Unusual distribution of mutations associated with serial bottleneck passages of human immunodeficiency virus type 1

Abstract

Repeated bottleneck passages result in fitness losses of RNA viruses. In the case of human immunodeficiency virus type 1 (HIV-1), decreases in fitness after a limited number of plaque-to-plaque transfers in MT-4 cells were very drastic. Here we report an analysis of entire genomic nucleotide sequences of four HIV-1 clones derived from the same HIV-1 isolate and their low-fitness progeny following 7 to 15 plaque-to-plaque passages. Clones accumulated 4 to 28 mutations per genome, with dominance of A --> G and G --> A transitions (57% of all mutations) and 49% nonsynonymous replacements. One clone-but not three sibling clones-showed an overabundance of G --> A transitions, evidencing the highly stochastic nature of some types of mutational bias. The distribution of mutations along the genome was very unusual in that mutation frequencies in gag were threefold higher than in env. Particularly striking was the complete absence of replacements in the V3 loop of gp120, confirmed with partial nucleotide sequences of additional HIV-1 clones subjected to repeated bottleneck passages. The analyses revealed several amino acid replacements that have not been previously recorded among natural HIV-1 isolates and illustrate how evolution of an RNA virus genome, with regard to constant and variable regions, can be profoundly modified by alterations in population dynamics.

FIG. 1

Scheme of passages of HIV-1 clones subjected to plaque-to-plaque transfers in MT-4 cells. Clonal populations (HIV-1 isolated from individual plaques) are depicted as filled squares. The experimental procedures and the origins of natural HIV-1 isolate S61 and clones B1 to K1 are given in reference and in Materials and Methods. HIV-1 clones are indicated by letters followed by a number which gives the total number of plaque-to-plaque transfers undergone by the clone. Infectious virus could not be rescued from viral populations B15, E13, G7, H13, and J15 as described in reference .

FIG. 2

Location of mutations found in HIV-1 clones D15, G7, I15, and K15, relative to their parental counterparts. The upper part indicates HIV-1 genes and regulatory regions based on the compilation of Korber et al. (35). The four horizontal bars in the center of the figure indicate the positions of mutations along the genome (9.1 kb) in the four clones analyzed (from top to bottom, D15, G7, I15, and K15 [described in Materials and Methods]); vertical lines within these bars indicate one, two, or three mutations, according to thickness. Mutations were found at positions 35, 171, 377, 379, 570, 584, 760, 807, 833, 988, 1128, 1161, 1188, 1351, 1467, 1545, 1578, 1596, 1810, 1863, 1875, 1937, 1961, 1966, 2145, 2174, 2329, 2668, 2804, 3068, 3114, 3129, 3945, 4458, 5239, 5270, 5342, 5422, 5684, 5686, 6588, 6655, 6670, 7962, 8095, 8890, 8900, and 8989 according to the numbering of HIV-1 isolate HXB2 (35). The gp120-coding region of env has been enlarged to depict the positions of variable loops V1 to V5; two mutations affected the V1-coding region. The two shaded rectangles correspond to genome positions 1337 to 1598 (gag [left shaded rectangle]) and 7071 to 7333 (env [right shaded rectangle]), which have been sequenced for a number of additional HIV-1 clones to confirm the asymmetric distribution of mutations (see text). The mutations found in these additional clones are not included in this scheme. The bottom part shows the three arbitrary regions into which the HIV-1 genome was divided to illustrate the bias in the distribution of mutations along the genome. Procedures used for nucleotide sequence determination are described in Materials and Methods, and the oligonucleotide primers are listed in Table 1.