The devices were tested for their frequency response:

1
Development of a Simple Inexpensive Bulk Acoustic
Wave (BAW) Nanosensor for Cancer Biomarkers
Detection of Secreted Sonic Hedgehog from
Prostate Cancer Cells
Christopher Corso1, Anthony Dickherber2, Payal
Shah3, Alexandra Migdal3, Milton W. Datta3,
Sumana Datta3, and William Hunt2
1Department of Biomedical Engineering, Georgia
Institute of Technology, Atlanta, GA
30332 2Department of Elecrical and Computer
Engineering, Georgia Institute of Technology,
Atlanta, GA 30332 3Departments of Pathology and
Urology, Winship Cancer Institute, Emory
University, Atlanta, GA 30322
Abstract 8866

• Methodology (Continued)
• BAW Array Fabrication and Testing
• The metallic electrode arrays were fabricated on
  a 3-layer, piezoelectric Ta2O5 and SiO2 stack on
  a Si wafer by RF Sputtering

Abstract Introduction  Acoustic wave device
technology and standard photolithographic
processes can be employed to produce small
inexpensive sensors as disposable assay platforms
for cancer biomarkers.  We are developing a Bulk
Acoustic Wave (BAW) sensor chip containing 16
independent sensors that can be conjugated with
antibodies and used as a biosensor in complex
solutions such as serum or blood.  Advanced
prostate cancers have increased levels of Sonic
Hedgehog (SHH) production and signaling, and
therapeutics that target the SHH pathway stop
prostate cancer cell proliferation.  A biosensor
that could detect SHH could be used in prostate
cancer diagnosis and treatment. Methods  We
have developed a standardized BAW sensor chip
platform that can be used as a biosensor in
complex solutions such as blood or serum. While
it is not yet ready for testing, progress towards
the finished sensor chip is moving rapidly and a
testable product is expected soon. A Quartz
Crystal Microbalance (QCM) is a BAW device with
an identical mode of physical operation to our
chip, and has been implemented by us to prove the
viability of the approach. The QCM platform was
conjugated with anti-Sonic Hedgehog antibodies,
and the resulting BAW-SHH biosensor was used to
assay for the presence of SHH in conditioned
medium from LNCaP prostate cancer cells.
 Resultant sensitivities and detection range were
calculated for the BAW-SHH biosensor.Results
 SHH antibodies were efficiently coupled to the
BAW sensor through a self-assembled monolayer
(SAM) alkane-thiol crosslinker. This
immobilization procedure is a simple 7-step
process that takes less than 10 hours.  The
QCM-SHH biosensors were capable of detecting SHH
in undiluted conditioned medium in a repeatable
manner.  Sensor reaction curves were notable for
detection above the noise background for the
undiluted sample and will be further tested with
serially diluted samples until the detection
limit is reached. In the undiluted samples, the
average resonant frequency shift was 80 Hz which
corresponds to roughly 100 ng of SHH bound to the
device. Conclusions  Here we propose the first
use of a bulk acoustic wave sensor chip for the
detection of a cancer biomarker in complex media.
 We have created an inexpensive BAW sensor chip
whose production cost is a few cents per chip.
 These chips can be rapidly and efficiently
conjugated to antibodies and used for the
detection of circulating antigens in complex
solutions such as blood or serum.  The proof of
concept was shown with a quartz crystal
microbalance device. This is demonstrated through
the conjugation of anti-SHH antibodies and the
subsequent detection of SHH in conditioned medium
from prostate cancer cells. While the sensitivity
of the quartz crystal microbalance is high
(1x10-9 g per 1 Hz shift), theoretical
calculations show that the bulk acoustic wave
sensor chips will have a much higher sensitivity
than the QCM devices (1x10-15 g per 1 Hz shift).
Additionally, the array formation of our devices
allows for immediate and efficient repeatability
of the test as well as the possibility for
statistical analysis of the results. The ability
to detect low levels of Sonic Hedgehog in serum,
whether combined with or independent of serum PSA
levels, could be used to diagnose aggressive
prostate cancers or monitor response to treatment.
Introduction Acoustic wave device technology and
standard photolithographic processes can be
employed to produce small inexpensive sensors as
disposable assay platforms for cancer biomarkers.
 We have developed a Bulk Acoustic Wave (BAW)
piezoelectric sensor chip containing an array of
8 independent sensors that can be conjugated with
antibodies and used as a biosensor in complex
solutions such as serum or blood.  Advanced
prostate cancers have increased levels of Sonic
Hedgehog (SHH) protein production and signaling.
Therapeutics that target the SHH pathway have
been shown to stop prostate cancer cell
proliferation.  A biosensor that could detect SHH
could be used towards prostate cancer diagnosis
and treatment.

• The devices were tested for their frequency
  response
• Uncoated
• Coated only with antibodies Anti-SHH and
  Anti-FITC (pre-LNCap conditioned medium exposure)
  
• After being exposed to 5 ul of LNCap conditioned
  medium for 20 minutes followed by a buffer wash.
• The frequency responses at each stage were
  recorded and analyzed.

• Results
• The QCM-SHH biosensors were capable of detecting
  SHH in undiluted conditioned medium in a
  repeatable manner (n5).  In the undiluted
  samples, the average resonant frequency shift was
  80 Hz for a conditioned medium sample of 70
  microliters.

Injection Start
Injection End
Freq. Shift

• ELISA assays for purified SHH demonstrated a
  detection sensitivity to less than 0.2 ng of
  purified SHH (blue curve). When SHH was added to
  conditioned medium detection was reduced, but
  ELISA assays detected less than 0.3 ng. Of note,
  when the LNCAP conditioned medium used in the BAW
  studies was assayed, the SHH was noted to be
  present at less than 0.2 ng/70 microliter sample,
  with a detected total concentration of less than
  6 ng SHH for the conditioned medium used in the
  sensor studies.

Principle of a piezoelectric acoustic wave
immunosensor
Governing equations
Surface perturbation
Target input
Molecular recognition
Output quantity (electrical)

• The QCM sensor detection (80-Hz change to detect
  less than 200 picograms SHH protein) demonstrates
  the sensitivity of the system. While
  non-specific binding could compound the results,
  a control FITC antibody was used to subtract
  background binding. Subsequent array studies
  using multiple different SHH antibodies will
  allow us to confirm the specificity of the
  frequency shifts and calculate more accurate
  sensitivities.

Sauerbrey equation 1
Bio-molecules, chemicals
Immuno-reaction (binding events)
Frequency, phase, V, I, etc.
Surface property changes (mass, stiffness, etc.)
Hunt equation 2
Non-specific target
AT-cut quartz plate (0.17mm thick)
?mass density, µstiffness, f frequency, ?m
mass loading, hf film thickness, A sensing
area, µq and ?q shear stiffness and density of
the quartz crystal
Specific target (antigen)
13.67 mm
(Not drawn to scale)
Gold electrode (100 nm thick)
5.1 mm
Antibody

• The BAW Array design and fabrication was
  successful and the array is pictured below. The
  average frequency shift for SHH sensors is 1.27
  MHz as compared to a shift of 0.48 MHz of the
  reference sensor. The results were collected from
  56 separate devices. The SHH-based change of 0.8
  MHz is an approximately 10,000 fold increase in
  sensor sensitivity.

• The wavelength (?) is fixed by lithography
  process.

Cross-linker
?? ?µ
? f
Quartz Crystal Microbalance
Device surface
Surface property change
Frequency shift

• Methodology
• QCM Testing (Proof of concept)
• A reference sensor was coated with Anti-FITC
  antibodies (5 µl/ml) using a Self-Assembled
  Monolayer (SAM) as a cross-linking mechanism. The
  target sensor was coated with anti-SHH (5 µl/ml)
  antibodies.
• The sensors were driven at their resonant
  frequency while LNCap conditioned medium was
  injected at a flow rate of 0.2 ml/min. The
  injection was stopped and the volume of 70 µl of
  conditioned medium was static in the flow chamber
  incident with the two sensors. A buffer wash
  provided the washing of unbound non-specific
  particles from the surface.
• The transient resonant responses were collected
  and analyzed
• ELISA Testing
• ELISA plates were coated with anti-SHH antibody
  (clone N-19) at 11,000 in PBS. Either purified
  SHH diluted in PBS or conditioned medium was
  incubated, and subsequently the plated were
  washed. A second anti-SHH antibody (clone H-160)
  was added at 120 and incubated and subsequently
  washed. A final anti-mouse HRP antibody was
  added and the plate developed after color
  activation. Plates were subsequently read for
  absorbance (ABS).

Freq. Shift

• The Anti-SHH devices showed an average frequency
  shift of 1.27 MHz (n32) or approx. 0.3 of the
  average resonant frequency of 405 MHz
• The Anti-FITC devices showed an avg. frequency
  shift of 0.48 MHz (n24) which is approx. 0.12
  of the avg. resonant frequency.

• Conclusions
• Here we demonstrate the first use of a bulk
  acoustic wave sensor chip for the detection of a
  cancer biomarker in complex media.  We have
  created an easily fabricated, inexpensive BAW
  sensor chip whose production cost is a few cents
  per chip.  These chips can be rapidly and
  efficiently conjugated to antibodies and used for
  the detection of circulating antigens in complex
  solutions such as blood or serum.  The array
  formation of our devices allows for immediate and
  efficient repeatability of the test as well as
  the possibility for statistical analysis of the
  results. The ability to detect low levels of
  sonic hedgehog in serum, whether combined with or
  independent of serum PSA levels, could be used to
  diagnose aggressive prostate cancers or monitor
  response to treatment.