Functional characterization of a portion of the Moloney murine leukemia virus gag gene by genetic footprinting

Abstract

Retroviral Gag proteins perform important functions in viral assembly, but are also involved in other steps in the viral life cycle. Conventional mutational analysis has yielded considerable information about domains essential for these functions, yet many regions of gag remain uncharacterized. We used genetic footprinting, a technique that permits the generation and simultaneous analysis of large numbers of mutations, to perform a near-saturation mutagenesis and functional analysis of 639 nucleotides in the gag region of Moloney murine leukemia virus. We report here the resulting functional map defined by eight footprints representing regions of Moloney murine leukemia virus gag, some previously uncharacterized, that are essential for replication. We found that significant portions of matrix and p12 proteins were tolerant of insertions, in contrast to the N-terminal half of capsid, which was not. We analyzed 30 mutants from our library by using conventional methods to validate the footprints. Six of these mutants were characterized in detail, identifying the precise stage at which their replication is blocked. In addition to providing the most comprehensive functional map of a retroviral gag gene, our study demonstrates the abundance of information that can be gleaned by genetic footprinting of viral sequences.

Fig. 1.

Scheme for generating and analyzing mutations. (A) Library of mutations. Members of the library contained a single insertion at a random position in gag within pNCA, the proviral construct. (B and C) Selection for gag function. 293T cells were transfected with the library. Virus was harvested from these cells and used to infect Rat2 cells at low moi. Virions were harvested and used to infect a second set of Rat2 cells at low moi, thus completing at least one cycle of replication. Low molecular weight (LMW) DNA prepared from infected Rat2 cells constituted selected DNA. (D) Analysis by PCR. Unselected DNA and selected DNA served as templates. Primer P1 corresponded to the insertion; P2 primed from defined sites in gag. PCR products run on a sequencing gel produced a ladder of bands, each band corresponding to at least one independent mutant. Mutants that failed selection gave rise to footprints. Comparing the sizes of missing bands with bands of known size allowed precise mapping of essential regions in gag.

Fig. 2.

Genetic footprinting of MoMLV gag. Footprinting analyses of gag spanning nucleotides 1380–1946. (A) C terminus of MA and part of p12 (nucleotides 1380–1550). (B) Rest of p12 and p12/CA border (nucleotides 1543–1715). (C) N terminus of CA (nucleotides 1705–1946). DNA from the unselected and selected libraries was used as a template for PCR. A primer complementary to the insertion was paired with one of many radiolabeled primers complementary to defined sites in gag. These sites lie within and just outside the region of mutagenesis. Lanes U contain PCR products made from unselected libraries, and lanes S contain products from selected libraries. Products were run adjacent to sequencing reactions (not shown) to allow the precise sizing of bands. A diagrammatic representation of bands is shown alongside the lanes. Red bars, read-through mutations; pink bars, mutants with stop codons; blue bars, viable mutants; and absence of blue bars, footprints. This figure is the result of several PCRs, with many primers, priming in both directions. The gels shown are representative of those used in the complete analysis. D and E show longer exposures of the areas outlined by dashed boxes in B and C. Bands not seen in shorter exposures are marked by asterisks. The nucleotide sequence of the relevant region of gag is depicted next to the bands to indicate the exact location of the corresponding insertions.

Fig. 3.

Functional map of gag. The nucleotide sequence of the mutagenized region of gag is shown, with the amino acid sequence directly below. Colored bars are described in the legend to Fig. 2. Footprints whose extent was greater than three amino acids in length are marked by horizontal lines and assigned a letter code (A–H). Where the boundaries of a footprint are ambiguous because of a paucity of insertions in that region, the horizontal lines are dashed. The location of the 30 mutants analyzed by conventional methods are denoted by the nucleotide number immediately 5′ to the insertion. Mutants subjected to further detailed analysis (see Fig. 5) have a box drawn around their nucleotide number.

Fig. 4.

Validation of footprinting analysis by conventional methods. Mutant proviral DNAs were individually introduced into 293T cells. Medium was harvested 48 h later (posttransfection) and assayed for RT activity (indicating virion assembly and release). This supernatant was used to infect Rat2 cells, and RT activity was assayed 48 h later (postinfection). Values for RT activity are relative to wild type, which was set at 100%. The right-hand column shows whether these mutants were located within (–) or outside (+) footprints.

Fig. 5.

Stepwise conventional analysis of a few mutations. (A) MoMLV life cycle, focusing on specific steps that were analyzed for each mutant. (B–I) Data from each mutant are displayed in vertical rows, with each subsequent step represented directly below the previous one. Wild-type (WT) virus, gag deletion mutant (Δgag), and RT deletion mutant (ΔRT virus) were used as controls. Mock controls underwent identical treatments, except no proviral DNA was used in the transfection. (B) Expression of viral proteins in transfected cells. Cell lysates from transfected cells were analyzed by Western blots using anti-CA antibody. Arrows indicate positions of Gag precursor Pr65^gag and CA. (C) Virion proteins. Virions harvested from transfected cells were analyzed by Western blots as in B.(D) Viral assembly and particle release. Cell supernatants were collected 48 and 72 h posttransfection and analyzed for RT activity. (E) PCR amplification of minus-strand strong-stop DNA [(–)ss DNA]. Primers amplify a 124-bp stretch of the R-U5 region. (F) PCR amplification of gag DNA, using one primer within the insertion and a second in the CA region. For WT virus and for plasmid pNCA, which do not contain insertions, no PCR product was seen. (G) PCR amplification of the LTR junction. Arrows indicate the position of the LTR-junction product. For the pNCA plasmid, primers amplify the β-lactamase gene located in the region between the two LTRs (arrowhead). (H) Amplification of mtDNA. (I) Viral infectivity measured by RT activity. Culture supernatants from infected Rat2 cells were harvested at day 4 and day 6 postinfection and assayed for RT activity.