Supplementary Materials for

Transcription

1 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. (J.H.W.); (J.N.I.); (A.B.) This PDF file includes: Materials and Methods Supplementary Text Figs. S1 to S15 Tables S1 to S4 Published 7 August 215, Science 349, 628 (215) DOI: /science.aab556

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 53 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 1 resin (1-2 mesh, sodium form, Bio-Rad) using the batch method. 3 ml of 29.5 C ±.1 of 5 mm buffer was introduced to the reaction chamber (also held at 29.5 C ±.1). Phosphate buffer was used for ph 7.5 and CAPSO buffer was used for ph 1.. The reaction chamber was sealed with the Clark electrode, removing all non-dissolved air from the system and allowed to equilibrate for 5 1 minutes. Kinetics experiments began upon the injection of 1 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.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 1 st 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 (3). 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, TL Ac C, 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 2

3 with 5% 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. TL Ac C 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 2, 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 cm 2 ), 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 (~5 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 (~1 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 2 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 5 mm acetate + 15 mm KNO3 buffer solution for ph 3.3 and 5.5, and (ii) a 5 mm phosphate + 15 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 1.. Increasing the concentration of the 3

4 catechol compounds increases the autoxidation rate, as is expected for pseudo 1 st 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, 1.6 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 <.5. 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 <.5 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 <.5 (with the exception of ph 5.5 ph 7.5 at 6 minutes of contact time). Additionally, the null hypothesis is accepted for ph 3.3 ph 5.5 (except at 6 minutes of contact time), indicating that there is no statistical difference between the TLC-mediated adhesion at ph 3.3 and ph

5 A 3.3 mm 2,3-DHBA 3.3 mm 3,4-DHBA 3.3 mm 4-MC 3.3 mm Catechol 5 mm ph 1. CAPSO Buffer at 29.5 C B 5 mm ph 7.5 Phosphate Buffer at 29.5 C Fig. S1. Autoxidation of 4-MC, Catechol, 2,3-DHBA and 3,4-DHBA at ph 1 (A.) and ph 7.5 (B.). A. 3.3 mm of each catechol compound, ph 1. in 5 mm CAPSO buffer at 29.5 C. B. 1. mm 4-MC, catechol and 3.3 mm 2,3-DHBA, 3,4-DHBA and catechol at ph 7.5 in 5 mm phosphate buffer at 29.5 C. Oxygen consumption was monitored using a Clark electrode. 5

6 Fig. S2. Synthesis Scheme for Tren-Lys-Cam. Reaction Conditions: (a) KOH,, 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. 6

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 ~5 μl. Error bars for the 1 min data points (±12 mn m -1 ) have been omitted for visual clarity. The anionic mica surfaces used in these experiments are ~ 1 cm 2 and contain ~ 2.1 x 1 14 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 1 14 molecules) is roughly equivalent to the number of negatively charged mica sites in the system (~ 4.2 x 1 14 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.5 values are highlighted in grey. See supplemental text for description. Contact Time 2 min 1 min 3 min 6 min ph 3.3 ph 5.5 P =.1274 P =.952 P =.6216 P =.158 ph 3.3 ph 7.5 P =.1 P =.62 P =.14 P =.129 ph 5.5 ph 7.5 P =.2 P =.492 P =.135 P =

8 Fig. S4. Synthetic Tren-based Siderophore Analogs. Numbered carbons correspond to NMR data in Figs. S1-S13 and Tables S3-S4. 8

9 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 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 25 mm, Vydac). A gradient elution was performed from 1% nanopure H2O with.5% trifluoroacetic acid to 1% MeOH with.5% trifluoroacetic acid. The rate of gradient transition was optimized for each individual compound. The eluent was monitored at 215 nm. 9

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-Lys Ac -Cam Tren-Cam Fragment [ M + H ] + Fragment [ M + H ] + Fragment [ M + H ] + Parent Ion Parent Ion Parent Ion loss of DHB loss of DHB loss of DHB loss of DHB-Lys loss of DHB-Lys Ac loss of DHB-Lys Ac - loss of single arm loss of single arm loss of single arm loss of DHB-Lys + DHB loss of DHB-Lys Ac + DHB loss of DHB-Lys Ac + DHB - loss of 2x DHB-Lys loss of 2x DHB-Lys Ac loss of 2x DHB-Lys Ac - single arm single arm single arm 18.7 DHB-Lys DHB-Lys Ac DHB-Lys Ac - Lys Lys Ac - Lys Ac - Tren-Dab-Cam* Tren-Lys-Pam* Tren-Lys-Bam* Fragment [ M + H ] + Fragment [ M + H ] + Fragment [ M + H ] + Parent Ion Parent Ion Parent Ion loss of DHB loss of Phenol loss of Benzyl loss of DHB-Dab loss of Phenol-Lys loss of Benzyl-Lys loss of single arm - loss of single arm 6.33 loss of single arm loss of DHB-Dab + DHB - loss of Phenol-Lys + Phenol loss of Benzyl-Lys + Benzyl loss of 2x DHB-Dab loss of 2x Phenol-Lys loss of 2x Benzyl-Lys single arm - single arm single arm DHB-Dab Phenol-Lys Benzyl-Lys Dab 11.6 Lys 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. 1

11 Chrys-TREN large post-hplc BUT31314MI 38 (5.833) Sm (SG, 2x4.); Cm (34:35) A Chrys-TREN large scale-up post-hplc MSMS 939 BUT31314MB2 TOF MS ES+ 21 (.44) Sm (SG, 2x4.); Cm (13:41) 6.39e3 1 B TOF MSMS 939.ES % % TREN-Dab-CAM HPLC Ultra test m/z m/z BUT11414MA (.917) Sm 3(SG, 2x4.); 4 Cm (49:6) TREN-Lys-PAM 11 1TOF MS 2 ES+ 3HPLC 4Peak e3 1 BUT12814MD 61 (1.165) AM (Top,4, Ht,5.,.,1.); Sm (Mn, 2x1.); Sb (1,4. ); Sm (SG, 2x4.); Cm (61:85) e3 1 % C % TREN-Lys-BAM HPLC Peak 3 m/z m/z BUT72214MA (2.679) 3 AM (Top,4, 4 Ht,5.,.,1.); Sm 6 (Mn, 2x1.); 7Sb (1,4. 8); Sm (SG, 92x4.); Cm 1(126:141) (Ac)3TREN 11 1 purified BUT2214MY 3.6e3 86 (1.636) Sm (SG, 2x4.); Cm (4:98) TOF MS ES e3 1 E D F % % (Ac)3TREN purified MSMS 165 m/z BUT2214MY (.844) Sm (SG, 8 2x4.); 9 Sm (SG, 2x4.); 1 Cm (33:126) TOF 7 MSMS 165.ES G m/z % m/z 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-Lys Ac -Cam. 11

12 Table S3. NMR Data for Tren-Lys-Cam, Tren-Dab-Cam, and Tren-Lys-Pam. NMR ( 1 H on a Varian Unity Inova 6 MHz spectrometer and 13 C on a Varian Unity Inova 5 MHz spectrometer) was taken in D2O or. See Figs. S8-S11 for NMR spectra. Tren-Lys-Cam Tren-Dab-Cam Tren-Lys-Pam Position* δ C δ H (J in Hz) δ C δ H (J in Hz) δ C δ H (J in Hz) TREN , m Water Water , CH , m 36.33, CH 2 Water 38.49, CH 2 Water DHBA , C , C , C , C , C , C , C , C , CH 7.3, s , C , C , C , CH 6.93, d (7.31) , CH 6.95, d (7.49) , CH 7.35, m , CH 6.7, t (7.74) 118.5, CH 6.71, t (7.65) 129.1, CH 7.23, t (7.65) , CH 7.9, d (7.88) , CH 7.37, d (8.2) , CH 6.93, d (7.68) Lysine , C , C , C , CH 4.29, m 5.28, CH 4.54, m 53.27, CH 4.35, m , CH , m 25.25, CH 2 2.4, m 3.51, CH , m , CH 2 1.3, m 29.56, CH , m 22.6, CH , m , CH , m , CH , m , t (7.66) , m * Positions numbers are indicated in Fig. S4. 12

13 Table S4. NMR Data for Tren-Lys-Bam, Tren-Lys Ac -Cam, and Tren-Cam, NMR ( 1 H on a Varian Unity Inova 6 MHz spectrometer and 13 C on a Varian Unity Inova 5 MHz spectrometer) was taken in D2O or. 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-Lys Ac -Cam Tren-Cam Position* δ C δ H (J in Hz) δ C δ H (J in Hz) δ C δ H (J in Hz) TREN , m Water 3.84, m , CH , m 38.74, CH 2 Water 3.69, m DHBA , C , C , C , C , C , C , CH 7.69, d (7.3) 149.8, C , C , CH 7.45, t (7.65) , C , C , CH 7.57, t (7.19) , CH 6.93, d (7.65) , CH 6.87, d (7.65) , CH 7.45, t (7.65) , CH 6.7, t (7.18) 118.2, CH 6.57, t (7.43) , CH 7.69, d (7.3) , CH 7.42, d (7.3) , CH 6.89, d (7.65) Lysine , C , C , CH 4.31, m 53.59, CH 4.4, m , CH , m 31.4, CH , m , CH , m 23.2, CH , m , CH 2 1.6, m 29.27, CH , m , t (7.3) 2.99, m , C , CH , s - - * Positions numbers are indicated in Fig. S4. 13

14 ChrysTREN_small_post_HPLC_ A H 2 O Normalized Intensity TRENLysCAM 13C_ Chemical Shift (ppm).35.3 B, 15, 1 Normalized Intensity TFA 6 5 7,8, Chemical Shift (ppm) Fig. S8. NMR data for Tren-Lys-Cam. A. 1 H NMR Data for Tren-Lys-Cam. NMR (6 MHz) in D2O with enlarged aromatic region. B. 13 C NMR Data for Tren-Lys-Cam. NMR (5 MHz) in. Trifluoroacetic acid (TFA) originates from RP-HPLC purification. The 13C resonance for C1 is evident in the 2D nmr spectrum at ppm, Fig. S9. 14

15 F1 Chemical Shift (ppm) F2 Chemical Shift (ppm) H7-C4 H9-C4 H8-C4 H9-C7,C8 H8-C9,C7 H7-C9,C8 H9-C6 H7-C6 H8-C6 H9-C5 H8-C5 H7-C5 H9-C3 H8-C3 H11-C3 H11-C F2 Chemical Shift (ppm) F1 Chemical Shift (ppm) H11-C13 H15-C13 H12-C13 H11-C12 H15-C12 H14-C12 H12-C11 H14-C13 H12-C14 H13-C14 H13-C12 H14-C15 H13-C15 H13-C F2 Chemical Shift (ppm) Fig. S9. 1 H- 13 C HMBC NMR for Tren-Lys-Cam. NMR (6 MHz) in. Enlarged Regions of the 1 H- 13 C HMBC NMR for Tren-Lys-Cam are in bottom panel. The spectrum is annotated with the correlations between specific carbons and hydrogens F1 Chemical Shift (ppm) 15

16 TREN-Dab-CAM_pure A H 2 O, 1, Normalized Intensity Impurity Chemical Shift (ppm).2 TRENDabCAM C13_12315 B, 15, ,8,9 2 Normalized Intensity TFA Impurity Impurity Chemical Shift (ppm) Fig. S1. NMR data for Tren-Dab-Cam. A. 1 H NMR Data for Tren-Dab-Cam. NMR (6 MHz) in with enlarged aromatic region. B. 13 C NMR Data for Tren-Dab- Cam. NMR (5 MHz) in. Trifluoroacetic acid (TFA) originates from RP-HPLC purification. 16

17 TREN_Lys_PAM_pure Normalized Intensity A , 2 H 2 O TRENLysPAM 13C_12515 Chemical Shift (ppm).45.4 B, 15, 1.35 Normalized Intensity TFA 4 8 9, Chemical Shift (ppm) Fig. S11. NMR data for Tren-Lys-Pam. A. 1 H NMR Data for Tren-Lys-Pam. NMR (6 MHz) in with enlarged aromatic region. B. 13 C NMR Data for Tren-Lys- Pam. NMR (5 MHz) in. Trifluoroacetic acid (TFA) originates from RP-HPLC purification. 17

18 .11 TREN-Lys-BAM_pure_72414 Normalized Intensity A H 2 O TrenLysBam_13C Chemical Shift (ppm) B, 15, 1.15 Normalized Intensity TFA Chemical Shift (ppm) Fig. S12. NMR data for Tren-Lys-Bam. A. 1 H NMR Data for Tren-Lys-Bam. NMR (6 MHz) in D2O with enlarged aromatic region. B. 13 C NMR Data for Tren-Lys-Bam. NMR (5 MHz) in. Trifluoroacetic acid (TFA) originates from RP-HPLC purification. 18

19 Ac3_TREN_ A 17 Normalized Intensity H 2 O ,2 14,13 12 Normalized Intensity TRENLysAcCAM C13_12215 Chemical Shift (ppm) B, 15, 1 17, ,1,16 2 7,8, Chemical Shift (ppm) Fig. S13. NMR data for Tren-Lys Ac -Cam. A. 1 H NMR Data for Tren-Lys Ac -Cam. NMR (6 MHz) in with enlarged aromatic region. B. 13 C NMR Data for Tren-Lys Ac - Cam. NMR (5 MHz) in. Trifluoroacetic acid (TFA) originates from RP- HPLC purification. 19

20 A B C 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 (5 mm acetate buffer + 15 mm KNO3) at ph 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 (5 mm acetate buffer + 15 mm KNO3) at ph 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 (5 mm acetate buffer + 15 mm KNO3) at ph 3.3. Measurements are shown for 1-9 moles and 1-8 moles of TLB injected into the gap solution between the surfaces. At 1-9 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 1-8 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. 2

21 A B C 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 (5 mm acetate buffer + 15 mm KNO3) with an additional 1-9 moles of TL Ac C injected into the solution between the surfaces. TL Ac C 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 TL Ac C. B. SFA force-distance profile for two mica surfaces interacting in a buffer at ph 3.3 (5 mm acetate buffer + 15 mm KNO3) with an additional 1-9 moles of Tren-Cam injected into the solution between the surfaces. TC does not significantly adsorb onto mica, and the resulting forcedistance 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 1-9 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 1-9 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. 21