Distinct Pathway of Human T-Cell Leukemia Virus Type 1 Gag Punctum Biogenesis Provides New Insights into Enveloped Virus Assembly

Abstract

The assembly of virus particles is a crucial aspect of virus spread. For retroviruses, the Gag polyprotein is the key driver for virus particle assembly. In order to produce progeny virus, once Gag is translated, it must translocate from the location in the cytoplasm where it is synthesized to the plasma membrane and form an oligomeric lattice that results in Gag puncta. The biogenesis of mature Gag puncta can trigger the budding process, resulting in virus particle production. While some aspects of the dynamics of Gag oligomerization and particle biogenesis have been observed with human immunodeficiency virus type 1 (HIV-1), the process of Gag punctum biogenesis remains poorly understood, particularly for other retroviruses. Here, we have conducted the most detailed studies thus far on Gag punctum biogenesis for human T-cell leukemia virus type 1 (HTLV-1). Using mEos2 photoconvertible fluorescent proteins and total internal reflection fluorescence microscopy (TIRF), we have found that HTLV-1 Gag was recruited to Gag puncta primarily from the plasma membrane. This was in stark contrast to HIV-1 Gag, which was recruited from the cytoplasm. These observations imply fundamental differences among retroviruses regarding the orchestration of Gag punctum biogenesis, which has important general implications for enveloped virus particle assembly.IMPORTANCE This report describes the results of experiments examining the pathway by which the human retroviral Gag protein is recruited to sites along the inner leaflet of the plasma membrane where Gag punctum biogenesis occurs. In particular, clever and sensitive experimental methods were devised to image in living cells fluorescently labeled Gag protein derivatives from human T-cell leukemia virus type 1 (HTLV-1) and human immunodeficiency virus type 1 (HIV-1) at the plasma membrane. The photoconvertible fluorescent protein mEos2 was strategically utilized, as the fluorescence emission of Gag at the plasma membrane could be differentiated from that of cytosolic Gag. This experimental strategy allowed for the determination of the Gag recruitment pathway into Gag puncta. For HTLV-1 Gag, puncta recruited Gag primarily from the plasma membrane, while HIV-1 Gag was recruited from the cytoplasm. These observations represent the first report of HTLV-1 particle biogenesis and its contrast to that of HIV-1. The observed differences in the Gag recruitment pathways used by HTLV-1 and HIV-1 Gag provide key information that is useful for informing the discovery of novel targets for antiretroviral therapies directed at eliminating virus infectivity and spread.

Keywords: Gag; deltaretroviruses; oligomerization; retroviruses; virus assembly.

FIG 1

HTLV-1 Gag-mEos2 and HIV-1 Gag-mEos2 in living cells. (A) Schematic depiction illustrating the selective photoconversion (PC) of Gag-mEos2 at the plasma membrane (PM). (B) TIRF images taken with 488-nm excitation (green excitation [GR:Exc]) and 561-nm (red excitation [RD:Exc]) of a cell expressing HIV-1 Gag-mEos2 before PC (top) and after PC (bottom). (C) TIRF images of cells expressing HTLV-1 Gag-mEos2 prior to PC (top) display features of the cell near the plasma membrane for 488-nm excitation only. After PC (bottom), bright puncta and diffuse fluorescence along the plasma membrane are visible in images collected with both 488-nm and 561-nm excitation. Because of the large range of intensities in these images, the color scales of the images in panels B and C were computationally saturated. The same scale was maintained before and after PC for each imaging condition to facilitate direct comparison.

FIG 2

Analysis of Gag punctum biogenesis by using mean-squared displacement analysis. (A) Images of a growing punctum taken with a time interval of 2.5 min. The rightmost panel includes the punctum trajectory (magenta line) for the complete image series. (B) The mean-squared displacement (MSD) of the punctum trajectory was calculated as a function of the time difference between images. The MSD can be approximated by a linear fit, indicating that this punctum was undergoing Brownian motion while growing. The slope of the fit determines the diffusion coefficient. (C) Average diffusion coefficients (D) of both stationary and growing puncta produced by HTLV-1 Gag (1:3 labeled/unlabeled) and HIV-1 Gag (1:3 labeled/unlabeled).

FIG 3

HIV-1 Gag puncta analysis following photoconversion. (A) The green color fraction of puncta from a set of HeLa cells transfected with HIV-1 Gag-mEos2 immediately after photoconversion (PC) was plotted in a histogram. (B) Histogram of the green intensity fraction for newly appearing puncta 15 min after PC. (C and D) Green color distribution of puncta from HeLa cells transfected with HIV-1 Gag-mEos2 and unlabeled HIV-1 Gag (1:3 ratio) immediately after PC (C) and of new puncta identified 15 min after PC (D). (E and G) Cropped TIRF images of a location of a future new punctum produced by HIV-1 Gag-mEos2 excited at 488 nm (E) and 561 nm (G) immediately after PC. (F and H) Cropped TIRF images of the same location taken 15 min later at 488 nm (F) and 561 nm (H). The white boxes in panels E to H indicate the locations of a newly appearing punctum recorded during the experiment, which is identified upon 488-nm excitation. GR:Exc, green excitation; RD:Exc, red excitation.

FIG 4

HTLV-1 Gag puncta analysis following photoconversion. (A) The green color fraction of puncta from a set of HeLa cells transfected with HTLV-1 Gag-mEos2 immediately after photoconversion (PC) was plotted in a histogram. (B) Histogram of the green intensity fraction for newly appearing puncta 15 min after photoconversion (PC). (C and D) Green color distribution of puncta from HeLa cells transfected with HTLV-1 Gag-mEos2 and unlabeled HTLV-1 Gag (1:3 ratio) immediately after PC (C) and of new puncta identified 15 min after PC (D). (E and G) Cropped TIRF images of a location of a future punctum newly produced by HTLV-1 Gag-mEos2 excited at 488 nm (E) and 561 nm (G) immediately after PC. (F and H) Cropped TIRF images of the same location taken 15 min later at 488 nm (F) and 561 nm (H). The white boxes indicate the locations of a newly appearing punctum recorded during the experiment, which fluoresces both at 488-nm (F) and 561-nm (H) excitation.

FIG 5

HIV-1 Gag puncta recruit Gag from the cytoplasm. (A and B) Images of a cell transfected with labeled and unlabeled HIV-1 Gag (1:3 ratio) and collected with 488-nm excitation were cropped to display a growing punctum located within the white box as a function of imaging time for 488-nm excitation (A) and 561-nm excitation (B). (C) The integrated intensity of the growing punctum is shown for 488-nm excitation (green symbols) and 561-nm excitation (red symbols) as a function of time. The solid lines describe exponential charging curves with a time constant of ~2 min. (D) Cartoon depicting the recruitment of cytoplasmic HIV-1 Gag-mEos2 into a budding particle.

FIG 6

HTLV-1 Gag puncta recruit Gag from the plasma membrane. (A and B) Images of a cell transfected with labeled and unlabeled HTLV-1 Gag (1:3 ratio) and collected with 488-nm excitation were cropped to display a growing punctum located within the white box as a function of imaging time for 488-nm excitation (A) and 561-nm excitation (B). (C) The integrated intensity of the growing puncta is shown for 488-nm excitation (green symbols) and 561-nm excitation (red symbols) as a function of time. The solid lines describe exponential charging curves with a time constant of ~5 min. (D) Cartoon depicting the recruitment of nonpunctate HTLV-1 Gag-mEos2 from the plasma membrane into a budding viral particle.

FIG 7

Model of HTLV-1 and HIV-1 Gag punctum biogenesis. (A) HTLV-1 Gag was found to be recruited to Gag puncta primarily from the plasma membrane (with a minority population recruited from the cytoplasm). (B) HIV-1 Gag was recruited to growing HIV-1 Gag puncta primarily from the cytoplasm.