Structure Determination of the Fcs Domain of Polycomb Repressive Complex 1 from Drosophila Melanogaster

Polycomb group (PcG) proteins play an important regulatory role in gene expression. These proteins are crucial in stem cell regulation and normal development in all organisms. The FCS domain is part of a multimeric polycomb protein complex called Polycomb repressive complex 1 (PRC1). Based on previous studies, the domain has been hypothesized as a link binding PRC1 to polylinker DNA. This protein/DNA interaction allows other proteins to associate with the repression complex, creating a higher order repressed chromatin structure. This study seeks to determine the structure of the FCS domain via protein crystallization that utilizes a fusion protein construct. The fusion protein methodology attaches the FCS domain to maltose binding protein (MBP), promoting crystallization. The FCS-MBP protein was over-expressed and purified to more than 95% purity followed by observation of FCS-MBP protein crystals. This study determined an adequate expression system, purified the FCS-MBP protein, and observed and refined crystal growth in the FCS-MBP construct. Further optimization of the crystals is required to analyze and determine the FCS domain structure. INTRODUCTION Gene expression is the process by which genetic material is used as a template for the manufacturing of gene products including proteins. Proteins play an integral part in every functional cell and, as expected, gene expression is a complex and highly regulated process. Regulatory processes monitor genetic expression according to environmental factors and cellular signals, thereby allowing cells to express a unique combination of genetic products to suit functional needs. Each cell possesses its own individual function that contributes to the needs of an organism. A mutation that alters genetic expression or diverts a cell from its expected purpose often affects an organism adversely. Similarly, organisms are affected unfavorably by additional information such as extra chromosomes. These changes disrupt the delicate balance of genetic information essential for normal function and generally result in unwanted consequences such as cancer or cellular/organismal death. Gene silencing, a mechanism exhibited by all cells, functions to maintain the genetic balance in organisms. Proteins assist in the silencing process by a variety of mechanisms, many of which are still speculative. Recently much work has been done on regulatory proteins known as Polycomb proteins. Studies have shown that Polycomb proteins play an integral part in dosage compensation in X-linked genes in humans (1). Also, more than 150 genes involved with cell growth and proliferation have been identified that may be subjected to Polycomb protein repression (2). It is because of their important regulatory role in cellular development, stem cells, and cancer that much attention has been directed towards better understanding Polycomb proteins (3,4). Polycomb group (PcG) proteins arise from conserved DNA sequences found in all organisms and known as Polycomb group genes. PcG proteins aggregate into large multiprotein complexes that work on chromatin, creating higher order structures silencing targeted genes over C. TONG: STRUCTURE DETERMINATION OF THE FCS DOMAIN 2 many mitotic divisions. Two multimeric complexes, Polycomb repressive complex 1 (PRC1) and Polycomb repressive complex 2 (PRC2), have been most abundantly studied because of their intrinsic collaborative behavior. PRC1 and PRC2 are composed of several individual PcG proteins, all contributing to the overall function of the repressive complexes. PRC2 core members include the enhancer of zeste [E(z)], extra sex combs (Esc), the suppressor of zeste 12 [Su(z)12], and P55 (5). PRC1 contains the core proteins, polyhomeotic (Ph), posterior sex combs (Psc) (Bmi-1 in humans), RING1, and polycomb (Pc) (10,11). PRC2 possesses methyltransferase activity attributed to the SET domain within E(z) with specificity for K9 and K27 of histone 3 (6-9). This histone methytransferase activity establishes a binding site for PRC1 recruiting the repression complex to the targeted gene. PRC1 is then responsible in compacting the targeted genes, as marked by PRC2, creating a repressed chromatin structure (12). Although known to exhibit these actions, the exact mechanisms involved in the formation of the repressed chromatin structure are undiscovered. PRC1 has been known to inhibit chromatin remodeling enzymes which could be attributed to a possible mechanism (13). It has also been observed that PRC1 is coupled with transcription factors such as TBP, TFIIF, and TFIIB resulting in an alternative method of repression (14, 15). Thanks to the seemingly multiple capabilities of PRC1 one can isolate functional domains, determine the structure of such sequences, and piece together the individual proteins to propose a mechanism that explains PRC1’s unique abilities. This study focuses on a 30 amino acid sequence called the FCS domain located in various subunits in PRC1. The FCS domain is a conserved sequence in all organisms. It is named appropriately after the homologous sequence of phenylalanine (F), cysteine (C), and serine (S) amino acids prevalent in all variations of the FCS domain. Because of the arrangement of the amino acids, the sequence is assumed to exhibit a Zn-ribbon fold. This Zn-binding domain has been unappreciated in the past, but with recent evidence of the FCS domain’s unique activities, the FCS domain has proven to be essential to the overall repressive function of PRC1. Interestingly the FCS domain is able to exhibit non-sequence-specific binding to DNA via the conserved first two cysteine residues (16). The zinc finger motif of TFIIA is known to bind to RNA and DNA (17). The sequences of the FCS domain are homologically similar to those of the zinc finger motif of TFIIA. Because of these similarities, a proposed FCS Zn-ribbon motif is possible; however, the actual FCS structure is still unknown. Based on recent studies, one may hypothesize that the FCS domain is a link binding PRC1 to polylinker DNA, allowing other proteins to associate with the repression complex creating the higher order repressed chromatin structure. Although much literature defends this hypothesis, there is no physical evidence of the actual mechanism used by FCS in gene silencing. Structural analysis of a protein often gives insight into the behavior of a molecule. Proposed models provide bases for understanding mechanisms, functions, and interactions of biomolecules. Indeed, structures have been an essential tool for understanding molecules, from Rosalind Franklin’s X-ray crystallography, the foundation for Watson and Crick’s DNA structure, to recent discoveries about the mechanisms of C-RING1B of PRC1 (5). Because of the importance of understanding the structure of the FCS domain, various human and Drosophila FCS domains were studied. Preliminary studies of the FCS domain by Professor Chongwoo Kim and his lab showed that the NMR was an inadequate technique for determining the structure of the FCS domain (C. Kim, personal communication, June 8, 2008). This conclusion was based primarily on problematic structure evaluations. Attempts to crystallize the FCS domain following NMR failed because of the extremely soluble nature of the domain. This study reexamines the FCS domain’s structure by using an alternative crystallographic approach. Two versions of the human FCS domain, hPh1 and hPh3, and one version in Drosophila melanogaster, sex comb on midleg (Scm) were purified and used for this study based on prior success with FCS domains (C. Kim, personal communication, June 8, 2008). TCNJ JOURNAL OF STUDENT SCHOLARSHIP VOLUME XII APRIL, 2010 3 This alternative method used protein fusion to couple a binding protein to the FCS domain. Maltose binding protein (MBP) was chosen because of its success in previous structural studies (18-20). With this new approach, two versions of the FCS-MBP sequences of hPh1, hPh3 and Scm with two different linker sequences were cloned, providing six distinctly different FCS-MBP samples. The FCS-MBP crystals produced by this study, a small but crucial domain of PRC1 could provide insight into fundamental mechanisms of PcG protein repressed chromatin structures and their function in organismal regulation and development. EXPERIMENTAL PROCEDURES Protein Cloning and Expression. hPh1 (residues 796-828), hPh3 (residues 781-813), and SCM (residues 59-130) amino acids were attached to MBP amino acid sequence with a His-tag (6H) via two different linker sequences. The sequences were cloned via restriction enzymes into pETMxt and pBADMxt vectors. The pBADMxt and pETMxt vectors were transformed into ARI814 and BL21(DE3) pLysS E. coli cells respectively (21). The cells were added to a starter culture of Luria-Bertani (LB) media with Ampicillin solution. Chloramphenicol was also added to the BL21(DE3) pLysS culture to maintain the pETMxt plasmid. These starter cultures were grown overnight on an incubated shaker at 250 RPM and 37 ̊C. The overnight starter cultures were used to inoculate several 1L volumes of LB media with proper antibiotics. The cells were grown and induced with 20% Arabinose for ARI814 cells and Isopropyl-β-D-Thiogalactopyranoside for BL21(DE3) pLysS cells. The growth culture cells were harvested and frozen at -80 ̊C. Protein Purification. A typical protein purification protocol involved re-suspending the cells from the harvested culture in a buffer consisting of 50mM Tris (pH8), 200 mM NaCl, 25 mM imidazole (pH 7.5), 10mM β-mercaptoethanol, and 1 mM PMSF. The cells were lysed by sonication, centrifuged in AvanatiJ20: JA20/15K, and separated into supernatant and cellular debris. All proteins were extracted from the supernatant by means of Ni affinity chromatography (NiSepharose). Tobacco etch virus (TEV) was used to cut the His-tag from the fusion protein. This was followed by ion exchange chromatography (HiTrap SP and HiTrap Q column). Protein solutions were concentrated using a Millipore stirred Ultrafiltration unit to approximately 30 mg/ml. Figure 1 schematically represents the protein constructs following purification with their approximate masses. C. TONG: STRUCTURE DETERMINATION OF THE FCS DOMAIN 4 Figure 1. Schematic drawings (not to scale) of varying FCS domains expressed by their appropriate vectors. Below the illustrations are the approximate molecular weights of the protein products. Yellow box represents MBP. The blue box represents the varying linker sequences. Pink box is the FCS domain of hph1, hph3, and Scm.