RNAi as Antiviral Therapy: The HIV-1 Case - RNA Interference from Biology to Therapeutics - page 224

Combinatorial drug/RNAi approaches are essential to restrict HIV-1 evolution
and to prevent viral escape that consequently will lead to therapeutic failure [ 102 ] .
A variety of strategies have been described for multiplexing of shRNA cassettes in
a single therapeutic vector. As repeat sequences should be avoided in the lentiviral
vector to prevent recombination-mediated deletions, the multiple shRNA cassettes
generally use separate polymerase III promoters or a combination of polymerase II
and III promoters [ 103 ]. Multiplexed siRNAs can also be expressed from a single
transcript. We and others have developed extended-shRNAs that are processed into
two or maximally three functional siRNAs [ 104- 106 ] . Another strategy uses truly
long-hairpin RNAs (lhRNAs) that should encode numerous siRNAs [ 107- 109 ] .
A disadvantage of the lhRNA approach is that it is a priori unknown what siRNA
molecules will be produced and whether these molecules will be active inhibitors
[ 99 ], and it was recently demonstrated that such constructs express a very low siRNA
level [ 105 ]. Polycistronic miRNA transcripts have also been developed [ 110 ] .
Various groups have reported toxicity of shRNAs [ 111- 114 ], which can perhaps be
solved by inserting the siRNA sequence into a natural miRNA backbone [ 112 ] .
Conditional expression of the siRNA molecules will increase the safety of a ther-
apeutic vector. For instance, one would like to avoid shRNA expression in trans-
duced hematopoietic stem cells that still have to undergo hematopoiesis, a process
that is particularly susceptible to changes in the RNAi machinery. Tissue-specific
miRNA expression has been described for several organs, including the liver [ 115 ] .
Another option is the design of constructs that are induced by HIV-1 infection [ 89 ] .
Selective expression in HIV-1 susceptible cells would be an elegant way to restrict
putative saturation and off-target effects. Another option is the use of inducible gene
expression systems such as the doxycycline-controlled Tet system [ 116, 117 ] . While
shRNAs are generally expressed from polymerase III promoters, miRNAs are
expressed from polymerase II promoters. These polymerase II systems are better
equipped for tissue-specific or drug-regulated gene expression.
Comprehensive reviews on combinatorial RNA approaches are available [ 118,
119 ]. Other types of inhibitory RNA molecules can be added to the RNAi-inducing
antiviral regimes, and we already mentioned the anti-HIV aptamer-siRNA conjugate
transcript [ 97 ]. The currently ongoing phase I clinical trial at the City of Hope
employs a lentiviral vector that encodes a TAR-decoy, CCR5-ribozyme, and shRNA
that targets the HIV-1 genome in the tat-rev region [ 95, 96 ] . The TAR-decoy is a
small nucleolar RNA molecule that absorbs the viral Tat protein, which will prevent
the Tat-TAR interaction that is essential for enhanced viral promoter activity [ 120 ] .
The ribozyme cleaves the CCR5-encoding mRNA to cause reduced expression of
this important HIV-1 receptor on the cell surface [ 121 ] . Alternative antiviral RNA
molecules include antisense transcripts [ 122, 123 ] , decoys [ 124 ] , ribozymes [ 121 ] ,
and aptamers [ 125 ]. A new addition to this arsenal is an antisense molecule that can
elicit transcriptional gene silencing of the viral LTR promoter [ 126 ] . The novel RNAu
method is based on the expression of a modified U1 small nuclear RNA that blocks
polyadenylation of the targeted mRNA, which is subsequently degraded [ 127, 128 ] .