Improving the Signal-to-Noise Performance of Molecular Diagnostics

Jan 21, 2009 - Darby Kozak, Peter Surawski, Kurt M Thoren, Chieh-Yu Lu, Lionel Marcon, and Matt Trau*. Centre for Nanotechnology and Biomaterials, Lev...
0 downloads 0 Views 614KB Size
Download PDF
360

Biomacromolecules 2009, 10, 360–365

Improving the Signal-to-Noise Performance of Molecular Diagnostics with PEG-Lysine Copolymer Dendrons Darby Kozak, Peter Surawski, Kurt M Thoren, Chieh-Yu Lu, Lionel Marcon, and Matt Trau* Centre for Nanotechnology and Biomaterials, Level 5 East, Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St. Lucia, QLD 4072, Australia Received October 8, 2008; Revised Manuscript Received December 4, 2008

The synthesis, characterization, and use of dendron-like poly(ethylene glycol)-lysine (PEG-Lys) copolymers as an intermediate layer for biomolecular diagnostic signal enhancement is presented. Solid phase Fmoc-peptide synthesis was used to synthesize polymers with one, two, and three PEG-Lys comonomer units in both a linear and first and second-generation dendronic structure directly onto organosilica microspheres. The microsphere surface loadings (number of free amine sites) were modified and quantified through an innovative use of the protecting groups of coupled amino acids. Surfaces with 0.1-100% of the original loading corresponding to 0.3-270 nmol/m2 of free amines were achieved. The influence of polymer structure and surface loading (grafting density) on the signal-to-noise of the microsphere-based molecular diagnostic was assessed measuring the difference in the signal of a model protease digestion assay and reduction in the nonspecific adsorption of bovine serum albumin. Increasing the polymer grafting density and the addition of dendronic branching were both found to increase the assay signal and reduce the nonspecific protein adsorption.

Introduction Biomarker-based proteomic and genomic diagnostic platforms are increasingly being used for the diagnosis of adverse health conditions1 and the early detection of life-threatening diseases such as cancer.2 These diagnostic assays, however, are currently limited by their high percentage of false-positive diagnoses which gives rise to the need for additional often costly and timeconsuming tests, and increased and unnecessary patient anxiety.3 Therefore, considerable focus has been given to improving assay confidence through a variety of means. Biologically, assay confidence and performance can be enhanced through the use of more sensitive, specific and multiple biomarkers and complementary biomolecule assay probes.4 Another easy and effective means of improving assay confidence is through reducing the signal-to-noise generated by the assay platform. This can be achieved in part through choice of assay analysis technique and by incorporating an intermediate grafted polymer layer between the assay platform surface and immobilized biomolecule probes. Grafted polymer layers have been shown to be highly effective in decreasing the assay background “noise” generated by nonspecifically adsorbed biomolecules and to aid in the display of immobilized biomolecule probes.5 These antifouling polymer layers, however, frequently reduce the surface concentration of probe molecules, leading to reduced assay signals. Herein we describe the use of a new polymer structure for improving the signal-to-noise ratio of molecular diagnostic assays. Dendronic intermediate copolymer layers composed of alternating poly(ethylene glycol) and lysine monomer units (PEG-Lys) were synthesized onto an organosilica microsphere based molecular diagnostic surface with the goal of increasing signal probe densities, while maintaining minimum nonspecific biomolecule adsorption noise. PEG is one of the most common and widely used synthetic polymers in biological applications. It is considered to be * To whom correspondence should be addressed. Tel.: +61 7 334 64173. Fax: +61 7 334 63973. E-mail: [email protected].

biologically inert and provides an effective steric barrier against nonspecific biomolecule interactions. Consequently, PEG structures have been applied as coatings for biosensor surfaces6 and to increase the in vivo longevity of injected macromolecules.7 In such applications, the performance of the PEG coatings or structures often improves with increasing PEG density,8 which is dependent upon both the molecular weight of the polymer and the number of polymers grafted per unit area.9 Hyperbranched, star, dendrimeric, and dendronic polymer structures have been seen as attractive materials for biological applications as they form ordered and controllable polymer structures of high polymer and terminal functional site densities.10,11 These unique properties give rise to increased steric repulsion and biomolecule probe attachment sites, making branched polymers very effective intermediate layers for enhancing assay signal-to-noise ratios. PEG hyperbranched polymer structures are therefore believed to offer numerous advantages in biological applications where linear PEG is already used.12-15 A number of recent synthetic strategies have been undertaken to create more controllable dendrimer-like PEG structures. PEG dendrimers are typically synthesized in a two step repeating process of free radical polymerization of PEG followed by the introduction of a branching moiety.16 Recently, Berna et al.17 showed that traditional Fmoc peptide synthesis could be used to synthesize a first and second-generation PEG-Lysine copolymer dendrimer and dendron in solution. This step-by-step synthetic strategy provided products possessing highly controlled polymer structures of low polydispersity. Although primarily applied for cell-uptake studies, the potential use of such structures in antifouling applications was noted. Herein, we present the synthesis and application of PEG dendrons for improving the signal-to-noise ratio of an organosilica microsphere molecular diagnostic. Using Fmoc peptide synthesis, 1, 2 and 3 PEG-Lys comonomer units of linear and dendronic structure were successfully synthesized directly onto microsphere support surfaces. These microspheres have been

10.1021/bm8011314 CCC: $40.75  2009 American Chemical Society Published on Web 01/21/2009

PEG-Lysine Copolymer Dendrons

previously used as both an organic synthetic and a bioassay support18 that can be optically encoded19,20 to create high throughput multiplexed bioassays.21,22 The immobilized copolymers were subsequently modified with a pentapeptide sequence and used as a particle-based enzymatic assay. The effects of polymer loading and structure on the enzymatic assay signal enhancement and reduction in nonspecific protein adsorption were examined.

Biomacromolecules, Vol. 10, No. 2, 2009

361

Table 1. Modified Surface Loadings and Schematic Representation of the Three-Unit Linear and Dendronic PEG-Lysine Copolymers with the Trypsin Cleavable V Pentapeptide Sequences via Monomer-by-Monomer Peptide Addition onto the Microspheresa

Materials and Methods Organosilica Microsphere Synthesis and Amine Loading. Organosilica microspheres (4.60 ( 0.32 µm, measured by optical microscopy) were synthesized using 3-mercaptopropyl trimethoxysilane (MPS), purchased from Lancaster (U.K.), according to the protocol outline in Miller et al.23 The microspheres were amine modified by reacting with 3-aminopropyl trimethoxysilane (APS) and triethyleamine (TEA) in ethanol for 2 h under constant agitation. APS incorporation into the microspheres was examined as a function of increasing APS concentration to microsphere weight (0-200% w/w) using a constant TEA concentration (10% w/w). Microsphere surface amine composition was measured by XPS based on the ratio of the atomic percentage of nitrogen to silica, using a Kratos Axis ULTRA X-ray photoelectron spectrometer. Amine and thiol surface loadings of microspheres modified with 20% w/w of APS were determined by coupling FmocLys(Mtt)-OH onto the microsphere surfaces and measuring the resulting methyltrityl (Mtt) concentration in solution when the Mtt protecting group was cleaved. Mtt concentrations were quantified by UV-vis absorption at 453 nm, using a Cary 4300 UV-vis Spectrometer calibrated with increasing concentrations of Mtt deprotected FmocLys(Mtt)-OH free in the deprotection solution. Surface amine and thiol groups were individually quantified by selectively cleaving the thiolester Fmoc-Lys(Mtt)-OH bonds with basic solution treatment, for example, 30% v/v piperidine/DMF solution for 15 min, prior to measuring the Mtt concentration. Modified Amine Loading, PEG-Lys Copolymer, and Pentapeptide Synthesis. Microspheres functionalized with 20% w/w of APS (nominal amine loading ∼2.5 µmol/g, as determined by ninihydrin24 and Lys-Mtt coupling) were used for copolymer and peptide synthesis modification. All of the free acid coupling reactions were performed on 100 mg of microspheres in anhydrous peptide grade dimethylformamide (DMF) by standard 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethylaminium hexafluorophosphate (HBTU) Merrifield peptide chemistry; 20 molar excess of free acid reagent (Fmoc-Gly-OH (297.3 g/mol), Boc-Gly-OH (175.2 g/mol), Fmoc-Lys(Mtt)-OH (624.8 g/mol), FmocNH-PEG27-COOH (1544.8 g/mol), or Atto 350-COOH (326.33 g/mol)) to microsphere free amine loading were added in a 1:0.98:1 ratio of free acid reagent to HBTU to N,N-diisoproplyethylamine (DIPEA), respectively. Coupling reactions were conducted under constant agitation at room temperature for 30 min in a sealed Eppendorph tube. Peptide Fmoc protecting groups were cleaved using a 30% v/v piperidine/DMF solution for 15 min. The highly acid labile Mtt protecting group of lysine was selectively removed using a 30 min exposure to a solution of 5% v/v trifluroacetic acid in DCM under constant agitation. The resulting free amines of the lysine side chain not used during linear copolymer synthesis were subsequently capped via acetylation (overnight exposure of microspheres to 20 mol equiv of acetic anhydride in DMF under constant agitation). All of the amino acids, PEG monomers, free acid reagents and HBTU were purchased from Novabiochem, DIEPA and piperidine were purchased from Auspep. The fluorescent dye Atto 350-COOH was purchased from AttoTech (Germany) and the solvents were of peptide synthesis grade, purchased from Laboratory-Scan Analytical Sciences. The microsphere surfaces were modified to tailor the number of free amines available for polymer growth by reacting a mixture of FmocGly-OH and Boc-Gly-OH onto the microspheres. Decreasing the molar percentage of Fmoc-Gly-OH to Boc-Gly-OH from 50, 30, 20, 15, 10,

a Microsphere surface loading was tailored (0.1-50% original loading) via selective blocking of the surface by coupling increasing molar percentages of Boc to Fmoc protected glysine. Loadings of 10 and 50% were used as a platform for the trypsin protease assay.

5, 1, 0.5 to 0.1% reduced microsphere amine loading. PEG-Lys copolymers were synthesized monomer-by-monomer onto the FmocGly amines by alternating between Fmoc-PEG27-OH and FmocLys(Mtt)-OH. Unreacted amines after each amino acid coupling were capped by acetylation. Microsphere amine loading of the original, loading modified and throughout the polymer syntheses steps were quantified after each Fmoc-Lys(Mtt)-OH addition. The production of either linear or dendronic polymer structures was determined by either capping the deprotected amine side chain of lysine with acetic anhydride or leaving it a free amine group. Microsphere samples with 10% and 50% loading having no polymer, a 3-unit linear or a second-generation dendron polymer were modified with a short, fluorescently labeled pentapeptide sequence Gabba-Gly-Gly-LysVGly-Gly-Gabba-FL. A “hit” fluorophore (FL ) Atto 350) was added to the N-terminus of the peptide as a means of monitoring the enzymatic (trypsin) digestion of the peptide substrates. Schematic representations of the copolymer and peptide structures synthesized onto the microspheres are shown in Table 1. “Signal” (Trypsin Assay) to “Noise” (BSA Adsorption) Measurements. Sequencing grade modified porcine trypsin was purchased from Promega Corporation (U.S.A.). The 20 µg enzyme pellet was reconstituted in 50 µL of resuspension buffer (50 mM acetic acid). A total of 450 µL of 50 mM ammonium carbonate buffer (NH4CO3; pH 7.8, 0.02% Tween 20 (w/w)) was then added to the enzyme suspension and allowed to equilibrate to room temperature. A total of 1 mg of microspheres were dispersed in 1100 µL of 50 mM ammonium carbonate buffer (pH 7.8, 0.02% Tween 20 (w/w)). A total of 5 µL of the trypsin solution was added to the particle suspension and incubated at 37 °C. At the time points T0 (before enzyme added), 5, 30, 60, and 240 min, 50 µL aliquots of the particle suspension were removed and dispersed in 100 µL of ice-cold, acidified (pH 1.8), 0.1% sodium dodecyl sulfate (SDS) solution for flow cytometric analysis. Proteolysis of the peptide sequences were directly measured by flow cytometry as a reduction in microsphere fluorescence, normalized to the original fluorescence intensities, and indirectly by fluorometry as an increase in the supernatant fluorescence. The nonspecific adsorption of BSA onto the copolymer modified microspheres was measured by flow cytometry following the procedure and theory outlined in Kozak et al.25 Briefly, 100 µL of fluorescently tagged BSA of increasing concentrations (up to 2 mg/mL) were incubated with 1 mg of APS modified microspheres for 2 h. The adsorption of the fluorescently tagged protein onto the microsphere surface results in a linear increase in the microspheres fluorescence which is used as a flow cytometric calibration curve. The effects of surface amine loading and copolymer modified microspheres on nonspecific protein

362

Biomacromolecules, Vol. 10, No. 2, 2009

Kozak et al.

Figure 1. Tailoring microsphere surface loading: Microsphere free amine loading as a function of the original amine loading available for “blank” surfaces (b), the addition of a single PEG (4), two PEGs (2), and three PEGs (1) linearly, and for the growth of a second-generation dendron PEG (0) and third-generation dendron PEG (9). Dashed line illustrates the observed linear increase of quantified amine loading with increasing free amine surface tailoring, percentage of original loading.

adsorption were tested by incubating 0.25 mg of the microspheres with 25 µL of 2 mg/mL fluorescent BSA for 2 h. The resulting adsorbed amount of protein was calculated from the mean relative fluorescence intensity of the support microspheres. Flow cytometric measurements were made on a Dako Cytometations MoFlo with three laser excitation lines and 6 fluorescence detectors, excluding the forward and side scatter detectors. The laser and corresponding detectors used were a iCyt Visionary Bioscience Inc. (emission: 488 nm, power: 100 mW) with detection at 530 ( 20 nm and 580 ( 10 nm. Fluorescent histograms were composed of at least five thousand particles and were analyzed using Summit V4.1 software.

Results and Discussion Quantification of Tailored Diagnostic Surface Amine Loading. Amine modified silica surfaces are widely used as a substrate for the growth or attachment of biomolecules for molecular diagnostics. Quantification of the amine and subsequent biomolecule loading of diagnostics is seen as an important factor in improving assay confidence. Typically, amine modification is achieved through the exposure of a silica surface to an amine-functionalized organosilane, such as APS, in the presence of an organic base catalyst. MPS microspheres were amine modified with increasing concentrations of APS to examine amine loading of the microsphere support. Even at high APS reaction concentrations, greater than 100 times the microsphere mass, the monomer composition of the microsphere surface remained approximately 20%, as calculated from XPS measurements (Supporting Information). This indicates that the microsphere surface is composed of a mixture of amine and thiol groups which can compete and impact on subsequent biomolecule immobilization loading. Although representative of the surface loading, XPS measurements are neither quantitative nor surface exterior specific. To overcome this, the coupling and selective cleavage of Fmoc-Lys(Mtt)-OH to the microsphere exterior was used to both quantify and differentiate between the two nucleophilic functional groups (i.e., thiols and primary amines) available at the microsphere surface. In the presence of an activated ester derivative of Fmoc-Lys(Mtt)-OH, both thiols and primary amines will react, producing thioester and amide products, respectively. Both of these products are stable in neutral to acidic solvents. Exposure of Fmoc-Lys(Mtt)-OH functionalized microspheres to acidified DCM thereby releases the yellow colored Mtt cation without cleaving amide or thioester bonds. This

intense chromophore can then be used as a measure of the total concentration of thiols and amines available for Fmoc-Lys(Mtt)OH attachment. By treating the Fmoc-Lys(Mtt)-OH functionalized microspheres with an organic base, the thioester bonds are selectively cleaved, leaving the more stable amide bonds intact. Consequently, the amount of Mtt cation released from base-treated surfaces is indicative of the amine surface loading. The difference between the untreated and base-treated surfaces thereby provides a means of determining the concentrations of the thiol and amine functional groups available at the microsphere surface. This spectroscopic test was used to probe the microsphere surface functional groups and was also applied over the course of the polymer growth to monitor synthesis yields and assess final polymer density. Fmoc-Lys(Mtt)-OH conjugation was used to establish the functional loading of MPS microspheres modified with 20% w/w of APS. These microspheres were found to possess an external amine and thiol loading of 0.49 ( 0.03 and 0.41 ( 0.05 µmol/g, respectively. Using the N2 adsorption BET surface area measurements of the microspheres (0.9 m2/g) and simple molecular modeling, the theoretical surface loading was calculated to be 2.5 and 2.1 µmol/m2 of amines and thiols, respectively. This is assuming Fmoc-Lys(Mtt)-OH has a radius of 7 Å, occupies an area of 1.54 nm2 and that the average distance between Si-O-Si bonds of the microsphere is 3.2 Å. The technique was validated by an independent amine loading measurement by ninhydrin test which gave a microsphere loading of 3 µmol/m2. Spectroscopic trityl analysis of FmocLys(Mtt)-OH conjugation further indicated that the exterior of the microspheres display APS and MPS monomer groups in roughly equimolar amounts. This correlates with the findings of Corrie et al.24 that the APS monomer is incorporated into the microspheres with an associated concentration gradient. Customized surface loadings were achieved by coupling specific molar mixtures of acid and base liable glycine (Boc and Fmoc protected, respectively) onto the surface prior to polymer and biomolecule modification. As the polymer synthesis was Fmoc directed, the amine surface loading was reduced or “poisoned” by increasing the molar percentage of Boc-Gly in the initial coupling solution. The resulting microsphere free amine loading as a function of this customizing of the surface loading is shown in Figure 1. As expected, the free amine density of these tailored surfaces increases linearly proportional to the percentage of loading modification (illustrated by the

PEG-Lysine Copolymer Dendrons

Biomacromolecules, Vol. 10, No. 2, 2009

363

Table 2. XPS Percentages of C-O (286.5 eV) Peaks of the Carbon 1s Spectra for Copolymer Linear and Dendronic Growth on Micospheres with 10 and 50% Modified Amine Loadingsa microsphere loading

no. polymer

1 linear comonomer unit

2 linear comonomer units

1st gen. dendron

3 linear comonomer units

2nd gen. dendron

50% 10%

4 2

37 14

19

63 49

51 24

74 67

a

Each growth step represents the addition of a PEG-Lys comonomer unit in either a linear or dendronic branching motif.

dashed line in Figure 1) to a maximum amine loading density of 270 nmol/g. From these customized loading surfaces, linear and dendronic PEG-Lys copolymers were synthesized and the resulting terminal amine loadings were investigated. Grafting one to three consecutive PEG-Lys comonomer units in a linear conformation onto the microsphere surface resulted in an initial linear increase in amine loading equal to the original tailored surface loading (Figure 1). Each of the loadings then went through a transition where additional surface sites did not correspond to an increase in polymer loading. The amine loading transition for the polymers were observed to correlate to twice the polymer radius of gyration, being Rg ) (n1/2l)/6, where n ) number of monomer units and l is the monomer length. Polymer overlap and steric exclusion for the grafting of linear copolymers with 1, 2, and 3 comonomer units is expected to occur at 45, 22, and 15% of the original amino acid loading, respectively. Assuming that a 100% grafting density would correspond to a microsphere amine loading of 0.5 µmol/m2. In comparison to surface-attached linear PEG chains, the growth of dendronic polymer structures induced a very rapid increase in free amine loading. The resulting maximum amine loading was greater than, and occurred before, that of linear polymers with a comparable number of subsequent comonomer unit additions (generations). Dendrimer and dendron structures are attractive for improving signal or functional intensity because the number of end groups increase as a function of Bg, where B is the number of branching arms per generation (g). As expected in this study, where each lysine residue results in two branches being formed, the second-generation dendron loading was approximately twice that of the first-generation dendron and four times that of the linear polymer of the same generation. Dendronic copolymers were also observed to go through a more distinct transition at much lower tailored surface loadings, approximately 5% loading for the second-generation dendron compared to 15% loading for the three unit linear PEG. This is due to the branched dendronic structure giving rise to smaller and denser polymer structure compared to linear PEG, sterically limiting the number of dendritic polymers formed on the surface. Linear and dendronic polymers synthesized on micospheres with modified amine loadings of 10 and 50% were characterized by XPS. Modifying a surface with a PEG layer has been shown to increase the C-O (286.5 eV) peak intensity of the XPS C 1s spectra, relative to the amount of PEG on the surface.8,13 As expected, increasing the microsphere loading (from 10 to 50%), the number of PEG-Lys comonomer additions (from 1 to 3), and introducing polymer branching (first- to second-generation dendrons) all gave rise to an increase in the C-O peak intensity, as given in Table 2. Although the C-O percentages for the 50% were greater than the 10% loaded surfaces, for similar polymer structures, they were less than the theoretical 5-fold increase. Similar effects were observed with increasing polymer length and dendronic branching. These effects are likely due to the comonomer chemical structure, unlike pure PEG structures, giving rise to an increase in both the C-O and C-C (285 eV) peaks along with steric limitations during polymer synthesis. As the surface loading or polymer size increases the interspatial

distance between free amine groups becomes smaller than the radius of gyration of the reactive species being coupled. This results in incomplete, truncated, polymer structures, reducing the amount of PEG on the surface. Assessing Diagnostic Signal-to-Noise. The impact of the grafted PEG-Lys copolymer structure on improving the signal of molecular diagnostic assays was examined via a bead-based proteolysis assay. A short pentapeptide sequence (GGKGG) representing a preferred substrate to the protease trypsin was synthesized directly onto the terminal amine functional groups of the PEG-Lys copolymers. Microsphere samples with 10 and 50% loading modified with no polymer, a three-unit linear or a second-generation dendron polymer (as illustrated in Table 1) were compared. The extent and qualitative rate of proteolysis was measured through the decrease in the mean relative fluorescence intensity of the peptide functionalized microspheres and indirectly by the increase in the supernatant fluorescence. As shown in Figure 2A,B, the dendronic PEG-Lys copolymers displayed the greatest improvement in the tryspin assay signal. Peptide sequences grown directly onto the microsphere surface showed no change in fluorescence upon trypsin digestion. This inability of trypsin to digest the surface grown sequences is likely due to steric constraints of the enzyme to access the surface bound peptides along with the nonspecific adsorption of trypsin on the surface further blocking peptide access. Dendronic polymer coated surfaces gave rise to higher initial peptide signals than the linear copolymers, being 965 and 475, and 1000 and 590 MFI for the 10 and 50% dendronic and linear polymer loaded surfaces, respectively. Dendronic modified surfaces also exhibited a greater change in trypsin digestion signal. Flow cytometric and fluorometry measurements of the change in on-bead and supernatant fluorescence both indicated that the greatest peptide digestion occurred on the dendronic polymer modified surfaces. The improved proteolysis of the dendronic polymer peptide substrate is believed to be due in part to the higher polymer density of the dendronic structure giving rise to an increased enzyme stability through reduced nonspecific interactions with the assay surface and the increased peptide loading giving rise to an outward “display” of the peptide into solution. Complete digestion of microsphere-bound peptides however, was never achieved. Incomplete digestion, indicated by the retention of microsphere fluorescence after prolonged exposure to protease, is most likely the result of fluorophore attachment to truncated polymers, peptides or other particle functional groups producing fluorescent products unrecognizable by trypsin. The antifouling characteristics associated with both polymer loading and structure was tested by measuring the adsorption of BSA onto the PEG-Lys copolymer-functionalized microspheres. Serum proteins are known to readily adsorb onto assay surfaces, potentially blocking the access of biomarkers to surface immobilized probe molecules, reducing assay signal and inducing false negative results. Additionally, target proteins may also adsorb to the surface without the involvement of the probe biomolecule, leading to false positive outcomes. Nonspecific

364

Biomacromolecules, Vol. 10, No. 2, 2009

Kozak et al.

Figure 2. Influence of polymer structure on assay signal: Trypsin cleavable pentapeptide modified microspheres with surface loadings of 10% (open figures) and 50% (closed figures) with no polymer spacer (O) and (b) a three-unit linear PEG-Lys copolymer (4) and (2) and a secondgeneration dendronic structure (0) and (9) monitored by (A) flow cytometry of microsphere fluorescence and (B) fluorimetry measurements of the cleaved fluorescently tagged peptide in the assay supernatant. Dashed lines represent general data trends.

grafting densities the second-generation dendronic PEG-Lys was the most effective in preventing protein adsorption, reducing nonspecific protein adsorption by 85%.

Conclusions

Figure 3. Influence of polymer structure on assay noise: Nonspecific BSA adsorption onto surface loading modified “blank” surfaces (b), linear copolymer with a single PEG-Lys (4), linear with two PEG-Lys copolymer units (2), and linear with three PEG-Lys copolymer units (1). A first-generation PEG-Lys copolymer dendron (0) and a secondgeneration PEG-Lys copolymer dendron PEG (9). Dashed lines represent general data trends.

protein adsorption on the copolymer modified microspheres was quantified by flow cytometry, on a particle-by-particle basis, taking an average of 5000 particles per sample. The adsorption of BSA onto particles of increasing amine loading and Lys-PEG copolymer modified microspheres are shown in Figure 3. Microspheres with tailored amine loading all gave similar BSA adsorbed amounts (2.0 mg/m2) irrespective of the increasing amine loading. Linear copolymer addition of 1, 2, and 3 comonomer units all exhibited a similar decrease in BSA adsorption with increasing grafting density. At low grafting densities (0.1-1% original loading) all of the copolymers showed a marked (approximately 60%) reduction in BSA adsorption. The antifouling efficacy of the surface-bound polymers appeared to be volume fraction dependent and independent of the copolymer structure. Consequently, the dendronic copolymers, which possess a greater polymer volume fraction compared to linear structures, provided the greatest protection against BSA adsorption, especially at low grafting densities. The first-generation dendron and three-unit linear copolymers, which were predicted to have similar polymer fractions, showed similar protein adsorption profiles. At higher

Linear and dendronic-like PEG-Lys copolymers were synthesized directly onto organosilica microsphere supports displaying customized surface amine loadings. A novel method of quantifying the surface composition and modified amine loadings of the organosilica microspheres was achieved by spectroscopically measuring the cleaved Mtt chromophore concentration of coupled Fmoc-Lys(Mtt)-OH on the microspheres. The effects of polymer structure and surface loading on the signal-to-noise ratio of the microsphere based assay were examined in two parts: (i) changes in enzymatic digestion assay signal intensity and (ii) the reduction in nonspecific adsorption noise of proteins onto the assay surface. Of the polymers tested, dendronic polymers exhibited the greatest improvements in assay performance. These polymers increased assay signal intensity and decreased the amount of nonspecifically adsorbed proteins on the assay surface. Consequently, dendronic like PEG-Lys copolymers hold promise for improving the performance of a wide range of assay materials where low grafting loading high density PEG is required. Acknowledgment. This project was financially supported by the Australian Research Council (FF0455861). Supporting Information Available. XPS analysis of APS modified MPS microspheres indicated that the APS monomer was incorporated into the microspheres rather than forming an APS multilayer film on the surface. The MPS microspheres were found to be comprised of a maximum of 20% APS monomer. This material is available free of charge via the Internet at http:// pubs.acs.org.

References and Notes (1) Rostom, A.; Dube, C.; Cranney, A.; Saloojee, N.; Sy, R.; Garritty, C.; Sampson, M.; Zhang, L.; Yazdi, F.; Mamaladze, V.; Pan, I.; Macneil, J.; Mack, D.; Patel, D.; Moher, D. The diagnostic accuracy of serologic tests for celiac disease: A systematic review. Gastroenterology 2005, 128 (4), S38-S46. (2) Hellstrom, I.; Raycraft, J.; Hayden-Ledbetter, M.; Ledbetter, J. A.; Schummer, M.; McIntosh, M.; Drescher, C.; Urban, N.; Hellstrom,

PEG-Lysine Copolymer Dendrons

(3) (4)

(5)

(6)

(7) (8) (9) (10) (11) (12) (13) (14)

K. E. The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer Res. 2003, 63 (13), 3695–3700. Hartwell, L.; Mankoff, D.; Paulovich, A.; Ramsey, S.; Swisher, E. Cancer biomarkers: A systems approach. Nat. Biotechnol. 2006, 24 (8), 905–908. McIntosh, M. W.; Drescher, C.; Karlan, B.; Scholler, N.; Urban, N.; Hellstrom, K. E.; Hellstrom, I. Combining CA 125 and SMR serum markers for diagnosis and early detection of ovarian carcinoma. Gynecol. Oncol. 2004, 95 (1), 9–15. Gong, P.; Grainger, D. W. Nonfouling surfacessA review of principles and applications for Microarray capture assay designs. Methods in Molecular Biology; Humana Press Inc.: Totowa, NJ, 2007; pp 5992. Vermette, P.; Meagher, L. Interactions of phospholipid- and poly(ethylene glycol)-modified surfaces with biological systems: relation to physico-chemical properties and mechanisms. Colloids Surf., B 2003, 28 (2-3), 153–198. Greenwald, R. B.; Choe, Y. H.; McGuire, J.; Conover, C. D. Effective drug delivery by PEGylated drug conjugates. AdV. Drug DeliVery ReV. 2003, 55 (2), 217–250. Chen, A.; Kozak, D.; Battersby, B. J.; Trau, M. Particle-by-particle quantification of protein adsorption onto poly(ethylene glycol) grafted surfaces. Biofouling 2008, 24 (4), 267–273. Szleifer, I. Polymers and proteins: Interactions at interfaces. Curr. Opin. Solid State Mater. Sci. 1997, 2 (3), 337–344. Cloninger, M. J. Biological applications of dendrimers. Curr. Opin. Chem. Biol. 2002, 6 (6), 742–748. Svenson, S.; Tomalia, D. A. Commentary - Dendrimers in biomedical applicationssReflections on the field. AdV. Drug DeliVery ReV. 2005, 57 (15), 2106–2129. Guillaudeu, S. J.; Fox, M. E.; Haidar, Y. M.; Dy, E. E.; Szoka, F. C.; Frechet, J. M. J. PEGylated dendrimers with core functionality for biological applications. Bioconjugate Chem. 2008, 19 (2), 461–469. Sofia, S. J.; Premnath, V.; Merrill, E. W. Poly(ethylene oxide) grafted to silicon surfaces: Grafting density and protein adsorption. Macromolecules 1998, 31 (15), 5059–5070. Yeh, P. Y. J.; Kainthan, R. K.; Zou, Y. Q.; Chiao, M.; Kizhakkedathu, J. N. Self-assembled monothiol-terminated hyperbranched polyglycerols on a gold surface: A comparative study on the structure, morphology, and protein adsorption characteristics with linear poly(ethylene glycol)s. Langmuir 2008, 24 (9), 4907–4916.

Biomacromolecules, Vol. 10, No. 2, 2009

365

(15) Benhabbour, S. R.; Sheardown, H.; Adronov, A. Protein resistance of PEG-functionalized dendronized surfaces: Effect of PEG molecular weight and dendron generation. Macromolecules 2008, 41 (13), 4817– 4823. (16) Feng, X. S.; Taton, D.; Chaikof, E. L.; Gnanou, Y. Toward an easy access to dendrimer-like poly(ethylene oxide)s. J. Am. Chem. Soc. 2005, 127 (31), 10956–10966. (17) Berna, M.; Dalzoppo, D.; Pasut, G.; Manunta, M.; Izzo, L.; Jones, A. T.; Duncan, R.; Veronese, F. M. Novel monodisperse PEG-dendrons as new tools for targeted drug delivery: Synthesis, characterization and cellular uptake. Biomacromolecules 2006, 7 (1), 146–153. (18) Marcon, L.; Kozak, D.; Battersby, B. J.; Chappell, K. J.; Fairlie, D. P.; Young, P.; Trau, M. A dual-purpose synthetic colloidal platform for protease mapping: substrate profiling for Dengue and West Nile virus proteases. Anal. Biochem. 2008, 376 (1), 151–153. (19) Battersby, B. J.; Lawrie, G. A.; Johnston, A. P. R.; Trau, M. Optical barcoding of colloidal suspensions: Applications in genomics, proteomics and drug discovery. Chem. Commun. 2002, (14), 1435–1441. (20) Lawrie, G. A.; Battersby, B. J.; Trau, M. Synthesis of optically complex core-shell colloidal suspensions: Pathways to multiplexed biological screening. AdV. Funct. Mater. 2003, 13 (11), 887–896. (21) Corrie, S. R.; Lawrie, G. A.; Trau, M. Quantitative analysis and characterization of biofunctionalized fluorescent silica particles. Langmuir 2006, 22 (6), 2731–2737. (22) Surawski, P. P. T.; Battersby, B. J.; Lawrie, G. A.; Ford, K.; Ruhmann, A.; Marcon, L.; Kozak, D.; Trau, M. Flow cytometric detection of proteolysis in peptide libraries synthesised on optically encoded supports. Mol. BioSyst. 2008, 4 (7), 774–778. (23) Miller, C. R.; Vogel, R.; Surawski, P. P. T.; Jack, K. S.; Corrie, S. R.; Trau, M. Functionalized organosilica microspheres via a novel emulsion-based route. Langmuir 2005, 21 (21), 9733–9740. (24) Corrie, S. R.; Vogel, R.; Keen, I.; Jack, K.; Kozak, D.; Lawrie, G. A.; Battersby, B. J.; Fredericks, P.; Trau, M. A structural study of hybrid organosilica materials for colloid-based DNA biosensors. J. Mater. Chem. 2008, 18 (5), 523–529. (25) Kozak, D.; Chen, A.; Trau, M. Profiling protein-surface interactions of multicomponent suspensions via flow cytometry. Langmuir 2008, 24 (4), 1204–1211.

BM8011314