With the aim of developing new strategies in antiviral therapy and vaccination our work has been based upon interdisciplinary cooperation and application of a broad spectrum of methods within the fields of retrovirology, molecular and cell biology, biochemistry, protein structure determination, as well as immunology. The overall research goal is to contribute to the development of novel anti-retroviral strategies by deciphering the structural constraints and molecular mechanisms, governing the function of viral proteins and their cellular counterparts.
We are also investigating the molecular structure and function of small virus proteins like Vpr, Vpu and p6 from HIV-1, and PB1-F2 from IAV that govern specific regulatory functions in virus replication.
We are currently working on the following main research topics:
One of our research objectives is to decipher the role of the ubiquitin proteasome system (UPS) in MHC class I (MHC-I) antigen presentation. CD8+ cytotoxic T lymphocytes (CTLs) play a critical role in the immune response to viral infections. Understanding the mechanisms involved in the presentation of viral determinants may be important for optimizing the induction of anti-viral CTL responses, for example during vaccination. In the course of studies initiated to examine how MHC-I peptides are generated from metabolically stable viral proteins, like HIV-1 Gag, it was demonstrated that so called defective ribosomal products (DRiPs) constitute a large fraction of newly synthesized proteins. DRiPs, which were until then only predicted theoretically, are polyubiquitinated short-lived proteins that never attain their native structure due to errors in transcription, translation or immediate post-translational processes necessary for proper protein folding. The DRiP-pathway provides a reasonable explanation for the paradox that MHC-I determinants of stable proteins, e.g. viral structural proteins, are presented in association with MHC-I molecules on the cell surface for detection by CTLs immediately after infection of target cells. Therefore, the discovery of the DRiP-pathway opened a new field in the research of antiviral and anti-tumor T-cell responses and thus might be important for the optimization of current vaccination strategies.
If the stability of a given antigen inversely correlates with the efficiency of its entry into the MHC-I pathway, vaccine development should be focused on the expression of mainly short-lived antigens with high access rates to the UPS, rather than the over-expression of a natural and metabolically stable antigen. We challenged this working hypothesis using HIV-1 Gag as a viral model antigen that is also a major target for vaccine development.
We sought to identify a naturally occurring sequence motif within Gag that regulates its entry into the DRiP-pathway. As the PTAP late assembly (L)-domain motif in the C-terminal p6 domain of Gag has been shown to negatively regulate the ubiquitination of Gag, we analyzed the correlation between ubiquitination and MHC-I presentation of PTAP-deficient Gag. Intriguingly, mutation of PTAP not only reduces the release of virus-like particles, but also increases polyubiquitination of Gag and, consistently, enhances MHC-I presentation of the Gag-derived SL epitope. Although the half-life of the PTAP mutant was only mildly reduced, the entry into the DRiP-pathway was significantly increased. Altogether, these results indicate that besides driving virus release the PTAP motif regulates the entry of Gag into the DRiP- and thus into the MHC-I pathway. Though there are no naturally occurring PTAP-mutants of HIV-1 identified so far, mutations of PTAP might enhance the immunogenicity of Gag and should thus be considered for the improvement of vaccine development.
In cooperation with the group of Prof. Dr. Schuler (Department of Dermatology, University Medical Center Erlangen), we investigated if the immunogenicity of the tumor-associated antigen MelanA can be enhanced by the introduction of a degradation signal. MelanA is used as a vaccine antigen for immunotherapy of malignant melanoma using autologous, antigen-charged dendritic cells.
N-terminal stable in frame fusion of ubiquitin (Ub) has been shown to target the fusion protein for proteasomal degradation. The pathway known as the Ub fusion degradation (UFD) was mainly studied on cytosolic proteins so far. It has been described that the degradation of such a fusion protein is mediated by polyubiquitination of specific lysine residues within the fused Ub moiety. We could show that fusion of the non-cleavable UbG76V variant to the N-terminus of MelanA results in rapid proteasomal degradation via the endoplasmic reticulum-associated degradation (ERAD) pathway and, consequently, leads to an increased MHC-I antigen presentation of this transmembrane protein. While lysine residues within Ub are dispensable for these effects, the presence of one single lysine residue, irrespectively of its location along the fusion protein, is sufficient to induce degradation of MelanA. These findings are in contrast to the conventional wisdom about the UFD mechanism and indicate a new concept to target a protein into the UPS for enhanced MHC-I antigen presentation.
A major project focuses on the role of the HIV-1 p6 Gag-protein as part of the Gag precursor protein in the late processes of virus replication, especially assembly and budding.
Viruses like HIV-1 hijack a cellular system known as the endosomal sorting complexes required for transport (ESCRT) in order to promote their release from the cell surface. This cellular machinery that is involved in cytokinesis and in the vacuolar protein sorting of endocytosed plasma membrane proteins regulates the abscission of membrane stalks.
The HIV-1 p6 Gag protein regulates the final abscission step of nascent virions from the cell membrane by the action of its two L-domains. Although p6 is located within one of the most polymorphic regions of the HIV-1 gag gene, the 52 amino acid peptide binds to at least two cellular budding factors of the ESCRT-complex, Tsg101 (tumor susceptibility gene product 101) and ALIX (ALG-2 interacting protein X), via its highly conserved primary PTAP and secondary YPXnL L-domain, respectively. p6 is substrate for phosphorylation, ubiquitination and sumoylation, and mediates the incorporation of the HIV-1 accessory protein Vpr into viral particles. As expected, known functional domains mostly overlap with several conserved residues in p6. We investigated the importance of the highly conserved serine residue in position 40 (S40) which is phosphorylated by the atypical phosphokinase C (aPKC) and until now not assigned to any known function of p6.
Consistent with previous data we found that mutation of S40 has no effect on the ALIX mediated rescue of HIV-1 L-domain mutants. However, the only mutation that preserves the overlapping pol open reading frame, S40F, reduces virus replication in T-cell lines and in human lymphocyte tissue cultivated ex vivo and significantly reduces the specific infectivity of released virions. Furthermore, it was observed that mutation of S40 selectively interferes with the cleavage between capsid (CA) and the spacer peptide SP1 in Gag. This deficiency in CA-SP1 processing led to an irregular morphology of the virus core and the formation of an electron dense extra core structure. This observation could be confirmed by the group of our cooperation partner Carol Carter (Stony Brook University, Stony Brook (NY), USA), who found that S40F, a mutation frequently occurring in patients with treatment failure, causes formation of filopodia-like structures enabling cell to cell transmission of the virus.
By analyzing synthetic p6-peptides by surface plasmon resonance (SPR) spectroscopy we could show that p6 directly interacts with a cytoplasmic model membrane through both N-terminal and C-terminal regions. Most interestingly, the S40F exchange leads to an increased membrane binding of p6 most likely by hydrophobic interactions.
By performing membrane flotation assays we found that in HIV-1 expressing cells the mutant S40F, but not the conservative mutation to Asp (S40D) or Asn (S40N), leads to an increased membrane association of Gag. This correlates with an enhanced ubiquitination of Gag which is predominantly constituted of K48-linked polyubiquitin chains. Consequently, the S40F mutant augments the entry of Gag into the UPS and thus enhances MHC-I antigen presentation and Gag-mediated T-cell activation. Others could show that the membrane resident ubiquitin E3 ligase Cul7 is able to ubiquitinate Gag. This finding led us to the hypothesis that the S40F mutant might also be ubiquitinated by this or other E3 ligases due to prolonged membrane association.
As an explanation for the enhanced membrane association of Gag, we found by performing high resolution NMR analysis that only S40F, together with Tyr-36, induces the formation of a hydrophobic patch within the C-terminal α-helix of p6 which might account for the augmentation in membrane association of Gag.
We further analyzed the role of p6 in membrane association and found that the mutation of the highly conserved glutamic acids within p6 to alanine (E0A), like the S40F mutant, leads to an enhanced polyubiquitination and subsequent entry of Gag into the UPS and thus an increased MHC-I antigen presentation of Gag derived epitopes. In addition, like for the S40F mutant, the CA-SP1 processing is impaired, also resulting in loss of infectivity. Furthermore, the replication capacity of the E0A mutant was significantly decreased. The mutant also displays defective virus release that could not be rescued by ALIX overexpression. Most strikingly, the E0A mutant exhibits elevated membrane association which might be due to removal of the negatively charged Glu residues. In wt situation these negative charges might contribute to the repulsion of p6 from the plasma membrane. To verify this hypothesis, we created mutants where all Glutamic acids were exchanged with aspartic acids (E0D) that altered the side chain of the amino acid but maintained the overall charge of p6. Intriguingly, this mutant behaves like the wt in terms of ubiquitination, MHC-I antigen presentation and membrane association of Gag, leading to the assumption that the negative charges but not the nature of the side chains of the glutamic acids are important for the functionality of p6 during late steps replication.
The cumulative data support a model in which p6, in addition to matrix, acts as a membrane targeting domain of Gag, either by hydrophobic (S40F) or electrostatic (E0A) interactions with the inner leaflet of the plasma membrane.
Deciphering the molecular mechanism of small regulatory virus proteins provides knowledge about specific host-virus interactions and should open the path to the development of innovative antiviral strategies. The major focus of this long term project has been the biochemical, structural and functional characterization of the HIV-1 accessory protein Vpr and its cellular binding factors. More recently we have initiated functional analysis of another HIV-1 accessory protein, Vpu. Further-more, we have been studying structure and function of the L-domain active HIV-1 p6 Gag protein.
In the last years, we intensively investigated the HIV-1 accessory protein Vpr (viral protein R) of HIV-1. The 96 amino acid Vpr has multiple functions in HIV-1 pathogenesis, including virion incorporation, nuclear translocation of the HIV-1 preintegration complex, induction of cell cycle arrest at the G2/M phase, and the regulation of apoptosis.
Vpr, which predominantly localizes in the nucleus, in mitochondria and in the cytoplasm, is known to interact with members of the nuclear pore complex, especially nucleoporin, and alters its permeability during induction of apoptosis. It was shown previously by others that the accumulation of Vpr at the nuclear lamina is dependent on the integrity of its N-terminal helix-α1, which is also important for its capacity to induce G2 arrest. Furthermore, caspases, especially caspase 3, are known to regulate the diffusion limit of the nuclear pore complex. As Vpr is a nucleo-cytoplasmic shuttling protein, the importance of the nuclear transport of Vpr and the permeability of the nuclear pore for the function of Vpr was analyzed. Treatment of HeLa cells with a caspase 3 inhibitor, the tetrapeptide Z-DEVD, caused accumulation of Vpr at the nuclear lamina. As caspases regulate the pore morphology during induction of apoptosis, altered permeability may result from a change in pore complex assembly caused by the caspase 3 inhibition. Consistently, HIV-1 infected Jurkat T-cells showed an increased induction of G2 arrest when treated with Z-DEVD. Mutation of Pro-35 in the N-terminus of Vpr, which is important for the formation of helix-α1, abrogated Z-DEVD mediated enhancement of G2 arrest. Moreover, the induction of Vpr mediated apoptosis is reduced by Z-DEVD, as it is also compromised by mutating Pro-35. Altogether, this indicates that Vpr, at least partially, induces apoptosis via this specific caspase pathway and that the nucleo-cytoplasmic shuttling of Vpr is dependent on the permeability of the nuclear pore.
HIV patients manifest adipose dysfunction characterized by accelerated lipolysis, hepatosteatosis, dyslipidemia, insulin resistance, and hyperglycemia. However, the in vivo mechanisms whereby HIV infection induces those defects in human adipose disorders have not been reported. Thus, in cooperation with the lab of Dr. Ashok Balasubramanyam (Baylor College of Medicine, Houston, TX, USA) the pathogenic role of Vpr in HIV-associated adipose dysfunction was investigated. It could be shown that Vpr released from HIV-1 in tissue reservoirs, can disrupt PPAR/GR (peroxisome proliferator-activated receptor/glucocorticoid receptor) co-regulation and cell cycle control to produce adipose dysfunction and hepatosteatosis. Confirmation of this mechanism in patients could pave the way for targeted treatment with small-molecule inhibitors of Vpr, GR antagonists, or PPAR agonists.
The HIV-1 accessory protein Vpu is an 81-amino-acid oligomeric type 1 integral membrane phosphoprotein, which is encoded exclusively in HIV-1 and related simian immunodeficiency viruses (SIV), but not in HIV-2. Vpu has been shown to induce degradation of the CD4 receptor by the ER-associated protein degradation (ERAD) pathway and to enhance virus particle release from the plasma membrane. Randomization of the transmembrane (TM) domain prevents Vpu`s ion channel formation and impairs its ability to regulate virus release. This suggested a causal relation between the ion channel activity of Vpu and its augmentation of virus release. Recently, it was shown that Vpu enhances HIV-1 virion release by counteracting a cellular restriction factor termed CD317 (also known as tetherin, BST-2 or HM1.24). Thus, in order to analyze whether ion channel activity of Vpu correlates with viral particle release, several TM mutants were generated. The highly conserved residues Ala-14 and Ala-18 in the TM domain of Vpu were analyzed for their ability to form ion channels and to counteract CD317.
Mutation of Ala-14 and Ala-18 to asparagine impairs the efficiency of surface down-regulation of CD317. However, both mutants still exhibit ion channel activity in 293T cells, indicating that channel activity of Vpu is not sufficient to support virus release. Furthermore, to assess its relevance in CD317 counteraction, we mutated Ser-23, which is essential for Vpu's ion channel activity, to alanine (S23A). Intriguingly, the S23A mutant still efficiently interacts with CD317, and thus supports virus release in the presence of CD317. Taken together, our data suggest that the ion channel activity of Vpu is not critical for counteraction of CD317.
Moreover, it was previously shown that Vpu also induces downregulation of the coactivating NK cell receptor, the NK, T-cell, B-cell antigen (NTB-A), from the cell surface in order to evade lysis of HIV-1 infected cells by NK cells. Pulse-chase analyses revealed that HIV-1 Vpu affects the glycosylation pattern of NTB-A by a mechanism that is distinct from the Vpu induced downregulation of CD4 and tetherin. In the presence of Vpu, only the high mannose form of NTB-A was detectable, suggesting that Vpu prevented the formation of the mature form of NTB-A. This phenomenon is associated with the ability of Vpu to downregulate cell surface NTB-A by retention of NTB-A within the Golgi-compartment. Furthermore, the Vpu-mediated effect on NTB-A glycosylation is highly conserved among Vpu proteins derived from HIV-1 and SIV and corresponds to the level of downregulation of NTB-A. Together, our results suggest that the reduction of NTB-A from the cell surface is associated with the Vpu-mediated effect on the glycosylation pattern of newly synthesized NTB-A molecules.
Recently, it was shown that Vpu induces downregulation of cell surface CD155, a ligand for the DNAM-1 activating receptor of NK and CD8+ T cells, to evade NK cell-mediated immune response. In order to analyze the effect of Vpu on the surface expression of CD155, Vpu TM and cytoplasmic deletion mutants were analyzed in HeLa cells, which constitutively express endogenous CD155. The results suggested that the TM domain of Vpu, particularly Ala-10, Ala-14 and Ala-18, is crucial for cell surface downregulation of CD155. Moreover, Vpu induces accumulation of CD155 in perinuclear compartments indicating that Vpu might inhibit trafficking of CD155 to the cell surface. Thus, Vpu seems to subvert NK cell responses against HIV-1 infected T cells by modulation of multiple receptors necessary for NK cell activation.