www.sciencemag.org/content/349/6248/628/suppl/DC1

Supplementary Materials for

Adaptive synergy between catechol and lysine promotes wet adhesion by surface salt displacement Greg P. Maier, Michael V. Rapp,

J. Herbert Waite,* Jacob N. Israelachvili,* Alison Butler*

*Corresponding author. E-mail: herbert.waite@lifesci.ucsb.edu (J.H.W.); jacob@engineering.ucsb.edu (J.N.I.); butler@chem.ucsb.edu (A.B.)

Published 7 August 2015, Science 349, 628 (2015) DOI: 10.1126/science.aab0556

This PDF file includes:

Materials and Methods Supplementary Text Figs. S1 to S15 Tables S1 to S4

2

Materials and Methods

Materials

2,3-dihydroxybenzoic acid (2,3-DHBA), catechol, 3,4-dihydroxybenzoic acid (3,4-

DHBA), dicyclohexylcarbodiimide (DCC), and tris(2-aminoethyl)amine (TREN) were

purchased from Aldrich. 4-methylcatechol (4-MC) was purchased from Acros Organics.

Benzyl bromide, palladium on carbon, and benzoic acid were purchased from Alfa Aesar.

N-hydroxysuccinimide (NHS) was purchased from Fluka. Triethylamine, sodium

phosphate dibasic, potassium hydroxide, and trifluoroacetic acid (TFA) were purchased

from Fisher. 3-hydroxybenzoic acid was purchased from TCI. CAPSO buffer was

purchased from Research Organics. Phosphate buffer was purchased from Fisher. H-

Lys(Z)-OH, H-Dab(Boc)-H, and H-Lys(Ac)-OH were purchased from Bachem. Unless

otherwise stated, all chemicals were used as received without further purification or

modification.

Microbial Siderophore Isolation, Purification, and Characterization

Cyclic trichrysobactin (CTC) was isolated, purified, and characterized using

previously published methods (17).

Autoxidation of Catechol Analogs

The autoxidation of 4-Methylcatechol (4-MC), catechol, 3,4-dihydroxybenzoic acid

(3,4-DHBA), and 2,3-dihydroxybenzoic acid (2,3-DHBA) was tracked using the

following procedure. A 5300 Biological Oxygen Monitor (Yellow Springs Instruments)

equipped with a Clark electrode was used to track oxidation kinetics through the

consumption of dissolved molecular oxygen. Buffer solutions were sparged with

compressed air to ensure the same starting concentration of dissolved molecular oxygen

for all experiments. Trace metal was removed from buffers with Chelex 100 resin (100-

200 mesh, sodium form, Bio-Rad) using the batch method. 3 ml of 29.5˚C ± 0.01 of 50

mM buffer was introduced to the reaction chamber (also held at 29.5˚C ± 0.01).

Phosphate buffer was used for pH 7.5 and CAPSO buffer was used for pH 10.0. The

reaction chamber was sealed with the Clark electrode, removing all non-dissolved air

from the system and allowed to equilibrate for 5 – 10 minutes. Kinetics experiments

began upon the injection of 10 µl of a catechol analog solution with a metal free injection

device and the percent oxygen remaining was recorded for ten minutes. All catechol

analog solutions were prepared immediately prior to use. 4-Methylcatechol and catechol

were dissolved in 0.5M HCl to prevent oxidation. 3,4-Dihydroxybenzoic acid and 2,3-

dihydroxybenzoic acid were dissolved in ethanol to maximize solubility. All experiments

were done in pseudo 1st order conditions with catechol analog in excess.

Siderophore Analog Synthesis

Trencam (TC) was synthesized according to previously published methods (16).

Tren-Lys-Cam was synthesized using well known peptide bond formation and protecting

group chemistries (30). The remaining homologs were synthesized by variations of the

synthesis scheme for TLC. See Fig. S2 for details of the synthesis procedure. See Fig.

S3 for the structures of TLC, TDC, TLP, TLB, TLAcC, and TC. The synthesis of TDC

used H-Dab(Boc)-OH in place of H-Lys(Z)-OH in step c. An additional final step was

performed in the synthesis of TDC to remove the Boc protecting group. This was done

3

with 50% trifluoroacetic acid in DCM at room temperature for 2 hours. The synthesis of

TLP was done using 3-hydroxybenzoic acid as the starting material in place of 2,3-

DHBA in step a. The synthesis of TLB used benzoic acid as a starting material rather

than 2,3-DHBA. The absence of hydroxyl groups obviated the need for step a. TLAcC

was synthesized using H-Lys(Ac)-OH in place of H-Lys(Z)-OH for step c.

Surface Force Apparatus (SFA) Technique and Measurements for the Natural

Siderophore and Analogs

The full details of the SFA technique are elaborated elsewhere (18). All

measurements were performed with a SFA 2000, manufactured by SurForce LLC. in

Santa Barbara, California. Briefly, for each experiment, two mica surfaces are prepared

by gluing a piece of freshly-cleaved, back-silvered mica (~1 cm2), of equal mica and

silver thicknesses, onto cylindrical glass disks (radius ~2 cm), with the pristine mica

surface facing upward. The two mica surfaces are installed into the SFA, with the pristine

mica surfaces facing each other. The surfaces are brought close together and small

droplets of aqueous buffer are injected between the surfaces (~50 μL total volume).

Normal force-distance measurements are then performed between the two surfaces in

aqueous solution. The contact area between the mica surfaces is verified as free from

asperities or contaminants based on the interferometric profile of the contact zone and the

measured forces, which are well documented for mica interacting in aqueous solutions

(19-21). Following, a small amount (~10 μL) of siderophore, or analog, in aqueous buffer

is injected into the gap solution between the surfaces, and the system is allowed to

equilibrate for 20 minutes as the siderophores adsorb to the mica surfaces. While

remaining at the same contact position, force-distance measurements are then performed

between the mica surfaces in the siderophore solutions. The aqueous solutions used in

SFA experiments were: (i) a 50 mM acetate + 150 mM KNO3 buffer solution for pH 3.3

and 5.5, and (ii) a 50 mM phosphate + 150 mM KNO3 buffer solution for pH 7.5. The

force-distance data shown in the main text and the Supplementary Information are

representative of measurements performed over at least 4 separate experiments for each

molecule and solution condition. The adhesion values, Fad, and compressed film

thicknesses, DT, are reported as the sample mean and standard deviation.

Supplementary Text

Mfp-Type vs. Siderophore-Type Catechol Autoxidation Rates

Catechol autoxidation by dissolved molecular oxygen is a pH dependent process that

accelerates at higher pH. The reaction proceeds through a series of one electron

oxidations and therefore, superoxide and semiquinone are present as intermediates. The

reaction mechanism for the autoxidation of catechol is not well defined.

The rate of oxygen consumption for the series of catechol compounds correlates

with the strength of electron donating or withdrawing group (see Fig. S1). The electron

donating methyl group of 4-MC promotes the fastest oxidation rate of the compounds

measured. Autoxidation of unsubstituted catechol is slower. Addition of an electron

withdrawing carboxylic acid, e.g., 3,4-DHBA, further slows the autoxidation rate. This

correlation holds true at both pH 7.5 and pH 10.0. Increasing the concentration of the

4

catechol compounds increases the autoxidation rate, as is expected for pseudo 1st order

conditions.

The autoxidation rate of 2,3-DHBA is even slower than 3,4-DHBA; these isomeric

compounds differ only in the position of the electron withdrawing substituent.

Intramolecular hydrogen bonding in 2,3-DHBA, between the 2-hydroxyl group and the

carbonyl oxygen atom, further protects the catechol from oxidation and raises the first

catecholic hydroxyl pKa above that of 3,4-DHBA, 10.06 and 8.82, respectively. The pH

of the wet adhesion environment must be significantly lower than the first pKa of the

catechol hydroxyl groups to ensure a bidentate H-bond with the target surface.

Ultimately, the subtle molecular differences make the common 2,3-DHBA catechol in

siderophores significantly more oxidation resistant and enlarge the pH window over

which these siderophore analogs bind to target surfaces.

ANOVA Analysis of pH dependent TLC Adhesion

We have performed a statistical analysis of the pH and contact time dependence of

TLC-mediated adhesion, shown in Fig. 2C. At each contact time, analysis of variance

(ANOVA) was performed between the data sets for pH 3.3, 5.5, and 7.5 (n ≥ 4); the data

sets were found to be statistically significant (i.e. the null hypothesis was rejected) at all

contact times at a P-value < 0.05. Pairwise t-tests were then run between each of the three

pH conditions at each contact time. The P-values for each t-test are summarized in the

table below, with P-values < 0.05 highlighted in grey. From this analysis, we can

conclude that the TLC-mediated adhesion between mica surfaces at pH 3.3 and pH 5.5 is

statistically different from the adhesion at pH 7.5 for a P-value < 0.05 (with the exception

of pH 5.5—pH 7.5 at 60 minutes of contact time). Additionally, the null hypothesis is

accepted for pH 3.3—pH 5.5 (except at 60 minutes of contact time), indicating that there

is no statistical difference between the TLC-mediated adhesion at pH 3.3 and pH 5.5.

5

Fig. S1. Autoxidation of 4-MC, Catechol, 2,3-DHBA and 3,4-DHBA at pH 10 (A.) and

pH 7.5 (B.). A. 3.3 mM of each catechol compound, pH 10.0 in 50 mM CAPSO buffer at

29.5°C. B. 10.0 mM 4-MC, catechol and 3.3 mM 2,3-DHBA, 3,4-DHBA and catechol at

pH 7.5 in 50 mM phosphate buffer at 29.5°C. Oxygen consumption was monitored using

a Clark electrode.

3.3 mM 2,3-DHBA

3.3 mM 3,4-DHBA

3.3 mM Catechol

3.3 mM 4-MC

50 mM pH 10.0 CAPSO Buffer at 29.5°C

A

50 mM pH 7.5 Phosphate Buffer at 29.5°C

B

6

Fig. S2. Synthesis Scheme for Tren-Lys-Cam. Reaction Conditions: (a) KOH, DMSO,

Benzyl Bromide, 4 hours. (b) NHS, DCC, anhydrous THF under N2, overnight. (c) THF,

H2O, Et3N, H-Lys(Z)-OH, overnight. (d) NHS, DCC, anhydrous THF under N2,

overnight. (e) Et3N, TREN, anhydrous DCM under N2, overnight. (f) EtOH, 3% HOAc,

Pd/C, overnight.

7

Fig. S3. The TLC-mediated adhesion force (and energy) between two mica surfaces in

buffered solution as a function of the number of moles of TLC injected into the

intervening gap solution between the mica surfaces. The total volume of intervening

solution between the two mica surfaces is typically ~50 μL. Error bars for the 10 min data

points (±12 mN m-1) have been omitted for visual clarity. The anionic mica surfaces used

in these experiments are ~ 1 cm2 and contain ~ 2.1 x 1014 negatively charged sites per

surface. The adhesion force between mica surfaces peaks when the total number of TLC

molecules injected between the surfaces (~ 6 x 1014 molecules) is roughly equivalent to

the number of negatively charged mica sites in the system (~ 4.2 x 1014 total sites

between the two surfaces). Increasing the number of moles of TLC injected into the

system causes a slight decrease in the adhesion force, a result of over-adsorption of TLC

molecules to the mica surface.

Table S1. P-values from the pairwise t-tests between the TLC-mediated adhesion data

sets at pH 3.3, 5.5, and 7.5. P ≤ 0.05 values are highlighted in grey. See supplemental

text for description.

Contact Time 2 min 10 min 30 min 60 min

pH 3.3—pH 5.5 P = 0.1274 P = 0.0952 P = 0.6216 P = 0.0158 pH 3.3—pH 7.5 P = 0.00001 P = 0.0062 P = 0.0014 P = 0.0129 pH 5.5—pH 7.5 P = 0.0002 P = 0.0492 P = 0.0135 P = 0.1954

8

Fig. S4. Synthetic Tren-based Siderophore Analogs. Numbered carbons correspond to

NMR data in Figs. S10-S13 and Tables S3-S4.

9

Fig. S5. Reverse Phase HPLC Purification of the Tren-based Siderophore Analogs. RP-

HPLC was carried out on a C4 preparative column (22 mm i.d., x 250 mm, Vydac). A

gradient elution was performed from 100% nanopure H2O with 0.05% trifluoroacetic acid

to 100% MeOH with 0.05% trifluoroacetic acid. The rate of gradient transition was

optimized for each individual compound. The eluent was monitored at 215 nm.

RP-HPLC of Tren-Lys-Cam RP-HPLC of Tren-Dab-Cam

RP-HPLC of Tren-Lys-Pam RP-HPLC of Tren-Lys

Ac-Cam

RP-HPLC of Tren-Cam

10

Table S2. ESI Mass Spectrometry (MS) and ESIMS/MS Fragmentation of Tren-based

Siderophore Analogs. See Fig. S7 for ESI-MS and ESI-MS/MS spectra.

Tren-Lys-Cam Tren-LysAc-Cam Tren-Cam

Fragment [ M + H ] + Fragment [ M + H ] + Fragment [ M + H ] +

Parent Ion 939.5 Parent Ion 1065.44 Parent Ion 555.2

loss of DHB 803.48 loss of DHB 929.49 loss of DHB 419.2

loss of DHB-Lys 675.39 loss of DHB-LysAc 759.4 loss of DHB-LysAc -

loss of single arm 632.35 loss of single arm 716.35 loss of single arm 376.15

loss of DHB-Lys + DHB 539.38 loss of DHB-LysAc + DHB 623.39 loss of DHB-LysAc + DHB -

loss of 2x DHB-Lys 411.28 loss of 2x DHB-LysAc 453.26 loss of 2x DHB-LysAc -

single arm 308.16 single arm 350.17 single arm 180.07

DHB-Lys 265.13 DHB-LysAc 307.14 DHB-LysAc -

Lys 129.1 LysAc - LysAc -

Tren-Dab-Cam* Tren-Lys-Pam* Tren-Lys-Bam*

Fragment [ M + H ] + Fragment [ M + H ] + Fragment [ M + H ] +

Parent Ion 855.31 Parent Ion 891.47 Parent Ion 843.53

loss of DHB 719.26 loss of Phenol 771.46 loss of Benzyl 739.51

loss of DHB-Dab 619.25 loss of Phenol-Lys 643.37 loss of Benzyl-Lys 611.41

loss of single arm - loss of single arm 600.33 loss of single arm 568.37

loss of DHB-Dab + DHB - loss of Phenol-Lys + Phenol 523.35 loss of Benzyl-Lys + Benzyl 507.38

loss of 2x DHB-Dab 383.19 loss of 2x Phenol-Lys 395.26 loss of 2x Benzyl-Lys 379.29

single arm - single arm 292.15 single arm 276.18

DHB-Dab 237.06 Phenol-Lys 249.12 Benzyl-Lys 233.13

Dab 101.06 Lys 129.1 Lys -

* Only ESI-MS was carried out

Fig. S6. Typical Fragmentation of the Tren-based Siderophore Analogs. The

fragmentation typically occurs at amide bonds between the 2,3-dihydroxybenzamide and

lysine, at the amide bonds between the lysine and the Tren scaffold, and at the center

nitrogen of the Tren scaffold.

11

Fig. S7. ESI Mass Spectrometry (MS) and ESIMS/MS of the Tren-based Siderophore

Analogs. A. ESI-MS of Tren-Lys-Cam. B. ESI-MS/MS of Tren-Lys-Cam. C. ESI-MS

of Tren-Dab-Cam. D. ESI-MS of Tren-Lys-Pam. E. ESI-MS of Tren-Lys-Bam. F. ESI-

MS of Tren-LysAc-Cam. G. ESI-MS/MS of Tren-LysAc-Cam.

Chrys-TREN large post-HPLC

m/z100 200 300 400 500 600 700 800 900 1000 1100

%

0

100

BUT031314MI 308 (5.833) Sm (SG, 2x4.00); Cm (304:350) TOF MS ES+ 6.39e3470.24

129.10

411.27137.02368.23

939.48

675.37

470.73

481.22

632.33

676.37

677.39

940.49

961.47

962.47

Chrys-TREN large scale-up post-HPLC MSMS 939

m/z100 200 300 400 500 600 700 800 900 1000 1100

%

0

100

BUT031314MB2 21 (0.404) Sm (SG, 2x4.00); Cm (13:41) TOF MSMS 939.00ES+ 267939.50

675.39

308.16

265.13

632.35368.23

411.28 539.38

676.39

803.49

940.50

941.53

942.54

TREN-Dab-CAM HPLC Ultra test

m/z100 200 300 400 500 600 700 800 900 1000 1100

%

0

100

BUT110414MA 49 (0.917) Sm (SG, 2x4.00); Cm (49:60) TOF MS ES+ 2.49e3855.31

383.19

237.06

137.00366.17

365.19

619.25

428.14 602.23

585.20

620.26

633.26

856.31

857.31

858.31

TREN-Lys-PAM HPLC Peak 5

m/z100 200 300 400 500 600 700 800 900 1000

%

0

100

BUT102814MD 61 (1.165) AM (Top,4, Ht,5000.0,0.00,1.00); Sm (Mn, 2x1.00); Sb (1,40.00 ); Sm (SG, 2x4.00); Cm (61:85)3.87e3446.23

249.12

891.47643.37446.74

644.37

892.47

TREN-Lys-BAM HPLC Peak 3

m/z100 200 300 400 500 600 700 800 900 1000

%

0

100

BUT072214MA 141 (2.679) AM (Top,4, Ht,5000.0,0.00,1.00); Sm (Mn, 2x1.00); Sb (1,40.00 ); Sm (SG, 2x4.00); Cm (126:141)3.60e3843.53

611.41

422.26

233.13

379.29336.25

422.76

423.27568.37

612.41

613.42739.51

844.54

845.55

871.57

(Ac)3TREN purified

m/z100 200 300 400 500 600 700 800 900 1000 1100 1200 1300

%

0

100

BUT022014MY 86 (1.636) Sm (SG, 2x4.00); Cm (40:98) TOF MS ES+ 5.63e31087.51

1065.52

555.24

487.23118.98555.74

781.39556.24

1088.51

1089.52

1103.48

1109.50

(Ac)3TREN purified MSMS 1065

m/z100 200 300 400 500 600 700 800 900 1000 1100 1200 1300

%

0

100

BUT022014MY1 44 (0.844) Sm (SG, 2x4.00); Sm (SG, 2x4.00); Cm (33:126) TOF MSMS 1065.00ES+ 57.21065.47

759.41

350.17

126.09

716.35

410.25

760.40

1064.36761.41

1066.42

1067.46

A B

C D

E F

G

12

Table S3. NMR Data for Tren-Lys-Cam, Tren-Dab-Cam, and Tren-Lys-Pam. NMR (1H

on a Varian Unity Inova 600 MHz spectrometer and 13C on a Varian Unity Inova 500

MHz spectrometer) was taken in D2O or DMSO. See Figs. S8-S11 for NMR spectra.

Tren-Lys-Cam Tren-Dab-Cam Tren-Lys-Pam

Position* δC δH (J in Hz) δC δH (J in Hz) δC δH (J in Hz)

TREN

1 Obscured by

DMSO 3.67, m

Obscured by DMSO

Obscured by Water

Obscured by DMSO

Obscured by Water

2 38.53, CH2 3.46, m 36.33 , CH2 Obscured by

Water 38.49, CH2

Obscured by Water

DHBA

3 168.77, C - 168.370, C - 166.59, C -

4 115.67, C - 115.47, C - 135.19, C -

5 148.41, C - 147.88, C - 114.46, CH 7.30, s

6 145.90, C - 145.59, C - 157.13, C -

7 118.63, CH 6.93, d (7.31) 118.41, CH 6.95, d (7.49) 117.98, CH 7.35, m

8 118.19, CH 6.70, t (7.74) 118.05, CH 6.71, t (7.65) 129.01, CH 7.23, t (7.65)

9 117.94, CH 7.09, d (7.88) 117.77, CH 7.37, d (8.02) 118.17, CH 6.93, d (7.68)

Lysine

10 172.44, C - 172.31, C - 172.39, C -

11 52.72, CH 4.29, m 50.28, CH 4.54, m 53.27, CH 4.35, m

12 30.74, CH2 1.65, m 25.25, CH2 2.04, m 30.51, CH2 1.75, m

13 22.44, CH2 1.30, m 29.56, CH2 2.85, m 22.60, CH2 1.35, m

14 26.48, CH2 1.57, m - - 26.47, CH2 1.55, m

15 Obscured by

DMSO 2.86, t (7.66) - -

Obscured by DMSO

2.75, m

16 - - - -

17 - - - - -

* Positions numbers are indicated in Fig. S4.

13

Table S4. NMR Data for Tren-Lys-Bam, Tren-LysAc-Cam, and Tren-Cam, NMR (1H on

a Varian Unity Inova 600 MHz spectrometer and 13C on a Varian Unity Inova 500 MHz

spectrometer) was taken in D2O or DMSO. See Figs. S12-S13 for NMR spectra. The

nmr spectra for Tren-Cam is in agreement with previously published data (16), and thus

not included herein.

Tren-Lys-Bam Tren-LysAc-Cam Tren-Cam

Position* δC δH (J in Hz) δC δH (J in Hz) δC δH (J in Hz)

TREN

1 Obscured by

DMSO 3.62, m

Obscured by DMSO

Obscured by Water

Obscured by DMSO

3.84, m

2 38.66, CH2 3.47, m 38.74, CH2 Obscured by

Water Obscured by

DMSO 3.69, m

DHBA

3 166.71, C - 169.52, C - 169.86, C -

4 133.89, C - 116.17, C - 115.12, C -

5 127.61, CH 7.69, d (7.30) 149.08, C - 148.93, C -

6 128.17, CH 7.45, t (7.65) 146.49, C - 146.04, C -

7 131.41, CH 7.57, t (7.19) 119.28, CH 6.93, d (7.65) 118.85, CH 6.87, d (7.65)

8 128.17, CH 7.45, t (7.65) 118.81, CH 6.70, t (7.18) 118.02, CH 6.57, t (7.43)

9 127.61, CH 7.69, d (7.30) 118.56, CH 7.42, d (7.30) 117.48, CH 6.89, d (7.65)

Lysine

10 172.54, C - 169.44, C - - -

11 53.52, CH 4.31, m 53.59, CH 4.40, m - -

12 30.68, CH2 1.75, m 31.40, CH2 1.73, m - -

13 22.80, CH2 1.36, m 23.02, CH2 1.29, m - -

14 26.67, CH2 1.60, m 29.27, CH2 1.37, m - -

15 Obscured by

DMSO 2.90, t (7.30)

Obscured by DMSO

2.99, m - -

16 - - 169.33, C - - -

17 - - 23.52, CH3 1.76, s - -

* Positions numbers are indicated in Fig. S4.

14

Fig. S8. NMR data for Tren-Lys-Cam. A. 1H NMR Data for Tren-Lys-Cam. NMR (600

MHz) in D2O with enlarged aromatic region. B. 13C NMR Data for Tren-Lys-Cam.

NMR (500 MHz) in DMSO. Trifluoroacetic acid (TFA) originates from RP-HPLC

purification. The 13C resonance for C10 is evident in the 2D nmr spectrum at 172.4 ppm,

Fig. S9.

ChrysTREN_small_post_HPLC_031514

7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

Chemical Shift (ppm)

0

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

Norm

alized Inte

nsity

H2O

7 9 8 11

15

12

14

13 1

2

TRENLysCAM_DMSO_13C_012615

180 160 140 120 100 80 60 40 20 0

Chemical Shift (ppm)

0

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

No

rma

lize

d I

nte

nsity

3

5

6

7,8,9

4 11

DMSO, 15, 1

12

14

13

2

TFA

9 7 8 A

B

15

Fig. S9. 1H-13C HMBC NMR for Tren-Lys-Cam. NMR (600 MHz) in DMSO.

Enlarged Regions of the 1H-13C HMBC NMR for Tren-Lys-Cam are in bottom panel.

The spectrum is annotated with the correlations between specific carbons and hydrogens.

12 11 10 9 8 7 6 5 4 3 2 1 0

F2 Chemical Shift (ppm)

0

20

40

60

80

100

120

140

160

180

200

F1

Che

mic

al S

hift

(ppm

)

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5

F2 Chemical Shift (ppm)

104

112

120

128

136

144

152

160

168

176

184

F1

C

hem

ica

l S

hift (p

pm

)

H9-C4

H9-C7,C8

H7-C4 H8-C4

H7-C9,C8 H8-C9,C7

H9-C5

H9-C6 H7-C6

H7-C5

H8-C5

H8-C6

H9-C3 H8-C3

H11-C3

H11-C10

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0

F2 Chemical Shift (ppm)

5

10

15

20

25

30

35

40

45

50

55

F1

C

hem

ica

l S

hift (p

pm

)H11-C12

H11-C13 H15-C13 H12-C13

H15-C12

DMSO

H12-C11

H14-C15

H14-C12

H14-C13

H13-C14

H13-C12

H13-C15

H13-C11

H12-C14

16

Fig. S10. NMR data for Tren-Dab-Cam. A. 1H NMR Data for Tren-Dab-Cam. NMR

(600 MHz) in DMSO with enlarged aromatic region. B. 13C NMR Data for Tren-Dab-

Cam. NMR (500 MHz) in DMSO. Trifluoroacetic acid (TFA) originates from RP-HPLC

purification.

TREN-Dab-CAM_pure_DMSO_110614

7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

Chemical Shift (ppm)

0.05

0.10

0.15

No

rma

lize

d I

nte

nsity

9 7 8

11

H2O, 1, 2

13

DMSO

12

Impurity

TRENDabCAM_DMSO_C13_012315

180 160 140 120 100 80 60 40 20 0

Chemical Shift (ppm)

0

0.005

0.010

0.015

0.020

No

rma

lize

d I

nte

nsity

10 3

5

6

7,8,9

4

11

DMSO, 15, 1

2

13

12

TFA

Impurity

Impurity

9 7 8

A

B

17

Fig. S11. NMR data for Tren-Lys-Pam. A. 1H NMR Data for Tren-Lys-Pam. NMR

(600 MHz) in DMSO with enlarged aromatic region. B. 13C NMR Data for Tren-Lys-

Pam. NMR (500 MHz) in DMSO. Trifluoroacetic acid (TFA) originates from RP-HPLC

purification.

TREN_Lys_PAM_pure_DMSO_110114

7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

Chemical Shift (ppm)

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

No

rma

lize

d I

nte

nsity

7

9

11

1, 2

H2O

DMSO

15

12

14

13

TRENLysPAM_DMSO_13C_012515

180 160 140 120 100 80 60 40 20 0

Chemical Shift (ppm)

0

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

No

rma

lize

d I

nte

nsity

10

3

6

4

8 9,7 5

11

DMSO, 15, 1

2

12

14

13

TFA

7

5 8 9

5 8

A

B

18

Fig. S12. NMR data for Tren-Lys-Bam. A. 1H NMR Data for Tren-Lys-Bam. NMR

(600 MHz) in D2O with enlarged aromatic region. B. 13C NMR Data for Tren-Lys-Bam.

NMR (500 MHz) in DMSO. Trifluoroacetic acid (TFA) originates from RP-HPLC

purification.

TREN-Lys-BAM_pure_072414

8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0

Chemical Shift (ppm)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.11

No

rma

lize

d I

nte

nsity

5+9 8+6

7

H2O

11

2

1

15

12

14

13

TrenLysBam_13C_DMSO_012815

180 160 140 120 100 80 60 40 20 0

Chemical Shift (ppm)

0.05

0.10

0.15

0.20

No

rma

lize

d I

nte

nsity

10

3

4

7

6+8 7+9

11

DMSO, 15, 1

2

12

14

13

TFA

5+9

7

8+6 A

B

19

Fig. S13. NMR data for Tren-LysAc-Cam. A. 1H NMR Data for Tren-LysAc-Cam. NMR

(600 MHz) in DMSO with enlarged aromatic region. B. 13C NMR Data for Tren-LysAc-

Cam. NMR (500 MHz) in DMSO. Trifluoroacetic acid (TFA) originates from RP-

HPLC purification.

Ac3_TREN_DMSO

7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5

Chemical Shift (ppm)

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

No

rma

lize

d I

nte

nsity

11 1,2

15

DMSO 17

12

14,13

9 7 8

TRENLysAcCAM_DMSO_C13_012215

160 140 120 100 80 60 40 20 0

Chemical Shift (ppm)

0

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

0.055

0.060

0.065

No

rma

lize

d I

nte

nsity 3,10,16

5 6

7,8,9

4

11

DMSO, 15, 1

12

2

14

17,13

H2O

9 7 8

A

B

20

Fig. S14. SFA force-distance interaction of tren-homologs. A. SFA force-distance

interaction for the TDC-mediated adhesion between two mica surfaces in buffer (50 mM

acetate buffer + 150 mM KNO3) at pH 3.3. 10-9 moles of TDC were injected into the gap

solution between the surfaces. The inset displays the molecular structure of TDC. B.

SFA force-distance interaction for the TLP-mediated adhesion between two mica

surfaces in buffer (50 mM acetate buffer + 150 mM KNO3) at pH 3.3. 10-9 moles of TLP

were injected into the gap solution between the surfaces. The inset displays the molecular

structure of TLP. C. SFA force-distance interaction for the TLB-mediated adhesion

between two mica surfaces in buffer (50 mM acetate buffer + 150 mM KNO3) at pH 3.3.

Measurements are shown for 10-9 moles and 10-8 moles of TLB injected into the gap

solution between the surfaces. At 10-9 moles, TLB does not significantly adsorb onto

mica, and the resulting force-distance profile appears the same as for a buffer solution in

the absence of siderophore. However, at 10-8 moles, TLB does adsorb to the mica surface,

resulting in a decrease in the compressed film thickness and an increase in the measured

adhesion force. The inset displays the molecular structure of TLB.

A

B C

21

Fig. S15. SFA force-distance interactions of modified amine tren-homologs. A. SFA

force-distance profile for two mica surfaces interacting in a buffer at pH 3.3 (50 mM

acetate buffer + 150 mM KNO3) with an additional 10-9 moles of TLAcC injected into the

solution between the surfaces. TLAcC does not significantly adsorb onto mica, and the

resulting force-distance profile appears the same as for a buffer solution in the absence of

siderophore. The inset displays the molecular structure of TLAcC. B. SFA force-distance

profile for two mica surfaces interacting in a buffer at pH 3.3 (50 mM acetate buffer +

150 mM KNO3) with an additional 10-9 moles of Tren-Cam injected into the solution

between the surfaces. TC does not significantly adsorb onto mica, and the resulting force-

distance profile appears the same as for a buffer solution in the absence of siderophore.

The inset displays the molecular structure of TC. C. SFA force-distance profile for two

mica surfaces interacting in de-ionized water (pH 5.5) and with an additional 10-9 moles

of Tren-Cam injected into the water between the surfaces. In pure de-ionized water, a

long-ranged electrostatic repulsion between the mica surfaces is measured on approach;

upon separation of the surfaces, a moderate adhesion force is measured due to van der

Waals forces between the surfaces. After injecting 10-9 moles of TC into the water

between the surfaces, the siderophore molecules adsorb, resulting in a 1-2 nm thick TC

film (multilayers). The TC film mediates a moderate adhesive force between the surfaces

(that cannot be described by van der Waals forces) and notable bridging between the

surfaces is observed during separation. The inset displays the molecular structure of TC.

A B

C