356
Rare Gag Variants Selected in B*57/5801 HIV-1 Elite Controllers Induce de novo CTL Responses and Reduce Viral Replication Capacity
Toshiyuki Miura*1, M Brockman1, A Schneidewind1, A Raothod1, B Block1, F Pereyra1, T Allen1, M Lobritz2, E Arts2, and B Walker1
1Massachusetts Gen Hosp, Boston, US and 2Case Western Reserve Univ, Cleveland, OH, US
Background: HIV-1-specific cytotoxic T lymphocytes (CTL)
play an important role in control of infection. HLA class I alleles B*57/5801
are associated with slower disease progression and lower plasma viral loads,
and over 40% of HIV-1 elite controllers with viral loads <50 copies/mL
express B*57. But, differences between elite controllers and viremic B57+
subjects have not been described. The B57-restricted Gag epitope TW10 is
targeted during acute infection, and the ability of HIV to escape CTL pressure
in this region may be a key determinant of pathogenesis.
Methods: Full-length Gag sequences were obtained
from plasma of B*57/5801+ subjects. Interferon (IFN) -γ responses
to TW10 wild type and variants peptides were measured by enzyme-linked
immunosorbent spot assay (ELISpot). NL4-3 variants were generated by
mutagenesis, and viral replication capacity was measured using a GFP reporter T
cell line. Viral competition assays were performed by co-culture with a
vif-synonymous mutant NL4-3 strain and measured by TaqMan polymerase chain
reaction (PCR).
Results: Gag sequences were obtained from 22 elite
controllers and 32 viremic subjects expressing B*57/5801. Of 32 (87.5%) viremic
subjects, 28 encoded the typical TW10 CTL escape mutation T242N. In contrast,
15 of 22 (68.2%) elite controllers displayed T242N. Unusual mutations in TW10
were observed in the other elite controllers. G248D was seen in 2 elite
controllers and 2 viremic subjects, and is present at 0.8% in the Los Alamos
National Laboratory (LANL) database. N252H, which is adjacent to TW10, was seen
in 3 of these subjects, and N252R was observed in 1 elite controller. Also, viral
load in the 2 viremic subjects was extremely low (91.5 and 164.5 copies/mL). The
combination of D248 plus N252 reduced viral replication capacity to 23.5% of
wild type NL4-3, while H252 and R252 partially compensated D248 to 45.0% and 72.9%
of wild type, respectively. Notably, we saw a stronger ELISpot response to the
G248D variant peptide than to wild type in these individuals. Furthermore, we observed
that G248D could not co-exist with T242N. In addition, we found I247M, G248T/A,
and M250I in 3 elite controllers, none of which are seen in the LANL database,
and we confirmed that the combination of I247M, G248A, and M250I also reduced
viral RC significantly.
Conclusions: Escape mutations are frequently observed
in the immunodominant B*57/B5801-restricted CTL epitope TW10 in plasma of elite
controllers. Rare G248D mutations were seen in subjects with extremely low viral
loads. This substitution was recognized by de novo CTL but also diminished
viral replication capacity. Generation of unusual substitutions in the region
of TW10 may be an important factor in controlling viremia in B*57/5801 elite
controllers.
|