Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
stm.sciencemag.org/cgi/content/full/12/547/eaaz2878/DC1
Supplementary Materials for
Molecular and functional extracellular vesicle analysis using nanopatterned
microchips monitors tumor progression and metastasis
Peng Zhang, Xiaoqing Wu, Gulhumay Gardashova, Yang Yang, Yaohua Zhang, Liang Xu*, Yong Zeng*
*Corresponding author. Email: [email protected] (L.X.); [email protected] (Y. Zeng)
Published 10 June 2020, Sci. Transl. Med. 12, eaaz2878 (2020)
DOI: 10.1126/scitranslmed.aaz2878
The PDF file includes:
Materials and Methods Fig. S1. Investigation of surface wettability and ink composition for colloidal inkjet printing. Fig. S2. Optimization of drop spacing for the stacked coins printing. Fig. S3. High-quality colloidal inkjet printing. Fig. S4. NTA plots of EVs isolated from breast cancer cell lines and patient plasma by UC. Fig. S5. Specificity of sEV immunoisolation with the EV-CLUE chips. Fig. S6. Optimization of capture antibodies for immunodetection of sEV-MMP14. Fig. S7. Verification of MMP14 expression on EVs derived from breast cancer cells. Fig. S8. Optimization of the sEV MMP14 activity assay. Fig. S9. Knockdown of MMP14 reduces MDA-MB-231 cell invasion. Fig. S10. Detection of the invasiveness of pancreatic cancer cells. Fig. S11. Validation of antibodies for sEV analysis in the mouse models. Fig. S12. Growth of subcutaneously xenografted tumors in 16 mice. Fig. S13. Time-lapse, multiparametric measurements of circulating sEVs in 16 mice using the EV-CLUE technology. Fig. S14. Box-whiskers charts with overlapping data points of time-lapse analysis of circulating sEVs in individual mice. Fig. S15. Comparison of MMP14 expression and activity of plasma sEVs in the mice developing primary tumors with or without lung metastasis. Fig. S16. Scatterplots of the SUM signatures for detecting breast cancer from the control. Fig. S17. Detection of the combined group of invasive and metastatic IDC from the control and DCIS groups in the training cohort. Fig. S18. Confusion matrices from discriminant classification of the patient cohorts. Fig. S19. Correlation circle derived from discriminant analysis of the training cohort. Fig. S20. Correlation between the sEV MMP14 expression and activity measured for all samples of the training and validation cohorts. Fig. S21. Detection of the combined group of advanced cases from the control and DCIS groups in the validation cohort.
Fig. S22. NTA plots of EVs isolated from patient plasma. Fig. S23. Correlation between the measurements of EVs in patient plasma with the EV-CLUE chip and NTA. Fig. S24. Characterization of UC-purified EVs from the sample set used in fig. S22. Fig. S25. Comparison of the EV-CLUE nanochip and standard microplate ELISA for analysis of breast cancer samples. Fig. S26. Representative images of H&E and IHC staining of MMP14 in patient-matched primary tumor tissues. Table S1. Antibodies and ELISA kits used in this research. Table S2. Summary of two cohorts of women controls and patients. Table S3. Statistical analyses of diagnostic performance of sEV markers for the training and validation cohorts. References (69, 70)
Other Supplementary Material for this manuscript includes the following: (available at stm.sciencemag.org/cgi/content/full/12/547/eaaz2878/DC1)
Data file S1 (Microsoft Excel format). Original data for the mouse model studies in Figs. 4 and 5. Data file S2 (Microsoft Excel format). Original data for the on-chip measurements of patient samples in Figs. 6 and 7.
Materials and Methods
Reagents and Materials. 10% (w/w) monodispersed silica colloids were purchased from
Bangs Laboratories Inc. (3-Mercaptopropyl) trimethoxysilane (3-MPS), 4-Maleimidobutyric
acid N-hydroxysuccinimide ester (GMBS) were purchased from Sigma-Aldrich.
Formamide was obtained from Fisher Scientific. The ELISA kits for MMP14, MMP15 and
MMP16 were ordered from R&D Systems and contain capture antibody, standard protein,
and detection antibody. Streptavidin conjugated β-Galactosidase (SβG),
Fluorescein-di-β-D-galactopyranoside (FDG), and VybrantTM CM-Dil cell staining solution
were purchased from Life Technologies. The FRET peptide substrate of MMP14
(SensoLyte 520) was ordered from AnaSpec Inc. The detailed information of antibodies
used in our studies was listed in table S1 below. 1× PBS solution and SuperBlock buffer
were from Mediatech, Inc and ThermoFisher Scientific, respectively. All other solutions
were prepared with deionized water (18.2 MV-cm, Millipore). SβG and FDG were
dissolved in PBS working solution (PBSW) at pH 7.4 which contain 0.5 mM
DL-dithiothreitol (Sigma-Aldrich), 2 mM MgCl2 (Fluka Analytical), and 0.5% bovine serum
albumin (BSA, Sigma-Aldrich).
Colloidal inkjet printing. A piezoelectric drop-on-demand inkjet printer DMP-2850
(Fujifilm Dimatix) was used for colloidal printing with a cartridge (Model No. DMC-11610)
that supports 10 pL droplets. The printer head consists of 16 nozzles in a row and the
operation of each nozzle can be controlled individually. The center-to-center drop spacing
is adjustable in one-micron increments within a 5 to 254 μm range. The patterns were
designed with AutoCAD and then converted into the bitmap images for printer input. Prior
to printing, the glass substrate was cleaned thoroughly with DI water under sonication. For
“stacked coins” printing mode, only one nozzle was used and the temperature of the substrate
was set to 50 oC to ensure that the evaporation time of single drop is less than the drop jetting
period. The drop size was adjusted by controlling the voltage of cartridge, which was
optimized to be 15 kV. For multi-layer printing, the inter-layer delay was 60 s.
Fabrication of EV-CLUE chip. Two-layer PDMS chips were fabricated by multi-layer
soft lithography according to our established protocol. Briefly, silicon wafers were cleaned
with piranha solution and spin-coated with 30 µm thick SU-8 2025 photoresist (MicroChem).
The SU-8 microstructures were fabricated onto the wafers from the photomasks, following
the protocols recommended by the manufacturer. Prior to use, the SU-8 molds were treated
with trichloro(1H,1H,2H,2H-perfluorooctyl)silane (Sigma-Aldrich) under vacuum for 8 h.
To fabricate the pneumatic layer, 30 g mixture of PDMS base and curing agent at a 7:1 ratio
was poured on the mold and cured in the oven at 70 oC for 2 h. The PDMS pieces were
peeled off from the mold, cut, and punched to make pneumatic connection holes.
Meanwhile, the fluidic layer was prepared by spin-coating the mold with 5 g mixture of
PDMS base and curing agent at a ratio of 15:1 at 1000 rpm for 45 s, followed by curing on a
70 oC hotplate for 30 min. The pneumatic layer was then manually aligned with the bottom
fluidic layer under a stereomicroscope and permanently bonded by baking in the 70 oC oven
overnight. The printed 3D nanopatterns were treated with 5% 3-MPS in ethanol for 1 h,
followed by heating at 80 oC for half an hour to stabilize the nanostructures. The coated
nanopatterns were then treated with 0.28 mg/mL GMBS for 0.5 h, which was used as a linker
to immobilize antibody. Using a patterning chip, the nanopatterns were washed with PBS
and then 0.1 mg/mL anti-CD81capture antibody was flowed through and incubated for 1 h at
room temperature. After washing with PBS, the patterning chip was removed and the
modified nanopatterns were then aligned and assembled with a flow-channel chip to construct
the complete microfluidic system. Finally, the channel surface was blocked with 5% BSA
for 1 h and stored at 4 °C before use. For SEM characterization, the nanopatterns were
coated with ~5 nm gold using a high resolution ion bean coater and then imaged with an FEI
Versa 3D Dual Beam scanning electron microscope.
sEV ELISA and activity assays on chip. The lyophilized standard EVs of COLO-1,
MCF7 and MDA-MB-436 cell lines were purchased from HansaBioMed, Ltd and
reconstituted in water prior to use. 5-10 μL samples (purified EVs or 5x diluted plasma)
were added into the inlet of each unit on the EV-CLUE chip and pneumatically pumped
through at an average flow rate of ~0.1 μL/min in a “stop-flow” manner (70). After
immuno-capture of sEVs, unbounded species were washed with 10 μL PBS. For ELISA
detection, specific biotinylated detection antibodies (a cocktail of CD9 and CD63, or MMP14,
20 μg/mL) were injected and reacted for 1 h. Excess antibodies were washed by PBS and
SβG prepared in PBSW buffer (20 ng/mL) was introduced as the reporter enzyme. After
another 10 min washing with 10 μL SuperBlock buffer, FDG in PBSW (500 μM) were
injected into the chamber and reacted in dark for 0.5 h before imaging readout. For the
parallel enzymatic activity assays, PBS was injected instead of the detection antibody and
SβG used for the sEV ELISA. The FRET peptide substrate of MMP14 was then injected
into the activity assay chambers and reacted for 1 h before imaging. Fluorescence images
were taken using a Zeiss Axiovert A1 inverted fluorescence microscope equipped with a
LED excitation light source (Thorlabs). Digital images were processed using ImageJ (NIH,
http://rsbweb.nih.gov/ij/) to quantify the fluorescence intensity.
Characterization of surface-captured sEVs followed our established protocols (23). Briefly,
for SEM, sEVs were fixed with 2.5% glutaraldehyde in PBS for 30 minutes and 1% osmium
tetroxide for 15 minutes, and then rinsed with water for 10 minutes. The samples were
dehydrated in ethanol with gradually increasing fraction (30%, 50%, 70%, 95% and 100%)
for 2×10 min each, coated with a gold thin film, and then examined with an FEI Versa 3D
Dual Beam SEM. Confocal imaging were done with an Olympus 3I spinning disk confocal
epifluorescence TIRF inverted microscope. Image stacks were taken in a 1 μm interval
along the z-axis, which ranged from the bottom of nanostructures to the top of flow channel.
The obtained image stacks were fitted into 3D view photography using SlideBook version
5.5.
Cell lines and culture conditions. Human cancer cell line MDA-MB-231 and MIAPaCa2
were purchased from American Type Culture Collection. To generate HuR knockout
sublines, MDA-MB-231 and MIAPaCa2 cells were infected with LentiCRISPRv2 lentiviral
vector (Addgene) to stably express control sgRNA or HuR sgRNAs. The cells were then
under puromycin selection for two weeks and single clones were generated. These cell lines
were cultured in DMEM (Mediatech) supplemented with 10% fetal bovine serum (FBS;
Sigma-Aldrich), 1% Glutamine (Mediatech), 1% antibiotics (Mediatech) in a 5% CO2
humidified incubator at 37 oC. For sEV studies, cells were grown in culture media that
contained 10% FBS depleted of EVs (Gibco). At confluency, cell medium was collected
and immediately used for sEV isolation.
Ultracentrifugation isolation of EVs. The supernatant of cell culture media was
centrifuged at 4 oC at 2,000× g for 10 min to remove large cell debris, 10,000 g for 45
minutes to remove large vesicles, and at 100,000 g for 2 h to pellet EVs. The supernatant
was carefully removed and EV pellets were then resuspended in 10 mL of PBS for washing
and collected again with UC at 4 oC for 60 min at 110,000 g in Beckman Coulter Quik-Seal
Centrifuge Tubes. After aspiration of the supernatant, EV pellet was resuspended in 100 µL
PBS. The aliquots of isolated EVs were stored at -80 oC.
Western Blot analysis. Western blotting was performed using 4-12% precast
polyacrylamide slab mini-gels (Tris-glycine pH 8.3) with Blot Module (Bio-Rad), following
the standard protocol. 30 μg cell lysate or ~1010 EVs were pretreated with RIPA lysis buffer
with protease inhibitors on ice for 45 min and heated at 72 °C for 10 min after adding equal
volume of 2× loading buffer. The electrophoresis was carried out at 125 V for 2 hrs, and
then gels were electro-transferred to the cellulose membranes (0.2 μm) at 25 V for 2.5 hrs.
The NC membrane was first blocked with Odyssey Blocking Buffer (PBS), then incubated
overnight at 4 °C in primary antibodies (table S1): rabbit anti-MMP14 (1:500), mouse
anti-HuR (1:500), mouse anti-α-tubulin (1:1000), and mouse anti-CD81 (1:1000). The
membranes were washed 3 times for 10 min each (1×PBS, 0.5% Tween 20, pH 7.4) and then
incubated with anti-mouse or anti-rabbit IRDye 680 (1:7500) or 800(1:15000) from LI-COR
for 60 minutes at room temperature. After that, the washing step was repeated three times.
Imaging was performed using an Odyssey Fc Imaging System (LI-COR Biosciences).
Cell invasion assay. To analyze cell invasion, Corning BioCoat Matrigel Invasion
Chambers were used. 1105 cells in 0.5 mL of serum free medium were seeded in
Matrigel-coated upper chambers and incubated for 22 h at 37°C, 5% CO2 atmosphere. Cells
were fixed with 95% methanol and stained with 0.1% crystal violet. Non-invading cells
were removed from the upper surface of the membrane by cotton swabs. Cells that invaded
were visualized and photographed with EVOS FL cell imaging systems (Life Technologies)
under 4 and 20 magnification.
Animal experiments. In the experimental metastasis model, 1106 2LMP cells stably
expressing luciferase were injected into tail veins of 4-week old female nude mice (Charles
River). Bioluminescence imaging was taken weekly to monitor tumor burden at lung.
Specifically, mice were interperitoneally injected 150 mg/kg D-luciferin dissolved in PBS,
anesthetized and imaged in MS FX PRO small animal imaging systems. At each time point
specified in the main text, ~50 L blood was collected from tail veins of mice in the
EDTA-coated tubes (Microvette 100 K3E, Sarstedt AG Co.) to prepare plasma samples for
microfluidic sEV profiling. In the spontaneous metastasis model, 0.5 106 4T1 cells stably
expressing luciferase were injected into #4 mammary fat pad of 4-week old female BALB/c
mice. Primary tumor sizes were measured using a caliper twice a week. Tumor volume
was calculated using the formula: (length × width2)/2. Bioluminescence imaging was taken
weekly to monitor the primary tumor and the metastasis burden at lung. ~50 L blood was
collected from tail veins of mice weekly with the EDTA-coated tubes and plasma was
immediately prepared for on-chip sEV analysis.
Patient specimen and clinical EV analysis. De-identified plasma samples from patients
with breast cancer and cancer-free individuals with accompanying clinical information (table
S2) were obtained from the KU Cancer Center’s Biospecimen Repository Core Facility
(BRCF). Blood specimens (plasma samples) were collected from women enrolled under the
repository’s Institutional Review Board (IRB)–approved protocol (HSC #5929) and
following U.S. Common Rule. Once the patient provides written, informed consent in
accordance with the BRCF’s IRB protocol, blood was collected by BRCF staff and processed
for long-term storage at -80˚C. In this study, most patients had ductal carcinomas, which is
more prevalent in breast cancer (>80%) compared to lobular carcinomas, which account for
only ~10-15% of cases. We estimated the required sample size for evaluating diagnostic
accuracy via comparing the area under a ROC curve (AUC) with a null hypothesis value of
0.5. For conventional characterization of the samples, EVs were purified by UC and then
characterized by NTA sizing, Bradford assay, and Western blot, following the protocols we
established (71). For microfluidic analysis, 6-μL plasma was diluted with PBS by 5 times to
prevent channel clogging and used without any further pre-treatment. SEV assay and data
acquisition followed the same processes as for EV standards. For EV analysis with the
standard ELISA method, an ExoTEST Ready-to-Use Kit purchased from HansaBioMed, Ltd
was used, following the established protocol (23). Briefly, 10 μL plasma samples were
10-fold diluted to 100 μL and then added into each well of a 96-well plate. The microplate
was incubated on a microplate shaker at room temperature for 30 min and then put in a 4 °C
fridge for overnight incubation. The plate was washed with 200 μL of washing buffer per
well for 3 times and incubated with 100 μL of the biotinylated anti-MMP14 detection
antibody (2 μg/mL) for 30 min at room temperature and 2 h at 4 °C. The plate was washed
for 3 times, incubated with 100 μL of 1:5000 diluted HRP-streptavidin conjugate (15 min at
room temperature and 1 h at 4 °C), washed for 3 times, and then incubated with 100 μL of
chromogenic substrate solution in dark for 10 min at room temperature. After adding 100
μL of stop solution, absorbance at 450 nm was measured on a CYTATION 5 imaging reader
(BioTek) and subtracted by the background measured with PBS.
Histological analysis of patient-matched tissues. H&E and IHC staining was performed
in the Histology Laboratory at the KU Cancer Center according to the following procedure.
Four micron paraffin sections were mounted on Fisherbrand Superfrost slides and baked for
60 min at 60 °C then deparaffinized. Epitope retrieval was performed using a Biocare
Decloaking Chamber (pressure cooker) under pressure for 5 min, using pH 6.0 Citrate buffer
followed by a 10 min cool down period. Endogenous peroxidase was blocked with 3%
H2O2 for 10 min followed by incubation with a specific primary antibody for 30 min: 1:400
dilution of monoclonal MMP14 (Clone # 5H2) (R&D, MAB918). This was followed by
Envision+ anti-mouse secondary (Dako) for 30 min and DAB+ chromogen (Dako) for 5 min.
IHC staining was performed using the IntelliPATH FLX Automated Stainer at room
temperature. A light hematoxylin counterstain was performed, following which the slides
were dehydrated, cleared, and mounted using permanent mounting media.
Supplementary Figures
Fig. S1. Investigation of surface wettability and ink composition for colloidal inkjet
printing. (A) Printing of continuous patterns on a hydrophobic glass substrate modified by
silane chemistry. The printed small droplets merged into larger discrete droplets due to the
surface tension of the aqueous colloidal solution and low wettability of the substrate (left).
20% formamide was added to reduce the surface tension, which deposited larger discrete
patterns (right). (B) Printing on an untreated hydrophilic glass substrate produced
continuous colloidal particle patterns with significant distortion because of strong wetting on
the surface. Particles were found to accumulate sharply on the contact line due to the coffee
ring effect (left). While addition of up to 40% formamide reduced the coffee ring effect, it
resulted in worse patterning resolution owing to the lower surface tension and strong wetting
on the surface (middle and right). The printing procedure was adapted from (31, 32).
Scale bars: 100 μm.
Fig. S2. Optimization of drop spacing for the stacked coins printing. A series of drop
spaces of (A) 5 μm, (B) 10 μm, (C) 15 μm, and (D) 20 μm were tested for 10-cycle printing
of an array of sinusoidal stripes with 5% (w/w) 1-μm silica suspension. Increasing the
spacing improves the printing resolution, but results in more sparse distribution of the
assembled colloidal particles. Thus, 10 μm was chosen as the optimal spacing, considering
the balance between the resolution and the morphology of the printed colloidal micropatterns.
The printing procedure was detailed in the Methods section. Scale bars: 100 μm.
Fig. S3. High-quality colloidal inkjet printing. A large-area, complex colloidal crystal
pattern of 1 μm silica colloids printed on a microscope slide displayed uniform structural
colors at different angles across the entire pattern.
Fig. S4. NTA plots of EVs isolated from breast cancer cell lines and patient plasma by
UC. The plots were obtained by averaging five repeated measurements and the mean EV
sizes were determined to be 113 nm, 105 nm, 116 nm and 146 nm for MCF-7, MDA-MB-438,
MDA-MB-231 and plasma, respectively. 100 mL conditioned culture media were
processed to yield purified EVs in 1 mL PBS, while 1 mL plasma was used to produce a 50
μL suspension of EVs in PBS.
Fig. S5. Specificity of sEV immunoisolation with the EV-CLUE chips. UC-purified EVs
from various breast cancer cell lines were fluorescently stained and spiked in healthy plasma
at 106 μL-1. The spiked samples were measured on the chips coated with BSA or anti-CD81
mAb to determine the capture efficiencies. Error bars indicate one S.D. (n = 3).
Fig. S6. Optimization of capture antibodies for immunodetection of sEV-MMP14.
UC-purified vesicles (105 μL-1) from various breast cancer cell lines were captured on the
chip coated with individual antibodies as indicated, followed by sandwich detection using the
MMP-14 mAb. Error bars, one S.D. (n = 3).
Fig. S7. Verification of MMP14 expression on EVs derived from breast cancer cells. (A)
Western blot analysis of UC-purified EVs (30 μg) using anti-MMP14 mAb. CD81 was
assayed in each sample as the loading control. (B) Analysis of proteolytic activity of
MMP14 on EVs using a standard microplate-based activity assay. Equal amount of purified
EVs from each cell line (1 μg) was mixed with 10 μL MMP14 FRET probe in PBS and
reacted at 37 oC for 1h, followed by fluorescence detection on a microplate reader (Cytation 5,
BioTek). Error bars, one S.D. (n = 3).
Fig. S8. Optimization of the sEV MMP14 activity assay. (A) Verification of the specificity
of the MMP14 FRET peptide probe using a standard microplate-based activity assay.
Standard proteins of soluble and membrane-type MMPs were assayed with the MMP14 probe,
following the protocol recommended by the manufacturer. Signals were normalized against
that of MMP14. (B) Effects of chemical treatment by 4-aminophenylmercuric acetate
(APMA) on the standard microplate activity assay of recombinant pro-MMP14 protein. (C)
Effects of APMA treatment on the microplate-based MMP14 activity assay of UC-purified
EVs of various breast cancer cell lines. (D) Optimization of the enzymatic reaction time to
improve the detection of sEV MMP14 activity using the EV-CLUE chip. UC-purified
MDA-MB-231 EVs (106 μL-1) were captured on chip by anti-CD81 and incubated with the
fluorogenic peptide probe for time-lapse detection of the fluorescence signals. The optimal
reaction time was determined to be 60 min. Error bars, one S.D. (n = 3).
Fig. S9. Knockdown of MMP14 reduces MDA-MB-231 cell invasion. (A) Knockdown of
MMP14 in MDA-MB-231 cells with non-targeting control (NC) siRNA or MMP14-targeting
siRNA. siRNA knockdown was verified by RT-qPCR analysis of HuR and MMP14
mRNAs in total RNA extracted from the cells. GAPDH was used as housekeeping gene.
Error bars: one S.D. (n = 3). (B) Representative images of cell invasion assay in
MDA-MB-231 cells transfected with NC siRNA and MMP14 siRNA. Scale bars: 200 μm.
Fig. S10. Detection of the invasiveness of pancreatic cancer cells. (A) Matrigel invasion
assays showing reduced invasiveness for two HuR knockout (KO) clones than the parental
MIAPaCa2 cell line and the control KO clone. Scale bars: 200 μm. (B) Radar plot of the
nanochip analyses of the total sEV (CD9&CD63), MMP14 expression and activity of sEVs
derived from MIAPaCa2 and the control and HuR KO clones. 10 μL, 106 μL-1 purified EVs
were used in each assay on a chip and the measured signal intensity was corrected by the
corresponding background levels. (C) Correlation of the measured sEV MMP14 activity
and the number of invading cells counted in the Matrigel assays. The linear fitting was
performed using the Deming regression model at the 95% confidence level. In all cases,
anti-CD81 mAb was used for sEV capture and each measurement was done in triplicate.
Error bars: one S.D. (n = 3).
Fig. S11. Validation of antibodies for sEV analysis in the mouse models. (A) Detection of
total sEVs in the plasma from the control and human tumor-xenografted mice using the
anti-mouse and anti-human antibody pairs, respectively. Anti-CD81 was used for capture
and a mix of anti-CD9 and CD63 mAbs for detection. Human antibodies detected sEVs in
the plasma collected from a human tumor-xenografted mouse, but not in the plasma of a
control mouse. Mouse antibodies detected sEVs in the plasma collected from both control
and xenografted mice. (B) Detection of MMP14 protein on sEVs in the plasma from the
control and human tumor-xenografted mice using the anti-mouse and anti-human antibody
pairs, respectively. Human antibodies detected sEV MMP14 in the plasma of a human
tumor-xenografted mouse, but not in the plasma of a control mouse. Mouse antibodies
detected similar levels of sEV MMP14 in the plasma collected from both control and
xenografted mice. Statistical two-sample comparisons were conducted with the two-tailed
Student’s t-test with the significance level at P < 0.05. N.S., not significant. Error bars
indicate one S.D. (n = 3).
Fig. S12. Growth of subcutaneously xenografted tumors in 16 mice. 0.5 × 106 mouse
4T1-Luc breast cancer cells were injected into the #4 mammary fat pad of 4-week old female
BALB/c mice. Size of primary tumor in each mouse was measured with a caliper and the
tumor volume was calculated using the modified ellipsoidal formula: (length × width2)/2. P
values were determined by one-way repeated measures ANOVA with post-hoc Tukey’s
multiple comparisons tests at the significance level of P < 0.05.
Fig. S13. Time-lapse, multiparametric measurements of circulating sEVs in 16 mice
using the EV-CLUE technology. ~50 L blood was collected from each mouse via tail vein
at the indicated time points. 6 L plasma prepared from the blood was diluted by 5 times
and assayed directly on different chips to determine the mean and S.D. (error bars, n = 3).
Fig. S14. Box-whiskers charts with overlapping data points of time-lapse analysis of
circulating sEVs in individual mice. The box indicates the range of 25-75th percentile and
the Whisker range is set to be the 5-95th percentile of the mean values of (A) the total level
(CD9&CD63), (B) MMP14 expression, and (C) MMP14 activity of plasma sEVs measured
by the EV-CLUE chips. Middle lines indicate the median, while open squares symbol the
mean. Pairwise comparison P values were determined by the Tukey’s multiple comparisons
test at the significance level of P < 0.05.
Fig. S15. Comparison of MMP14 expression and activity of plasma sEVs in the mice
developing primary tumors with or without lung metastasis. (A) MMP14 expression, (B)
MMP14 activity. The box indicates the range of 25-75th percentile and the Whisker range is
set to be the 5-95th percentile. Middle lines indicate the median and open squares represent
the mean. Statistical difference was determined by two-way ANOVA followed by the
Tukey’s multiple comparisons test at the significance level of P < 0.05.
Fig. S16. Scatterplots of the SUM signatures for detecting breast cancer from the
control. SUM1 denotes the unweighted sum of the total sEV level (CD9&CD63) and sEV
MMP14 activity, while SUM2 is the unweighted sum of the sEV MMP14 expression and
activity levels. Error bars indicate the mean and one standard error of the mean.
Two-group comparison was done by non-parametric, two-tailed Mann–Whitney U-test at the
significance level of P < 0.05.
Fig. S17. Detection of the combined group of invasive and metastatic IDC from the
control and DCIS groups in the training cohort. Error bars in the dot plots indicate the
mean and one s.e.m. of the groups. P values were determined by Kruskal-Wallis one-way
ANOVA with post hoc Dunn’s test for pairwise multiple comparisons. All statistical
analyses were performed at 95% confidence level.
Fig. S18. Confusion matrices from discriminant classification of the patient cohorts. The
sEV markers were assessed individually and in various combinations. The training cohort
data was used to train the discriminant function model, which was then tested with the
independent validation cohort for multi-class diagnosis of breast cancer.
Fig. S19. Correlation circle derived from discriminant analysis of the training cohort. It
shows the relationship between the original variables (i.e., three sEV markers) with the two
canonical variables derived from the discriminant analysis that together capture 99.93% of
the variance.
Fig. S20. Correlation between the sEV MMP14 expression and activity measured for all
samples of the training and validation cohorts. Deming linear fitting was performed at the
95% confidence level. Error bars indicate one S.D. (n = 3).
Fig. S21. Detection of the combined group of advanced cases from the control and DCIS
groups in the validation cohort. Error bars in the dot plots indicate the mean and one s.e.m.
of the groups. P values were determined by Kruskal-Wallis one-way ANOVA with post
hoc Dunn’s test for pairwise multiple comparisons. All statistical analyses were performed
at 95% confidence level.
Fig. S22. NTA plots of EVs isolated from patient plasma. 10 samples for each BrCa
subgroup were randomly selected from the training and validation cohorts. For each sample,
0.6 mL plasma was processed to yield 30 μL purified EVs in PBS, which was diluted by 600
times for NTA. Measured EV sizes were presented as mean ± S.D.
Fig. S23. Correlation between the measurements of EVs in patient plasma with the
EV-CLUE chip and NTA. The results of NTA counting of UC-purified plasma EVs were
presented in Figure S22. The chip assay results of the total sEV levels (CD9 and CD63
combined) in the same plasma samples were adopted from Figs. 6 and 7. The linear curve
was obtained by Deming linear fitting at the 95% confidence level. Error bars indicate one
S.D. (n = 3).
Fig. S24. Characterization of UC-purified EVs from the sample set used in fig. S22. n =
10 each group. (A) The abundance and (B) mean size were measured by the NanoSight
NTA and (C) the total protein levels of EVs was by measured by the Bradford assay. The
error bars in the dot plots indicate the mean and one s.e.m. P values were determined by
Kruskal-Wallis one-way ANOVA with post hoc Dunn’s test for pairwise multiple
comparisons at the significance level of P < 0.05.
Fig. S25. Comparison of the EV-CLUE nanochip and standard microplate ELISA for
analysis of breast cancer samples. (A) SEV MMP14 levels measured by the two methods
for a subset of the cancer-free controls, DCIS, IDC, and locally metastatic patients (n = 15
each). The color intensity displays the mean signals that were averaged over triplicate
measurements and normalized against the 99th percentile of the signal range after
background subtraction. (B) The data sets from the two methods revealed a very good
correlation for the relatively high concentrations (i.e., >1100 a.u. for the nanochip assay).
The error bars represent one S.D. (n = 3). (C) ROC curves for evaluating the diagnostic
performance of sEV MMP14 measurements by two methods for breast cancer. (D) Scatter
plots of the sEV MMP14 levels measured by the two methods for differentiating individual
groups at progressing disease stages. Kruskal-Wallis one-way ANOVA with post hoc
Dunn’s test was used for the overall and pairwise multiple comparisons. The middle line and
error bar represent the mean and one s.e.m., respectively. All statistical analyses were
performed at 95% confidence level.
Fig. S26. Representative images of H&E and IHC staining of MMP14 in
patient-matched primary tumor tissues. 10× magnification. Barely perceptible to weak
cytoplasmic staining of MMP14 was observed in 100% tumor cells of the DCIS tissues from
patients #46 and #53. Weak to focal moderate MMP14 staining in 100% tumor cells was
observed for the IDC patient #72 with no definitive nuclear staining. For comparison,
tumor-adjacent normal breast tissue from the same IDC patient was assayed, which showed
absence of MMP14 staining. Weak to moderate cytoplasmic staining in 100% tumor cells
and weak to moderate nuclear staining in approximately 10% of tumor cells was observed for
the locally metastatic IDC patient #90.
Table S1. Antibodies and ELISA kits used in this research.
Antibody/Kit Vendor Catalog No. Clone Host Reactivity
Anti-CD9 (biotin) Ancell 156-030 C3-3A2 Mouse Human
Anti-CD63 (biotin) Biolegend 353018 H5C6 Mouse Human
Anti-CD81 (biotin) Ancell 302-030 1.3.3.22 Mouse Human
Anti-CD81 Ancell 302-820 1.3.3.22 Mouse Human
Anti-MMP14 Cell Signaling 13130S D1E4 Rabbit Human
Anti-HuR Santa Cruz sc-5261 3A2 Mouse Human
Anti-α-tubulin Sigma-Aldrich T5168 B-5-1-2
Mouse Human
MMP14 ELISA kit R&D Systems DY918-05 Human
MMP15 ELISA kit LifeSpan Biosciences LS-F21276 Human
MMP16 ELISA kit LifeSpan Biosciences LS-F12055 Human
Anti-CD81 LifeSpan Biosciences LS-C108453 Poly Rabbit Mouse
Anti-CD9 (biotin) LifeSpan Biosciences LS-C204846 EM-04 Rat Mouse
Anti-CD63 (biotin) LifeSpan Biosciences LS-C316997 Poly Rabbit Mouse
Anti-MMP14 (biotin) LifeSpan Biosciences LS-C686694 Poly Rabbit Mouse
IgG (FITC) Life Technologies 34-152-110413 Poly Goat Mouse
Table S2. Summary of two cohorts of women controls and patients.
Training Cohort (n = 30)
ID# Age
Range Clinical Diagnosis
Pathological Stages
ER/PR/HER2
1 55-60 Cancer free N/A N/A
2 25-30 Cancer free N/A N/A
3 55-60 Cancer free N/A N/A
4 60-65 Cancer free N/A N/A
5 50-55 Cancer free N/A N/A
6 50-55 Cancer free N/A N/A
7 65-70 Cancer free N/A N/A
8 65-70 Cancer free N/A N/A
9 75-80 DCIS, Right 0 ER 0%, PR 0%, Ki-67 15%
10 50-55 DCIS, Left 0 ER 0%, PR 0%, HER2 35%
11 50-55 DCIS, Right 0 ER 100%, PR 96%, Ki-67 2%
12 45-50 DCIS, Left 0 ER 98%, PR 15%, Ki-67 10 %, HER2 1+
13 60-65 DCIS, Left 0 ER 99%, PR 24%, Ki-67 6%
14 45-50 DCIS, Left 0 ER100%, PR98%, Ki-67 4%
15 50-55 DCIS, Right 0 ER 100%, PR 60%, Ki-67 3%
16 50-55 DCIS, Left 0 ER100%, PR0%, ki-67 1%
17 60-65 IDC, Right IIA ER100%, PR98%, HER2 negative
18 60-65 IDC, Left IA ER100%, PR100%, HER2 1+, Ki-67 6%
19 55-60 IDC, Right N/A ER99%, PR77%, HER2 0+, Ki-67 2%
20 65-70 IDC, Left IA ER98%, PR75%, HER2 1+, Ki-67 5%
21 55-60 IDC, Right IA ER100%, PR100%, HER2 0+, Ki-67 3%
22 55-60 IDC, Left IIA ER98%, PR65%, HER2 negative
23 55-60 IDC, Right IA ER100%, PR32%, HER2 1+, Ki-67 1%
24 55-60 Metastatic IDC, Left IIB ER 100%, PR 0%, HER2 equivocal on FISH, Ki-67
28%
25 45-50 Metastatic IDC, Right IIA ER 100%, PR 99%, HER2 Negative, Ki-67 8%
26 55-60 Metastatic IDC, Left IIA ER 100%, PR 68%, HER2 1+, Ki-67 4%
27 70-75 Metastatic IDC, Left IIB ER 91.96%, PR 96.28%, HER2: 0, Ki-67: 1.65%
28 55-60 Metastatic IDC, Right IIB ER 99%, PR15%, HER2 0, Ki-67 20%
29 40-45 Metastatic IDC, Right IIB ER >90%, PR 60%, HER2 IHC: 2+, Ki-67: 30%
30 55-60 Metastatic IDC, Right IIIA ER100%, PR90%, HER2 0, Ki-67 2%
Validation Cohort (n = 70)
ID# Age
Range Clinical Diagnosis
Pathological Stages
ER/PR/HER2
31 60-65 Cancer free N/A N/A
32 60-65 Cancer free N/A N/A
33 60-65 Cancer free N/A N/A
34 50-55 Cancer free N/A N/A
35 65-70 Cancer free N/A N/A
36 55-60 Cancer free N/A N/A
37 45-50 Cancer free N/A N/A
38 50-55 Cancer free N/A N/A
39 55-60 Cancer free N/A N/A
40 70-75 Cancer free N/A N/A
41 65-70 Cancer free N/A N/A
42 50-55 Cancer free N/A N/A
43 45-50 DCIS, Left 0 ER/PR +
44 65-70 DCIS, Left 0 ER 0%, PR 0%, Ki-67 18%
45 65-70 DCIS, Left 0 ER/PR -
46 55-60 DCIS, Right 0 ER 100%, PR 100%, Ki-67 1%
47 60-65 DCIS, Left 0 ER 9%, PR 10%
48 60-65 DCIS, Right 0 ER/PR -
49 55-60 DCIS, Right 0 ER 35%, PR 15%, Ki-67 5%
50 50-55 DCIS, Left 0 ER 99%, PR 54%
51 45-50 DCIS, Right 0 ER 100%, PR 100%, Ki-67 5%
52 75-80 DCIS, Left 0 ER 100%, PR 100%, HER2 1+, Ki-67 7%
53 60-65 DCIS, Right 0 ER 99%, PR 2%, HER2 n/a, Ki-67 22%
54 50-55 DCIS, Right 0 ER 100%, PR 85%, Ki-67 4%
55 60-65 DCIS, Right 0 ER100%, PR98%, Ki-67 7%
56 40-45 DCIS, Left 0 ER 99%, PR 94%, Ki-67 22%
57 50-55 DCIS, Right 0 ER 0%, PR 0%, Ki-67 28%
58 65-70 DCIS, Right 0 ER 100%, PR 80%
59 65-70 DCIS, Right 0 ER 100%, PR 57%, Ki-67 14%
60 50-55 DCIS, Left 0 ER/PR0%, Ki-67 24%
61 70-75 IDC, Left IA ER100%, PR100%, HER2 0+, Ki-67 1%
62 45-50 IDC, Right IA ER100%, PR98%, HER2 2+, Ki-67 18%
63 60-65 IDC, Left N/A ER89%, PR83%, HER2 2+, Ki-67 19%
64 55-60 IDC, Right IA ER100%, PR94%, HER2 1+, Ki-67 9%
65 75-80 IDC, Left IA ER 100%, PR 39%, HER2 negative
66 50-55 IDC, Left IIA ER 100%, PR 98%, HER2 1+, Ki-67 3%
67 60-65 IDC, Right IA ER 100%, PR 89%, HER2 1+, Ki-67 7%
68 65-70 IDC, Left IIB ER 99%, PR 99%, HER2 1+, Ki-67 2%
69 60-65 IDC, Left IIA ER100%, PR90%, HER2 0+, Ki-67 4%
70 50-55 IDC, Right N/A ER100%, PR94%, HER2 1+, Ki-67 4%
71 45-50 IDC, Right N/A ER98%, PR98%, HER2 negative
72 60-65 IDC, Right IA ER100%, PR85%, HER2 1+, Ki-67 1%
73 65-70 IDC, Left IA ER73%, PR44%, HER2 1+, Ki-67 6%
74 70-75 IDC, Right IA ER100%, PR85%, HER2 1+, Ki-67 12%
75 50-55 IDC, Left and Right IA ER99%, PR99%, HER2 2+, Ki-67 20%
76 45-50 IDC, Right IA ER 98%, PR 0.5%, HER2 1+, Ki-67 4%
77 60-65 IDC, Left IA ER 99%, PR 85%, HER2 1+, Ki-67 3%
78 60-65 IDC, Left IA ER 99%, PR 65%, HER2 1+, Ki-67 9%
79 55-60 IDC, Left IA ER 99%, PR 2%, HER2 1+, Ki-67 1%
80 50-55 IDC, Right IA ER95%, PR25%, HER2 1+, Ki-67 5%
81 60-65 Metastatic IDC, Left IIA ER 100%, PR 98% | HER2 1+ | Ki-67 5%
82 80-85 Metastatic IDC, Right IIIA ER 99%, PR 8%, HER2 - not amplified on FISH,
Ki-67 15%
83 65-70 Metastatic IDC, Left IIB ER 100%, PR 0%, HER2 not amplified by FISH,
Ki-67 1%
84 40-45 Metastatic IDC, Right IIB ER 99%, PR 90%, HER2 1+, Ki-67 15%
85 75-80 Metastatic IDC, Right IIB ER 100%, PR 100%, HER2 2+, FISH 0.9 Negative,
Ki-67 16%
86 65-70 Metastatic IDC, Left IIIC ER 100%, PR 2%, HER2 Negative, Ki-67 5%
87 70-75 Metastatic ILC, Left IV ER 100%, PR 95%, HER2 neu +1, Ki-67 5%
88 45-50 Metastatic IDC, Left;
IDC, Right IIB
Left, ER 99%, PR 99%, HER2 -, Ki-67 15% Right, ER 98%, PR 100%, HER2 -, Ki-67 3%
89 45-50 Metastatic ILC, Left IIIB ER 80%, PR 80%, HER2 1+
90 40-45 Metastatic IDC, Right IIA ER 99%, PR 85%, HER2 1+, Ki-67 28%
91 40-45 Metastatic IDC, Right IIIA ER 0%, PR 0%, HER2 0, Ki-67 80%
92 50-55 Metastatic IDC, Left IIA ER 95%, PR 95%, Her2 1+, Ki-67 5%
93 50-55 Metastatic IDC, Left IIB ER 99%, PR 99%, HER2 1+, Ki-67 10%
94 40-45 Metastatic IDC, Right IIA ER99%, PR99%, Her2 1+, Ki-67 1%
95 40-45 Metastatic IDC, Left IIIA ER98%, PR98%, HER2 1+, Ki-67 5%
96 60-65 Metastatic IDC, Left IIA ER/PR0%, HER2 3+,Ki-67 10%
97 50-55 Metastatic IDC, Right IIB ER 100%, PR 63%, HER2 1+, Ki-67 13%
98 55-60 Metastatic IDC, Right IIA ER 90%, PR <1%, HER2 negative, Ki-67 40%
99 55-60 Metastatic IDC, Right IIA ER99%, PR19%, Her2 1+, Ki-67 6%
100 70-75 Metastatic IDC, Left IIA ER 100%, PR 17%, HER2 1+, Ki-67 7%
DCIS: ductal carcinoma in situ; IDC: invasive ductal carcinoma; ILC: Invasive lobular carcinoma.
Table S3. Statistical analyses of diagnostic performance of sEV markers for the training
and validation cohorts. 95% CIs are indicated in parentheses.