HDX_SI_04032022_clean - ja2c01991_si

Supporting Information for

Alkyne

tagged Raman probes for local environment

sensing by

ydrogen

euterium

exchange

Xiaotian Bi

†

, Kun Miao

†

and Lu Wei*

Division of Chemistry and Chemical Engineering, California Institute of

Technology,

Pasadena, California 91125, United States

† These authors contributed equally: Xiaotian Bi, Kun Miao

*Corresponding author. Email:

lwei@caltech.edu

Materials and Methods

Chemicals

All

chemicals

were purchased from Sigmal

Aldrich

unless otherwise specified

Ethynyl

2′

deoxyuridine

U, CAS# 61135

9) was purchased from TCI

America.

Ethynyl

deoxycytidine

(EdC, CAS#

69075

) was purchased

from Cayman Chemicals.

Ethynyl Urid

ine

(EU, CAS#

69075

) was

purchased from Sigma

Aldrich.

Ethynyl

deoxyuridine 5'

triphosphate

EdUTP)

was purchased from

Jena Bioscience.

BCECF, AM (2',7'

Bis

Carboxyethyl)

(and

Carboxyfluorescein, Acetoxymethyl Ester)

and BCECF,

free acid were purchased from ThermoFisher scientific.

Pwo DNA

olymerase

was

purchased

from Roche

Exonuclease I

(ExoI)

purchased from Sigma

Aldrich.

Nigericin sodium salt

(CAS#

28643

) was purchased from Sigma

Aldrich.

Propargyl

holine

romide (PCho, CAS#

111755

) was synthesized according

Briefly,

propargyl bromide (80 wt. % solution in toluene

, Sigma

Aldrich

)

added dropwise to

a stirring solution of

dimethylaminoethanol

(Sigma

Aldrich)

in anhydrous THF

in ice bath

under argon gas protection. The

reaction

mixture

was

slowly warmed up to room temperature and stirred

overnight. The

resulting

white solids were filtered and washed extensively with cold anhydrous

THF to obtain

pure

PCho

Buffer preparation

r solution experiments, w

e used different buffering systems according to the final

pD region to prepare D

O buffers. (Buffer range: citric acid

HPO

, 2.6

7.6;

NaH

HPO

, 6.2

8.2; CAPSO

CAPSO sodium salt,

8.9

10.3)

. We prepared

O buffer

by directly diluting corresponding buffer

wder

the same amount

of D

O as that

for the

O system.

keep ion strength consistent

for

all buffers

to avoid the influence

the

alkyne

HDX kinetics, w

e added extra NaCl to

those

buffers

with less salt concentrations

. The detailed buffer recipes are shown below:

pD= 5.3 (46.4 mM citric acid, 107.2 mM Na

HPO

); pD= 6.2 (33.9 mM citric acid,

132.2 mM Na

HPO

); pD= 6.6 (27.25 mM

citric acid, 145.5 mM Na

HPO

); pD=

7.0 (17.65 mM citric acid, 164.7 mM Na

HPO

); pD= 7.6 (DPBS); pD= 7.9 (3 mM

NaH

, 30 mM Na

HPO

, 60 mM NaCl); pD= 9.4 (28mM CAPSO, 7 mM CAPSO

sodium salt, 143 mM NaCl); pD= 10.4 (19 mM CAPSO, 16 mM CAPSO sodium

salt,

134 mM NaCl); pD= 10.4 (10 mM CAPSO, 25 mM CAPSO sodium salt, 125

mM NaCl).

The final pD values were determined by the pH

meter. For salt

concentration experiment, 2 M NaCl

was added into the DPBS

O buffer solution

For

BCECF

experiments,

high K

conditions were used to meet the requirement of

nigericin.

The detailed

high K

buffer recipes are shown below:

High K

O buffer

pD = 7.64 (120 mM KCl, 5 mM NaCl, 10 mM Na

HPO

, 1.8 mM KH

in D

)

;

6.88, 7.11, 7.24 buffers were made by adjusting pD of pD

7.64 buffer

with

drops of

0.1 M HCl in D

pD=

8.01 buffer was made by adjusting pD of pD

7.64

with drops of

0.1 M NaHCO

in D

High K

O buffer

pH = 7.35 (120 mM KCl,

5 mM NaCl, 10 mM Na

HPO

, 1.8 mM KH

in H

;

pH = 6.8

6.95

, 7.

buffers were made by adjusting pH of pH=7.64 buffer using 0.1 M HCl in H

pH/pD determination

pH reading for all D

buffers

and H

O buffers were acquired on

pH meter

(Mettler Toledo

FiveEasy Plus FP20

, with LE 410 sensor)

at room temperature (~

°

The pH meter was calibrated with standard solutions (pH

4.01, 7.00, 10.01

Mettler Toledo) before measuring the custom

made buffers.

As we discussed in

the main manuscri

pt, to avoid confusion, we show all the pD values in our

manuscript from direct pH meter reading (i.e. the

meter reading

) for all D

O buffers.

As shown below, the calibration factor is small and would not influence any of our

conclusions.

The offset between pH

meter reading

and pD value

was obtained by

two control

experiments. First, we compare

the pH

meter reading

of DPBS

O solution (7.35

considered as real pD if

O is

replaced

with D

) to that of DPBS

O solution

(7.6

, considered as pH

meter reading

Second, w

e used the pH

sensitive

ratiometric

fluorophore

BCECF to provide

additional calibration between pH

meter reading

and pD

value

reported through BCECF ratios

In brief, we dissolved BCECF acid in water

and made a 2 mM stock solution. We

then

diluted the stock BCECF solution in

the

igh K

O/D

O buffers

with

adjusted

meter reading

(directly

reading from

the

meter

)

to be in

the range 6.8

8.1.

The final BCECF concentration is

We then

acquired fluorescence images using 445

and 488

excitation lasers (ZEISS

LSM 980). The ratios of fluorescence intensity at 488 nm over

that at

445 nm were

plotted against respective

reading

from the

meter

for H

buffers

and D

buffers

(Figure S7

)

The offset

between

linearly fitted

ratio vs pH/pD

curve

0.25

The

reading from the pH

meter

for

O buffers

shown as

meter reading

(and

are

reported as the

D values in our manuscript

)

The determined relationship

consistent

with

the

above

two

different experiments:

meter reading

–

0.25

DFT calculation

DFT calculations were performed using the Gaussian09 software. Structures were

optimized and then characterized using frequency calculations at the B3LYP/6

311(G)++(d,p) level of

theory.

Spontaneous Raman Spectroscopy

Spontaneous Raman spectra were acquired using an upright confocal Raman

spectrometer (Horiba Raman microscope; Xplora plus). A 532 nm YAG laser was

used to illuminate the sample with a power of 12 mW through a 100×,

N.A. 0.9

objective (MPLAN N; Olympus). Data acquisition was performed with 10 s

integration by the LabSpec6 software. For whole spectra recording, background

was subtracted by measuring signal from the same solution without probe

molecules. The spectra sho

wn in

Figure

1a and S1a are normalized to

the

alkyne

peak.

EdU, EdC and EU were dissolved into DMSO to make 100 mM stock solutions.

PCho was dissolved into H

O to make

M stock solution. For measurement, EdU,

EdC and EU are 1:10 diluted into correspondin

g H

buffers

or D

O buffers, while

PCho is 1:50 diluted into corresponding H

buffers

or D

O buffers.

Model molecules (4

fluorophenylacetylene

(CAS# 766

, methy

ethynylbenzoate

(CAS# 3034

, 4

ethynylbenzaldehyde

(CAS# 63697

ethynyl

nitrobenzene

(CAS# 937

Ethynylanisole

(CAS#

768

)

were dissolved into DMSO to make 100 mM stock solutions. For measurement in

DMSO

O system, model molecules were diluted into

the

1:1 DMSO

(

DPBS

=7.6)

solution

to ensure good dissolvability for all model molecules

with corresponding dilution factors

For measurement in methanol

OD system,

model molecules were 1:10 diluted into methanol

(CAS#

1455

)

All data

are confirmed by

at least t

hree

sets of independent experiments.

Stimulated Raman Scattering (SRS) Microscopy

A picoEmerald laser system (Applied Physics and Electronics) was used as the

light source for SRS microscopy.

Briefly, i

t produces 2 ps

pump (tunable from 770

–

990 nm, bandwidth 0.5 nm, spectral bandwidth ~ 7 cm

) and Stokes (1031.2

nm, spectral bandwidth 10 cm

)

pulse

with 80 MHz repetition rate. Stokes beam

is modulated at 20 MHz by an internal electro

optic modulator. The spatia

lly and

temporally overlapped Pump and Stokes beams are introduced into an inverted

laser

scanning microscope (FV3000, Olympus), and then focused onto

the sample

by a 25X water objective (XLPLN25XWMP, 1.05 N.A., Olympus). Transmitted

Pump and Stokes beams

are collected by a high N.A. condenser lens (oil

immersion, 1.4 N.A., Olympus) and pass through a bandpass filter (893/209

BrightLine, 25mm, Semrock) to filter out Stokes beam. A large area (10×10 mm)

Si photodiode (S3590

09, Hamamatsu) is used to measure

the pump beam

intensity. A 64 V reverse

biased DC voltage is applied on the photodiode to

increase the saturation threshold and reduce response time. The output current is

terminated by a 50

terminator and pre

filtered by a 19.2

23.6

MHz band

pass

filter

(BBP

21.4+, MiniCircuits) to reduce laser and scanning noise. The signal is

then demodulated by a lock

in amplifier (SR844, Stanford Research Systems) at

the modulation frequency. The in

phase X output is fed back to the Olympus IO

interface box (FV30

ANAL

OG) of the microscope. 30 μs time constant is set for

the lock

in amplifier. Correspondingly, 80 μs pixel dwell time is used, which gives

a speed of 21.3 s/frame for a 512

512

pixel image

, with two frame

averaging.

Laser powers are monitored throughout

image acquisition by an internal power

meter and power fluctuations are controlled within 1%.

The power for the pump

and Stocks beam is about 25 mW and 2

0 mW, respectively.

bit greyscale

images were acquired by Olympus Fluoview 3000 software.

To minimi

ze th

line

pattern issue

likely

due to

an interfering Radio frequency (RF)

picked up by

our

lock

in amplifier detection (demodulation at 20 MHz), we have optimized our

alignment and replaced a few

bandpa

filters between

our

photodiode and lock

in amplifier

For EdU and PCho

measurement,

the wavelengths of pump laser

for SRS

are

845.9 an

d 844.6 nm, respectively. For ratiometric imaging of

EdU and PCho

after

the

exchange, the wavelengths of pump laser

for SRS

are 8

55.5

and 85

4.5

nm,

respectively. Off

resonance images were taken

under

851.3 nm pump

wavelength

For

O diffusion

measurement, the

pump

wavelength

820.5

For EdU/EdU

dimer spectra recording

the wavelengths of pump laser

tuned from

843.9

847.9 nm

with

0.5 nm interval

For two

color ratiometric

imag

ing

EdU and

PCho

during

the

exchange

the

wavelengths of pump laser

for SRS

nd SRS

are 8

and

855

nm, respectively.

Cell culture

, sample preparation

and

alkyne

HDX

cells

For all SRS imaging experiments

ultured HeLa

CCL2 (ATCC)

cells were seeded

onto coverslips (12mm, #1.5,

Fisher)

with a density of 1 × 10

/mL in 4 well plate

with 0.3 mL DMEM culture medium (DMEM+10%FBS+1% penicillin

streptomycin)

for 20 h at 37 °C and 5% CO

Prior to imaging, c

overslips were collected and

attached

to a

microscope slide (1mm thick, VWR)

with

imaging spacer (

0.12mm

thick,

Sigma

Aldrich)

For the EdU experiment,

DMEM culture medium was then changed to DMEM

medium (FBS

fre

, Gibco

) for 20

22 h for cell cycle synchronization. After

synchronization,

the

medium was replaced back to DMEM culture medium and

EdU (10 mM stock in DPBS) was simultaneously added to a concentration of 100

M for

h. Then 4% PFA was added for 20 min for f

ixation. After that, DPBS

was used to wash away PFA and fixed cells could be stored in DPBS

for

several days.

For UV

irradiat

ion on live

cells, cells were put inside the

biosafety cabinet (BSC

)

with UV

(254 nm)

on for one hour.

Morphologies were quickly check

with no

severe abnormality under a transmission light microscope.

The cells were fixed

4% PFA

immediately after the UV irradiation.

For all the fixed cells

alkyne

HDX experiments, corresponding D

O buffers were

used

to wash the cells three times and then the coverslip was taken out to make

an imaging chamber filled with

designating

O buffers for SRS imaging.

For live

cell BCECF experiment

, cells were first loaded with 2 μM BCECF

AM in

HBSS for 20 min in the CO

incubator and were subsequently treated with 10

igericin in pH

7.35

igh K

O buffer for 5 min. Cells were switch

igh

O b

uffers

containing

M nigericin

with

designated

and put onto the

microscope slide. The image was collected on an inverted confocal microscope

(ZEISS LSM 980) using 40x water immersion objective (NA 1.2) with either 445 or

488 nm excitation laser. Both images were

quired

using

the

same PMT which

was s

et to collect photon

from 530

570 nm.

Control experiment

ere

done

under

similar conditions

yet

with the addition of DMSO

instead of

nigericin

in all the

buffer

solutions

For live

cells

alkyne

HDX experiment

for pD sensing

HeLa cells were first

incubated with 10 μM nigericin in high K

O buffer for 5 min at 37 ̊C. The high

O buffer was then removed, and cells were washed with high K

buffer

with

10 μM n

igericin of different pD

value

s. We start

timing as the cells were

washed with hig

h K

buffers. The coverslip with cells was placed onto a

microscope slide with spacer filled with 10 μM

igericin high K

buffers.

Control experiment

were

done

under similar conditions

but with

the addition of

DMSO

instead of

nigericin

in all

the

buffer

solutions

For the PCho and EdU experiment, DMEM culture medium was changed to DMEM

medium (FBS

free, Gibco) for synchronization. After synchronization, medium was

replaced back to DMEM culture medium by simultaneously adding both

propargylchol

ine (100 mM stock in DPBS) and EdU (10 mM stock in DPBS) to the

culture medium

with

final

concentration of 1 mM and 100

M, respectively, for

24 h.

For

the live

cells

alkyne

HDX experiments,

home

made DMEM

buffer

through

dissolving

DMEM pow

er into D

was

used to wash the cells three

times

hen the coverslip was taken out to make an imaging chamber filled with

the DMEM

O buffer

for SRS imaging.

ynthesis and purification of dsDNA and ssDNA

Pwo DNA polymerase, an enzyme showing good performance to a

ccept modified

nucleotides

was

used to incorporate EdUTP into DNA through PCR process.

typical PCR reaction contained

~100 ng plasmid template, 0.05 mM

each of

the

forward and reverse primers,

2.5

U polymerase and 10x polymerase buffer

with

magnesium

00μM of dNTPs (dATP, dCTP and dGTP

NEB)

instead of dTTP

and

00μM modified

dUTP. The reactions were done in an overall volume of 50 μL

with the addition of

UltraPure™ DNase/RNase

Free Distilled Water

(Invitrogen)

For ssDNA synthesis, asymmetric PCR

was

used. The concentration of

the

orward primer and

the

reverse

primer is 0.001 mM and 0.05 mM

respectively

while other conditions are

the

same with

that

indicated

above.

PCR

experiments

were performed on

Bio

Rad C1000 thermal cycler

rimers

and

the

emplate

employed for PCR experiments:

Forward primer

5 ́

GGAAATCGGTACTGGCTTTCCATTCGAC

reverse primer

3 ́

GTGAGTTAAAGTTGTACTCGAGTTTGTGTCCG

Template

(

sequence of

746

contains

382 T

GAAATCGGTACTGGCTTTCCATTCGACCCCCATGATGGTTCCGTTCAACTA

GCAGACCATTATCAACAAAATACTCCAATTGGCGATGGCCCTGTCCTTTTACC

AGACAACCATTACCTGTCGACACAATCTGTCCTTTCGAAAGATCCCAACGAA

AAGCGTGACCACATGGTCCTTCTTGAGTTTGTAACTGCTGCTGGGATTACAC

ATGGCATGGATGAGCTCTACAAAGGCGGTGGGTCGGGCGGGGGCTCCC

GGGGGTGGCGGTTCATGATCAGGTGGAGGGTCAGGGGGCGGATCAATGAG

CAAAGGAGAAGAACTTTTCACTGGAGTTGTCCCAATTCTTGTTGAATTAGATG

GTGATGTTAATGGGCACAAATTTTCTGTCCGTGGAGAGGGTGAAGGTGATGC

TACAAACGGAAAACTCACCCTTAAATTTATTTGCACTACTGGAAAACTACCTG

TTCCATGGCCAACACTTGTCACTACTCTGACCTATGGTGTTCAATG

CTTTTCC

CGTTATCCGGATCACATGAAACGGCATGACTTTTTCAAGAGTGCCATGCCCG

AAGGTTATGTACAGGAACGCACTATATCTTTCAAAGATGACGGGACCTACAAG

ACGCGTGCTGAAGTCAAGTTTGAAGGTGATACCCTTGTTAATCGTATCGAGTT

AAAAGGTATTGATTTTAAAGAAGATGGAAACATTCTCGGACACAAACTCGAGT

ACAACTTTAACTCAC

PCR products were pur

ified

with

a Monarch PCR & DNA Cleanup Kit

(

NEB

) and

confirmed

gel electrophoresis

and

anger

sequencing

. The products of PCR

reaction

were separated by

% agarose

(

UltraPure™ Low Melting Point Agarose

Invitrogen)

gel electrophoresis.

dsDNA and ssDNA were purified with a

Monarch

Gel Extraction Kit

(NEB).

Exonuclease I

(

Thermo Scientific

20 U/μL)

was used to

digest ssDNA in 37

°C

, which was confirmed by

gel electrophoresis

as well

Before

enzyme digestion, 5 min

°C

heat

shock was used for denaturation.

The

images

were recorded with

Bio

Rad Gel and Blot Imaging Systems

The combined PCR

products (~50 tubes per sample) were loaded into

Microcon

10kDa Centrifugal

Filter

(

Millipore

Sigma

) and washed with

UltraPure™ DNase/RNase

Free Distilled

Water

three times to remove the remaining salts

and EDTA

the

elusion buffer

through buffer exchange.

Then the PCR products in pure water were concentrated

into ~1

uL through vacufuge

(

Vacufuge Plus Concentrator, Eppendorf

). The

concentrated PCR products were diluted into DPBS

/ DPBS

O + 2 M NaCl

buffer solutions

for

alkyne

HDX kinetics measurement

ormation and purification of

EdU dimer

For solution samples, EdU

(or thymine)

was dissolved into water to make saturated

solution

. After frozen into ice, the EdU

(or thymine)

solution was put into a

homemade dry ice chamber and was irradiated with a UV lamp

(254nm

, UVLS

EL Series UV Lamp, 4 Watt

)

for 6 hours.

Analytical HPLC

coupled with mass

spectrometry (LC

MS)

for the UV

irradiated EdU product

was performed on Agilent

90 infinity LC system

using ZORBAX RRHD Eclipse Plus C18, 95Å, 2.1 x 50

mm, 1.8 μm column with Agilent 6140

Series Quadrupole LCMS

MSD/Mass Spectrometer System. The mobile phase is

water

(0.1% AcOH)

and acetonitrile with

running method of gradient

95% acetonitrile (1 ml/min,

min for total running time). The data shown in the supplementary

figures

(Fig

ure

S6e

)

are the absorption

(

254

nm) intensity

traces.

The

aqueous

reaction

mixture (100 mg) post UV

radiation was first concentrated

under reduced pressure

at 40 ̊C and was then load on a reverse phase Biotage

cartridge (12g SNAP Ultra C18). The

lash column chromatography was

automated by the Biotage Isolera System, with acetoni

trile and water as the mobile

phase

. The flushing gradient was set

to 5% to 40% ace

tonitrile over 10 column

volume (CV) and the EdU dimer should be the second UV active portion eluted

around

the

second to third CV. The collect portion was confirmed by nor

mal phase

TLC. EdU dimer would have an R

value of 0.15 and EdU has an R

of 0.

when

running with pure Ethyl Acetate.

The purified EdU dimer was confirmed through

high

resolution mass

spectrometry

(calculated exact mass:

505.

1571

detected

mass:

505.1575

)

and dissolved into H

O to prepare a stock solution for further

alkyne

HDX measurement

Immuno

fluorescence

staining

he UV

irradiated cells

and control cells

were

first fixed and

treated with 0.2%

Triton

100 in 1x DPBS (

no calcium, no

magnesium

ibco)

at room temperature

for

min.

Then the Triton

100 was

removed,

and cells were washed with 1x

DPBST (DPBS + 0.1% Tween 20) three times. Cells were incubated with

1% BSA,

22.52 mg/mL glycine in PBST

at room temperature for 60 min

for blocking.

After

washed with PBST, cells were incubated with

1:300 diluted a

nti

hymine

imer

antibody

(

Mouse monoclonal

~2 mg/mL

MilliporeSigma

)

0.1% BSA

, PBST

overnight at 4°C.

The

primary antibody solution

was

removed,

and cells

were

washed

three times with

DPBST. Then cells were incubated with 3

% BSA

PBST

at room

temperature for 60 min

for blocking.

After washed with PBST, cells were

incubated with 1:400 diluted

Goat anti

Mouse IgG (H+L),

(

Invitrogen

Superclonal™ Recombinant Secondary Antibody, Alexa Fluor 647

1 mg/mL

)

0.1%

BSA

, PBST

for

2 hours

room temper

ature. The

secondary

antibody solution

was

removed,

and cells

were

washed

three times with

DPBST. Fluorescent imaging

was conducted immediately after sample preparation through

the same Olympus

FV3000 confocal microscope with CW laser excitation (640

nm, Coherent OBIS

LX laser) and standard bandpass filter sets

Data processing

All spectra

wer

processed using LabSpec6 software. Spectral baselines were

subtracted. Peak centers and intensity were read out by Gaussian peak fitting. All

images were processed using ImageJ software. Corresponding off

resonance

images were subtracted.

For precise t

1/2

fitting, we obtained the time

zero intensity (i.e. I

º

H)) by

the

normalization of I

º

H) and I

º

D),

the

intensity measured from the

first exchange data point at the alkyne

H and the alkyne

D channel

, respectively.

º

H) is

defined as I

º

H)+I

º

D)/r, in which r is the

intensity

correction

ratio

defined as

dividing

I(R

º

the alkyne

probe solution after

equilibrium in D

O (i.e. finished with exchange), by I(R

º

the corresponding

alkyne

probe solution in

O with the same concentration without any exchange.

This strategy was used for both the spontaneous Raman measurement of solution

samples and the SRS imaging recording of cell samples.

I(R

º

and

I(R

º

)

from spontaneous Raman measurements

were read out by Gaussian peak

fitting

I(R

º

and

I(R

º

)

from SRS measurements (also refer as SRS

and SRS

) were read out from all the nucleus regions in one

field

of view based

on SRS images.

erivation

lkyne

HDX

kinetics

The rate

determining step (RDS) in HDX between alkyne

tagged probes (R

≡

H) and catalytic base OD

is shown in Eq. (1) below

It includes three elementary

steps: a) diffusional collision, b) equilibrium

redistribution of the hydrogen in the

intermediate state, and c) dissociation.

(1)

We can assume that: 1)

the

temperature remains constant; 2) the reaction is

diffusion

limited; 3) the intermediates are in steady states.

So the

concentrations

of the two intermediates remain the same in the reaction

, shown in Eq. (2).

[

(

≡

(

∙

)

]

[

(

≡

∙

(

)

]

(2)

i.e.

푘

[

푅

−

퐶

≡

퐶

−

퐻

]

[

푂퐷

]

−

푘

[

(

푅

−

퐶

≡

퐶

−

퐻

∙

푂퐷

)

]

푘

[

(

푅

−

퐶

≡

퐶

∙

퐻

푂퐷

)

]

−

푘

[

(

푅

−

퐶

≡

퐶

−

퐻

∙

푂퐷

)

]

푘

[

(

푅

−

퐶

≡

퐶

−

퐻

∙

푂퐷

)

]

−

푘

[

(

푅

−

퐶

≡

퐶

∙

퐻

푂퐷

)

]

−

푘

[

(

푅

−

퐶

≡

퐶

∙

퐻

푂퐷

)

]

we can get the concentration of the intermediate shown in Eq. (3).

[

(

푅

−

퐶

≡

퐶

∙

퐻

푂퐷

)

]

[

푅

−

퐶

≡

퐶

−

퐻

]

[

푂퐷

]

(

)

The overall rate (k) for transferring a proton from R

≡

H to OD

then becomes:

푘

[

≡

]

푘

[

(

푅

−

퐶

≡

퐶

∙

퐻

푂퐷

)

]

[

푅

−

퐶

≡

퐶

−

퐻

]

[

푂퐷

]

(

)

Since the reaction is diffusion

limited, there are some restrictions on rate constants:

푘

≪

푘

≪

푘

≈

푘

the overall rate

can be simplified as Eq. (5).

푘

[

푅

−

퐶

≡

퐶

−

퐻

]

[

푂퐷

]

(

)

In addition,

(

≡

(

)

(

(+,

)

≫

, so Eq. (5) can be further simplified

as Eq. (6).

푘

∙

(

5=>=?

≡

)

(

BCCD;8=?

AEF

)

∙

[

푂퐷

]

∙

[

푅

−

퐶

≡

퐶

−

퐻

]

(

)

is the diffusion

limited collision constant, upper

bounded by 10

. As

fluctuation of OD

concentration is negligible during HDX, the

RDS is considered

as a pseudo

first

order reaction. Since the acceptor pK

is much smaller than the

donor pK

, the corresponding exchange half

life (t

1/2

) is reduced to Eq. (

푡

푝

퐾

−

푝퐷

푙푔

(

푙푛

)

−

푝

퐾

(

퐻푂퐷

)

푝

퐾

(

퐷

푂

)

(

)

Here pK

designates the donor pK

, i.e.

푝퐾

(

푅

−

퐶

≡

퐶

−

퐻

)

. Taking the pK

alkynyl hydrogen as 20

25 and the pK

of HOD as ~15, we estimated the t

1/2

alkyne

HDX in physiological pD (7.6) to be on the order of minutes

For example,

f we used 20 for pK

alkynyl hydroge

n, the calculated t

1/2

could be 174 s.

Figure S1

Characterizations of hydrogen

deuterium exchange on terminal

alkyne groups

(

alkyne

HDX)

by spontaneous Raman spectroscopy.

Chemical structures and the corresponding spontaneous

Raman spectra of EdC

(10 mM) and EU (10 mM) solutions. Pink

shade highlight the terminal alkynes.

Spontaneous

Raman peaks of alkyne in

EU (10 mM, red

) and EdC (10 mM, black)

solutions before (solid lines) and after (dashed lines) HDX

Figure S2

The alkyne

HDX rates for

EdC and

EU solutions

) The Kinetics

trace of the spontaneous Raman spectra for the decrease of alkyne

peaks

and

the increase of alkyne

D peaks from EU (10 mM) during

alkyne

HDX in

the

pD=7.6

O buffer solution.

) Exponential curve fitting of normalized alkyne peak

intensit

ies

for EU (10 mM) from (

The a

verage of t

1/2

over t

hree

independent

measurements is also shown.

) The kinetics trace of the spontaneous Raman

spectra for the decrease of alkyne

peaks

and the increase of alkyne

D peaks

from EdC (20 mM) during

alkyne

HDX in

the

pD=7.6 D

O buffer solution.

)

Exponential curve fitting of normalized alkyne peak intensit

ies

for EdC (20 mM) in

(

The a

verage of t

1/2

over three independent measurements is also shown.

Figure

Fast D

2

O diffusion across cells in

solutions with different

osmolarities.

Time

trace

pontaneous Raman spectra

in the nuclear region of

a fixed cell in the DPBS

O solution. The solution spectrum is taken in the

surrounding region

without

cells

from

the same sample.

) SRS image of

vibrational peak (2490 cm

) for

the same set of

fixed cells in DPBS

O buffer

solution

after

3 min

(b) and 62 min (c) incubation.

d) Spontaneous Raman spectra

in the nuclear region

of a fixed cell in the

high

osmolarity

solution (

DPBS

O + 2

M NaCl

)

. The solution spectrum is taken in the surrounding region

without

cells

from

the same sample.

) SRS image of

vibrational peak (2490 cm

) for

the

same set of

fixed cells in DPBS

O + 2 M NaCl buffer solution a

fter

3 min

(e) and

63 min

(f) incubation

. Scale bar: 20

Figure S4

Representative data sets for SRS imaging

based

alkyne

HDX

kinetics

on EdU

labeled cells

) Exponential curve fitting of normalized

alkyne

peak intensit

ies

for EdU

corporated cells

pD=7.6 D

O buffer solution

)

Representative r

atiometric imaging

of SRS

/ SRS

(right)

generated by dividing

the SRS imag

at the alkyne

channel

(2224 cm

)

(left) by

ose

at the alkyne

D channel (1992 cm

)

(middle)

for cells

immersed in DPBS

O from a series of

alkyne

HDX time points (473 min, 1097 min, 1916 min)

Figure S

The alkyne

HDX kinetics

for

EdU and EdU

lab

led structures at

the

regular

(DPBS

2

O, D

, black

)

and high

(DPBS

2

O+2M NaCl, S, red)

salt

concentration.

) Exponential curve fitting of normalized alkyne peak intensit

ies

for EdU (10 mM)

(a);

EdU

corporated cells (b);

EdU

(10 mM)

(c)

;

EdU

labeled

dsDNA (d); and

EdU

labeled ssDNA (e)

pD=7.6 D

O buffer solution (black)

and pD=7.6 D

O buffer solution with 2M NaCl (red)

, respectively

f) DNA

gel

electrophoresis

for confirming the EdU

incorporated products of dsDNA and

ssDNA.

eft: 1 kb DNA ladder; right: EdU

labeled PCR products, containing 746

bp dsDNA and 746 nt ssDNA

DNA

gel electrophoresis

confirm

ing

the presence

EdU

sDNA

and EdU

ssDNA with

Exonuclease I

(ExoI) digestion.

eft: control

PCR products; right: PCR products after

Exonuclease I

(ExoI) d

igestion

Figure S6

Confirmation for the

induced

dimer formation

Spontaneous

Raman spectra of thymine solution

(yellow)

and thymine solution irradiated for 6

hours

(purple)

gray

boxed

regions show the featured

Raman spectra

changes

for thymine dimer formation, consistent with

what

was

reported

Ref.

)

Immuno

fluorescence

staining image

with anti

T dimer

Alexa Fluor 647

for control

cells without UV irradiation

(control

)

and

irradiated cells

(c)

d) HPLC trace

of EdU solution

(

light

green)

, EdU solution irradiated for 6 hours

(light purple)

and

purified EdU dimer

(magenta)

. Arrow

ed peaks indicate

the corresponding

color

coded

UV absorption traces in (

e) UV absorption traces of peaks indicated by

color

coded

arrow

in the HPLC trace

(

)

the

EdU solution before and after

irradiation.

Figure S

Calibration of solution and

ntracellular p

D by

ratiometric

fluorescent pH sensor

BCECF

a) Linear relationship between

ratios

(

488

/ F

445

)

taken at the 488 nm

excitation

and

445 nm excitation

of BCECF

both H

(light

blue)

and D

(orange)

buffers

with varying

reading results

from the

pH meter

over

the

physiological

relevant

range

)

Representative r

atiometric

imag

(

488

445

)

for BCECF in live cells in pD=

7.11

DPBS

O buffer

without (b, control,

average ratio 2.93±0.38)

and with nigericin (

average ratio

2.55±0.31

)

Representative

atiometric imag

es (

488

/ F

445

)

for BCECF in live cells in pD=7.64

DPBS

O buffer,

without (

, control,

average ratio 3.01±0.31)

and with nigericin

(

average ratio

3.34±0.44

Scale bar: 20