CRISPR-Cas-mediated Multianalyte Synthetic Urine Biomarker Test for Portable Diagnostics

Summary

This protocol describes a CRISPR-Cas-mediated, multianalyte synthetic urine biomarker test that enables point-of-care cancer diagnostics through the ex vivo analysis of tumor-associated protease activities.

Abstract

Creating synthetic biomarkers for the development of precision diagnostics has enabled detection of disease through pathways beyond those used for traditional biofluid measurements. Synthetic biomarkers generally make use of reporters that provide readable signals in the biofluid to reflect the biochemical alterations in the local disease microenvironment during disease incidence and progression. The pharmacokinetic concentration of the reporters and biochemical amplification of the disease signal are paramount to achieving high sensitivity and specificity in a diagnostic test. Here, a cancer diagnostic platform is built using one format of synthetic biomarkers: activity-based nanosensors carrying chemically stabilized DNA reporters that can be liberated by aberrant proteolytic signatures in the tumor microenvironment. Synthetic DNA as a disease reporter affords multiplexing capability through its use as a barcode, allowing for the readout of multiple proteolytic signatures at once. DNA reporters released into the urine are detected using CRISPR nucleases via hybridization with CRISPR RNAs, which in turn produce a fluorescent or colorimetric signal upon enzyme activation. In this protocol, DNA-barcoded, activity-based nanosensors are constructed and their application is exemplified in a preclinical mouse model of metastatic colorectal cancer. This system is highly modifiable according to disease biology and generates multiple disease signals simultaneously, affording a comprehensive understanding of the disease characteristics through a minimally invasive process requiring only nanosensor administration, urine collection, and a paper test which enables point-of-care diagnostics.

Introduction

Despite the significant effort to identify tumor biomarkers such as shed proteins and DNA, the cancer diagnostic field has been strained by their low abundance or rapid degradation in circulation¹. As a complementary strategy, bioengineered synthetic biomarkers that selectively respond to disease features to generate amplified signals represent new avenues towards accurate and accessible diagnostics^2,3. To aid detection, these synthetic biomarkers harness tumor-dependent activation mechanisms such as enzymatic amplification to produce analytes with improved signal-to-noise ratio⁴. Herein, a class of cancer-associated enzymes, proteases, are leveraged to activate injectable nanoscale sensors to release disease reporters detectable from the biofluids such as blood or urine^5,6. In light of tumor heterogeneity, developing a panel of protease-activated sensors allows for multianalyte tests that combine different protease cleavage events into a 'disease signature' to assess cancer incidence and progression in a more specific, multiplexed manner.

Protease-activated synthetic biomarkers have been developed that comprise peptide substrates conjugated to the surface of an inert carrier⁷. When injected in vivo, these peptides are carried to the tumor whereupon enzymatic cleavage by tumor proteases release reporters into blood or urine for detection. Multiplexed detection with protease-activated synthetic biomarkers requires each synthetic biomarker within a cocktail to be labeled with a unique molecular barcode. To this end, various approaches have been developed, including mass barcodes and ligand-encoded reporters^8,9,10. As opposed to these methods of multiplexing which may be limited to a few different signal possibilities, DNA barcoding affords many more combinations in accordance with the high complexity and heterogeneity of human disease states. To expand the multiplexity of synthetic biomarkers, the sensors are barcoded by labeling each reporter with a unique DNA sequence for detection via CRISPR-Cas nuclease to amplify the biofluidic signal ex vivo. These single-stranded DNA (ssDNA) barcodes are designed to bind to complementary CRISPR guide RNAs (crRNAs), activating the target-triggered collateral nuclease activity of CRISPR-Cas12a¹¹. This nuclease activity can be employed to cleave a reporter DNA strand detected through fluorescence kinetics or using paper strips.

In addition to molecular amplification via proteases (in vivo) and CRISPR-Cas (ex vivo), another key design feature of protease-activated synthetic biomarkers involves harnessing nanomaterial pharmacokinetics to increase diagnostic signal concentration in biofluids¹⁰. One approach is the use of a nanoparticle carrier to increase the circulation time of surface-conjugated peptide substrates. A polyethylene glycol (PEG) dendrimer is selected as a nanocarrier with relatively small hydrodynamic diameter and multivalency to increase delivery to tumors. While small enough to promote tumor delivery, the size of the PEG carrier is larger than the ~5 nm size cut-off of the kidney glomerular filtration barrier so that only cleaved peptide substrates can be cleared into the urine, taking advantage of size filtration by the kidneys¹². In this protocol, the multiple-step workflow is outlined for the synthesis and application of DNA-barcoded activity-based nanosensors in a preclinical murine model, highlighting the setup of the CRISPR-Cas-mediated, multianalyte synthetic urine biomarker test, which has been employed by this group to classify disease status in murine models of multiple cancer types¹³. Owing to the versatile design principle, all three functional components of the nanosensor - the nanocarrier (PEG polymer), the stimuli-responsive module (protease-activated substrate), and the biofluidic reporter (DNA barcode) - can be precisely interchanged according to application-specific needs, allowing for modularity by tailoring the target and release specificities.

Protocol

All animal studies are approved by the Institutional Animal Care and Use Committee (IACUC) at the authors' institution. Standard animal care facilities including housing chambers, sterile hoods, anesthetization, and ethical endpoint euthanization are required to properly carry out these experiments. All experiments are conducted in compliance with institutional and national guidelines and supervised by the veterinarian staff at the institution. Female BALB/c mice, used for the experiments, are obtained from a commercial source (see Table of Materials) and ranged in age from 6 to 8 weeks at the start of the study. Sequences for custom-synthesized DNA, crRNA, FRET-based peptide substrate probes, and sensor peptides are provided in Supplementary Table 1.

1. Protease-activated peptide substrate selection

1. Collect and prepare tissue samples from healthy lung or lung tissue with tumor nodules following a previously published report¹³.
   1. Homogenize tissues in pre-chilled phosphate buffered saline (PBS, pH 7.4) (200 mg tissue/mL PBS) with tissue dissociation tubes for the automated dissociation of tissues into single-cell suspensions.
   2. Centrifuge tissue homogenates at 6,000 x g for 5 min at 4 °C. Retain the supernatant in a new tube.
   3. Centrifuge the supernatant at 14,000 x g for 25 min at 4 °C.
   4. Measure the protein concentration using bicinchoninic acid (BCA) protein assay kit (see Table of Materials).
   5. Add PBS to the sample to prepare a 0.33 mg/mL solution.
2. Assess the proteolytic activity following the steps below.
   1. In a 384-well plate, add 5 µL, 6 µM FRET-based peptide substrate probes to each well. For each probe, perform the reaction in triplicate.
      NOTE: FRET-based peptide substrate probe contains a short peptide sequence (6-8 amino acids, designed to be specific to a target protease or group of proteases) terminated with a fluorophore and quencher pair. Two FRET probes are described in Supplementary Table 1 as an example. The full library of probes can be found in Hao et al.¹³.
   2. Centrifuge the well plate for 10 s at room temperature at 180 x g to ensure the probes are at the bottom of the plate.
   3. Add 25 µL 0.33 mg/mL tissue sample or 40 µM recombinant protease¹³ to each well.
      NOTE: The final concentration of the tissue sample or recombinant protease may need to be optimized according to their intrinsic proteolytic activity. For instance, highly proteolytic tissues such as intestines may require higher dilution factors to allow for the monitoring of initial enzymatic cleavage.
   4. Centrifuge the well plate briefly at 180 x g (at room temperature) to mix the probes and tissue lysates.
   5. Immediately begin detecting probe cleavage by measuring fluorescence with the plate reader at 37 °C every 2 min for 1 h (λ_ex: 485 nm; λ_em: 535 nm) to monitor the cleavage.
      NOTE: Use the appropriate excitation and emission wavelengths for specific fluorophore and quencher pairs. In the case of this study, parameters (λ_ex: 485 nm and λ_em: 535 nm) are set for the FAM fluorophore.
   6. To analyze the fluorescent measurement data, utilize the Python (see Table of Materials) package for enzyme kinetics analysis available at https://github.com/nharzallah/NNanotech-Kinetic. This script calculates the initial reaction velocity (V₀) using the slope of the linear fit of the first 8-10 initial time points.

2. Sensor formulation and characterization

1. Synthesize the conjugate of DNA and protease-activated peptide (PAP).
   1. Incubate 20-mer phosphorothioated DNA reporters with 3'-DBCO group (1.1 eq., see Table of Materials) with azide-terminated PAPs with C-terminus cysteine end in 100 mM phosphate buffer (pH 7.0) at room temperature for >4 h. If leaving overnight, incubate at 4 °C.
   2. Purify the product on a high-performance liquid chromatography (HPLC) system equipped with a column ideal for biomolecules (see Table of Materials). Set the HPLC gradient to start from 5% A buffer (0.05% TFA in H₂O), keep isocratic for 20 min, and reach 80% B buffer (0.05% TFA, 99.95% acetonitrile) at 65 min with a flow rate of 0.3 mL min^-1, and collect the conjugate product with retention time at ~31 min.
   3. Validate the purified conjugates by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry using α-cyano-4-hydroxycinnamic acid as matrix¹⁴(see Table of Materials).
2. Synthesize the DNA-encoded activity-based nanosensors.
   1. Dissolve 2 mg of multivalent PEG (40 kDa, 8-arm, see Table of Materials) with maleimide-reactive group in 1 mL of 100 mM phosphate buffer (pH 7.0) and filter (cutoff: 0.2 µm).
   2. Add the cysteine-terminated DNA-peptide conjugates (2 eq., see Table of Materials) to the PEG and react at room temperature for >4 h. If leaving overnight, incubate at 4 °C.
   3. Remove the unconjugated materials using size exclusion chromatography with a commercially available dextran-agarose composite matrix column on a fast protein liquid chromatography (FPLC) (see Table of Materials). Run samples in PBS and monitor absorbance at 260 nm for DNA and 280 nm for peptide.
   4. Concentrate the synthesized nanosensors with centrifugal filter tubes (MWCO = 10 kDa) according to the manufacturer's recommended speed (see Table of Materials).
   5. Quantify the concentration of DNA using ssDNA assay kit (see Table of Materials) and a plate reader at λ_ex: 485 nm and λ_em: 535 nm. Store the nanosensors at 4 °C.
3. Characterize the DNA-encoded activity-based nanosensors.
   1. Measure the hydrodynamic particle size of DNA-encoded nanosensors by dynamic light scattering (DLS).
      NOTE: The expected size range of nanosensors is 15-50 nm, with an average size of 20-30 nm. If a limited size range is desired, FPLC (in step 2.3) can be used to isolate narrower fractions with different molecular weights.
   2. Concentrate the nanosensors to 0.5 mg/mL (by DNA concentration) with centrifugal filter tubes (MWCO = 10 kDa) and load the samples onto a carbon film-coated copper grid mounted on a cryoholder. Observe the morphology using cryogenic transmission electron microscopy (200 kV, magnification of 10,000-60,000)¹⁴.

3. Sensor injection and urine collection

1. Collect urine for baseline measurement.
   1. Place the mouse into a custom housing chamber (see Supplementary Figure 1) with a 96-well plate as the base.
   2. Restrain the mouse and apply gentle pressure on the bladder to void any remaining urine onto the plate.
   3. Pipette the collected urine (~100-200 µL) from the 96-well plate into a 1.5 mL tube after replacing the mouse to the normal housing.
2. Establish preclinical murine tumor model.
   1. Inoculate 6 to 8-week-old BALB/c female mice by intravenous injection with luciferase-expressing MC26-Fluc cell line (100k cells/mouse) (see Table of Materials). Monitor tumor progression weekly using an in vivo fluorescence imaging system.
      NOTE: Visible tumor burden indicated by luminescent signal occurs approximately in week 2 of injection of this particular cell line. Carefully check on tumor-bearing animals during the tumor progression on a regular basis.
3. Inject nanosensors at different time points after tumor implantation.
   1. Prepare an injection solution (200 µL maximum volume) containing nanosensors at a concentration of 1 nmol by DNA barcode in sterile PBS.
   2. Inject 200 µL sensor solution in PBS into each experimental mouse intravenously.
4. Collect urine samples from healthy control and tumor-bearing mice at 1 h after sensor injection as described in steps 1.1-1.3.
   ​NOTE: Fresh urine samples can be processed to DNA barcode analysis directly or frozen immediately on ice.

4. CRISPR detection of DNA barcodes: fluorescence-based

1. Use fresh urine samples or defrost frozen samples on ice. Centrifuge urine samples at 800 x g for 5 min at room temperature.
2. Combine the reagents in Supplementary Table 2, add the Cas12a enzyme (see Table of Materials) last and gently mix the reaction by pipetting up and down. Incubate the reaction at 37 °C for 30 min.
3. Run the reporter reaction, as shown in Supplementary Table 3, in triplicate using the product of step 2. Add the reaction from Step 2 last and quickly bring to the plate reader.
4. Detect LbaCas12a activation by measuring fluorescence with a plate reader at 37 °C every 2 min for 3 h (λ_ex: 485 nm and λ_em: 535 nm) to monitor the cleavage kinetics of the DNA reporter.
5. To analyze the fluorescent measurement data, utilize the Python package for enzyme kinetics analysis available at https://github.com/nharzallah/NNanotech-Kinetic. This script calculates the initial reaction velocity (V₀) using the slope of the linear fit of the first 8-10 initial time points.

5. CRISPR detection of DNA barcodes: paper-based

1. Centrifuge urine samples at 800 x g for 5 min at room temperature.
   NOTE: Run the urine samples for fluorescence-based and paper-based CRISPR detection in parallel.
2. Combine the reagents in Supplementary Table 2. Incubate at 37 °C for 30 min.
   NOTE: This incubation step is identical to that for the fluorescence-based CRISPR detection.
3. Run the reporter reaction using FAM-biotin labeled DNA reporter for lateral flow assay on a paper strip (see Table of Materials). Combine the reagents in Supplementary Table 4 in a 96-well plate using the product of step 2. Cover with aluminum foil and incubate at 37 °C for 1 h.
   NOTE: Select the optimal incubation time according to the real-time kinetics monitoring in the fluorescence-based CRISPR detection assay described above.
4. To a fresh well of a 96-well plate, add 80 µL of PBS. Add 20 µL of sample from step 3 to this well.
5. Place one lateral flow paper strip to each well and wait until the liquid reaches the top of the strip (<5 min). Look for the appearance of the control and/or sample band(s) on the paper strip.
6. Take a picture of the lateral flow strip and quantify band intensity using ImageJ.

Representative Results

Nominating protease-activated peptide substrates
To design sensors which will reflect changes in the proteolytic activity of the tissue, protease activity in the tissue is first characterized using a library of peptide probes¹³ (Figure 1). Fresh and frozen tissue samples can provide substantial information about the proteolytic activity of the tumor microenvironment by combining tissue samples with FRET probes designed to detect substrate cleavage. A library of FRET probes with a FAM fluorophore and a CPQ-2 quencher is incubated with tissue samples. Probes with the greatest difference in cleavage rate between sham tissue and tumor tissue, as measured by fluorescence spectrophotometry, are selected as peptide linkers for use in the in vivo nanosensors (Figure 1A). To better understand the physiology of the tumor environment and effectively compare protease expression profiles to activity profiles, the FRET probes may be incubated with recombinant proteases to associate substrate cleavage with specific protease activity, as demonstrated with two example probes in Figure 1B.

Sensor formulation and characterization
The nanosensors are composed of a polymeric core, peptides, and single-stranded DNA barcodes, as depicted in Figure 2A. The polymeric cores are 40 kDa, PEG dendrimers which act as carriers for up to 8 peptide-DNA conjugates (Figure 2B). The peptides, ~1.4 kDa, connect the core and DNA with an amino acid sequence designed to be cleaved through proteolytic activity. The single-stranded DNA barcode is 20 base pairs in length, ~6.8 kDa, and chemically modified for improved stability in vivo. Following conjugation, the peptide-DNA construct is purified via HPLC, allowing for separation of the conjugated and free constituents (Figure 2C). Mass spectrometry analysis indicates the conjugate has a molecular weight of 8.283 kDa, which is expected given the constituent molecular weights for the peptide and DNA barcode of 1.4 kDa and 6.8 kDa, respectively. The peptide-DNA conjugate is further conjugated to the PEG dendrimer. The resulting sensor is purified using FPLC to remove free peptide-DNA which exhibits a prolonged retention time compared with the PEGylated ones (Figure 2D). The size of the sensor can be characterized via dynamic light scattering and cryogenic transmission electron microscopy, which indicate an increase in diameter following peptide-DNA addition to the core from ~8 nm to ~20 nm (Figure 2B). Variance in particle size may be due to the flexibility of the constituent chains and a variable number of peptide-DNA arms conjugated to each PEG dendrimer.

Preclinical murine disease model
In vivo work to test the sensors involves establishing tumors for 3 weeks in the lungs of BALB/c mice, injecting the nanosensors, and collecting urine 1 h after sensor injection (Figure 3A). A baseline urine sample is also taken prior to tumor inoculation to compare with samples after disease induction. In order to assess the efficacy of the sensor in detecting malignancy, luciferase-expression colorectal cancer cell line MC26 (MC26-Fluc) is injected intravenously, resulting in lung tumor nodules visible in in vivo luminescence imaging. Hematoxylin and eosin staining¹³ reveals histopathology of the lung tissue from tumor-bearing and sham control mice, respectively, and evidence of tumor formation at 11- and 21-day post-injection, as indicated by black arrows in Figure 3B.

CRISPR activation by chemically stabilized DNA
To optimize Cas12a activation via ssDNA barcodes and complementary crRNA pairing, different oligonucleotide sequences and lengths have been tested. Activation of Cas12a is assessed by measuring the increase in fluorescence over time due to Cas12a cleavage of bystander DNA reporter with a FRET pairing. An increase in cleavage rate is associated with higher concentrations of chemically modified ssDNA barcode, as shown with a representative crRNA/barcode pair (Figure 4B). Importantly, a linear relationship between concentration and activation, as displayed in Figure 4B, enables the readout to be reflective of DNA barcode presence in the urine and indirectly reflective of proteolytic activity in vivo. Further, multiple different crRNA sequences, as detailed in Supplementary Table 1, can be used for Cas12a activation, which is critical for enabling multiplexing. DNA activators have been selected for the construction of in vivo sensors based on their similarity in assay performance. The chemically modified ssDNA which is critical for in vivo stability of the synthetic biomarker demonstrated Cas12a activation, in comparison with unmodified dsDNA and ssDNA (Figure 4C).

Multiplexed urinary detection of chemically stabilized DNA barcodes
Multiple crRNA-modified single-stranded DNA (ssDNA) activator pairs have been verified for their orthogonality between different sequences. This allows for simultaneous readout in multiple well assays, as depicted in Figure 5A. Additionally, modified DNA molecules in unprocessed urine can be detected using a colorimetric readout on lateral-flow paper strips, as shown in Figure 5B. The presence of the leading 'sample band' indicates the release of detectable FAM through reporter cleavage after Cas12a is triggered by the urinary DNA. When Cas12a is activated by the DNA activator in mouse urine, it cleaves the fluorescein (FAM)-biotin-paired oligonucleotide reporter, releasing the FAM molecule, which is then detectable on the 'sample band'. Uncleaved reporters are captured on the 'control band' due to the binding of biotin to streptavidin. Nanosensors are selected for in vivo experiments and constructed using protease-activatable peptides and distinct DNA barcodes. Peptides are nominated through ex vivo tissue profiling (Figure 1A) and DNA barcodes are selected based on similar Cas12a activation (Figure 4B).

Figure 1: Identification of peptide substrates activated by dysregulated proteases in colorectal cancer for sensor construction. (A) FRET-paired peptide substrates, each consisting of a peptide sequence flanked by a FAM fluorophore and a CPQ-2 quencher, are screened against recombinant proteolytic enzymes or tissue homogenates. (B) Representative protease cleavage kinetics of FRET-paired peptide substrates (Probe 1 & 2). The figure is adapted from Hao et al.¹³. Please click here to view a larger version of this figure.

Figure 2: Characterization of DNA-barcoded activity-based nanosensor with a polymeric PEG core. (A) DNA-barcoded nanosensors comprise polymeric nanocarrier (8-arm PEG) functionalized with protease-activated peptides barcoded with oligonucleotides. (B) Dynamic light scattering analysis shows an increase of particle size from 8.3 nm (PEG core only) and 13 nm (functionalized sensor). (C) HPLC purification of peptide-DNA conjugate. The conjugate is analyzed in mass spectrometry and shows expected molecular weight. (D) FPLC purification of the sensor shows separation of functionalized sensor and unbounded peptide-DNA conjugate. The figure is adapted from Hao et al.¹³. Please click here to view a larger version of this figure.

Figure 3: In vivo disease induction and sensor utilization. (A) Timeline of longitudinal tumor monitoring with nanosensor at different time points over tumor development. (B) Histological lung staining of BALB/c mice bearing CRC lung tumors at 11- and 21-day after tumor inoculation and saline-injected Sham control mice. Scale bar = 200 µm. Organs were fixed, embedded in paraffin, and stained with hematoxylin and eosin. The study was done with n = 3 mice per time point and images from a representative animal are shown. Arrows indicate tumor nodules in the lung. The figure is adapted from Hao et al.¹³. Please click here to view a larger version of this figure.

Figure 4: Cas12a activation by chemically stabilized DNA activators. (A) Activation of Cas12a by a representative ssDNA barcode through binding with complementary crRNA results in trans-cleavage of bystander DNA in a dose-dependent manner, as measured by fluorescence intensity via spectrophotometer. (B) The initial reaction velocity (V₀) is determined from the slope of the curve at the beginning of a reaction in (A) and plotted to determine the linear range of assay performance. (C) Cas12a activation by double-stranded, single-stranded, and chemically-modified DNA activators. The figure is adapted from Hao et al.¹³. Please click here to view a larger version of this figure.

Figure 5: Multiplexed CRISPR-Cas12-mediated DNA barcode readout with two detection options. (A) Trans-cleavage rates of Cas12a upon activation of different modified ssDNA activator-crRNA pairs are determined in the Cas12a fluorescent cleavage assay. Assays are performed with urine samples collected from mice injected with 1 nmol of modified ssDNA activator after 1 h of i.v. administration. The figure is adapted from Hao et al.¹³. (B) Paper readout of urine sample-activated Cas12a with colorimetric changes appear at different locations of the paper strip. The top band is from cleaved FAM DNA reporter and the bottom band is from uncleaved FAM-biotin reporter. Please click here to view a larger version of this figure.

Supplementary Figure 1: Urine collection chamber to place over a 96-well plate. Mouse is temporarily placed inside the cylinder for urination into the 96-well plate. Please click here to download this File.

Supplementary Table 1: Oligo and peptide sequences for sensor formulation and CRISPR assay. Please click here to download this File.

Supplementary Table 2: Reagents for fluorescence-based and paper-based Cas12a and urine sample reaction. Please click here to download this File.

Supplementary Table 3: Reagents of reporter reaction for fluorescence-based CRISPR detection. Please click here to download this File.

Supplementary Table 4: Reagents of reporter reaction for paper-based CRISPR detection. Please click here to download this File.

Discussion

Presented here is a highly customizable platform for multiplexed cancer detection with a portable urine test that assesses disease-associated proteolytic activity using a minimally invasive injected sensor. When activated by tumor proteases, peptide substrate cleavage is amplified via DNA barcode release into the urine. The synthetic DNA reporters in a urine sample can be read out by a secondary CRISPR-Cas-mediated enzymatic amplification using fluorometric detection or a simple paper-based test. DNA barcoding is an attractive method for labeling synthetic biomarkers to afford multiplexity. Improving the stability of the DNA barcodes through chemical modification is imperative to maintaining nanosensor stability in vivo. Specifically, DNA barcode detection via CRISPR-Cas12a can be activated using fully phosphorothioate-modified DNA barcodes with a stabilized backbone. The chemical modification reduces the speed of CRISPR-Cas12a activation relative to its native DNA counterpart, however, due to its increased stability. Further, inserting terminal modifications to the crRNA or testing different modified oligonucleotides in the DNA barcodes may further increase the sensitivity of the CRISPR-mediated DNA barcode detection¹⁵, thereby improving the limit of detection (LOD) of cancer diagnostics.

In addition to molecular amplification, another key strategy to attain the LOD required for early cancer detection involves taking advantage of size filtration by the kidneys to concentrate the synthetic DNA reporters in urine. To this end, it is critical to optimize the size of the DNA barcodes. 20-mer crRNA-complementary DNAs exhibited optimal CRISPR-Cas12a-mediated DNA barcode readout from the urine samples. This optimal length may differ when applied to different CRISPR nucleases or preclinical animal models.

To maintain the reproducibility of this detection system, it is important to conduct a thorough characterization of the injectable nanosensors to minimize batch-to-batch variation. Variation in the number of peptide-DNA conjugates per core will alter the amount of DNA barcodes released into the urine and therefore affect the final readout. Similarly, careful injection of the sensor intravenously will help consistency in sensor dosing. In order to ensure appropriate urine collection of at least 50 µL of urine volume from each animal at a given time point, maintaining tumor-bearing animal hydration is critical. This can be supported by keeping the mice on non-wetting water gel or injecting PBS subcutaneously one hour prior to urine collection¹⁶.

Timing of the urine collection and duration of CRISPR-Cas-mediated DNA barcode readout are critical factors for assay consistency. Barcodes in circulation undergo characteristic single-exponential concentration decay after intravenous injection and size-dependent renal filtration from the blood, peaking 1 h after administration in mouse models¹³. This time point may vary in a different animal model. For the two readout formats of CRISPR-Cas activation assays (fluorescent vs. paper-based lateral flow assay), it is important to track the fluorescence-based kinetics prior to the lateral flow, allowing for the optimal end-point to read the CRISPR-Cas cleavage product.

This approach is advantageous due to the incorporation of dual signal amplification steps from proteolytic and CRISPR-Cas-mediated enzymatic activation, and the multiplexing capability to capture the complexity of disease. Expanding the multiplexity of the sensors will necessitate careful material dosing as well as expanding the throughput of the readout to maintain portability and ease of use. This strategy requires a thorough characterization of protease activity in multiple models and the specificity of the proteolytic profile toward a particular disease, or disease stage. Further, while sensor injection is less invasive than many of the diagnostic alternatives such as biopsy, intravenous injection may rely on trained phlebotomists for sensor administration, potentially limiting the use case to disease monitoring rather than initial detection in a clinical setting. In addition to systemic injection, the DNA-barcoded nanosensors can be further engineered with formulations that can be applied via noninvasive oral, inhaled and topical delivery, to allow for portable disease detection and easy self-monitoring of disease progression and therapy assessment.

Materials

Name	Company	Catalog Number	Comments
10x NEB Buffer 2.1	New England Biolabs	B6002SVIAL
20-mer phosphorothioated DNA reporters with 3’-DBCO group	IDT		Custom DNA
Agilent 1100 High Performance Liquid Chromatography system with Vydac 214TP510 C4 column	Agilent		HPLC
ÄKTA fast protein liquid chromatography (FPLC)	GE Healthcare		FPLC
Amicon ultracentrifuge tubes (MWCO = 10 kDa)	EMD millipore		Various volumes available
Azide-terminated PAPs with C-terminus cysteine	CPC Scientific		Custom peptide
crRNAs	IDT		See Supplementary Table 1
Cryogenic transmission electron microscopy	JEM-2100F	JEOL	cyroTEM
Cysteine terminated DNA-peptide conjugates	CPC Scientific		Custom peptide
Dynamic light scattering (DLS)			DLS
EnGen LbaCas12a (Cpf1), 100 µM	New England Biolabs	M0653T
Experimental animals	Taconic Biosciences	BALB/cAnNTac	6–8 weeks of age
gentleMACS C tubes	Miltenyi Biotec	130-093-237	tissue homogenization
HybriDetect Universal Lateral Flow Assay Kit	Miltenyi Biotec	MGHD 1
Matrix-assisted laser desorption/ionization–time of flight (MALDI–TOF) mass spectrometry	Bruker		Microflex MALDI–TOF
MC26-Fluc cell line			Kenneth K. Tanabe Laboratory, Massachusetts General Hospital
multivalent PEG (40 kDA, 8-arm) with maleimide-reactive group	JenKem	A10020-1 / 8ARM(TP)-MAL-40K,1 g
Python, Version 3.9			https://www.python.org/
Quant-iT OliGreen ssDNA Assay Kit and Quant-iT OliGreen ssDNA Reagent	Invitrogen	O11492	ssDNA assay kit
ssDNA FAM-T10-Quencher and  FAM-T10-Biotin reporter substrates	IDT		Custom DNA
Superdex 200 Increase 10/300 GL column	GE Healthcare	GE28-9909-44	For FPLC
Tecan Infinite Pro M200 plate reader	Tecan
ThermoFisher Pierce BCA Protein Assay Kit	ThermoFisher Scientific	23225