Mechanism of reduction in titers from lentivirus vectors carrying large inserts in the 3'LTR

Abstract

Self-inactivating (SIN) lentiviruses flanked by the 1.2-kb chicken hypersensitive site-4 (cHS4) insulator element provide consistent, improved expression of transgenes, but have significantly lower titers. The mechanism by which this occurs is unknown. Lengthening the lentiviral (LV) vector transgene cassette by an additional 1.2 kb by an internal cassette caused no further reduction in titers. However, when cHS4 sequences or inert DNA spacers of increasing size were placed in the 3'-long terminal repeat (LTR), infectious titers decreased proportional to the length of the insert. The stage of vector life cycle affected by vectors carrying the large cHS4 3'LTR insert was compared to a control vector: there was no increase in read-through transcription with insertion of the 1.2-kb cHS4 in the 3'LTR. Equal amount of full-length viral mRNA was produced in packaging cells and viral assembly/packaging was unaffected, resulting in comparable amounts of intact vector particles produced by either vectors. However, LV vectors carrying cHS4 in the 3'LTR were inefficiently processed following target-cell entry, with reduced reverse transcription and integration efficiency, and hence lower transduction titers. Therefore, vectors with large insertions in the 3'LTR are transcribed and packaged efficiently, but the LTR insert hinders viral-RNA (vRNA) processing and transduction of target cells. These studies have important implications in design of integrating vectors.

Figure 1

Viral titers of LV vectors with inserts into the 3′LTR were inversely proportional to the length of the LTR insert. (a) Schematic representation of the LV vectors. All vectors were based on sBG, a SIN LV vector carrying the β-globin gene, β-globin promoter and the locus control region elements HS2, HS3, and HS4. Different fragments of the cHS4 site were inserted in the U3 region of the sBG 3′LTR (shown above the sBG vector). Similar-sized inserts were made by replacing the region downstream of cHS4 core with inert DNA spacers from the lambda phage DNA (shown below the sBG vector). (b) Viral titers of insulated vectors decreased as the length of the insulator insert increased. Titers reflect concentrated vector made concurrently for all vectors in each experiment (n = 4). All titers were significantly lower than the titers of the control vector sBG (P < 0.01; one-way analysis of variance). (c) A 650-bp sequence of cHS4, optimized for insulator activity through a structure–function analysis. A vector containing this 650-bp fragment (sBG⁶⁵⁰) was found to have approximately twofold lower titers than the uninsulated vector sBG (n = 3). (d) Titers fell with insertion of increasing length of an inert DNA spacer downstream of the core. Titers of insulated LV vectors (hatched bars) are similar to those containing inert DNA spacers in the LTR (open bar) in four independent experiments. The titers of sBG with a 400-bp spacer were slightly higher (*P < 0.05). (e) The sBG^2C vector, carrying tandem repeats of the cHS4 core recombined with high frequency. A schematic representation of the vectors sBG-I and sBG^2C proviruses, when intact, or when the core elements recombine with loss of one or two cores with the region probed and restriction site of the enzyme used (AflII) is shown. The size of the expected band is shown adjacent to each vector cartoon. The right panel is the Southern blot analysis showing a single 8 kb expected band for sBG-I transduced MEL cell population, and two bands in the sBG^2C-transduced MEL cell population, representing sBG^2C with either loss of one or both cores. cHS4, chicken hypersensitive site-4; CMV, cytomegalovirus; LTR, long terminal repeat; LV, lentiviral; MEL, mouse erythroleukemia; SIN, self-inactivating.

Figure 2

Similar amounts of viral RNA were produced from the insulated and uninsulated vectors in packaging cells. Northern blot analysis on the 293T packaging cells after transfection with sBG and sBG-I vectors showed the expected length viral RNA. The membrane was hybridized with a ³²P-labeled β-globin probe (top panel) and 18S (bottom panel) as a loading control. An expected 7.3 and 8.5-kb band corresponds to sBG and sBG-I viral RNA were detected. The 18S and 28S rRNA were nonspecifically probed with this probe. No extraneous recombined bands were detected with either vector. The phosphoimager quantified ratios of viral RNA and 18S rRNA of both vectors are listed below the lanes and show no difference in the amount of vRNA between the two vectors.

Figure 3

Vector production was not impaired by insertion of cHS4 in the 3′LTR. (a) Reverse transcriptase activity in sBG and sBG-I viral supernatants is similar (23 ± 5 versus 27 ± 3; n = 3, P > 0.5). (b) p24 levels detected in the concentrated viral preparation is the same with sBG and sBG-I. (2.9 ± 0.5 × 10⁵ versus 1.7 ± 0.5 × 10⁵; n = 3, P > 0.1). (c) Dot-blot analysis of viral RNA extracted from sBG and sBG-I viral supernatant shows similar amounts of viral RNA packaged into virions in both vectors. Note that four different dilutions of viral RNA were loaded in duplicate for the two vectors. The membrane was hybridized with a ³²P-labeled β-globin probe. Only one of two representative experiments is shown. (d) Phosphoimager quantification of two independent experiments was plotted and showed similar amounts of viral RNA in sBG and sBG-I virions (1.9 ± 0.7 × 10⁶ versus 1.9 ± 0.6 × 10⁶; n = 2, P > 0.5). cHS4, chicken hypersensitive site-4; cpm, counts per minute; LTR, long terminal repeat; RT conc., reverse transcriptase concentration; vRNA, viral RNA.

Figure 4

Kinetic of reverse transcription and nuclear translocation in lentivirus vector carrying insulator element in the LTR. (a) Schema of the LV reverse transcription and nuclear translocation process are illustrated. On the right a summary of qPCR assays performed to analyze several steps of the process. Thin line: RNA; thick line: DNA. Open boxes: polypurine tract (PPT). Open circle: primer-binding site (pBS). The 3′ LTR DNA insert is illustrated in the first strand transfer diagram. The positions of the qPCR assays are shown. DNA from MEL cells after infection with sBG and sBG-I viral vector was collected at different time points after infection and analyzed by qPCR. Solid line: sBG. Dashed line: sBG-I. (b) Kinetic of reverse transcription before the first strand transfer (R/U5) shows no difference between the two vectors. (c,d) After the first strand transfer (U3/R and ψ) there is a decrease in reverse transcription efficiency in presence of the insulator (n = 3). cDNA, complementary DNA; LTR, long terminal repeat; qPCR, quantitative PCR.

Figure 5

Insertion of cHS4 in the LTR affected viral integration. Linear viral cDNA circularizes and is the form that integrates; 1-LTR and 2-LTR circles represent abortive integration products from homologous recombination and nonhomologous end-joining, respectively. The 1-LTR and 2-LTR circles are therefore used as markers of nuclear translocation. (a) There are reduced 2-LTR circles, analyzed by qPCR on DNA extracted from MEL cells infected at different times after infection with viral vector suggesting reduced nuclear translocation or nonhomolgous end-joining. (b) Southern blot analysis of MEL cells 72 hours after infection with same amount of sBG and sBG-I vector. StuI digestion of genomic DNA allows identification of 1-LTR circles, 2-LTR circles, linear DNA, and integrated DNA (a smear) for sBG and sBG-I. Expected band sizes are shown for both vectors. While linear, 1-LTR and 2-LTR circles are seen in the sBG lane, no linear DNA or 2-LTR circles are detected in the sBG-I lane. However, 1-LTR circles are almost as prominent as in the sBG lane. The relative ratios of linear, 1-LTR and 2-LTR circles suggest increased recombined abortive integration products with the sBG-I vector, and hence result in inefficient integration. (c) sBG- and sBG-I-transduced MEL cells show intact proviral integrants (7.5 and 8.0 kb, respectively). There was an eightfold difference in phosphoimager counts of the two bands. Vector copy number per cell was also quantified by qPCR and is depicted below the lanes. cDNA, complementary DNA; cHS4, chicken hypersensitive site-4; LTR, long terminal repeat; MEL, mouse erythroleukemia; qPCR, quantitative PCR.

Figure 6

Hypothesis of mechanism by which insulator sequence decrease viral titer. In wild-type HIV, linear cDNA molecules translocate to the nucleus where a small percentage undergoes recombination and end-joining ligation to form 1-LTR and 2-LTR circles, respectively. Only the linear form is the immediate precursor to the integrated provirus. In the case of insulated LV vectors, we show an increase in 1-LTR circle formation, probably due to the presence of a larger U3 sequence that could facilitate an increase in homologous recombination. This process depletes the amount of viral DNA available for integration as well as the amount of 2-LTR circle formation. The decreased amount of DNA available for integration could explain the loss in titers for LV vector carrying large inserts in the LTR. cDNA, complementary DNA; LTR, long terminal repeat; LV, lentiviral.