Supplemental Material Promoter-Specific Dynamics Of

Supplemental Material

Promoter-specific dynamics of TATA-binding protein association with the human genome Yuko Hasegawa and Kevin Struhl

Table of Contents • S upplemental methods • Supplemental Figures S1 to S10 • Supplemental Tables S1 to S3

SUPPLEMENTAL M ETHODS

Plasmid construction

The DNA fragment coding 3xHA was prepared by annealing oligos (Supplementary Table 2,) and inserted between the BamHI and EcoRI sites of pBudCE4.1 (Invitrogen). The ERT2 DNA fragment was PCR amplified from pCAG-Cre-ERT2 (Addgene #14797) using 1Fw and 1Rv primers and inserted between SalI and BamHI sites of pBudCE4.1-3xHA. ERT2-3HA fragment was then PCR amplified using 2Fw and 2Rv primers and inserted between BmtI and BamHI sites of pmCherry-C1 (Clontech). The TBP coding sequence was PCR amplified from human cDNA and was cloned into SalI site of this plasmid containing ERT2-3HA, yielding pCMV-TBP-ERT2-3HA.

Cell culture pCMV-TBP-ERT2-3HA construct was transfected to HEK293 (ATCC) cells, and the stable line was established by G418 selection. Cells were routinely cultured in phenol-red free DMEM containing 10% FBS.

Nuclear and nucleolar fractionation Cells cultured in 6 cm dish were washed twice with ice-cold PBS twice, then scraped and collected in 1.5 ml tubes. After brief centrifugation, cells were suspended in 250 µl of CE buffer (10 mM Hepes-KOH (pH7.5), 1.5 mM MgCl2, 10 mM KCl, 0.05% NP-40) and incubated for 3 min on ice. Cells were again centrifuged for 3 min at 3,000 rpm, and the pellet were suspended in 250 µl of RIPA buffer (50 mM Tris-HCl (pH8.0), 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 0.5% Na-Deoxycholate, 0.1% SDS) and sonicated (Misonix 3000,

level 2, on 30 sec off 30 sec, total 2 min). After centrifugation for 12,000 rpm for 10 min, the supernatant was collected and used as the nuclear fraction. For quantification of the protein, we performed western blotting and detected TBP-ERT2 and endogenous TBP using anti-TBP antibody (Abcam #ab51841). The signals were detected by LAS3000 imaging system and analyzed by ImageJ software. To isolate nucleoli, cells in five 150-mm dishes were trypsinized and collected in the 15 ml tubes, then suspended in ice-cold CE buffer without detergent (10 mM Hepes-KOH

(pH7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT) and incubated for 5 min on ice. The cell suspension was transferred to pre-chilled 7 ml Dounce tissue homogenizer, and homogenized to obtain nuclei. After centrifuge and removed supernatant, the nuclear pellet was suspended in 3 ml of S1 buffer (0.25 M Sucrose, 10 mM MgCl2) and then put on the top of 3 ml S2 buffer

(0.35 M Sucrose, 0.5 mM MgCl2). The tube was centrifuged at 2,500 rpm for 5 min at 4oC, and after that supernatant was removed. The pellet was suspended in 3 ml of S2 buffer and sonicated for 3 x 10 sec using a probe sonicator. The sonicated solution was layered onto 3 ml

o of S3 buffer (0.88 M Sucrose, 0.5 mM MgCl2), and centrifuged for 10 min at 3,500 rpm, 4 C. The pellet containing nucleolus was washed with 500 µl of S2 buffer, spun down at 3,500 rpm for 5 min and then used for the assay.

Sample preparation during the time course TBP-ERT2 nuclear translocation was induced by 4OHT (Sigma-Aldrich, H7904) addition at the final concentration of 100 nM to culture medium. Cells were retained in the CO2 incubator for indicated times. For western blotting, cells were immediately put on ice, and proceed to the subcellular fractionation. For ChIP, culture medium was removed and then exchanged by fixing solution (see below) at room temperature to avoid over crosslinking. This step takes 2.5 minutes, thus we summed up this time gap and incubation time and used as real time points (7.5, 12.5, 17.5, 32.5, 62.5, 92.5, 362.5, 1442.5 min) for simulating turnover rate.

Chromatin immunoprecipitation Chromatin was prepared from two 150 mm dishes. After removing the culture medium, cells were fixed at room temperature with fixing solution (1% formaldehyde, methanol-free in DMEM without serum) for 10 min. Fixation was stopped by adding 125 mM glycine. Cells were washed with ice-cold PBS twice, and then scraped from dishes and collected. Cells were suspended in 2 ml Lysis Buffer 1 (50 mM Hepes-KOH (pH7.5), 140 mM NaCl, 1mM EDTA, 10% Glycerol, 0.5% NP-40, 0.25% Triton X-100, with protease inhibitor and phosphatase inhibitor), rotated for 10 min at 4°C. After collecting cells by centrifugation, cells were re- suspended in 2 ml Lysis Buffer 2 (10 mM Tris-HCl (pH8.0), 200 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, with protease inhibitor and phosphatase inhibitor) and rotated for 10 min at 4°C. Cells were collected by centrifugation and re-suspended in 1.3 ml Lysis Buffer 3 (10 mM Tris- HCl (pH8.0), 100 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 0.1% Sodium deoxycholate, 0.5% N-Lauroylsarcosine, with protease inhibitor and phosphatase inhibitor), and then sonicated (Misonix 3000 sonicator, level 2, ON 30 sec, OFF 30 sec, 7 cycles (total sonication time is 3.5 min)). Majority of the fragment sizes were approximately 200-500 bp. Cell lysates were centrifuged at 12,000 rpm for 10 min at 4°C after adding Triton X-100 (1% final concentration), and the supernatants were used for the analysis. The protein concentration of the lysates was measured by BCA Protein Assay Kit (Thermo Scientific) and the amounts of the lysates yielding 1.3 µg total protein were used for immunoprecipitation. The volume of the lysates were adjusted to 1 ml by adding Lysis Buffer 3. For sequencing samples, yeast chromatin prepared from HA-TBP expressing strain were added at the constant ratio (0.039 µg total protein in the lysate, that corresponds to 3% of total human protein amount in the lysates) to the human cell lysates as a spike-in control. 1/10 volumes of the mixed lysates were kept as input samples. 0.3 µg of anti-HA antibody (Santa Cruz #sc-7392) or 3 µg of anti-TBP antibody (Abcam #ab51841) was added to the lysates, and the mixtures were rotated overnight at 4°C. Immunoprecipitation was performed at 4°C for 3 hr with 50 µl protein G dynabeads (Thermo Fisher). Beads were washed five times with ice-cold Wash buffer 1 (50 mM Hepes-KOH

(pH7.6), 500 mM LiCl, 1 mM EDTA, 1% NP-40, 0.7% Sodium deoxycholate) and rinsed with TE (10 mM Tris-HCl (pH8.0), 1 mM EDTA) with 50 mM NaCl. Immunoprecipitates were eluted by incubating the beads in 75 µl Elution buffer (50 mM Tris-HCl (pH 8.0), 10 mM EDTA, 1% SDS) at 65°C for 15 min and collecting the supernatants. This step was repeated twice. The eluted samples were incubated overnight at 65°C for de-crosslinking. After treating with RNase A and Proteinase K, DNA was purified with PCR purification kit (QIAGEN) by adding 100 µl water. 55.5 µl of eluted DNA from IP samples and 2.5 µl of eluted DNA (diluted to 1/10) from input samples were used for the library preparation. For qPCR analysis, 3.3 µl of eluted DNA from IP samples and four dilution series of input samples for standards were used. Primers are listed in Supplementary Table 2.

Yeast chromatin preparation HA-TBP yeast strain was grown in YPD at 30 °C to an 0.4 OD600 and fixed for 20 min at room temperature with final 1% formaldehyde. Then the fixation was quenched with final 250 mM glycine for 5 min at room temperature. Cells were collected and washed with PBS once, and were disrupted in Mini Bead Beater (BioSpec Products, Inc.) with 3 cycles of 5 min run at maximum speed and 1 min rest on ice following suspending in 400 µl of FA lysis buffer (50 mM Hepes-KOH (pH7.5), 150 mM NaCl, 1% Triton X-100, 0.1% Sodium deoxycholate, 0.1% SDS). After centrifugation and removal of the supernatant, the pellet was suspended in 480 µl of FA lysis buffer and sonicated (Misonix 3000 sonicator, level 6, ON 10 sec, OFF 10 sec, 21 cycles (total sonication time is 3.5 min) x 6 times with 1 min interval). The sample was centrifuged at maximum speed at 4°C for 20 min, and then the supernatant was collected and used as yeast chromatin.

Library preparation and sequencing Sequencing library preparation was performed with NEBNext Ultra DNA library prep kit (NEB #E7370S) and NEBNext Multiplex Oligos (NEB #E7335S) following the

manufacturer’s instructions. PCR cycles for amplification of adaptor-ligated DNA were 9 cycles for input samples and 14 cycles for IP samples. Large fragments were removed using 0.4 x AMPure beads (Beckman Coulter). Libraries were sequenced on NextSeq 500.

ChIP-seq computational analysis W e concatenated the geno me sequen ces of human (hg38) a nd cerevis iae (sacCe r3) to gener ate a combined genom e sequence . We added “_s ” to al l yeas t chromosom e name s to avoi d chromo some name duplic ation, and built a c ustom Bowtie2 inde x from this combined g enome s equence. Sequen ced rea ds were align ed to th is custom library wit h default p arameters (-- s ensi tive). The resul ting S AM files were then spli t into the two files con taining re ads mapp ed t o human chromosomes and mapped to yeast chromosome. Duplicates were removed by P icar d, and only reads exceed mapping qua lity 30 were retained exc ept for s pecific a nalysis ( repetitive genes including ribosomal DNA). The number of reads mapped to human or yeast g enomes and that pa ssed the qua lity filter are listed in Supplementa ry Tab le 1. We ca lculated R RPM (ref erence adjusted RPM) by dividi ng the nu m ber of the reads mapped to human g eno me loc i by t he tot al reads number mapped to yeast genome (the number s in the MQ>30 c olumn in Supplementary Table 1, and that in the MQ>0 column for repetitive gene analysis). T his RR PM v alue was use d as the intensit y of the C hIP si gn al in this st udy. P eak calling was p erformed using MACS2 with a P- value thresho ld of 0.01. We extracted 13,148 peak s tha t w ere calle d at least two sam ples in the later time po ints (60, 90, 360, 1440 m in). Peak a nnotation was performed using UROPA wit h cu stom GTF file combined G RCh38.94 and t RNA GTF files downloaded from En sembl and U CSC g enome b rowser respe ctively.

5 A No treat 5 10 15 30 60 90 min Cyt Nuc Cyt Nuc Cyt Nuc Cyt Nuc Cyt Nuc Cyt Nuc Cyt Nuc TBP-ERT2

Endogenous TBP

Tubulin

B No treat 5 10 15 30 60 90 360 1440 min

TBP-ERT2

Endogenous TBP

C EtOH 4OHT (90 min) Cyto Nuc S3 NO(x20) Cyto Nuc S3 NO(x20) TBP-ERT2

TBP Nuclear fraction

Fibrillarin S3

Tubulin Nucleolar

Supplemental Fig. S1. TBP-ERT2 localizes to the nucleus upon tamoxifen treatment. (A) Western blotting analysis of TBP-ERT2, endogenous TBP, Tubulin alpha (cytoplasmic marker), and Histone H3 (nuclear marker) in the cytoplasmic (Cyt) and the nuclear (Nuc) fraction following the given time after tamoxifen treatment. (B) Western blotting analysis of TBP-ERT2 and endogenous TBP in nuclear fraction. (C) Western blotting analysis of TBP-ERT2, endogenous TBP, Fibrillarin (nucleolar marker), and Tubulin alpha (cytoplasmic marker) in the cytoplasmic (Cyt), nuclear (Nuc), S3, and nucleolar (NO) fraction before (EtOH) and 90 min after tamoxifen treatment (4OHT). Nucleolar fraction was concentrated to 20-fold. A B

1440min 0.38 0.55 0.72 0.74 0.77 0.76 0.78 0.81 1

360min 0.36 0.54 0.71 0.73 0.78 0.80.8110.81 Promoter (84.07%)

5’ UTR (0.3%) 90min 0.37 0.55 0.72 0.75 0.79 0.810.81 0.78 3’ UTR (0.65%) 1st Exon (0.33%) value 60min 0.36 0.53 0.71 0.73 0.78 10.8 0.80.76 1.0 Other Exon (0.7%) 0.8 1st Intron (3.61%) 30min 0.42 0.59 0.76 0.7710.780.790.780.77 Other Intron (3.33%) 0.6 Downstream (<=3kb) (0.41%) 15min 0.42 0.59 0.75 10.770.730.750.730.74 Distal Intergenic (6.59%) 0.4

10min 0.45 0.62 10.750.760.710.720.710.72

5min 0.4510.620.590.590.530.550.540.55

0min 10.450.450.420.420.360.370.360.38

0min 5min 10min 15min 30min 60min 90min 360min1440min

C D Randomly selected regions TBP-ERT2 peaks 3000 Frequency 1000 0 TBP-ERT2 targets TBP-ERT2 0 5 10 15 20

ordered by RRPM at 1440 min TBP-ERT2 signal ratio (0 min vs 1440 min) IP_0min IP_5min IP_10min IP_15min IP_30min IP_60min IP_90min IP_360min IP_1440min

0 50 100 150 200 RRPM

Supplemental Fig. S2. TBP-ERT2 ChIP signal gradually increases after tamoxifen treatment. (A) Location of TBP-ERT2 peaks. (B) Spearman correlation of ChIP-seq signal at peaks between samples at a given time point after tamoxifen induction. (C) ChIP-seq signal (RRPM: Reference-adjusted RPM) at each target site through time-course. (D) Histogram of the frequency (number) of regions or TBP-ERT2 peaks having indicated ChIP-seq signal ratio between 0 min and 1440 min samples. A B 1.2 Turnover rate 1 5 1 1 0.8 0.2 0.8 0.1 1 0.6 1.2 0.05 0.8 0.6 1 0.025 0.4 0.6 0.8 0.4 0.0125 0.6 0.4 0.00625 0.4 0.2 0.2 Relative ChIP signal 0.003125

Relative protein amount 0.2 0.2 0 0 0.0015625 0 50 100 150 200 250 300 350 0 10 20 30 40 50 60 70 80 90 0 0 0 200 400 600 800 1000 1200 1400 0 200 400 600 800 1000 1200 1400 Time after induction (min) Time after induction (min)

C 500 D 8 450

400

350 7

300

250 6

Halftime 200

150 (Occupancy)

2 5 100

50 Log

0 4 -1 -0.5 0 0.5 1 1.5 2 2.5 3 -Log10(Turnover rate) 25 50 75 100 125 150 175 200 225 250 275 300 Halftime Others Peaks annotated as Pol III promoter

Supplemental Fig. S3. ChIP signal increasing is reﬂected in the binding turnover rate. (A) Change of the nuclear TBP-ERT2 amount over the time-course. Markers are real data measured by western blotting (means of the three inde- pendent analyses) and the line is simulated change of the protein amount based on the equation describing nuclear translocation of the protein (Methods). Small window indicates the enlargement of the data at the early time points. (B) Simulated increase of the relative ChIP signal with various Lambda (binding turnover rate) values. Small window indicates the enlargement of the data at the early time points. (C) Relation between Lambda (binding turnover rate) and halftime (time point where the relative ChIP signal reaches to 0.5). (D) Scatter plot of TBP-ERT2 peaks in log2 occupancy vs halftime. Green color indicates peaks annotated as pol III genes. Black are the others. Cutoff (Resnorm = 0.4) 8

4 (Occupancy of TBP-ERT2) 2 Log

0.0 2.5 5.0 7.5 10.0 Resnorm (Goodness of fitting)

Supplemental Fig . S 4. C utoff fo r pea ks tha t w e s et ba sed o n G oodness-of-ﬁtting. A TBP consensus INR BREu BREd

1.00 1.00 1.00 1.00

0.75 0.75 0.75 0.75

0.50 0.50 0.50 0.50

0.25 0.25 0.25 0.25 Fast Middle Slow Cumulative frequency 0.00 0.00 0.00 0.00 0 300 600 900 1200 0 200 400 600 800 0 300 600 900 1200 0 100 200 300 400 500 600 Match score Match score Match score Match score

B TBP consensus INR BREu BREd 1000 1500 500 600 400 750 1000 300 400 500 200

Match score 500 200 250 100

0 0 0 0 Fast Middle Slow Fast Middle Slow Fast Middle Slow Fast Middle Slow

Supplemental Fig. S5 . Analysi s o f motif s i n Po l II core promote r. (A ) Cumulative frequencie s o f match score o f hit sequences (forward direction) to each motif in ±100 bp region (±10 bp region for INR) from gene start site. (B) Boxplots of the highest match score of hit sequences (forward direction). C A o p ( S Slow class promoters h f r HOXD13 HOXA10 HOXB13 HOXA13 up

o tt LMX1B Name T

Barhl1 CDX1 p m (+/- 100 bp from TSS) TBP F pl : // o

em m t TBP consensus - TBP consensus + e i n

e rs luster di c e m

n n

( e F g

t -s

a i

s Consensus sequence TBP m l ui he

STATAAAWRSVVVBN KRCCACGCCCMCYT Fig o t t e r i

f DGYMATAAAAH .

E-v s NNTAAWTGNB . o CYMATAAAAH CCAATAAAAH

GYMATAAAA

i g NYAATTAA n 6 CDX1 a

S . l

u o K l o e c no w

w e c XA13 O H 0 n la

n .05

t ss r T i m

F ,

bindin

. XD13 O o H

50 h m t m 0 o

l t ).

g e

(

E-value Fisher m

2.8e-38

A Barhl1 f 1.9e-2 4.3e-3 1.1e-3 1.0e-4 3.8e-7 1.5e-8 1.5e-8

r o

)

( o

± t

Th m i 10 f

li XB13 O H b st S). r p i c

o f he

( f o

B B

d m

)

XA10 O i H

Probability 0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 s n

SS).

o s c f lo a

whi -500 t w

n LMX1B c ch

Position ofbestsiteinsequence(from TSS, bp) la

o ss -400 f

bindin

p e r

be 0 1 2 3 -300 m g st

o c

m t

luster e

hi o rs

2−2 2−1 1−2 1−1 t t -200

i f

s s i t

ea e

a s re rc

-100

s he

igni

ﬁc

equen 0

an Cen t l y 100 LMX1B MA0703.1 Barhl1 MA0877.1 HOXA10 MA0899.1 HOXB13 MA0901.1 CDX1 MA0878.1 HOXD13 MA0909.1 HOXA13 MA0650.1 TBP MA0108.2 t

en c i Mo e r s. i c

200 h

(

C e d

)

i n cc 300

S u lo p=2.5e-10 p=1.4e-11 p=4.3e-14 p=6.2e-23 p=1.3e-46 p=9.5e-8 p=1.1e-15 p=4.0e-17 r r w en

400

c c la e ss s 500 Strong TBP consensus containing TATA-less

0.45 0.45 Fast Fast Middle Middle Slow Slow 0.40 0.40

0.35 0.35 AT contents AT contents

0.30 0.30

-200 bp -100 bp 0 +100 bp +200 bp -200 bp -100 bp 0 +100 bp +200 bp Distance from TSS Distance from TSS

Supplemental Fig. S7 . TATA-less po l II promoters i n Slow class ten d t o have highe r A T conten t. Averag e AT-content s in the promoters containing strong TBP consensus (left) or TATA-less promoters (right). Promoters that have strong T BP consensus (P-value < 0.0001, PWMtools) or TATAWAW (RSAT, pattern matching) are categorized as Strong TBP consensus containing (n=1459, Fast n=895, Middle n=415, Slow n=149). Rest were categorized in TATA-less (n=4164, Fast n=3201, Middle n=700, Slow n=263). Shadow areas represent 95% conﬁdence intervals for the means. A B C D p=0.88 Pausing index (Prox/GB) 15 p=3.9e-05 15 p=0.0036 10 PRO-seq signal Prox GB 10 Prox Gene body 5 10 5 0 5 (PRO-seq signal)

(PRO-seq signal) 0 2 2 −5 Log PRO-seq signal Log 0 −5 PRO-seq pausing index Gene Group Group Group E 15 15 15 Sense High&Fast Sense High&Middle Sense High&Slow Antisense Antisense Antisense 10 10 10

5 5 5

PRO-seq signal 0 PRO-seq signal 0 PRO-seq signal 0

-1 kb -0.5 kb 0 +0.5 kb +1 kb -1 kb -0.5 kb 0 +0.5 kb +1 kb -1 kb -0.5 kb 0 +0.5 kb +1 kb Distance from TSS Distance from TSS Distance from TSS 15 15 15 Sense Sense Sense Antisense Low&Fast Antisense Low&Middle Antisense Low&Slow 10 10 10

5 5 5

PRO-seq signal 0 PRO-seq signal 0 PRO-seq signal 0

-1 kb -0.5 kb 0 +0.5 kb +1 kb -1 kb -0.5 kb 0 +0.5 kb +1 kb -1 kb -0.5 kb 0 +0.5 kb +1 kb Distance from TSS Distance from TSS Distance from TSS

F p=0.0026 p=0.34 Antisense/Sense ratio 5

Sense 0

−5 As PRO-seq signal (Antisense/Sense Ratio) 2 Gene Log Group

Supplemental Fig. S8 . Polymerase pausin g doe s no t associat e wit h th e bindin g turnove r rat e o f TBP , an d Hig h an d Slow class Pol II promoters show strong transcription in sense direction. (A) Schematic illustration of the proximal region (Prox: from -500-bp to +500-bp of TSS) and the gene body region (GB: from +500-bp of TSS to TES). (B) PRO-seq signal (Woo et al. 2018*) in the proximal region. (C) PRO-seq signal in the gene body (GB) region. (D) Pausing index. P-values are t-test results. (E) PRO-seq signal in sense (blue) and anti-sense (red) direction. (F) Schematic illustration of the region

(from -500-bp to +500-bp of TSS) for counting PRO-seq signal (left), and log2 ratio between anti-sense and sense PRO-seq signal (right). P-values are t-test results.

*Woo YM, Kwak Y, Namkoong S, Kristjansdottir K, Lee SH, Lee JH, Kwak H. 2018. TED-Seq Identifies the Dynamics of Poly(A) Length during ER Stress. Cell Rep 24: 3630-3641 e3637. E-value=9.1e-12

E-value=9.3e-07

E-value=2.6e-06

E-value=3.7e-05

E-value=6.3e-05

E-value=3.5e-22

E-value=4.1e-20

E-value=4.4e-18

E-value=3.9e-14

Supplemental Fig. S9. Discovered motifs enriched in high TBP occupancy and fast TBP turnover class of promoters that are strongly transcribed (Pol II-ser5 ChIP signal of Fig. 4D > 1; n = 219) compared to high occupancy and slow turnover with similar transcriptional activity (n = 125). A 1500

1000

500 Match score

0 Fast Middle Slow

B TBP consensus (pvalue < 0.0001) TBP consensus (pvalue < 0.001) 100 100

75 75 58.3% 62.9% 54.3% 54.4% 53.8% 66.7% Frequency 92.4% 0 100% 91.7% 91.4% 100% 96.5% 50 50 1 2 >3 Percentage Percentage 28.6% 25 25 20% 33.9% 33.9% 25% 33.3% 8.6% 14.3% 9.4% 9.9% 8.3% 8.6% 7.6% 3.5% 8.3% 8.3% 8.6% 0 0 2.9% 2.3% 2.3% Fast_F Fast_R Middle_FMiddle_R Slow_F Slow_R Fast_F Fast_R Middle_F Middle_R Slow_F Slow_R

Supplemental Fig. S10 . TB P consensu s mot if i n Po l II I genes. (A ) Boxplo ts o f th e highe st match scor e of h it sequence s in forward direction. (B) Percentages of the po l II I promoters (pseudogene s wer e removed , n=217 , Fast n=11 , Midd le n=35, Slow=171) that have hit sequences to TBP consensus motif (left: P-value < 0.0001, right: P-value < 0.001, PWMtools). % in the % in the % in the % in the Sample Total reads Mapped 0 times total reads Mapped total reads Mapped to hg38 mapped reads Mapped to sacCer3 mapped reads input_0min 32,834,384 544,285 1.7% 32,290,099 98.3% 32,251,821 99.9% 38,278 0.1% input_5min 40,468,937 565,707 1.4% 39,903,230 98.6% 39,828,313 99.8% 74,917 0.2% input_10min 34,332,896 531,837 1.5% 33,801,059 98.5% 33,735,832 99.8% 65,227 0.2% input_15min 45,183,241 640,534 1.4% 44,542,707 98.6% 44,471,017 99.8% 71,690 0.2% input_30min 54,189,936 727,886 1.3% 53,462,050 98.7% 53,330,184 99.8% 131,866 0.2% input_60min 43,575,407 678,361 1.6% 42,897,046 98.4% 42,787,995 99.7% 109,051 0.3% input_90min 39,602,962 646,229 1.6% 38,956,733 98.4% 38,866,732 99.8% 90,001 0.2% input_360min 41,574,884 680,870 1.6% 40,894,014 98.4% 40,797,127 99.8% 96,887 0.2% input_1440min 44,255,025 680,279 1.5% 43,574,746 98.5% 43,482,779 99.8% 91,967 0.2% IP_0min 61,463,578 1,412,705 2.3% 60,050,873 97.7% 56,694,056 94.4% 3,356,817 5.6% IP_5min 48,153,540 970,311 2.0% 47,183,229 98.0% 45,245,093 95.9% 1,938,136 4.1% IP_10min 62,715,387 1,127,539 1.8% 61,587,848 98.2% 59,532,301 96.7% 2,055,547 3.3% IP_15min 46,491,870 1,030,356 2.2% 45,461,514 97.8% 43,406,502 95.5% 2,055,012 4.5% IP_30min 59,597,348 1,242,744 2.1% 58,354,604 97.9% 56,383,307 96.6% 1,971,297 3.4% IP_60min 56,181,699 1,250,257 2.2% 54,931,442 97.8% 52,472,684 95.5% 2,458,758 4.5% IP_90min 55,323,612 1,373,512 2.5% 53,950,100 97.5% 51,364,702 95.2% 2,585,398 4.8% IP_360min 62,691,237 1,596,832 2.5% 61,484,366 98.1% 58,894,000 95.8% 2,590,366 4.2% IP_1440min 64,460,688 1,206,871 1.9% 63,129,110 97.9% 60,192,825 95.3% 2,936,285 4.7%

Supplemen ta l Tab le S 1 . Num bers o f rea d s mapp e d t o h um a n gen ome (hg38 ) a n d yea st genome (sacCer 3) Experiment Primer Name Sequence Plasmid construction 3xHA Fw GATCCTACCCTTATGACGTCCCCGACTACGCTGGTTACCCTTATGACGTCCCCGACTATGCCGGCTACCCCTACGATGTCCCGGATTACGCATGAG Plasmid construction 3xHA Rv AATTCTCATGCGTAATCCGGGACATCGTAGGGGTAGCCGGCATAGTCGGGGACGTCATAAGGGTAACCAGCGTAGTCGGGGACGTCATAAGGGTAG Plasmid construction 1Fw ATTCCTGCAGGTCGACTCTGCTGGAGACATGAGAGCTGC Plasmid construction 1Rv CATAAGGGTAGGATCCGGCCGCAGCTGTGGCAGGGAAACC Plasmid construction 2Fw GAACCGTCAGATCCGCTAGCGTCGACGGACCATCTGCTGGAGACATG Plasmid construction 2Rv TAGATCCGGTGGATCCGAATTCTCATGCGTAATCCGGGAC qPCR qPCR-45S Fw CCGGGGGAGGTATATCTTT qPCR qPCR-45S Rv CCAACCTCTCCGACGACA qPCR qPCR-5S Fw CCTGAACGCGCCCGATCTC qPCR qPCR-5S Rv AAGCCTACAGCACCCGGTATTC qPCR qPCR-RAB5B Fw GGTCAGTAGCGGCCTTCCTA qPCR qPCR-RAB5B Rv GAGGCGGGTCAAAGTGAAAC qPCR qPCR-RPS9 Fw TCTCGTGACGTTTTCACGCACC qPCR qPCR-RPS9 Rv CTCCCTACTTCTCAGGTTTCCACT qPCR qPCR-chr8NC Fw CCATCCCACTAGGGAGAAAA qPCR qPCR-chr8NC Rv ATGCAACAGCTGTGAGGTGA

Supplemental Table S2 . Primer list GEO accession Experiment Cell line Reference GSE133729 TBP-ERT2 ChIP-seq HEK293 This study DRX000062 (SRA) MNase-seq HEK293 Sugano et al. 2010. GSE31477 (sample GSM935534) pol II ChIP-seq (Ab: 8WG16) HEK293 ENCODE project GSE97827 (sample GSM2579073) pol II Ser5-P ChIP-seq HEK293 Fong et al. 2017. GSE97827 (sample GSM2579070) pol II Ser7-P ChIP-seq HEK293 Fong et al. 2017. GSE97827 (sample GSM2579063) pol II Ser2-P ChIP-seq HEK293 Fong et al. 2017. GSE31477 (sample GSM935302) KAT2A(GCN5) ChIP-seq HeLa-S3 ENCODE project GSE127427 (sample GSM3634283) EP300 ChIP-seq HeLa-S3 ENCODE project GSE31477 (sample GSM935511) SMARCA4(BRG1) ChIP-seq HeLa-S3 ENCODE project GSE32465 (sample GSM803455) TAF1 ChIP-seq HeLa-S3 ENCODE project GSE37403 (sample GSM917814) SNAPC1 ChIP-seq MCF10A Baillat et al. 2012. GSE32970 (sample GSM1008573) DNase-seq HEK293T ENCODE project GSE118739 (sample GSM3346248) POLR3A ChIP-seq HEK293 Choquet et al. 2019. GSE103719 (sample GSM2779677) PRO-seq HEK293 Woo et al. 2018.

Supplemental Table S3. GEO accession numbers of publicly available datasets used in this study