Scripps Florida Logo

The Disney Lab

Projects:

Multidimensional Library Screening (Library versus Library), Development of 2-Dimensional Combinatorial Screening (2DCS). Our lab developed a high throughput method to identify RNA motif-small molecule partners, or “paired” spaces.   Named Two-Dimensional Combinatorial Screening (2DCS), a chemical library is covalently immobilized onto an agarose-coated glass slide to afford small molecule microarrays (Fig. 2).  Arrays are then probed for binding to a radioactively labeled RNA library displaying a discrete secondary structure that is commonly found in biological RNA such as a hairpin or an internal loop, for example.  Array hybridization is also completed under conditions of high oligonucleotide stringency such that only interactions of the ligands to the hairpin and the internal loops are selected.  The individual RNA motifs that bind to each ligand are identified by excising the ligand bound RNA from the array surface, amplifying the RNA by RT-PCR, cloning, and sequencing.  Sequencing data are structure modeled to identify the exact RNA motifs that bind a ligand. 

2DCS has several advantages as compared to conventional methods that identify RNA binders by screening a validated RNA drug target against a chemical library.  First, 2DCS probes a large number of interactions in a single experiment.  For example, testing an array-immobilized 500-member library for binding an RNA randomized in six positions (46 or 4,096 members) probes 2,048,000 (4,096 ´ 500) interactions in a single experiment.  This diversity is likely to identify new RNA binders that would go undetected if only a single or a few biologically important RNAs were probed.  Second, 2DCS identifies the RNA motifs preferred by a ligand.  Third, the RNA motif-ligand partners identified by 2DCS are portable.   Thus, RNA motif-ligand interactions may facilitate the rational design of ligands targeting large RNAs that contain similar motifs.

Fig 1: The Disney group has developed Multidimensional, library versus library small molecule screening assays to identify the optimal RNA motif targets for small molecules.

Fig 1:  The Disney group has developed Multidimensional, library versus library small molecule screening assays to identify the optimal RNA motif targets for small molecules.

Computational Chemistry and Rational Drug Design

I. RNA-Privileged Space Predictor (RNA-PSP) Program. In order to determine the privileged RNA space that bound to each ligand from the 2DCS experiment, statistical analysis of the selected structures must be completed.  In these studies, z-scores and two-tailed p-values are computed by comparing the feature of the RNA motifs selected from the library to the entire library.  To streamline these analyses, we developed a computer program called RNA-Privileged Space Predictor (RNA-PSP, Fig. 3) that takes raw sequencing data, extracts the RNA loops that bind ligands and compared their individual occurrence to their occurrence in the original library of structures that were hybridized with the array.  RNA-PSP helps facilitate a more rigorous and facile statistical analysis of the RNA motif space that is selected to bind a ligand.  The program could also have uses in the statistical analysis of a wide variety of nucleic acid selections.  These studies are described in the following publications:  Paul DJ, Seedhouse SJ, and Disney MD.  Two-dimensional combinatorial screening and the RNA Privileged Space Predictor (RNA-PSP) efficiently identify aminoglycoside-RNA hairpin loop interactions. Nucleic Acids Research, (2009), 37, 5894-5907; Velagapudi SP, Seedhouse S, French J, Disney MD. Defining the RNA Internal Loops Preferred by Drug-Like Ligands via Two-Dimensional Combinatorial Screening.  JACS, (2011), 133, 10111-8; Tran T, Disney MD. Two-dimensional combinatorial screening of a bacterial rRNA A-site-like motif library: defining privileged asymmetric internal loops that bind aminoglycosides.  Biochemistry (2010), 49, 1833-1842; Velagapudi SP, Seedhouse SJ, Disney MD. Structure-activity relationships through sequencing (StARTS) defines optimal and suboptimal RNA motif targets for small molecules.  Angewandte Chemie International Edition, English (2010), 49, 3816-3818.

Fig. 3: Schematic illustration of RNA-PSP and Chem-PSP computer programs to identify privileged RNA and small molecule space

Fig. 3:  Schematic illustration of RNA-PSP and Chem-PSP computer programs to identify privileged RNA and small molecule space

II. Structure-Activity Relationships through Sequencing (StARTS). StARTS is an approach to couple computation and experiment to allow the accurate prediction of the affinity of RNA motif–ligand partners identified by two-dimensional combinatorial screening (2DCS). StARTS uses information from the sequences of the RNA motifs selected to bind a ligand. Sequences are statistically analyzed using the RNA Privileged Space Predictor (RNA-PSP) program to determine features (for example, 5’GC steps, Fig. 4) in the selected sequences that occur with >95% confidence. The confidence intervals are associated with a Z-score, with a larger value corresponding to a higher confidence level. Each selected RNA motif can have multiple significant features. Therefore, the sum of the Z-scores for all features is also computed. These data are then plotted against the measured binding affinities and can be fit to an inverse first-order equation, which allow prediction of the affinity of the ligand for any RNA library member. StARTS has the potential to streamline the identification and scoring of both optimal and suboptimal RNA motif–ligand partners selected by 2DCS.  These studies are described in the following publications:  Velagapudi SP, Seedhouse S, French J, Disney MD. Defining the RNA Internal Loops Preferred by Drug-Like Ligands via Two-Dimensional Combinatorial Screening.  JACS, (2011), 133, 10111-8; Velagapudi SP, Seedhouse SJ, Disney MD. Structure-activity relationships through sequencing (StARTS) defines optimal and suboptimal RNA motif targets for small molecules.  Angewandte Chemie International Edition, English (2010), 49, 3816-3818.

III.Privileged Chemical Space Predictor Program (PCSP) Program Modularly assembled combinatorial libraries are often used to identify ligands that bind to and modulate the function of a protein or a nucleic acid. Much of the data from screening these compounds, however, is not efficiently utilized to define structure-activity relationships (SAR, Fig. 3). If SAR data are accurately constructed, it can enable the design of more potent binders. We developed a computer program called Privileged Chemical Space Predictor (PCSP) that statistically determines SAR from high-throughput screening (HTS) data and then identifies features in small molecules that predispose them for binding a target. Features are scored for statistical significance and can be utilized to design improved second-generation compounds or more target-focused libraries.  These studies are described in the following publication:  Seedhouse SJ, Labuda LP, Disney MD. The Privileged Chemical Space Predictor (PCSP): a computer program that identifies privileged chemical space from screens of modularly assembled chemical libraries.  Bioorganic Medicinal Chemistry Letters (2010), 20, 1338-1343.

Fig. 4: Schematic of StARTS and its predictive power for RNA motif-ligand interactions. This computational approach allows for the facile assignment of optimal and sub-optimal RNA motif-small molecule binders. This information is then used in rational drug or chemical probe design.

Fig. 4:  Schematic of StARTS and its predictive power for RNA motif-ligand interactions.  This computational approach allows for the facile assignment of optimal and sub-optimal RNA motif-small molecule binders.  This information is then used in rational drug or chemical probe design.

IV. Rational Drug Design - Using the RNA Motif-Ligand Database – Rational Design of Ligands Targeting Toxic RNAs.  The power of this “bottom-up” that couples the results of 2DCS with genome mining has been illustrated in the rational design of a modularly assembled compound that binds the RNA that causes myotonic dystrophy type 2 (DM2) and modulates its function.  The RNA binding module used to facilitate rational design was identified by 2DCS, which determined that 6’-N-5-hexynoate kanamycin A bound 2 ´ 2 pyrimidine-rich internal loops with the highest affinity.  This type of loop is present many times in the DM2 RNA.   DM2 is caused by a repeat expansion of rCCUG in an intron of the zinc finger 9 protein.  When the repeat is long enough, it folds into a hairpin structure that displays multiple copies of the internal loops that preferentially bind 6’-N-5-hexynoate kanamycin A (Fig. 5).  The repeat binds the RNA splicing regulator muscleblind protein, compromising its normal function.  We developed a modular assembly strategy to target multiple motifs simultaneously using the kanamycin A derivative as the ligand module (Fig. 5).   In this strategy, small molecule modules are anchored onto peptoids displaying either azides using a Huisgen dipolar cycloaddition reaction (HDCR).  The HDCR was used for these studies because ligand modules in 2DCS are anchored onto agarose arrays using the same chemistry.  Peptoids were used to display RNA binding modules because both the spacing and valency of ligand modules can be precisely controlled. The studies resulted in the development of a series of cell-permeable ligands that bind the toxic rCCUG repeats with higher affinity and specificity than muscleblind.

In addition, we have developed rational methods to enable the construction of bioactive small molecule ligands that target a variety of toxic RNAs that are present in several untreatable neurological and orphan diseases.  They include: Kennedy’s disease, the Spinocerebellar ataxias, Huntington’s disease, Fragile X syndrome and a host of others.  In many of these cases, not only do we use directly the information contained in the RNA motif-ligand database but also we utilized similarity searching and virtual screening to rationally optimize the initial leads into potently bioactive small molecules (Fig. 6)

Fig 5: The rational and modular design of small molecules targeting the r(CCUG) expansion that causes DM2. A similar approach has been used to target r(CUG) expansions in DM1 and r(CAG) expansions present in a variety of diseases. Designer small molecules are much higher affinity and are much more specific binders to RNA than proteins.

Fig 5:  The rational and modular design of small molecules targeting the r(CCUG) expansion that causes DM2.  A similar approach has been used to target r(CUG) expansions in DM1 and r(CAG) expansions present in a variety of diseases.  Designer small molecules are much higher affinity and are much more specific binders to RNA than proteins.  

Fig 6: By synergistically merging the information contained in the RNA motif-ligand database, bioactive small molecules can be optimized by similarity searching.

Fig 6:  By synergistically merging the information contained in the RNA motif-ligand database, bioactive small molecules can be optimized by similarity searching. 

Structural Biology of RNA.  Our group is heavily involved in the rational design of small molecules that target RNA and in solving the structures of these small molecules in complex with their RNA targets.  These studies will not only allow us to understand the molecular and atomic level interactions that drive association of complexes but will also allow us to rationally design improved small molecules that target RNA.  Previously, we have solved a variety of RNA structures that are involved in Myotonic Dystrophy Type 1 and Fragile X-associated tremor ataxia and Fragile X syndrome (Fig. 7).  One interesting feature that we have shown is that in many of these RNA-mediated diseases, the RNA structures in the absence of bound ligand can sample multiple conformations.  The dynamic nature of the RNA target can provide opportunities for small molecules to recognize a variety of states in the RNA, i.e. induced fit or conformational selection.  The results of these studies have been reported in:  Kumar A, Park H, Fang P, Parkesh R, Guo M, Nettles K, Disney MD.  Crystal Structure of the Triplet Repeat in Myotonic Dystrophy Reveals Heterogeneous 1x1 Nucleotide UU Internal Loop Conformations.  Biochemistry (2011), 50, 9928-9935; Kumar A, Fang P, Park H, Guo M, Nettles K, Disney MD.  Crystal Structure of the Repeating CGG Motif Found in the RNA that Causes Fragile X Syndrome.  ChemBioChem (2011), 12, 2140-2142; Parkesh R, Fountain M, Disney MD. NMR Spectroscopy and Molecular Dynamics Simulation of r(CCGCUGCGG)2 Reveal a Dynamic UU Internal Loop Found in Myotonic Dystrophy Type 1. Biochemistry, (2011), 50, 599-601.

Fig 7: Two crystal structures that have been refined of a model duplex of r(CUG)exp in DM1.

Fig 7:  Two crystal structures that have been refined of a model duplex of r(CUG)exp in DM1. 

Development of novel antibacterials that evade known modes of resistance.  Antibiotic resistance is a threat to human health and the development of novel antibacterials that evade known modes of resistance are critical to remain one step ahead of resistant bacteria.  The Disney group has developed two tools that can be used to deal with this problem.  By using microarray technologies, our group has developed sensitive platforms to enable the rapid screening of new antibiotics for being substrates of resistance enzymes.  By scanning a variety of chemical spaces around a known antibacterial scaffold using this technology, we can gain fast access to important information regarding how the chemical structure of an antibiotic affects modification by resistance-causing enzymes.  In a second synergistic approach, the Disney group has developed chemoenzymatic routes to quickly construct modified antibiotics that are traditionally rather laborious to synthesize.  By merging these two approaches, one can gain rapid access to diversified antibacterial structures while at the same time quickly probe these structures for evading known modes of resistance (Fig. 8).   These studies have been disclosed in the following publications: Disney MD and Barrett OJ.  An aminoglycoside microarray platform for directly monitoring and studying resistance.  Biochemistry (2007), 46, 11223-11230; Barrett OJ, Pushechnikov A, Wu M, and Disney MD.  Studying aminoglycoside modification by the acetyltransferase class of resistance-causing enzymes via microarray. Carbohydrate Research, (2008), 343, 2924-2931; Tsitovich PB, Pushechnikov A, French JM, Disney MD. A chemoenzymatic route to diversify aminoglycosides enables a microarray-based method to probe acetyltransferase activity.  ChemBioChem (2010), 11, 1656-1660.  Inside cover article.

Fig. 8: Schematic of a facile microarray-based method to probe antibiotic resistance.

Fig. 8:  Schematic of a facile microarray-based method to probe antibiotic resistance.

___________

Home
___________

Projects
___________

Lab Members
___________

Publications
___________

Press Release
___________

Funding
___________

Positions
___________

Software
___________

Philanthropy
___________

Contacts
___________