Elecrtonic Supplementary Information Application of quantum dot barcodes prepared using biological self-assembly to multiplexed immunoassays Sakandar Rauf, Andrew Glidle and Jonathan M Cooper Department of Electronics and Electrical Engineering, Oakfield Avenue, University of Glasgow, UK G12 8LT Experimental Section Materials: Dynabeads M-28 streptavidin (2.8µm), streptavidin- and biotin-conjugated CdSe/ZnS quantum dots (QD65-biotin, QD65-streptavidin, QD655-streptavidin and QD565-streptavidin) were purchased from Invitrogen, UK. Rabbit IgG (cat#i56), human IgG (cat#456), mouse IgG (I5381) and goat IgG (cat#i5256) were all of reagent grade and purchased from Sigma-Aldrich, UK. Protein A-biotin (cat# P2165), protein G-biotin (cat#p845) and IgG free bovine serum albumin (BSA) (cat# A985) were purchesd from Sigma-Aldrich, UK. Fluorescein isothiocyanate (FITC) lablled anti-goat IgG whole molecule developed in rabbit (cat# F-7367), anti-mouse IgG (Fab specific, cat# F5262), anti-human IgG whole molecule developed in goat (cat# F3512) were purchesd from Sigma-Aldrich. FITC-labelled anti-rabbit IgG (cat# F25) was purchased from Dako, UK. 1mM phosphate buffer, PBS, ph7.4 (adjusted to a constant ionic strength of 25mM using NaCl and containing.2% sodium azide) was prepared using ultrapure MilliQ water (Ω>18 MΩ.cm). Three different PBS buffers were used. PBS buffer without BSA, PBS with.15% BSA and PBS wth 2%BSA. Aliquots of all IgGs, anti-iggs-fitc, protein A-biotin and protein G-biotin were prepared in PBS using flash freezing with liquid nitrogen before being stored at 2 C. Before using, aliquots were equilibrated at room temperature. Preparation of quantum dot barcodes using layer-by-layer assembly of QD-biotin and QDstreptavidin conjugates: Scheme S1 shows the quantum dot barcode production using the layer-by-layer assembly of quantum dot-biotin (QD-biotin) and quantum dot streptavidin (QD-streptavidin) conjugates. To make the first layer of QD65-biotin conjugate (B 65 ), streptavidin coated magnetic beads ( ) (2µl~1.4 1 6 beads) from stock solution were added to 1mM PBS ph 7.4 (3µl) containing.15% BSA. To this, B 65 (2µl) was added from the 1
stock solution (2.2µM) and incubated for one hour, agitated using a vortex mixer at a low speed. After one hour, magnetic beads were separated from the quantum dot suspension using a magnet (MagnaRack, DNA Research Innovations, UK). These beads were then washed four times with the same PBS to remove any non-specifically adsorbed QD65-biotin conjugate. The beads were then re-suspended in PBS buffer (3µl). These beads were named B 65. In order to deposit a second layer of quantum dots, QD655-streptavidin conjugate (S 655 ) (4µl from 1µM stock) was added to the bead suspension and incubated for one hour. 1 Streptavidin coated magnetic bead( ) ~ 2.8µm + Quantum dot 65-biotin conjugate (B 65 ) (15-2nm) B 65 + Quantum dot 655-streptavidin conjugate (S 655 ) (15-2nm) Quantum dot 655-streptavidin conjugate (S 65 ) (15-2nm + B 65 S 655 B 65 B 65 + B 65 S 655 (B 65 S 65 ) 1 B 65 S 655 B 65 S 655 (B 65 S 65 ) 2 --- B 65 S 655 (B 65 S 65 ) n Scheme S1. Shows the production of quantum dot barcodes using layer-by-layer assembly of QD-biotin and QDstreptavidin conjugates onto a streptavidin coated magnetic bead. Note that figures are not to scale. (Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission). 1 2
After separation, the beads were washed four times with PBS buffer and re-suspended in PBS (3µl) to make the first barcode B 65 S 655. To make subsequent layers of quantum dots, barcoded magnetic beads containing B 65 S 655 were incubated in QD65-biotin and QD65- streptavidin conjugates. In this manner, using the layer-by-layer assembly of biotinstreptavidin, B 65 S 655, B 65 S 655 (B 65 S 65 ) 1 were prepared. For the preparation of B 65 S 65 and B 65 S 565, QD65-biotin, QD65-streptavidin and QD565-streptavidin conjugates were used following the same procedure, as described above. We had previously prepared three different barcodes of the same series B 65 S 655, -B 65 S 655 (B 65 S 65 ) 1 and -B 65 S 655 (B 65 S 65 ) 2. Now, in order to show the broad applicability of the technology we prepared a new sequence of beads, compared to the sequence reported in Scheme S1 1. Although the numbers of layers differed, given that the size of biotin or streptavidin conjugate is ~1-12nm this was considered negligible when compared to the size of the bead (2.8μm). Single bead fluorescence spectroscopy measurements: For single bead fluorescence spectroscopic measurements, a Nikon microscope (Microphot-SA) (NA=.85, X4 objective) was coupled with a Triax 32 spectrometer (Jobin Yvon) equipped with a Symphony CCD camera. Figure S1 shows the schematic diagram for single bead fluorescence measurements. A B-2A long pass emission filter set (Nikon) was used to excite the quantum dots in the blue region (45-49 nm) with a dichroic mirror DM51 and 515 nm long pass emission filter. For the detection of FITC, a band pass filter (5-55nm, Omega Optical) was used to distinguish between the FITC and non-fitc barcode. micrographs were collected using a true colour camera. 3
1μm Light source CCD camera Spectrometer Emission filter Dichroic mirror Variable aperture Excitation filter Microscope Objective 4, NA=.85 Sample Figure S1. Schematic diagram of single bead fluorescence measurement set-up. (Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission). 1 Multiplexed Immunoassay: To perform the multiplexed immunoassay, following steps were performed. (i) Immobilization of Protein A-biotin or Protein G-biotin: Protein A-biotin was immobilized on barcode B 65 S 655, B 65 S 655 (B 65 S 65 ) 1 and B 65 S 65. Briefly, 5µl (~2.3 1 5 beads ) from a 3µl barcode stock was added to an eppendorff tube. The barcoded beads were collected using the magnet and PBS buffer was discarded. To this, a 15µl aliquot of protein A-biotin (1mg/ml) was added and incubated for 45 min. The beads were agitated every 5-1 min using a pippete in order to ensure the mixing of barcoded beads in the protein A-biotin solution. After this the barcodes were collected using the magnet and excess of protein A- biotin was removed. After washing, with the PBS, these barcoded beads were redispersed in 1µl PBS. A similar procedure was used for the immobilization of protein G-biotin on the barcode B 65 S 565. 4
(ii) Immobilization of Different IgGs: The barcoded beads B 65 S 655, B 65 S 655 (B 65 S 65 ) 1 and B 65 S 65 with protein A-biotin were used for the immobilization of rabbit IgG, human IgG and mouse IgG respectively. The barcoded beads with protein A-biotin were collected by using the magnet and the PBS buffer was discarded. To the barcoded beads B 65 S 655, B 65 S 655 (B 65 S 65 ) 1 and B 65 S 65, onto which protein A-biotin had been immobilized, 3µl of rabbit IgG (1mg/ml), human IgG (1mg/ml) and mouse IgG (1mg/ml) were added respectively and incubated for 45 min. These barcoded beads were mixed every 5-1 min using a pippette. As before, these barcodes were then washed exhaustively to remove excess of IgGs using PBS which contain 2% BSA. The same procedure was followed to immobilize goat IgG (1mg/ml) on the barcode B 65 S 565 using protein G-biotin. Accordingly, a specific IgG was bound to each of the different barcodes. (iii) Detection of Different Immunoglobulins (IgGs): The barcoded beads containing different IgGs were mixed in an equal proportion and collected using a magnet. This mixture was then redispersed in 3µl of PBS buffer containing 2% BSA. Equal concentration (1.5mg/ml) of stock solutions of all four FITC-labelled anti-iggs (anti-rabbit IgG-FITC, anti-human IgG-FITC, antimouse IgG-FITC and anti-gorat IgG-FITC) were made. In the first assay, a cocktail (4µl) of four FITC-labelled anti-iggs (1µl from each stock solution) was added to this mixture and incubated for 45min in the dark. The solutions were gently agitated by using a vortex mixer. After this, the beads were washed exhaustively with PBS containing 2% BSA to remove excess of FITClabelled anti-iggs. In the second assay, the procedure of the immunoassay was the same but antimouse IgG-FITC was not added to the cocktail and three other FITC-labelled IgGs were added. In the third assay, anti-human IgG-FITC was not added to the mixture and, again the three other FITC-labelled IgGs were added. These beads were then analyzed using single bead fluorescence spectroscopy. 2% BSA was added to the reaction mixture to block non-specific adsorption or cross-reactivity of different IgGs. Tween 2 was not used as it was found to destabilize the quantum dot conjugates resulting in a loss of activity in the the barcode. 5
Figure S2 shows the single bead fluorescence spectra of four different barcodes before and after the immunoassay. It can be seen that a shoulder appeared at around 53nm in the spectrum of each barcode, after the attachment of the specific anti-igg-fitc conjugate on the barcode. Control Experiment: As a control experiment, barcode B 65 S 565 (without the immobilization of any IgG) was incubated with a cocktail (4µl) of four FITC-lablled anti-iggs (1µl from each) for 45min using a vortex mixer. After exhaustive washing with PBS containing 2% BSA, single bead fluorescence spectroscopy was used to measure the fluorescent signal. Figure S3 shows that no FITC signal was observed in the barcode spectra indicating that FITC-lablled anti-iggs bind 2 B 65 S 655 B 65 S 655 + Anti-rabbit IgG FITC 2 B 65 S 655 (B 65 S 65 ) 1 B 65 S 655 (B 65 S 65 ) 1 + Anti-human IgG FITC 2 B 65 S 65 2 B 65 S 565 B 65 S 65 + Anti mouse IgG FITC B 65 S 565 + Anti-goat IgG FITC Figure S2. Single bead fluorescence spectra of four different barcodes measured before and after performing multiplexed immunoassay. Rabbit IgG was immobilised on barcode -B 65 S 655 (red), human IgG was immobilised on barcode B 65 S 655 (B 65 S 65 ) 1 (blue), mouse IgG was immobilised on barcode -B 65 S 65 (pink) and goat IgG was immobilised on barcode -B 65 S 565 (green). All the spectra are on the same scale. 6
only in the presence of the corresponding IgGs. Due to the presence of streptavidin on the barcode surface which provides a coating that shows minimal non-specific binding to proteins, FITClablled anti-iggs did not adsorb on the barcode surface. The blocking of the beads with 2% BSA plays a key role to block the non-specific adsorption of proteins. Before incubation After incubation Figure S3. Control experiment showing the single bead spectra of the barcode B 65 S 565 before and after incubation in a cocktail of four different FITC-labelled anti-iggs. The shoulder at around 53nm did not appear after incubation of the barcode in the FITC-labelled anti-iggs. Note this a representative spectra. In this control experiment 25 single bead spectra were averaged. The limit of detection for the immunoassay was estimated by immobilising human IgG onto the barcode -B 65 S 655 and different concentrations of the anti-human IgG-FITC were added to a known concentration of the barcoded beads, containing human IgG. Using simple linear regression of this standard addition curve, we estimated the limit of detection to be the equivalent concentration at three standard deviations above the mean of the background (141 au), namely ~ 2ng/ml or 133 pm (data not shown). References 1. S. Rauf, A. Glidle, and J. M. Cooper, Advanced Materials, 29, 21, 42. 7