My laboratory has had a long-standing interest in regulation of gene expression and cell proliferation as they relate to cancer. Our early work focused on the transcriptional regulation of genes that lack a TATA element in their promoter, which was once thought to be a canonical feature. Genes regulated by TATA-less promoters comprise approximately 80% of the genome and include genes involved in regulation of many metabolic processes, DNA replication, DNA repair, and apoptosis, as well as growth factors and their receptors, oncogenes and tumor suppressors.
Promoters for these genes are GC-rich and most contain multiple sites that bind the transcription factor Sp1. We and others demonstrated that in TATA-less promoters, Sp1 functions to control transcription initiation and recruit the general transcription machinery through protein-protein interactions with a component of the TBP-containing general transcription factor, TFIID. Although a ubiquitous transcription factor, regulation of a large number of genes in response to a wide array of signals has been ascribed to Sp1. Sp1 is a 785 aa protein containing 96 serine and 66 threonine residues that are extensively post-translationally modified by phosphorylation and O-linked glycosylation; it is also modified by acetylation, sumoylation and ubiquitylation. These modifications affect not only DNA binding, but also Sp1 activity and its interactions with other factors.
Sp1 and the DNA Damage Response
Several years ago, a graduate student in the lab discovered that Sp1 is significantly phosphorylated in response to DNA damage. Sp1 is phosphorylated at many different sites by different kinases to modulate its activity in response to multiple signals. Much of our current work is focused on phosphorylation in response to DNA damage and the role of Sp1 in the cellular response to damage. Eleven of the 96 Ser residues in Sp1 are SQ sequences clustered in the glutamine-rich transactivation domains; S/TQ cluster domains (SCDs) are characteristic of proteins phosphorylated by ATM/ATR in response to DNA damage.
We have found that Sp1 is phosphorylated by ATM on several Ser residues in response to reactive oxygen species (ROS) generated by DNA damage and that its phosphorylation is involved in the increased sensitivity to DNA damage observed in cells depleted of Sp1. Phosphorylation on S101 is required for additional phosphorylation, i.e., its phosphorylation primes for additional phosphorylation, and we have made an antibody that specifically detects Sp1 phosphorylated on S101. We have shown by immunofluorescence/confocal microscopy and chromatin immunoprecipitation that phospho-Sp1 is localized to sites of DNA damage and that its phosphorylation is dependent on the presence of Nbs1, a key component of the MRN complex that recruits ATM to DSB sites.
We are also studying the role of Sp1 in the induction of apoptosis after DNA damage. Sp1 is preferentially degraded by caspases at higher levels of damage. Degradation of Sp1 is associated with induction of apoptosis, and blocking its caspase-mediated cleavage (by mutation D183A) protects cells from damage-induced apoptosis. The caspase-mediated cleavage results in stabilization of the remaining Sp1 that encompasses the DNA binding domain (Sp1183-785), and release the N-terminal portion (Sp11-182), which is capable of rescuing DNA damage sensitivity and mediating chromatin modifications. Sp1 mediates increased H3K18 acetylation through recruitment of p300 and decreased H4K16 acetylation through recruitment of HDAC1/2 at DSBs, which allows 53BP1 binding to H4K20me. Together these data indicate that Sp1 facilitates recruitment of 53BP1 to break sites, to promote NHEJ. Sp1-pS101 co-localizes with 53BP1 and shows no overlap with BRCA1 at IRIF.
There are several studies underway and planned to establish the clinical significance of our findings. Sp1 overexpression has been reported in several cancers and is an indicator of poor prognosis; however, there are no reports of specific mutations in Sp1 in tumors.
Sp1 and Chromosomal Instability
Whole chromosomal instability (CIN) is a dynamic and continual gain or loss of whole chromosomes during cell division. It is associated with poor patient outcomes in multiple cancer types, as well as tumor heterogeneity and resistance to multiple chemotherapeutics, underscoring its clinical significance. Despite its prevalence and clinical importance, the exact mechanism(s) that leads to CIN remain to be determined. We have found that depletion of Sp1 results in abnormal chromosome alignment along the metaphase plate, micronuclei and aneuploidy, all of which are phenotypes consistent with CIN.
One potential mechanism that may explain the relationship between Sp1 depletion and CIN is centrosome amplification, a phenomenon shown to contribute to CIN. Previously, we reported that Sp1 localizes to the centrosome and that depletion of Sp1 results in centrosome amplification. Further, this process appears to be independent of Sp1’s function as a transcription factor. A truncated form of Sp1 lacking the N-terminal 182 amino acids, which retains 91% of Sp1’s transcriptional activity, fails to localize at the centrosome and fails to rescue the supernumerary centrosome phenotype we see in Sp1 depleted cells. We have found that the 182 amino acid N-terminal portion of Sp1 alone can localize to the centrosome. Importantly, this mutant is not transcriptionally active, indicating that Sp1 transcriptional activity is not required for its centrosomal localization. Further, RNAi-mediated depletion of full length Sp1 and re-expression of this N-terminal portion can rescue the supernumerary centrosome phenotype. Taken together, these data indicate that Sp1 regulates centrosome number in a non-transcriptional manner.
We have also discovered that Sp1 localizes to the centromere during mitosis. This localization is dependent on ATM activity, and does not require the Sp1 DNA binding domain. Loss of Sp1 results in disrupted centrochromatin, including changes in histone modifications and transcription of α-satellite arrays. Further, loss of Sp1 results in defects in centromeric cohesion, as well as a decrease in centromeric protein A (CENP-A) deposition at the centromere, suggesting that Sp1 may be important for maintaining the structure and function of this region. We hypothesize that Sp1 contributes to faithful chromosome segregation through a novel function(s) at the centromere during mitosis, thereby maintaining chromosomal stability.
Herpes simplex virus-1 (HSV-1) infects the cornea, oral epithelium and esophagus. Herpes keratitis (HK) is the leading cause of both cornea-derived and infection-associated blindness in the developed world. Clinical management of HSV-1 infection largely relies on the use of nucleoside analog antiviral drugs, which target the HSV-1 DNA polymerase enzyme. Despite the availability of effective antivirals, some patients develop refractory disease, drug-resistant infection and topical toxicity. Multidrug-resistant HSV-1 is particularly common and serious in the growing immune-compromised population, and is confounded by the lack of therapeutic target diversity of the currently used antivirals. Novel treatment modalities may offer a unique advantage in the management of such cases.
We demonstrate that HSV-1 infection of corneal epithelial or oral keratinocyte cells results in rapid activation of ataxia telangiectasia mutated (ATM), an apical kinase in the DNA damage response. Pharmaceutical inhibition of ATM causes potent suppression of HSV-1 replication in cultured cells and in explanted human and rabbit corneas. Furthermore, ATM inhibition reduces stromal keratitis and herpes labialis in the mouse eye and lip without causing appreciable toxicity. Investigations of the underlying molecular events revealed the presence of a specialized viral mechanism, whereby the immediate early gene product ICP4 interacts with the viral genome to produce early and robust activation of ATM. This activation is shown to be independent of the presence of DNA lesions and does not appear to be mediated by the viral protein ICP0. The downstream significance of ATM activation remains to be determined. Inhibition of checkpoint kinase 2 (Chk2), a direct target of ATM, causes pronounced suppression of HSV-1 replication in corneal epithelial cells and in explanted corneas, as well as in oral epithelial cells in vitro and infection of the mouse lip in vivo, suggesting that the pro-viral effect of ATM activation may be mediated by Chk2. We have further determined that ATM activity is essential for virus circularization upon nuclear entry, an event that precedes rolling circle replication. These findings warrant further investigation into the DNA damage response as an area for therapeutic intervention in herpetic diseases.
Diabetic Retinopathy and Sp1
Sp1 was the first transcription factor shown to be O-glycosylated; however, the sites of glycosylation in response to specific signals have not been thoroughly mapped and the functional significance of glycosylation remains a mystery. We have established a model of diabetic retinopathy (DR) using retinal endothelial cells exposed to high glucose and found that increased transcription of VEGF, which has been implicated in DR, is mediated through the O-linked glycosylation of Sp1. We are currently mapping sites of O-glycosylation, and the effect of agents that block glycosylation on Sp1-dependent functions.
Back to Top