Eliminating the Animal Species Constraints in Antibody Selection for

Apr 27, 2017 - Eliminating the Animal Species Constraints in Antibody Selection for Multicolor Immunoassays. Yao Sun and Xiaohu Gao ... Related Conten...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/bc

Eliminating the Animal Species Constraints in Antibody Selection for Multicolor Immunoassays Yao Sun and Xiaohu Gao* Department of Bioengineering, University of Washington, William H. Foege Building N530M, 3720 15th Avenue NE, Seattle, Washington 98195, United States S Supporting Information *

ABSTRACT: Molecular profiling on the single-cell level helps unveil the mystery of gene expression patterns in individual cells at subcellular resolution, enabling discovery of small but meaningful variations that are often overlooked at the population level. Similar to other immunoassays, the most common and economical protocols are developed by combining primary antibodies (1′Abs) and fluorophore-labeled secondary antibodies (2′Abs). The selection of 1′ and 2′ Abs, however, has been limited by the availability of animal species, consequently resulting in low multiplexing capability. Here we report the development of preassembled Ab pairs using 1′Abs all from the same animal species. We show that multiple molecular targets can be simultaneously labeled without cross reactivity. This simple and general self-assembly technology eliminates the animal species constraints in multicolor immunoassays, offering exciting new opportunities for a wide range of biomedical and clinical applications.

■

INTRODUCTION Immunoassays are one the most popular bioanalytical methods in both basic research and clinical diagnosis, attributed to their high specificity, high sensitivity, and the broad selection of antibodies.1−3 Over the years, a number of assay formats have been developed such as immunofluorescence (IF),4,5 immunohistochemistry (IHC),6,7 enzyme-linked immunosorbent assay (ELISA),8,9 and Western blotting.10,11 Using IF, which utilizes fluorophore-labeled antibodies to image antigens in cells, as an example, it allows direct visualization of the presence, abundance, and distribution of important molecular events with the aid of fluorescence microscopy. Despite the simple and robust protocol, development of fluorescent antibody probes has been a key limiting factor for multiplexed studies. Two general approaches exist at this time for antibody labeling, direct labeling of the 1′Ab (single-step) and indirect labeling by combining unmodified 1′Ab with fluorescently labeled 2′Ab (two-step), both of which have distinct advantages as well as disadvantages. Direct labeling of 1′Abs with fluorophores enables multiple 1′Abs raised in the same animal species to be labeled with different fluorophores and shortens the cell staining assay time. At the same time, covalent conjugation of fluorophores to 1′Abs is a very © XXXX American Chemical Society

complicated and costly practice involving multiple chemical reaction and purification steps (Figure 1a).6,12 In addition, commercially available 1′Abs often come with stabilizer proteins that have to be removed to avoid interferences with bioconjugation.6 In contrast, the indirect labeling approach typically takes two steps for target detection (unmodified 1′Ab binding with the target, followed by fluorophore-labeled 2′Ab binding with the 1′Ab), thus avoiding the complicated chemistry (Figure 1b). Indeed, with both the 1′Abs and 2′Ab-fluorophore conjugates broadly available, freeing the end users from running multistep chemical reactions, the two-step indirect labeling has become a more popular approach in biology and medicine.13,14 A major limitation, however, is its limited multiplexing capability. When multiple targets are imaged in the same sample, the 1′Abs have to be raised in different animal species to avoid cross reactivity. Common animals for raising antibodies include rabbit, mouse, goat, and horse, with the former two being the most popular. Ideally, the labeling technology should have the strengths of both direct Received: March 22, 2017 Revised: April 18, 2017

A

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

Figure 1. Schematic illustration of single-step, two-step, and SASMI-based immunoassays. (a) Single-step immunoassay. QD-1′Ab bioconjugates are prepared through complex chemical reactions and multiple rounds of purification. 1′Ab raised in the same animal species can be labeled with different QDs and used in the single-step multiplexed staining. This approach is costly and labor intensive. (b) Two-step immunoassay. Cells are first incubated with 1′Abs raised in different animal species, washed thoroughly, then incubated with QD-2′Ab against different animal species. No chemical reaction is involved, but the multiplexing capability is limited by the animal species. (c) Immunoassay based on SASMI technology. 1′Abs from single animal species are mixed with QD-2′Ab of different colors separately to achieve single-step, purification-free assembly of functional QD2′Ab-1′Ab probes. Nonspecific IgG from the same animal species as the 1′Ab is added to prevent cross reactivity between different QD colors when pooled into a multicolor cocktail. The nonspecific IgG is used at a significantly higher concentration than the 1′Ab to dominate the binding with QD-2′Ab. The multicolor staining cocktail is used for one-step multiplexed staining. This mix-and-use approach combines the advantages of the conventional single-step and two-step immunoassays.

complicated chemical reactions and purifications. When multicolor 1′Ab-QD bioconjugates are mixed together for multiplexed immunoassays, however, major cross reactivity among different targets are expected (Figure S1. Organic dyelabeled 2′Ab shares the same problem; see Figure S2). The cross reactivity is a result of two processes. First, when 1′Ab and QD-2′Ab are mixed, complete capture of the 1′Ab (depletion of free 1′Ab) is very difficult even when excess amount of QD-2′Ab is used. Second, due to the noncovalent nature of 1′Ab and 2′Ab binding, a small portion of 1′Abs may dissociate from the QDs and reassociate with QDs of a different color during cell staining (incubation time typically 1−2 h). These processes associated with biomolecular interactions and purification-free staining conditions introduce cross reactivity (nonspecific binding). As a result, the approach of preassembling 1′Ab and 2′Ab-fluorophore is not used in current practices despite its simplicity. We have devised a simple solution to this problem, addition of high-concentration nonspecific IgG (Ab with no specificity to any targets). Compared to the minute amount of free 1′Abs (either due to the initial incomplete association or due to dissociation during staining), the nonspecific IgG competes for any available binding sites in the 2′Ab-fluorophore, eliminating the possibility of mismatched reassociation (Figure 1c). For proof of concept, in the current study the single animal species-based multicolor imaging (SASMI) technology is demonstrated in immunofluorescence staining of cells, while a similar protocol should apply to other immunoassays.

and indirect labeling (specifically, the multiplexing capability of direct labeling and the simplicity and low cost of indirect labeling) while avoiding the aforementioned limitations. To eliminate the constraints in antibody selection and enable simple, low-cost, on-demand production of fluorescently labeled 1′Abs, we propose to hybridize the conventional onestep staining (Figure 1a) and two-step staining process (Figure 1b) by preassembling the 1′Ab with fluorophore-labeled 2′Ab in solution. As schematically illustrated in Figure 1c, 1′Ab and QD-2′Ab bioconjugates are prepared by a simple incubation step. Note that although the design works for all types of fluorophores (see Supporting Information of organic dyelabeled 2′Ab), here we mainly use QDs to demonstrate the concept because of their photostability that allows quantitative imaging.15−17 Due to the natural binding affinity between the 1′Ab and 2′Ab, the 1′Ab docks onto the QD surface, leading to formation of fluorescently labeled 1′Abs ready for immunostaining. More importantly, this self-assembly process is not affected by carrier proteins typically present in commercial 1′Ab samples (e.g., BSA),18,19 unlike the direct chemical conjugation approach discussed above. Note that the term “selfassembly” refers the formation of individual 1′Ab-QD bioconjugates through IgG recognition rather than orderly arrangements of a large number of building blocks that are often seen in materials sciences. For example, when N 1′Abs all raised in mouse against different targets are separately incubated with N color QDs that all have rabbit-anti-mouse (r-anti-m) 2′Abs on them, N color QD-2′Ab-1′Ab bioconjugates can be easily prepared without B

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

■

RESULTS AND DISCUSSION The performance of preassembled QD-2′Ab-1′Ab probes was first evaluated through single-color staining of three molecular targets in formalin-fixed HeLa cells (targets stained separately, Figure 2). The three model targets used in this study are Lamin

Figure 3. Multiplexed staining using SASMI. HSP-90, Lamin A, and Ki-67 are simultaneously labeled with preassembled QD-2′Ab-1′Ab probes emitting at 525, 585, and 625 nm, respectively. (a) Composite image showing all three colors. (b) Spectrally unmixed individual colors shown in separate channels (hyperspectral imaging). (c) Quantitative analysis of the staining intensity. The average staining intensities of the three model targets measured in multiplexed staining (yellow bars) are consistent with those in the single-color staining (green bars, based on data from Figure 2a). The fluorescence intensity of individual QDs is adjusted to compensate for differential QD brightness.18,19 Error bars represent standard deviation of average staining intensity of three different images of each specimen. Scale bar, 50 μm.

Figure 2. Single-color staining using SASMI. (a) Three model targets (HSP-90, Lamin A, and Ki-67, from left to right) are labeled with preassembled QD-2′Ab-1′Ab probes emitting at 525, 585, and 625 nm, respectively, in a one-step staining procedure. (b) Negative control experiments skipping the addition of 1′Ab in the probe assembly. Virtually no fluorescence is detected. (c) Positive-control cell labeling using the conventional two-step staining procedure. Scale bar, 50 μm.

experiment. HSP-90 was chosen as the staining target due to its high expression level in HeLa cells. When the 1′Ab against HSP-90 was added to a mixture of equal amounts of QD5852′Ab and QD625-2′Ab probes, both QD585 and QD625 channels showed strong fluorescence signals (Figure 4a,d). When the 1′Ab was first assembled with one QD-2′Ab (incubation for 30 min) followed by addition to cells in the presence of the nonspecific IgG and the second color QD-2′Ab, virtually no staining is detected for the second QD color (Figure 4b,c,d). This quantitative imaging assay directly proves that the large excess of nonspecific IgG completely blocks all the free QD2′Ab probes, leaving no chance to the small amount of initially unbound or later dissociated 1′Ab to reattach with another QD2′Ab. As a control experiment, we also quantitatively evaluated the impact of the nonspecific IgG on our SASMI technology. The same dual-color experiments (as shown in Figure 4) were repeated without addition of the nonspecific IgG. When both QD-2′Ab probes are preassembled with anti-HSP-90 1′Ab at the same time, strong target staining can be observed in both channels (Figure 5a,d). As shown in Figure 5b,c,d, however, strong fluorescence is not only seen for the first-color QD-2′Ab (initially incubated with the 1′Ab), but also for the secondcolor QD-2′Ab probes that are added at a later time. This significant cross reactivity problem is largely due to unlabeled 1′Abs or 1′Ab with unoccupied binding sites during the initial self-assembly step rather than 1′Ab dissociated during cell incubation. The molar ratio of QD-2′Ab to 1′Ab used in this work is 1.5:1, insufficient to saturate all the 1′Abs binding sites (one 1′Ab can bind with several QD-2′Abs). Increasing the concentration of the first-color QD-2′Ab would help reduce the cross reactivity, but it is not an option practically, because QD staining requires an optimal concentration of QDs (higher concentration results in nonspecific binding). Even if higher

A, HSP-90, and Ki-67, representing both nuclear and cytoplasmic targets, and the 1′Abs are all raised in rabbit. Note that the form of 2′Ab in the QD-2′Ab conjugates is F(ab′)2 fragment, while the same design should apply to other types such as whole IgG and Fab. Control experiments without the addition of primary antibodies rule out nonspecific binding of QDs (Figure 2b). Comparing the results obtained with the 1-step SASMI staining (Figure 2a) and the conventional 2-step staining (Figure 2c), the staining patterns are highly consistent, indicating that the specificity and affinity of 1′Ab are preserved through the self-assembly between 1′Abs and QD-2′Abs. To demonstrate the feasibility of multicolor staining using SASMI, cells were incubated with the three previously mentioned molecular probes in parallel. HSP-90, Lamin A, and Ki-67 are simultaneously labeled with QD-2′Ab-1′Ab probes emitting at 525, 585, and 625 nm, respectively (Figure 3a), and imaged using Hyperspectral imaging (HSI), which is capable of separating multicolor QD stains into different channels for quantitative assessment.20 As shown in Figure 3b, the staining patterns for all three targets in the parallel staining process are consistent with those in the single-color staining experiments shown in Figure 2a, corroborating the multiplexing capability of the SASMI technology. Furthermore, the relative target expression profiles are identical between the multicolor and single-color staining experiments, confirming the robustness of quantitative signal analysis combining QDs and HSI (Figure 3c). Following the qualitative multicolor staining studies, we set out to quantitatively confirm the absence of cross reactivity among different targets, which directly determines the suitability of SASMI technology for multicolor immunoassays. To do so, we designed a dual-color competitive staining C

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

Figure 5. Cross reactivity studies of QD-2′Ab-1′Ab probes in the absence of nonspecific IgG. (a) When QD585-2′Ab and QD625-2′Ab are incubated together with anti-HSP-90 1′Ab, both probes bind to 1′Ab, producing characteristic HSP-90 staining patterns in both channels. At the same time, regardless of whether QD585 (b) or QD625 (c) is preassembled with anti-HSP-90 1′Ab, in the absence of nonspecific rabbit IgG, both the preassembled QDs and the competitor QDs show specific staining. (d) Staining intensities of each QD channel are comparable for the three staining experiments, due to unsaturated 1′Ab during the initial assembly step. The brightness of individual QD channels is normalized in order to directly compare staining intensity (QD585 scaled up by a factor of 1.1 relative to QD625 due to differential QD brightness). Error bars represent standard deviation of average staining intensity of three different images of each specimen. Scale bar, 50 μm.

Figure 4. Quantitative cross reactivity assessment for SASMI. (a) Preassembled probes made by coincubation of QD585-2′Ab and QD625-2′Ab with anti-HSP-90 1′Ab produce characteristic HSP-90 staining patterns in both channels. (b) QD585 preassembled with antiHSP-90 1′Ab for 30 min then mixed with QD625 and the nonspecific IgG only produces strong staining signals in the 585 nm channel. (c) Same as (b), except the sequence of QD addition is reversed. (d) Staining intensity of preassembled probe (QD585 in b and QD625 in c) is comparable to that obtained with dual-color staining experiments (a). HSI is used to unmix true-color images (left column) into QD585 (middle column) and QD625 (right column) channels, and perform quantitative signal analysis of the staining intensity (d). Brightness of individual QD channels is normalized in order to directly compare staining intensity. The intensity of QD585 channel is scaled up by a factor of 1.1 relative to QD625 to compensate for differential QD brightness. Error bars represent standard deviation of average staining intensity of three different images of each specimen. Scale bar, 50 μm.

by adding high-concentration nonspecific IgGs, a simple and clever idea to eliminate cross reactivity and consequently the constraint of selecting Abs from different animal species. Previously, we have demonstrated that protein A can be used as an adaptor protein for Ab conjugation.18,19 However, due to the intrinsic limitations of protein A, the conjugation only works for rabbit Abs and a few mouse Ab subtypes. In addition, due to noncovalent binding, the immunostaining process should be finished within 2 h to avoid significant probe dissociation. When a new adaptor protein (e.g., 2′Abs and protein G) or a new 1′Ab is introduced, these detailed experimental conditions must be repeated, posing a practical challenge for solving real biological and medical problems. In the current work, the introduction of nonspecific IgG (or IgGs if 1′Abs are raised in several animals) completely eliminates these concerns, because the small amount of unbound or dissociated 1′Abs are outcompeted by the nonspecific IgGs of higher concentration. Here, we demonstrated the SASMI technology in immunofluorescence cell staining, and we expect it will work in a similar fashion in virtually all other immunoassays.

concentrations of QDs were used, it would not completely eliminate cross reactivity. We show this problem by using dyelabeled 2′Ab, for which a much higher concentration can be used without causing significant nonspecific binding. As shown in Figure S3, at a molar ratio of 2′Ab-dye/1′Ab of 6, significant cross reactivity was still observed. The direct contrast between Figures 4 and 5 clearly shows the importance of nonspecific IgG in suppressing cross reactivity. It acts as a scavenger that cleans the available binding sites in all QD-2′Ab regardless of their color, thereby eliminating potential cross reactivity when multicolor probes are pooled into a cocktail for multiplexed immunoassays. In summary, we have developed a general, scalable antibody labeling technology that combines the multiplexing capability of 1′Ab-fluorophore conjugates and the simplicity and costeffectiveness of indirect labeling using 1′Ab and 2′Abfluorophore. The self-assembly process of immobilizing 1′Ab onto QD-2′Ab enables preparation of fluorescently labeled 1′Ab on-demand, but introduces cross reactivity when multiple targets are stained simultaneously. We have solved this problem D

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

■

EXPERIMENTAL PROCEDURES Cell Culture and Processing. HeLa cells (ATCC) were grown in glass-bottom 24-well plates (Greiner Bio-One) for 2− 3 days to a density of about 80−90%. Humidified atmosphere at 37 °C with 5% CO2 was maintained. MEM culture medium with Earle′s salts and L-glutamine (GIBCO) was supplemented with 10% fetal bovine serum (PAA Laboratories) and 0.6% penicillin−streptomycin (10 000 U mL−1, GIBCO). After reaching the desired confluency, cells were rinsed with Trisbuffered saline (TBS), fixed with 4% formaldehyde/0.05% Triton X-100/TBS for 20 min at 37 °C, permeabilized with 2% DTAC/TBS (dodecyltrimethylammonium chloride, SigmaAldrich) for 20 min at RT, followed by 0.25% Triton X-100/ TBS for 5 min at RT. The fixed and permeabilized cells were washed three times with TBS before storing in TBS at 4 °C. Cell Imaging and Quantitative Analysis. Inverted fluorescence microscope (IX-71, Olympus) equipped with a true-color charge-coupled device (QColor5, Olympus) and an HSI camera (Nuance, 420−720 nm spectral range, CRI, now PerkinElmer) was used to examine cells attached to the coverslip bottom of 24-well plates. Low-magnification images were obtained with a 20× dry objective (numerical aperture (NA) 0.75, Olympus) and high-magnification with a 100× oilimmersion objective (NA 1.40, Olympus). Wide UV filter cube (330−385 nm band-pass excitation, 420 nm long-pass emission, Olympus) was used for imaging of all QD probes, FITC LP cube (460−500 nm band-pass excitation, 510 nm long-pass emission, Chroma) for Alexa Fluor 488 probes, Rhodamine LP cube (530−560 nm band-pass excitation, 572 nm long-pass emission, Chroma) for Rhodamine Red probes. Nuance image analysis software was used to unmix the obtained hyperspectral images based on the reference spectra of individual QD colors. The brightness and contrast of each channel were automatically adjusted in the pseudocolor composite image for best visual representation. For quantitative analysis of each QD channel, regions of interest that include QD staining were automatically identified using the Nuance software while excluding “blank” nonstained areas. Average signal intensity was recorded for each QD channel. The same process was performed on three low-magnification images taken from different areas of the same specimen to obtain an average. Differential brightness of QD probes was quantified via bulk fluorescence measurement of QD solutions, and appropriate correction factors were applied in quantitative analysis to compensate for differential QD brightness. Two-Step Single-Color IF. Polyclonal rabbit antibodies against Lamin A (Sigma-Aldrich), HSP90 (Sigma-Aldrich), and Ki-67 (Abcam) were used for all immunostaining. QD-2′Ab conjugates (Qdot goat F(ab′)2 anti-rabbit IgG conjugates (H +L), Invitrogen) with emission peaks centered at 525, 585, 605, and 625 nm were used as fluorescent probes. All buffers were prepared with deionized water (>18 MΩ cm). The blocking buffer was composed of 2% BSA (from 10% bovine serum albumin in TBS, Thermo Scientific), 0.1% casein (from 5% solution, Novagen), and 1× TBS. The staining buffer was composed of 6% BSA in 1× TBS. The washing buffer was composed of 1% BSA, 0.1% casein, and 1× TBS. For two-step IF, fixed cells were blocked with the blocking buffer for 1 h and incubated with 300 μL 1 μg mL−1 primary antibodies in the staining buffer for 1 h. Cells were then washed three times with TBS and incubated with 300 μL 3 nM QD2′Ab in staining buffer for 1 h in the dark. Extra QD probes

were removed by washing cells with washing buffer twice and TBS three times. Fluorescence imaging was performed afterward. Single-Step Single-Color or Multicolor IF. For singlestep IF, cells were blocked with the blocking buffer for 1 h as described in the two-step IF section. Concurrently, QD-2′Ab1′Ab probes were prepared by incubating 3 μL 1 μM QD-2′Ab with 1.5 μL 0.2 mg mL−1 1′Ab for 30 min at RT. Then, 1.5 μL 5 mg mL−1 nonspecific rabbit IgG (Thermo Scientific) was added to incubate for another 5 min at RT. Each QD-2′Ab1′Ab probe was prepared in a separate microcentrifuge tube. For single-color staining, each probe was diluted to a final volume of 300 μL with the staining buffer and applied separately to preblocked cells. For parallel multiplexed staining, all probes of different colors were combined in a single tube, diluted to 300 μL with the staining buffer, and immediately applied to preblocked cells. After staining for 2 h, cells were washed twice with washing buffer and three times with TBS. Fluorescence imaging was done immediately following the staining. QD-2′Ab-1′Ab Probe Cross Reactivity Studies. Probe cross reactivity studies with nonspecific rabbit IgG were carried out in the same manner as the single-step dual-color staining described already, except that one QD-2′Ab was incubated with 1′Ab and nonspecific rabbit IgG sequentially, while the other one was not. Following QD1-2′Ab-1′Ab probe assembly and the addition of nonspecific rabbit IgG, QD2-2′Ab was added. The mixture was diluted to 300 μL in the staining buffer and added to the preblocked cells. As a reference, to demonstrate the capability of QD-2′Ab probes to efficiently compete for 1′Ab in solution, both QD-2′Ab probes were incubated together with 1′Ab and nonspecific rabbit IgG sequentially, diluted to 300 μL in the staining buffer, and applied to preblocked cells. After staining for 2 h, cells were washed using the same method as described earlier and imaged. HSI was used to unmix and quantify the staining intensity of individual QD signals. QD585 and QD625 were used for this study. For quantitative analysis, correction factor of 1.1 was applied to QD585 signal to normalize it against QD625. To show the importance of nonspecific IgG addition, the probe cross reactivity studies without nonspecific IgG were performed in the same manner as described earlier, except that no nonspecific rabbit IgG was employed. The cell imaging and signal analysis were carried out in the same fashion. Dye-2′Ab Based Multiplexed IF and Cross Reactivity Studies. To demonstrate the general applicability of SASMI technology, similar multiplexed immunoassays and cross reactivity studies were carried out using dye-2′Ab probes. Alexa Fluor 488 (AF488) and Rhodamine Red (RR) labeled 2′Ab (Alexa Fluor 488 and Rhodamine Red-X AffiniPure Fab fragment goat anti-rabbit IgG, Fc fragment specific, Jackson ImmunoResearch) were used as fluorescent probes. Cells were blocked with blocking buffer for 30 min. Concurrently, dye2′Ab-1′Ab probes were prepared by incubating 3 μL 0.2 mg mL−1 dye-2′Ab with 1.5 μL 0.2 mg mL−1 1′Ab for 30 min at RT. Then 3 μL 5 mg mL−1 nonspecific rabbit IgG (Thermo Scientific) was added to incubate for another 5 min at RT. Each dye-2′Ab-1′Ab probe was prepared in a separate microcentrifuge tube. For parallel multiplexed staining, the two probes were combined in a single tube, diluted to 300 μL with staining buffer, and immediately applied to preblocked cells. After staining for 1 h, cells were washed three times with PBS. Fluorescence imaging was done immediately following staining. E

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

(5) Surovtseva, Y. V., Jairam, V., Salem, A. F., Sundaram, R. K., Bindra, R. S., and Herzon, S. B. (2016) Characterization of Cardiac Glycoside Natural Products as Potent Inhibitors of DNA DoubleStrand Break Repair by a Whole-Cell Double Immunofluorescence Assay. J. Am. Chem. Soc. 138, 3844−3855. (6) Xing, Y., Chaudry, Q., Shen, C., Kong, K. Y., Zhau, H. E., Chung, L. W., Petros, J. A., O’Regan, R. M., Yezhelyev, M. V., Simons, J. W., et al. (2007) Bioconjugated quantum dots for multiplexed and quantitative immunohistochemistry. Nat. Protoc. 2, 1152−1165. (7) Sankaran, V. G., Xu, J., Ragoczy, T., Ippolito, G. C., Walkley, C. R., Maika, S. D., Fujiwara, Y., Ito, M., Groudine, M., Bender, M. a, et al. (2009) Developmental and species-divergent globin switching are driven by BCL11A. Nature 460, 1093−1097. (8) de la Rica, R., and Stevens, M. M. (2012) Plasmonic ELISA for the ultrasensitive detection of disease biomarkers with the naked eye. Nat. Nanotechnol. 7, 821−824. (9) Rissin, D. M., Kan, C. W., Campbell, T. G., Howes, S. C., Fournier, D. R., Song, L., Piech, T., Patel, P. P., Chang, L., Rivnak, A. J., et al. (2010) Single-molecule enzyme-linked immunosorbent assay detects serum proteins at subfemtomolar concentrations. Nat. Biotechnol. 28, 595−599. (10) Welinder, C., and Ekblad, L. (2011) Coomassie Staining as Loading Control in Western Blot Analysis. J. Proteome Res. 10, 1416− 1419. (11) Bakalova, R., Zhelev, Z., Ohba, H., and Baba, Y. (2005) Quantum Dot-Based Western Blot Technology for Ultrasensitive Detection of Tracer Proteins. J. Am. Chem. Soc. 127, 9328−9329. (12) Goldman, E. R., Anderson, G. P., Tran, P. T., Mattoussi, H., Charles, P. T., and Mauro, J. M. (2002) Conjugation of Luminescent Quantum Dots with Antibodies Using an Engineered Adaptor Protein To Provide New Reagents for Fluoroimmunoassays. Anal. Chem. 74, 841−847. (13) Lachner, M., O’Carroll, D., Rea, S., Mechtler, K., and Jenuwein, T. (2001) Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410, 116−20. (14) Herker, E., Harris, C., Hernandez, C., Carpentier, A., Kaehlcke, K., Rosenberg, A. R., Farese, R. V., and Ott, M. (2010) Efficient hepatitis C virus particle formation requires diacylglycerol acyltransferase-1. Nat. Med. 16, 1295−1298. (15) Alivisatos, P. (2004) The use of nanocrystals in biological detection. Nat. Biotechnol. 22, 47−52. (16) Medintz, I. L., Uyeda, H. T., Goldman, E. R., and Mattoussi, H. (2005) Quantum dot bioconjugates for imaging, labelling and sensing. Nat. Mater. 4, 435−446. (17) Zrazhevskiy, P., Sena, M., and Gao, X. (2010) Designing multifunctional quantum dots for bioimaging, detection, and drug delivery. Chem. Soc. Rev. 39, 4326. (18) Zrazhevskiy, P., and Gao, X. (2013) Quantum dot imaging platform for single-cell molecular profiling. Nat. Commun. 4, 1619. (19) Zrazhevskiy, P., True, L. D., and Gao, X. (2013) Multicolor multicycle molecular profiling with quantum dots for single-cell analysis. Nat. Protoc. 8, 1852−1869. (20) True, L. D., and Gao, X. (2007) Quantum Dots for Molecular Pathology. J. Mol. Diagn. 9, 7−11.

Probe cross reactivity studies with nonspecific rabbit IgG were carried out in the same manner as the QD-based studies. Specifically, dye1-2′Ab was incubated with 1′Ab and nonspecific rabbit IgG sequentially, while dye2-2′Ab was not. Following dye1-2′Ab-1′Ab probe assembly and the addition of nonspecific rabbit IgG, dye2-2′Ab was added. The mixture was diluted to 300 μL in staining buffer and added to the preblocked cells. As a reference, to demonstrate the capability of dye-2′Ab probes to efficiently compete for 1′Ab in solution, both dye-2′Ab probes were incubated together with 1′Ab and nonspecific rabbit IgG sequentially, diluted to 300 μL in staining buffer, and applied to preblocked cells. After staining for 1 h, cells were washed three times with PBS and imaged. Probe cross reactivity studies without nonspecific IgG were carried out in the same fashion omitting the nonspecific IgG.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.bioconjchem.7b00156. Two-color staining using QD-2′Ab-1′Ab or dye-2′Ab1′Ab probes in the absence of nonspecific IgG. Cross reactivity study using dye-2′Ab probes (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: 206-543-6562. ORCID

Xiaohu Gao: 0000-0002-6054-0530 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported in part by the NIH (R21CA192985) and the Department of Bioengineering at University of Washington. We thank Junwei Li, Wanyi Tai, and Pavel Zrazhevskiy for fruitful discussions.

■

ABBREVIATIONS IF, immunofluorescence; QD, quantum dot; 1′Ab, primary antibody; 2′Ab, secondary antibody; BSA, bovine serum albumin; TBS, Tris-buffered saline; PBS, phosphate-buffered saline; HSI, hyperspectral imaging

■

REFERENCES

(1) Schweitzer, B., Wiltshire, S., Lambert, J., O’Malley, S., Kukanskis, K., Zhu, Z., Kingsmore, S. F., Lizardi, P. M., and Ward, D. C. (2000) Immunoassays with rolling circle DNA amplification: A versatile platform for ultrasensitive antigen detection. Proc. Natl. Acad. Sci. U. S. A. 97, 10113−10119. (2) Herr, A. E., Hatch, A. V., Throckmorton, D. J., Tran, H. M., Brennan, J. S., Giannobile, W. V., and Singh, A. K. (2007) Microfluidic immunoassays as rapid saliva-based clinical diagnostics. Proc. Natl. Acad. Sci. U. S. A. 104, 5268−5273. (3) Deiss, F., LaFratta, C. N., Symer, M., Blicharz, T. M., Sojic, N., and Walt, D. R. (2009) Multiplexed Sandwich Immunoassays Using Electrochemiluminescence Imaging Resolved at the Single Bead Level. J. Am. Chem. Soc. 131, 6088−6089. (4) Lin, J.-R., Fallahi-Sichani, M., and Sorger, P. K. (2015) Highly multiplexed imaging of single cells using a high-throughput cyclic immunofluorescence method. Nat. Commun. 6, 8390. F

DOI: 10.1021/acs.bioconjchem.7b00156 Bioconjugate Chem. XXXX, XXX, XXX−XXX