Supplementary Information for "Rapid Analysis of the DNA Binding Specificities of Transcription Factors with DNA Microarrays"

Supplementary Methods

(a)    Synthesis of the DNA for the Rpn4 Proof-of-Principle Microarrays

(b)    Microarray Preparation for the Rpn4 Proof-of-Principle Microarrays

(c)    PBM Data Quality Control Filters

(d)    Electrophoretic Mobility Shift Assays (EMSAs)

(a)  Synthesis of the DNA for the Rpn4 PBM Proof-of-Principle Microarrays 

    The following Cy3-labeled oligonucleotide (Operon) was spotted at 10 uM in 150 mM K2HPO4, pH 9.0 for alignment purposes: 5' TCAGAACTCACCTGTTAGAC 3'. This oligo was spotted to create a series of alignment spots, essentially as described previously (Bulyk et al., Proc. Natl. Acad. Sci. U.S.A. 2001 Jun 12; 98(13):7158-7163.) The following set of oligonucleotides was synthesized (Operon) so as to represent various positive and negative control spots for binding by the yeast transcription factor Rpn4:

(1) 5' NNNNNNNNNNTTCTTCTTCTTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(2) 5' NNNNNNNNNNNTCNTCNTCNTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(3) 5' NNNNNNNNNNTTCTTCTTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(4) 5' NNNNNNNNNNNTCNTCNTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(5) 5' NNNNNNNNNNTTCTTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(6) 5' NNNNNNNNNNNTCNTCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(7) 5' NNNNNNNNNNCTCATCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(8) 5' NNNNNNNNNNCTCATCCTCATCNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(9) 5' NNNNNNNNNNWTTTGCCACCNNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(10) 5' NNNNNNNNNNNNNYRCCRCYRNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(11) 5' NNNNNNNNNNNNNCGCCACCNNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(12) 5' NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(13) 5' NNNNNNNNNNTGTCCTACTGCTNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(14) 5' NNNNNNNNNNYYYYYYYYYYYYNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(15) 5' NNNNNNNNNNYYYYYYYYYNNNNNNNNNNTCAAGTCAATCGGTCC 3'
(16) 5' NNNNNNNNNNYYYYYYNNNNNNNNNNTCAAGTCAATCGGTCC 3'

The following 16mer was synthesized with a 5' amino linker and HPLC-purified (Operon) and used as a universal primer: 5' GGACCGATTGACTTGA 3'. Each of the 12 unmodified full-length oligonucleotides listed above was combined with the amino-tagged 16mer in a 2:1 molar ratio in a Sequenase reaction using 20 uM 16mer. The completed extension reactions were exchanged into 150 mM K2HPO4, pH 9.0 using CentriSpin-10 spin columns (Princeton Separations, Inc.). The resulting samples were transferred to a 384-well plate for arraying.

(go to top of page)

(b)  Microarray Preparation for the Rpn4 PBM Proof-of-Principle Microarrays

    Glass slides (Gold Seal) were cleaned for 0.5 to 2 hrs in 2 N nitric acid.  After rinsing in distilled water, the slides were soaked in distilled water for 5 to 15 minutes, and then washed once with acetone.  The slides were silanized by immersing them for 15 minutes in a solution of 1% aminopropyl-methyl-diethoxysilane (Fluka) dissolved in 95% acetone.  After washing the slides twice in acetone, the slides were baked for 30 minutes at 75°C.  The surface of the slides was then activated by placing the slides in a solution of 0.5% 1,4-diphenylene-diisothiocyanate (PDC) (Fluka) dissolved in a solution consisting of 40 ml pyridine and 360 ml anhydrous N,N-dimethylformamide for 2 to 4 hrs.  The slides were then washed twice with methanol, twice with acetone, and stored in a dessicator until use.  A custom-built arraying robot equipped with piezo-electric printheads was used to print the microarrays.  After printing, the microarrays were incubated overnight at room temperature, then for 1 hr at 37°C in a humidity chamber containing 300 mM K2HPO4, pH 9.0.  The rest of the PDC surface was inactivated by a 10 min incubation in 1% ammonium hydroxide/0.1% SDS/200 mM NaCl.  After washing in 4xSSC, the slides were neutralized in 6xSSPE/0.01% Triton X-100, washed twice in 4xSSC, then washed in 2xSSC and spun dry in a clinical centrifuge.  Slides were stored in a closed box at room temperature until use.

(go to top of page)

(c)  PBM Data Quality Control Filters 

Total # of spots on microarrays = 8192

(1) We considered only those spots with known sequence. We removed empty, blank, and control spots, and also spots for which the Research Genetics primers did not map well to the claimed intergenic region, thus possibly leading to poor PCR quality.

Total # of possible spots considered = 6814


(2) We employed a number of spot filters based on the SybrGreen I data.  The following numbers of spots passed each of the following filters, applied in the following order:
    1. Raw data spot quality (flags; intraspot SD/mean; 50% pixels > B+2SD):  6683
    2. Variability (inter-array SD/mean <= 1):  6683
    3. Length < 1500:  6654
    4. Intensity (SybrGreenI intensity >= 10000):  6625
    5. Density (SybrGreenI/length >= 61; this is 10% of the array average):  6494
   
Total # of spots with acceptable SybrGreen I data = 6494


(3) We employed a number of spot filters based on the PBM data:
1.    Raw data spot quality (flags; intra-spot SD/median > 2)
2.    Variability (inter-array SD/mean > 1)
3.    Must have SybrGreen data in order to calculate log ratio

dataset
 array 1 (# spots filtered out due to reasons #1 above)
 array 2 (# spots filtered out due to reasons #1 above)
 array 3 (# spots filtered out due to reasons #1 above)
 total # spots removed (required data in >= 2 out of 3 arrays)
 additional # spots removed due to reasons #2 above (i.e., due to variability)
 additional # spots filtered out due to SybrGreen I filters #3-5 above
SybrGreen I
 150
 244
 132
 131
 0
 189
Rap1
 40
 94
 511
 23
 21
 -
Abf1
 1107
 320
 439
 280
 100
 -
Mig1
 126
 95
 92
 65
 9
 -



(4) We removed duplicate spots. In cases where there were identical spots, we kept the one with PBM data and higher absolute SybrGreen I signal intensity.

Total # of possible non-duplicate spots = 6723

dataset non-duplicate spots with SybrGreen I (and PBM) data
 average SD/mean
SybrGreen I
 6449 (95.9%)
 0.130
Rap1
 6431 (95.7%)
 0.323
Abf1
 6142 (91.4%)
 0.282
Mig1
 6442 (95.8%)
 0.202


(go to top of page)

(d)  Electrophoretic Mobility Shift Assays

Complementary biotinylated DNA oligonucleotides, each 45 bp in length, were synthesized such that they contained the predicted Rap1 binding site, flanked by its native flanking sequence from the given intergenic region. A positive control probe containing a known Rap1 binding site and a negative control probe lacking a Rap1 binding site were also synthesized and used in EMSAs.

The oligonucleotides were diluted in TE, pH 8.0 to a working stock of 10 uM. The working stocks for each of the two complementary oligonucleotides (designated "-U" for biotinylated "upper", and "-L", for "lower") were combined in the following ratios, respectively: 1:2 for both control probes and iYPL221W, and 1:5 for the probe iYLL051C. The solutions were then brought to a final volume of 20 ul with TE, pH 8.0. Each such pair of single-stranded complementary oligonucleotides were annealed in a thermocycler according to the following thermocycling protocol, resulting in approximately 1 uM final concentration of each biotinylated, double-stranded oligonucleotide:

(1) 94°C for 3 minutes
(2) Ramp down 1°C in 10 seconds (0.1°C per second)
(3) Hold at current temperature for 1 minute
(4) Repeat Steps 2 and 3 to 37°C
(5) 3°C Hold

The annealed oligonucleotide probes were filtered using Millipore's Microcon YM-100 (100,000 MWCO) before use in EMSAs.

The sequences of the oligonucleotides that we used for the Rap1 EMSA probes were as follows:

    Positive Control Rap1 binding site:

    Rap1-iYHL038C-U
        5'ACTTTCACTAAATACACCCATACACACCAATCTTGGATTCTACTC 3'

    Rap1-iYHL038C-L
        5'GAGTAGAATCCAAGATTGGTGTGTATGGGTGTATTTAGTGAAAGT 3'

    Negative Control Rap1 binding site:

    Rap1-iYLR288C-U
        5'TCCGTATGTTTATGTTGCTATTTTGATGTAAATAAAAAAGTTGAA 3'

    Rap1-iYLR288C-L
        5'TTCAACTTTTTTATTTACATCAAAATAGCAACATAAACATACGGA 3'

    Interesting Probes:

    Significant PBM p-value with poor match to PBM derived Rap1 binding site:

    Rap1-iYLL051C-U
        5'AAACAATGCCCACTAGCCGGGGTGTACGGGACCTTAAATCTAAGT 3'

    Rap1-iYLL051C-L
        5'ACTTAGATTTAAGGTCCCGTACACCCCGGCTAGTGGGCATTGTTT 3'

    Significant PBM p-value with poor match to TRANSFAC derived Rap1 binding site:

    Rap1-iYPL221W-U
        5'GGGCATACTTTACGGGGTGCACGGATTTTAGCAGTCTTTTTCTTT 3'

    Rap1-iYPL221W-L
        5'AAAGAAAAAGACTGCTAAAATCCGTGCACCCCGTAAAGTATGCCC 3' 


EMSAs were performed according to manufacturer’s protocols for the LightShift© Chemiluminescent EMSA Kit (Pierce). The "+" lanes correspond to EMSA reactions that contained approximately 350 nM DNA probe and approximately 29 nM GST-His6-Rap1. The 'no protein' EMSA reactions ("-" lanes) contained approximately 350 nM DNA probe and no Rap1 protein. All EMSA binding reactions contained 1x LightShift Binding Buffer, 0.05 ug/ul poly(dI-dC) nonspecific DNA competitor, 2.5% glycerol, 50 mM KCl, 0.2 ug/ul BSA, 0.05% NP-40, and 0.5 mM zinc acetate. The 20 ul binding reactions were allowed to sit at room temperature for 1 hour. Subsequently, 2.2 ul of Novex® 5X Hi-Density TBE Sample Buffer (Invitrogen) was added to the reactions, and 12.5 ul of the reaction was run on a 6% polyacrylamide DNA retardation gel (Invitrogen) in 0.5X TBE at 100 V for 50 minutes. The contents of the gel were transferred to a Biodyne® B Pre-cut Modified Nylon Membrane, 0.45 uM (Pierce), and UV-crosslinked to the membrane at 120 uJ/cm2.  The membrane was then treated with developing buffers (Lightshift Blocking Buffer with stabilized Streptavidin-Horse Radish Peroxidase conjugate, Wash Buffer, Substrate Equilibration Buffer, Luminol/Enhancer Solution and Peroxide Solution) according to manufacturer’s protocols.  The blot was promptly exposed to Kodak film for ½ second, which was then developed.

  (go to top of page)