Discovery of new antifungal chemo-therapeutics for humans is limited by the large degree of conservation among eukaryotic organisms. In recent years, the histone acetyl-transferase Rtt109 was identified as the sole enzyme responsible for an abundant and important histone modification, histone H3 lysine 56 (H3K56) acetylation. In the absence of Rtt109, the lack of acetylated H3K56 renders yeast cells extremely sensitive to genotoxic agents. Consequently, the ability to sustain genotoxic stress from the host immune system is crucial for pathogens to perpetuate an infection. Because Rtt109 is conserved only within the fungal kingdom, I reasoned that Rtt109 could be a novel drug target. My dissertation first establishes that genome stability provided by Rtt109 and H3K56 acetylation is required for Candida albicans pathogenesis. I demonstrate that mice infected with rtt109 -/- cells experience a significant reduction in organ pathology and mortality rate. I hypothesized that the avirulent phenotype of rtt109 -/- cells is due to their intrinsic hypersensitivity to the genotoxic effects of reactive oxygen species (ROS), which are utilized by phagocytic cells of the immune system to kill pathogens. Indeed, C. albicans rtt109 -/- cells are more efficiently killed by macrophages in vitro than are wild-type cells. However, inhibition of ROS generation in macrophages renders rtt109 -/- and wild-type yeast cells equally resilient to killing. These findings support the concept that ability to resist genotoxic stress conferred by Rtt109 and H3K56 acetylation is a virulence factor for fungal pathogens and establish Rtt109 as an opportune drug- target for novel antifungal therapeutics. Second, I report the discovery of a specific chemical inhibitor of Rtt109 catalysis as the initial step in the development of a novel antifungal agent. We established a collaboration with the Broad Institute (Cambridge, MA) to perform a high-throughput screen of 300,000 compounds. From these, I identified a single chemical, termed KB7, which specifically inhibits Rtt109 catalysis, with no effect on other HAT enzymes tested. KB7 has an IC50 value of approximately 60 nM and displays noncompetitive inhibition regarding both acetyl-coenzyme A and histone substrates. With the genotoxic agent camptothecin, KB7 causes a synergistic decrease in C. albicans growth rate. However, this effect is only observed in an efflux-pump mutant, suggesting that this compound would be more effective if it were better retained intracellularly. Further studies through structure-activity relationship (SAR) modifications will be conducted on KB7 to improve its effective cellular concentration.

The Rip2 kinase contains a caspase recruitment domain (CARD) and has been implicated in the activation of the transcriptional factor NF-кB downstream of Nod-like receptors. However, how Rip2 mediates innate immune responses is still largely unclear. We show that Rip2 and IKK-γ become stably polyubiquitinated upon treatment of cells with the Nod2 ligand, muramyl dipeptide. We demonstrate a requirement for the E2 conjugating enzyme Ubc13, the E3 ubiquitin ligase Traf6 and the ubiquitin activated kinase Tak1 in Nod2-mediated NF-кB activation. We also show that M. tuberculosisinfection stimulates Rip2 polyubiquitination. Collectively, this study revealed that the Nod2 pathway is ubiquitin regulated and that Rip2 employs a ubiquitin-dependent mechanism to achieve NF-кB activation. We also demonstrate that intraphagosomal M. tuberculosis stimulates the cytosolic Nod2 pathway. We show that upon Mtb infection, Nod2 recognition triggers the expression of type I interferons in a Tbk1- and Irf5-dependent manner. This response is only partially impaired by the loss of Irf3 and therefore, differs fundamentally from those stimulated by bacterial DNA, which depends entirely on this transcription factor. This difference appears to result from the unusual peptidoglycan produced by mycobacteria, which we show is a uniquely potent agonist of the Nod2/Rip2/Irf5 pathway. Thus, the Nod2 system is specialized to recognize bacteria that actively perturb host membranes and is remarkably sensitive to Mycobacteria, perhaps reflecting the strong evolutionary pressure exerted by these pathogens on the mammalian immune system.

The importance of insulin delivery and action is best characterized in Type 2 Diabetes, a disease that is becoming a pandemic both nationally and globally. Obesity is a principal risk factor for Type 2 Diabetes, and adipocyte function abnormalities due to adipose hypertrophy and hyperplasia, have been linked to obesity. Numerous reports suggest that the intracellular and systemic consequences of adipocyte function abnormalities include adipocyte insulin resistance, enhanced production of free fatty acids, and production of inflammatory mediators. A hallmark of adipocyte insulin sensitivity is the stimulation of glucose transporter isoform 4 (GLUT4) trafficking events to promote glucose uptake. In the Type 2 diabetic and insulin resistant states the mechanism behind insulin-stimulated GLUT4 trafficking is compromised. Therefore, understanding the role of factors involved in glucose-uptake in adipose tissue is of great importance. Studies from our laboratory suggest an important role for the unconventional myosin, Myo1c, in promoting insulin-mediated glucose uptake in cultured adipocytes. Our observations suggest that depletion of Myo1c in cultured adipocytes results in a significant reduction in the ability of adipocytes to take up glucose following insulin treatment, suggesting Myo1c is required for insulin-mediated glucose uptake. A plausible mechanism by which Myo1c promotes glucose uptake in adipocytes has been suggested by further work from our laboratory in which expression of fluorescently-tagged Myo1c in cultured adipocytes induces significant membrane ruffling at the cell periphery, insulin-independent GLUT4 translocation to the cell periphery, and accumulation of GLUT4 in membrane ruffling regions. Taken together Myo1c seems to facilitate glucose uptake through remodeling of cortical actin. In the first part of this thesis I, in collaboration with others, uncovered a possible mechanism through which Myo1c regulates adipocyte membrane ruffling. Here we identified a novel protein complex in cultured adipocytes, comprising Myo1c and the mTOR binding partner, Rictor. Interestingly our studies in cultured adipocytes suggest that the Rictor-Myo1c complex is biochemically distinct from the Rictor-mTOR complex of mTORC2. Functionally, only depletion of Rictor but not Myo1c results in decreased Akt phosphorylation at serine 473, but depletion of either Rictor or Myo1c results in compromised cortical actin dynamic events. Furthermore we observed that whereas the overexpression of Myo1c in cultured adipocytes causes remarkable membrane ruffling, Rictor depletion in cells overexpressing Myo1c significantly reduces these ruffling events. Taken together our findings suggest that Myo1c, in conjunction with Rictor, modulates cortical actin remodeling events in cultured adipocytes. These findings have implications for GLUT4 trafficking as GLUT4 has been previously observed to accumulate in Myo1c-induced membrane ruffles prior to fusion with the plasma membrane. During our studies of adipocyte function we noticed that current siRNA electroporation methods present numerous limitations. To silence genes more effectively we employed a lentivirus-mediated shRNA delivery system, and to standardize this technology in cultured adipocytes we targeted Myo1c and MAP4K4. Using this technology we were able to achieve clear advantages over siRNA oligonucleotide electroporation techniques in stability and permanence of gene silencing. Furthermore we showed that the use of lentiviral vectors in cultured adipocytes did not affect insulin signaling or insulin-mediated glucose uptake events. Despite our inability to use lentiviral vectors to achieve gene silencing in mice we were able to achieve adipose tissue-specific gene silencing effects in mice following manipulation of the lentiviral conditional silencing vector, and then crossing resulting founders with aP2-Cre mice. Interestingly however, only founders from the MAP4K4 conditional shRNA vector, but not founders from the Myo1c conditional shRNA vector, showed gene knockdown, possibly due to position-effect variegation. Taken together, findings from these studies are important because they present an alternative means of achieving gene silencing in cultured adipocytes, with numerous advantages not offered by siRNA oligonucleotide electroporation methods. Furthermore, the in vivo, adipose tissue-specific RNAi studies offer a quick, inexpensive, and less technically challenging means of achieving adipose tissue-specific gene ablations relative to traditional gene knockout approaches.

Crucial to a cell’s ability to migrate is the organization of its plasma membrane and associated proteins in a polarized manner to interact with and respond to its surrounding environment. Cells interact with the extracellular matrix (ECM) through specialized contact sites, including podosomes and invadopodia. Tumor cells use F-actin-rich invadopodia to degrade ECM and invade tissues; related structures, termed podosomes, are sites of dynamic ECM interaction and degradation. We show here that supervillin (SV), a peripheral membrane protein that binds F-actin and myosin II,reorganizes the actin cytoskeleton and potentiates invadopodial function. Overexpressed SV increases the number of F-actin punctae, which are highly dynamic and co-localize with markers of podosomes and invadopodia. Endogenous SV localizes to the cores of Src-generated podosomes in COS-7 cells and with invadopodia in MDA-MB-231 cells. EGFP-SV overexpression increases the average amount of matrix degradation; RNAi-mediated downregulation of SV decreases degradation. Cortactin, an essential component of both podosomes and invadopodia, binds SV sequences in vitro and contributes to the formation of EGFP-SV induced punctae. Additionally, SV affects cortactin localization,which could provide a mechanism for SV action at invadopodia. The formation of cholesterol-rich membrane rafts is one method of plasma membrane organization. A property of membrane rafts is resistance to extraction with cold Triton X-100 and subsequent flotation to low buoyant densities. The actin cytoskeleton has been implicated in many signaling events localized to membrane rafts, but interactions between actin and raft components are not well characterized. Our laboratory isolated a heavy detergent resistant membrane fraction from neutrophils, called DRM-H, that contains at least 23 plasma membrane proteins. DRM-H is rich in cytoskeletal proteins, including fodrin, actin, myosin II, as well as supervillin. DRM-H also contains proteins implicated in both raft organization and membrane-mediated signaling. DRM-H complexes exhibit a higher buoyant density than do most DRMs (referred to as DRM-L), which are deficient in cytoskeletal proteins. By using similar purification methods, I find that COS-7 cells also contain cytoskeleton-associated DRMs. In addition, when transfected into COS-7 cells, estrogen receptor (ER)α associates with DRM-H, while ERβ is seen in both DRM-L and DRM-H populations, suggesting a role for DRM-H in nongenomic estrogen signaling. Thus, the cytoskeleton-associated DRM-H not limited to hematopoietic cells and could constitute a scaffold for membrane raftcytoskeleton signaling events in many cells. Taken together, our results show that SV is a component of cytoskeleton-associated membrane rafts as well as podosomes and invadopodia, and that SV plays a role in invadopodial function. SV, with its connections to both membrane rafts and the cytoskeleton, is well situated to mediate cortactin localization, activation state, and/or dynamics of matrix metalloproteases at the ventral cell surface for proper matrix degradation through invadopodia. The molecular dissection of invadopodia formation and function may contribute to a greater understanding of in vivo invasion, and thus, tumor cell metastasis.

Cu, Zn superoxide dismutase (SOD1) is a dimeric, β-sandwich, metalloenzyme responsible for the dismutation of superoxide. Mutations covering nearly 50% of the amino acid sequence of SOD1 have been found to acquire a toxic gain-of-function leading to amyotrophic lateral sclerosis. A hallmark of this disease is the presence of insoluble aggregates containing SOD1 found in the brain and spinal cord. While it is unclear how these aggregates or smaller, precursor oligomeric species may be the source of the toxicity, mutations leading to increased populations of unstable, partially folded species along the folding pathway of SOD1 may be responsible for seeding and propagating aggregation. In an effort to determine the responsible species, we have systematically characterized the stability and folding kinetics of five well studied ALS variants: A4V, L38V, G93A, L106V and S134N. The effect of the amino acid substitutions was determined on a variety of different constructs characterizing the various post-translational maturation steps of SOD1: folding, disulfide bond formation and Zn binding. Zn was found to bind progressively tighter along the folding pathway of SOD1, minimizing populations of monomeric species. In contrast, ALS variants were found to have the greatest perturbation in the equilibrium populations of the folded and unfolded state for the most immature, disulfide-reduced metal-free SOD1. In this species, at physiological temperature, four out of five ALS variants were >50% unfolded. Finally the energetic barriers in the folding and unfolding reaction were studied to investigate the unusually slow folding of SOD1. These results reveal that both unfolding and refolding are dominated by enthalpic barriers which may be explained by the desolvation of the chain and provide insights into the role of sequence in governing the folding pathway and rate.

Human Immunodeficiency Virus Type 1 (HIV-1) is the causative agent of Acquired Immunodeficiency Syndrome (AIDS), currently the leading cause of death from infectious diseases. Since HIV-1 co-opts the host cellular machinery, the study of cellular factors involved is a rational approach in discovering novel therapeutic targets for AIDS drug development. In this thesis, we present studies on two such proteins. APOBEC3G is from the family of cytidine deaminases known to keep endogenous retroviruses and retrotransposons at bay to maintain stability of the human genome. APOBEC3G targets Vif-deficient HIV-1 particles and renders them noninfectious, partially through deaminase-dependent hypermutation of the provirus during reverse transcription. APOBEC3G largely localizes in mRNA processing (P) bodies, cytoplasmic structures involved in RNA metabolism. Here we explore the significance of APOBEC3G localization in P bodies. We found that disrupting P bodies does not affect virion incorporation of endogenous APOBEC3G, implying that the APOBEC3G fraction in P bodies is not directly involved in the production of nascent, non-infectious particles. We also study UPF1, another host protein encapsidated by HIV-1. It is an essential protein mainly studied for its role in nonsense-mediated decay (NMD) pathway and belongs to the same helicase superfamily as MOV10, a recently identified antiviral factor. We found that UPF1 is incorporated in HIV-1 virions in a nucleocapsid-dependent manner and is required for single-cycle infectivity at an early, post-entry step of the viral life cycle. This novel function of UPF1 most likely does not involve NMD since depletion of UPF2 does not affect viral infectivity.

KCNE1 (E1) peptide is the founding member of the KCNE family (1-5), which is a class of type I transmembrane ß-subunits. KCNE1 peptides assemble with and modulate the gating, ion conducting properties and pharmacology of a variety of voltage-gated K+ channel a-subunits, including KCNQ1 (Q1). Mutations that interfere with the function of either E1 and/or Q1 and disrupt the assembly and trafficking of KCNE1- KCNQ1 channel complexes give rise to diseases such as Romano-Ward (RW) and Jervell Lange Nielsen Syndrome (JLNS), two different forms of Long QT Syndrome (LQTS). Using enzymatic deglycosylation assays, immunofluorescence techniques and quantitative cell surface labeling, we showed that KCNE1 peptides are retained in the early stages of the secretory pathway as immaturely N-linked glycosylated proteins. KCNE1 co-assembly with KCNQ1 leads to E1 progression through the secretory pathway and glycan maturation, resulting in cell surface expression. N-linked glycosylation of some membrane proteins is critical for proper folding, co-assembly and subsequent trafficking through the biosynthetic pathway. Previous studies have shown that genetic mutations that disrupt one of the two N-linked glycosylation sites on KCNE family members lead to LQTS (T7I, KCNE1 and T8A, KCNE2) (Schulze-Bahr et al., 1997; Sesti et al., 2000a; Park et al., 2003). Having confirmed that KCNE1 proteins acquire N-linked glycans, we examined the kinetics and efficiency of N-linked glycan addition to KCNE1. We showed that KCNE1 has two distinct N-linked glycosylation sites. The N-terminal sequon is a traditional co-translational site. The internal sequon (which is only ~ 20 residues away from the N-terminal sequon) acquires N-linked glycans primarily after protein synthesis (post-translationally). Surprisingly, mutations that prevent N-glycosylation at the cotranslational site also reduce the glycosylation efficiency of post-translational glycosylation at the internal sequon, resulting in a large population of unglycosylated KCNE1 peptides that are retained in the early stages of the secretory pathway and do not reach the cell surface with their cognate K+ channel. We showed that KCNE1 post-translational N-glycosylation in the endoplasmic reticulum is a cellular mechanism that ensures E1 proteins acquire the maximal number of glycans needed for proper channel assembly and trafficking. Our findings provide a new biogenic mechanism for human disease by showing that the JLNS mutation, T7I, not only inhibits glycosylation of the N-terminal sequon, but also indirectly prevents the glycosylation of the internal sequon, giving rise to a large population of assembly incompetent hypoglycosylated KCNE1 peptides. To further investigate the two N-linked glycosylation sites on KCNE1, we generated structure-function deletion scans of KCNE1 and performed positional glycosylation scanning mutagenesis. We examined the glycosylation pattern of glycosylation mutants in an effort to define the glycosylation window important for proper KCNE1 assembly and trafficking. Our findings suggested a nine amino acid periodicity to serve as a desirable glycosylation site and a better substrate for N-glycosylation. Appendix II shows work on the characterization of the C-terminally HA-tagged KCNE1 protein, which was used throughout the experiments presented in Chapter II, Chapter III and Chapter IV. Analysis of the C-terminally HA-tagged KCNE1 protein revealed that in heterologous expression systems KCNE1 had an internal translational start site, a methionine at position 27. A proteolytic cleavage site was also identified at the arginine cluster spanning residues 32 through 38 bearing the two known Long QT mutations (R32H and R36H) (Splawski et al., 2000; Napolitano et al., 2005). My work in Professor Craig C. Mello’s lab during the first four years of my graduate study is presented in Appendix I. The highly conserved Wnt/Wingless glycoproteins regulate many aspects of animal development. Wnt signaling specifies endoderm fate by controlling the fate of EMS blastomere daughters in 4-cell stage Caenorhabditis elegans embryos. A suppressor genetic screen was performed using two temperature sensitive alleles of mom-2/Wnt to identify additional regulators of the Wnt/Wingless signaling pathway during C. elegans endoderm specification. Five intragenic suppressors and three extragenic suppressors of mom-2/Wnt embryonic lethality were identified. We cloned ifg-1, eIF4G homologue, as one of the extragenic suppressors suggesting an intriguing connection between the Wnt signaling pathway and the translational machinery.

Obesity has increased globally in epidemic proportions and as have the associated disorders. Insulin resistance that could further lead to type 2 diabetes is a major obesity associated dysfunction. Studies using insulin resistant mouse models and observations from human subjects exhibiting insulin resistance provide evidence for ectopic lipid deposition in organs like liver, muscle and heart as one of the major risk factors for developing insulin resistance. These observations suggest that deregulated adipose function to sequester and store excess energy as fat, could lead to insulin resistance. Furthermore, several studies have demonstrated adipose tissue dysfunction leading to inflammation and related syndromes. Interestingly, a mouse model with transgenic expression of glucose transporter in the adipose tissue exhibited improved glucose tolerance and increased insulin sensitivity despite development of obesity, upon high fat feeding. Thus mechanisms that improve adipose function could alleviate insulin resistance and associated diseases. Mitogen activated protein kinase kinase kinase kinase 4 (MAP4K4) was identified in our laboratory as a negative regulator of adipocyte function. Interestingly, siRNA mediated knockdown of MAP4K4 promoted PPARγ protein expression. Additionally, silencing of MAP4K4 increased adipocyte triglyceride content. Because MAP4K4 is a negative regulator of PPARγ expression and adipocyte function, understanding the mechanism by which MAP4K4 regulates PPARγ expression is of interest. Thus, for the first part of this thesis, I characterized the signaling pathways utilized by MAP4K4 to regulate PPARγ expression in cultured adipocytes. Here I show that MAP4K4 regulates PPARγ expression through regulation of its protein translation. siRNA mediated MAP4K4 gene silencing stimulated PPARγ protein synthesis without changing its mRNA transcription or its protein degradation. This increase in PPARγ protein translation was due to an increase in the activity of mammalian target of rapamycin (mTOR). The increase in PPARγ protein expression mediated by mTOR activation was a specific effect of the 4E-BP1 phosphorylation that leads to its inactivation and was not a general increase in mTOR activity towards all of its substrates. Finally, adenovirus mediated over expression of MAP4K4 inhibited mTOR activation, and suppressed PPARγ protein translation. For the second part of this thesis, I assessed the role of MAP4K4 in adipocytes in vivo. To accomplish this, a lentivirus mediated shRNA construct was generated to attenuate MAP4K4 expression selectively in the mouse adipose tissue. First we demonstrate that the MAP4K4 shRNA construct is able to efficiently silence the expression of MAP4K4 in vitro when co-expressed with Cre recombinase. Furthermore, we show that following modification of the lentiviral conditional vector that was introduced into a mouse embryo at one cell stage, and crossing the resulting founders with aP2-Cre mice, adipose tissue specific MAP4K4 gene silencing was achieved. Moreover, shRNA mediated gene silencing is a faster and an inexpensive means of achieving tissue specific gene knockdown relative to the available traditional gene knockout approaches. Utilizing these adipose specific MAP4K4 gene knockdown mice, I reveal that MAP4K4 silencing enhanced fat mass as well as PPARγ expression significantly. This is accompanied by improved whole body insulin sensitivity. Furthermore, when challenged with high fat diet, adipose-specific MAP4K4 silenced mice exhibit enhanced adiposity with decreased lean mass. Moreover, adipocyte cell size and triglyceride content are significantly increased. Interestingly, despite increased adiposity, hepatic insulin sensitivity is significantly improved leading to decreased glucose output. Thus MAP4K4 is an important regulator of adipocyte function that mediates whole body glucose homeostasis, through a mechanism that is yet to be identified.

Viral proteases have been shown to be effective targets of anti-viral therapies for human immunodeficiency virus (HIV) and hepatitis C virus (HCV). However, under the pressure of therapy including protease inhibitors, the virus evolves to select drug resistance mutations both in the protease and substrates. In my thesis study, I aimed to understand the mechanisms of how this protease−substrate co-evolution contributes to drug resistance. Currently, there are no approved drugs against dengue virus (DENV); I investigated substrate recognition by DENV protease and designed cyclic peptides as inhibitors targeting the prime site of dengue protease. First, I used X-ray crystallography and subsequent structural analysis to investigate the molecular basis of HIV-1 protease and p1-p6 substrate coevolution. I found that co-evolved p1-p6 substrates rescue the HIV-1 I50V protease’s binding activity by forming more van der Waals contacts and hydrogen bonds, and that co-evolution restores the dynamics at the active site for all three mutant substrates. Next, I used aprotinin as a platform to investigate DENV protease–substrate recognizing pattern, which revealed that the prime side residues significantly modulate substrate affinity to protease and the optimal interactions at each residue position. Based on these results, I designed cyclic peptide inhibitors that target the prime site pocket of DENV protease. Through optimizing the length and sequence, the best inhibitor achieved a 2.9 micromolar Ki value against DENV3 protease. Since dengue protease does not share substrate sequence with human serine proteases, these cyclic peptides can be used as scaffolds for inhibitor design with higher specificity.

Animals have evolved complicated and overlapping mechanisms to guard against the development of cancer and infection by pathogenic organisms. ARF, a potent tumor suppressor, positively regulates p53 by antagonizing p53’s negative regulator, MDM2, which in turn results in either apoptosis or cell cycle arrest. ARF also has p53-independent tumor suppressor activity. The CtBP transcriptional co-repressors promote cancer cell survival and migration/invasion. CtBP senses cellular metabolism via a regulatory dehydrogenase domain, and is a target for negative regulation by ARF. ARF targets CtBP to the proteasome for degradation, which results in the up regulation of proapoptotic BH3-only proteins, and p53-independent apoptosis. CtBP inhibition by ARF also up regulates PTEN, reducing cancer cell motility, making CtBP a potential therapeutic target in human cancer. The CtBP dehydrogenase substrate 4-methylthio-2-oxobutyric acid (MTOB) can act as a CtBP inhibitor at high concentrations, and is cytotoxic to cancer cells from a wide variety of tissues. MTOB induced apoptosis was independent of p53, and correlated with the de-repression of the pro-apoptotic CtBP repression target Bik. CtBP over-expression, or Bik silencing, rescued MTOB-induced cell death. MTOB did not induce apoptosis in mouse embryonic fibroblasts (MEFs), but was increasingly cytotoxic to immortalized and transformed MEFs, suggesting that CtBP inhibition may provide a suitable therapeutic index for cancer therapy. In human colon cancer cell peritoneal xenografts, MTOB treatment decreased tumor burden, and induced tumor cell apoptosis. To verify the potential utility of CtBP as a therapeutic target in human cancer the expression of CtBP and its negative regulator ARF was studied in a series of resected human colon adenocarcinomas. CtBP and ARF levels were inversely-correlated, with elevated CtBP levels (compared with adjacent normal tissue) observed in greater than 60% of specimens, with ARF absent in nearly all specimens exhibiting elevated CtBP levels. Targeting CtBP with a small molecule like MTOB may thus represent a useful and widely applicable therapeutic strategy in human malignancies. ARF has long been known to respond to virally encoded oncogenes. Recently, p14ARF was linked to the innate immune response to non-transforming viruses in mice. Therefore a wider role for the ARF pathway in viral infection was considered. Previous studies linking p53 to multiple points of the Human Immunodeficiency Virus-1 (HIV-1) life cycle suggested that ARF may also play a role in the HIV life cycle. In this study the interdependency of ARF and HIV infection was investigated. ARF expression was determined for a variety of cell types upon HIV infection. In every case, ARF levels exhibited dynamic changes upon HIV infection-in most cases ARF levels were reduced in infected cells. The impact of ARF over-expression or silencing by RNAi on HIV infection was also examined. Consistently, p24 levels were increased with ARF overexpression, and decreased when ARF was silenced. Thus ARF and HIV modulate each other, and ARF may paradoxically play a positive role in the HIV life cycle.