The Runx family of transcription factors performs an essential role in animal development by controlling gene expression programs that mediate cell proliferation, growth and differentiation. The work described in this thesis is concerned with understanding mechanisms by which Runx proteins support this program of gene expression within the architectural context of the mammalian cell nucleus. Multiple aspects of nuclear architecture are influenced by Runx2 proteins including sequence-specific DNA binding at gene regulatory regions, organization of promoter chromatin structure, and higher-order compartmentalization of proteins in nuclear foci. This work provides evidence for several functional activities of Runx2 in relation to architectural parameters of gene. expression for the control of cell growth and differentiation. First, the coordination of SWI/SNF mediated chromatin alterations by Runx2 proteins is found to be a critical component of osteoblast differentiation for skeletal development. Several chromatin modifying enzymes and signaling factors interact with the developmentally essential Runx2 C-terminus. A patent-pending microscopic image analysis strategy invented as part of this thesis work - called intranuclear informatics - has contributed to defining the C-terminal portion of Runx2 as a molecular determinant for the nuclear organization of Runx2 foci and directly links Runx2 function with its organization in the nucleus. Intranuclear informatics also led to the discovery that nuclear organization of Runx2 foci is equivalently restored in progeny cells following mitotic division - a natural perturbation in nuclear structure and function. Additional microscopic studies revealed the sequential and selective reorganization of transcriptional regulators and RNA processing factors during progression of cell division to render progeny cells equivalently competent to support Runx2 mediated gene expression. Molecular studies provide evidence that the Runx proteins have an active role in retaining phenotype by interacting with target gene promoters through sequence-specific DNA binding during cell division to support lineage-specific control of transcriptional programs in progeny cells. Immunolocalization of Runx2 foci on mitotic chromosome spreads revealed several large foci with pairwise symmetry on sister chromatids; these foci co-localize with the RNA polymerase I transcription factor, Upstream Binding Factor (UBFl) at nucleolar organizing regions. A series of experiments were carried out to reveal that Runx2 interacts directly with ribosomal DNA loci in a cell cycle dependent manner; that Runx2 is localized to UBF foci within nucleoli during interphase; that Runx2 attenuates rRNA synthesis; and that this repression of ribosomal gene expression by Runx2 is associated with cell growth inhibition and induction of osteoblast-specific gene expression. This thesis has identified multiple novel mechanisms by which Runx2 proteins function within the hierarchy of nuclear architecture to control cell proliferation, growth and differentiation.

The insulin/IGF-1 signaling pathway plays a pivotal role in life span regulation in diverse organisms. In Caenorhabditis elegans, a PI 3-kinase signaling cascade downstream of DAF-2, an ortholog of the mammalian insulin and insulin-like growth factor-1 (IGF-1) receptor, negatively regulates DAF-16/forkhead transcription factor. DAF-16 then regulates a wide variety of genes involved in longevity, stress response, metabolism and development. DAF-16 also receives signals from other pathways regulating life span and development. However, the precise mechanism by which DAF-16 directs multiple functions is poorly understood. First, in Chapter II, we demonstrate that JNK is a novel positive regulator of DAF-16 in both life span regulation and stress resistance. Our genetic analysis suggests that the JNK pathway acts in parallel with the insulin-like signaling pathway to regulate life span and both pathways converge onto DAF-16. We also show that JNK-1 directly interacts with and phosphorylates DAF-16. Moreover, in response to heat stress, JNK-1 promotes the translocation of DAF-16 into thc nucleus. Our findings define a novel interaction between the stress response pathway (JNK) and the master regulator of life span (DAF-16), and provide a mechanism by which JNK regulates longevity and stress resistance. Next, in Chapter III, we focus on the downstream targets of DAF-16. Here, we used a modified chromatin immunoprecipitation (ChIP) method to identify direct target promoters of DAF-16. We cloned 103 target sequences containing consensus DAF-16 binding sites and randomly selected 33 targets for further analysis. The expression of majority of these genes is regulated in a DAF-16-dependent manner. Moreover, inactivation of more than 50% of these genes significantly altered DAF-16-dependent functions such as longevity, fat storage and dauer diapause. Our results show that the ChIP-based cloning strategy leads to greater enrichment of DAF-16 target genes, compared to previous studies using DNA micro array or bioinformatics. We also demonstrate that DAF-16 is recruited to multiple promoters to coordinate regulation of its downstream target genes. In summary, we identified the JNK signaling pathway as a novel input into DAF-16 to adapt animals to the environmental stresses. We also revealed a large number of novel outputs of DAF-16. Taken together, these studies provide insight into the complex regulation by DAF-16 to control diverse biological functions and eventually broaden our understanding of aging.

The overall objective of this study was to identify novel proteins of the nuclear matrix in order to contribute to a better understanding of nuclear structure and organization. To accomplish this, a monoclonal antibody specific for the nuclear matrix was used to screen a human λgt11 expression library. Several cDNAs were isolated, cloned, sequenced, and shown to represent NuMA, the nuclear mitotic spindle apparatus protein. Further characterization of the gene and RNA was undertaken in an effort to obtain information about NuMA. The NuMA gene was present at a single site on human chromosome 11q13. Northern and PCR analysis of NuMA mRNA showed a major 7.2 kb transcript and minor forms of 8.0 and 3.0 kb. The minor forms were shown to be alternatively spliced although their functional significance is not yet understood. Immunofluorescence microscopy demonstrated that NuMA oscillates between the nucleus and the microtubule spindle apparatus during the mitotic cell cycle. NuMA appeared as a 200-275 kDa protein detectable in all mammalian cells except human neutrophils. To determine whether NuMA's changes in intracellular distribution correlated with post-translational modifications, the protein's phosphorylation state was examined through the cell cycle using highly synchronized cells. NuMA was a phosphoprotein in interphase and underwent additional phosphorylation events in mitosis. The mitotic phosphorylation events occurred with similar timing to lamin B (G2/M transition) and were concomitant with NuMA's release from the nucleus and its association with the mitotic spindle. However, the mitotic phosphorylation occurred in the absence of spindle formation. Dephosphorylation of NuMA did not correlate with reassociation with the nuclear matrix but occurred in two distinct steps after nuclear reformation. Based on the timing of these events, phosphorylation may playa role in nuclear processes. In conclusion, the work in this dissertation identified NuMA, a nuclear matrix protein and showed that it is phosphorylated during the cell cycle and may be important for nuclear events such as nuclear organization, transcription, or initiation of DNA replication at G1/S.

p38 mitogen-activated protein (MAP) kinases represent a subgroup of MAP kinases that respond to environmental stress and inflammatory cytokines. p38 MAPK is activated by two upstream kinases, MKK3 and MKK6, by dual phosphorylation on threonine and tyrosine in conserved kinase subdomain VII. Until recently the relative roles of MKK3 and MKK6 have remained unclear. I have undertaken two strategies in an effort to understand the importance of MKK3 as a p38 MAPK activator. First, I cloned and characterized the murine mkk3 gene and determined the structure of the 5'-terminus. Comparison of the murine and human mkk3 genes revealed that the mouse gene encodes a single MKK3 isoform, MKK3b, and the human gene encodes two isoforms, MKK3a and MKK3b. Comparison of the mouse and human mkk3 genes suggests that expression of MKK3a and MKK3b is regulated from different promotors. Analysis of the mkk3 promoter demonstrates that muscle specific expression of murine MKK3b is controlled, in part, by the transcription factors MEF2 and MyoD. Second, I have utilized a gene targeting strategy to disrupt the murine mkk3 gene and to examine the effect on p38 MAPK signaling. I found that there is a p38-specific signaling defect in MKK3 deficient primary mouse embryo fibroblasts (MEF) which correlates with deficits in interleukin (IL)-1 and IL-6 production in response to tumor necrosis factor-α (TNFα) stimulation. In addition there is a defect in TNFα mediated expression of TNFα and macrophage inflammatory proteins (MIP) 1α, MIP1β and MIP2. p38 MAPK-specific signaling defects were also observed in lipopolysaccharide (LPS) stimulated mkk3 (-/-) macrophages. Additionally, mkk3 (-/-) macrophages exhibit defects in LPS and CD40-ligand (CD40L) stimulated IL-12 biosynthesis. Similar data were obtained from CD40L-stimulated mkk3 (-/-) dendritic cells. I also observe that interferon (Ifn)-γ production is diminished during T-helper-1 (TH1) differentiation of CD4+ T-cells derived from mkk3 (-/-) mice. Taken together these data demonstrate a crucial role for p38 MAPK activation by MKK3 in response to the inflammatory cytokine, TNFα and during a TH1 inflammatory response.

Exposure to external stimuli promotes a variety of cellular responses including changes in morphology, gene expression and cell division status. These responses are promoted by signaling pathways composed of modules that are conserved from lower to higher eukaryotes. In Saccharomyces cerevisiae response to the external stimuli provided by mating pheromone is governed by the pheromone response pathway. This pathway is composed of a G protein coupled receptor/heterotrimeric G protein (Gαβγ) module and a MAP kinase cascade. Activation of this pathway allows the heterotrimeric G protein βγ dimer (Gβγ) to recruit polarity proteins to promote changes in cell morphology and to activate signaling through the MAP kinase cascade. Here we investigate the regulation of these pheromone-induced responses. We first examine how an asymmetric polarization response is generated. Normally, a gradient of pheromone serves as a spatial cue for formation of a polarized mating projection, but cells can still polarize when pheromone is present uniformly. Here we show that an intact receptor/Gαβγ module is required for polarization in response to both a gradient and uniform concentration of pheromone. Further investigation into regulation of Gβγ by Gα revealed that the two interaction interfaces between Gα and Gβ have qualitatively different roles. Our results suggest that one interface controls signaling whereas the other governs coupling to the receptor. Overall our results indicate that communication between the receptor and Gαβγ is required for proper polarization. We then examine how G1 CDKs regulate MAP kinase signaling. Response to pheromone is restricted to the G1 stage of the cell cycle. Once cells commit to a round of division they become refractory to mating pheromone until that round of division is complete. One contributor to this specificity involves inhibition of signaling through the MAP kinase cascade by G1 CDKs, but it was not known how this occurs. Here, we show that the MAP kinase cascade scaffold Ste5 is the target of this inhibition. Cln/CDKs inhibit signaling by phosphorylating sites surrounding a small membrane-binding domain in Ste5, thereby disrupting the membrane localization of Ste5. Furthermore, we found that disrupting this regulation allows cells to arrest at an aberrant non-G1 position. Our findings define a mechanism and a physiological benefit for restricting pheromone-induced signaling to G1. This thesis describes findings related to generation of an asymmetric polarization response, heterotrimeric G protein function, and coordination of differentiation signaling with cell division status. Lessons learned here might be applicable to the regulation of polarization and differentiation responses in other systems as the signaling modules are conserved.

A common phenotype observed in most cancers is chromosomal instability. This includes both structural and numerical chromosomal aberrations, which can promote carcinogenesis. The fusion gene CBFB/MYH11 is created by the structural chromosomal inversion(16)(p13.1q22), resulting in the fusion protein CBFβ-SMMHC, which blocks differentiation in hematopoietic progenitor cells. This mutation alone, however, is not sufficient for transformation, and at least one additional cooperating mutation is necessary. The role of wildtype Cbfb in modulating the oncogenic function of the fusion protein Cbfβ-SMMHC in mice was examined. Transgenic mice expressing the fusion protein, but lacking a wild-type copy of Cbfb, were created to model the effects of these combined mutations. It was found that wild-type Cbfb is necessary for maintaining normal hematopoietic differentiation. Consequently, complete loss of wild-type Cbfb accelerates leukemogenesis in Cbfb/MYH11 mice compared to mice expressing both the fusion and wild-type proteins. While there is no evidence in human patient samples that loss of wild-type Cbfb expression cooperates with the fusion protein to cause transformation, it is apparent from these experiments that wild-type Cbfβ does play a role in maintaining genomic integrity in the presence of Cbfβ-SMMHC. Experiments have also shown that loss of Cbfb leads to accumulation of hematopoietic progenitor cells, which may acquire additional cooperating mutations. Not unlike CBFB/MYH11, the human papillomavirus (HPV) E6 and E7 proteins are not sufficient for cellular transformation. Instead, high risk HPV E7 causes numerical chromosomal aberrations, which can lead to accumulation of additional cooperating mutations. Expression of HPV-16 E7 and subsequent downregulation of the retinoblastoma protein (Rb) has been shown to induce polyploidy in human keratinocytes. Polyploidy predisposes cells to aneuploidy and can eventually lead to transformation in HPV positive cells. There are several possible mechanisms through which E7 may lead to polyploidization, including abrogation of the spindle assembly checkpoint, cleavage failure, abrogation of the postmitotic checkpoint, and re-replication. Rb-defective mouse and human cells were found to undergo normal mitosis and complete cytokinesis. Furthermore, DNA re-replication was not found to be a major mechanism to polyploidization in HPV-E7 cells upon microtubule disruption. Interestingly, upon prolonged mitotic arrest, cells were found to adapt to the spindle assembly checkpoint and halt in a G1-like state with 4C DNA content. This post-mitotic checkpoint is abrogated by E7-induced Rb-downregulation leading to S-phase induction and polyploidy. This dissertation explores two examples of the multi-step pathway in human cancers. While certain genes or genetic mutations are often characteristic of specific cancers, those mutations are often not sufficient for transformation. The genetic or chromosomal abnormalities that they produce often stimulate the additional mutations necessary for oncogenesis. The studies with Cbfb/MYH11 and HPV E7 further exemplify the significance of numerical and structural chromosomal aberrations in multi-step carcinogenesis.

Circadian clocks are biological timekeepers that help maintain an organism’s behavior and physiological state optimally timed to the Earth’s day/night cycle. To do this, these internal pacemakers must accurately keep track of time. Equally importantly, they must be able to adjust their oscillations in response to external time cues to remain properly synchronized with the environment, and correctly anticipate environmental changes. When the internal clock is offset from its surrounding day/night cycle, clinically relevant disruptions develop, ranging from inconveniences such as jet-lag to more severe problems such as sleep disorders or mood disorders. In this work, I have used the fruit fly, Drosophila melanogaster, as a model organism to investigate how light and temperature can synchronize circadian systems. My initial studies centered on an intracellular photoreceptor, CRYPTOCHROME (CRY). CRY is a blue light photoreceptor previously identified as a major component of the primary light-input pathway into the Drosophila circadian clock. We used molecular techniques to show that after light-activation, CRY binds to the key circadian molecule TIMELESS (TIM). This interaction irreversibly targets TIM, but not CRY, for degradation. Further studies characterizing a newly isolated cry mutant, crym, showed that the carboxyl-terminus of CRY is not necessary for CRY’s ability to impart photic information to the molecular clock. Instead, the C-terminus appears to be necessary for normal CRY stability and protein-protein interactions. Thus, we conclude that in contrast to previous reports on CRYs of other species, where the C-terminal domain was required for transduction of photic information, the C-terminus of DrosophilaCRY has a purely modulatory function. During the second part of my dissertation work, I focused my studies on circadian thermoreception. While the effects of light in synchronization of the Drosophilaclock to environmental cycles have been extensively characterized, significantly less is known about temperature input pathways into the circadian pacemaker. I have used two approaches to look at how temperature affects the circadian system. First, I conducted a series of behavioral analyses looking at how locomotor rhythms can be phase-shifted in response to temperature cycles. By examining the behavior of genetically ablated flies, we determined that the well-characterized neurons controlling morning and evening surges of activity during light/dark cycles are also implicated in morning and evening behaviors under temperature cycles. However, we also find evidence of cells that contribute to modulating afternoon and evening behavior specifically under temperature cycles. These data contribute to a growing number of studies in the field suggesting that pacemaker cells may play different roles under various environmental conditions. Additionally, we provide data showing that intercellular communication plays an important role in regulating circadian response to temperature cycles. When the morning oscillator is absent or attenuated, the evening cells respond abnormally quickly to temperature cycles. My work thus provides information on the roles of different cell groups during temperature cycles, and suggests that beyond simply synchronizing individual oscillating cells, intercellular network activity may also have a role in modulating proper response to environmental time cues. Finally, I present some preliminary work looking at effects of temperature on known circadian molecules. Using a combination of in vivo and cell culture techniques, I have found that TIM protein levels decrease at higher temperatures. My cell culture data suggest that this is a proteasome-independent degradation event. As TIM is also a key molecule in the light-input pathway, the stability of TIM proteins may be a key point of integration for light and temperature input pathways. While additional research needs to be conducted to confirm these effects in vivoin wild-type flies, these preliminary results identify a possible avenue for further study. Taken together, my work has contributed new data on both molecular and neuronal substrates involved in processing light and temperature inputs into the Drosophila circadian clock.

Most organisms, from cyanobacteria to humans are equipped with circadian clocks. These endogenous and self-sustained pacemakers allow organisms to adapt their physiology and behavior to daily environmental variations, and to anticipate them. The circadian clock is synchronized by environmental cues (i.e. light and temperature fluctuations). The fruit fly, Drosophila melanogaster, is well established as a model for the study of circadian rhythms. Molecular mechanisms of the Drosophilacircadian clock are conserved in mammals. Using genetic screens, several essential clock proteins (PER, TIM, CLK, CYC, DBT, SGG and CK-II) were identified in flies. Homologs of most of these proteins are also involved in generating mammalian circadian rhythms. In addition, there are only six neuronal groups in the adult fly brain (comprising about 75 pairs of cells) that express high levels of clock genes. The simplicity of this system is ideal for the study of the neural circuitry underlying behavior. The first half of this dissertation focuses on a genetic screen designed to identify novel genes involved in the circadian light input pathway. The screen was based on previous observations that a mutation in the circadian photoreceptor CRYPTOCHROME (CRY) allows flies to remain rhythmic in constant light (LL), while wild type flies are usually arrhythmic under this condition. 2000 genes were overexpressed and those that showed a rhythmic behavior in LL (like crymutants) were isolated. The candidate genes isolated in the screen present a wide variety of biological functions. These include genes involved in protein degradation, signaling pathways, regulation of transcription, and even a pacemaker gene. In this dissertation, I describe work done in order to validate and characterize such candidates. The second part of this dissertation focuses on identifying the pacemaker neurons that drive circadian rhythms in constant light (LL) when the pacemaker gene period is overexpressed. We found that a subset of pacemaker neurons, the DN1s, is responsible for driving rhythms in constant light. This attractive finding reveals a novel role for the DN1s in driving behavioral rhythms under constant conditions and suggests a mechanism for seasonal adaptation in Drosophila.

The Innate immune system is the first line of defense against invading microbial pathogens. It is a fast-acting and non-antigen-specific defense system, which employs germline encoded surveillance systems capable of responding to a broad-spectrum of pathogens. The innate immune system involves a variety of immune cells, which express different profiles of surveillance or detection receptors. Upon sensing pathogens, these receptors trigger cell signalling to turn on transcription of inflammatory cytokines, chemokines, anti-microbial peptides and type I Interferons. These effectors have direct effects on the control of pathogen load and also activate the adaptive immune system, which is ultimately required to clear infections. The type I interferons (IFNs) are the principal cytokines strongly induced during infection with viruses and are required for direct control of viral replication and modulation of cells of the adaptive immune response. The signalling pathways induced in order to activate type I IFNs are dependent on the interferon regulatory factors (IRFs). Striving for survival, microbes have evolved various strategies to subvert/impair these critical defense molecules. In this thesis work, I have used Human Metapneumoviruses (HMPVs), a relatively newly described family of paramyxoviruses as model viruses to explore the role of pattern recognition receptors (PRRs) and the IRF family of transcription factors in the innate immune response. These studies revealed that the recognition of HMPV viral pathogen-associated molecular patterns (PAMPs) by immune cells is different in different cell types. Retinoic acid-inducible gene-I (RIG-I), a cytosolic RNA helicases senses HMPV-A1 virus for triggering type I IFN activation by detecting its 5’- triphosphate viral RNA in most human cells, including cell lines and primary monocytes. An exception to these findings was plasmacytoid dendritic cells (PDCs), where Toll-like receptor (TLR)-7 is the primary sensor involved in detecting HMPV viruses. By comparing the innate immune response to two HMPV strains, we found that these two closely related strains had very different immune stimulatory capabilities. HMPV-1A strain triggered type I IFNs in monocytes, PDCs and cells of epithelial origin. In contrast, a related strain, HMPV-B1 failed to trigger IFN responses in most cell types. Our studies suggested that the phosphoprotein (P) of HMPV-B1 could prevent the viral RNA from being detected by RIG-I, thus inhibiting the induction of type I IFN production in most cell type examined. This finding adds to our understanding of the mechanisms by which viruses are sensed by surveillance receptors and also unveils new means of viral evasion of host immune responses. Although IRFs are extensively studied for their role in regulating type I IFN activation, especially in TLR and RIG-I like receptor (RLR) signalling pathways upon viral infection, a clear understanding of how this family of transcription factors contributes to anti-viral immunity was lacking. Studies conducted as part of this thesis revealed that in addition to IRF3 and IRF7, which play a central role in anti-viral immunity downstream of most PRRs (e.g. TLRs, RLRs, DNA sensors), the related factor IRF5 was also an important component of innate anti-viral defenses. Using IRF5-deficient mice we studied in detail the role of IRF5 in coordinating antiviral defenses by examining its involvement in signalling downstream of TLRs. These studies led us to examine the role of IRF5 in the regulation of type I IFNs as well as inflammatory cytokines in different cell types. While most TLRs that induced IFNβ showed normal responses in IRF5-deficient mice, CpG-B-induced IFNβ production in CD11c+CDCs isolated from mouse spleen but not those generated in vitro from bone marrow required IRF5. This was in contrast to responses with lipopolysaccharide (LPS) or polyriboinosinic polyribocytidylic acid (polyIC), ligands for TLR4 and 3, respectively. Moreover, we found that in contrast to IRF3 and/or IRF7, IRF5 was important in coordinating the expression of inflammatory cytokines such as TNFα downstream of some TLRs. In addition to our studies to examine the requirement for IRF5 in TLR signaling, we also showed that muramyl peptide (MDP) from Mycobacterium tuberculosis (Mtb) could activate type I IFNs via IRF5. This was the first evidence linking IRF5 to a non-TLR-driven pathway. IRF5 activation in this case was downstream of a novel nucleotide-binding oligomerization domain containing (NOD)-2/receptor-interacting serine-threonine kinase (RIP)-2 signaling pathway. Collectively, the studies outlined in this thesis have assisted in providing a framework to understand the role of TLRs, RLRs and IRFs in the immune response to paramyxoviruses and have unveiled new mechanisms of activation of the IRFs as well as new mechanisms by which pathogens subvert or evade these important innate defense mechanisms.

The KCNE1 and KCNE3 type I transmembrane-spanning β-subunits assemble with the KCNQ1 voltage-gated K+ channel to afford membrane-embedded complexes with dramatically different properties. Assembly with KCNE1 produces the very slowly activating and deactivating IKs current that shapes the repolarization phase of cardiac action potentials. Genetic mutations in KCNQ1 or KCNE1 that reduce IKs current cause long QT syndrome and predispose affected individuals to potentially fatal cardiac arrhythmias. In contrast, complexes formed between KCNQ1 and KCNE3 produce rapidly activating and mostly voltage-independent currents, properties that are essential for function in K+ recycling and Cl−secretion in gastrointestinal epithelia. This thesis addresses how these two homologous accessory peptides impart their distinctive effects on KCNQ1 channel gating by examining two important protein regions: 1) a conserved C-terminal motif in the β-subunits themselves, and 2) the voltage sensing domain of KCNQ1 channels. Sequences in both the transmembrane domain and C-terminus of KCNE1 and KCNE3 have been identified as contributing to the divergent modulatory effects that these β-subunits exert. The homology of transmembrane-abutting C-terminal residues within the KCNE family and the presence of long QT-causing mutations in this region highlight its importance. A bipartite model of modulation was proposed that suggests the transmembrane domain of KCNE1 is passive, allowing the C-terminal domain to control modulation. Chapter II builds on this model by investigating the effect of mutating specific amino acids in the KCNE1 C-terminal domain. Point mutants that produce ‘high impact’ perturbations in gating were shown to cluster in a periodic fashion, suggesting an alpha-helical secondary structure that is kinked by a conserved proline residue and interacts with the Q1 channel complex. In Chapter III, the voltage sensing domain of Q1 channels is examined in the presence of either KCNE1 or KCNE3. To determine the influence of these two peptides on voltage sensing, the position of the S4 voltage sensor was monitored using cysteine accessibility experiments. In the slowly opening KCNQ1/KCNE1 complexes, voltage sensor activation appears to occur much faster than the onset of current, suggesting that slow channel activation is not due to slowly moving voltage sensors. KCNE3, on the other hand, shifts the voltage sensor equilibrium to favor the active state, producing open channels even at negative voltages. Taken together, these findings provide mechanistic detail to illustrate how two homologous peptides radically alter the gating properties of the same K+ channel and present a structural scaffold to map protein-protein interactions.