Phosphopeptide Adsorption on Zirconium Phosphate and Zirconium Phosphonate Solids

MISSING IMAGE

Material Information

Title:
Phosphopeptide Adsorption on Zirconium Phosphate and Zirconium Phosphonate Solids
Physical Description:
1 online resource (117 p.)
Language:
english
Creator:
Pollard, Emily J
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
TALHAM,DANIEL R
Committee Co-Chair:
FANUCCI,GAIL E
Committee Members:
POWELL,DAVID HINTON
POLFER,NICOLAS CAMILLE
GOWER,LAURIE B

Subjects

Subjects / Keywords:
isotherm -- phosphate -- phosphonate -- phosphopeptide -- zirconium
Chemistry -- Dissertations, Academic -- UF
Genre:
Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Phosphorylation is one of the most common post-translational modification of proteins, and it serves a regulatory or signaling function in numerous biological processes such as metabolism, gene expression, and differentiation. Studying the tryptic digests of proteins is an important aspect of understanding these processes. Mass spectrometry is the tool of choice for the study of phosphopeptides, but the substoichiometric amounts of phosphopeptides, the poor ionization of them in typical processes, and signal suppression from non-phosphopeptides necessitates an enrichment step before mass spectrometry analysis.  Typical phosphopeptide enrichment uses immobilized metal affinity chromatography (IMAC) or metal oxide affinity chromatography (MOAC) to separate phosphopeptides for focused study. To successfully employ these systems, it is often necessary to pre-treat protein digests or work near pH 3 in efforts to minimize competitive, acidic non-phosphopeptide binding. Even still, enrichment is often incomplete either losing some phosphopeptides or retaining some non-phosphopeptides. Increasing and applying an understanding of the enrichment mechanism on IMAC or MOAC or developing a new material with improved properties could improve enrichment efforts.  Titania, a MOAC material, is perhaps the most common phosphopeptide enrichment material in use today, but its enrichment mechanism is not well understood. Zirconium phosphate has been reported once as a viable enrichment material, but little is known about its enrichment mechanism. Zirconium phosphonate particles have not been previously reported for this purpose. This work discusses research efforts to probe the enrichment mechanism on titania and to develop and explore new enrichment materials derived from zirconium phosphate and zirconium phosphonate. It will include quantitative and semi-quantitative comparisons of the binding of model peptides to titania, zirconium phosphate, and zirconium phosphonate.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Emily J Pollard.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: TALHAM,DANIEL R.
Local:
Co-adviser: FANUCCI,GAIL E.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-06-30

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045754:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 PHOSPHOPEPTIDE ADSORPTION ON ZIRCONIUM PHOSPHATE AND ZIRCONIUM PHOSPHONATE SOLIDS By EMILY J. POLLARD A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Emily J. Pollard

PAGE 3

3 To my family, especially my husband, who made this possible, and all for the glory of God

PAGE 4

4 ACKNOWLEDGMENTS I would l ike to express my gratitude to my family for all of their support throughout my graduate studies. Of special note is my husband whose day in and day out efforts made my degree progression possible. My parents were exceptionally helpful during the busiest times of my graduate career traveling long distances to lend a hand. My children, who bring so much joy to my life, always turned my bad days into good days with their bright smiles, contagious giggles, and amusing antics. Many thanks go out to my advis or and my research group for their instruction and aide in my research; I could not have come this far without them. I would also like to thank the Air Force Institute of Technology and the National Science Foundation (NSF) who funded my graduate work and the NSF award 10400016, which provided our department with the MALDI TOF TOF referred to in this work. Last, but not least, I would like to say thank you to all of my friends and extended family members both near and far who gave me moral and spiritual s upport thought my time in graduate school. DISCLAIMER: The views expressed in this document are those of the author and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Phosphopeptide Significance and the Need for Enrichment ................................ ... 14 Common Enr ichment Materials ................................ ................................ ............... 14 Zirconium Phosphate and Zirconium Phosphonates ................................ ........... 16 Isoelectric Points ................................ ................................ ................................ ..... 19 Adsorption Isotherms ................................ ................................ .............................. 21 This Work ................................ ................................ ................................ ................ 23 2 ISOELECRIC POINTS OF ENRICHMENT MATERIALS ................................ ........ 30 Isoelectric Points ................................ ................................ ................................ ..... 30 Experimental Section ................................ ................................ .............................. 32 Synthesis of Materials ................................ ................................ ...................... 33 Characterization of Materials ................................ ................................ ............ 34 Zeta Potential Titrations ................................ ................................ .................... 35 Results and Discussion ................................ ................................ ........................... 36 Characterization of Materials ................................ ................................ ............ 36 Isoelectric Point Determination ................................ ................................ ......... 37 Summary ................................ ................................ ................................ ................ 40 3 ADSORPTION STUDIES ................................ ................................ ........................ 46 Separation and Adsorption ................................ ................................ ..................... 46 Experimental Section ................................ ................................ .............................. 48 Materials ................................ ................................ ................................ ........... 48 Peptide Preparation ................................ ................................ .......................... 48 Peptide Adsorption ................................ ................................ ........................... 49 Adsorption onto TiO 2 ................................ ................................ .................. 49 ZrP ................................ ................................ ................ 49 Adsorption onto Zr OBP ................................ ................................ ............. 49 Peptide mass balance by UV absorbance spectroscopy ........................... 50 Results ................................ ................................ ................................ .................... 51 Adsorption onto TiO 2 ................................ ................................ ........................ 51

PAGE 6

6 ZrP ................................ ................................ ...................... 52 Adsorption onto Zr OBP ................................ ................................ ................... 53 Discussion ................................ ................................ ................................ .............. 53 Binding Capacity ................................ ................................ .............................. 54 Binding Efficiency ................................ ................................ ............................. 55 Selectivity ................................ ................................ ................................ ......... 56 Bindin g Mechanisms ................................ ................................ ........................ 57 Summary ................................ ................................ ................................ ................ 60 4 EXPLORATORY WORK INTO DESORPTION AND ENRICHMENT ..................... 67 Overview ................................ ................................ ................................ ................. 67 Experimental Section ................................ ................................ .............................. 67 ZrP ................................ ................................ ............... 68 Determining the limit of mass dependence ................................ ................ 68 Testing enrichment independent of adsorbent mass ................................ 68 Mock Enrichment on Titania ................................ ................................ ............. 69 Mass Spectrometry after Mock Enrichment ................................ ...................... 69 Enrichment an Casein Digest ................................ .. 70 Results and Discussion ................................ ................................ ........................... 70 Mass Independence ................................ ................................ ......................... 70 Mock Enrichment with UV Detection ................................ ................................ 71 ZrP T as the adsorbent ................................ ................................ ............. 71 ZrP T as the adso rbent ................................ ........... 73 TiO 2 as the adsorbent ................................ ................................ ................ 75 Mock Enrichment with MS Detection ................................ ................................ 76 MALDI ZrP ................................ ...... 76 MALDI TOF MS after mock enrichment on Zr OBP ................................ ... 77 Simple Enrichment with MS Detecti on ................................ ............................. 78 Summary ................................ ................................ ................................ ................ 78 5 CONCLUSIONS ................................ ................................ ................................ ..... 92 APPENDIX A RAW DATA ................................ ................................ ................................ ............. 94 B FREUNDLICH ADSORPTION ISOTHERMS ................................ ........................ 108 LIST OF REFERENCES ................................ ................................ ............................. 113 BIOGRAPHIC AL SKETCH ................................ ................................ .......................... 117

PAGE 7

7 LIST OF TABLES Table page 2 1 The isoelectric point the standard deviation for phosphopeptide enrichment materials. ................................ ................................ ................................ ............ 45 3 1 The Q max (nmol/mg) and K L (nmol 1 ) values of the Langmuir fits the standard error for the interaction of pDSIP with TiO 2 ZrP, and Zr OBP. ......... 66 3 2 The Q max (nmol/mg) and K L (nmol 1 ) values of the Langmuir fits the standard error for the interaction of DSIP with TiO 2. and Zr OBP. ....................... 66 3 3 Binding capacity considered by mass and by surface area for pDSIP at pH 3.0 onto the materials. ................................ ................................ ........................ 66 3 4 Estimations of the charge for peptides or the si values considered in the work. ................................ ................................ ........... 66 4 1 The approximate point of mass independence is tabulated when we can reasonably estimate it from th e data in Figures 4 1 an d 4 2. .............................. 91 B 1 The 1/ n (unitless), and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the interaction of pDSIP with TiO 2 ZrP. ...................... 112 B 2 The 1/ n and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the interaction of DSIP with TiO 2 ZrP. ................................ ....... 112 B 3 The 1/ n and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the interaction of pDSIP and DSIP with Zr OBP. ................................ 112

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Schematics showing the surface s of phos phopeptide enrichment materials and the possible mecha nism by which enrichment occurs for IMAC, MOAC, ZrP ................................ ................................ ................................ .......... 24 1 2 ZrP. ................................ ....... 25 1 3 Schematic representation of idealized structu re of the pillared material zirconium biphenylenebis(phosphonate), Zr(PO 3 Ph Ph PO 3 ) ........................... 26 1 4 casein (400 fmol) analyzed by MALDI TOF MS .................. 27 1 5 casein analyzed by MALDI TOF MS ............. 28 1 6 The overlap of the phosphopeptides observed in mouse liver lysate after enrichment ZrP, TiO 2 and Fe 3+ IMAC ................................ .................... 29 1 7 A schematic showing the composition of the electrical double layer and the ................................ ................ 29 2 1 ZrP and Zr OBP. Absorbance signal was normalized to one. ................................ ................................ .............................. 41 2 2 ZrP and Zr OBP. We observe peaks in ZrP diffraction that are consistent with published work. ............................ 41 2 3 SEM of materials ................................ ................................ .............................. 42 2 4 of the materials with respect to pH. ................................ ........................... 42 2 5 Our method for determining the IEP is illustrated. ................................ .............. 43 2 6 Schemati c showing the hexa gonal surface and edge surfaces. ........................ 44 3 1 Langmuir adsorption isotherms for the interaction of pDSIP with TiO 2 at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the st andard deviation for n = 3 sets. ................................ ................................ ....... 62 3 2 Langmuir adsorption isotherm for the interaction of DSIP with TiO 2 at pH 3.0. Error bars represent the standard deviation for n = 2 sets. .............................. 62 3 3 ZrP at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard deviation from n=2 to n=5 sets. ................................ ........................... 63

PAGE 9

9 3 4 ZrP at pH 3.0. The data cannot be fit with the Langmuir isotherm. Error bars represent the standard deviation for n = 4 sets. ................................ ................................ ....... 63 3 5 ZrP at pH 2.0 shown with overlaid Langmuir fitting curve for pDSIP. The inset shows the Langmuir equation parameters Q max (n mol/mg) and K L (nmol 1 ). .................. 64 3 6 Langmuir adsorption isotherms for the interaction of pDSIP (black) and DSIP (red) with Zr OBP at pH 3.0. ................................ ................................ ............... 64 3 7 A comparison of the Langmuir adsorption isotherms for the adsorbents TiO 2 ZrP (red), and Zr OBP (blue) with the adsorbates pDSIP (solid) and D SIP (dashed) in pH 3.0 buffer. ................................ ................................ ... 65 3 8 The modeled peptide charge as a function of pH (right axis) is shown overla described in Chapter 2. ................................ ................................ ...................... 65 4 1 ZrP T Approximately 3 ZrP T is necessary to completely bind pDSIP at pH ZrP T is necessary at pH 7.4 ................................ 80 4 2 The mass dependent adsorption of pDSIP and DSIP onto TiO 2 The data shown ref lect the amount of peptide in solut ion after an 18 h interaction. ......... 81 4 3 ZrP T ........................ 82 4 4 ZrP T The majority of the peptide does not bind to the particles and is thus observed in the unbound fraction.. ................................ ................................ ............................... 83 4 5 Full ZrP T With ten times more adsorbent, it is possible to almost completely adsorb pDSIP onto ZrP T under pH 7.4 conditions. ................................ ................................ ......... 84 4 6 Full enri ZrP T With more adsorbent available, we see an increase in the non specific binding of DSIP ZrP T .. ................................ ................................ ................................ ....... 85 4 7 Full enrichment sequence study o f pDSIP onto TiO 2 ................................ ......... 86 4 8 Full enrichment sequence study of DSIP onto TiO2. ................................ .......... 86 4 9 MALDI TOF MS of the unbound fraction after th ZrP T in pH 3.0 Trizma buffer. Two peaks indicating the presence of pDSIP can be observed. ................................ ................................ ................................ 87

PAGE 10

10 4 10 MALDI TOF MS of the elution of pDSIP using 300 mM ABC after the ZrP T ................................ ................................ ........ 87 4 11 MALDI TOF MS of the unbound fraction after the interaction of pDSIP with Zr OBP. ................................ ................................ ................................ .............. 88 4 12 MALDI TOF MS of the elution fraction after the interaction of pDSIP with Zr OBP. ................................ ................................ ................................ ................... 88 4 13 MALDI casein. The phosphopeptides are labeled. ................................ ................................ ................................ ............... 89 4 14 MALDI casein ZrP. No phosphopeptides are observed. ........................... 89 4 15 MALDI casein tryptic ZrP. No phosphopeptides are observed. ................................ ...... 90 4 16 MALDI TOF MS of the elution fraction after the inter casein tryptic ZrP. The phosphopeptides are labeled. ................................ ........ 90 A 1 TiO2; pDSIP; pH 3.0, 4.5, 7.4; Set A; file 2013 03 14. ................................ ........ 95 A 2 TiO2; pDSIP; pH 3.0, 4.5, 7.4; Set B; file 2013 03 15. ................................ ........ 96 A 3 TiO2; pDSIP; pH 3.0, 4.5, 7.4; Set C; file 2013 03 15. ................................ ........ 97 A 4 ZrP; pDSIP; pH 3.0, 4.5; Set A; file 2013 03 06. ................................ ............. 98 A 5 ZrP; pDSIP; pH 3.0, 4.5; Set B; file 2013 03 07. ................................ ............. 99 A 6 ZrP; pDSIP ; pH 3.0, 4.5; Set C; file 2013 03 07. ................................ ........... 100 A 7 ZrP; pDSIP; pH 3.0; Sets D,E; files 2013 02 05, 2013 02 07. ....................... 101 A 8 ZrP; pDSI P; pH 4.5; Set D; file 2013 03 22. ................................ .................. 102 A 9 ZrP; pDSIP; pH 7.4; Sets A C; files 2013 03 15, 2013 03 22. ....................... 103 A 10 TiO 2 ; DSIP; pH 3.0; Sets A B; file 2013 05 23. ................................ ................. 104 A 11 ZrP; DSIP; pH 3.0; Sets A B; files 2013 05 23, 2013 05 29. ......................... 105 A 12 Zr OBP; pDSIP; p H 3.0; Sets A C; files 2013 04 09, 2013 04 19. Samples B 4 and B 5 were removed due to a pipetteing error. ................................ .......... 106 A 13 Zr OBP; DSIP; pH 3.0; Sets A B; files 2013 05 31. ................................ .......... 107

PAGE 11

11 B 1 Freundlich adsorption isotherms for the interaction of pDSIP with TiO 2 at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard deviation for n = 3 sets. ................................ ................................ ..... 109 B 2 Freundlich adsorption isotherm for the interaction of DSIP with TiO 2 at pH 3.0. Error bars represent the standard deviation for n = 2 sets. ..................... 109 B 3 ZrP at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard deviation from n=2 to n=5 sets. ................................ ......................... 110 B 4 ZrP at pH 3.0. Error bars represent the standard deviation for n = 4 sets. ..................... 110 B 5 Freundlich adsorption isotherms f or the interaction of pDSIP (black) and DSIP (red) with Zr OBP at pH 3.0. ................................ ................................ .... 111

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirement s for the Degree of Doctor of Philosophy PHOSPHOPEPTIDE ADSORPTION ON ZIRCONIUM PHOSPHATE AND ZIRCONIUM PHOSPHONATE SOLIDS By Emily J. Pollard December 2013 Chair: Daniel R. Talham Major: Chemistry Phosphorylation is one of the most common post transl ational modification of proteins, and it serves a regulatory or signaling function in numerous biological processes such as metabolism, gene expression, and differentiation. Studying the tryptic digests of proteins is an important aspect of understanding these processes. Mass spectrometry is the tool of choice for the study of phosphopeptides, but the substoichiometric amounts of phosphopeptides, the poor ionization of them in typical processes, and signal suppression from non phosphopeptides necessitates an enrichment step before mass spectrometry analysis. Typical phosphopeptide enrichment uses immobilized metal affinity chromatography (IMAC) or metal oxide affinity chromatography (MOAC) to separate phosphopeptides for focused study. To successfully emp loy these systems, it is often necessary to pre treat protein digests or work near pH 3 in efforts to minimize competitive, acidic non phosphopeptide binding. Even still, enrichment is often incomplete either losing some phosphopeptides or retaining some non phosphopeptides. Increasing and applying an understanding of the enrichment

PAGE 13

13 mechanism on IMAC or MOAC or developing a new material with improved properties could improve enrichment efforts. Titania, a MOAC material, is perhaps the most common phosphop eptide enrichment material in use today, but its enrichment mechanism is not well understood. Zirconium phosphate has been reported once as a viable enrichment material, but little is known about its enrichment mechanism. Zirconium phosphonate particles have not been previously reported for this purpose. This work discusses research efforts to probe the enrichment mechanism on titania and to develop and explore new enrichment materials derived from zirconium phosphate and zirconium phosphonate. It will include quantitative and semi quantitative comparisons of the binding of model peptides to titania, zirconium phosphate, and zirconium phosphonate.

PAGE 14

14 CHAPTER 1 INTRODUCTION Phosphopeptide Significance and the Need for Enrichment Phosphorylation is a common post translational modification of proteins. It plays a key regulatory roll in biology as it may control important cellular activities such as cell division, growth, differentiation, metabolism, and apoptosis. 1 5 Kinases and phosphatases, which respective ly phosphorylate and de phosphorylate proteins, signal changes for these cellular activities and, when malfunctioning, may have a role in the development of some cancers. 1 5 Studying the phosphorylation characteristics of proteins will provide insight in to these important biological mechanisms. This transient modification is observed primarily on amino acids serine (S), threonine (T), or tyrosine (Y) at approximately the ratio of 90:10:0.05, respectively. 1 However, estimates on the extent of protein pho sphorylation are varied. The high end speculates that almost all eukaryotic proteins are phosphorylated at some point, 1 and about 30% of proteins might be phosphorylated at a given time. 4 Phosphorylated forms of proteins are present at substoichiometric concentrations. 5 The low ratio of the phosphorylated to the non phosphorylated forms of a given protein and the transient nature of the modification make it difficult to study the phosphoproteome and drives the need for enrichment. Common Enrichment Mater ials Mass spectrometry (MS) is the method of choice for studying the phosphoproteome. It is sensitive enough to detect fmol levels of protein under the appropriate conditions, and it can identify unknown materials using tandem technologies. 2 Typically, p roteins are enzymatically digested to yield peptides for MS

PAGE 15

15 abundant non phosphorylated peptides making phosphopeptides difficult to detect. Enrichment techniques are used to separate phosphorylated and non phosphorylated peptides before MS analysis to improve detection. Several enrichment techniques exist; immobilized metal affinity chromatography (IMAC) and metal oxide affinity chromatography (MOAC) are the most comm on. 1 5 IMAC systems complex phosphate via immobilized metals such as Fe 3+ and Ga 3+ held to a support by chelating ligands such as iminodiacetic or ni trilotriacetic acid (Figure 1 1A ). 1 5 When a mixture of peptides are exposed to the IMAC material, both the phosphate on a phosphopeptide and the carboxylate on acidic amino acid side chains are able to displace the ligands surrounding the metal center and can coordinate to the metal center (Figure1 1D ). In order to improve the enrichment of phosphopeptides researchers using IMAC will frequently invoke a methyl esterification reaction for the peptide mixture prior to IMAC. In methyl esterification, the carboxylic acids are converted to methyl esters by the addition of methanolic HCl to a solution of digest ed peptides. 6 While this step reduces non specific binding of acidic non phosphopeptides, methyl esterification does not prevent non specific binding, and it introduces an additional handling step that can result in the loss of material. Additionally, IM AC suffers from metal leaching which can also contribute to the loss of phosphopeptides or the contamination of sample with metal ions. 2 MOAC materials such as TiO 2 have a positively charged surface unde r acidic conditions (Figure 1 1B ). Under the same conditions, phosphopeptides and acidic peptides can have a negative charge. Thus an electrostatic mechanism controls MOAC

PAGE 16

16 enrichment (Figure 1 1E ). 5 If the pH is sufficiently low, around pH 3, carboxylates will protonate and become neutral thereby reduci ng non specific binding to some extent. However, some non specific binding still occurs. 1 2, 4, 5 Additionally, since the metals in MOAC are covalently bound to in the metal oxide structure, the materials are much less prone to leaching. 2 5 Zirconium b ased forms exist for both IMAC and MOAC materials. ZrO 2 is an alternative MOAC material to TiO 2 and Zr 4+ can be chelated by a phosphonate to make a Zr 4+ IMAC material. 2 7 9 These materials demonstrate different, and in some cases perhaps improved, enri chment capabilities compared to their more traditional counterparts. An explanation for this observation has not been provided, but instead it raises some interesting questions about the role the metal might play in the enrichment mechanism. The role of zirconium in phosphopeptide enrichment must be explored further. Zirconium Phosphate and Zirconium Phosphonates ZrP) was first synthesized by Clearfield and Stynes in 1964. 10 It is a layered material with the formula Zr(HPO 4 ) 2 H 2 O. Each layer consists of a plane of zirconium atoms, each bridged by six phosphate groups with three from above and three from below (Figure 1 2). Three oxygen atoms from each phosphate group bind to three different zirconium atoms, and the fourth oxygen exists as a hydroxyl group that extends into the interlayer space. Thus an octahedral zirconium sheet is formed at the center of the layer with tetrahedral p hosphate sheets on either ZrP are held together by van der Waals forces. The interlayer space houses one water molecule per unit cell. About two decades after the ZrP, Dines et al. demonstrated pillared zirconium

PAGE 17

17 bisphosphonic acids. 11 13 These materials have a similar ZrP except that an aryl group or alkyl chain takes the place of two hydroxyl groups on two phosphates and connects the layers of the material; an example is shown in Figure 1 3. 14 Clearfield provides a compendium of metal phosphonate ZrP and a large number of zirconium phosphonates. 15 ZrP as a phosphopeptide enrichment material. 16 In their wo ZrP to TiO 2 and Fe 3+ IMAC using MALDI TOF MS to analyze their enrichments. They used tryptic digests of purified phosphoproteins casein, dilutions of the same with tryptic digest of bovine serum albumin (a non phosphoprot ein), mouse liver cell protein extracts and leukemia cell protein extracts. The authors noted several interesting observations. First, the post synthetic ZrP with additional zirconyl chloride increased the affinity of the material for phos phopeptides (Figure 1 4). In Figure 1 4 C we see every expected casein and good exclusion of non phosphopeptides from the spectra. Second, the inclusion of free zirconyl chloride in the system prevented the bind ZrP while the inclusion of FeCl 3 largely did not (Figure 1 5). We think that the free Zr 4+ present in the experimental conditions for Figure 1 5 B bound to the phosphates of the phosphopeptides and thus prevented the phosphopep ZrP had no bias for different amino acid sites of phosphorylation. Fourth, more ZrP than with TiO 2 or Fe 3+ IMAC, however some phosphopeptides were uniquely enr iched by each material (Figure 1 6). For this experiment, proteins were extracted from mouse liver lysate and were digested with

PAGE 18

18 trypsin according to a previously published protocol. 17 Similar results were observed for the REH human leukemia cell line (d ata not shown here). Fifth, non phosphopeptides ZrP surface. In spite of these important observations, the authors do not offer a possible mechanism that could explain their observations other than the suggestion that Zr may be t he cause. zirconium and a phosphonate in several materials for the enrichment of phosphopeptides. These materials coordinate zirconium using a phosphonate intrinsically available o r chemically added onto the surface of each of the following materials: a surface modified porous silicon wafer, surface modified polymeric beads, monolithic particles, surface modified magnetic Fe 3 O 4 /SiO 2 core/shell nanoparticles, and bi functionalized per iodic mesoporous organosilicas 18 22 consistently shows good qualitative selectivity for phosphopeptides over non phosphopeptides and sensitivity down to the fmol level in some cases. ZrP as a phosphopeptide enrichment material with an exploration of the zeta potential of the particle in suspension and a specific emphasis on the adsorption step in the overall enrichment process. Our work will define the isoelec ZrP and detail ZrP(Figure 1 1 C F ) similar to the mechanism of traditi onal IMAC and highlighting the role of zirconium in allowing phosphopeptide adsorption and preventing non phosphopeptide adsorption. In cases where zirconium is not fully coordinated within the

PAGE 19

19 material, only the phosphate on a phosphopeptide is strong en ough to displace the ligands surrounding the zirconium center. Competing carboxylates on non phosphopeptides are not strong enough to displace the ligands around. This should reduce non specific binding, which is a common problem in IMAC. Our proposed m echanism explains the good selectivity reported previously. 16 We present data in support of our proposed mechanism. Additionally, we introduce a known zirconium phosphonate particle synthesized bisphosphonic acid as a novel phosphopeptide enr ichment material. 23 This material was chosen to increase the availability of zirconium on the surface of the particle, a point that will be discussed in Chapter 2. Our zirconium phosphonate material should have similar enrichment capability to the variou s zirconium anchored by 18 22 Additional ly, it should have a ZrP. We explore the ZrP and compare both of our synthesized materials to commercially available TiO 2 S ince the enrichment mechanism with TiO 2 is electrostatic, and we suspect a coordinate covalent mechanism on our zirconium phosphate/phosphonate materials, we expect, and in fact observe, differences in the pH dependent binding of these materials. In order to understand the effect of pH on the particles, we begin our study with the exploration of isoelectric points for the materials of interest in this work. Isoelectric Points The first aspect of this study will consist of the characterization of the materi als including the intrinsic electrokinetic properties of the particles in suspension. Suppose a particle exists in aqueous suspension with some innate charge on the surface. Counter

PAGE 20

20 ions and water molecules will align at the surface forming a tight layer the Stern layer, which ends at the inner Helmholtz plane (Figure 1 7). A second layer forms, the diffuse layer, with less tightly bound counter ions and water molecules. The potential within this layer gradually decreases as you move away from the part icle surface until it reaches zero. These two layers together are considered the electrical double layer. If a particle in suspension is put in motion, all the molecules within the Stern layer and those some distance from the surface in the diffuse layer will move with the particle. The boundary between molecules th at move with the particle and those that do not has several names: the outer Helmholtz plane, the slipping plane, or the shear plane. The zeta in suspension. 2 4 26 suspension is likely to be stable. 25 terial. In some cases, such as adjusting the stability of a suspension, it can be useful to titrate in pH can change the sign of the suspension is expected to be stable.

PAGE 21

21 ntial (Figure 1 There are some cases, however where this assumption might not be accurate. One such case exists when the surface is electrokinetically anisotropic. 26 Since th some distance from the surface of the particle, in principle it represents a weighted average of the surface potentials of various sites present on the particle. Nonetheless, charged particle in suspension. as a means for estimating the surface potential of a material in suspension at different pH values. As we study the adsorption of peptides o nto the various materials across a range of pH values, we will be able to comment on the binding mechanism from a standpoint of whether or not electrostatic attraction is contributing. Adsorption Isotherms Enrichment can be considered a multi step process that includes adsorption, washing, desorption, and regeneration. 27 Characteristics of a successful enrichment system include adsorption of the desired analyte, reasonable separation of the desired analyte from competing components of the mixture, and good recovery of the desired analyte for further work. Additionally, it is beneficial if the system can be regenerated and used more than once. The first step of enrichment and the first characteristic of successful enrichment centers around adsorption. Chr onologically, adsorption occurs first, and without successful adsorption of the desired analyte enrichment cannot occur. Adsorption of competing components can also occur; if it occurs, the second characteristic, separation, can be brought about by the wa shing step where non

PAGE 22

22 specifically adsorbed components are removed and the desired analyte remains adsorbed. Alternatively, if non specific binding can be avoided during the adsorption step, two characteristics of successful enrichment, analyte adsorption and separation from competition, can be achieved in a single step. The final characteristic of successful enrichment, good recovery of the desired analyte, occurs during the desorption step. This step can also be called the elution step. The second as pect of this study will consist of the focused study on the adsorption step of model peptides onto novel and known phosphopeptide enrichment materials. The adsorption step was chosen because it occurs first and because it can provide two characteristic fe atures of successful enrichment, adsorption and separation. We will conduct batch adsorption studies and model binding with adsorption isotherms, which have been widely used to study adsorbents and adsorbates. 28 34 Using adsorption isotherms, we can quan titatively compare adsorption using the metrics binding capacity, efficiency, and favorability, and we can also quantitatively compare separation during the adsorption step as the metric selectivity. Adsorption isotherms provide for a focused and quantita tive study of the adsorption step of the larger chromatographic process, but alone they are not capable of predicting a full chromatographic system. 27 Additionally, our adsorption isotherms reflect the adsorption of a peptide from a single component buffe red solution whereas chromatographic enrichment processes handle complex mixtures of peptides. However, even under these conditions, adsorption isotherms still provide useful information for material/method development of more complete systems. 35 38 We w ill

PAGE 23

23 use adsorption isotherms as a means to study the first step of enrichment on phosphopeptide enrichment materials. This Work enrichment materials with adsorption isotherm data across a range of pHs to understand the conditions that favor enrichment and probe the binding mechanism that controls the enrichme nt process. We test a relatively recently demonstrated enrichment ZrP and a novel enrichment material Zr OBP, and we compare them to a well established enrichment material TiO 2 The materials are tested against a model phosphopeptide and a non phosphorylated analogue of the same sequence under strongly ionic conditions, and the results are modeled with adsorption isotherms revealing quantitative binding parameters. These parameters are then compared to describe the differences and similarities between the materials.

PAGE 24

24 Figure 1 1. Schematics showing the surface s of phosphopeptide enrichment materials ( A C ) and the possible mechanism by which enrichment occurs ( D F ) for ZrP (left to right). A) B) C) D) E) F)

PAGE 25

25 Figure 1 ZrP. Zirconium is green and has an octahedral geometry. Phosphorus is blue and has a tetrahedral geometry. Oxygen is red, and hydrogen is whi te. The interstitial water molecules are omitted for clarity. Input parameters from Troup and Clearfield. 39

PAGE 26

26 Figure 1 3. Schematic representation of idealized structure of the pillared material zirconium biphenylenebis(phosphonate), Zr(PO 3 Ph Ph PO 3 ). Reprinted with permission from Wang, Z.; Heising, J. M.; Clearfield, A. J. Am. Chem. Soc. 2003 125 10375 10383. Copyright 2003 American Chemical Society.

PAGE 27

27 Figure 1 4. A casein (400 fmol) analyzed by MALDI TOF MS: A ) direct analysis, B ZrP, C ZrP treated with excess ZrOCl 2 prior to use. P # indicates a phosphopeptide; an asterisk marks a phosphopeptide with an oxidized M residue. Reprinted with permission from Xu, S.; Whitin, J. C.; Yu, T. T.; Zhou, H.; Sun, D.; Sue, H.; Zou, H.; Cohen, H. J.; Zare, R. N. Anal. Chem. 2008 80 5542 5549 Copyright 2008 American Chemical Society.

PAGE 28

28 Figure 1 casein ana lyzed by MALDI TOF MS after using A ZrP, B ZrP and ZrOCl 2 C ZrP and FeCl 3 P # indicates a phosphopeptide; an asterisk marks a phosphopeptide with an oxidized M residue. Reprinted with permission from Xu, S.; Whitin, J. C. ; Yu, T. T.; Zhou, H.; Sun, D.; Sue, H.; Zou, H.; Cohen, H. J.; Zare, R. N. Anal. Chem. 2008 80 5542 5549 Copyright 2008 American Chemical Society.

PAGE 29

29 Figure 1 6. The overlap of the phosphopeptides observed in mouse liver lysate after enrichment with ZrP, TiO 2 and Fe 3+ IMAC. Adapted with permission from Xu, S.; Whitin, J. C.; Yu, T. T.; Zhou, H.; Sun, D.; Sue, H.; Zou, H.; Cohen, H. J.; Zare, R. N. Anal. Chem. 2008 80 5542 5549 Copyright 2008 American Chemical Society. Figure 1 7. A schema tic showing the composition of the electrical double layer and the Reprinted from Minerals Engineering 23, Alvarez Silva, M.; Uribe Salas, A.; Mirnezami, M.; Finch, J. A. The point of zero charge of ph yllosilicate minerals using the Mular Roberts titration technique, 383 389. 26 Copyright 2010, with permission from Elsevier.

PAGE 30

30 CHAPTER 2 ISOELECRIC POINTS OF ENRICHMENT MATERIALS Isoelectric Points Metal oxide affinity chromatography (MOAC) has become one o f the most common techniques for phosphopeptide enrichment. Yet, the mechanism by which this enrichment occurs is largely unknown One possibility is that an electrostatic attraction exists between a positively charged metal oxide surface and a negatively charged phosphate. 40 Researchers have observed that lower pH, where TiO 2 is positively charged and competing carboxylates are neutralized, improves the results. 1 Another possibility is that a coordinate covalent bond is formed between the metal center a nd the phosphate; other researchers have observed a bidentate binding mode of phosphates to metal oxides. 41 Additionally, some reports have shown preferential differences in phosphopeptide enrichment using ZrO 2 compared to TiO 2 without an explanation of w hy this might be 42 ZrP have shown promise, however a discussion of enrichment mechanism with this material is lacking. 16 Clearly, the mechanism of these enrichments needs to be studied further especially for new enrich ment materials such as zirconium phosphate and zirconium phosphonate. We have probed the electrokinetic potential of phosphopeptide enrichment materials by charge material. Knowing the IEP of phosphopeptide enrichment materials will give us insight into enrichment mechanisms.

PAGE 31

31 Kosmulski has published a series of reviews outlining the IEP of metal oxides, 24 43 46 and in another review Kosmulski specifically discusses the published distributions of the IEP for TiO 2 and other common metal oxides. 47 In this work, he finds the universal IEP of TiO 2 to be about 5.8. A universal IEP, as d escribed by Kosmulski, is independent of specific material specimen, detailed solution composition, and specific experimental conditions such as specific instrument or sample pretreatment. For ZrO 2 another common MOAC material, no universal IEP is found; rather the published results span a range of 4 8. 47 ZrP. 48 Beyond our aim of understanding the IEP within the frame of phosphopeptide enrichment, it is important to determine the IEP of materials as a contribution to colloid science where this information relates to the stability of the suspension 49 and the effect of pH changes on the uptake of ions. 50 The IEP of a colloid can be determined using any of several techniques. Two techniques of note are electrokinetic and potentiometric titrations. Between these techniques, there are differences in the way the IEP is determined and in what the value represents. The IEP is the pH at which no net charge exists at the slipping plane around the particle. P otentiometric techniques more accurately find the point of zero charge (PZC) instead of the IEP. The PZC is the pH at which the surface charge of the material is zero. Potentiometric techniques usually compare the behavior of a dispersion with the repeat ed additions of strong acid or base to a blank handled in the same way. The difference between the dispersion and the blank reveal the surface charge density of the dispersion as a function of pH and thus allows researchers to determine the PZC. 51 Electr

PAGE 32

32 pH using electroosmosis, electrophoresis, or electroacoustics. All electrokinetic measurements involve detecting the motion of either the solvent in the presence of a surface of interest (electr oosmosis) or the particle of interest in suspension the potential at the slipping plane of the electrical double layer above a charged surface using well establishe d models. In most cases, the IEP and the PZC of materials are approximately the same. However, this might not the case with anisotropic materials such as talc, chlorite, and serpentine 26 52 ZrP. Electrokinetic measurements consider the potenti al at the slipping plane, and therefore the technique reports an average of the potential contributing moieties on the surface. When a material is anisotropic, the contributions from one surface type can be very different from the other, and a shift in th e ratio of the different surface moieties could have a large effect on the IEP. In this work, the IEP of phosphopeptide enrichment materials are determined by determining the IEP of phosphopeptide enrichment materials in this work, we will be able to make inferences about the mechanism of phosphopeptide enrichment in future work. Similar strategies of using IEPs to make mechanistic inferences have been reported previously. 53 56 Experimental Section Titanium(IV) oxide nanopowder (TiO 2 ), zirconium(IV) oxide nanopowder (ZrO 2 ), and 1,8 octanediphosphonic acid (OBP) where purchased from Aldrich. Zirconyl chloride octahydrate was purchased from Acros Organics. Potassium chlor ide, potassium hydroxide, hydrochloric acid and o phosphoric acid 85% were purchased

PAGE 33

33 from Fisher. All materials were used as received. Water was purified to a resistivity of pure filtration system. Synthesis of Materials Zirconium phosphate (Zr(HPO 4 ) 2 H 2 ZrP) was prepared similarly to the reflux method previously described by Clearfield and Stynes. 10 Zirconyl chloride octahydrate (3 mmol) was dissolved in 2.5 mL of water, and 7.5 mL of 4 M phosphoric acid was heated to 97 C. The zirconyl chloride solution was quickly added to the heated phosphoric acid solution (1:10 Zr:P reaction ratio), and the resulting precipitant was allowed to reflux at 97 C for one day. The particles were recovered by centrifuga tion for 10 min at 11,300 x g, and the supernatant was discarded. The particles were washed with 90 mL of water three times. The washed particles were dried overnight at 65 C, and the dried particles were ground into a fine white powder with mortar and ZrP was post synthetically treated with excess zirconium ions by suspending particles 1 mg/mL in a solution of 20 mM zirconyl chloride octahydrate overnight with vigorous mixing. The treated particles were recovered by c entrifugation, washed with 300 mL water twice, dried overnight at 65 C, ZrP. Neat Zr(O 3 P (CH 2 ) 8 PO 3 ), hereafter known as Zr OBP, has been prepared previously by Hong et al where they formed a microcrystalline solid using fluorine complexation to aide in microcrystal development. 23 In this work, treated Zr OBP was ZrP. Zirconyl chloride octahydrate (0.37 mmol) was dissolved in 96 mL water, and 64 mL conta ining 0.73 mmol OBP was heated to 97 C. The zirconyl chloride solution was quickly added to the heated phosphonic acid solution (1:3.9 Zr:P reaction ratio), and the resulting precipitant was allowed to reflux at 97 C for one day.

PAGE 34

34 The particles were rec overed by centrifugation for 10 min at 11,300 x g, and the supernatant was discarded. The particles were washed with 180 mL of water three times. The washed particles were dried overnight at 65 C, and the dried particles were ground into a fine white po wder with mortar and pestle. The neat material was post synthetically treated with excess zirconium ions by suspending particles 1 mg/mL in a solution of 20 mM zirconyl chloride octahydrate overnight with vigorous mixing. The treated particles were recov ered by centrifugation, washed with 300 mL water twice, dried overnight at 65 C, and ground into a fine white powder with mortar and pestle. Characterization of Materials The Fourier transfer infrared (FTIR) spectra of the synthesized particles was collec ted using a Nicolet 6700 Thermo Scientific spectrophotometer from 4000 500 cm 1 at a resolution of 1 cm 1 as the average of 16 scans. The particles were mounted in a pressed KBr pellet, and the background absorbance of KBr was subtracted. Scanning electr on microscopy (SEM) micrographs of the samples were obtained using a Hitachi S 4000 FE SEM located at the Major Analytical Instrument Center (MAIC) at the University of Florida. SEM samples were mounted on silica slides and coated with carbon prior to ima ZrP and neat Zr OBP, powder X ray substrate with double sided a 50 at a step size of 0.008 and a scan rate of 10.2 sec per step. For treated Zr OBP, powder XRD was collected on a Bruker DUO diffractometer with CuK a an I m S power source, multi la yered mirror optics, and an APEXII CCD area detector (1024x1024 detection format) positioned 150 mm from the sample. Powder samples

PAGE 35

35 were packed into 0.3 mm diameter boron rich thin walled capillary tubes purchased from the Charles Supper Company. Diffracti on patterns were collected between 5 and 50 degrees with a 600 sec/image collection time. The Ma group of University of South Florida kindly completed gas sorption measurements on a Micromeritics ASAP 2020 surface area analyzer using nitrogen at 77 K. Co mplete Analysis Laboratory, Inc. in Parsippany, NJ conducted elemental analysis. Zeta Potential Titrations Titration experiments were carried out on a Malvern Zetasizer Nano ZS equipped with an MPT 2 autotitrator and an auto degas unit using a Malvern fo lded capillary cell. The pH probe was calibrated prior to each session. Samples were suspended in 10 mL of 1 mM KCl and sonicated for 20 min prior to titration. Titrations were carried out at room temperature (23.0 23.5 C). The materials were titrated by the addition of acid or base across the IEP of the material using a program specific to each material. TiO 2 was suspended approximately 1 mg/mL and was titrated from pH 3.5 to 8.0 with 0.5 step size and 0.1 target pH threshold using 0.25 M HCl, 0.025 M KOH, and 0.25 M KOH. ZrO 2 was suspended approximately 1 mg/mL and was titrated from pH 8.5 to 5.0 with 0.5 step size and 0.1 target pH threshold using 0.25 M HCl, 0.025 M HCl, and 0.25 M ZrP was suspended approximately 1 mg/mL and was titrated from resting pH to pH 1 .0 with 0.5 step size and 0.1 target pH threshold using 1.0 M HCl. Treated ZrP was suspended approximately 1 mg/mL and was titrated from resting pH to pH 1.5 with 0.5 step size and 0.1 target pH threshold using 1.0 M HCl. Neat Zr OBP was suspended ap proximately 0.2 mg/mL and was titrated from pH 3.5 to 1.0 with 0.5 step size and 0.1 target pH threshold using 1 M HCl. Treated Zr OBP was suspended

PAGE 36

36 approximately 0.2 mg/mL and was titrated from pH 5.0 to 2.0 with 0.5 step size and 0.2 target pH thresho ld using 1 M HCl, 0.25 M HCl, and 0.25 M KOH. Results and Discussion Characterization of Materials ZrP and Zr OBP, and we have characterized both materials with FTIR, powder XRD, SEM, nitrogen sorption, and elemental analysis. ZrP and Zr OBP (Figure 2 1) are consistent w ith the published spectra. For neat and ZrP, the peaks in the region 1330 900 cm 1 represent the stretching modes of phosphate. The broad absorbance from 3600 2900 cm 1 the two sharp peaks at ~3598 and ~3511 cm 1 and the feature observed at ~3 155 cm 1 are attributed to ZrP in different configurations (zeolitic or H bonded water) 57 For neat and treated Zr OBP, the phosphonate stretches are observed as an unresolved single peak centered at ~1040 cm 1 Asymmetric an d symmetric stretches of CH 2 are observed near ~2931 and ~2857 cm 1 respectively. Symmetric bending of CH 2 is observed near ~1467 cm 1 and CH 2 scissoring is observed near ~1409 cm 1 23 For both materials, we do not observe significant differences related to the treatment of the particles. Powder XRD for the synthesized samples are shown in Figure 2 2. Some peaks are observed for ZrP, and their positions are consistent with literature reports for the material. The poor signal to noise ratio and larger peak breadth can be attributed to small particles, limited amount of sample, or more amorphous material. For Zr OBP, we observe attributed to pillaring of the phosphonic acid. We do not observe peaks at higher

PAGE 37

37 angles, which would indicate organized structure of a shorter inter atomic distance; only th SEM mic ZrP (Figure 2 3A C and E F ) are consistent with aggregated platelets of approximately 100 nm in length, as reported previously. 58 SEM micrographs of Zr OBP (Figu re 2 3 D and G H ZrP in size and shape. Comparatively, TiO 2 nanopowder particles are less than 25 nm in diameter. Surface area analysis by nitrogen sorption shows BET surface areas of 86.3 m 2 /g ZrP a nd 182.3 m 2 /g for treated Zr OBP. Comparatively, TiO 2 has a surface area by BET of 45 55 m 2 /g, according to the manufacturer. Elemental analysis of Zr OBP N is consistent with the theoretical weight percentages for the molecular formula; the theoretical pe rcentages for Zr, P, O, C, and H are 25.24, 17.14, 26.56, 26.59, and 4.46 respectively, and the experimental values are 25.19, 17.01, 26.86, 26.48, and 4.54 respectively. The error is approximately 0.4%. Experimental oxygen percentages are determined by mass balance from the observed values of the other four elements. The slight increase of mass percentages of oxygen and hydrogen, and the corresponding slight decrease in C, P, and Zr, between theoretical to experimental values are most likely due to the incorporation of water in the lattice. This is confirmed by the modest absorbance near 3400 cm 1 in the FTIR spectra (Figure 2 1). Isoelectric Point Determination 4. Raw data was binned into x axis categories, and the data points plotted here represent the mean x and y position of raw data. The lines between the points represent point a to point b str aight line connections. The x error bar represents the standard deviation of pH values

PAGE 38

38 axis bin. The IEP of each material can be visualized as the point at which the titration visualized in Figure 2 5. The black data points represent the mean value for repeated titrations of the material on the Zetasizer as described in Figure 2 4. The x axis and y purpose of determining the IEP, the x axis error is ignored because it is sufficiently ) were drawn through all pairs of points and the x axis; the x axis intercept represents a possible IEP. The material IEP reported recorded below and presented here (Figure 2 5, blue points) represents the mean plus or minus the standard deviation (blue e rror bars) of the possible IEPs for the given pair of data points. Using the method described above, we calculated the IEPs of phosphopeptide enrichment materials (Table 2 1). The IEP we found for TiO 2 is within the range of the universal IEP, and the IEP we found for ZrO 2 falls within the broad range of IEPs reported previously. 47 ZrP is in agreement with the value published by Mercado et al. 48 ZrP with zirconium ions, we observe a dramatic shift in the IEP f rom ca. pH 1 to ca. pH 3. Interestingly, the IEP found for ZrP is similar to that reported for zirconium pyrophosphate. 48 However the FTIR (Figure 2 1) and XRD (Figure 2 ZrP and not zirconium pyr ophosphate. 57 59 This indicates our treatment with excess zirconium ions has reduced the number of acidic sites on the surface of the material without disrupting the overall structure. We observe similar behavior for neat and

PAGE 39

39 treated Zr OBP. This is a reflection of the similar surface chemistries of the zirconium phosphate and the zirconium phosphonate. In spite of the large IEP shifts we observe upon treatment, we do not observe an ZrP, roughly 3% of the zirconium present in a particle is positioned on the surface, and in post synthetic treatme nt we are only adding zirconium to phosphoric acid not already fully coordinated to three zirconium atoms. Thus, it is not surprising that we cannot detect a change by elemental analysis We propose that the dramatic shift observed in the IEP upon treat ment is rooted in the anisotropic nature of the particle surface. When we synthesize ZrP and Zr OBP, the phosphate/phosphonate are in excess and the reaction is carried out in an acidic aqueous environment, therefore on the edge of these particles and an y defects in the faces, zirconium is likely coordinated to water or phosphate/phosphonate that is not fully coordinated to three zirconium atoms. In the treatment step, we can add zirconium into these spaces. The dramatic IEP shifts we observe from our neat materials to our treated materials suggest we have changed the ratio of the two surface moieties by increasing the Zr and decreasing the P. We have also observed incomplete treatment by a smaller shift in the IEP (data not shown). If the IEP of the approximately 2.6, the treatment has failed, and the material must be treated again with excess zirconium ions. The treatment failure is related to poor mixing conditions, and after treating a second time with better mixing conditions, the IEP approaches 3. This observation further supports the proposed theory of shifting the ratio of surface sites with an increase in Zr and a decrease in P.

PAGE 40

40 Summary ZrP and Zr OBP in neat and treated forms. The characterizations of our neat materials match what has been published previously with the exception of the XRD of our Zr OBP. We attribute this to having, by design, a more amorphous mat erial than Hong et al 23 The FTIR, SEM, and XRD do not change appreciably between the neat and treated forms of the materials; this demonstrates that ZrP, TiO 2 and ZrO 2 are in agreement with published values. The IEP of neat Zr OBP is ZrP, as one would expect. The dramatic shift we observe in the IEP from neat to treated materials is attributed to a shift in the ratio of Zr to P sites on the surface o f these anisotropic materials. We observe IEPs in three clusters: neat zirconium phosphate/phosphonate around pH 1, treated zirconium phosphate/phosphonate around pH 3, and TiO 2 and ZrO 2 between pH 6 and 7. In subsequent work, we will use the knowledge o f the IEP for each material to make inferences about the mechanism of phosphopeptide enrichment.

PAGE 41

41 Figure 2 1. FTIR spectra of neat and treated ZrP and Zr OBP. Absorbance signal was normalized to one. The peaks centered around 1000 cm 1 are attributed to phosphate/phosphonate stretching. CH stretching is observed in Zr OBP just under 3000 cm 1 Broad water stretching is observed 3500 3000 c m 1 Additional interstitial water stretching is observed above 3500 cm 1 ZrP. Figure 2 ZrP and Zr OBP. We observe peaks in ZrP diffraction that are consistent with published work. The diffraction for Z r OBP shows only an ordered interlayer spacing al low angles indicating a large d spacing.

PAGE 42

42 Figure 2 3. SEM of materials: A ZrP, D ) treated Zr OBP, E F ZrP, and G H ) neat Zr OBP. Figure 2 espect to pH. A) B) C) D) E) F) G) H)

PAGE 43

43 Figure 2 5. Our method for determining the IEP is illustrated. TiO 2 (left) and Zr OBP T (right) are shown as representative examples. Using the nearest data points with error on either side of 0 mV we draw nine lines between the point s and the points the standard deviation. We calculate the average pH value of the nine points where these lines cross 0 mV. This mean the standard deviation is reported.

PAGE 44

44 Figure 2 6. Schematic showing the hexagonal surface (top and bottom) and edge surface (left) for A ZrP and B ) Zr OBP before and after treatment with excess Zr ions. a) b) A) B)

PAGE 45

45 Table 2 1. The isoelectric point the standard deviation for phosphopeptide enrichment materials. TiO 2 ZrO 2 ZrP T ZrP N Zr OBP T Zr OBP N 6.3 0.1 7.0 0. 1 3.1 0.4 1.1 0.3 3.2 0.2 1.2 0.3

PAGE 46

46 CHAPTER 3 ADSORPTION STUDIES Separation and Adsorption Liquid chromatography is a long standing technique invoked for the separation of components in solution. The essence of the technique is separating compone nts of a mixture in space. 60 61 A liquid chromatography separation can be divided into four processes: adsorption, washing, desorption, and regeneration. 27 Adsorption is the process of mixture components interacting with the separation medium. Washing is the process of removing unwanted components that have also bound to the material. Elution is the process of recovering the component of interest, and regeneration is the process of restoring the stationary phase to its original form for subsequent use. In this work we have fo ZrP and Zr OBP (described in Chapter 2), and we now present a study of the adsorption step of phosphopeptides and non phosphopeptides onto these materials in comparison to a common phosphopeptide enrichment mate rial TiO 2 Specifically, we will compare the binding capacity and binding efficiency for the adsorption of model peptides onto phosphopeptide over the non phosphopeptide. To t his end, we have prepared adsorption isotherms for a variety of materials, peptides, and conditions. These factors of interest can be quantified by the parameters of fitting model equations to experimental data and subsequently comparing fitting parameter s quantitatively. One of the most common models for adsorption is the Langmuir adsorption isotherm. The Langmuir model assumes monolayer coverage onto a homogeneous surface and provides parameters for binding efficiency and binding capacity. 62 Selectivi ty can be

PAGE 47

47 defined in several ways. One option is the ratio of the initial slopes of the adsorption isotherm. 63 64 Another option is a ratio of the distribution coefficients. 65 By calculating the selectivity quotient, we will have a semi quantitative me thod to compare the abilities of these materials to preferentially bind phosphopeptides over non phosphopeptides. This chapter will focus on the adsorption of two model peptides, one phosphopeptide and one non phosphopeptide, onto three enrichment materia ls at different pH values in a buffered solution. By fitting the adsorption data with isotherm models, we will be able to compare properties that help to define the adsorption step in an enrichment system. Using this data, we will consider the binding m echanisms that drive the adsorption of peptides onto each of the adsorbents. The binding mechanism that controls MOAC enrichment is debated. One possibility is that an electrostatic attraction exists between a positively charged metal oxide surface and a negatively charged phosphate. 40 Researchers have observed that lower pH, where TiO 2 is positively charged and competing carboxylates are neutralized, improves the results. 1 Another possibility is that a coordinate covalent bond is formed between the met al center and the phosphate; other researchers have observed a bidentate binding mode of phosphates to metal oxides. 41 We present MOAC enrichment with an electrostatic binding mechanism. Conversely, the binding mechanism that controls adsorption onto zir conium phosphate and zirconium phosphonate is unexplored. We present a coordinate covalent binding mechanism for the adsorption of phosphopeptide onto zirconium phosphate and zirconium phosphonate. The phosphate group of a phosphopeptide is capable of di splacing weaker ligands around zirconium to coordinate with the metal center. The carboxylate group on competing acidic non

PAGE 48

48 phosphopeptides is unable to coordinate under the same conditions. This could lead to preferential binding of phosphopeptides over non phosphopeptides in competitive binding situations such as those encountered in phosphopeptide enrichment experiments. Experimental Section Materials ZrP and Zr OBP were as described in Chapter 2; all work detailed in this chapter was done with the treated form of the adsorbents. TiO 2 was purchased from Aldrich. All other chemicals were purchased from Sigma or Fisher and used as received unless desc or greater using a Barnstead E pure filtration system. A buffer capable of buffering the pH range 2.2 10.6 was prepared at 200 mM strength using a 200 mM concentration of each of the five following components: 2 (N morpholino)ethanesulfonic acid sodium salt (MES), 2 amino 2 (hydroxymethyl) 1,3 propanediol (Tris base), glycine, acetic acid, and KCl. The buffer was adjusted to the desired pH using concentrated HCl or 10 M KOH. Peptide Prepa ration Phosphorylated delta sleep inducing peptide (pDSIP, sequence WAGGDA(pS)GE) and delta sleep inducing peptide (DSIP, sequence WAGGDASGE) were purchased from Mimotopes (Clayton, Victoria, Australia). Upon receipt of the lyophilized material, it was di ssolved in 0.1% aqueous acetic acid and distributed into aliquots were frozen at 20 C until use. Upon use, the aliquot was thawed and brought up to volume with the desire d buffer. There was no detectable pH change from the original buffer solution to the buffer peptide solution.

PAGE 49

49 Peptide Adsorption Adsorption onto TiO 2 The interaction of pDSIP with TiO 2 at pH 3.0 and at pH 4.5 was carried out in buffer using 1 mg of partic les (Cahn C 33 microbalance) and an amount of peptide ranging from 4.6 77.4 nmol; a blank with an equivalent amount of particles and 0 nmol peptide and a five point standard curve of peptide were run in parallel. Particles were appropriate amount of peptide was added to the particle suspension. The total volume using additional buffer as necessary. The interaction was mixed overnight at room temperature. For the interaction of pDSIP with TiO 2 at pH 7.4, 5 mg of TiO 2 was used; for the interaction of DSIP with TiO 2 at pH 3.0, 1 mg of particles and an amount of pe ptide ranging from 9.3 77.4 nmol were used. All other conditions remained the same. ZrP ZrP were carried out in the same manner as described above for TiO 2 ZrP was ZrP and a peptide range of 13.7 t o 114.3 nmol was used. All other interaction conditions remained the same. Adsorption onto Zr OBP The interaction of Zr OBP with pDSIP consisted of 0.4 mg adsorbent and a peptide range of 6.8 114.3 nmol in pH 3.0 buffer. The interaction with DSIP consist ed of 1.0 mg of particles and a peptide range of 9.3 to 77.4 nmol. All other conditions remained the same as those described above.

PAGE 50

50 Peptide mass balance by UV absorbance spectroscopy The interaction suspensions were centrifuged 5 min at 12k RPM on a Herml e microcentrifuge Z230MR equipped with a Hermle 220.59 VO4 rotor. One hundred microliters of the supernatant was transferred to a Corning UV transparent 96 well microplate purchased from Fisher. The absorbance of the supernatant was measured over the ran ge 200 500 nm using a Tecan Safire microplate spectrophotometer. The well volume of The absorbance value at 278 nm was attributed to peptide, and the absorbance at 500 nm was considered the background and was used for correcting baseline shifts. The amount of material in solution ( C e ) was determ ined using a standard curve prepared separately for each experiment. The amount of material bound per mass of particles ( q e ) was determined by mass balance (Equation 3 1) where C i is the amount of peptide added in nmol and m is the mass of particles in mg. (3 1) Data was plotted as q e with respect to C e and was fit with the Langmuir adsorption isotherm (Equation 3 2) using Origin 8.5; q e is the amount of peptide bound per mass of particles, C e is proportional to the con centration in solution, Q max is proportional to the adsorption capacity under the conditions, and K L is the relative Langmuir equilibrium constant and represents binding efficiency. (3 2)

PAGE 51

51 Results We studied the interacti on of two model peptides pDSIP and DSIP with three adsorbents TiO 2 ZrP, and Zr OBP. The data was fit with the Langmuir adsorption isotherm. Using the parameters of the Langmuir adsorption isotherm, we compare differences in adsorption as a result of peptide phosphorylation, adsorbent, and pH. We also attempted to f it the data with Freundlich adsorption isotherms; these fits were poorer than the Langmuir adsorption isotherm fits in all cases. Details on the Freundlich adsorption isotherm and the fitted data are available in Appendix B. Adsorption onto TiO 2 TiO 2 is o ne of the most commonly used phosphopeptide enrichment materials. We have studied adsorption onto TiO 2 as a basis for comparison to alternative enrichment materials. Figure 3 1 shows mean data points fit with Langmuir adsorption isotherms for the interac tion of pDSIP with TiO 2 from sets of data at pH values 3.0, 4.5, and 7.4. The isotherm at pH 3.0 shows a steep initial slope and approaches an asymptote near 40 nmol/mg. The parameters of the fit are shown in Table 3 1. The parameter K L quantifies the s teepness of the initial slope and relates to the binding efficiency the adsorbent has for the adsorbate. The Q max value corresponds well to the asymptote we observe in the data plot. As we move to higher pH, we observe a decrease in the slope (Figure 3 1 ) and the corresponding K L (Table 3 1). The results show that the binding of pDSIP onto TiO 2 is most favorable at pH 3.0. We studied the binding of DSIP onto TiO 2 at pH 3.0 for comparison. Figure 3 2 shows the mean data points fit with the Langmuir adso rption isotherm. We observe a gradual slope, but we do not observe an asymptote. The parameters (Table 3 2) give a very small K L value and a Q max value with a large error. The size of the error is directly

PAGE 52

52 related to the data not extending far enough to observe an asymptote. The results indicate the binding efficiency of DSIP onto TiO 2 at pH 3.0 is quite low and an adsorption maximum has not been satisfactorily defined. We compare the binding onto TiO 2 at pH 3.0 of pDSIP versus DSIP. The binding effici ency parameter K L is an order of magnitude larger for pDSIP than for DSIP indicating the binding of the phosphopeptide is significantly better, an important characteristic for a phosphopeptide enrichment material. ZrP ZrP was recently reported as a phosphopeptide enrichment material 16 We ZrP. Figure 3 3 shows mean data points fit with the Langmuir adsorption isotherm for the interaction of ZrP from sets of data at pH values 3.0, 4.5, and 7.4. We observe a moderate initial slope at pH 3.0 and an adsorption maximum above 30 nmol/mg. The initial slope and adsorption maximum decrease as the pH increases. The parameters (Table 3 1 ) quantify the same observations with the best binding efficiency and capacity values at ZrP. Figure 3 ZrP at pH 3.0. We do not observe appreciable ads orption below 60 nmol, and the data cannot be fit with the Langmuir adsorption isotherm. The results indicate that the binding efficiency is very poor, and we cannot define and adsorption maximum. The non ZrP at pH 3.0. ZrP at pH 2.0. Figure 3 5 shows this interaction. Compared to the results at pH 3.0, we observe an increase in ZrP. The data for DSIP a t pH 2.0

PAGE 53

53 demonstrate minimal binding out to 80 nmol of peptide. The appreciable adsorption of ZrP in combination with minimal adsorption of DSIP under the same conditions at both pH 3.0 and pH 2.0 have some interesting implications as we cons ider this adsorbent as a phosphopeptide enrichment material; these will be discussed in the discussion section below. Adsorption onto Zr OBP We explored peptide adsorption onto the zirconium phosphonate Zr OBP, a ZrP Figure 3 6 shows the mean data points fit with the Langmuir adsorption isotherm for the interaction of pDSIP and DSIP with Zr OBP at pH 3.0. For the interaction with pDSIP, we observe a steep initial slope corresponding to good binding efficiency and a high asymptote corresponding to a high binding capacity (Table 3 1). For the interaction with DSIP, we observe a more gradual initial slope and an asymptote indicating a binding capacity at about a third of the binding capacity of pDSIP onto the same mat erial (Table 3 2). Collectively, the superior binding of pDSIP onto Zr OBP compared to the binding of DSIP onto Zr OBP indicate the material could serve as a phosphopeptide enrichment material; we will discuss this further below. Discussion We have prepa red Langmuir adsorption isotherms when possible for the interaction of adsorbents TiO 2 ZrP, and Zr OBP with the adsorbates pDSIP and DSIP, two model peptides. In order to compare and contrast these materials with one another we will use the parameters of the Langmuir adsorption isotherm of binding capacity ( Q max ) and binding efficiency ( K L ). We also define selectivity the ability of an

PAGE 54

54 adsorbent to bind the phosphopeptide over the non phosphopeptide as a means to compare the phosphopeptide enrichment capability of the adsorbents. Binding Capacity The binding capacity is a measure o f how much adsorbate can bind to the adsorbent under the experimental conditions. It can be thought of as the number of binding sites for the adsorbate on the adsorbent, which is dependent on the environment. The binding constant is quantified by the par ameter Q max in the Langmuir adsorption isotherm. The parameters from the Langmuir adsorption isotherms for the adsorption of pDSIP are tabulated in Table 3 1. In all these cases, the highest binding capacity is observed at pH 3.0. We can compare the capa city graphically in Figure 3 7, which shows the fits for the data at pH 3.0 on a common scale. We observe the highest capacity with Zr OBP; TiO 2 an ZrP have the same capacity within error. ZrP at pH 2.0, the capacity increased (Figure 3 5). We attribute the superior binding capacity of Zr OBP to the surface chemistry of the material ( Figure 2 6 B ). Theoretically, the surface of Zr OBP consists completely of ZrP has Zr available and the remainder of the surface is comprised of the hydroxyl groups of phosphoric acid (Figure 2 6 A ). Additionally, the binding capacity is measured in nanomoles of peptide per milligram of adsorbent. We have the surface area of the particles (Chapter 2), and so we have estimated the binding capacity as nanomoles of peptide per square meter of adsorbent (Table 3 3). When we compare the binding capacity using the alternative units, Zr OBP has approximately the same capacity as TiO 2 This suggests that the surface of both of these materials is fully available to bind

PAGE 55

55 phosphopeptide whe ZrP is available. Thus, the apparent superiority for the binding capacity of Zr OBP over TiO 2 is diminished when we consider the surface area of the materials. Additionally, our calculation suggests that the full ZrP is not accessible for binding phosphopeptide. Binding Efficiency The binding efficiency is a measure of how well the adsorbate binds to the adsorbent under the experimental conditions. It is quantified by the parameter K L in the Lan gmuir adsorption isotherm. The parameters from the Langmuir adsorption isotherm for the adsorption of pDSIP are tabulated in Table 3 1. In all these cases, the most efficient binding is observed at pH 3.0. We can compare the efficiency graphically in Fi gure 3 7, which shows the fits for the data at pH 3.0 on a common scale. We observe the most efficient binding with Zr OBP followed by TiO 2 ZrP. We attribute the superior efficiency of Zr OBP to the availability of Zr metal centers over the entire surface of the material (Figure 2 6 B ). Conversely, TiO 2 does not offer free metal centers or Zr to bind phosphopeptides, and although ZrP has Zr metal centers available on the surface of the treated material, only a fraction of the surface has Zr available and the remainder of the surface is comprised of the hydroxyl groups of phosphoric acid (Figure 2 6 A ). A proposed binding mechanis m and the related importance of Zr metal centers will be discussed below. ZrP at pH 2.0 increases above the efficiency of Zr OBP at pH 3.0 (Figure 3 5). However, this data represents a single replicate, and we do not have comparable data for Zr OBP at pH 2.0. Additional experiments for both zirconium based materials should be done at or even below pH 2. Unfortunately, the buffering capacity of our buffer does not extend

PAGE 56

56 below pH 2.2, and an alternative experimental system should be developed to explore these conditions further. We also compare the binding efficiencies for the adsorbate DSIP; these Langmuir adsorption isotherm parameters are available in Table 3 2. We observe the same trend a s with pDSIP the highest efficiency for binding DSIP is seen with Zr OBP followed by TiO 2 ZrP. Since we are considering these adsorbents as phosphopeptide enrichment materials and DSIP is a non phosphopeptide, it would be ideal for the adsorb ZrP demonstrates the preferred binding efficiency toward DSIP. Selectivity An important characteristic of phosphopeptide enrichment materials is the ability to separate phosphopeptides f rom non phosphopeptides. This can occur during the adsorption step if the adsorbent preferentially binds the phosphopeptide over the non phosphopeptide; this is called selectivity. Selectivity can be defined in several ways. 63 65 Like Banerjee et al. an d Liebl et al. we use the initial slope of the adsorption isotherm, which is also the binding efficiency parameter K L We calculate the selectivity of a given adsorbent as the quotient the K L for binding pDSIP over the K L for binding DSIP. It is a ratio of the binding efficiencies for the phosphopeptide over the non phosphopeptide. We calculate the selectivity quotient for TiO 2 at 35 and for Zr OBP at 6. The selectivity is ca. fivefold better for TiO 2 compared to Zr OBP. This can be attributed to the mo re favorable binding of DSIP to Zr OBP versus DSIP to TiO 2 Since we cannot ZrP, we ZrP using K L ; this indicates the

PAGE 57

57 selectivity is very good on this adsorbent, perhaps three or more or ders of magnitude better than either TiO 2 or Zr ZrP to the surface chemistry of the material. Although the surface chemistry of Zr OBP is ZrP, the selectivity is not comparable. This is lik ely related to significantly more surface area available for binding on Zr OBP that led to more binding of DSIP. In this sense, perhaps binding capacity and selectivity are opposing characteristics. In order to better understand our selectivity observati ons and to increase the relevancy for phosphopeptide enrichment materials, it would be necessary to attempt competitive enrichment experiments where both phosphopeptides and non phosphopeptides are exposed to the material simultaneously. However, based on our observations in this ZrP has the best selectivity; this could be an indication that it would outperform TiO 2 and Zr OBP in phosphopeptide enrichment. Binding Mechanisms To explore potential binding mechanisms with the experimental methods and data we have presented here, we approach the discussion from a standpoint of considering the feasibility of an electrostatic adsorption mechanism. To begin, we must first establish an estimate of the peptide charge across a range of pH. We already have a n estimate of the particle charge across a broad pH range (chapter2). The IEP of a peptide (pI) and charge state of a peptide across a range of pH can be estimated with readily available online calculators. However, these calculators do not account for t he acidic nature of phosphate groups. Gauci et al. developed a java script that calculates the pI of modified peptides, 66 but this provides only a pI and no graphical representation of the charge across a pH range. Therefore, a simple Matlab script was k indly developed by Zheng, Z. to display the predicted charge of phosphopeptide across a

PAGE 58

58 range of pH. 67 All peptide charge calculators are limited by the accuracy of the pK a values input and in the assumption that the pK a for any group is independent of th e surrounding environment. Therefore, these calculators are especially weak for short sequences and outside the pH range of 4 8. 68 Nonetheless, the developed calculator provides an estimation of the peptide charge for DSIP and pDSIP, shown overlaid with 8). The pI of DSIP is estimated to be 3.0; and the pI for pDSIP is estimated to be 2.1. Given this information, we consider the adsorption isotherms and adsorption parameters as a function of pH in o rder to assess the possibility of an electrostatically conditions tested is available in Table 3 4. At pH 7.4 all peptides and particles are negatively charged. We do no t observe any appreciable pDSIP adsorption under these conditions, and we can attribute this to electrostatic repulsion between peptide and particle surface or the unavailability of binding sites due to charge states. At pH 4.5, TiO 2 is positively charged ZrP is negatively charged, and pDSIP is negatively charged. We see appreciable adsorption of pDSIP onto TiO 2 ZrP. For TiO 2 this observation supports the theory of electrostatic attraction between peptide and ZrP ele ctrostatic attraction does not seem plausible given that both peptide and surface are likely to be negative. At pH 3.0 TiO 2 ZrP are positively charged, pDSIP is negatively charged, and DSIP is at its estimated pI indicating charge balance rather tha n a lack of charge. With TiO 2 at pH 3.0, we observe increased binding efficiency, capacity, and favorability for the adsorption of pDSIP compared to pH 4.5. We observe less favorable adsorption of DSIP compared to

PAGE 59

59 pDSIP under these conditions. This furt her supports the theory of electrostatic ZrP at pH 3.0, we observe favorable binding of pDSIP and unfavorable binding of DSIP. If we consider only the pH 3.0 data with the modeled charges of the peptides and what we know about the surf ZrP, the interaction at pH 3.0 could be electrostatic. However, it is unreasonable to suspect a different mechanism is in control based only on a shift in pH, and the data at pH 4.5 does not ZrP i s positively charged, DSIP is positively charged, and pDSIP is near but below its pI and thus might also have a positive charge. We see improved adsorption parameters for pDSIP compared to pH 3.0 and worsened adsorption parameter for DSIP. Like the data at pH 4.5, the data at ZrP and pDSIP further refutes the idea of electrostatic interaction controlling the binding mechanism. The decrease in adsorption for DSIP could be attributed to electrostatic repulsion, which could further contribute t o an underlying ZrP. Our data supports the theory of an electrostatic binding mechanism onto TiO 2 as evident by binding between the positively charged surface and negatively charged peptides. Additional stu dies with TiO 2 at pH 2.5 where DSIP is positively charged and pDSIP is negatively charged and at pH 1.0 where pDSIP is also positively charged would provide more complete insight regarding the electrostatic binding mechanism. However, as mentioned previou sly, the limit of the buffering range for our buffer is pH 2.2, and an alternative system must be developed in order to run these experiments. Unlike with TiO 2 our data does not support an electrostatic binding mechanism ZrP are negatively charged yet we observe

PAGE 60

60 ZrP are positively charged, and we also observe binding. Again, f uture study of DSIP binding at pH 4.5 and pDSIP binding at pH 1.0 would be instructive in further supporting or refuting this theory. As an alternative to electrostatic binding, we propose a coordinate covalent interaction is the primary binding mechani sm whereby the phosphate groups of phosphopeptides are able to displace weaker ligands and coordinate to available Zr ZrP (Figure 1 1 F ). Conversely, carboxylate groups of non phosphopeptides are not strong enough to displace the ligands on Zr resulting in the unfavorable adsorption mechanism we observe. This mechanism explains the ZrP. Based on the limited data we have for Zr OBP, we cannot make an accurate assessment of the binding mecha nism that controls peptide adsorption. Based on the ZrP and Zr OBP discussed in previous chapters, we would expect similar mechanisms for both of these materials. However we see a surprising amount of DSIP adsorption onto Zr OBP, which could be related to the increased surface area on Zr OBP. More experiments at different pH would help us understand the mechanism of adsorption onto Zr OBP. Summary We have demonstrated a new potential phosphopeptide enrichment material in p articulate Zr ZrP and TiO 2 The Q max ( nmol/mg) for Zr OBP far exceeds those for the other materials we tested, but when considered in the unit nmol/m 2 the adsorption capacity of Zr OBP is comparable to that for TiO 2 The select ivity for Zr OBP at pH 3 is close to that of TiO 2 but it could be improved by exploring conditions with lower pH.

PAGE 61

61 Our findings support the theory of electrostatic interaction of peptides with the surface of TiO2, and they suggest an alternative mechanis m is involved in the binding of ZrP results in excellent selectivity at pH 3.0 and pH 2.0. We attribute this to a coordinate covalent binding mechanism where only phos phate groups, and not carboxylate groups, are basic enough to displace the ligands around Zr. Additional adsorption isotherm work is required to strengthen this theory, and isothermal titration calorimetry would further support it. We expect the surface chemistry of Zr OBP to ZrP, and with this relationship, we also expect a similar binding mechanism for phosphopeptides. With refinement of synthetic conditions and exploration of alternative phosphonic acids, we may be able to pair the full surface availability of Zr ZrP.

PAGE 62

62 Figure 3 1. Langmuir adsorption isotherms for the interaction of pDSIP with TiO 2 at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard de viation for n = 3 sets. Figure 3 2. Langmuir adsorption isotherm for the interaction of DSIP with TiO 2 at pH 3.0. Error bars represent the standard deviation for n = 2 sets.

PAGE 63

63 Figure 3 3. Langmuir adsorption isotherms for the interaction of pDSIP wi ZrP at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard deviation from n=2 to n=5 sets. Figure 3 ZrP at pH 3.0. The data cannot be fit with the Langm uir isotherm. Error bars represent the standard deviation for n = 4 sets.

PAGE 64

64 Figure 3 ZrP at pH 2.0 shown with overlaid Langmuir fitting curve for pDSIP. The inset shows the Langmuir equation parameters Q max (nmol/mg) and K L (nmol 1 ). Figure 3 6. Langmuir adsorption isotherms for the interaction of pDSIP (black) and DSIP (red) with Zr OBP at pH 3.0. ZrP pH 2.0 Q max 46.9 1.1 K L 0.60 0.08

PAGE 65

65 Figure 3 7. A comparison of the Langmuir adsorption isotherms for the adsorbent s TiO 2 (black) ZrP (red), and Zr OBP (blue) with the adsorbates pDSIP (solid) and ZrP the fit shown is not a Langmuir adsorption isotherm but rather a linear fit through the origin and the data points. Figure 3 8. The modeled peptide charge as a function of pH (right axis) is shown described in Chapter 2.

PAGE 66

66 Table 3 1. The Q max (nmol/mg) and K L (nmol 1 ) values of the Langmuir fits the standard error for the interaction of pDSIP with TiO 2 ZrP, and Zr OBP. TiO 2 ZrP Zr OBP pH 3.0 Q max 43 3 42 6 167 16 K L 0.21 0.05 0.06 0.02 0.24 0.08 pH 4.5 Q max 43 6 30 7 not tested K L 0.11 0.04 0 .04 0.02 pH 7.4 Q max 22 7 6 3 K L 7E 3 3E 3 0.02 0.02 Table 3 2. The Q max (nmol/mg) and K L (nmol 1 ) values of the Langmuir fits the standard error for the interaction of DSIP with TiO 2. and Zr OBP. The ZrP can not be fit with the Langmuir adsorption isotherm. TiO 2 ZrP Zr OBP pH 3.0 Q max 113 47 undefined 75 3 K L 6E 3 3E 3 undefined 4.1E 2 4E 3 Table 3 3. Binding capacity considered by mass and by surface area for pDSIP at pH 3.0 onto the mat erials. Q max (nmol/mg) Surface area (m 2 /g) Q max (nmol/m 2 ) TiO 2 43 45 55 782 956 ZrP 42 86 490 Zr OBP 167 182 918 Table 3 values considered in the work. 2.0 3.0 4 .5 7.4 pDSIP + DSIP + 0 TiO 2 + + + ZrP + + Zr OBP + +

PAGE 67

67 CHAPTER 4 EXPLORATORY WORK INTO DESORPTION AND ENRICHMENT Overview Two characteristics are essential to phosphopeptide enrichment: good separation of phosphopeptides from non phosphopeptides and the ability to recover phosphopeptides for further analysis. Enrichment can be related to chromatography, which consists of four steps: adsorption, washing, desorption, and regeneration. 27 Adsorption was covered extensively in Cha pter 3. Washing is the desorption process of non specifically bound non phosphopeptides. Desorption, in order to disambiguate, will be referred to hereafter as elution. Elution is the desorption process of specifically bound phosphopeptides. Regenerati on is the process by which the adsorbent is returned to its original state so that it can be used again. The work in the previous chapter demonstrated the adsorption step for phosphopeptides and non phosphopeptides onto TiO 2 ZrP, and Zr OBP, but adsor ption alone does not equate to enrichment. In this chapter we will demonstrate mock enrichment as we show in sequence the adsorption and desorption of pure model peptides. We will also demonstrate a simple enrichment scenario using an enzymatically diges casein. Experimental Section All materials and instruments are the same as mentioned previously. Additionally, Trizma buffer was purchased from Sigma Aldrich while trifluoroacetic acid (TFA) and ammonium bicarbonate (ABC) were purcha sed from Fisher.

PAGE 68

68 ZrP Determining the limit of mass dependence ZrP T ) was prepared and characterized as described in Chapter ZrP T ranging from 0.1 18 mg were mixed overnight at room temperature with 26 .4 nmol of peptide in buffer at pH values 3.0, 4.5, and 7.4. The suspensions were centrifuged. The amount of peptide in solution was determined by UV absorbance spectroscopy in comparison to a standard control. Testing enrichment independent of adsorben t mass Peptide solutions were reconstituted from frozen aliquots containing 264 nmol of ZrP T particles (2.0 4 were carried out in quadruplicate by adding 26.4 nmol of peptide in solution to the particle suspensions and mixing overnight at room temperature. The suspensions were centrifuged, and the supernatant was retained for analysis. The particles were washed temperature with mixing. The suspensions were centrifuged, and the supernatant was for 2 h at room temperature with mixing. Th e suspensions were centrifuged, and the supernatant was retained for analysis. overnight at room temperature with mixing. The suspensions were centrifuged, and the supernatant was ret ained for analysis.

PAGE 69

69 At pH 3.0 and 7.4, a similar experiment was carried out using 19.8 ZrP T room temperature. All other conditions remained the same. Positive cont rols were prepared in duplicate with 26.4 nmol of peptide and 0 mg ZrP T Negative controls were prepared in quadruplicate with 2.0 2.2 mg or 19.8 ZrP T handled identically to the experimental samples. The amount of peptide in the supernatant of each fraction in comparison to the a Tecan Safire microplate spectrophotometer, and Corning UV trans parent 96 well microplates (Fisher). Mock Enrichment on Titania The limit of mass dependence with TiO 2 ZrP T The enrichment independent of adsorbent mass with TiO 2 was carried out identically ZrP T except only p H 3.0 was considered. Mass Spectrometry after Mock Enrichment ZrP T and Zr OBP T particles were suspended 5 mg/mL in aqueous solution of approximately 0.5 mg of particles was combin mM Trizma buffer with 100 mM NaCl adjusted to pH 3.0. The remainder of the model enrichment remained the same as described above with the exception that only the first wash was performed. MALDI TOF mass spectra were obtained using an AB Sciex TOF/TOF 5800 ( Framingham, MA ) in reflection mode. Data were acquired from 600 1500 m/z. Spectra

PAGE 70

70 of model peptides pDSIP (929.3 m/z) and DSIP (849.3 m/z) were sought by combining the supernatant of fractions with matrix cyano 4 hydroxycinnamic acid (Sigma Aldrich) and spotting the mixture on a MALDI plate. Casein Digest casein (Sigma Aldrich) using trypsin (Sigma Aldrich), ZrP T and th e enrichment procedure followed published protocol. 16 MS analysis was competed as described above except that the data range was expanded to 600 4000 m/z. Results and Discussion ZrP T and TiO 2 are explored beyond the adsorpti on step as we consider the desorption of non phosphopeptides during the wash steps and the desorption of phosphopeptides during the elution step. First, we must ensure sample to sample variation in the mass of adsorbent does not affect the results. Mass I ndependence In Figure 4 1 we show the plot of peptide in solution as a function of the mass of ZrP T in the sample. A value of 100% represents 26.4 nmol of peptide. When an increase in mass no longer results in a decrease of peptide in solution, mass in dependence has been achieved. Mass independence for the binding of pDSIP was ZrP T at pH 3.0 and 6 mg at pH 4.5. We estimate 20 mg would be necessary at pH 7.4 based on the data available at 18 mg. At pH 3.0, relatively little DSIP bound with mass independence occurring at or below the lowest mass tested, 0.5 mg. At pH 4.5 and pH 7.4, the data were more scattered, and we could not determine the point where mass independence is achieved. These results are summarized in Table 4 1. We chose to conserve adsorbent and proceed with additional

PAGE 71

71 studies knowing that mass dependence could contribute to error in the data. For the same reason we chose to use 2 mg of adsorbent at both pH 3.0 and pH 4.5 in subsequent studies. Twen ty milligrams of adsorbent was necessary at pH 7.4. In Figure 4 2 we show the plot of peptide in solution as a function of the mass of TiO 2 in the sample. At pH 3.0 and pH 4.5, mass independence for the binding of pDSIP was achieved with as little as 1 2 mg of TiO 2 ; at pH 7.4, 4 mg or more was necessary. For DSIP, at pH 3.0 approximately 4 mg or more of TiO 2 was necessary to achieve mass independence; at pH 4.5, 10 mg is necessary. We observe only minimal DSIP binding at pH 7.4, but mass independence may be occurring with as little as 2 mg of TiO 2 These results are summarized in Table 4 1. Since the goal of subsequent work was to maximize pDSIP adsorption while minimizing DSIP adsorption 2 mg was chosen as the constant mass for further pH 3.0 and pH 4. 5 studies, and 10 mg was chosen as the constant mass for further pH 7.4 studies. Unfortunately, further studies at pH 4.5 and pH 7.4 were never realized due to unmitigated problems with the pH 3.0 work (see below). Mock Enrichment with UV Detection ZrP T as the adsorbent Using the known points of mass independence, we conducted a focused study of the four ZrP. Figure 4 3 shows the percentage of peptide in the supernatant of each fraction after the adsorption, wash, and desorptio n steps for the ZrP T At pH 3.0 and pH 4.5 we observe near complete adsorption of the peptide. At pH 7.4, a large amount of peptide remains in the supernatant after the adsorption step. This can be attributed to i nsufficient mass of material (Figure 4 1). In the wash steps, minimal amount of pDSIP is removed from the

PAGE 72

72 particles and present in the supernatant. In the elution step, more than 50% of the peptide adsorbed under pH 3.0 or pH 4.5 conditions is recovered from the particles. We do not observe pDSIP in the elution step of the pH 7.4 adsorption sequence because the majority of peptide did not bind initially. Furthermore, it appears that if pDSIP bound at pH 7.4 is removed in the first wash. The results in dicate that under pH 3.0 and pH ZrP T remains bound during the wash steps, and is recovered in the elution step. Using the same mass of particles, we tested the adsorption/desorption of DSIP ZrP T under the same conditions. Figure 4 4 shows the percentage of peptide in supernatant of each fraction. At all adsorption pH conditions tested, DSIP remains in the supernatant after the adsorption step. If any, only a minimal amount of peptide is observed in the subsequent steps. The results indicate that DSIP does not adsorb onto ZrP T Collectively, the results in Figures 4 3 and 4 ZrP T could be a good phosphopeptide enrichment material. We note the recovery of phosphopeptide in the elu tion step as a demonstration of this critical aspect of an enrichment material. While these results do not demonstrate the separation of phosphopeptides from non phosphopeptides under competitive binding conditions, we can infer from the results Zr P T would be able to selectively bind phosphopeptide over non phosphopeptide under such conditions. Further work under competitive binding conditions should be ZrP T will likely be able to separate phosphopeptides from non phosphopeptides during the adsorption step and

PAGE 73

73 ZrP T a good adsorbent option for phosphopeptide enrichment work. ZrP T as the adsorbent The earlier mass independence results (Figure 4 ZrP T is required for complete adsorption of pDSIP under pH 7.4 conditions. We completed a four step mock enrichment experiment similar to the one described above with 20 mg of ZrP T under pH 3.0 and pH 7.4 adsorption conditions (Figure 4 5). We observe near complete adsorption of pDSIP at both pH 3.0 and pH 7.4. A negligible amount of pDSIP is observed in either wash step. The first elution does not result in the desorption of pDSIP, however we ob serve ca. 50% of the pDSIP in the second elution of the pH 3.0 adsorption sequence and ca. 75% of the pDSIP in the second elution of the pH 7.4 adsorption sequence. We were intrigued by the lack of peptide recovery after the first elution. When we teste d the pH of the elution supernatant, the pH was neutral We attribute this to residual 1% TFA after the second wash. The amount of residual solution was higher in this experiment than in previous experiments because the increased mass, and therefore volu me, of particles trapped more residual solution than the lower mass experiment. Thus, we found it necessary to perform a second elution, and we chose 1 M KOH. After the second elution, the pH of the supernatant was at the expected pH 10, and we were able to recover an appreciable amount of pDSIP. The results indicate that ZrP T phosphopeptide can be adsorbed at pH 7.4 and recovered during a subsequent elution step. ZrP T we tested the adsorptio n/desorption of DSIP with the adsorbent under pH 3.0 and pH 7.4 conditions. Figure 4 6 shows the percentage of

PAGE 74

74 peptide in the supernatant of each fraction. We observe ca. 60% of the DSIP in the pH 3.0 adsorption supernatant and ca. 80% of the DSIP in the pH 7.4 supernatant after adsorption. A negligible amount of peptide is observed in either wash step, and within the projected error of the results ( 20%), no peptide signal is observed in the elution steps. The results indicate that an increase in mas s of adsorbent results in an increase in the binding of the non phosphopeptide. The binding of the non phosphopeptide appears to be favored at the lower pH value. This indicates binding is more favorable at pH 3.0 than at pH 7.4. The relative lack of pe ptide signal in the subsequent steps could indicate that the DSIP that does bind to the adsorbent is not being removed. However, it seems more reasonable to conclude that the conditions of the wash and elution steps have not been optimized for the experim ental setup. Additionally, the negative controls appear to be inappropriate for the system which is causing significant negative signal in ZrP T under these conditions, but the wash and elution steps need to be further explored. Combined, the results in Figures 4 5 and 4 ZrP T will readily bind pDSIP to a greater extent than it will bind DSIP. If competitive binding were attempted, the ZrP T will bind pDSIP prefere ntially over DSIP, a characteristic referred to as selectivity. If we compare these results to those with one tenth of the adsorbent (Figures 4 3 and 4 4), we see complete adsorption of pDSIP at pH 3.0 with both masses of adsorbent. We observe complete adsorption of pDSIP at pH 7.4 only with 20 mg of adsorbent, which is consistent with the mass independence results (Figure 4 1). The

PAGE 75

75 binding of DSIP increases with increased mass of adsorbent at both pH values tested. Collectively, the results suggest th at increased mass could reduce the selectivity under competitive binding scenarios. This could be mitigated by carefully selecting the mass of adsorbent or by improving the desorption of non phosphopeptide during the wash step prior to the desorption of p hosphopeptide during the elution step. Overall, ZrP T led to complete adsorption of phosphopeptide at pH 7.4 while maintaining some selectivity for the phosphopeptide over the non phosphopeptide. TiO 2 as the adsorbent We attempted simil ar four step mock enrichment experiments on the well documented MOAC enrichment material TiO 2 ZrP T Figure 4 7 shows the percentage of peptide in the supernatant of each fraction for the interaction of pDSIP with TiO 2 after ad sorption in pH 3.0 buffer. We observe minimal peptide in the supernatant after the adsorption step and both wash steps. There is a large amount of error in the first wash step, which we attribute to the increased presence of particles in the microplate w ells during the UV absorbance measurement. During the elution, ca. 25% of pDSIP is recovered. Between all four steps we observe only ca. 40% of the expected 100% of peptide signal. This indicates that either pDSIP remains on TiO 2 after all four steps or that the detection of the peptide during the steps shown by this UV absorbance method is inadequate. Overall, the relative amount of peptide signal in each step is in line with our expectations in that the only appreciable signal from pDSIP is in the elu tion step. The low percentage of peptide recovered in the elution step must me addressed further and will be discussed further below. We also tested the adsorption/desorption of DSIP on TiO 2 at pH 3.0. Figure 4 8 shows the percentage of peptide in the su pernatant of each fraction. We observe 25%

PAGE 76

76 of the peptide in the supernatant after the adsorption step. In the following steps, we do not observe any appreciable amount of peptide in the supernatant. As we saw for the interaction of pDSIP with TiO 2 (Fig ure 4 7), the majority of the DSIP is not accounted for, which again suggests that either the DSIP remains on the TiO 2 after all four steps or the detection of the peptide during the steps by this UV absorbance method is inadequate. Since we are using a w ell documented phosphopeptide enrichment material, it seems unlikely that both pDSIP and DSIP remain on the adsorbent after the wash and elution steps. We conclude that the detection method we have used is inadequate for studying the full interaction on T iO 2 even though the same method ZrP T The results on TiO 2 (Figures 4 7 and 4 8) show the expected relative presence of peptide (DSIP in the unbound, pDSIP in the elution), but not at the expected percentages. For this reason, we did not pursue similar studies at pH 4.5 or pH 7.4. Instead, we shifted our focus to the adsorption step (Chapter 3). Mock Enrichment with MS Detection Mass spectrometry (MS) is a very common method of analysis for phosphopeptides after e nrichment. As such, we wanted to demonstrate MS analysis of the model peptides after adsorption and elution in similar mock enrichment studies as those described above. MALDI ZrP Figure 4 9 shows the MALDI TOF spectrum of the supernatant after the ZrPT. We observe [M+H] + and [M+Na] + peaks indicating pDSIP is still present in the supernatant after the adsorption step. Corresponding UV absorbance of the same supernatant indicates ca. 5% of the pe ptide remains in the

PAGE 77

77 supernatant (data not shown). The interaction took place in 10 mM Trizma buffer with 100 mM NaCl and adjusted to pH 3 before use. It was later determined that the buffer ZrP T pulled the pH of the solution down to ca. pH 2.7. The presence of pDSIP in the unbound fraction does not indicate that adsorption has not occurred but rather suggests that some peptide is still present in the supernatant after the adsorption step. Figure 4 10 shows the M ALDI ZrP T using 300 mM ABC following the adsorption described above. The dominant peaks in the spectrum are attributed to the molecular ion peak or the sodium or potassium adducts. The results indicate that ZrP T ZrP T could serve as a phosphopeptide enrichment material and confirm the signal observed in the elution by UV absorbance (Figure 4 3) is rightly a ttributed to pDSIP. MALDI TOF MS after mock enrichment on Zr OBP We completed a mock enrichment using the adsorbent Zr OBP T ZrP T ; all other conditions remained the same. Figure 4 11 shows the MALDI TOF spectrum of the supernatant after the ad sorption step. The phosphopeptide pDSIP is not detected. This indicates good adsorption of pDSIP onto Zr OBP T Figure 4 12 shows the subsequent spectrum after elution. The dominant peaks in the spectrum are attributed to the molecular ion peak or the s odium or potassium adducts. The results indicate that we are able to elute pDSIP from Zr OBP T and suggest Zr OBP T could serve as a phosphopeptide enrichment material.

PAGE 78

78 Simple Enrichment with MS Detection The preceding results do not demonstrate competiti ve binding since a purified peptide as adsorbed and later removed. We repeated the enrichment experiment demonstrated by Xu et al. with the enrichment of the phosphopeptides present in the casein digest (Figure 4 13) and the supernatants after adsorption (Figure 4 14), washing (Figure 4 15), and elution (Figure 4 16). The spectrum of the digest shows the presence of both expected phosphopeptides at low intensity. Neither phosphopeptide peak is observed in the spectra of the supernatant of the adsorption or wash steps. Both phosphopeptide peaks are observed in the elution spectrum. The results we observe are consistent with the previously published report by Xu et al. ZrP T c an be used as a phosphopeptide enrichment material. Summary ZrP T as detected by UV ZrP T and Zr OBP T was confirmed to be pDSIP by MAL DI TOF MS. Furthermore, we placed the UV ZrP T suggesting that the adsorbent would serve well as a phosphopeptide enrichment material. Finally, we were able to repeat the enrichment experiment demonstrated by Xu et al. casein were enriched from the tryptic digest of ZrP T and Zr OBP T could serve as phosphopeptide enrichment mat erials. Further work needs to be done to refine the desorption steps: the wash step for non phosphopeptides and the elution step for phosphopeptides. Particularly in the work

PAGE 79

79 with TiO 2 new experimental methods must be developed to track the peptide throu ghout the process since we cannot account for 100% of the peptide introduced to the system. This might be corrected by using different wash and elution buffers, or an alternative detection approach might be required. This shortcoming not withstanding, th e results are promising that all the materials tested in this chapter can serve as phosphopeptide enrichment materials, and their effectiveness could be improved or better understood by further exploration of the conditions and detection methods.

PAGE 80

80 Figure 4 ZrP T Approximately 3 ZrP T is necessary to completely bind pDSIP at pH ZrP T is necessary at pH 7.4. The data for DSIP at this stage in inconclusive, but we can see that limited amounts of DSIP will bind under conditions favorable for the adsorption of pDSIP. We expect the error to be about 20%.

PAGE 81

81 Figure 4 2. The mass dependent adsorption of pDSIP and DSIP onto TiO 2 The data shown reflect the amount of peptide in solution after an 18 h interaction. Only 1 2 mg of TiO 2 is necessary to completely adsorb all pDSIP at pH 3.0 and 4.5. Increasing to 5 10 mg of TiO 2 is necessary to completely adsorb all pDSIP at pH 7.4. The adsorption of DSIP requires more mass of particles than for pDSIP and has a more defined pH trend. We expect the error to be about 15%.

PAGE 82

82 Figure 4 ZrP T Near complete binding is observed at pH 3.0 and 4.5, and 50 75% recovery is observed in the elution step for the same setup. At pH 7.4, minimal binding is observed because there is not enough adsorbent available.

PAGE 83

83 Figure 4 ZrP T The majority of the peptide does not bin d to the particles and is thus observed in the unbound fraction. Within error, almost no peptide is observed in the following steps. There is some indication that the first wash in 0.1% TFA removed DSIP bound at pH 7.4 and second wash in 1% TFA removed D SIP bound at pH 3.0 or 4.5.

PAGE 84

84 Figure 4 ZrP T With ten times more adsorbent, it is possible to almost completely adsorb pDSIP onto ZrP T under pH 7.4 conditions. Following the adsorption at pH 7.4, approximately 75% of the peptide is recovered in the second elution step with 300 mM ABC.

PAGE 85

85 Figure 4 ZrP T With more adsorbent available, we see an increase in the non specific binding of DS IP ZrP T The results do not show sufficient recovery of the bound peptide in the following washing or elution steps. Exceptionally large error is present in the elution steps. The large error was caused by signal scattering problems in the negati ve controls.

PAGE 86

86 Figure 4 7. Full enrichment sequence study of pDSIP onto TiO 2 We could not achieve good mass balance due to error in the measurements. We diverted our attention to understanding the adsorption step more completely (Chapter 3). Figure 4 8. Full enrichment sequence study of DSIP onto TiO2. We could not achieve good mass balance. After several attempts including changing parameters, we diverted our attention to the adsorption isotherms (Chapter 3).

PAGE 87

87 Figure 4 9. MALDI TOF MS of the unbound fraction after the interaction of pDSIP with ZrP T in pH 3.0 Trizma buffer. Two peaks indicating the presence of pDSIP can be observed. Figure 4 10. MALDI TOF MS of the elution of pDSIP using 300 mM ABC after the Z rP T Multiple peaks indicating the presence of pDSIP are observed including the hydrogenated, sodiated, and potassiated molecular ion peaks.

PAGE 88

88 Figure 4 11. MALDI TOF MS of the unbound fraction after the interaction of pDSIP with Zr OBP. Figure 4 12. MALDI TOF MS of the elution fraction after the interaction of pDSIP with Zr OBP.

PAGE 89

89 Figure 4 13. MALDI casein. The phosphopeptides are labeled. Figure 4 14. MALDI TOF MS of the unbound fraction after the interaction o casein ZrP. No phosphopeptides are observed.

PAGE 90

90 Figure 4 15. MALDI casein ZrP. No phosphopeptides are observed. Figure 4 16. MALDI TOF MS of the elu casein ZrP. The phosphopeptides are labeled.

PAGE 91

91 Table 4 1. The approximate point of mass independence is tabulated when we can reasonably estimate it from the data in Figures 4 1 and 4 2. The point of mass independence is the point at which no additional peptide will bind even if more mass of adsorbent were available. The abbreviation unk is used for unknown when it seems unreasonable to assign an approximate value based on the data available. ZrP T TiO 2 pH pDSIP DSIP pDSIP DSIP 3.0 2 mg 0.5 mg 0.5 mg 4 mg 4.5 6 mg unk 0.5 mg unk 7.4 20 mg unk 4 mg unk

PAGE 92

92 CHAPTER 5 CONCLUSIONS ZrP and Zr OBP and characterized them in co mparison to published examples of these determined the IEP of the materials we prepared and commercial phosphopeptide enrichment materials. We found three clusters of IEPs: ca. pH 6.5 for metal oxides, ca. ZrP and neat Zr ZrP and Zr OBP. These clusters are a testament to different surface chemistries in the materials. We attribute the large shift in IEP from neat materia l to treated material to a shift in the ratio of Zr to P sites on the surface of the materials. We conducted a study of the adsorption of a phosphopeptide and a non for the materials and a prediction of charge as a function of pH for the pair of model peptides. We changed the pH of the interaction in an effort to elucidate any pH charge as fu nctions of pH with the pH dependence of the adsorption results. Our data supports the theory that TiO 2 has an electrostatic binding mechanism. In the same way, ZrP does not have an electrostatic binding mechanism. We propose coordinate covale ZrP. We ZrP has exceptional selectivity of the model phosphopeptide over the model non phosphopeptide. This can also be attributed to a coordinate covalent binding mechanism where only the p hosphate on a phosphopeptide is basic enough to displace the ligands around a Zr center. Conversely, a carboxylate is not basic enough, and thus

PAGE 93

93 we do not observe any non specific binding of non phosphopeptide. We found Zr OBP has the highest binding cap acity of the materials tested. We attribute this to full surface ZrP and TiO 2 Ultimately through refinement of Zr OBP synthesis or with an alternative zirconium phosphonate it may be possible to combine the exceptional selectivity we ZrP and the increased amount of available surface area we observe with Zr OBP. We also demonstrated the ability to recover phosphopeptide during the elution step of mock enrichment and simple enrichment experiments. This provides assurance that the tested materials are suitable phosphopeptide enrichment materials because the desired adsorbate can be removed from the adsorbents for further analysis. Several tiers of future work are conceivable hereafter. First, the exploration of a few more pH would further solidify our proposed binding mechanisms. In some cases, this will require either the use of a different buffer to control pH or external pH control. Second, additional zircon bisphosphonic acids to prepare different solids could yield zirconium phosphonate particles with lower binding efficiency toward non phosphopeptide thus improving the selectivity while maintaining t he increased adsorption capacity. Third, the desorption step, which is essential to effective enrichment, should be explored further in order to improve the removal of non phosphopeptides during the wash step and to improve the recovery of the phosphopept ide during the elution step. Fourth, the work should move forward toward competitive binding scenarios and realistic samples. This will require the development of an alternative analysis method.

PAGE 94

94 APPENDIX A RAW DATA In this appendix we provide the raw d ata for the adsorption isotherm measurements. The category is consistent down each column and is thus labeled only in the first row.

PAGE 95

95 Figure A 1. TiO 2 ; pDSIP; pH 3.0, 4.5, 7.4; Set A; file 2013 03 14.

PAGE 96

96 Figure A 2. TiO 2 ; pDSIP; pH 3.0, 4.5, 7.4; Set B; file 2013 03 15.

PAGE 97

97 Figure A 3. TiO 2 ; pDSIP; pH 3.0, 4.5, 7.4; Set C; file 2013 03 15.

PAGE 98

98 Figure A ZrP; pDSIP; pH 3.0, 4.5; Set A; file 2013 03 06.

PAGE 99

99 Figure A ZrP; pDSIP; pH 3.0, 4.5; Set B; file 2013 03 07.

PAGE 100

100 Figure A ZrP; pDSIP; pH 3 .0, 4.5; Set C; file 2013 03 07.

PAGE 101

101 Figure A ZrP; pDSIP; pH 3.0; Sets D,E; files 2013 02 05, 2013 02 07.

PAGE 102

10 2 Figure A ZrP; pDSIP; pH 4.5; Set D; file 2013 03 22.

PAGE 103

103 Figure A ZrP; pDSIP; pH 7.4; Sets A C; files 2013 03 15, 2013 03 22.

PAGE 104

104 Figu re A 10. TiO 2 ; DSIP; pH 3.0; Sets A B; file 2013 05 23.

PAGE 105

105 Figure A ZrP; DSIP; pH 3.0; Sets A B; files 2013 05 23, 2013 05 29.

PAGE 106

106 Figure A 12. Zr OBP; pDSIP; pH 3.0; Sets A C; files 2013 04 09, 2013 04 19. Samples B 4 and B 5 were removed due to a p ipetteing error.

PAGE 107

107 Figure A 13. Zr OBP; DSIP; pH 3.0; Sets A B; files 2013 05 31.

PAGE 108

108 APPENDIX B FREUNDLICH ADSORPTION ISOTHERMS The data in Chapter 3 were also fit with the Freundlich adsorption isotherm. Equation B 1 is the Freundlich equation where the F reundlich constant K F describes the adsorption capability (similar to capacity) and 1/ n is a reflection of the adsorption intensity. When 1/ n < 1, binding is favorable. When 1/ n = 1, the plot is linear. When 1/ n > 1, binding is unfavorable. 62 Unlike th e Langmuir model, the Freundlich model allows for multilayer adsorption onto a potentially heterogeneous surface. As such, there is no adsorption maximum by the Freundlich model. (B 1) The figures in this appendix show t he Freundlich adsorption isotherms for the data in Chapter 3. The tables in this appendix tabulate the parameters K F and 1/n from the Freundlich adsorption isotherms. In most cases, the Langmuir adsorption isotherm was a better fit for the data. The onl ZrP; in this case, the Langmuir model could not fit the data. On the other hand, the Freundlich model could provide a fit for this data, but the fitting parameters have significant error associated with them.

PAGE 109

109 Figure B 1. Freundlich adsorption isotherms for the interaction of pDSIP with TiO 2 at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars represent the standard deviation for n = 3 sets. Figure B 2. Freundlich adsorption isotherm for the in teraction of DSIP with TiO 2 at pH 3.0. Error bars represent the standard deviation for n = 2 sets.

PAGE 110

110 Figure B 3. ZrP at pH 3.0 (black), pH 4.5 (red), and pH 7.4 (blue). Error bars re present the standard deviation from n=2 to n=5 sets. Figure B 4. ZrP at pH 3.0. Error bars represent the standard deviation for n = 4 sets.

PAGE 111

111 Figure B 5. Freundlich adsorption isother ms for the interaction of pDSIP (black) and DSIP (red) with Zr OBP at pH 3.0.

PAGE 112

112 Table B 1. The 1/ n (unitless), and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the interaction of pDSIP with TiO 2 ZrP. TiO 2 ZrP pH 3 .0 1/ n 0.36 0.06 0.55 0.09 K F 11.0 2.1 4.0 1.3 pH 4.5 1/ n 0.47 0.10 0.57 0.12 K F 6.5 2.2 2.4 1.0 pH 7.4 1/ n 0.85 0.06 0.72 0.18 K F 0.2 0.1 0.2 0.1 Table B 2. The 1/ n and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the interaction of DSIP with TiO 2 ZrP. TiO 2 ZrP pH 3.0 1/ n 0.90 0.05 11 8 K F 0.9 0.1 4E 20 1E 18 Table B 3. The 1/ n and K F (nmol/mg)(mg) n values of the Freundlich fits the standard error for the inter action of pDSIP and DSIP with Zr OBP. pDSIP DSIP pH 3.0 1/ n 0.30 0.09 0.52 0.04 K F 49.3 14.2 6.6 1.0

PAGE 113

113 LIST OF REFERENCES (1) Tichy, A.; Salovska, B.; Rehulka, P.; Klimentova, J.; Vavrova, J.; Stulik, J.; Hernychova, L. J. Proteomics 201 1 74 2786 2797. (2) Dunn, J. D.; Reid, G. E.; Bruening, M. L. Mass Spectrom. Rev. 2010 29 29 54. (3) Han, G.; Ye, M.; Zou, H. Analyst 2008 133 1128 1138. J. Anal. Chem. 2010 65 1295 1310. ( 5) Leitner, A.; Sturm, M.; Lindner, W. Anal. Chim. Acta 2011 703 19 30. (6) Ficarro, S. B.; McCleland, M. L.; Stukenberg, P. T.; Burke, D. J.; Ross, M. M.; Shabanowitz, J.; Hunt, D. F.; White, F. M. Nat Biotech 2002 20 301 305. (7) Wang, H.; Duan, J.; Xu, H.; Zhao, L.; Liang, Y.; Shan, Y.; Zhang, L.; Liang, Z.; Zhang, Y. J. Sep. Sci. 2011 34 2113 2121. (8) Xue, Y.; Wei, J.; Han, H.; Zhao, L.; Cao, D.; Wang, J.; Yang, X.; Zhang, Y.; Qian, X. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2009 87 7 757 764. (9) Lu, J.; Li, Y.; Deng, C. Nanoscale 2011 3 1225 1233. (10) Clearfield, A.; Stynes, J. A. J. Inorg. Nucl. Chem. 1964 26 117 129. (11) Dines, M. B.; DiGiacomo, P. M. Inorg. Chem. 1981 20 92 97. (12) Dines, M. B.; Digiacomo, P. M.; Callah an, K. P.; Griffith, P. C.; Lane, R. H.; Cooksey, R. E. In Derivatized Layerd M(IV) Phosphonates ; Miller, J. S., Ed.; Chemically Modified Surfaces in Catalysis and Electrocatalysis; American Chemical Society: Washington, D.C., 1982; (13) Dines, M. B.; Co oksey, R. E.; Griffith, P. C.; Lane, R. H. Inorg. Chem. 1983 22 1003 1004. (14) Wang, Z.; Heising, J. M.; Clearfield, A. J. Am. Chem. Soc. 2003 125 10375 10383. (15) Clearfield, A. In Metal Phosphonate Chemistry ; Karlin, K. D., Ed.; Prog. Inorg. Chem. ; Wiley: New York, 1998. (16) Xu, S.; Whitin, J. C.; Yu, T. T.; Zhou, H.; Sun, D.; Sue, H.; Zou, H.; Cohen, H. J.; Zare, R. N. Anal. Chem. 2008 80 5542 5549. (17) Jin, W.; Dai, J.; Zhou, H.; Xia, Q.; Zou, H.; Zeng, R. Rapid Commun. Mass Spectrom. 2004 18 2169 2176.

PAGE 114

114 (18) Dong, J.; Zhou, H.; Wu, R.; Ye, M.; Zou, H. J. Sep. Sci. 2007 30 2917 2923. (19) Wang, P.; Zhao, L.; Wu, R.; Zhong, H.; Zou, H.; Yang, J.; Yang, Q. J. Phys. Chem. C 2009 113 1359 1366. (20) Zhao, L.; Wu, R.; Han, G.; Zhou, H.; Ren, L. ; Tian, R.; Zou, H. J. Am. Soc. Mass Spectrom. 2008 19 1176 1186. (21) Zhou, H. J.; Xu, S. Y.; Ye, M. L.; Feng, S.; Pan, C. S.; Jiang, X. G.; Li, X.; Han, G. H.; Fu, Y.; Zou, H. F. J. Proteome Res. 2006 5 2431 2437. (22) Feng, S.; Ye, M.; Zhou, H.; Jia ng, X.; Jiang, X.; Zou, H.; Gong, B. Mol. Cell. Proteomics 2007 6 1656 1665. (23) Hong, H. G.; Sackett, D. D.; Mallouk, T. E. Chem. Mater. 1991 3 521 527. (24) Kosmulski, M. J. Colloid Interface Sci. 2002 253 77 87. (25) Hunter, R. J. In Zeta Potenti al in Colloid Science: Principles and Applications ; Academic Press : 1981. (26) Alvarez Silva, M.; Uribe Salas, A.; Mirnezami, M.; Finch, J. A. Minerals Eng 2010 23 383 389. (27) Mao, Q. M.; Hearn, M. T. W. Biotechnol. Bioeng. 1996 52 204 222. (28) Ston e, M. C.; Tao, Y.; Carta, G. J. Chromatogr., A 2009 1216 4465 4474. (29) Watkins, R.; Weiss, D.; Dubbin, W.; Peel, K.; Coles, B.; Arnold, T. J. Colloid Interface Sci. 2006 303 639 646. (30) Lin, F.; Chen, C.; Chen, W.; Yamamoto, S. J. Chromatogr., A 20 01 912 281 289. (31) Goobes, R.; Goobes, G.; Campbell, C. T.; Stayton, P. S. Biochemistry (N. Y. ) 2006 45 5576 5586. (32) Maharjan, P.; Hearn, M. T. W.; Jackson, W. R.; De Silva, K.; Woonton, B. W. J. Chromatogr., A 2009 1216 8722 8729. (33) Perez A lmodovar, E. X.; Carta, G. J. Chromatogr., A 2009 1216 8339 8347. (34) Zhou, X.; Li, W.; Shi, Q.; Sun, Y. J. Chromatogr., A 2006 1103 110 117. (35) Lin, X.; Wu, J.; Fan, J.; Qian, W.; Zhou, X.; Qian, C.; Jin, X.; Wang, L.; Bai, J.; Ying, H. J. Chem. Te chnol. Biotechnol. 2012 87 924 931. (36) Baker, H. M.; Massadeh, A. M.; Younes, H. A. Environ. Monit. Assess. 2008; 2009 157 319 330.

PAGE 115

115 (37) Valero Navarro, .; Gmez Romero, M.; Fernndez Snchez, J. F.; Cormack, P. A. G.; Segura Carretero, A.; Fernnde z Gutirrez, A. J. Chromatogr., A 2011 1218 7289 7296. (38) Bak, H.; Thomas, O. R. T.; Abildskov, J. J. Chromatogr., B 2007 848 131 141. (39) Troup, J. M.; Clearfield, A. Inorg. Chem. 1977 16 3311 3314. (40) Leitner, A. TrAC, Trends Anal. Chem. 2010 29 177 185. (41) Connor, P. A.; McQuillan, A. J. Langmuir 1999 15 2916 2921. (42) Kweon, H. K.; Hakansson, K. Anal. Chem. 2006 78 1743 1749. (43) Kosmulski, M. J. Colloid Interface Sci. 2004 275 214 224. (44) Kosmulski, M. J. Colloid Interface Sci. 2006 298 730 741. (45) Kosmulski, M. J. Colloid Interface Sci. 2009 337 439 448. (46) Kosmulski, M. J. Colloid Interface Sci. 2011 353 1 15. (47) Kosmulski, M. Colloids Surf. Physicochem. Eng. Aspects 2003 222 113 118. (48) Mercado, A.; Farfn Tor res, E. M.; Regazzoni, A. E.; Blesa, M. A. Colloids Surf. Physicochem. Eng. Aspects 2003 218 277 284. (49) Hubbard, A. T. In Encyclopedia of surface and colloid science, v3 ; Marcel Dekker : New York, 2002. (50) Kosmulski, M.; Maczka, E.; Jartych, E.; Rose nholm, J. B. Adv. Colloid Interface Sci. 2003 103 57 76. (51) Ltzenkirchen, J.; Preocanin, T.; Kovacevic, D.; Tomiscic, V.; Lovgren, L.; Kallay, N. Croat. Chem. Acta 2012 85 391 417. (52) Fuerstenau, D. W.; Pradip Adv. Colloid Interface Sci. 2005 114 115 9 26. (53) Wang, H.; Niu, C. H.; Yang, Q.; Badea, I. Nanotechnology 2011 22 145703. (54) Adamczyk, Z.; Bratek Skicki, A.; Dabrowska, P.; Nattich Rak, M. Langmuir 2011 28 474 485. (55) Nunes, A. P. L.; Peres, A. E. C.; de Araujo, A. C.; Valado, G E. S. J. Colloid Interface Sci. 2011 361 632 638. Powder Technol. 2013 233 146 154.

PAGE 116

116 (57) Ali, A.; hanna, A.; Gad, A. Phosphorus Research Bulletin 2008 22 32 40. (58) Sun, L.; Boo, W. J.; Sue, H.; Clearfield, A. New J. Chem. 2007 31 39 43. (59) Ch. Srilakshmi; Ramesh, K.; Nagaraju, P.; Lingaiah, N.; Sai Prasad, P. S. Catalysis Letters 2005 106 115 122. (60) Giddings, J. C. In Unified separation science ; Wiley : 1991. (61) Poole, C. F. In The Essence of Chromatography ; London : 2003. (62) El Khaiary, M. I. J. Hazard. Mater. 2008 158 73 87. J. Am. Chem. Soc. 2009 131 3875 3877. (64) Liebl, M. R.; Senker, J. C hem. Mater. 2013 25 970 980. (65) Li, H.; Li, G.; Li, Z.; Lu, C.; Li, Y.; Tan, X. Appl. Surf. Sci. 2013 264 644 652. (66) Gauci, S.; van Breukelen, B.; Lemeer, S. M.; Krijgsveld, J.; Heck, A. J. R. Proteomics 2008 8 4898 4906. (67) Zheng, Z. Unpublis hed work 2011 (68) Wilkins, M.; Gasteiger, E.; Bairoch, A.; Sanchez, J.; Williams, K.; Appel, R.; Hochstrasser, D. Methods Mol. Biol. 1999 112 531 552.

PAGE 117

117 BIOGRAPHICAL SKETCH Emily graduated from Colorado State University with her Bachelor of Science degr ee in chemistry in May of 2007. At that time she commissioned into the United States Air Force and took her first assignment at Tyndall AFB, Florida. During this assignment, she co mpleted a Professional Science m chemical s ynthesis from Illinois Institute of Technology in December of 2009. In the fall of 2010, she began her research at the University of Florida, Department of Chemistry under the direction of Dr. Dan Talham. She completed her work toward her Doctor of Philo sophy degree in analytical chemistry in August and graduated in December of 2013. Upon completion of her PhD, Emily returned to regular active duty service in the Air Force.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EFTFWQ716_6BTS04 INGEST_TIME 2014-05-13T21:28:47Z PACKAGE UFE0045754_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES