Ultrasensitive ImmunoMag-CRISPR lateral flow assay for point of-care testing of urinary biomarkers

Abstract

Rapid, accurate, and non-invasive detection of biomarkers in saliva, urine or nasal fluid is essential for the identification, early diagnosis and monitoring of cancer, organ failure, transplant rejection, vascular diseases, autoimmune disorders, and infectious diseases. We report the development of an Immuno-CRISPR based lateral flow assay (LFA) using antibody-DNA barcode complexes with magnetic enrichment of the target urinary biomarkers CXCL9 and CXCL10 for naked eye detection (ImmunoMag-CRISPR LFA). An intermediate approach involving a magnetic bead-based Immuno-CRISPR assay (ImmunoMag-CRISPR) resulted in a limit of detection (LOD) of 0.6 pg/mL for CXCL9. This value surpasses the detection limits achieved by previously reported assays. The highly sensitive detection method was then re-engineered into an LFA format with an LOD of 18 pg/mL for CXCL9, thereby enabling non-invasive early detection of acute kidney transplant rejection. ImmunoMag-CRISPR LFA was tested on 42 clinical urine samples from kidney transplant recipients, and the assay could determine 11 positive and 31 negative urinary samples through a simple visual comparison of the test line and the control line of the LFA strip. The LFA system was then expanded to quantify CXCL9 and CXCL10 levels in clinical urine samples from images. This approach has the potential to be extended to a wide range of point-of-care tests (POCTs) for highly sensitive biomarker detection.

Graphical Abstract

Rapid, sensitive, and non-invasive biomarker detection from saliva, urine or nasal fluid is critical for early detection of infectious disease, cancer, organ failure, transplant rejection, coronary and vascular disease and autoimmune disease.^1–8 Innovations in sensing and detection platforms have advanced both point-of-care testing (POCT) and point-of-use (POU) technology.^9–12 One such innovation, clustered regularly interspaced short palindromic repeats/CRISPR-associated (CRISPR/Cas), has been harnessed for genome editing, gene regulation, imaging, and detection.^13–16 In particular, indiscriminate single-stranded DNA cleavage activity of some CRISPR/Cas nucleases upon guide RNA-dependent DNA binding is attracting attention in new biomarker sensing.^16–19

A major challenge in developing activatable CRISPR/Cas-based detection is to broaden its use beyond nucleic acid targets, and to detect other biomarkers, including proteins, organic compounds, and metal ions.^20–26 For example, in the case of protein detection, aptamers or antibodies can be used as capture or detection probes and coupled with CRISPR/Cas-based detection system to achieve high sensitivity and specificity to identify target proteins.^24–29 Another challenge is optimizing the amplification step, which is crucial for achieving high sensitivity and specificity.³⁰ Traditional PCR amplification method is time-consuming and requires temperature-controlled devices. Additionally, they can pose a risk of contamination by aerosols, which can compromise the accuracy of the sensor.³⁰ To overcome this problem, alternative isothermal amplification methods have been explored, such as hybridization chain reaction (HCR),^31,32 T7 transcription amplification,^28,29 and multiplication of CRISPR/Cas recognition sites. HCR uses DNA hairpins to generate long, branched DNA structures at room temperature, without the need for specialized equipment or enzymes.^31,32 RNA amplification by T7 RNA polymerase transcription using DNA templates bound to antigen-antibody assemblies can enhance CRISPR/Cas collateral cleavage activity for protein detection.^28,29 Such trans-cleavage activity is enabled by fluorophore-quencher (FQ) reporters as signaling molecules and a nucleic acid amplification step, such as PCR (polymerase chain reaction),^33,34 LAMP (loop-mediated isothermal amplification)^35,36 and RPA (recombinase polymerase amplification),^37–40 to achieve attomolar or greater sensitivity. Overall, this technique offers a highly sensitive and specific way to amplify and detect nucleic acid signals, which has important applications in fields such as medical diagnostics and genetic analysis.

We previously developed a Cas12a-coupled Immuno-CRISPR assay²⁵ to detect the urinary chemokine ligand 9 (CXCL9), which is a biomarker for non-invasive early detection of acute transplant rejection and cancer.^41–45 The Immuno-CRISPR assay showed high sensitivity without an amplification step; however, it requires specialized equipment for fluorescence detection. Conversely, lateral flow assays (LFAs) offer the ability to detect biomarkers in an equipment-free, low-cost, and user-friendly manner.^12,46–48 Recently, smartphone-based LFA quantification is increasing the accessibility and affordability of LFA testing, as it can obviate the need for specialized optical equipment.^19,49,50 Unfortunately, LFAs are approximately three orders of magnitude less sensitive than, for example, enzyme-linked immunosorbent assay (ELISA), which allows signal amplification by enzymes, and quantitative PCR (qPCR) including DNA amplification and fluorescence detection with high sensitivity.^51,52 Given that the limit of detection (LOD) of conventional horseradish peroxidase (HRP)-based ELISA for CXCL9 detection (100 pg/mL)²⁵ is close to the threshold for borderline kidney transplant rejection (200 pg/mL),⁴³ this low LOD poses a significant challenge for use of an LFA.

In the current study, we developed an LFA-compatible, naked-eye detection methodology for urinary biomarker detection. We selected CXCL9 and the related chemokine ligand 10 (CXCL10)^44,45 as target urinary biomarkers. To increase the detection sensitivity of CXCL9 and CXCL10 using LFA, we used magnetic beads to enrich target biomarkers bound to a capture antibody conjugated to a self-assembled DNA barcode complex. This complex has multiple Cas12a recognition sites, which can activate Cas12a trans-cleavage activity approximately 8-fold higher than using a single DNA barcode.²⁵ Collateral FAM-biotin (FB) reporters were then cleaved by activated Cas12a/crRNA complexes without a preamplification step. The detection sensitivity of urinary CXCL9 and CXCL10 using ImmunoMag-CRISPR LFA was evaluated and compared to traditional ELISA and LFA without magnetic enrichment (Immuno-CRISPR LFA).²⁵ ImmunoMag-CRISPR LFA was then used to evaluate 42 clinical urine samples by visual inspection and quantification of urinary biomarkers to establish performance on patient samples.

EXPERIMENTAL SECTION

Materials.

EnGen Lba Cas12a (Cpf1) and buffer for Cas12a were purchased from New England Biolabs (Ipswich, MA). Oligonucleotides with biotin modification for DNA barcode preparation, CRISPR RNA (crRNA), FAM-quencher (FQ) reporter and FAM-biotin (FB) reporter for Cas12a were synthesized by Integrated DNA Technologies (Coralville, IA) (Table S1). Dynabeads MyOne with carboxylic acid, Dynabeads M-280 with streptavidin and streptavidin were purchased from Thermo Fisher Scientific (Waltham, MA). Polystyrene based magnetic particles were purchased from Magsphere (Pasadena, CA). 1-Ethyl-3-(3-(dimethylamino)propyl) carbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), 2-(N-morpholino) ethanesulfonic acid (MES), phosphate buffered saline (PBS) and Tween 20 were purchased from Sigma-Aldrich (St. Louis, MO). Streptavidin-conjugated HRP, substrate reagents for HRP activity, recombinant human CXCL9 and CXCL10 and their capture and detection antibodies were purchased from R&D Systems (Minneapolis, MN). Synthetic urine was purchased from Medica Corporation (Bedford, MA). Lateral flow assay kit was purchased from Milenia Biotec GmbH (Gießen, Germany). All solutions were prepared with purified water by a Milli-Q purification system from Millipore (Burlington, MA).

Preparation of antibody-conjugated magnetic beads.

We used three types of magnetic beads: carboxylate functionalized, streptavidin-coated, and polystyrene-coated. Antibody was immobilized onto the carboxylate functionalized magnetic beads using EDC/NHS chemistry. Carboxylate functionalized magnetic beads (2 mg) were mixed with EDC (25 mg/mL) and NHS (25 mg/mL) in MES buffer (50 mM, pH 6.0) under slow tilt rotation at room temperature for 30 min, followed by washing with MES buffer by magnetic capture and resuspension. The capture antibody in PBS (1 mL, 200 μg/mL) was added to EDC/NHS-treated magnetic beads. The capture antibody was also added to non-treated polystyrene-coated magnetic beads. In the case of streptavidin-coated magnetic beads, biotinylated capture antibody was used. The mixtures were then incubated under slow tilt rotation at room temperature for 30 min, and at 4°C overnight. To block each mixture, 1 mL of BSA solution in PBS was added and incubated under slow tilt rotation at room temperature for another 30 min, followed by washing 3-times with washing buffer (PBS with Tween 20) by magnetic capture and resuspension.

Antibody loading on each magnetic bead was estimated using a similar procedure to that in the literature.^53,54 Briefly, protein concentrations of the prepared capture antibody solution and the supernatants from three PBS washes were determined using a Pierce Micro-BCA protein assay kit from Thermo Fisher Scientific. The amount of antibody immobilized onto each magnetic bead was calculated by subtracting the total protein determined in the supernatants from the initial protein concentration in the antibody solution.

Magnetic bead-based immunoassays for urinary biomarker detection.

We prepared two magnetic bead-based immunoassays: fluorescence-based ImmunoMag-CRISPR assay and ImmunoMag-CRISPR LFA for urinary biomarker detection. First, 1 mL of CXCL9 or CXCL10 solution in synthetic urine was added to each mixture and incubated under slow tilt rotation at room temperature for 20 min, followed by washing 3-times with washing buffer by magnetic capture and resuspension. A detection antibody-HRP or DNA barcode complex was prepared by assembly with streptavidin as previously reported.²⁵ One milliliter of detection antibody-streptavidin-HRP or DNA barcode complexes was then added to each mixture and incubated under slow tilt rotation at room temperature for 20 min, followed by washing 3-times with washing buffer by magnetic capture and resuspension. After concentrating the mixture to 50 μL by magnetic capture of the beads, the resulting material was stored at 4°C until use. HRP activity of a detection antibody-HRP complex was measured to characterize and optimize the magnetic bead-based immunoassay. After adding substrate solution for HRP (1 mM of 3,3’,5,5’-tetramethylbenzidine and 3 mM of H₂O₂), the samples were incubated at room temperature for 30 min. The absorbance was then measured at 655 nm. Experiments were performed in triplicate, and standard deviations were indicated.

We also performed the traditional HRP-based ELISA and the plate-based Immuno-CRISPR assay in the form of LFA without using magnetic beads (Immuno-CRISPR LFA). Sandwich assemblies with capture antibody, antigen and detection antibody-DNA barcode complex were prepared in a 96-well plate as previously reported.^25,41 The HRP-based ELISA was tested by measuring HRP activity. After adding substrate solution for HRP (1 mM of 3,3’,5,5’-tetramethylbenzidine and 3 mM of H₂O₂), the samples incubated at room temperature for 30 min. The absorbance was then measured at 655 nm. Total assay time for the HRP-based ELISA is around 4 hours. Experiments were performed in triplicate, and standard deviations were indicated.

Fluorescence-based assay in solution using FQ (fluorophore-quencher) reporter.

To test the ImmunoMag-CRISPR assay with a detection antibody-streptavidin-DNA barcode complex, trans-cleavage activity of the Cas12a complex was measured using an FQ reporter-based assay.^25,39 Briefly, the Cas12a complex was pre-assembled by mixing Cas12a (1 μM) and crRNA (1.2 μM) in the buffer for Cas12a (50 mM NaCl, 10 mM Tris-HCl, 10 mM MgCl₂ and 100 μg/ml recombinant albumin) at room temperature for 30 min. The FQ reporter (1 μM) was then mixed with the diluted Cas12a complex (0.5 μM). After adding the prepared magnetic bead solutions (50 μL) to the reaction cocktails (50 μL), the mixtures were incubated at room temperature for 30 min. The fluorescence signal from excitation at λ_ex = 485 nm and emission at λ_em = 530 nm was measured using a plate reader (SpectraMax M5, San Jose, CA). Total assay time from magnetic enrichment for the fluorescence-based ImmunoMag-CRISPR assay is around 90 min. Experiments were performed in triplicate, and standard deviations were indicated. The limit of detection (LOD) was calculated from each immunoassay-based standard curve by determining the concentration with a fluorescence signal consistent with 2√2 times the standard deviation of the blank.⁵⁵

LFA using FB (fluorophore-biotin) reporter.

To test the ImmunoMag-CRISPR LFA and Immuno-CRISPR LFA, the trans-cleavage activity of the Cas12a complex was also measured using a LFA with an FB reporter.³⁹ We prepared the reaction cocktails with Cas12a complex (0.5 μM) and FB reporter (1 μM) in the same way as for the fluorescence-based assay. After adding the prepared magnetic bead solutions (50 μL) to the reaction cocktails (50 μL), the mixtures were incubated at room temperature for 30 min. For the plate-based Immuno-CRISPR LFA, reaction cocktails (50 μL) were then added to the prepared wells and incubated at room temperature for 30 min. Fifty microliters of the resulting solution were mixed with 50 μL of the assay buffer from the lateral flow assay kit and transferred to another 96-well plate. Then, a test strip was placed upright and removed from the well after 3 min. Total assay time from magnetic enrichment for the fluorescence-based ImmunoMag-CRISPR assay is around 93 min.

Photographs of LFA strips were then taken using a smartphone to quantify the concentrations of target proteins on the control and test lines using ImageJ software (NIH).³⁵ Briefly, rectangular regions in the control line, test line and background were selected, inverted, and converted into a 32-bit image, followed by measuring its average grey scale intensity. The intensities of test and control lines were background subtracted, followed by estimating the ratio of the test to control line signal intensities (T/C ratio). The LOD was calculated from each standard curve by determining the CXCL9 or CXCL10 concentration with a T/C ratio consistent with 2√2 times the standard deviation of the blank.⁵⁵

Evaluation of clinical samples.

We tested and analyzed 42 urine samples from kidney transplant recipients participating in the Clinical Trials in Organ Transplantation (CTOT)-19 trial⁵⁶ by the lateral flow assay described above. The CTOT urine samples were obtained at the time of kidney allograft biopsies as per study protocol, and the sample collection was approved by institutional review boards (IRBs) at participating centers.⁵⁶ The samples were aliquoted and stored at −20°C. These isolated urines were used as received and without pretreatment. The urine samples were de-identified and tested blinded to biopsy diagnosis. All samples collected also were analyzed for urinary CXCL9 and CXCL10 using sandwich ELISA as published for comparison.⁴¹

RESULTS AND DISCUSSION

Principle of ImmunoMag-CRISPR LFA.

To improve the sensitivity for detecting urinary CXCL9, we combined a magnetic bead-based Immuno-CRISPR assay with a paper-based lateral flow assay (LFA) to develop the ImmunoMag-CRISPR LFA (Figure 1). Target biomarkers in urine, such as CXCL9 and CXCL10, are captured and concentrated by antibody-coated magnetic beads, followed by the formation of a sandwich structure with a detection antibody-conjugated DNA barcode complex having multiple Cas12a recognition sites.²⁵ Visual signals in a paper-based LFA are then generated when the Cas12a/crRNA complex, which is activated by a specific DNA barcode sequence, cleaves FB reporter substrates.

Figure 1.

Schematic illustration for detecting urinary CXCL9 or CXCL10 using ImmunoMag-CRISPR assay with antibody–DNA barcode conjugates of Cas12a/crRNA complex and collateral FB substrates. (1) Sandwich structure with a capture antibody on magnetic bead, a target biomarker, and a detection antibody-streptavidin (SA)–DNA barcode complex formed by the antigen–antibody interaction, (2) followed by adding the Cas12a/crRNA complex. (3) The cleaved FAM and biotin can be detected in the test line and control line, respectively, of an LFA strip.

Construction of magnetic bead-based immunoassay.

To optimize the detection of urinary CXCL9 and CXCL10 with the magnetic bead-based immunoassay, we immobilized the capture antibody on three types of magnetic beads (Figure S1a); carboxylate-functionalized, streptavidin-coated and polystyrene-coated magnetic beads, via covalent attachment, biotin-streptavidin binding, and simple adsorption, respectively. To evaluate the detection sensitivity of each magnetic bead-based immunoassay, we used a detection antibody-HRP complex (Ab-HRP). The signal-to-noise ratio (SNR) was calculated by dividing the activity in the presence of the target biomarker by the activity in the absence of the target biomarker. Through the evaluation of HRP activity, the magnetic bead-based immunoassay using carboxylate-functionalized magnetic beads showed the best performance (Figure 2a) for detecting CXCL9 in synthetic urine when compared to assays using streptavidin-coated and polystyrene-coated magnetic beads (Figure S1b). The capture antibody loading of carboxylate-functionalized magnetic beads was measured to be 5.3 μg/mg bead, which is higher than that of both streptavidin-coated and polystyrene-coated magnetic beads (Figure S1c). Based on this, the improved SNR of magnetic bead-based immunoassay using carboxylate-functionalized magnetic beads is most likely due to the enhanced loading of capture antibody on the magnetic bead, and a lower background signal, compared to polystyrene-coated magnetic beads. We then optimized the magnetic enrichment time of CXCL9 and the binding time of the Ab-HRP to 20 min each and chose 30 min as the enzymatic reaction time (Figure S2).

Figure 2.

ImmunoMag-CRISPR assay using antibody-HRP complex (a) and antibody-DNA barcode complex for detection of CXCL9 (a and b) and CXCL10 (c) in synthetic urine. Calibration curves of ImmunoMag-CRISPR assay using antibody-HRP complex (d) and antibody-DNA barcode complex for detection of CXCL9 (d and e) and CXCL10 (f) at concentrations from 0 to 1000 pg/mL in synthetic urine. Error bars represent means and SDs from three independent experiments. Signal-to-noise (SNR) is the ratio of measured activity to background activity in the absence of target biomarker. The LOD was calculated from each standard curve by determining the CXCL9 or CXCL10 concentration with a fluorescence signal consistent with 2√2 times the standard deviation of the blank.

CXCL9 and CXCL10 detection using ImmunoMag-CRISPR assay in solution.

The Cas12a-coupled ImmunoMag-CRISPR assay combines capture antibody-immobilized magnetic beads and detection antibody-streptavidin-DNA barcode (Ab-DNA barcode) complexes. Di-biotinylated DNA barcodes were designed such that Cas12a trans-cleavage activity is not affected by modification of the DNA barcode (Figure S3), which shows no difference between the di-biotinylated and single biotinylated cases, and complexation can be performed with streptavidin and biotinylated detection antibodies. FQ reporter is used as a substrate to measure trans-cleavage activity of Cas12a activated by the DNA barcode complexes. The SNR of Cas12a-coupled ImmunoMag-CRISPR assay (170) (Figure 2b) was higher than that of the HRP-coupled magnetic bead-based immunoassay (67) (Figure 2a), indicating that magnetic bead-based immunoassay incorporated the Ab-DNA barcode complex. Moreover, the lower background signal of the Cas12a-coupled assay compared to the HRP-based assay (Figures S2b and c) contributed to the former having a higher SNR than the HRP-based assay (Figures 2a and b). We also applied this Cas12a-coupled assay to detect CXCL10 (Figure 2c) and observed an SNR for CXCL10 (85), which was also functional.

The ImmunoMag-CRISPR assay was then evaluated under conditions of varying CXCL9 concentrations spiked into synthetic urine and compared to a magnetic bead-based immunoassay using an Ab-HRP complex. The LODs of CXCL9 using the ImmunoMag-CRISPR assay with Ab-DNA barcode complex was estimated to be 0.6 pg/mL, which is more than 13-fold lower than that of the magnetic bead-based immunoassay using Ab-HRP complex (Figures 2d and e). This is because ImmunoMag-CRISPR assay is a highly sensitive fluorescence detection method with high S/N. The LOD of our previous well plate-based Immuno-CRISPR assay was 14 pg/mL.²⁵ Hence, the LOD of ImmunoMag-CRISPR assay is more than 20-times lower than that of the well plate-based Immuno-CRISPR assay as a result of the enhanced capture and concentration of target CXCL9 proteins by magnetic enrichment. The sensitivity of ImmunoMag-CRISPR assay for detection of CXCL9 is significantly improved over chemiluminescence-based single-plex micropatterned aqueous two-phase (LOD: 28 pg/mL)⁵⁷ and AlphaLISA (LOD: 20 pg/mL)⁵⁸ assays. The ImmunoMag-CRISPR assay was then applied to detect CXCL10, which resulted in an LOD of 2.4 pg/mL (Figure 2f), which is nearly 10-fold lower than that of the HRP-based ELISA (Figure S4).

Development of ImmunoMag-CRISPR LFA for CXCL9 and CXCL10 detection.

To combine a paper-based LFA system with ImmunoMag-CRISPR for the detection of CXCL9 and CXCL10, FB reporter was used as a signaling molecule to visualize the trans-cleavage activity of Cas12a using a commercially available LFA strip.^35,38 Intact FB reporter molecules are captured at the streptavidin-coated control line, while a visual signal is generated at the anti-rabbit antibody-coated test line via the trans-cleavage activity of Cas12a, which is activated by DNA barcode complexes (Figure 1).

ImmunoMag-CRISPR LFA was first tested as a function of CXCL9 concentration spiked into synthetic urine (Figures 3a and b). We also compared ImmunoMag-CRISPR LFA with a lateral flow assay combined with the well plate-based Immuno-CRISPR assay (Immuno-CRISPR LFA) as a control (Figure 3c).²⁵ The LOD of each assay was estimated by evaluating the ratio of the test to control line signal intensity (T/C ratio) for various concentrations of CXCL9 or CXCL10 using ImageJ density analysis (Figures 3d, e and f).^35,59,60 The LOD of ImmunoMag-CRISPR LFA for CXCL9 was calculated to be 18 pg/mL (Figure 3d). In contrast, the LOD of 96 well plate-based Immuno-CRISPR LFA for CXCL9 detection was 570 pg/mL (Figure 3f). This significant enhancement of detection sensitivity resulted from enrichment of the target biomarker (CXCL9) and reduction of non-specific binding in ImmunoMag-CRISPR LFA (Figure S5a). A urinary CXCL9 level of 200 pg/mL is a threshold value for early detection of kidney transplant rejection using the traditional HRP-based ELISA assay,⁴³ and corresponds to a T/C ratio of 0.9 ± 0.1 in the ImmunoMag-CRISPR LFA (Figures 3a and d). We then applied ImmunoMag-CRISPR LFA to detect CXCL10 spiked into urine and observed a similar trend as with CXCL9 detection (Figures 3b and e). We also confirmed no cross-reactivity between urinary CXCL9 and CXCL10 (1000 pg/mL each) in the ImmunoMag-CRISPR LFA (Figure S5a), as well as the ImmunoMag-CRISPR assay (Figure S5b). Specifically, CXCL9 with the CXCL10 capture antibody and CXCL10 with CXCL9 capture antibody do not result in a signal above background.

Figure 3.

Visual inspections (a, b and c) and their respective calibration curves (d, e and f) of ImmunoMag-CRISPR LFA (a, b, d and e) and Immuno-CRISPR LFA (c and f) with antibody-DNA barcode complexes, Cas12a/crRNA complex and collateral fluorophore-biotin reporters for detection of CXCL9 (a, c, d and f) and CXCL10 (b and e). The ratio of the test to control line signal intensity (T/C ratio) was evaluated by using ImageJ (NIH). The LOD was calculated from the standard curve of each LFA system generated by determining the CXCL9 or CXCL10 concentration with a T/C ratio consistent with 2√2 times the standard deviation of the blank.

Finally, we tested patient urine samples (n = 42), which were obtained from the Clinical Trials in Organ Transplantation (CTOT)-19 trial⁵⁶ using ImmunoMag-CRISPR LFA for the detection of both CXCL9 and CXCL10 (Figures 4 and S4). ImmunoMag-CRISPR LFA samples with T/C ratio ≥ 0.9, corresponding to HRP-based ELISA at least 200 pg/mL CXCL9 were classified as positive, while T/C ratio < 0.35, corresponding to HRP-based ELISA < 100 pg/mL (LOD value) were classified as negative. Intermediate ranges in the two assays were classified as borderline positive. Based on these cutoff ranges, 11 samples in ImmunoMag-CRISPR LFA were classified as positive and 12 samples in HRP-based ELISA were classified as positive. The extra positive with HRP-based ELISA was sample 20, which had T/C = 0.85 determined by ImmunoMag-CRISPR LFA and 220 pg/mL CXCL9 determined by HRP-ELISA, both values near the positive cutoff values for the two methods (red dots in Figure 5).

Figure 4.

Comparison of ImmunoMag-CRISPR LFA with antibody-DNA barcode complex and traditional HRP-based ELISA for detection of CXCL9 in 42 randomly chosen urine samples obtained from patients participating in the CTOT-19 trial. T/C ratio was evaluated by using ImageJ (NIH). ImmunoMag-CRISPR LFA results were classified as positive (+), borderline positive (+/−), or negative (−) based on T/C ratios of 0.9 or greater, 0.35 to 0.9, and less than 0.35, respectively. HRP-based ELISA results were also classified as positive, borderline positive, or negative based on concentrations (pg/mL) of 200 or greater, 100 to 199, and less than 100, respectively. ND, not detected (less than LOD for the given assay).

Figure 5.

Comparison of the CXCL9 concentrations of clinical samples evaluated by HRP-based ELISA and T/C ratios evaluated by ImmunoMag-CRISPR LFA. ND, not detected (less than LOD for the given assay).

The increased sensitivity of the ImmunoMag-CRISPR LFA vs. HRP-ELISA is evident in the detectable (> LOD) signal for the former vs. the latter. Specifically, for 0.20 < T/C < 0.35, five samples had detectable levels of CXCL9. These all correspond to CXCL9 < 100 pg/mL, and hence, undetectable as being below LOD for the HRP-ELISA assay. Interestingly, for the borderline positive samples, only five are detected by HRP-ELISA vs. 11 for ImmunoMag-CRISPR LFA. This suggests that the ImmunoMag-CRISPR LFA has a greater dynamic range and can better distinguish between borderline positive and negative urinary CXCL9 levels.

Broadly similar results were obtained for CXCL10 detection, although there is no clearly accepted positive cutoff value in the literature. Based on comparison of ImmunoMag-CRISPR LFA and HRP-ELISA for CXCL10 (Figure S6), the former had 22 samples showing > LOD detectable signal vs. 17 for the latter, once again reflecting the increased sensitivity of the ImmunoMag-CRISPR LFA platform. In aggregate, these results suggest that ImmunoMag-CRISPR LFA has the potential for very early-stage identification of conditions that are suspicious of worsening injury in kidney transplant recipients.

CONCLUSIONS

We have developed the ImmunoMag-CRISPR LFA assay platform, which combines magnetic enrichment of target urinary biomarker proteins with visual detection in solution and on a lateral flow test strip. The LOD for solution-phase CXCL9 using ImmunoMag-CRISPR assay was 0.6 pg/mL, which is a 20-fold improvement over traditional well plate-based ELISA. Extending to the ImmunoMag-CRISPR LFA platform, we demonstrated an LOD of 18 pg/mL via naked-eye detection. ImmunoMag-CRISPR LFA determined 11 positive, 11 borderline positive and 20 negative urinary samples from kidney transplant recipients through a simple visual comparison of the test line and the control line of LFA strip. Importantly, of the 20 clinically negative samples, five samples had detectable levels of CXCL9 (above LOD on the LFA). Similar results were obtained for CXCL10. This suggests that very early stage CXCL9 and CXCL10 levels can be detected using the ImmunoMag-CRISPR LFA platform.

Given the opportunity for POCT, when coupled with a smartphone-based intensity measurement system, it may be possible for home measurements by kidney transplant patients spanning several months to assess time-based increases in their urinary chemokine levels to alert their physicians. Therefore, ImmunoMag-CRISPR LFA provides for a rapid, selective, and highly sensitive detection methodology for point-of-care testing of CXCL9 and CXCL10. This technique can be further combined with a smartphone-driven LFA quantification technique^19,49,50 and adapted to patient point-of-use home testing in future clinical analyses.

As the assay is further improved, the capture antibody can be engineered to increase the binding affinity and/or capture antibody loading can be increased using magnetic nanoparticles with a high surface area per unit mass. Microfluidic designs can be developed for improved bead handling, and rapid and automated serial steps can be performed with antigen, antibody-conjugated DNA barcodes, Cas12a complex and the fluorogenic substrate, thereby leading to even further streamlined point-of-care and potentially point-of-use applications.