Poly-Carboxylated Dextran as a Multivalent Crosslinker: Synthesis

Subscriber access provided by Washington University | Libraries

Interface Components: Nanoparticles, Colloids, Emulsions, Surfactants, Proteins, Polymers

Poly-Carboxylated Dextran as a Multivalent Crosslinker: Synthesis and Target Recognition of the Antibody-Nanoparticle Bioconjugates in PBS and Serum. Filip Kunc, Colin Moore, Rachel E. Sully, Andrew J. Hall, and Vladimir Gubala Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b03833 • Publication Date (Web): 28 Feb 2019 Downloaded from http://pubs.acs.org on March 2, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Poly-Carboxylated Dextran as a Multivalent linker: Synthesis and Target Recognition of the Antibody-Nanoparticle Bioconjugates in PBS and Serum. Filip Kunca, Colin Mooreb, Rachel E. Sullyc, Andrew J. Hallc and Vladimir Gubalac*. a) National Research Council Canada, 100 Sussex Drive, Ottawa, Ontario, K1N 0R6, Canada b) Italian Institute of Technology, 30 Via Morego, 16163 Genoa, Italy c) Medway School of Pharmacy – Universities of Greenwich and Kent, Anson Building, Central Avenue, Chatham ME4 4TB, UK ABSTRACT: Nanoparticles functionalized with antibodies on their surface are used in wide range of research applications. However, the bioconjugation chemistry between the antibodies and the surface of nanoparticles can be very challenging, often accompanied by several undesired effects such as nanoparticle aggregation, antibody denaturation or poor target recognition of the surface-bound antibodies. Here we report on a synthesis of fluorescent silica nanoparticle-antibody conjugates (NP-Ab), in which poly-carboxylated dextran is used as the multivalent linker. Firstly, we present a synthetic methodology to prepare poly-carboxylated dextrans with molecular weights of 6kDa, 40kDa and 70kDa. Secondly, we used the water-soluble, poly-carboxylated dextrans as a multivalent spacers/linkers to immobilize antibodies onto fluorescent silica nanoparticles. The prepared NP-Ab conjugates were tested in a direct binding assay format in both PBS buffer and whole serum to investigate the role of the spacer/linker on the capacity of the NP-Ab to specifically recognize their target in ‘clean’ and also in complex media. We have compared the Dextran conjugates with two standards: a) NP-Ab with antibodies attached on the surface of nanoparticles through classical physical adsorption method and b) NP-Ab where an established poly(amidoamine) (PAMAM) dendrimer was used as the linker. Our results showed that the polycarboxylated 6 kDa dextran facilitates antibody immobilization efficiency of nearly 92%. This was directly translated into the improved molecular recognition of the NP-Ab, which was measured by direct binding assay. The signal-to-noise ratio in buffered solution for the 6kDa Dextran NP-Ab conjugates was 81, nearly 3-times higher than that of PAMAM G4.5 conjugates and 9-times higher than the physically adsorbed NP-Ab sample. In whole serum, the effect of 6kDa Dextran was more hindered due to the formation of protein corona but the signal-to-noise ratio was at least double that of the physically adsorbed NP-Ab conjugates.

Introduction The advances in controllable nanoparticle (NP) design over the past two decades has led to predictions that nanotechnology will revolutionize medicine, particularly the fields of biosensing and drug delivery1-4. In both cases, the carefully engineered are expected to perform well and with good reproducibility in both simple aqueous solutions (i.e. buffered solutions) as well as in complex media (i.e. whole blood, serum, culture medium etc.)5, 6. This is however, quite challenging to achieve. It is expected that NPs will provide a specific function in various biological processes, which means they have to be equipped with unique ligands that will recognize their specific receptors7. One of the most common strategies for generating a selective nanoprobe is to functionalize the nanomaterial with antibodies, thus imparting the capacity for specific target recognition8, 9. For biosensing, this allows for the detection of disease biomarkers using immunoassays or point-of-care devices and the ability to assess the wellness of a patient. In the case of drug delivery, typically cancer therapy, the use of targeted-NPs have been proposed as a way of ensuring the therapeutic nanomaterial specifically recognizes overexpressed receptors on the cell membrane and therefore localizes in the disease site10, 11. However, despite the undoubted potential of these specialized NPs it is clear that very few examples have been translated from the laboratory to commercial success; a sentiment echoed in literature by Juliano12 and Venditto13. Huge variety of nanoparticles have been reported in the literature in the past decade14, varying in the material from which they are made, their size, shape, chemico-physical properties and surface chemistry among other parameters. Many of the standard methods for functionalizing NPs with biomolecules lack robustness and reproducibility, which means scientists often resort to elaborate and/or expensive strategies to generate antibody-coated NPs (NPAb). This in turn leads to limited scope for large scale NP-Ab production and a difficulties in commercial translation5. The most frequently used method of generating NP-Ab is by attaching proteins to the NPs via a linker molecule8. The choice of linking agent is fundamental to overall functionality of the NP-Ab as it plays an important role in (i) imparting the desired functional groups to the NP surface to allow for as-easy-as-possible protein attachment and (ii) preventing possible protein denaturation by acting as a spacer between the antibody and the particle surface. However, the introduction of the linking agent to the ‘plain’ NP surface (Figure 1) can negatively affect the colloidal stability of the NP-intermediate and subsequently cause the resulting NP-Ab to aggregate and precipitate rapidly from solution. We have previously showed that this was the case when multivalent carboxy-

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 19

terminated poly(amidoamine) (PAMAM, generation 4.5) dendrimers were employed to conjugate proteins to dye-doped silica NPs15, 16. Specifically, it was demonstrated that attaching negatively-charged EDC/sulfoNHS-activated PAMAM to the negatively charged surface of silica NPs lead to stable PAMAM-coated colloids, and subsequently allowed for stable NP-Ab to be produced. In contrast, the use of uncharged EDC/NHS activated PAMAM led to the aggregation of the subsequent PAMAM-coated silica NPs because of reduced NP-NP electrostatic repulsion and lower NP solubility5. While this EDC/sulfoNHS activated PAMAM method enabled the reliable production of stable NP-Ab, the major drawback was the relatively high cost of the functionalized dendritic macromolecule, which was attributed to its multistep synthesis and laborious purification. To this end, we have endeavored to develop an alternative linker that exhibits the multivalency and COOH-functionality of dendrimers for straightforward EDC/(sulfo)NHS coupling reactions. One macromolecule that is a potentially well-suited alternative to PAMAM is dextran; an inexpensive polymeric carbohydrate that has been successfully applied to create functionalized surfaces in several, commercially available biosensor devices17-19, but in its poly-carboxylated form has remained overlooked as a linker in NPantibody conjugation. The cost of material and reagents associated with the production of poly-carboxylated dextran is approximately 50 times lower (per gram) when compared to PAMAM dendrimer. Generally, the functional derivatives of polysaccharides benefit from the inherent biocompatibility of the template carbohydrate20. Carboxylated derivatives of dextran have been prepared for applications in cancer immunotherapy21, 22 or formulation of prodrugs for inflammatory bowel disease treatment20, 23. In order to engineer a comparable macromolecule to PAMAM dendrimer, we have developed a poly-carboxylated form of dextran by reacting the dextran polymer with succinic anhydride. Three dextran polymers with different molecular weights/sizes were used in this study: 6kDa, 40kDa and 70kDa. These three poly-carboxylated dextrans were used as linkers to attach goat-anti-human IgG antibodies to dye-doped silica NPs. In terms of hydrodynamic radius (i.e. the apparent 3D size), 40 kDa dextran is well comparable to PAMAM G4.5 dendrimer, which was subject of our previous studies. The higher MW dextran (70kDa) and the lower MW dextran (6kDa) were chosen in this study to assess how the size of the polymer influences the overall colloidal stability of the resulting antibody-coated NP.

Figure 1: A schematic description of Ab-NPs preparation. We have used direct-binding immunoassay format to compare the capacity of all the dextran-linked NP-Ab and PAMAM-linked NP-Ab to specifically recognize human IgG in PBS buffer and serum. The rationale was that the PBS buffer represents a ‘clean’ medium, which is frequently used in biomedical diagnostic applications or in basic academic research. The serum, on the other hand is a ‘complex’ medium, which represents more accurate environment for nanoparticles that are meant to be used for targeted drug delivery applications or used in point-of-care biosensors where sample pre-processing steps are not possible24. Results and discussion Dextran is as a simple polysaccharide molecule bearing relatively poorly reactive hydroxyl groups that cannot facilitate reliable conjugation of antibodies. As a material, it is frequently just used to provide a biocompatible shell around various types of nanomaterials that are studied in vitro or in vivo16, 25. Our strategy was to modify it with high number of more reactive groups such as carboxylic acids that can be further used for common EDC/(sulfo)NHS coupling. Commercially available poly-carboxylated dextrans are usually prepared by nucleophilic substitution reaction with chloroacetic acid (i.e. to prepare carboxymethylated dextran)26, however these possess relatively low degree of carboxylation (only 5-10% of -OH groups per dextran molecule are converted to -COOH). Our initial attempts to increase the yield of modification with more reactive 3-bromopropionic acid was not successful because of the excessive formation of side product via initial de-homogenization of 3-brompropionic acid followed by Michael addition (ESI, Figure S.1). Therefore, succinic anhydride (SA) was selected to be used in anhydrous DMF as an aprotic


Page 3 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

environment (Figure 2). The use of a cyclic anhydride has offered a convenient method to modify dextran with -COOH groups. The reaction conditions were initially optimized using 40 kDa dextran. Firstly, the number of -OH groups present in the 40 kDa dextran polymer (ESI, Table S.1) were estimated, followed by the application of SA in three different molar ratios to OH groups (0.1×, 2×, and 5×) in order to vary the degree of carboxylation.The products (40Dex - 0.1×, 2×, and 5×) were characterized by FTIR and 1H NMR that qualitatively confirmed the presence of newly introduced succinate moiety (ESI, Figures S.2, S.3). 1H NMR spectra of succinated products showed resonances that were broadened, presumably due to decreased molecular tumbling probably caused by the gain in molecular weight and the change in the surface chemistry of the macromolecule. We have also attempted to quantitatively measure the number of succinate moieties by gravimetric comparison of the succinated and unmodified dextran, and also by a simple acid-base titration (ESI, Table S.1); a method successfully applied to the analysis of carboxymethylated dextrans previously27. According to the acid-base titration, 2× molar excess of SA per OH group was found to introduce ≈ 178 carboxylic acids groups on a single molecule of 40 kDa dextran. Interestingly, this was a lower amount (by ≈ 40 %) than it was estimated from the gravimetry. This discrepancy was attributed to the fact that gravimetry requires all samples to be absolutely anhydrous to compare the gain/loss in weight, which is quite challenging with hygroscopic material such as dextran. Although, gravimetry overestimated the absolute number of -COOH groups per dextran, it was still a useful tool for relative comparison of the synthesized polycarboxylated dextran samples (i.e. 0.1×, 2×, and 5×). The conditions with 2× molar excess of SA per -OH group were deemed optimal, because the modified dextran remained readily soluble in aqueous media and the number of reactive carboxylic groups determined by acid-base titration (≈ 178) was relatively high and comparable to the model PAMAM G4.5 dendrimer (128 -COOH groups in generation 4.5). All subsequent dextran carboxylation reactions (for 6 kDa and also for 70 kDa dextran) were performed with 2× molar excess of SA per -OH group. For simplicity, we will refer to the polycarboxylated dextran moleclues as 6Dex, 40Dex and 70Dex when they are discussed below.

Figure 2: A reaction Scheme and the mechanism of dextran esterification with succinic anhydride. Dextran 6 kDa: n = 33, Dextran 40 kDa: n = 222, Dextran 70 kDa: n = 389 In the next step, we have compared the capacity of 40Dex and the previously established PAMAM G4.5 linker to conjugate goat anti-human-IgG antibodies to fluorescent silica NPs. Fluorescein-doped silica NPs were chosen as the model nanomaterial because these are well researched and applicable in bio-sensing and in-vivo applications28-30. Bioconjugation reaction between antibodies and nanoparticles is usually a multistep process, involving at least one or two purification/separation step(s)31. The resulting NP-Ab have the tendency to reversibly or irreversibly aggregate, which is a very undesired feature14. Therefore, one specific aim was to assess the effect of PAMAM G4.5 and 40Dex to maintain colloidal stability of the nanomaterial during and after the multi-step bioconjugation reaction (Figure 1). The colloidal stability at each reaction step was monitored by dynamic light scattering (ESI, Table S.1). The ‘plain’ silica nanoparticles (152.8 nm in diameter on average) were synthesized according to a well-known Stöber method32, the negatively charged surface of the silica NP was then treated with APTES to introduce amine functionality as a reactive handle for further conjugation steps. The successful grafting of -NH2 groups was accompanied by a change in zeta potential (change from -37.1 mV to -29.8 mV, Table 1) as well as by a minor increase (by 6.4 nm) in the hydrodynamic diameter, suggesting modest amine surface modification. Lower density of amine groups on the surface of silica NPs is more desirable for colloidal stability purposes33 and also because densely aminated silica NPs with overall positive charge pose greater toxicity than their negatively charged counterparts34. The second step in the bioconjugation reaction was the activation of the carboxy groups of the respective linkers (i.e. PAMAM G4.5 and 40Dex) to amine-reactive sulfo-NHS esters, and their subsequent reaction with the aminated silica NPs. Successful attachment of dextran and PAMAM linkage to the aminated-NP surface was confirmed by a significant increase in size in both respective nanoparticles (Table 1). Interestingly, both linkers increased the hydration sphere of the measured nanoparticles considerably. The overall size of the NP40Dex was higher than that of NP-PAMAM despite the fact that both linkers were approximately of the same size (~ 5nm in diameter). The difference can be attributed to the more-hygroscopic nature of dextran, which results in a nanoparticles with increased hydration sphere due to the increase in the hydrogen boding sites35. Nonetheless, the increase in the average hydrodynamic diameter could be also assigned to a small degree of agglomeration, measured by the polydispersity index. The APTES-coated NP (NP-APTES) were highly uniform and monodisperse with PDI= 0.049. The polydispersity index of NP-40Dex and NP-PAMAM increased to 0.137 and 0.150 respectively.


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 19

The final step in the bioconjugation reaction was to immobilize goat anti-human IgG antibodies on the ‘activated’ surface of linkercoated NPs. It is noteworthy to mention that when antibodies were conjugated on the NP surface, the average hydrodynamic diameter decreased. This is probably due to the fact that antibody-coated-nanoparticles have the tendency to be more colloidally stable and monodisperse (PDI values decreased to 0.105 and 0.066 respectively). This finding was in agreement with previously published work by us and other groups5, 36. The overall zeta potential of antibody coated NPs was more than -26 mV, which renders it sufficient for a stable colloidal dispersion. Table 1: Physiochemical characterization of NPs at various reaction steps. All Measurements are done in DI Water pH 7. Sample name

Hydrodynami c Diameter

PDI

Zeta Potential (mV)

(nm)

Plain NP

152.8 ± 0.2

0.066

37.1±0.6

NP-APTES

159.2±0.3

0.049

29.8±0.8

NP-40Dex

224.4±1.3

0.137

23.2±0.6

NP-40Dex-Ab

186.9±2.4

0.105

24.1±0.1

NP-PAMAM

204.3±1.1

0.150

18.5±0.8

NP-PAMAM-Ab

172.5±2.3

0.066

27.3±1.0

The colloidal stability of prepared NP-linker-Ab conjugates was further probed by transmission electron microscopy. No significant agglomerates/aggregates were observed, all prepared samples showed reasonably good degree of uniformity in shape and size. (Figure 3).


Page 5 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3: TEM analysis of FITC-SiO2 NPs: (a) NP-APTES as a starting material; after modification with goat anti-human IgG linked via: 40Dex (b), and PAMAM G4.5 (c) using EDC/sulfo-NHS chemistry. All NPs appear to be reasonably uniformed in size and shape, showing no apparent aggregation. The images show the nanoparticles in their ‘dry’ state, where the solvent (DI water) was removed after the samples were deposited on the copper/carbon TEM grid at 1 mg/mL. Using the optimized, established protocol with 2-fold excess of SA to -OH group, two additional dextran molecules with molecular weights of 6kDa and 70 kDa were carboxylated. The results are summarized in Table 2. Table 2: Determination of carboxylic group content on 70, 40 and 6 kDa dextrans modified with 2 x molar excess of SA per OH group by acid-base titration. Dextran linker

% of modified hydroxyl groups

COOH dextran

70Dex

27 %

320 ± 2

40Dex

27 %

178 ± 2

6Dex

37 %

36 ± 2

per

In each case, the reaction with 2× molar excess of SA resulted in a poly-carboxylated dextran, in which 27-37 % of the available OH groups were functionalized with -COOH groups. This corresponds to 1.1, 0.80, and 0.83 -COOH groups per glucose unit in 6Dex, 40Dex and 70Dex, respectively. The total number of antibodies that were immobilized via a linker (dextran or PAMAM) was measured by fluorescence. Linkercoated NPs were reacted with AlexaFluor647 (AF647)-labeled antibodies to produce NP-AF647Ab conjugates, using 6 kDa, 40 kDa and 70 kDa poly-carboxylated dextrans. PAMAM G4.5 (with 128 COOH groups) was also used as a reference linker. A control batch of NP-AF647Ab conjugates was prepared in the absence of any linker, simply by physically adsorbing AF647-labeled antibodies onto the NP surface. The conjugation efficiency and the stability of the NP-AF647Ab samples over the number of washing steps was quantified. This was done by measuring the signal of unbound AF647-labelled antibodies (proportional to the concentration of the AF647-labeled antibodies using a calibration curve) that remained in the supernatant following NP-Ab purification/washing steps. The data, presented in Figure 4, show the percentage of AF647-antibodies that were immobilized on the NP surface.

Figure 4: The relative linker immobilization efficiency is presented as the percentage of antibodies that remained attached during EDC/sulfo-NHS coupling and throughout the 4 washing steps. 100 % corresponds to the initial amount of antibodies used in the conjugation reaction. The grey bars (4th wash) represents the portion of antibodies immobilized on the surface after the final wash. The antibody adsorption/chemisorption, washing experiment was performed in three independent replicates (n=3). Figure 4 shows a relative comparison of the linker immobilization efficiency, using multiple Ab-NP conjugates prepared from three independent replicates of FITC-doped nanoparticles. In the case of physical adsorption method, the antibodies were initially adsorbed on the surface in high amount (86% from the solution was adsorbed) but they were continuously desorbed throughout the purification steps. The relatively large error bars also suggest that the physical adsorption process is difficult to control with acceptable


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 19

accuracy. In case of PAMAM, small degree of desorption was observed after the first washing step, however only a small degree of desorption of antibodies was observed in the remaining washes. 6Dex was the most effective linker molecule as it could conjugate 89 % of antibodies introduced to the reaction. 40Dex and 70Dex were able to capture 21 and 51 % of the initial amount of antibodies respectively, suggesting that the immobilization efficiency of dextran is not simply related to its size. Presumably, other factors such as macromolecule shape or charge play a significant role in the process. It is interesting to note that all samples in which Dextran was used as a linker showed relatively negligible desorption of antibodies from the NP surface throughout the conjugation and washing steps. Although it is highly desirable to produce NPs with high density of antibodies on their surface, it is more difficult to estimate how many antibodies are immobilized in their active mode and orientation. To investigate the activity of the complete antibody-linker-NP conjugates, these were tested in a direct binding immunoassay5. PBS buffer and whole fetal bovine serum (FBS) were used as the testing media. PBS buffer is a biocompatible, isotonic solution, frequently used in many in vitro studies. The FBS serum was chosen as an example of a complex medium, which allowed us to investigate the effects of the protein corona37, 38 and interference of other serum proteins on the binding capacity of the NP-Ab conjugates. This test was performed in simple 96-microwell plate format, where the target protein (human IgG) was adsorbed at the bottom of the well. The unoccupied surface was then blocked by BSA39. To test for non-specific binding in order to assess the signal-to-noise (S/N) ratio of NP conjugates, hemoglobin was used instead of IgG. Goat anti-human IgGs were conjugated to FITC-doped silica nanoparticles and the binding capacity of NP-Ab to IgG molecules was evaluated via the fluorescence (λex/λem, 495/535 nm). In Figure 5, the relative fluorescence intensities allow to compare the performance of antibody coated NP-conjugates in PBS and serum.


Page 7 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5: The activity of NP-Ab conjugates tested in a direct binding immunoassay. a) Specific binding – assessed using human IgG (60 μL, 50 mg/mL) adsorbed at the bottom of the microwell plate, b) Non-specific binding – hemoglobin (60 μL, 50 mg/mL) adsorbed at the bottom of the microwell plate as a control. Antibodies were conjugated with NP via simple physical adsorption (‘Adsorbed’) or through ‘PAMAM’ G4.5 dendrimer or carboxylated dextran linkers (‘6Dex’,’40Dex and ‘70Dex’). Note: The non-specific binding of NP-goat-anti-human-IgG conjugates to hemoglobin was almost negligible, the scale in figure 5B has a ‘break’ for better illustration of differences in non-specific binding among the samples. The first interesting observation was that the NP-Ab conjugates prepared by physical adsorption performed poorly both in PBS and serum. This was not surprising. Although physical adsorption provides high surface coverage of antibodies, the desorption of antibodies from the NP surface was also relatively high (Figure 4). The antibody desorption might even be accelerated in serum by proteins that have higher affinity for NP surface, thus replacing the physically adsorbed goat anti-human IgG antibodies. In addition, the adsorbed antibodies adopt random orientation on the NP surface and engage in protein-protein interactions, which negatively impacts their binding ability40. Interestingly though, the strongest signal in PBS was observed with NP-6Dex-Ab conjugate. This is consistent with the immobilization efficiency data presented in Figure 4, which indicated high surface coverage of antibodies on nanoparticles for NP6Dex-Ab. The fluorescence signal for NP-6Dex-Ab was approximately double that of NP-40Dex-Ab and NP-70Dex-Ab. In addition,


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 19

the fluorescence signal of NP-6Dex-Ab in PBS was more than 5 times higher than for the reference sample, NP-PAMAM-Ab. However, the direct binding assay performed in whole bovine serum shows a different trend. Herein, the signal intensity (proportional to the binding of FITC-doped NPs to human IgG) was significantly lower in all cases, indicating that the interference of the NPs with serum proteins was significant41. Presumably, a layer of protein corona (e.g. opsonins, albumines, fibrinogens etc.) was formed around the nanoparticles and similarly, serum proteins could also adsorb at the bottom of the 96-well plate, effectively blocking the human IgG molecules. The higher viscosity of serum also might have played some role, for example by slowing down the diffusion of the NPs in the sample and reducing the binding rate between the NP and the plate surface. Interestingly, NP-PAMAM-Ab sample performed slightly better in serum as it did in PBS. The qualities of PAMAM dendrimer as a multivalent linker were presented before in several in vitro and in vivo experiments using complex media such as cell culture or blood7, 31. The performance of NP-6Dex-Ab in serum closely followed that of NP-PAMAM-Ab, while the order amongst the dextran samples was following a clear trend NP6Dex-Ab> NP-40Dex-Ab>> NP-70Dex-Ab. It is likely that the higher molecular weight dextran molecules present themselves as ‘sticky’ molecules and they attract larger amount of serum proteins, i.e. create protein corona around the NPs. A layer of nonspecifically adsorbed serum proteins on the NP surface blocks the goat human-anti-IgG, which resulted in low fluorescence signal. To assess the binding specificity of the prepared NP-Ab conjugates, we have also performed direct binding immunoassay with Hemoglobin, instead of human IgG. The data, shown in Figure 5B indicate that the level of non-specific binding was negligible, and the signal reported in Figure 5A is due to a specific interaction between the goat anti-human-IgG on the NP surface and the humanIgG immobilized at the bottom of the 96-well plate. The signal intensities from Figure 5A (representing specific binding ‘signal’) and 5B (representing non-specific, background ‘noise’) were compared for each sample and summarized as a quantitative value in Table S.2 (ESI). This value represents the signal-to-noise ratio (S/N) and it is one of the crucial parameters that indicates the specificity of the NP-Ab conjugates. The highest S/N value in PBS was for the NP-6Dex-Ab (S/N=81), the highest value in serum was for NPPAMAM-Ab (S/N=52). The trend of decreasing performance in serum with the increased MW of the Dextran was again evident by the decrease in S/N (22 – 12 – 1.1 for NP-6Dex-Ab – NP-40Dex-Ab – NP-70Dex-Ab). In order to exclude the possibility that the data presented in Figure 5 were governed by the assay format and not the NP-Ab conjugate themselves, we have performed a simple direct binding assay test, in which the variable parameter was the amount of human-IgG adsorbed at the bottom of the 96-well plate. We have varied the concentration of human-IgG from 0-50 µg/mL and studied the specific binding of NP-6Dex-Ab and NP-PAMAM-Ab both in serum and PBS. The data presented in Figure 6 confirmed the development observed before. NP-6Dex-Ab performed better than NP-PAMAM-Ab in PBS across all the human IgG concentration range, and consistently with previous results, NP-PAMAM-Ab performed better in whole serum than in PBS. Presumably, this reversed trend could be caused by the subtle aggregation of NPs during the binding reaction. In our previous studies, we confirmed that the Ab-NPs remain colloidally stable after re-dispersion in both PBS and whole serum over the course of 4 hours at room temperature. This experiment presented in this work, however, was conducted over 16 hours at 4°C. The possible aggregation over the 16 hors time period could have depleted the concentration of free, monodisperse Ab-NPs in the solution. The aggregates were presumably removed during the purification steps, hence not contributing to the overall fluorescence signal. On the other hand, the aggregation phenomenon in whole serum is minimized, due to the abundance of serum proteins and the formation of protein corona that is known to stabilize colloids. The presented work will provide scientist with choices of several types of multivalent linkers – PAMAM and poly-carboxylated dextrans. The decisions which one to use might depend on multiple factors. For example, whether the intended application occurs in ‘clean’ or complex media (i.e. water, buffer, urine, serum, blood etc.), how much material is required (i.e. scalability) and also on the overall cost-value ratio. Although, NP-PAMAM-Ab outperformed NP-6Dex-Ab slightly in serum, the commercial cost of PAMAM G4.5 dendrimer is significantly higher than the cost associated with dextran carboxylation as presented here. The differences in cost are even more apparent at the multigram-scale quantities required for meaningful clinical trials in the industry.


Page 9 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 6: Testing NP-Ab conjugates (using PAMAM, 6Dex as linker molecules) in the direct binding immunoassay with different initial amounts of target human IgG (60 µL, 0-50 µg/mL) performed in: a) PBS, N=3 and b) whole serum, N=3. Conclusion In summary, we have presented a simple chemical method to prepare poly-carboxylated dextran and demonstrated the use of such material in the context of NP-antibody bio-conjugates. Poly-carboxylated dextran allowed for robust and straightforward EDC/(sulfo)NHS chemistry and therefore makes the bioconjugation approached shown here highly applicable and accessible to a wide range of research applications. Goat anti-human antibodies were conjugated to fluorescent silica NPs using poly-carboxylated dextran and such modified NP-Ab were compared with an established carboxy-PAMAM linking approach. We have shown that antibody-coated NPs using poly-carboxylated dextran outperformed those NP-Ab produced using PAMAM in a direct binding assay in PBS, mimicking condition that is employed during common bioassays signal transduction such as ELISA. In addition, target recognition of the antibody-coated NPs in whole serum using both linkers were comparable, thus suggesting the use of the polycarboxylated carbohydrate is preferable to PAMAM due to its lower cost and known biocompatibility. It is noteworthy to point out that the choice of the linker does have a significant effect on the surface chemistry of the nanomaterial. This can be utilized to improve material’s dispersibility and reduce undesired agglomeration. Therefore, our findings are presumably applicable and relevant also for researchers working with materials other than silica NPs and could become a widely employed approach for attaching biomolecules to other types of nanomaterials


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 19

Materials and methods Cyclohexane (anhydrous, 99.5 %), n-hexanol (anhydrous, ≥99 %), Triton-X 100, aminopropyl triethoxysilane (APTES) (97 %), tetraethyl orthosilicate (TEOS) (99.99 %), ammonium hydroxide solution (28 % w/v in water, ≥ 99.99 %), fluorescein isothiocyanate isomer (95 %), succinic anhydrite (SA) (98 %) poly(amidoamine) (PAMAM) dendrimer (ethylenediamine core, generation 4.5 solution, 5 % w/v in methanol), , dextran (from Leuconostoc spp., MW 6 kDa, 40 kDa, and 70 kDa, IgG from human serum (reagent grade, ≥95%), 4-morpholoneethanesulfonic acid (MES) (≥ 99 %), bovine serum albumin (BSA) (lyophilised powder, ≥ 98%), Tween 20, fetal bovine serum (FBS), bovine haemoglobin (lyophilised powder, ≥ 98%), (N-hydroxysuccinimide (98 %), and NaOH standard solution (0.1000 M) were purchased from Sigma Aldrich. Absolute ethanol, 1-ethyl-3(3-dimethylaminopropyl) carbodiimide HCl (EDC), phosphate buffer saline tablets (one tablet dissolved in 200 mL DI water yields 0.01 M phosphate buffer, pH 7.4) were purchased from Fisher Scientific. N-hydroxysulfosuccinimide sodium salt (> 98 %) was purchased from TCI. AlexaFluor647 labelled goat anti-human and goat anti-rabbit IgG (2 mg/mL) were purchased from Life Technologies. Transparent Nunc Maxisorb 96-well plates used in the immunoassay were purchased from Fisher Scientific. Deionised water (< 18 MU) was obtained from a Milli-Q system from Millipore. Carboxylation of Dextran with Succinic Anhydride. To introduce a high number of carboxyl groups to plain Dextran of various MWs (6kDa, 40 kDa or 70 kDa), the hydroxyl groups of the carbohydrate chain were partly esterified by succinic anhydride as previously reported on comparable template. Briefly, the dextran sample (0.5 g, 8.3 mmol of OH groups) was dissolved in anhydrous DMF (10 mL) followed by of LiCl (0.2 g, 4.7 mmol). The mixture was heated up to 90 ˚C and the appropriate amount of succinic anhydride (75 mg, 375 mg, 1.5 g, or, 3.75 g) together with pyridine (0.15 mL) were added. The amount of succinic anhydride used represents 0.1 x, 2 x or 5 x molar excess to the amount of hydroxyl groups. The temperature was set to 80 ˚C and the reaction was allowed to proceed for 3 h while stirring under dinitrogen atmosphere. After that the reaction mixture was poured over ice (25 g) and HCl (0.2 mL, 10 M) was added to stop the reaction and remove residual succinic anhydride and pyridine. The mixture was transferred into a Thermo Fisher Snake Skin dialysing bag (MWCO 3000) and dialysed against DI water for 4 days. After that, modified dextran was lyophilised and characterised by IR, 1H and 13C NMR Back-titration of carboxylated dextran. Prior the titration, DI water used was bubbled with nitrogen for 30 min in order to remove dissolved CO2. The exact concentration of HCl titrant solution (19.1815 mM) was determined by titration of NaOH standard solution (10 mL, 0.01 M) distributed by Sigma Aldrich. Then, a weighted sample of plain or carboxylated dextran (≈10 mg) was dissolved in 9 mL of DI water and standard NaOH solution (1 mL, 0.1000 M) was added. The sample stirred for 5 min to ensure quantitative conversion of carboxylic acid to carboxylate anions. After that, 2 drops of phenolphthalein indicator were added and the remaining OH- ions were titrated against HCl (19.1815 mM). Plain dextran (6, 40 or 70 kDa) was titrated as a blank. The number of carboxylic groups present on one dextran molecule was then calculated based on the difference in the consumption of titrant between carboxylated and plain dextran. Preparation of dye-doped SiO2 NPs. Fluorescein doped silica NPs were synthesised by modification of a procedure reported elsewhere3. Firstly, the FITC-APTMS conjugate was prepared by mixing fluorescein isothiocyanate (FITC) (39 mg, 0.1 mmol) with the stoichiometric amount of APTMS (23 µL, 0.13 mmol) in absolute ethanol (100 mL). The mixture was stirred at 40 ˚C for 24 h. After that, 10 mL of the mixture was collected and combined with 29 mL of absolute ethanol. Then TEOS (2.8 mL, 12.5 mmol) was added. The reaction was triggered by the addition of NH4OH (2.8 mL, 25 % (w/w)) water solution was added under vigorous stirring. The reaction was allowed to proceed overnight at RT, after that the nanoparticles were collected by centrifugation at 7800 rpm for 20 min and washed by redispersion in absolute ethanol (40 mL) (3 ×). Amine-modification of SiO2 NPs. The reactive amine groups were introduced on the surface of the prepared fluorescein-doped nanoparticles by APTES treatment in PBS 0.01 M reported previously4. Briefly, a pellet of silica NPs (15 mg) was dispersed in DI water (20 mL) by thorough sonication. Then, APTES (5.5 μL, 0.31 mmol) was added and the mixture was stirred for 2 hours. After that, NPs were isolated by centrifugation (14 000 rpm, 8 min) and washed by redispersion in ethanol (10 mL) 3 ×. Finally, the NPs were re-dispersed in ethanol (15 mL) and stored in the solution in the fridge, before they were used for further bio-conjugation with antibodies. Note that this form of storage should not exceed more than a week Conjugation of antibodies to NPs using PAMAM G4.5.The optimised method to conjugate antibodies to silica NPs via PAMAM linker was adapted from Moore et al.5 Briefly, EDC (73.6 mg, 0.38 mmol) was dissolved in PAMAM dendrimer (562 μL, 1μmol) methanol solution (5% w/v), which represented 0.128 mmol carboxylic groups. Separately, sulfo-NHS (42.6 mg, 0.2 mmol) was dissolved in DI water (250 μL) and the solution was added into the PAMAM dendrimer solution followed by HCl (16 μL, 1 N) and the mixture was topped up to 1 mL with DI water. The sample was then shaken for 20 min at 600 rpm using Co-Mix Laboratory Mixer. Meanwhile, previously prepared amino-modified NPs (2 mg) were isolated from the stock solution by centrifugation, washed with DI water, and redispersed in DI water (0.5 mL) by thorough ultrasonication. Then, the NPs dispersion was added into the activated dendrimer solution and the sample was allowed to shake for 15 min. After that, the sample was centrifuged (14 000 rpm, 8 min) the supernatant containing unreacted compounds was discarded and the dendrimer-coated and using mild sonication, NPs were re-dispersed in MES buffer pH 4.2 (1 mL, 0.1 M). AlexaFluor647 goat Anti-Human antibody solution (31.5 μL, 2 mg/mL) was added into dendrimer-coated NPs and was allowed to shake for 4 h at 600 rpm. To purify antibody-conjugated NPs afterwards the sample was centrifuged for 8 min at 14 000 rpm, mildly sonicated and re-dispersed in 1 mL PBS (4 ×). After each purification step, 0.3 mL


Page 11 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

of the supernatant solution was saved to quantify the amount of present antibodies via fluorescence (647 nm/677 nm) using calibration curve method. Conjugation of antibodies to NPs using carboxylated dextran. EDC (73.6 mg, 0.38 mmol) and previously freeze-dried carboxylated 6 kDa, 40 kDa or 70 kDa Dextran 2SA (34.2, 41.6, 53.0 or 40.8 mg respectively) was dissolved in 734 μL DI water. The amount of dextran used represented an equimolar amount (0.128 mmol) of carboxylic groups to PAMAM G4.5. The increase of MW of polymer in contrast to the original template (6, 40 or 70 kDa) was considered. Separately, sulfo-NHS (42.6 mg, 0.2 mmol) was dissolved in DI water (250 μL) and the solution was mixed with carboxylated dextran solution together with HCl (16 μL, 1 N). The rest of procedures was similar to the PAMAM conjugation procedure. Direct Binding Assays. Using Nunc Maxisorb 96 well plates, human IgG (60 μL, 50 mg/mL), bovine hemoglobin (60 μL, 50 mg/mL) or fetal bovine serum (FBS) (60 μL, 12.5% v/v)), all in 0.01 M PBS pH 7.4 (referred as ‘PBS’), were added to respective wells and left overnight at 4 ˚C. Human IgG wells served as a positive control, bovine haemoglobin/serum wells served as negative control. The plate was then cleaned by washing the wells with 120 μL 0.2% (v/v) Tween 80 in PBS and then PBS (2 ×). All wells were then blocked with 120 μL of 1% (w/v) BSA solution in PBS and incubated at 37 ˚C for 2 h. The same well cleaning procedure was used as before. The solid form antibody coated NPs (freeze-dried in 1% BSA) were re-dispersed in either PBS or whole FBS such that their concentration was 0.4 mg/mL. Antibody-coated NP sample (70 μL) was added to both types of well (human IgG coated and hemoglobin/FBS coated). The plate was then incubated overnight at 4 ˚C. The wells were cleaned as before and dried with a stream of nitrogen. PBS (60 μL) was added to the wells and fluorescence readings for FITC-doped NPs were recorded (495 nm / 530 nm, gain of 75). The experiment was performed in triplicate, all tested samples were independent conjugates from a large, single stock of FITC-loaded nanoparticles to keep the concentration of FITC constant. However, to account for the dye-leeching that might have occurred during the parallel conjugation and purifications steps, the fluorescence of all Ab-NP conjugates was normalized prior the microwell plate incubation at the initial concentration 0.4 mg/mL (which was within the linear region of the excitation/emission concentration dependence).

ASSOCIATED CONTENT Supporting Information Characterization data for the poly-carboxylate dextran synthesis such as NMR, FTIR, gravimetry, acid-base titration results and signal-tonoise ratio table are available as Supporting Information. The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author *[email protected]

Author Contributions The manuscript was written through contributions of all authors. / All authors have given approval to the final version of the manuscript. /

ACKNOWLEDGMENT We would like to acknowledge and thank to Medway School of Pharmacy, Universities of Kent and Greenwich for providing the funding for this project.

REFERENCES 1. Szelenyi, I., Nanomedicine: evolutionary and revolutionary developments in the treatment of certain inflammatory diseases. Inflamm. Res. 2012, 61 (1), 1-9. 2. Gubala, V.; Johnston, L. J.; Krug, H. F.; Moore, C. J.; Ober, C. K.; Schwenk, M.; Vert, M., Engineered nanomaterials and human health: Part 2. Applications and nanotoxicology (IUPAC Technical Report). Pure Appl. Chem. 2018, 90 (8), 1325-1356. 3. Giovaninni, G.; Moore, C. J.; Hall, A. J.; Byrne, H. J.; Gubala, V., pH-Dependent silica nanoparticle dissolution and cargo release. Colloid. Surface. B 2018, 169, 242-248. 4. Moore, C. J.; Giovannini, G.; Kunc, F.; Hall, A. J.; Gubala, V., Overloading' fluorescent silica nanoparticles with dyes to improve biosensor performance. J.Mater. Chem. B 2017, 5 (28), 5564-5572. 5. Moore, C. J.; Monton, H.; O'Kennedy, R.; Williams, D. E.; Nogues, C.; Crean, C.; Gubala, V., Controlling colloidal stability of silica nanoparticles during bioconjugation reactions with proteins and improving their longer-term stability, handling and storage. J. Mater. Chem. B 2015, 3 (10), 2043-2055. 6. Moore, T. L.; Rodriguez-Lorenzo, L.; Hirsch, V.; Balog, S.; Urban, D.; Jud, C.; Rothen-Rutishauser, B.; Lattuada, M.; PetriFink, A., Nanoparticle colloidal stability in cell culture media and impact on cellular interactions. Chem. Soc. Rev. 2015, 44 (17), 6287-305.


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 19

7. Tivnan, A.; Orr, W. S.; Gubala, V.; Nooney, R.; Williams, D. E.; McDonagh, C.; Prenter, S.; Harvey, H.; DomingoFernandez, R.; Bray, I. M.; Piskareva, O.; Ng, C. Y.; Lode, H. N.; Davidoff, A. M.; Stallings, R. L., Inhibition of Neuroblastoma Tumor Growth by Targeted Delivery of MicroRNA-34a Using Anti-Disialoganglioside GD(2) Coated Nanoparticles. PLoS One 2012, 7 (5). 8. Gubala, V.; Le Guevel, X.; Nooney, R.; Williams, D. E.; MacCraith, B., A comparison of mono and multivalent linkers and their effect on the colloidal stability of nanoparticle and immunoassays performance. Talanta 2010, 81 (4-5), 1833-1839. 9. Farka, Z.; Juriik, T.; Kovaar, D.; Trnkova, L.; Sklaadal, P., Nanoparticle-Based Immunochemical Biosensors and Assays: Recent Advances and Challenges. Chem. Rev. 2017, 117 (15), 9973-10042. 10. Tietjen, G. T.; Bracaglia, L. G.; Saltzman, W. M.; Pober, J. S., Focus on Fundamentals: Achieving Effective Nanoparticle Targeting. Trends Mol. Med. 2018, 24 (7), 598-606. 11. Spicer, C. D.; Jumeaux, C.; Gupta, B.; Stevens, M. M., Peptide and protein nanoparticle conjugates: versatile platforms for biomedical applications. Chem. Soc. Rev. 2018, 47 (10), 3574-3620. 12. Juliano, R., Nanomedicine: is the wave cresting? Nat. Rev. Drug. Discov. 2013, 12 (3), 171-2. 13. Venditto, V. J.; Szoka, F. C., Jr., Cancer nanomedicines: so many papers and so few drugs! Adv. Drug. Deliv. Rev. 2013, 65 (1), 80-8. 14. Gubala, V.; Johnston, L. J.; Liu, Z. W.; Krug, H.; Moore, C. J.; Ober, C. K.; Schwenk, M.; Vert, M., Engineered nanomaterials and human health: Part 1. Preparation, functionalization and characterization (IUPAC Technical Report). Pure Appl. Chem. 2018, 90 (8), 1283-1324. 15. Giovannini, G.; Kunc, F.; Piras, C. C.; Stranik, O.; Edwards, A. A.; Hall, A. J.; Gubala, V., Stabilizing silica nanoparticles in hydrogels: impact on storage and polydispersity. RSC Adv. 2017, 7 (32), 19924-19933. 16. Giovannini, G.; Warncke, P.; Fischer, D.; Stranik, O.; Hall, A. J.; Gubala, V., Improving colloidal stability of silica nanoparticles when stored in responsive gel: application and toxicity study. Nanotoxicol. 2018, 12 (5), 407-422. 17. Li, Z. J.; Narouz, M. R.; Munro, K.; Hao, B.; Crudden, C. M.; Horton, J. H.; Hao, H. X., Carboxymethylated DextranModified N-Heterocyclic Carbene Self-Assembled Monolayers on Gold for Use in Surface Plasmon Resonance Biosensing. ACS Appl. Mater. Interfaces 2017, 9 (45), 39223-39234. 18. Drake, A. W.; Tang, M. L.; Papalia, G. A.; Landes, G.; Haak-Frendscho, M.; Klakamp, S. L., Biacore surface matrix effects on the binding kinetics and affinity of an antigen/antibody complex. Anal. Biochem. 2012, 429 (1), 58-69. 19. Jonsson, C.; Aronsson, M.; Rundstrom, G.; Pettersson, C.; Mendel-Hartvig, I.; Bakker, J.; Martinsson, E.; Liedberg, B.; MacCraith, B.; Ohman, O.; Melin, J., Silane-dextran chemistry on lateral flow polymer chips for immunoassays. Lab Chip 2008, 8 (7), 1191-1197. 20. Varshosaz, J.; Emami, J.; Ahmadi, F.; Tavakoli, N.; Minaiyan, M.; Fassihi, A.; Mahzouni, P.; Dorkoosh, F., Preparation of budesonide-dextran conjugates using glutarate spacer as a colon-targeted drug delivery system: in vitro/in vivo evaluation in induced ulcerative colitis. J. Drug Target. 2011, 19 (2), 140-153. 21. Yuba, E.; Tajima, N.; Yoshizaki, Y.; Harada, A.; Hayashi, H.; Kono, K., Dextran derivative-based pH-sensitive liposomes for cancer immunotherapy. Biomater. 2014, 35 (9), 3091-3101. 22. Yuba, E.; Uesugi, S.; Miyazaki, M.; Kado, Y.; Harada, A.; Kono, K., Development of pH-sensitive Dextran Derivatives with Strong Adjuvant Function and Their Application to Antigen Delivery. Membranes 2017, 7 (3), 15. 23. Varshosaz, J.; Emami, J.; Tavakoli, N.; Fassihi, A.; Minaiyan, M.; Ahmadi, F.; Dorkoosh, F., Synthesis and evaluation of dextran-budesonide conjugates as colon specific prodrugs for treatment of ulcerative colitis. Int. J. Pharmaceut. 2009, 365 (1-2), 6976. 24. Gubala, V.; Harris, L. F.; Ricco, A. J.; Tan, M. X.; Williams, D. E., Point of Care Diagnostics: Status and Future. Anal. Chem. 2012, 84 (2), 487-515. 25. Shaterabadi, Z.; Nabiyouni, G.; Soleymani, M., High impact of in situ dextran coating on biocompatibility, stability and magnetic properties of iron oxide nanoparticles. Mater. Sci. Eng. C 2017, 75, 947-956. 26. Hermanson, G. T.; Hermanson, G. T., Bioconjugate Techniques, 2nd Edition. 2008. 27. Sun, Y. X.; Zhang, X. Z.; Cheng, H.; Cheng, S. X.; Zhuo, R. X., A low-toxic and efficient gene vector: Carboxymethyl dextran-graft-polyethylenimine. J. Biomed. Mater. Res. A 2008, 84A (4), 1102-1110. 28. Bagwe, R. P.; Yang, C.; Hilliard, L. R.; Tan, W., Optimization of Dye-Doped Silica Nanoparticles Prepared Using a Reverse Microemulsion Method. Langmuir 2004, 20 (19), 8336-8342. 29. Battistelli, G.; Cantelli, A.; Guidetti, G.; Manzi, J.; Montalti, M., Ultra-bright and stimuli-responsive fluorescent nanoparticles for bioimaging. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2016, 8 (1), 139-150. 30. Burns, A.; Sengupta, P.; Zedayko, T.; Baird, B.; Wiesner, U., Core/Shell Fluorescent Silica Nanoparticles for Chemical Sensing: Towards Single-Particle Laboratories. Small 2006, 2 (6), 723-726. 31. Gubala, V.; Crean, C.; Nooney, R.; Hearty, S.; McDonnell, B.; Heydon, K.; O'Kennedy, R.; MacCraith, B. D.; Williams, D. E., Kinetics of immunoassays with particles as labels: effect of antibody coupling using dendrimers as linkers. Analyst 2011, 136 (12), 2533-2541. 32. Stöber, W.; Fink, A.; Bohn, E., Controlled growth of monodisperse silica spheres in the micron size range. J. Colloid Interface Sci. 1968, 26 (1), 62-69. 33. Bagwe, R. P.; Hilliard, L. R.; Tan, W. H., Surface modification of silica nanoparticles to reduce aggregation and nonspecific binding. Langmuir 2006, 22 (9), 4357-4362. 34. Shahabi, S.; Treccani, L.; Dringen, R.; Rezwan, K., Modulation of Silica Nanoparticle Uptake into Human Osteoblast Cells by Variation of the Ratio of Amino and Sulfonate Surface Groups: Effects of Serum. ACS Appl. Mater. Interfaces 2015, 7 (25), 13821-13833.


Page 13 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

35. Hunger, J.; Bernecker, A.; Bakker, H. J.; Bonn, M.; Richter, R. P., Hydration Dynamics of Hyaluronan and Dextran. Biophys. J. 2012, 103 (1), L10-L12. 36. O’Connell, C. L.; Nooney, R.; McDonagh, C., Cyanine5-doped silica nanoparticles as ultra-bright immunospecific labels for model circulating tumour cells in flow cytometry and microscopy. Biosens. Bioelectron. 2017, 91, 190-198. 37. Docter, D.; Westmeier, D.; Markiewicz, M.; Stolte, S.; Knauer, S. K.; Stauber, R. H., The nanoparticle biomolecule corona: lessons learned - challenge accepted? Chem. Soc. Rev. 2015, 44 (17), 6094-6121. 38. Lundqvist, M.; Stigler, J.; Cedervall, T.; Berggård, T.; Flanagan, M. B.; Lynch, I.; Elia, G.; Dawson, K., The Evolution of the Protein Corona around Nanoparticles: A Test Study. ACS Nano 2011, 5 (9), 7503-7509. 39. Le, N. C. H.; Gubala, V.; Gandhiraman, R. P.; Daniels, S.; Williams, D. E., Evaluation of Different Nonspecific Binding Blocking Agents Deposited Inside Poly(methyl methacrylate) Microfluidic Flow-Cells. Langmuir 2011, 27 (14), 9043-9051. 40. Trilling, A. K.; Beekwilder, J.; Zuilhof, H., Antibody orientation on biosensor surfaces: a minireview. Analyst 2013, 138 (6), 1619-1627. 41. Monopoli, M. P.; Aberg, C.; Salvati, A.; Dawson, K. A., Biomolecular coronas provide the biological identity of nanosized materials. Nat. Nanotechnol. 2012, 7 (12), 779-786.


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1: A schematic description of Ab-NPs preparation. 273x115mm (150 x 150 DPI)


Page 14 of 19

Page 15 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 2: A reaction Scheme and the mechanism of dextran esterification with succinic anhydride. Dextran 6 kDa: n = 33, Dextran 40 kDa: n = 222, Dextran 70 kDa: n = 389 174x54mm (150 x 150 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3: TEM analysis of FITC-SiO2 NPs: (a) NP-APTES as a starting material; after modification with goat anti-human IgG linked via: 40Dex (b), and PAMAM G4.5 (c) using EDC/sulfo-NHS chemistry. All NPs appear to be reasonably uniformed in size and shape, showing no apparent aggregation. The images show the nanoparticles in their ‘dry’ state, where the solvent (DI water) was removed after the samples were deposited on the copper/carbon TEM grid at 1 mg/mL. 142x210mm (150 x 150 DPI)


Page 16 of 19

Page 17 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 4: The relative linker immobilization efficiency is presented as the percentage of antibodies that remained attached during EDC/sulfo-NHS coupling and throughout the 4 washing steps. 100 % corresponds to the initial amount of antibodies used in the conjugation reaction. The grey bars (4th wash) represents the portion of antibodies immobilized on the surface after the final wash. The antibody adsorption/chemisorption, washing experiment was performed in three independent repli-cates (n=3). 551x336mm (72 x 72 DPI)


Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5: The activity of NP-Ab conjugates tested in a direct binding immunoassay. a) Specific binding – assessed using human IgG (60 μL, 50 mg/mL) adsorbed at the bottom of the microwell plate, b) Nonspecific binding – hemoglobin (60 μL, 50 mg/mL) adsorbed at the bottom of the microwell plate as a control. Antibodies were conjugated with NP via simple physical adsorption (‘Adsorbed’) or through ‘PAMAM’ G4.5 dendrimer or carboxylated dextran linkers (‘6Dex’,’40Dex and ‘70Dex’). Note: The non-specific binding of NP-goat-anti-human-IgG conjugates to hemoglobin was almost negligible, the scale in figure 5B has a ‘break’ for better illustration of differences in non-specific binding among the samples. 109x174mm (150 x 150 DPI)


Page 18 of 19

Page 19 of 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 6: Testing NP-Ab conjugates (using PAMAM, 6Dex as linker molecules) in the direct binding immunoassay with dif-ferent initial amounts of target human IgG (60 µL, 0-50 µg/mL) performed in: a) PBS, b) whole serum. 109x158mm (150 x 150 DPI)


Poly-Carboxylated Dextran as a Multivalent Crosslinker: Synthesis

Recommend Documents