Tag Archives: NMR spectroscopy

Molecular mechanism of specificity in promoter DNA recognition

July 2, 2021Sciencec-myc, NMR spectroscopy, Promoter, Protein-DNA interaction, Structure, X-ray crystallographyNeel

Recognition of *c-myc* gene promoter by human RBMS1 protein. The figure shows RBMS1 protein scanning the DNA in search of its target sequence and binding to the sequenceâ€™s nucleotides in a specific manner. The surface view of the crystal structure of RBMS1 protein is shown bound with the bases of nucleotides of *c-myc* promoter. .

Decoding the codes of life stored in DNA is a challenge which is performed in a tightly regulated manner by the cell. Decoding involves reading the correct code at a right time by DNA binding proteins. The decoded information is then channeled through mRNA. This requires an efficient and highly specific interaction between protein and DNA that control some of the most important processes pertaining to cell survival and growth. Any dysregulation in the process can lead to malfunctioning of the cell and disease. Proteins search and bind specific sequence in the background of billions of bases in the genome. This happens through a combination of 1D sliding, 2D hopping, and 3D diffusion.

Understanding specificity of protein-DNA interactions is a long-standing question that has been attempted to understand many times by scientists all over the world. Thermodynamics and kinetics have always been discussed to be behind the matters concerning DNA binding specificity and affinity. The evidence, however, remains scarce when answering the crucial biological questions.

In this background we have worked on human RBMS1 protein that has been shown to directly regulate the c-myc gene expression levels in cancerous cells. In our study we have elucidate the molecular basis of RBMS1-promoter DNA interaction and understand the mechanism for specificity. The work provides the first structural and dynamics characterization of human RBMS1 protein, that controls the expression of c-myc proto-oncogene inside the human cell by its interaction with 7 base pair consensus sequence within the 21 bp promoter/ autonomous origin of replication region 2 kb upstream of c-myc proto-oncogene.

During the work we overcame different challenges. Bioinformatic studies have failed to correctly identify the domain boundary and this was leading to protein instability. It required a careful human analysis using some of fundamentals of molecular biophysics to redesign the construct. We extended the boundary by 5 residues which lead to expression of highly stable RBMS1 protein. This is a lesson for students who blindly trust bioinformatics results and artificial intelligence.

We performed extensive binding assay with different DNA sequences varying the bases and length of the sequences to answer the question of specificity. We determined the 2.57 Ã… crystal structure of RBMS1 in its promoter DNA bound state that provided atomic-resolution insight into specific binding of individual nucleotides of DNA with the protein. The NMR structure of free RBMS1 was solved, as the protein did not crystallize, most likely due to its inherent flexibility, which we confirmed through NMR relaxation dynamics. The protein undergoes a hinge-like motion in order to bind with the specific DNA which facilitated by flexibility in the linker region. The X-ray structure of RBMS1-c-myc promoter DNA complex and solution NMR structure of RBMS1 protein helped us to delineate the non-canonical binding mode of RBMS1 with the promoter DNA.

In a nutshell the mechanism of specificity of RBMS1 binding with promoter is driven by the thermodynamics and the dynamic domain re-organization, which is responsible for conferring specificity and affinity. The work has implications for understanding general mechanism of protein-DNA interactions that involves sliding, hopping and diffusion during stochastic target search process in a dense nucleus. In addition, the work is likely to aid in designing future anti-gene therapies.

—————

References

Aggarwal P* and Bhavesh NS* (2021) Hinge like domain motion facilitates human RBMS1 protein binding to proto-oncogene c-myc promoter. Nucleic Acids Res. 49, 5943-5955
Structure co-ordinates: Protein Data Bank (PDB) accession numbers 6M75 and 7C36.

Observation of a novel domain swapped dimer in RRM domain

June 18, 2021ScienceDND1, Domain swapped dimer, NMR spectroscopy, RRM, X-ray crystallographyNeel

2.3 Ã… resolution structure of human DND1-RRM2(PDB id 6LE1). The work was done by Ms. Pooja Kumari

Cell, the functional unit of life, stores codes of its functioning in long strings called DNA, which is made of different combination of four chemicals namely A, G, C and T. Specific stretches of DNA is called gene or coding region and thus a cell contains many genes. These genes code specifically to make proteins through a messenger entity called mRNA which decode the code. The proteins take birth at ribosomes inside cell as a string where each bead in one amino acid selected out of 20 naturally occurring ones. The proteins immediately folds, either assisted or in a non-assisted manner, to form a distinct three-dimensional shape. The three-dimensional organization dictates the functions of proteins, which are the workhorses of cell performing all functions in a very well-orchestrated manner. Depending on time, type and location of cell, one gene can code for different variant of proteins, called isoforms, that can have different shape and functions. As Francis Crick has said; â€œThe ultimate aim of biology is in fact to explain all biology in terms of physics and chemistry. Eventually one may hope to have the whole of biology â€œexplainedâ€ in terms of the level below it, and so on right down to the atomic levelâ€. In order to understand the function of proteins at atomic level it is absolutely essential to see their three-dimensional structures at atomic-resolution. X-ray crystallography, NMR spectroscopy, X-ray free electron laser (XFEL) and single-particle Cryo Electron Microscopy are used to determine the atomic-resolution three-dimensional structures of proteins.

Our group has been working on a class of RNA binding protein domain known as RNA Recognition Motif (RRM), or Ribonucleoprotein domain (RNP), is found to be the most abundant nucleic acid binding domain in higher vertebrates. In humans about 2% of the genome codes for RRM and more than 500 RRM containing proteins have been annotated. They are found in single copy, multiple copies or in conjunction with other domains in the same protein. Studies show that RRM domains not only recognize nucleic acid but are also engaged in protein-protein interactions, protein lipid interactions owing to their structural plasticity. RRMs typically have about 90 amino acids with a representative topology of Î²1-Î±1-Î²2-Î²3-Î±2-Î²4 , wherein the 4 stranded Î²-sheets are packed against the 2 Î±-helices. There has been several examples where length of loops, Î±-helix and Î²-sheet vary and additional secondary structured element at RRM extremities are well conserved. These are crucial for interaction allowing longer RNA recognition. In of our previous study we have discovered a novel mode of RNA interaction where an extended Î±1Î²2 loop of human TAF-15 protein recognised a stem-loop RNA.

Due to its unique amino acid composition we started working on human DND1 (Dead end protein homolog1) also known as DND MicroRNA-Mediated Repression Inhibitor 1. DND1 is an RNA binding protein containing two RRMs, RRM1, RRM2 domains, in tandem and a double stranded RNA binding domain at the C-terminal separated by 40 residues flexible linker. In humans, reports of dysregulation at DND1 or mutation in the gene have been associated with testicular cancer, tongue squamous cell carcinoma. We have determined atomic-resolution crystal structure of DND1-RRM2 and observed a novel domain swapped dimer formation like one seen in HIV-1 protease. We validated the domain swapped dimer formation in solution using nOe and relaxation data from NMR spectroscopy. Ours is the first report of domain swapped dimer formation in any RRM and it also points to the limitation faced by structure prediction tools. In conclusion there is no substitute to experimentally determined structure if we have to appreciate the diversity in proteins.

References

Kumari P* and Bhavesh NS* (2021) Human DND1-RRM2 forms a non-canonical domain swapped dimer. Protein Sci. 30, 1184-1195
Structure co-ordinates: Protein Data Bank (PDB) accession numbers 6LE1.

Structure of an intermediate state during protein unfolding

January 24, 2020ScienceNMR spectroscopy, Protein, Protein Folding, StructureNeel

Cover art (Designed by Anupam Patra) featured on 21st Jan 2020 issue of Biophysical Journal (Left). Dr. Harshesh Bhatt and Dr. Akshay Kumar Ganguly (Right)

Proteins are the unique molecules; they read the code of life in genes and their own codes are also hidden in genes. Therefore, they are called architect, pillars and workhorses of life. A vast majority of proteins requires a well-defined three-dimensional shape and flexibility for their specific and regulated function inside the cell. Hierarchy of protein structure starts with primary structure consisting of individual amino acid residues, which then organizes itself into helices or sheets to form secondary structures. The great Prof. G. N. Ramachandran was the first to codify geometry of the secondary structural elements in the form of backbone torsion angles. The iconic plot is called Ramachandran Plot. Proteins further fold these secondary structural elements into tertiary and quaternary structures.

Process of formation of these structures, called Protein Folding, starts immediately at their birth place called Ribosomes and most proteins emerge as perfectly folded native structure at the end of their complete synthesis. However, there are few which require helper protein molecules called chaperone to correctly shape them. Different kinds of protein have different lifespan and at the end of it they are degraded to individual units with the help of another set of protein molecules, the process called Protein Unfolding. Both folding and unfolding processes are very specific, rapid and follows a well-defined path. Levinthal Paradox described by Prof. Cyrus Levinthal concluded that had it been a random search process among all possible conformation it would take astronomical age to fold a protein. Any aberration in folding-unfolding processes leads to dysregulation and sometimes formation of aggregated state, resulting in disease conditions such as Alzheimer and Parkinson disease.

Prof. Christian B. Anfinsen had used ribonuclease A as a model showed that primary sequence of the protein under a particular environmental conditions (temperature, solvent concentration and composition, etc.) at which folding occurs, dictates formation of native structure, which is a unique, stable and kinetically accessible minimum of the free energy. Several studies have demonstrated that proteins unfold under different physical and chemical conditions in a test tube and refold back to their native confirmation upon removal of such conditions. This helped the experimentalists to follow folding-unfolding pathways at a high spatial and time resolution and understand mechanisms. Landscape funnel model of protein folding is currently widely accepted model that describes a protein’s journey through the crests and troughs of internal free energy and degrees of freedom to reach the folded state from rapidly exchanging ensemble. Proteins also take a trek through the funnel during their unfolding.

The key to understanding protein (un)folding mechanism, which is still a challenge, is high-resolution structural characterisation of all states along the funnel. In this regards solution-state NMR spectroscopy is unparalleled in providing atomic-resolution structural and dynamics information of all the states, folded, unfolded and intermediates including invisibles, along the funnel. In this work we have used a canonical RNA recognition motif (RRM) from ETR3 protein (involved in muscular dystrophy disease) to understand the unfolding mechanism of RRM containing protein. This is important as a similar motif in TDP-43 protein aggregated and eventually leads to neuromuscular disease conditions due to formation of non-native structural elements. We determined the atomic-resolution structure and dynamics of folded native state (at bottom of the funnel) and an unfolding intermediate at the middle of the funnel and performed structural and dynamics characterisation of the unfolded ensemble rapidly dancing the top of the funnel. Our data indicates that intermediate state has folded like structure but swollen and dynamics. It allows solvent to access its core more easily than the folded structure. The edges of secondary structural elements are the initiation sites of unfolding. Although the unfolded state is very dynamic and lacks any structural elements still it show bias for certain structural propensities which appears like keeping a memory of their folded state.

We believe that the atomic-resolution characterisation of unfolding pathway of a canonical RNA recognition motif is likely to help in understanding the unfolding events in other RRMs involved in disease causing conditions upon misfolding.

The study might also help in understanding those systems where protein is refolded from the purified inclusion bodies. In those cases there could be presence of minor population of folded-like but not fully folded native structure.

The work was primarily done by Dr. Harshesh Bhatt and Dr. Akshay Kumar Ganguly and FCS work was performed in collaboration with Dr. Sobhan Sen, School of Physical Sciences, Jawaharlal Nehru University (JNU), New Delhi. The cover art was designed by Mr. Anupam Patra, Ph.D. student.

References

Dill K.A. and Chan H.S. (1997) From Levinthal to pathways to funnels. Nat. Struct. Biol. 4: 10-19
Bhatt H, Ganguly AK, Sharma S, Kushwaha GS, Khan MF, Sen S and Bhavesh NS (2020) Structure of an unfolding intermediate of an RRM domain of ETR-3 reveals its native-like fold. Biophys. J. 118, 352-365.
Structure co-ordinates: Protein Data Bank (PDB) accession numbers 2MY7 and 2MY8. Sequence-specific resonance assignments: BioMagResBank (BMRB) IDs: 19316 and 19685.

First report of molecular mechanism of RNA recognition in Malaria parasite

December 12, 2018Scienceà¤ªà¥à¤°à¥‹à¤Ÿà¥€à¤¨, à¤ªà¥à¤²à¤¾à¤œà¥à¤®à¥‹à¤¡à¤¿à¤¯à¤®, à¤®à¤²à¥‡à¤°à¤¿à¤¯à¤¾, Malaria, NMR spectroscopy, Plasmodium, Protein, RNA, StructureNeel

Structure of the PfSR1-RNA complex. Dr. Akshay Kumar Ganguly and Ms. Garima Verma who did the work.

Among the vector borne diseases malaria affects the maximum population. In 2017 alone there were about 22 crore cases more than 4 lakh death worldwide with India accounting for 6% of them¹. Malaria is caused in humans when a female Anopheles mosquito transfer protozoan parasites of the genus Plasmodium and Plasmodium falciparum is the most lethal of the malarial parasites. The economic consequence of this widespread disease is also apparent from the fact that malaria endemic countries have much lower GDP growth than countries free from malaria². In recent years drug resistance in malaria has emerged as a major threat to the public health. It is thus, imperative to elucidate the ways in which this parasite thrives in its hosts as well as the various mechanisms it employs to sustain an infection, in order to be able to effectively combat this menace. Malaria parasite uses an intelligent mechanism called alternate splicing to generate protein diversity. Alternate Splicing is one of the evolved mechanisms in higher organisms to produce different protein from a single gene by differentially processing the transcribed mRNA. This gives the parasite an added advantage to evade the drugs. To understand this important cellular event in malaria parasite we delineated the molecular mechanism of RNA binding of one such protein called serine/arginine rich (SR) protein of P. falciparum (PfSR1)⁵, which is the first such study in apicomplexans. Using Nuclear Magnetic Resonance (NMR) spectroscopy we determined atomic-resolution solution structures of PfSR1 protein⁶ in free and RNA bound states.

Supported by thermodynamic quantification we found that RNA binding domains of PfSR1 protein have contrasting preference for RNA while first domain has preference for pyrimidine especially 5â€™cytosine, other prefers purines (A or G), possibly due to different charges of the surface of both domains.

Our work opens a new window to understand how malaria parasite is able to produce 20% more proteins from its 5700 genes^3-4 which makes it more complex than many of the organisms having more genes than it and easy to evolve into drug resistant one.

The work was primarily done by Dr. Akshay Kumar Ganguly and was actively supported by Ms. Garima Verma.

References

World Malaria Report 2018 (https://www.who.int/malaria/publications/world-malaria-report-2018/en/)
Gallup JL and Sachs JD. (2001).The economic burden of malaria. Am J Trop Med Hyg 64, 85-96
Gardner MJ, Hall N, Fung E, White O, Berriman M, Hyman RW, et al. (2002). Genome sequence of the human malaria parasite Plasmodium falciparum, Nature. 419 498-511.
Sorber K, Dimon MT and DeRisi JL. (2011) RNA-Seq analysis of splicing in Plasmodium falciparum uncovers new splice junctions, alternative splicing and splicing of antisense transcripts. Nucleic Acids Res 39, 3820-3835.
Eshar S, Allemand E, Sebag A, Glaser F, Muchardt C, Mandel-Gutfreund Y, et al. (2012). A novel Plasmodium falciparum SR protein is an alternative splicing factor required for the parasitesâ€™ proliferation in human erythrocytes, Nucleic Acids Res 40, 9903-9916.
Ganguly AK, Verma G and Bhavesh NS(2019) The N-terminal RNA recognition motif of PfSR1 confers semi-specificity for pyrimidines during RNA recognition.Â J. Mol. Biol. 431, 498-510
Structure co-ordinates: Protein Data Bank (PDB) accession numbers 2N3L and 2N7C. Sequence-specific resonance assignments BioMagResBank (BMRB) IDs: 25650, 25800, 25779.