T-cell dysfunctions in hu-PBL-SCID mice infected with human immunodeficiency virus (HIV) shortly after reconstitution: in vivo effects of HIV on highly activated human immune cells

Abstract

The state of activation of the immune system may be an important factor which renders a host more receptive to human immunodeficiency virus (HIV) and more vulnerable to its effects. To explore this issue with a practical in vivo model, we developed a modified protocol of HIV infection in hu-PBL-SCID mice. First, we assessed the time course of activation of human peripheral blood lymphocytes (hu-PBL) in the peritoneal cavity of SCID mice. At 2 to 24 h after the intraperitoneal injection into SCID mice, there was a clear-cut increase in the percentage of hu-PBL expressing early activation markers (CD69), concomitant with the release of soluble intercellular adhesion molecule-1 (sICAM-1) and the soluble interleukin-2 receptor (sIL-2R) and with the accumulation of mRNAs for a number of human cytokines. At 2 weeks, virtually all of the hu-PBL expressed the memory phenotype (CD45RO) and HLA-DR antigens as well. Cells collected from the SCID mouse peritoneum at 2 and 24 h after transplantation were fully susceptible to in vitro infection with HIV type 1 (HIV-1) in the absence of either IL-2 or mitogens. The injection of HIV into hu-PBL-SCID mice at 2 h after reconstitution resulted in a generalized and productive HIV infection of the xenochimeras. This early HIV-1 infection resulted in a dramatic depletion of human CD4+ cells and in decreased levels of sICAM-1 (in the peritoneal lavage fluid) as well as of sIL-2R and immunoglobulins M and A (in the serum). Enzyme-linked immunosorbent assay and/or reverse transcriptase PCR analysis showed higher levels of IL-4, IL-5, and IL-10 in the HIV-infected animals than in control hu-PBL-SCID mice, while gamma interferon levels in the two groups were comparable. When we compared the current model of HIV-1 infection at 2 weeks after the intraperitoneal injection of the hu-PBL in the SCID mice with the model described here, we found that the majority of immune dysfunctions induced in the 2-h infection of the xenochimeras are not inducible in the 2-week infection. This supports the concept that the state of activation of human cells at the moment of the in vivo infection with HIV-1 is a crucial factor in determining the immune derangement observed in AIDS patients. These results show that some immunological dysfunctions induced by HIV infection in AIDS patients can be mimicked in this xenochimeric model. Thus, the hu-PBL-SCID mouse model may be useful in exploring, in vivo, the relevance of hu-PBL activation and differentiation in HIV-1 infection and for testing therapeutic intervention directed towards either the virus or the immune system.