Category: Molecular DynaMicsposter

Mo02 contact name

matthew Wampole: matthew.wampole@jefferson.edu

Computational design of peptide nucleic acid conjugates for imaging oncogenic KRAS2 mRNA in-vivo

Matthew E. Wampole1, Jeffrey M. Sanders1, Mathew L. Thakur2,3, and Eric Wickstrom1,3 1Biochemistry & Molecular Biology, 2Radiology; 3Kimmel Cancer Center, Thomas Jefferson

University, Philadelphia PA 19107

Lung cancers often show mutation and overexpression of the epidermal growth factor receptor (EGFR) and K-Ras GTPase, a signaling molecule downstream of EGFR. Tyrosine kinase inhibitor (TKI) treatments have benefited some patients with particular EGFR mutations. However, activating mutations in KRAS2 oncogene lead to constitutive K-Ras activity independent of EGFR signaling, enabling resistance to TKIs. Imaging activated KRAS2 mRNA in lung cancer would rule out interventions targeting EGFR. We have demonstrated significant tumor image contrast in breast cancer and pancreatic cancer xenografts with peptide nucleic acid (PNA) dodecamers coupled to receptor-targeting peptides and imaging reporter moieties. KRAS2 mutations are common in the 12th and 13th codons in many cancers; for lung cancer the three mutants seen most frequently are G12D, G12V, and G12C. Multiple mutants complicate the development of a sequence-specific PNA for targeting the mutations, creating the need for specific probe sequences for each mutation. Hypoxanthine substitutions might allow a single PNA to bind multiple mutant mRNAs through wobble base pairing. We wish to predict effective PNA hybridization imaging agents prior to synthesis and testing.

Introduction

Conventional vs. Accelerated Molecular Dynamics Conclusions

RNA and PNA sequences used RNA

KRAS2 wildtype 5'-[GGAGCUGGUGGC]-3' G12C mutant 5'-[GGAGCUUGUGGC]-3' G12D mutant 5'-[GGAGCUGAUGGC]-3' G12V mutant 5'-[GGAGCUGUUGGC]-3'

PNA KRAS2 wildtype COOH-[CCTCGACCACCG]-NH2 Hypoxanthine COOH-[CCTCGACHACCG]-NH2 G12C COOH-[CCTCGAACACCG]-NH2 G12D COOH-[CCTCGACTACCG]-NH2 G12V COOH-[CCTCGACAACCG]-NH2

Background Peptide nucleic acids (PNA)

Ideal for binding to specific mutants

High binding affinity to complementary DNA or RNA

Differentiation of single base mismatch by high destabilization effect

High chemical stability to temperature and pH

High biological stability to nuclease and protease

SUPPORTED BY NIH CA148565 IP owned by EW/MLT, licensed to MTTI Contact: Matthew.Wampole@Jefferson.edu

Wobble Base Pairs

A base pair that can bind to two or three other bases

Important for binding to multiple mutants

Hypoxanthine binds with cytosine, adenine, and uracil

Structural Results

=C'-N1'-C2'-C3' =N1'-C2'-C3'-N4' =C2'-C3'-N4'-C5' =C3'-N4'-C5'-C' =N4'-C5'-C'-N1' =C5'-C'-N1'-C2'

Future Work

Continue AMD simulations

Develop force fields for gamma substitutions on PNA backbone

Polar plots of wildtype RNA with matching PNA sequence

Black: PNA/RNA(MD) Blue2: PNA/RNA(NMR) Green3: d-lys PNA/DNA(X-ray) Red4: PNA/PNA(X-ray)

Backbone angles agree with NMR/x-ray averages

References: 1Andreas W. Goetz; Mark J. Williamson; Dong Xu; Duncan Poole; Scott Le Grand; & Ross C. Walker*, J. Chem. Theory Comput., 2012, 8 (5), pp 1542-1555 2D.A. Case, T.A. Darden, T.E. Cheatham, III, C.L. Simmerling, J. Wang, R.E. Duke, R. Luo, R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. Seabra, J. Swails, A.W. Goetz, I. Kolossvai, K.F. Wong, F. Paesani, J. Vanicek, R.M. Wolf, J. Liu, X. Wu, S.R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews, M.G. Seetin, R. Salomon-Ferrer, C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, and P.A. Kollman (2012), AMBER 12, University of California, San Francisco. 3Brown S.C.; Thomson S.A.; Veal J.M.; Davis D.G. Science 1994, 265, 777-780. 4Menchise V.; De Simone G.; Tedeschi T.; Corradini R.; Sforza S.; Marchelli R.; Capasso D.; Saviano M.; Pedone C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 12021-6. 5Rasmussen H.; Kastrup J.S.; Nielsen J.N. Nielsen J.M.; Nielsen P.E. Nat. Struct. Biol. 1997, 4 98-101.

GPU vs. CPU Performance

Accelerated Molecular Dynamics Higher correlation than conventional MD More motion in the backbone angles as well as in base pair binding

Using GPUs for Molecular Dynamics (MD) calculations give workstations the processing power of High Performance Computers.

Amber 11 (Harold) & Amber 12 with nvcc 4.2 (Kollman)

PMEMD using Double Precision (CPU) or SPFP (GPU)

typical MD run: – &cntrl – imin=0,irest=1, – ntx=5, nstlim=5000000, – dt=0.002, ntc=2, – ntf=2, cut=8.0, – ntb=1, ntp=0, – taup=2.0, ntpr=50000, – ntwx=5000, ntwr=100000, – ntt=3, gamma_ln=2.0, – ig=-1, temp0=300.0, – /

Average number of atoms: 76312

Molecular Mechanics Poisson Boltzmann Surface Area (MMPBSA) estimated binding energies. Tm for the PNA/RNA hybrids were obtained using Circular Dichroism melting curves.

Harold (DoD HPC)

Kollman (local workstation)

Nodes 1344 1 Cores/Node 8 8

Operating System SLES 11 SP1 Ubuntu 12.04 Core Type Intel Xeon quad-core Nehalem Intel Xeon quad-core Nehalem

Core Speed 2.8 GHz 2.6 GHz Memory/Node 24 GBytes 48 GBytes

GPU NA 1x Tesla 1060C

1x GTX 680 1x Quadro FX 5800

Conventional Molecular Dynamics Poor correlation between Tm exp. and Tmcalc. MD trajectories display non-canonical base pairing in mismatched hybrids

Hypoxanthine can bind with G12D and G12V mutants

PNA force fields agree with experimental backbone angles

AMD binding energies correlate with experimental Tm

R² = 0.8458

600

640

680

720

760

65 70 75 80 85

T m(c

alc)

[c]

Tm(exp) [c]

R² = 0.0026

600

640

680

720

760

65 70 75 80 85

T m(c

alc)

[c]

Tm(exp) [c]