2009 Proposal--career: Single-molecule Analysis Of Genomic Dna And Chromatin In Eukaryotic Transcription

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Summary, Page 1 of 1

The long-term mission of our laboratory is to enable and make important discoveries in molecular cell biology by innovating new biophysical methods and culturing interdisciplinary research and education partnerships. The specific research goal of this 5-year proposal is to develop new methods for sitespecific single-molecule analysis of native DNA and chromatin extracted from yeast cells. Our long-term mission is also supported by our specific educational and broader impacts goals in this proposal, which will multiply the impact of our research goals. These include participation in Open Science, integration of research and university education, and community and local school outreach. Together, the research and education goals in this proposal will establish a successful career path for this PI as leader of an exciting biophysics research laboratory collaborating with leading chromatin and transcription biologists, successful university educator, and recruiter of underrepresented minorities to research careers. Intellectual Merit In order to understand the dynamics of chromatin remodeling during gene transcription, new biophysical tools are needed for single-molecule analysis of native chromatin. Therefore, we are developing a unique single-molecule method (see Fig. 1) for mapping proteins on native chromatin that will provide this capability. Our specific hypothesis is that optical tweezers unzipping of native chromatin will allow mapping of histones and polymerases with near base pair resolution on the same individual fiber. The specific aims have been designed to achieve this goal of native chromatin mapping by unzipping. Each aim can produce high-impact results and develop new useful biophysical techniques, independent of the success of the other aims:

nucleosome

Optical Trap

RNA Pol II ssDNA Coverglass

Figure 1 Proposed unzipping of single chromatin fibers with optical tweezers. Optical tweezers use laser light focused through a microscope objective to apply and measure small forces on microspheres attached to biomolecules. Monitoring the length of ssDNA and the unzipping forces will reveal the position of nucleosomes and polymerases with close to base pair resolution.

1. Develop Shotgun DNA mapping (SDM) as a new single-molecule method for identifying unknown DNA fragments. Shotgun DNA mapping is the ability to identify the genomic location of a random DNA fragment based on its naked DNA unzipping forces compared with simulated unzipping forces of a published genome. 2. Determine the unzipping signature for RNA Polymerase II (Pol II) transcription complexes. 3. Map native yeast chromatin by single-molecule unzipping. Broader Impacts 1. Open Science. The entire CAREER project will be conducted as Open Science. This will include Open Notebook Science, Open Access publishing, free availability of raw data and all stages of processed data, and sharing of all data acquisition and processing software. Furthermore, the proposal will be published at the time of its submission. 2. Integration of research with undergraduate and graduate education. Virtual reality physics demos that mirror our “real life” demos will be developed for “Second Life.” I will continue transforming Junior Lab course into an open science course and develop new biophysics experimental modules. 3. Community and school outreach in Albuquerque. PI participation in outreach and development of new partnerships with middle school teachers as well as encouragement of strong outreach participation by graduate and undergraduate researchers in the lab.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 1 of 15

I. Specific Aims (2)

DNA in eukaryotic cells exists as chromatin, of which the fundamental unit is the nucleosome . Nucleosomes consist of an octamer of histone proteins which control access of other proteins to DNA. Therefore, nucleosomes are barriers in processes such as transcription, replication, and DNA repair. In order to modulate this barrier, nucleosomes can be remodeled (moved or removed) or modified by a variety of post-translational modification (PTMs). During transcription, both remodeling and PTMs (3) regulate RNA Polymerase II (Pol II) . The key role in transcription combined with the fact that many histone PTMs are heritable to the next cell generation makes understanding chromatin remodeling during (4) transcription very important to cancer biology . However, Optical understanding these processes during gene transcription is Trap nucleosome currently limited by the inability to sensitively characterize RNA Pol II with high spatial resolution the positions of polymerases and nucleosomes on individual chromatin fibers in cells. ssDNA Therefore, we are developing a unique single-molecule method (see Fig. 1) for mapping proteins on native chromatin that will Coverglass provide this capability. Our specific hypothesis is that optical Figure 1 Proposed unzipping of tweezers unzipping of native chromatin will allow mapping single chromatin fibers with optical of histones and polymerases with near base pair resolution tweezers. Optical tweezers use laser on the same individual fiber. We have several reasons to light focused through a microscope objective to apply and measure small believe we can prove our hypothesis and shed light on these forces on microspheres attached to questions. First, The PI of this proposal is the co-inventor of the biomolecules. Monitoring the length technique for mapping protein binding by unzipping single DNA of ssDNA and the unzipping forces will molecules and an expert in constructing tools for manipulating reveal the position of nucleosomes (5-8) and polymerases with close to base single DNA molecules . It has been shown that bound pair resolution. proteins can be detected because the force required to unzip protein-bound DNA is substantially higher than for naked DNA. Second, It has recently been shown that this DNA unzipping method can map the positions of in vitro (9, 10) assembled mononucleosomes with close to base pair resolution . Third, we are collaborating with labs with extensive expertise in characterizing chromatin remodeling and Pol II elongation with ensemble (11-14) methods such as Chromatin Immunoprecipitation (ChIP) . The specific aims for this CAREER proposal have been designed to achieve our goal of native chromatin mapping by unzipping. Each aim can produce high-impact results and develop new useful biophysical techniques, independent of the success of the other aims: 1. Develop Shotgun DNA mapping (SDM) as a new single-molecule method for identifying unknown DNA fragments. Shotgun DNA mapping is the ability to identify the genomic location of a random DNA fragment based on its naked DNA unzipping forces compared with simulated unzipping forces of a (15) published genome. We have preliminary indication that it will work well in the yeast genome and it is an enabler of our goal of native chromatin mapping. In this aim, we will: 1.1 Validate SDM for identifying yeast genomic DNA fragments. We have created a set of unzipping constructs from a library of S. cerevesiae shotgun clones. We will use SDM to identify clones. SDM success rate will be characterized by subsequent DNA sequencing of the clones. 1.2 Improve DNA unzipping simulations and SDM matching algorithms. To work with larger sets of possible fragments (larger genome than yeast or more frequent cutter than XhoI), the sensitivity and specificity need to be increased. We will do so by increasing the sophistication and quality of the simulations and matching algorithms.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 2 of 15

1.3 Automate SDM data acquisition and analysis. SDM automation will be required for genomewide analyses and will greatly increase the practicality of the method. The goal of this aim is to fully eliminate the need for user input, once a micro-flow cell has been manually mounted to the stage. 2. Determine the unzipping signature for RNA Polymerase II (Pol II) transcription complexes. In this aim, we will determine the in vitro Pol II unzipping signature for use in identifying native complexes in Aim 3. Furthermore, new structural information about in vitro transcription complexes will be obtained. 2.1 Unzip stalled Pol II in vitro transcription complexes. Stalled complexes provided by the Adelman laboratory will be unzipped from both upstream and downstream directions. Individual traces will be aligned and two reference ―signatures‖ will be produced: one for each sense or antisense orientation of transcription. 2.2 Atomistically model unzipping of protein-DNA complexes. Molecular dynamics simulations of the unzipping of naked DNA and a model protein-DNA system will be carried out. Results will guide structural interpretation of Pol II data from aim 2.1 and will be used to initiate a separate project with the long-term goal of full atomistic simulations of nucleosome and Pol II unzipping. 3. Map native yeast chromatin by single-molecule unzipping. The ultimate goal of this proposal is high-resolution, single-molecule, site-specific mapping of nucleosome and polymerase positions on native chromatin. Various independent approaches will be taken towards native chromatin unzipping in this aim and significant progress towards the overall goal will be achieved, independent of success of the other aims. (16)

3.1 Unzipping of yeast plasmid chromatin. Yeast plasmid chromatin will be isolated with and (17, 18) without induction of the PHO5 gene encoded in the plasmid and ligated to unzipping constructs. Single-molecule unzipping will be used to determine the locations of the nucleosomes and compared with well-known locations from ensemble measurements. 3.2 Unzipping of native yeast chromatin from a single genomic site. Genomic chromatin containing the YLR454 gene will be digested with I-SceI and ligated onto unzipping constructs. Nucleosome and polymerase positions will be measured by single-molecule unzipping. Chromatin structure will be mapped in wild type, an spt16 mutant, a mutant deficient in histone H2B (12) ubiquitylation, and the double mutant . Correlated nucleosome and polymerase occupancy on individual chromatin fibers will be measured. 3.3 Shotgun chromatin mapping. Yeast native chromatin will be digested with XhoI and random fragments will be ligated onto unzipping constructs. Shotgun DNA mapping will be used to identify individual fragments. Nucleosome and polymerase positions will be identified at high resolution on each individual fiber. High throughput experiments will reveal genome-wide correlations between polymerases and histones and also identify divergent and cryptically initiated polymerases in wild (12) type, spt16 mutants, H2B-ubiquitylation mutants, and double mutants . New biophysical tools are needed for single-molecule analysis of native chromatin. Success in this project will develop such tools and will answer fundamental open questions in eukaryotic transcription. Furthermore, we will have developed a unique and powerful single-molecule genetics technique that will enable many future high-impact research avenues. These may include studying single-molecule mapping of epigenetic chromatin marks, DNA damage repair and replication, structural genome mapping by unzipping, single-molecule telomere mapping, and studies of alternative splicing by single-molecule cDNA unzipping. Given this wealth of possibilities and the strong collaborations formed, this CAREER project will lay a solid foundation for a successful research career.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 3 of 15

II. Background and significance A. Chromatin remodeling during transcription elongation During transcription, nucleosomes are removed from in front of the polymerase to prevent stalling of (19) elongation . Nucleosomes are then reassembled behind the polymerase, which is important for (20, 21) preventing improper transcription initiation within the gene (termed cryptic initiation, see Fig. 2) . This remodeling process is coordinated by histone chaperones that assist in chromatin reassembly. During transcription, lysine residues on the histone tails are marked with various post-translational modifications (22, 23) (PTMs), but their functions in transcription are not yet well understood . In order to understand these processes, we need to know with basepair precision the position of all polymerases and Reassembled nucleosomes on the gene. Furthermore, we promoter Nucleosomes need to know whether the nucleosomes are Pol completely assembled octamers or are II cryptic tetrameric or hexameric. Finally, we need to Transcription promoter know this information for polymerases and Figure 2 Nucleosome remodeling during transcription. Nucleosomes are nucleosomes on the same individual gene— evicted in front of the polymerase and reassembled behind the polymerase. Proper reassembly prevents initiation from cryptic promoters. so that their direct interactions can be seen. The main tool for studying remodeling during transcription is chromatin immunoprecipitation (ChIP). ChIP is a powerful method that can provide some of the needed information such as relative levels of polymerase and nucleosome occupancy, as well as information about PTMs and nucleosome structure (tetramer v. octamer). However, the spatial resolution of ChIP is limited to about 100 base pairs and it is unable to provide information about single chromatin fibers. Thus, we are working to provide a much needed single-molecule analysis method based on unzipping of chromatin fibers with optical tweezers. Mary Ann Osley’s lab has recently been studying the interaction of a particular histone chaperone, (24-27) (12) FACT , and a specific PTM, monoubiquitylation of histone H2B (H2B-ub) . Both FACT and H2B-ub (28) are important to cancer biology . The Osley lab has discovered that when both FACT and H2B-ub are not functioning, chromatin is improperly reassembled on active genes. The molecular nature of this misassembled chromatin is very difficult to decipher with ChIP or other ensemble assays. Answering questions about this process is a specific motivation for our single-molecule technique development, but we expect to be able to study chromatin structure during many important cellular processes and in any eukaryotic organism. B. Antisense Transcription The misassembled chromatin mentioned in the previous section can lead to improper Pol II transcription initiation from promoters within genes or elsewhere that would normally be blocked by nucleosomes. (29, 30) This is called cryptic initiation and it is rapidly becoming an important area of study in transcription. A particularly fascinating, and possibly crucial point of cryptic initiation is that polymerases may initiate in the antisense direction, and this has indeed very recently been shown to be widely (20) prevalent . Buratowski says in his perspective, ―The mystery is why RNA polymerase II traveling in one direction can produce RNAs thousands of nucleotides long, whereas polymerases moving in the opposite (29) direction don’t get very far.‖ One possible explanation is that antisense-oriented polymerases quickly encounter sense-oriented polymerases, and these head-to-head collisions effectively stall both complexes. The ability to detect sense orientation of Pol II complexes is difficult with ensemble (31) techniques. Due to the asymmetry of the Pol II-DNA complex (in contrast to the nucleosome) , it is

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 4 of 15

possible that single-molecule unzipping will be able to clearly indicate the transcriptional orientation of the Pol II complexes on individual chromatin fibers. This may open the door for answering critical open questions in Pol II transcription, such as: are head-to-head collisions of senseand antisense-oriented polymerases common in vivo? Is this an important component of gene regulation or misregulation? C. Mapping protein binding by single-molecule DNA unzipping In my dissertation work, I showed that positions of DNA-binding proteins could be mapped with high (5) resolution by unzipping single DNA molecules with optical tweezers . Fig. 1 shows conceptually how these experiments are carried out with optical tweezers—though instead of a chromatin fiber, I unzipped DNA plus site-specific binding proteins such as EcoRI. To Data from Shundrovsky, et al., Nature Structural enable these experiments, I designed and implemented a & Molecular Biology 13, versatile unzipping anchoring construct that allows for p. 549 (Not Koch Data) unzipping of any DNA molecule with a known 5’ or 3’ (5) overhang (see Fig. 4) . These two achievements are the foundation for our goal of mapping native chromatin by unzipping. A former labmate has used my versatile unzipping construct to unzip through reconstituted mononucleosomes and tetrasomes in the Wang lab. Some of this data is shown ©2006 Nature Publishing Group in Fig. 3. The authors clearly demonstrated that mononucleosome positions can be mapped with about 3 basepair precision. Furthermore, they were able to easily Figure 3 The protein mapping method developed by Koch has been shown to distinguish between octameric and tetrameric (9, 10) be capable of measuring the positions of nucleosomes . These results give us confidence that we nucleosomes with 3 bp precision. It was will be able to map nucleosome positions on native also shown possible to clearly distinguish chromatin. Because no such data exist for Pol II unzipping, between octames and tetramers. one of the specific aims of this proposal is to perform single-molecule unzipping of stalled RNA Polymerase II complexes. Success will provide strong evidence for the ability to map and identify polymerases and nucleosomes on individual native chromatin fibers.

5’

Dig

Nick

Biotin 5’

3’

Sticky end for ligation

Figure 4 The versatile unzipping construct from Koch et al. 2002 will be used in both aims. Any sticky end desired is easily created via choice of oligonucleotide sequence.

D. Single-molecule transcription and chromatin studies The power of single-molecule analysis for studying transcription and chromatin structure has been one of (9, 10, 32-47) the famous successes of single-molecule biophysics . Most of these studies have used purified proteins to reconstitute chromatin or in vitro transcription. As shown in Figure 5, these experiments reside in the ―single-molecule biochemistry‖ quadrant on the bottom left, as is true for single-molecule (48-54) manipulation experiments in general . Single-molecule biochemical experiments strongly complement the vast number of ensemble experiments in the two upper quadrants and have revealed exquisite details about the structures and biochemical processes involved in transcription and chromatin. By comparison, very few single-molecule experiments have been performed in the bottom-right quadrant. I have termed this quadrant ―single-molecule genetics,‖ because it presents the opportunity to leverage the many tools in ensemble genetics in combination with the power of single-molecule analyses. Of the (36, 44-47, few existing ―single-molecule genetics‖ studies, many have been chromatin or transcription studies 55) . Those studies have demonstrated the value of this untapped quadrant of experiments, but as of yet, no methods have been developed for site-specific, single-molecule, native chromatin analysis as we are

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 5 of 15

proposing. The wealth of potential discoveries in “single-molecule genetics” remains for the most part untapped, and is the main focus of our CAREER proposal. in vitro systems (Reconstituted from components)

in vivo systems (Intact, native molecules)

Ensemble Biochemistry N=trillions & in vitro systems

Ensemble Genetics N= trillions & in vivo systems

Example: in vitro transcription, analyzing RNA transcripts on gel

Example: Chromatin Immunoprecipitation (ChIP) analysis of polymerase and nucleosome occupancy

Single-molecule Biochemistry N=1 & in vitro systems

“Single-molecule Genetics” N=1 & in vivo systems

Example: single-molecule measurement of force versus velocity for RNA polymerase

Example: Our proposal for singlemolecule mapping of native chromatin fibers

Nmolecules=trillions Ensemble assays

Nmolecules =1 Single-molecule assays

The majority of the excellent single-molecule research is going on in this quadrant

We feel our niche in the singlemolecule research world will be in leading the way in this quadrant

Figure 5 Two by two matrix illustrating four classes of experiments in transcription biology. “Single-molecule genetics” is our terminology for the relatively untapped quadrant that uses single-molecule analysis on native molecules or cells.

III. Preliminary Studies A. Optical tweezers instrumentation and data analysis software Design and construction of our optical tweezers was completed in July of 2009, with the last key element being a high quality 4 Watt TEM00 diode-pumped solid state laser (Coherent) that was generously provided to us by Dr. Evan Evans. We had previously been attempting construction using a high-power visible (690 nm) laser diode system, due to the added safety and potential superiority of a 690 nm optical trap. However, we had to put this on hold due to the inability to acquire a diode with sufficiently high power and high beam quality. Other key elements of the system are shown in Figure 6. There are two critical software components of our optical tweezers system. The first is the optical tweezers feedback (5, 56) control software , which we have succeeded in upgrading to the newer DAQmx version of LabVIEW. This software provides the necessary feedback modules (force clamp; velocity clamp; loading rate clamp) for DNA unzipping and protein-DNA unzipping. The other critical software component is a suite of automated data analysis LabVIEW programs. These programs convert the raw OT data into ―force v. unzipping index,‖ and are further automated to identify locations and disruption forces of DNA binding (6) proteins . The suite has been upgraded to LabVIEW 8.5. The primary coders of the original software applications were the PI (Koch) and his colleague Dr. Richard Yeh during their Ph.D work. As of the time of this writing (July 2009), we have stretched single double-stranded DNA molecules with our system but have not yet unzipped DNA. We have preliminary calibration of the optical trap, showing a trap stiffness factor of 0.2 pN / nm per Watt of input laser power. We can input at least 2 W of laser power without damaging the objective, giving a stiffness of 0.4 pN / nm. We cannot directly verify this yet,

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 6 of 15

due to bandwidth limitations of the detector. The linear extent of the trap is about 100 nm, giving us a maximum force of at least 40 pN – well over the necessary 17 pN maximum unzipping force. We are currently finishing calibration and constructing an enclosure to minimize drift and noise.

7

2 1

3

B. Shotgun DNA Mapping We have made substantial progress towards the capability we are calling shotgun DNA mapping (SDM). SDM is the ability to identify the genomic location of a random DNA fragment based on its naked DNA unzipping forces compared with simulated unzipping forces of a published genome. SDM will be an important enabler of our ability to map native chromatin by unzipping without the need for site-specific genetic engineering. We call this proposed method shotgun chromatin mapping (SCM) and the method is outlined in Fig. 7. Native chromatin will be digested with restriction endonucleases and random fragments will be unzipped, providing high resolution positional information for nucleosomes and polymerases (see Aim 3). The high resolution information will then be assigned to an exact location in the genome based on the underlying SDM method. A very appealing aspect of this process is that it does not require genetic engineering or site-specific chromatin extraction.

4

6

5

Figure 6 KochLab optical tweezers setup, July 2009. (1) Coherent 4W diode-pumped solid state laser (2) Power modulater (BEOC) (3) Thorlabs Variable Beam Expander (4) Steering optics (5) Olympus IX-71 Microscope body (6) Piezo stage (Mad City Labs)(7) Quadrant photodiode (PhreshPhotonics)

SDM works because we can accurately predict the unzipping Figure 7. Overview of proposed method for shotgun DNA and chromatin forces for any known sequence of (57) mapping. We have recently achieved proof-of-principle results important DNA . Fig. 8 shows one for the “Global Genome Location,” part of the process (lower right). experimental unzipping trace (Koch (5) data , 2002) compared with two simulations: one from the correct sequence, and one from an incorrect sequence. The correct match is easily picked out by eye. We have developed an algorithm that quantifies the match between and experimental and simulated unzipping curve. The match score we use exp

is , where N is the number of points in matching window, F is the sim experimental unzipping force, F is the predicted force, FG=17.56 pN is the unzipping required for poly(dG), and FA = 10.23 pN for poly(dA). A perfect match gives m = 0, and the maximum possible mismatch is m = 1.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Correct Match, Score 0.2 18

Force (pN)

Force (pN)

Mismatch, Score 0.8 18

Simulated data Simulation OT Optical Data tweezers Data (Koch 2002)

12 0

B

1500

Unzipping fork index (bp)

12 0

data OT Simulated Data Optical tweezers Simulation Data (Koch 2002)

1500

Unzipping fork index (bp)

Figure 8 Experimental unzipping data compared with (A) matching and (B) non-matching simulation. The green window indicates the region from j=1200 to j=1700 where the match scores were computed. A lower match score indicates the better match.

Match Score

A

Project Description, Page 7 of 15

Match

File Number (Arb.) Figure 19 Compilation of match scores for a single experimental data set. The file number is an arbitrary, arising from the order in which the library simulations were loaded. A perfect match would have a score of zero, and the correct match can be seen as having the lowest score, very distinguishable from the incorrect matches.

We tested the algorithm using existing experimental unzipping data (5) for pBR322 plasmid . Our algorithm easily identified the pBR322 fragment out of a background of simulations of all ~2700 possible yeast XhoI digested fragments. Fig. 9 shows this result, with the correct match clearly standing out from the incorrect matches. Furthermore, we repeated this success for approximately 32 separate pBR322 data sets with zero failures. We are currently preparing a manuscript describing our proof-of-principle (15) results for SDM. These preliminary results give us confidence that we will be successful in performing SDM with yeast genomic DNA as outlined in Specific Aim 1. C. Molecular Constructs and E. coli, Yeast strains Shotgun Clones Specific Aim 1.1 is to demonstrate success of Shotgun DNA mapping in #4 Clone #1 #1 yeast genomic DNA. In order to do #2 #5 #3 #4 #5 so, we needed to first create clones pBlue that can be sequenced to validate the SDM results. We have “mix” successfully prepared a dozen shotgun clones of yeast genomic DNA fragments propagated in a SapI-digested NotI-digested pBluescript plasmid in E. coli. The shotgun clones shotgun clones clones were created by digesting a Dig-labeled “anchor” (Incomplete yeast genomic DNA prep and a segment for unzipping digestion lanes 2 & 6) construct (see figure 4) pBluescript prep with either XhoI, or a double digest of XhoI-EcoRI, Figure 10 – Agarose gel analysis of some of the successful shotgun clones. followed by ligation. Blue/white Lane 1 is 10 kb ladder. Lanes 2-6 are SapI digested plasmid miniprep of selection was used to pick colonies clones 1, 2, 4, 5 and pBluescript alone. (Lanes 2 and 6 did not digest completely.) Lanes 7-11 are NotI digestions of pBluescript, clones 1, 3, 4, 5. with incorporated yeast genomic Lane 13 & 14 are SapI and NotI digestion (respectively) of the “mixture” of DNA. Single colonies were used for colonies described in the text. Lane 16 is the PCR product for the dig creation of glycerol stocks and labeled anchor uses in the unzipping construct. miniprep DNA. Figure 10 shows digested miniprep DNA. The variable sizes indicates different fragments of yeast genomic DNA. Additionally, lanes 13-14 show digestion of miniprep DNA from a mixture of colonies. This DNA was created by combining 100’s of individual colonies into one growth media, and will be used as a further test case of the SDM method.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 8 of 15

Digested miniprep DNA has been ligated onto the versatile unzipping construct (Figure 4) and we have (5, 58) confirmed the ability to form single-molecule tethers from this DNA . We are thus now ready to proceed with Specific Aim 1 by unzipping the unknown fragments and testing the capabilities of SDM in identifying them. Yeast plasmid chromatin As described above in the Background and Significance, prior work has demonstrated that the position of in vitro reconstituted mononucleosomes can be precisely measure by unzipping with optical tweezers. We feel this is strong evidence that nucleosome positions on native chromatin can be measured by unzipping. As part of Specific Aim 3, we will seek to prove this hypothesis. In order to do so, we need a system with very well characterized in vivo nucleosome positions. The PHO5 gene promoter has 4 very (17, 18) well positioned nucleosomes that are thoroughly evicted upon gene induction . Furthermore, Korber et al. have shown that when the gene is placed in a small yeast plasmid, the chromatin structure is indistinguishable from that on the genomic PHO5 promoter. Therefore, we have set out to use the PHO5 plasmid system as an initial test of our native nucleosome mapping capabilities.

A

I-SceI site inserted here for our pDRP100 construct

B Plasmid size markers* Lanes 2-5

Wild-type (no signal) Lanes 6-9 pDRP100 chromatin digests Lanes 10-13 nicked linearized

supercoiled

* Mistakenly used plasmid before removal of pCU19 insert Figure 11 – (A) Diagram of pTAP5C from Korber et al. (2004). We inserted an I-SceI recognition in the DNAse hyper-sensitive region next to the EcoRI site, naming the plasmid pDRP100. The filled gray circles indicate the 4 nucleosomes evicted upon induction of the PHO5 gene. (B) Southern blot analysis of DNA run on an agarose gel after restriction endonuclease digestion of yeast nuclei. Probe hybridizes to Trp region of pDRP100. Lanes 2-5 are digested plasmid miniprep DNA (unfortunately we mistakenly used the plasmid before removing the pUC19 region used for propagation in E. coli). Lanes 6 -9 and 10-13 are for digested yeast chromatin. Absence of signal in lanes 6-9 (Wild-type W303A) and presence in 10-13 (pDRP100) demonstrates successful propagation of pDRP100 plasmid . Lane 10: I-SceI digest; Lane 11: I-SceI/BamHI; Lane 12:I-SceI/ClaI; Lane 13: ClaI. All lanes 10-13 show innefective digestion, with perhaps only ClaI showing increased linear DNA (see text).

We have engineered a unique I-SceI site near the EcoRI hypersensitive site in the pTAP5C PHO5 (17) plasmid (kindly provided by Korber et al., see figure 11A). This plasmid was then transformed into yeast, a strain which we are calling pDRP100. Propagation of the plasmid was confirmed by Southern blot, using a probe against the Trp marker in the plasmid. We have begun to assess the ability to digest the native chromatin and have preliminary indications of limited success. Yeast spheroplasts were generated by digesting the cell walls with lyticase. Nuclei were purified via centrifugation and resuspended in one of four restriction digest solutions: I-SceI; I-SceI+BamHI; I-SceI + ClaI; or ClaI. Figure 11B shows a Southern blot analysis of DNA purified after nuclei digestion. Clearly demonstrated

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 9 of 15

is the presence of the plasmid in the pDRP100 nuclei (no probe hybridizes in the wild type lanes). None of the enzymes digested the plasmid chromatin effectively. However, limited cutting may be seen by ClaI from the increase in linearized band in last visibile lane. This could be due to sub-optimal endonuclease digestion conditions or possibly due to altered PHO5 chromatin structure (and thus shifted hypersensitive sites) due to our insertion of the I-SceI site. This will be explored further as part of Specific Aim 3.

IV. Research Methods The specific aims for this CAREER proposal have been designed to achieve our goal of native chromatin mapping by single-molecule unzipping. The first specific aim is to develop “shotgun DNA mapping,” (SDM) which is our name for the ability to identify the genomic location of a random DNA fragment based on its naked DNA unzipping forces compared with simulated unzipping forces of a (15) published genome . SDM will allow us to perform chromatin mapping of yeast and other organisms without the need for site-specific genetic engineering. The second aim is to obtain the characteristic force pattern for unzipping through RNA Pol II complexes. There already exists excellent data on the (9, 10) unzipping force pattern of reconstituted mononucleosomes , but not for Pol II. Obtaining this pattern for in vitro Pol II complexes will be essential for identifying in vivo Pol II complexes by unzipping. The third aim is to map native yeast chromatin by single-molecule unzipping. We will leverage successes in the first two aims, to increase the value of success in the third aim. Specific Aim 1: Develop shotgun DNA mapping (SDM). rd

Lead graduate student: Larry Herskowitz (3 year Ph.D.) followed by future Ph.D. student. Sub Aim 1.1: Use a shotgun cloning strategy to prove that SDM works for yeast genomic DNA. So far, we have demonstrated with existing pBR322 data that the pBR322 fragment can easily be (15) identified from a background of the approximately 2700 yeast XhoI sites . We believe this is convincing evidence that shotgun DNA mapping will be successful when applied to yeast genomic DNA, but the possibility remains that there is something special about pBR322, or that there will be many difficult yeast genomic sequences. Thus, in this sub aim, we will prove that SDM works for XhoI-digested yeast genomic DNA. 1. Restriction digestion of yeast genomic DNA 2.

Clone random fragments into plasmid pBR322

As shown in Figure 12, we will validate the shotgun 3. Prepare unzipping constructs from miniprep DNA 4. Predict identity of clone based on shotgun DNA DNA mapping method in yeast via a shotgun cloning mapping by unzipping method. Steps 1, 2, and 3 have been completed as 5. Sequence clones, compare results, and determine discussed in Preliminary Studies. In step 4, we will success rate follow our standard published tethering and unzipping (5, 6, 59) protocol . Briefly, microspheres will be tethered Figure 12 Process for validating shotgun DNA mapping to glass surfaces via dig-antidig and biotin-streptavidin method for yeast genomic DNA linkages. Dozens of tethers will be selected by eye and unzipped automatically by feedback-controlled optical tweezers. We will use the existing upgraded DNA unzipping data acquisition and analysis (5, 6, 56) software that was used for our pBR322 proof-of-principle . We will use the shotgun DNA mapping (15) software application we have developed with possible improvements from sub-aim 1.2. Finally, in step 5, we will unblind our experiments by revealing the identity of the chosen clones via standard DNA sequencing. Overall, this method will be low-throughput, and we anticipate analysis of about 10 individual clones. The success rate we achieve (we are expecting errors only in the case of obviously ambiguous sequences) will predict the success rate of shotgun DNA mapping straight from genomic DNA or chromatin—in which case the cloning steps will be eliminated.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 10 of 15

Sub Aim 1.2: Improve the algorithm for shotgun DNA mapping. The purpose of this sub-aim is to increase the sensitivity and specificity of the shotgun DNA mapping algorithm we have developed. Improving sensitivity and specificity will allow us to work with larger reference libraries, e.g. from human genome fragments. We have three clear paths for achieving this: (a) improving the DNA modeling, (b) developing a more sophisticated and better matching algorithm, and (c) significantly improving optical tweezers data quality. (57)

Modeling DNA unzipping. We are currently using a very simple model for the DNA energetics . This simple model is analogous to the short-hand calculation for DNA melting temperatures, where only the AT versus GC content of the primer is accounted for. This simplification usually works remarkably well for DNA melting temperatures and it also works quite well for predicting DNA unzipping forces. However, DNA base-stacking energy is important and by accounting for it we can improve the modeling of the unzipping forces. Accounting for base-stacking energy is simple in principle as it only requires knowing (60) the energy values for the 10 nearest-neighbor interactions . The challenge we will address is optimizing the energy values from SantaLucia for our case of DNA melting by unzipping, which differs slightly from thermal melting. Our current model is a quasi-equilibrium thermodynamic model that does not utilize any of the kinetics information that is readily available in the data. We will work to implement a sequence-dependent kinetic (61) model of DNA unzipping. One such model is the Peyrard-Bishhop-Dauxois (PBD) model , a one(62) dimensional model of DNA that has been shown to successfully model DNA bubble formation kinetics (63) and DNA unzipping . We have met with the lead PI of the latter reference, Dr. Rasmussen of Los Alamos National Lab, and have obtained their Monte Carlo PBD unzipping code. Our first step will be to see if the PBD code can produce averaged force predictions that are comparable to the thermodynamical model. Next, we will look at kinetics predictions from the PBD model and develop data analysis methods that can extract kinetics information from the data. We will do this in collaboration with Dr. Evan Evans, a leader in the development of single-molecule bond kinetics analyses. Optimizing the matching algorithm. Our current matching algorithm uses a straight-forward calculation of the squared deviation of data from simulation. We believe it is likely we can improve the matching algorithm via a number of independent avenues. The first avenue is to account for slight length errors in experimental data due to microsphere size variation, drift, and other causes. We will allow for stretch and (9) shift in the data as in previous nucleosome mapping work . The next avenue is to include other independent match criteria, which when combined with our existing match scoring may greatly increase our sensitivity and specificity. There are several thermodynamical criteria we will look into, such as average fluctuations in force or unzipping index, and cross-correlation techniques. Furthermore, as discussed above, we will work to model and analyze the unzipping kinetics, which could dramatically improve the matching algorithm. Optimizing data quality. Drift and other noise in single-molecule unzipping data degrade the sensitivity and specificity of the SDM process. The rate of unzipping and the mode of feedback (velocity clamp; loading rate clamp) significantly impact these factors. For example, very slow stretch rate will optimize the amount of kinetics information and allow for significant averaging of the Brownian noise. However, drift will be much more significant for slow stretching. We will systematically study these effects in order to find an optimum unzipping feedback program. Furthermore, we will develop more sophisticated feedback algorithms—for example, dynamically switching from a velocity clamp to a force clamp in order to extract DNA opening / closing kinetics from a particular unzipping fork location. Sub Aim 1.3: Automate SDM data acquisition and analysis For our initial proof of principle, and indeed many envisioned implementations, low-throughput unzipping and analysis is sufficient to obtain

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 11 of 15

results. However, SDM automation will be required for genome-wide analyses and will greatly increase the practicality of the method. Many of the components of the data acquisition and processing stream have already been successfully automated. The goal of this aim is to fully eliminate the need for user input, once a micro-flow cell has been manually mounted on the optical tweezers instrument. Currently, there are three distinct steps that require user intervention and that need to be automated. (1) The first is the selection of a tether for unzipping, which is currently done by eye. We will adapt our existing LabVIEW bead tracking software in combination with an x-y stage and stepper motors to automatically detect tethers with the correct amount of Brownian motion and initiate feedback controlled unzipping. After unzipping the tether and storing the data, the stage will move to find the next tether. The find / unzip / store data process will repeat until the entire sample area has been scanned. (2) The next user intervention step is the selection of ―good‖ data for automatic conversion from raw data to ―force‖ versus ―unzipping index‖ data. Currently ―junk‖ tethers or tethers that prematurely break are identified and filtered out manually. We will develop algorithms to objectively filter the data automatically. (3) Finally, the algorithms for identifying nucleosomes and polymerases need to be developed along with means for accumulating and classifying this data. This is the most substantial step of the three, and will require substantial development time in coordination with the other specific aims. Specific Aim 2: Determine the unzipping signature for RNA Polymerase II transcription complexes. rd

Lead graduate student: Anthony Salvagno (3 year Ph.D.) followed by future Ph.D. student. (9, 10)

The unzipping signature of an in vitro assembled mononucleosome is already known and will be used for identifying nucleosomes during native chromatin unzipping. In this aim, we will determine the in vitro Pol II unzipping signature for use in identifying native complexes in Aim 3. Furthermore, new structural information about in vitro transcription complexes may be obtained. Sub Aim 2.1: Unzip through stalled Pol II in vitro transcription complexes. The lab of Dr. Karen Adelman has extensive experience with in vitro Pol II transcription and will provide us with elongating (13, 14, 64, 65) transcription complexes stalled via nucleotide depravation . These complexes are known to be very stable and thus can be shipped to our lab without dissociation. Importantly, the transcription complexes will be prepared so as to have defined 5’-overhangs either upstream or downstream of the elongating polymerase. These two different overhangs will allow us to unzip through the Pol II structure from both directions. In contrast to the symmetric nucleosome structure, we anticipate that the asymmetric elongating Pol II structure may provide distinct patterns depending on unzipping direction…and thus we hope to be able to distinguish sense from antisense transcription complexes in future in vivo mapping experiments. (5)

The unzipping anchor designed by the PI permits unzipping of virtually any nucleic acid with a known 5’ or 3’ overhang with only simple changes required to an adaptor duplex. Thus, transcription complexes will be ligated onto unzipping constructs using virtually the same methods as in the shotgun DNA (58) mapping experiments (Aim 1.1). Because it is highly stable, we do not expect the ligation process to disturb the Pol II complex. If we do encounter trouble, we will first make the unzipping construct and then initiate transcription on the full construct. This may offer further advantages by allowing for use of streptavidin beads during the transcription initiation and elongation. Following ligation, constructs will be (5, 6, 59) tethered with our standard protocol and dozens of individual complexes will be unzipped, each generating a Pol II disruption force pattern. We will generate computer algorithms to align these patterns and to generate a ―master pattern‖ that we can use for identifying Pol II complexes in aim 3. Two reference ―signatures‖ will be produced: one for each sense or antisense orientation of transcription.

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 12 of 15

Stalled complexes will first be prepared by nucleotide depravation using a G-less cassette. Following (66) this, transcription will be stalled via lac repressor binding to an intra-gene lac operator sequence . It is likely that the in vivo Pol II unzipping signature will be modified due to presence of different elongation factors and chaperones, but to what extent, we do not yet know. In vitro experiments with the lac repressor will provide information on how much the unzipping pattern changes due to interactions with other DNA-binding proteins. Sub Aim 2.2: Atomistic modeling of unzipping DNA and protein-DNA complexes. A big risk of our goal of native chromatin mapping by unzipping is that the extracted molecules may be too heterogeneous, and the complicated unzipping signals may be too difficult to decipher. We are minimizing this risk by the systematic approach we are taking in this proposal. Given that Pol II is an enzymatic machine with a wide array of interacting factors, it’s likely to present more of an identification problem than nucleosomes. The ability to do atomistic modeling of the unzipping of arbitrarily large protein-DNA complexes would allow for in silico predictions of unzipping force patterns of Pol II, nucleosomes, including the array of possible in vivo varieties of histone post-translational modifications (67) and transcription elongation factors . However, this is a highly ambitious goal and well beyond the scope of this CAREER proposal. It is a project that we are attempting to get funded to pursue in collaboration with the lab of Dr. Chris Lorenz, an expert in computational modeling. It also ties into our other funded project, in collaboration with Dr. Susan Atlas, which seeks to develop an atomistic model of the molecular motor kinesin. The goal of this sub-aim is to take small steps that will both help us initiate the larger project and will also help us to interpret our new Pol II in vitro unzipping data. The student on this project will use our highpriority free access to supercomputers at the UNM Center for Advanced Research Computing and will be assisted by Drs. Atlas and Lorenz. The first step will be to build a simulation of naked DNA, and implement molecular dynamics (MD) techniques for applying strain to the molecules that will cause unzipping. These results will be tied in with the kinetics modeling in specific aim 1.2. Following naked DNA MD simulations, a small model protein-DNA system with an existing crystal structure, such as EcoRI-DNA will be modeled and subjected to in silico MD unzipping. Results from these simulations may lend tremendous insight into the actual unzipping force patterns seen for nucleosomes and polymerases, even though we are not able to simulate those large structures. In support of the MD simulations, and kinetics modeling a number of naked DNA and model protein-DNA complexes will be unzipped. We expect to use these experiments to produce important biophysical results enabled by the PI’s prior (5, 6) research that have not yet been investigated. Specific Aim 3: Mapping native yeast chromatin by single-molecule unzipping. Co-lead graduate students: A. Salvagno & L. Herskowitz, followed by future graduate students. In these sub-aims, all steps prior to chromatin tethering and unzipping will be carried out in the lab of Dr. Osley under the guidance of postdoctoral associate Dr. Kelly Trujillo. The ultimate goal of this proposal is high-resolution, single-molecule, site-specific mapping of nucleosome and polymerase positions on native chromatin. Various independent approaches will be taken towards native chromatin unzipping in this aim, so that significant progress towards the overall goal can be achieved, independent of success in the other aims. Sub Aim 3.1: Unzipping of yeast plasmid chromatin. We have engineered a yeast plasmid containing (17) the PHO5 gene and an I-SceI restriction site and have transfected it into S cerevisiae. We will confirm (17) that I-SceI insertion did not change the chromatin structure by following the methods of Korber et al. (see Preliminary Studies). Yeast nuclei will be prepared with and without PHO5 induction and digested

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 13 of 15

(68)

with I-SceI . Native chromatin will be extracted (without sonication or cross-linking) and ligated onto the unzipping construct (Fig. 4). Because the I-SceI site is unique, the only possible ligation will be from the plasmid. If the I-SceI site is accessible, we feel this method will succeed, despite the ―needle in a haystack‖ problem of all the background chromatin. However if the I-SceI insertion destroys the standard PHO5 chromatin structure, I-SceI may no longer be accessible. If this is the case, or if other problems (16) arise we will pursue the plasmid chromatin purification methods of Clark before endonuclease (18, 69) digestion. The recombinase methods of Kornberg for endogenous PHO5 chromatin isolation are also appealing, though we have not yet initiated communication with the Kornberg lab. (5, 6, 59)

We will tether and unzip PHO5 chromatin using our standard protocol and determine nucleosome (9, 10) locations with close to basepair precision following methods established for reconstituted chromatin . Single-molecule nuclesome occupancy on the PHO5 promoter with and without induction of the gene will be compared with the known behavior from ensemble studies in order to validate our method. Furthermore, important questions about the nucleosome remodeling mechanism during transcription initiation can be answered with single-molecule measurements of native PHO5 chromatin, as recently (18) stated by Boeger et al . Sub Aim 3.2: Unzipping of native yeast chromatin from a single genomic site. To study native chromatin remodeling during transcription elongation, we will engineer an I-SceI restriction site upstream of an endogenous yeast gene under galactose control. We will use the YLR454 gene, which is 8 kilobases long and has well-characterized polymerase and nucleosome dynamics during transcription (21, 70) elongation . Genomic chromatin will be double-digested with I-SceI and NotI and ligated onto unzipping constructs complementary to the I-SceI overhang. We will initially attempt the process without sonication or cross-linking, and if necessary to solubilize the YLR454 will apply light sonication and / or cross-linking. Nucleosome positions on single chromatin fibers will be determined with close to basepair (9, 10) precision following methods established for reconstituted chromatin . Polymerase locations on the same fibers will be determined following methods developed in Aim 2 above. Chromatin structure will first be mapped in wild type I-SceI-YLR454, and compared with ensemble (21, 70, 71) measurements . Following this, I-SceI will be engineered into existing mutant strains in the Osley (12) lab: an spt16 mutant, a mutant deficient in histone H2B ubiquitylation (K123R), and the double mutant . Correlated nucleosome and polymerase occupancy will be measured in single fibers, which will answer (12) questions about the unusual chromatin structure in these mutants seen by ensemble assays . Based on Specific Aim 2, novel features of elongating polymerase may be identified, such as anti-sense oriented polymerases from cryptic initiation. Sub Aim 3.3: Shotgun chromatin mapping. Shotgun chromatin mapping (SCM) is the use of shotgun DNA mapping (SDM; Aim 1) as a means for identifying the underlying DNA sequence of random chromatin fibers. This is the ultimate goal of the proposal, and if successful it will allow mapping of polymerases and nucleosomes in any mutant yeast strain, without the need for engineering an ISceI site. Full success depends on success in Aims 1 and 2. However, even partial success will be very valuable and will be possible independent of the other aims. (For example, non-site-specific mapping, or nucleosome-only mapping.) The first step is to create unzippable chromatin fragments from yeast genomic chromatin. Yeast nuclei (68) will be prepared as described in preliminary studies and chromatin will be digested by XhoI . XhoI (15) recognizes approximately 1350 sites in the genome . However, many of these sites will be occluded by chromatin structure—thus there is an inherent bias in this step (which can be measured by SDM of genomic DNA following digestion). Strategies for reducing this bias will be pursued in the future, though likely it will be a persistent issue with SCM. The versatile unzipping construct (Fig. 4) with XhoI sticky end

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 14 of 15

will be ligated to the XhoI-digested soluble chromatin component. Because XhoI digestion produces selfcomplementary overhangs, concatamers will form during ligation. This is not expected to be a problem, but to eliminate concatamers we can first digest the chromatin with another enzyme and treat with calfintestinal alkaline phosphatase. (5, 6, 59)

Chromatin unzipping tethers will be formed using our standard procotol . Shotgun DNA Mapping (Aim 1) will be used to exactly identify each fragment, a process we are calling shotgun chromatin mapping (SCM). Because we are using native chromatin, we will need to generate a sophisticated sequence of feedback-controlled optical tweezers unzipping steps. Our control software is ideally suited for this and reconfiguring feedback sequences is easy and does not require modifications to the LabVIEW (56) code . We will use a feedback sequence as follows: (1) find tether center, (2) constant velocity stretching until j=500 basepairs are unzipped, (3) reverse direction and constant slow-velocity ―rezipping,‖ until j = 0 (4) reverse direction again and continue unzipping until tether breaks or piezo runs out of range. Steps 2 and 4 will produce data showing protein-DNA interactions. Nucleosome positions will be (9, determined with close to basepair precision following methods established for reconstituted chromatin 10) . Polymerase locations on the same fibers will be determined following methods developed in Aim 2 above. SDM will be applied to map these high-resolution positions to an exact genome location. This will be possible because in step 4 of the feedback sequence the first 500 bases will be free of protein, and thus will provide a naked DNA unzipping curve for use in SDM. With our existing software and hardware, it is possible but tedious to unzip over a thousand individual (5, 6) tethers . This will probably be sufficient for validating SCM as a new tool for high-resolution singlemolecule chromatin mapping. However, we want to explore genome-wide correlations between polymerases and histones and look for divergent and cryptically initiated polymerases. This will require high-throughput automation of data acquisition and analysis, which is specific aim 1.3. Because SCM does not require genetic engineering of each strain studied, we will be able to quickly generate and compare results in wild-type and strains deficient in FACT and H2B-ubiquitylation, as described in Background Section A.

V. Educational Integration and Broader Impacts Our lab believes strongly in our mission of advancing molecular cell biology, and therefore our goals are well aligned with those of the NSF for integrating research with education and maximizing the broader impacts of our research. Broadly communicating our results and methods, teaching at all levels, and training future leaders in biophysics research together will greatly multiply the impact of our research discoveries. Below, we describe our plans towards this goal in two areas: open science and integration of research with undergraduate education. One way of broadening our impact is in recruitment and training of underrepresented minorities to biophysics, and this is an underlying component of our goals below. We are at an advantage in this area, because of the unique demographics of the University of New Mexico and the surrounding area. (72) UNM is designated a Hispanic-Serving Institution by the US Dept. Ed. and other government agencies. (73) We participate in the UNM PREP program that recruits minorities to science and our lab employed three Hispanic researchers so far. Additionally, the local population consists of a high proportion of minorities underrepresented in science. I have participated in community scientific outreach since early in my graduate career and it remains an enjoyable and valued activity. During my first three years at UNM, I have judged the Central New Mexico Science and Engineering Challenge (Middle School Microbiology), the Cleveland Middle School Science Fair, and given a presentation on nanomanipulation in the biological sciences to local Middle and High School science teachers at the NNIN-sponsored summer ―nanocamp.‖

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

Project Description, Page 15 of 15

I plan to grow my own outreach, and continue to encourage, value, and reward community outreach by the graduate, undergraduate, and postdoctoral members of our lab.

Open Science Since early 2007, our lab has been actively participating in OpenWetWare (OWW), an Open Science site hosted by MIT (http://www.openwetware.org/wiki/Koch_Lab). The goal of open science is complete open sharing of all data and knowledge generated in the laboratory. As our lab has matured, we have taken many steps and a few leaps towards open science. These include ―open (74) (75) (76, 77) notebook science‖ via OpenWetWare , detailed protocol publishing , and with this CAREER (78) proposal, a fully-open research proposal . If funded, we are committed to carrying out the proposed research as openly as possible. This will include Open Notebook Science, Open Access publishing, (79) free availability of raw data and all stages of processed data , and sharing of all data acquisition and processing software. While we are committed to this mission, exactly what the best methods are for doing this is a difficult question that leaders in the open science movement are still figuring out. In the past year, I have been fortunate to meet some of these leaders who have already helped our lab tremendously in its goal for openness. I have attached letters (Cameron Neylon, Jean-Claude Bradley, and Andrew Lang) expressing continued willingness to help with these difficult issues. I have also formed a collaboration with two undergraduate students at UNM, Noel Fernando and Leslie Broyles, who are majoring in digital media arts. These students have offered assistance in our open science and education endeavors that would benefit from advanced digital media expertise. A specific idea is to make a video presentation that will explain our lab’s open science activities to the general public. Our ability to succeed in our open science goals will be greatly aided by the CAREER award and the stability and assurance it will provide our lab.

Integration of research with undergraduate education. Many of the classes I have taught or will teach are carried out in a lecture hall that has a wide array of fantastic hands-on physics demos. While teaching physics 102 (a non-math based course intended for non-science majors), I have noticed an unsurprising fact: students are very eager for an opportunity to play with the demos! Unfortunately, it is not practical for students to have this opportunity. I often assign quiz questions or homework problems that directly relate to demos we use in class. For example, the ―wave table‖ is a device made up of rods tied together with wire. Beautiful waves are created and concepts such as wave interference, standing waves, frequency, are illustrated. I believe that virtual reality is a very promising method for bringing these demos into the students’ hands. Thus, I have budgeted to employ an undergraduate computer engineer to build virtual versions of our demos in the virtual reality environment ―Second Life‖ (SL). Two (80) of the experts in leveraging SL for scientific purposes have offered their advice to me in getting this project going (see letter from Jean-Claude Bradley and Andrew Lang). Another course I teach is Junior Lab, which is the first ―real‖ physics lab course that physics majors at UNM enroll in. The main innovation with this course is that I host this course completely as Open (81) Science, on OpenWetWare . The students keep all of their notes electronically on OWW, along with their lab write-ups and formal reports. Additionally, all of the written feedback from me (with the exception of letter grades) is also carried out in public on OWW. The results have been very rewarding: of the 31 students, at least 6 of them continued to use OWW subsequent to Junior Lab in their own coursework and lab work. Furthermore, I think there are significant benefits to training students in open science at this early stage in their research careers and will ultimately make a large impact on the next generation of open scientists. In addition to the open science component to Junior Lab, I intend to add new biophysics modules to the course. Specific modules I would like to add are: tethered particle motion (TPM) analysis of single DNA molecules optical tweezers / magnetic tweezers experiments, and PDMS microfluidics. I have commitment from the chair of UNM Physics Dept. to support our goals of adding biophysics experiment modules with college and department funds when needed (see Chair letter).

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

References, Page 1 of 4

References Cited

1.

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

"Acknowledgments: I thank Andy Maloney, Anthony Salvagno, Larry Herskowitz, Brigette Black, Linh Le, Diego Ramallo, Laurie Hudson, Janet Oliver, Kelly Trujillo, Chris Lorenz, Evan Evans, Karen Adelman, and Mary Ann Osley for significant help in generation of this research plan. Thank you to the American Cancer Society for project seed funding through UNM’s IRG program." Kornberg, R. D., "Chromatin structure: a repeating unit of histones and DNA." Science 184, 868 (May 24, 1974). Fuchs, S., Laribee, R., Strahl, B., "Protein modifications in transcription elongation." Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1789, 26 (2009). Jones, P. A., Baylin, S. B., "The epigenomics of cancer." Cell 128, 683 (2007). Koch, S. J., Shundrovsky, A., Jantzen, B. C., Wang, M. D., "Probing protein-DNA interactions by unzipping a single DNA double helix." Biophys J 83, 1098 (Aug, 2002). Koch, S. J., Wang, M. D., "Dynamic force spectroscopy of protein-DNA interactions by unzipping DNA." Phys Rev Lett 91, 028103 (Jul 11, 2003). Koch, S. J., Thayer, G. E., Corwin, A. D., de Boer, M. P., "Micromachined piconewton force sensor for biophysics investigations." APPLIED PHYSICS LETTERS 89, 173901 (2006). Xia, D., Gamble, T. C., Mendoza, E. A., Koch, S. J., He, X., Lopez, G. P., Brueck, S. R., "DNA transport in hierarchically-structured colloidal-nanoparticle porous-wall nanochannels." Nano Lett 8, 1610 (Jun, 2008). Shundrovsky, A., Smith, C. L., Lis, J. T., Peterson, C. L., Wang, M. D., "Probing SWI/SNF remodeling of the nucleosome by unzipping single DNA molecules." Nature Structural & Molecular Biology 13, 549 (2006). Hall, M. A., Shundrovsky, A., Bai, L., Fulbright, R. M., Lis, J. T., Wang, M. D., "High-resolution dynamic mapping of histone-DNA interactions in a nucleosome." Nat Struct Mol Biol (Jan 11, 2009). Tsukuda, T., Fleming, A. B., Nickoloff, J. A., Osley, M. A., "Chromatin remodelling at a DNA double-strand break site in Saccharomyces cerevisiae." Nature 438, 379 (2005). Fleming, A. B., Kao, C. F., Hillyer, C., Pikaart, M., Osley, M. A., "H2B ubiquitylation plays a role in nucleosome dynamics during transcription elongation." Mol Cell 31, 57 (Jul 11, 2008). Muse, G., Gilchrist, D., Nechaev, S., Shah, R., Parker, J., Grissom, S., Zeitlinger, J., Adelman, K., "RNA polymerase is poised for activation across the genome." Nature Genetics 39, 1507 (2007). Zeitlinger, J., Stark, A., Kellis, M., Hong, J. W., Nechaev, S., Adelman, K., Levine, M., Young, R. A., "RNA polymerase stalling at developmental control genes in the Drosophila melanogaster embryo." Nat Genet 39, 1512 (Dec, 2007). "Herskowitz, Lawrence, Salvagno, Anthony, Maloney, R. Andy, Le, Linh, and Koch, Steven. Proof of principle for shotgun DNA mapping by unzipping. Available from Nature Precedings (2009)." Kim, Y., Shen, C.-H., Clark, D. J., "Purification and nucleosome mapping analysis of native yeast plasmid chromatin." Methods 33, 59 (2004). Korber, P., Luckenbach, T., Blaschke, D., Hörz, W., "Evidence for histone eviction in trans upon induction of the yeast PHO5 promoter." Molecular and Cellular Biology 24, 10965 (2004). Boeger, H., Griesenbeck, J., Kornberg, R. D., "Nucleosome retention and the stochastic nature of promoter chromatin remodeling for transcription." Cell 133, 716 (May 16, 2008). Studitsky, V. M., Walter, W., Kireeva, M., Kashlev, M., Felsenfeld, G., "Chromatin remodeling by RNA polymerases." Trends Biochem Sci 29, 127 (Mar, 2004). Core, L., Waterfall, J., Lis, J., "Nascent RNA Sequencing Reveals Widespread Pausing and Divergent Initiation at Human Promoters." Science, 1162228 (2008). Mason, P. B., Struhl, K., "The FACT Complex Travels with Elongating RNA Polymerase II and Is Important for the Fidelity of Transcriptional Initiation In Vivo." Mol Cell Biol 23, 8323 (November 15, 2003, 2003).

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

35. 36. 37. 38. 39. 40. 41. 42. 43.

44.

References, Page 2 of 4

Li, B., Carey, M., Workman, J. L., "The Role of Chromatin during Transcription." Cell 128, 707 (2007). Peterson, C. L., Laniel, M. A., "Histones and histone modifications." Curr Biol 14, R546 (Jul 27, 2004). Orphanides, G., LeRoy, G., Chang, C. H., Luse, D. S., Reinberg, D., "FACT, a factor that facilitates transcript elongation through nucleosomes." Cell 92, 105 (1998). Orphanides, G., Wu, W.-H., Lane, W., Hampsey, M., Reinberg, D., "The chromatin-specific transcription elongation factor FACT comprises human SPT16 and SSRP1 proteins." Nature 400, 284 (1999). Belotserkovskaya, R., Oh, S., Bondarenko, V., Orphanides, G., Studitsky, V., Reinberg, D., "FACT Facilitates Transcription-Dependent Nucleosome Alteration." Science 301, 1090 (2003). Saunders, A., Werner, J., Andrulis, E., Nakayama, T., Hirose, S., Reinberg, D., Lis, J., "Tracking FACT and the RNA Polymerase II Elongation Complex Through Chromatin in Vivo." Science 301, 1094 (2003). Espinosa, J. n. M. (2008), vol. 22, pp. 2743-2749. Buratowski, S., "TRANSCRIPTION: Gene Expression--Where to Start?" Science 322, 1804 (2008). Fleming, A. B., Kao, C.-F., Hillyer, C., Pikaart, M., Osley, M. A., "H2B ubiquitylation plays a role in nucleosome dynamics during transcription elongation." Mol Cell in press (2008). Gnatt, A. L., Cramer, P., Fu, J., Bushnell, D. A., Kornberg, R. D., "Structural basis of transcription: an RNA polymerase II elongation complex at 3.3 A resolution." Science 292, 1876 (Jun 8, 2001). Shaevitz, J. W., Abbondanzieri, E. A., Landick, R., Block, S. M., "Backtracking by single RNA polymerase molecules observed at near-base-pair resolution." Nature (Nov 23, 2003). Wang, M. D., Schnitzer, M. J., Yin, H., Landick, R., Gelles, J., Block, S. M., "Force and velocity measured for single molecules of RNA polymerase." Science 282, 902 (Oct 30, 1998). Brower-Toland, B. D., Smith, C. L., Yeh, R. C., Lis, J. T., Peterson, C. L., Wang, M. D., "Mechanical disruption of individual nucleosomes reveals a reversible multistage release of DNA." PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 99, 1960 (2002). Brower-Toland, B., Wacker, D. A., Fulbright, R. M., Lis, J. T., Kraus, W. L., Wang, M. D., "Specific contributions of histone tails and their acetylation to the mechanical stability of nucleosomes." Journal of Molecular Biology 346, 135 (2005). Cui, Y., Bustamante, C., "Pulling a single chromatin fiber reveals the forces that maintain its higher-order structure." Proc Natl Acad Sci U S A 97, 127 (Jan 4, 2000). Adelman, K., La Porta, A., Santangelo, T. J., Lis, J. T., Roberts, J. W., Wang, M. D., "Single molecule analysis of RNA polymerase elongation reveals uniform kinetic behavior." Proc Natl Acad Sci U S A 99, 13538 (Oct 15, 2002). Neuman, K. C., Abbondanzieri, E. A., Landick, R., Gelles, J., Block, S. M., "Ubiquitous transcriptional pausing is independent of RNA polymerase backtracking." Cell 115, 437 (Nov 14, 2003). Davenport, R. J., Wuite, G. J., Landick, R., Bustamante, C., "Single-molecule study of transcriptional pausing and arrest by E. coli RNA polymerase." Science 287, 2497 (Mar 31, 2000). Zlatanova, J., Leuba, S. H., "Stretching and imaging single DNA molecules and chromatin." Journal of Muscle Research and Cell Motility 23, 377 (2002). Leuba, S. H., Karymov, M. A., Tomschik, M., Ramjit, R., Smith, P., Zlatanova, J., "Assembly of single chromatin fibers depends on the tension in the DNA molecule: Magnetic tweezers study." PNAS 100, 495 (January 21, 2003, 2003). Zlatanova, J., Leuba, S., "Chromatin Fibers, One-at-a-time." Journal of Molecular Biology 331, 1 (2003). Revyakin, A., Ebright, R. H., Strick, T. R., "Promoter unwinding and promoter clearance by RNA polymerase: Detection by single-molecule DNA nanomanipulation." PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 101, 4776 (2004). Roopa, T., Shivashankar, G. V., "Direct Measurement of Local Chromatin Fluidity Using Optical Trap Modulation Force Spectroscopy." Biophys. J. 91, 4632 (2006).

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56.

57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67.

References, Page 3 of 4

Osheim, Y. N., Sikes, M. L., Beyer, A. L., "EM visualization of Pol II genes in Drosophila: most genes terminate without prior 3' end cleavage of nascent transcripts." Chromosoma 111, 1 (2002). Leuba, S. H., Yang, G., Robert, C., Samori, B., van Holde, K., Zlatanova, J., Bustamante, C., "Three-dimensional structure of extended chromatin fibers as revealed by tapping-mode scanning force microscopy." Proc Natl Acad Sci U S A 91, 11621 (Nov 22, 1994). Claudet, C., Angelov, D., Bouvet, P., Dimitrov, S., Bednar, J., "Histone octamer instability under single molecule experiment conditions." J Biol Chem 280, 19958 (May 20, 2005). Mathe, J., Visram, H., Viasnoff, V., Rabin, Y., Meller, A., "Nanopore Unzipping of Individual DNA Hairpin Molecules." Biophys. J. 87, 3205 (November 1, 2004, 2004). Evans, E., Ritchie, K., Merkel, R., "Sensitive force technique to probe molecular adhesion and structural linkages at biological interfaces." Biophys J 68, 2580 (Jun, 1995). Pant, K., Karpel, R. L., Rouzina, I., Williams, M. C., "Salt dependent binding of T4 gene 32 protein to single and double-stranded DNA: Single molecule force spectroscopy measurements." Journal of Molecular Biology 349, 317 (2005). Greenleaf, W. J., Woodside, M. T., Block, S. M., "High-resolution, single-molecule measurements of biomolecular motion." Annu Rev Biophys Biomol Struct 36, 171 (2007). Bustamante, C., Macosko, J. C., Wuite, G. J., "Grabbing the cat by the tail: manipulating molecules one by one." Nat Rev Mol Cell Biol 1, 130 (Nov, 2000). Allemand, J. F., Bensimon, D., Croquette, V., "Stretching DNA and RNA to probe their interactions with proteins." Curr Opin Struct Biol 13, 266 (Jun, 2003). Block, S. M., Asbury, C. L., Shaevitz, J. W., Lang, M. J., "Probing the kinesin reaction cycle with a 2D optical force clamp." Proc Natl Acad Sci U S A 100, 2351 (Mar 4, 2003). Yu, J., Xiao, J., Ren, X., Lao, K., Xie, S., "Probing Gene Expression in Live Cells, One Protein Molecule at a Time." Science 311, 1600 (2006). "We have made or data acquisition code available via sourceforge and described it on OpenWetWare. We will do the same with the analysis software, and all of our shotgun DNA mapping applications. http://openwetware.org/wiki/Koch_Lab:Publications/Drafts/Versatile_Feedback/Software/How_to_ obtain_the_software." Bockelmann, U., EssevazRoulet, B., Heslot, F., "Molecular stick-slip motion revealed by opening DNA with piconewton forces." PHYSICAL REVIEW LETTERS 79, 4489 (1997). "Public protocol available on our OpenWetWare site: http://openwetware.org/index.php?title=Koch_Lab:Protocols/Unzipping_constructs&oldid=212016 ." "Public protocol available on our OpenWetWare site: http://openwetware.org/index.php?title=Koch_Lab:Protocols/MicrosphereDNA_tethering/Glass%2C_dig%2C_biotin%2C_microsphere%2C_4kb_DNA&oldid=283740." SantaLucia, J., Jr., "A unified view of polymer, dumbbell, and oligonucleotide DNA nearestneighbor thermodynamics." PNAS 95, 1460 (February 17, 1998, 1998). Dauxois, T., Peyrard, M., Bishop, A. R., "Dynamics and Thermodynamics of a Nonlinear Model for DNA Denaturation." Physical Review E 47, 684 (Jan, 1993). Ares, S., Voulgarakis, N. K., Rasmussen, K. O., Bishop, A. R., "Bubble nucleation and cooperativity in DNA melting." Phys Rev Lett 94, 035504 (Jan 28, 2005). Voulgarakis, N. K., Redondo, A., Bishop, A. R., Rasmussen, K. O., "Probing the mechanical unzipping of DNA." Phys Rev Lett 96, 248101 (Jun 23, 2006). Adelman, K., Wei, W., Ardehali, M. B., Werner, J., Zhu, B., Reinberg, D., Lis, J. T. (2006), vol. 26, pp. 250-260. Adelman, K., Marr, M. T., Werner, J., Saunders, A., Ni, Z. Y., Andrulis, E. D., Lis, J. T., "Efficient release from promoter-proximal stall sites requires transcript cleavage factor TFIIS." Molecular Cell 17, 103 (2005). Deuschle, U., Hipskind, R. A., Bujard, H., "RNA polymerase II transcription blocked by Escherichia coli lac repressor." Science 248, 480 (Apr 27, 1990). Sotomayor, M., Schulten, K., "Single-molecule experiments in vitro and in silico." Science 316, 1144 (2007).

CAREER: Single-Molecule Analysis of Genomic DNA and Chromatin in Eukaryotic Transcription

68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79.

80. 81.

References, Page 4 of 4

Almer, A., Horz, W., "Nuclease hypersensitive regions with adjacent positioned nucleosomes mark the gene boundaries of the PHO5/PHO3 locus in yeast." Embo J 5, 2681 (Oct, 1986). Griesenbeck, J., Boeger, H., Strattan, J. S., Kornberg, R. D., "Affinity purification of specific chromatin segments from chromosomal loci in yeast." Molecular and Cellular Biology 23, 9275 (2003). Mason, P. B., Struhl, K., "Distinction and Relationship between Elongation Rate and Processivity of RNA Polymerase II In Vivo." Molecular Cell 17, 831 (2005). Fleming, A. B., Osley, M. A. (2008). "http://www.ed.gov/about/offices/list/ocr/edlite-minorityinst.html." "Postbaccalaureat Research and Education Program (PREP) @ UNM. http://biology.unm.edu/PREP/index.asp." "Wikipedia: Open Notebook Science http://en.wikipedia.org/wiki/Open_Notebook_Science." "KochLab members' open lab notebooks http://openwetware.org/wiki/Koch_Lab:Notebooks." "KochLab Protocols on OpenWetWare http://openwetware.org/wiki/Koch_Lab:Protocols." "A KochLab graduate student, Andy Maloney, has posted instructions on how to construct our OEM laser diode system for optical tweezers. http://openwetware.org/wiki/Koch_Lab:Research/How_to_build_your_own_laser_diode." "The technical portion of this proposal will be uploaded to the public document sharing site, Scribd. It will be available after submission via the site: http://www.pdfcoke.com/sjkoch4914." "Using start-up budget, we have acquired a 2 TB file server and thanks to Caleb Morse (a talented undergradaute ECE major), we already have a functioning system for sharing raw data files. Data files are copied to the server daily and become immediately publicly available. http://kochlab.org/files/data." Lang, A. S. I. D., Bradley, J.-C., "Chemistry in Second Life http://usefulchem.wikispaces.com/SLchemPaper." (2009). "UNM Junior Physics Lab on OpenWetWare: http://www.openwetware.org/wiki/Physics307L."