HIV-1 Nef AND HOST CELL PROTEIN KINASES

Kalle Saksela

Institute of Medical Technology, University of Tampere, P.O. Box 607, FIN-33101, Tampere, Finland

Received 11/10/97 Accepted 11/17/97

2. PHENOTYPIC EFFECTS OF Nef EXPRESSION

The growth of HIV in most transformed cell lines does not depend on a functional Nef gene. A positive effect of Nef on HIV replication kinetics can, however, be demonstrated under certain cell culture conditions, such as infection of resting peripheral blood mononuclear cells (PBMCs) which are subsequently incubated for a couple of days before stimulation of the culture to allow the virus to spread (5, 6). Another seemingly related phenotype reported for Nef in cell culture is its ability to increase the intrinsic infectious potential of HIV particles in single-round infection or end-point dilution assays (5). Interestingly, the lower infectivity of HIV particles which carry a deleted Nef gene can be complemented by expression of Nef from a separate vector during virus production, but not by ectopic Nef expression in the target cells of infection (7-11). This increased infectivity is not related to the success in entering into the target cells, but becomes apparent upon examination of the efficiency in completion of the reverse transcription of the viral genome.

It has been suggested that the mechanism by which Nef increases HIV infectivity involves packaging of Nef protein into the progeny virus particles. Indeed, an average of 10-100 Nef molecules per virion has been shown to be present in HIV particles, where most of it is cleaved by the HIV protease between the membrane-anchoring aminoterminus and the conserved Nef core domain (12, 13). However, the significance of the presence or cleavage of Nef in the virions remains to be demonstrated, and has been called into question by findings on behavior of certain mutant forms of Nef (14; H-G. Kräusslich, personal communication). Alternatively, if Nef protein incorporated into the virions is not important, one would have to postulate that in order to enhance particle infectivity Nef must somehow modify other viral proteins or cell-derived components of the virions. To this end, it has been suggested that expression of Nef increases the incorporation of a cellular serine kinase into the virions (15). On the other hand, decreasing the incorporation of certain host cell surface proteins into the envelopes of budding viruses could conceivable also be involved in increasing HIV particle infectivity.

Downregulation by a posttranslational mechanisms of the cell surface expression of the CD4 receptor for the virus (16-19) as well as the major histocompatibility complex class I (MHC I) molecules (20, 21) are two other known phenotypic effects attributed to Nef in vitro. Downregulation of CD4 requires a dileucine motif in its intracellular tail, and results in increased endocytosis and transport of CD4 into the lysosomal compartment in a manner that is independent of CD4 serine phosphorylation which normally triggers this event (20, 21). Downregulation of MHC I molecules, which do not contain a dileucine motif, also results in their accumulation in the lysosomes (21), but it is unclear how similar these two Nef-induced events are mechanistically. Unlike CD4 and MHC I, a number of other proteins whose cell surface expression is also regulated by endocytosis are not affected by Nef, indicating that this effect is relatively specific (21, 22). It has been speculated that downregulation of CD4 might enhance HIV replication by facilitating the release of virus progeny or by preventing a potentially hazardous superinfection of cells. On the other hand, this could be an unimportant by-product of some other more consequential event, for example liberation into the cytoplasm of Lck protein tyrosine kinase that is normally bound to the intracellular tail of CD4 (22-24). In the case of MHC I, suggested potential pathophysiological roles for its downmodulation by Nef involve decreasing the amount of MHC I molecules that get incorporated to the envelopes of virus progeny and perhaps more intriguingly, escape of HIV-infected cells from recognition and elimination by cytotoxic T cells.

It has been shown that downregulation of both CD4 and MHC I is unrelated to the Nef-induced increase in HIV particle infectivity (5, 20) as particles produced in cells lacking CD4 or MHC I also display this Nef-dependent phenotype. Moreover, we have demonstrated (25), and a number of studies have subsequently confirmed (26-29), that CD4 downregulation by Nef can be genetically separated by site-directed mutagenesis from its ability to enhance HIV replication in PBMCs and to increase HIV particle infectivity, in particular by changes involving amino acids forming a conserved proline-repeat (PxxP) motif of Nef (see later).

The failure to complement the lower infectious potential of Nef-deleted viruses by expression of Nef in the target cell of infection, the demonstrated facilitating effect of Nef on a step in retroviral life-cycle that occurs before de novo synthesis of viral proteins (including Nef), plus similar yields of virus obtained from cells transfected with HIV proviral plasmids regardless of whether they carry a functional Nef gene, have collectively led to an idea that Nef is a protein which is required during the very last steps of HIV life-cycle (virus particle formation) but manifests its function in the target cell during an early post-entry step before proviral integration. However, it would probably be unwise to ignore the fact that the elaborate sequential pattern of gene expression which primate lentiviruses have evolved to utilize (reviewed in ref. 30), appears to point to the opposite, suggesting a principal role for Nef during the early post-integration stage of HIV infection when it constitutes the majority of all viral proteins synthesized.

It is likely, however, that Nef has multiple functions which are important in distinct steps of viral life-cycle and in different aspects of HIV pathophysiology. A list of possible and partially overlapping pathogenic mechanisms, mainly based on extrapolation from the in vitro findings discussed elsewhere in this review, which might be involved in the Nef-induced HIV/SIV disease progression in vivo is presented in table 1.

Although it is probable that the critical in vivo function(s) of Nef is among or related to the listed mechanism, it is at the present difficult to speculate on the relative pathogenic importance of these scenarios. While the effect on HIV particle infectivity - perhaps the currently favored model of Nef action - may indeed be a significant function of Nef, there is no data to indicate that this would be crucial for its role in promoting disease progression in vivo. In fact, some observations hint to the possibility that such Nef-induced increase in HIV particle infectivity may not even account for the enhanced growth kinetics of Nef+ viruses compared to Nef- viruses in culture. For example, while the enhanced replicative kinetics provided by Nef can typically only be observed using the primary cell system described above, the increased HIV infectivity by Nef can be readily demonstrated with a number or transformed producer/target cell combinations. In this regard it is curious why successive rounds of production of apparently 5-to-50-fold more infectious virus do not amount to enhanced replication kinetics of Nef+ viruses in most cells. Also, although available parallel comparison data is scanty, the positive effect of Nef on single-round HIV infectivity appears to be much more dependent on the HIV strain used than the enhancing effect on viral growth kinetics in PBMCs. Furthermore, certain mutations made to Nef seem to affect these two properties in an uneven manner (H.-G. Kräusshlich, personal communication). One interesting possibility is that the positive effect of Nef on viral growth kinetics in PBMCs is relatively indirect, and perhaps related to the recent observation by Desrosiers and colleagues who reported that the substantial growth advantage of Nef+ SIV in H. saimiri-immortalized IL-2-dependent T cell clones correlates with endogenous IL-2 production induced by Nef in these cultures (31).

Such an indirect mechanism for Nef action on viral replication through altered host cell physiology is supported by the demonstrated potential of Nef to regulate cellular signal transduction pathways. The most extreme example of signaling abnormalities induced by Nef is its ability to promote malignant transformation. An unusually potent allele of SIV Nef has been shown to induce morphological transformation of immortalized 3T3 fibroblasts (32). More recently, HIV Nef was shown to cooperate with a cellular proto-oncogene product, protein tyrosine kinase Hck, in transforming another immortalized fibroblast cell line, Rat2 (33). Interestingly, this appears to involve deregulation by Nef of the normally tightly controlled kinase activity of Hck (see later). A large and growing number of effects by Nef on the activity of various components of cellular signaling cascades, such as second messengers and transcription factors, have been reported 28, 32, 34-47; A. Manninen and K.S., unpublished results). It is, however, not easy to build a coherent picture from these observations. One major problem is that often positive, negative, or lack of any effects have been reported regarding the same signaling step. In addition to the molecular clone of Nef used in these studies, such differences might be attributed to a number of variables, including the selection of the cell line and culture conditions, or the means, level and duration of Nef expression, all of which could profoundly modify the cellular responses to Nef due to their general effects on cellular physiology or specific effects on the activity or relative abundance of cellular partners of Nef. In line with such reasoning Baur and colleagues have reported that depending on experimental conditions the subcellular distribution of Nef may be different (plasma membrane vs. cytoplasm), and this can result in contrasting effects on cellular activation markers such as protein tyrosine phosphorylation and the activity of the transcription factors NF-kappaB and AP-1 (36).

Despite these unresolved issues, alteration of cellular signal transduction is a well-documented function of Nef, and an attractive candidate to explain its pathogenic effects in vivo. Modulating cellular activation could not only lead to increased virus replication and infectivity of viral particles produced in the infected cell, but could also promote spreading of HIV to neighboring cells through their paracrine stimulation. By altering the physiology and antiviral responses in the infected cells Nef could conceivably also modulate apoptosis in these as well as in cells involved in antiviral defense, or affect the balance of cytokine networks of the immune system in ways that may promote progression of the infection (table 1).