A Microfluidic Immunostaining System Enables Quality Assured and Standardized Immunohistochemical Biomarker Analysis

Abstract

Immunohistochemistry (IHC) plays an important role in biomarker-driven cancer therapy. Although there has been a high demand for standardized and quality assured IHC, it has rarely been achieved due to the complexity of IHC testing and the subjective validation-based process flow of IHC quality control. We present here a microfluidic immunostaining system for the standardization of IHC by creating a microfluidic linearly graded antibody (Ab)-staining device and a reference cell microarray. Unlike conventional efforts, our system deals primarily with the screening of biomarker staining conditions for quantitative quality assurance testing in IHC. We characterized the microfluidic matching of Ab staining intensity using three HER2 Abs produced by different manufacturers. The quality of HER2 Ab was also validated using tissues of breast cancer patients, demonstrating that our system is an efficient and powerful tool for the standardization and quality assurance of IHC.

Figure 1. Microfluidic immunostaining system for antibody (Ab) quality verification and standardization.

(A) Conventional immunostaining standardization process flow. A biomarker-specific cell or tissue sample that stained with optimized intensity is stored as a reference. When a new Ab is introduced, repetitive immunostaining is performed to subjectively match the staining intensity to the reference. (B) A positive control cell block section is immunostained with linearly graded concentrations of Ab. The staining intensity of the reference is compared with those resulting from Ab incubation at various Ab concentrations to identify the best match. (C) Images of the microchannels and cell sections resulting from the concentration gradient generating network of the microfluidic immunostaining system. Fluorescein isothiocyanate (FITC) solution was introduced into the inlet with distilled water. (D) A SKBR3 cell section was stained with graded concentrations of HER2 Ab and labeled with QD605 (left) and 3,3′-diaminobenzidine (right).

Figure 2. Comparison of the staining intensities between a batch process and a microfluidic process using HER2 Abs and QD605-IgG with SKBR3 cell sections.

(A,B) Staining results using the microfluidic device at 5, 15, 30, and 60 min. The data are mean ± s.d.; n = 3–4. (C) HER2 Ab was labeled with QD605 using a microfluidic immunostaining system (10 min) and a batch staining process (2 h); the concentration ranged from 0.375 ng mL⁻¹ to 24 ng mL⁻¹. (D) Quantitative analysis of the HER2 staining intensity between a batch process and microfluidic staining. The data are mean ± s.d.; n = 3–7. (E) Staining intensities of three HER2 Abs (Dako: 1.6 μg mL⁻¹, Abcam: 0.024 μg mL⁻¹, Thermo Fisher Scientific: 0.36 mg mL⁻¹ [for 1.000×]) at several concentrations using a 2 h batch process and a 10 min microfluidic process. The data are mean ± s.d.; n = 3–8 for 2 h batch process and n = 3–4 for microfluidic process.

Figure 3. Standardization of three HER2 Ab immunostaining and standardized tissue staining results.

Three HER2 Abs from different companies (Dako: 1.6 μg mL⁻¹, Abcam: 0.024 μg mL⁻¹, Thermo Fisher Scientific: 0.36 mg mL⁻¹ [for 1.000×]) were randomly labeled as LOTs #1, #2, and, #3, respectively, to ensure anonymity. The chip-based Ab-staining process was carried out for 10 min and the batch process was carried out for 2 h. (A–C) Based on the 1.000× intensity of LOT #2, 0.750× of LOTs #1 and #3 have the most similar intensity. The data are mean ± s.d.; n = 3. Red color shows the QD605-labeled cells in tissue sections and blue color comes from the autofluorescence of tissue sections caused by the excitation of ultraviolet. (D) Sections from the same breast cancer tissues were immunostained with selected concentrations of HER2 Abs. (E) Intensity quantification result of tissue sections with HER2 Abs. The data are mean ± s.d.; n = 3–15. The error bars are based on the QD605 intensity of the imaging spot only from the epithelial region in the same tissue section.

Figure 4. Fluorescence images of QD605s in SKBR3 cell sections labeled with a form of quantum dot-conjugated IgG (QD-IgG).

(A) Fluorescence images of QD605-conjugated goat anti-rabbit IgG incubated in a SKBR3 cell section using a microfluidic immunostaining system. (B–D) Comparison of fluorescence images stained using a microfluidic system (left) and a batch process (right). The incubation results from concentrations of 0.5×, 0.250×, and 0.125×. Red color shows the QD605-labeled cells and blue color comes from the autofluorescence of cell sections caused by the excitation of ultraviolet.

Figure 5. Concept of the Ab quality verification system and fluorescence images of breast cancer cells and breast cancer tissues.

(A) A schematic of the microfluidic immunostaining system and a cell array consisting of SKBR3 and MCF7. SKBR3 and MCF7 were used as positive and negative references for HER2, respectively. (B) Immunostaining results of HER2 Ab: LOT #1 in SKBR3 and MCF7 using a microfluidic system and batch incubation process. The data are mean ± s.d.; n = 3 for chip and n = 3–9 for batch. (C) HER2 Ab: LOT #2 was used as in panel B. (D) Immunostaining results of breast cancer tissues from four different patients using HER2 Abs of LOT#1 and LOT #2. (E) Surface plots based on the intensities of QD605; bottom images of panel D.