CNR - Institute of Neuroscience CNR
Institute of Neuroscience


Virotherapy by targeting citolytic viruses for the selective eradication of neural/glial cells latently infected by HIV

Redirecting the cellular tropism of alphatogaviruses for the the selective killing of neuro/glial cells latently infected by HIV-1

The research project concerns the construction of a virus with cytocidal action, caracterized by the capability to penetrate and selectively kill cells latently infected by HIV, which may result in the complete recovery of patients with AIDS. In fact, while it is thought that HAART therapy may virtually eliminate all circulating virus, for a complete and definitive recovery, it is necessary to get rid of the natural virus reservoir made by the pool of infected cells in which HIV genome is stably integrated.

To this aim, the Sindbis virus genome has been modified to alter its natural tropism, by selectively addressing the virus against target cells of both the immune (CD4-positive T-helper cells, macrophages and dendritic cells) and the nervous system (neurons, glial cells and astrocytes) expressing on their surface the gp120 molecule of HIV-1. The N-terminal portion of the structural surface E2 protein of Sindbis virus, which is responsible of the viral attachment to the target cells, has been replaced by the genome fragment coding for the D1 domain of the human CD4 glycoprotein, which is known to contain the binding site of gp120 of HIV-1.

The results of the analyses performed to characterize and determine the biological properties of the first virus we engineered, called Sindbis/D1-CD4, showed that Sindbis/D1-CD4 was probably produced by trasfected cells in a too low amount to be able to infect target cells. We hypothesized that the insertion of the D1 CD-4 domain into the Sindbis virus might have reduced the efficacy of the assembly process of the chimeric proteins.

In particular, we hypothesized that the first cloning might have hampered the proteolitic cleavage of E3 protein from the precursor pE3-E2, which is important for the expression on the surface of the E2 mature protein of Sindbis virus. We proceeded therefore to a new cloning strategy by increasing the distance between the human CD4 insertion and the proteolitic cleavage point and by preparing different constructs which might favor both membrane targeting and the correct folding of the D1 domain on viral surface.

Different genome fragments were prepared, derived from the human CD4 and containing the D1 domain, with or without the leader sequence (S), joined or not to the D2 domain. These fragments, which were called D1S, D1, D12S and D12, were used to engineer four different Sindbis viruses, which were designated SinD1S, SinD1, SinD1/2S, SinD1/2, respectively.

After infection, taking advantage of the apoptotic effect of the non-structural proteins of Sindbis virus, the new chimeric virions SinD1S, SinD1, SinD12S, SinD12 should selectively kill all HIV-1-infected cells expressing gp120 on their surface.


  • Rigano MM, Manna C, Giulini A, Pedrazzini E, Capobianchi M, Castilletti C, Di Caro A, Ippolito G, Beggio P, De Giuli Morghen C, Monti L, Vitale A, Cardi T (2009) Transgenic chloroplasts are efficient sites for high-yield production of the vaccinia virus envelope protein A27L in plant cellsdagger. Plant Biotechnol. J. 7:577-91.
  • Pozzi E, Zanotto C, Pacchioni S, De Giuli Morghen C, Radaelli A (2009) MHC-restricted cytotoxic T-lymphocyte assay: an improved method based on normal and SV40-immortalized rabbit epidermal target cells. J. Virol. Methods 155:77-81.


  • G. Palù, Dean, School of Medicine, University of Padova, Italy.
  • B. Combadière, H. Salpétrière, Paris, France.
  • C. Christiane Stahl-Hennig, German Primate Centre, Göttingen, Germany.


PI photo

Carlo De Giuli-Morghen

Contact information

email  E-mail

email  +39 02 5031 7062

Participating staff