Date of Award


Document Type


Degree Name

Master of Science (MS)


Environmental Science

First Advisor

Dr. Tahir A. Rizvi

Second Advisor

Prof. And rew Michael Lin dsay Leve r

Third Advisor

D r. Ahmed H. M. Hassa n Al - Marzouqi


The mechanism by which retroviruses preferentially encapsidate their unspliced genomic RNA among millions of both spliced viral and cellular mRNAs in the cytoplasm represents a function of great specificity. This selection process requires that the genomic RNA contain packaging determinants unique to its own RNA that can interact specifically with the packaging proteins of the viral particles, the Gag polyproteins. Knowing the exact sequences involved in packaging should provide basic insights into the mechanism of preferentially encapsidating the full length genomic RNA.

We have been interested in mapping the packaging determinants of the feline immunodeficiency virus (FIV), a lentivirus that is being considered as a potentially powerful gene delivery system for human gene therapy. Our initial studies have shown that the FIV packaging determinants are located as two discontinuous core regions within the 5' end of the viral genome (Browning et aI., 2003 a & b). The first region extends from the R/U5 in the 5' LTR to the first 120 bp of 5' UTR and the second consists of the first 100 bp of gag, while other regions of the genome may also be involved.

Studies undertaken in this thesis carried these observations further to determine whether the region in between the two core determinants was important for packaging or merely acted to maintain the spacing of the two core elements. Additionally, since other regions of the genome, especially the LTR, had been implicated as containing significant packaging determinants in other studies (Kemler et aI., 2002), we dissected the role of the LTR elements away from the untranslated region towards FIV RNA packaging. Towards this end, several series of small FIV transfer vectors were constructed either in the heterologous non-viral or homologous subgenomic context containing various combinations of the LTR, and/or UTR and gag and tested for their packaging potential in our well-established in vivo packaging assay. This was followed by analysis of the amount of transfer vector RNAs packaged directly into the virus particles using a semi-quantitative RT-PCR approach.

Test of the various transfer vector RNAs confirmed our earlier observation that the FIV packaging determinants are indeed discontinuous and spread out with the core packaging determinants residing within the first 150 bp UTR and 100 bp of gag (Chapter III). Furthermore, the intervening sequences between these two elements were not required either for vector RNA packaging or propagation (Chapter III). Analysis of the LTR elements revealed the presence of other packaging determinants of lesser strength than the core determinants in the 5' R/U5 and 3' U3/R regions of the viral LTRs (Chapter IV). Folding of the 5' end of the viral RNA using computer analysis software revealed the presence of complex stem loop structures. Correlation of the mutational analysis with the folding algorithms revealed the presence of a conserved stem loop in the 5' UTR that may serve as the principal packaging determinant of FIV. Interestingly, no consistent structural element could be identified within the first 100 bp of gag that could be responsible for the packaging potential of the gag region, suggesting that gag sequences may function at the primary sequence level, perhaps providing the intronic sequences needed to distinguish between genomic and sub genomic mRNAs. Taken together, these data should add to the increasing knowledge of how complex retroviruses package their genomic RNAs and help streamline the design of safer self-inactivating FIV–based vectors for human gene therapy.