 h 00 min - 12 h 00 min

salle des Thèses Chantal Rabourdin-Combes

Alessia Zamborlini (Université Paris Sud – France) : SIM-dependent SUMOylation of K595 together with unphosphorylated T592 is essential to render human SAMHD1 restriction-competent

SAMHD1 is a cellular triphosphohydrolase (dNTPase) that inhibits HIV-1 replication in non-cycling immune cells by reducing the levels of cellular dNTPs below a threshold required for efficient reverse transcription. As SAMHD1 is ubiquitously expressed, the mechanisms underlying the regulation of its cell cycle-specific restriction activity has been deeply investigated in recent years. It is now well established that residue threonine (T) 592 of SAMHD1 is phosphorylated by cyclin/CDK complexes in a cell cycle-dependent manner, while it is hypo-phosphorylated in non-cycling cells refractory to HIV-1. The role of phosphorylation in the down-regulation of SAMHD1 antiviral functions is supported by finding that mutation of T592 into a phosphomimetic residue (D or E) abolishes restriction. However, the SAMHD1 T592D/E variant efficiently hydrolyzes dNTPs indicating that SAMHD1-mediated restriction may not exclusively rely on its dNTPase activity, and/or is regulation may not only depend on phosphorylation of T592.

We have established that inhibiting the SUMOylation process favors the infection of macrophages by HIV-1 and HIV-2∆Vpx, but not HIV-2. The existence of a correlation between the sensitivity to both small molecule inhibitors of SUMOylation and SAMHD1-dependent restriction led us to investigate whether the SUMO conjugation plays a direct role in the antiviral mechanism of this protein. We found that SAMHD1 harbors three major SUMO targeted sites (K469, K595 and 622) and a surface-exposed SIM. Functional studies revealed that blocking SUMO attachment to K595 is sufficient to humper the antiviral, but not dNTPase, function of SAMHD1. Importantly, mutation of the SIM precludes both viral restriction and K595 SUMOylation, strengthening the requirement of SUMO conjugation to this residue for the establishment of a SAMHD1-dependent antiviral state. We also investigated whether a connection exist between phosphorylation of T592 and SUMOylation of the adjacent K595 residue, and found that the two modifications occur independently of one another. Interestingly, we observed that SUMOylation-deficient K595A SAMHD1 variant is hypophosphorylated on residue T592, but is nevertheless antivirally inactive. In addition, we found that combining mutations disrupting K595 SUMOylation with the phosphoablative T592A change abolished SAMHD1-mediated restriction.

In conclusion, our data demonstrate that SUMOylation of K595 and absence of T592 phosphorylation are prerequisites for SAMHD1 to be restriction competent.

Contact: Renaud MAHIEUX