Journal of Biomaterials and Nanobiotechnology, 2011, 2, 134-143
doi:10.4236/jbnb.2011.22017 Published Online April 2011 (http://www.scirp.org/journal/jbnb)
Microcantilever-Based Nanomechanical Studies of
the Orphan Nuclear Receptor Pregnane X
Receptor-Ligand Interactions
Kasey L. Hill, Pampa Dutta, Zhou Long, Michael J. Sepaniak
Department of Chemistry, University of Tennessee, Knoxville, USA.
Received September 10th, 2010; revised January 1st, 2010; accepted January 5th, 2011.
ABSTRACT
Human pregnane X receptor (PXR) is of vital importance in pharmaceutical and exogenous compound metabolism
within the body. This prov ides strong motivation for investiga ting this orphan recep tor’s activation by various pharma-
ceuticals, xenobiotics, and endocrine disrupting chemicals (EDCs). A nanomechanical transducer is developed to study
xenobiotic and EDC interaction s with the bioreceptor PXR’s ligand binding doma in (LBD). The combina tion of immo-
bilized LBD PXR with a nanostructured microcantilever (MC) platform allows for the sensitive, label-free study of li-
gand interaction with the receptor. PXR shows real-time, reversible responses when exposed to a specific pharmaceu-
tical, EDCs, and xenobiotic ligands. Three EDCs binding interactions are tested, which include phthalic acid, nonyl-
phenol, and bisphenol A, with PXR. PXR LBD was exposed to rifampicin, a potent PXR activator, with various injection
and recovery times to study their interaction. A two protein array of PXR and estrogen receptor
 (ER-  ) directly
compares the nanomechanical responses of these receptors with rifampicin on a single platform.
Keywords: Microcantilevers, Nanobiosensing, Pregnane X Receptor, Estrogen Receptor, Endocrine Disrup ting
Chemicals, Bioarray
1. Introduction
Humans are exposed to harmful chemicals and contami-
nants each day. It is essential that these toxins are re-
moved or detoxed from the body. These foreign com-
pounds or xenobiotics, which include environmental
toxins, endogenous hormones, steroids, pharmaceuticals,
and dietary supplements, trigger a line of defense me-
chanisms within the body. The family of cytochrome
P450 enzymes are the main xenobiotic defenders for
mammals. These enzymes include four families of
Cytochrome P450 monooxygenases: CYP1, CYP2,
CYP3, and CYP4 [1]. Cytochrome P4503A4 (CYP3A4),
a critical member of our defense system, makes the re-
moval of many unwanted xenobiotics possible [2].
CYP3A4 and its isoforms are highly involved in phar-
maceutical metabolism, playing a significant role in me-
tabolizing approximately 50% of drugs used today [3,4].
When xenobiotic ligands bind to a specific nuclear hor-
mone receptor, the interaction transcriptionally activates
CYP3A4 [5]. In 1998, a novel orphan human nuclear
receptor was identified and termed pregnane X receptor
(PXR), steroid and xenobiotic receptor (SXR), pregnane
activated receptor (PAR), and NR1I2 [6-9]. Herein, we
choose to use the term PXR.
PXR is activated by a broad array of structurally di-
verse xenobiotics [10]. This wide activation range is pos-
sible is due to a unique ligand binding domain (LBD) or
pocket, which can expand to fit a variety of sized ligands.
PXR’s LBD has two  strands that are not present in
other nuclear receptors and that allow its expansion [11].
This flexible, hydrophobic LBD allows PXR to be acti-
vated by a diverse range of synthetic and naturally occur-
ring chemicals making it an interesting candidate to serve
as a xenobiotic sensor [12].
Importantly, PXR is activated by numerous pharma-
ceuticals with diverse properties, functions, and struc-
tures and controls the expression of genes that are vital to
pharmaceutical metaolism [13,14]. PXR activation medi-
ates transcription of CYP enzymes, which are considered
drug metabolizing enzymes [1,12]. Since PXR activation
allows for regulation and expression of CYP3A this in-
teraction is critical to drug metabolism [2]. Pharmaceuti-
cal metabolism and interaction is vital to monitor and

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 135
Receptor-Ligand Interactions
prevent drug-drug interactions. Drug-drug interactions
can occur when co-administered drugs alter the efficacy
of one another. This usually occurs when one drug in-
creases or decreases the metabolism of another [3,14].
Contaminants could modify drug concentrations in vivo
and skew the prescribed therapeutic dose.
High throughput, sensitive screening and detection of
xenobiotics are critical to determine harmful toxins and
possible drug-drug interactions. A highly sensitive, in-
expensive, and rapid sensor utilizing the nuclear receptor
PXR has been developed and is demonstrated within.
Recently, biosensors based on microcantilevers (MCs)
have been used to detect and screen for many harmful
environmental contaminants using estrogen and thyroid
receptor proteins [16,17]. The high sensitivity and wide-
spread availability of inexpensive MCs has generated
intense interest in their use as chemical and biological
sensors [18-27]. Additionally, MCs can be used with
on-chip circuitry and in microcantilever arrays for high
throughput and simultaneous differential assays and bio-
affinity studies with a very small transducer foo tprint th at
potentially could be employed in the field.
A MC suitable for biosensing is modified on one side
with a nanostructured layer and a receptor phase that has
some degree of affinity for the analyte. By exploiting
PXR’s affinity for a diverse range of ligands, we are able
to screen for xenobiotics quickly and without extensive,
time-consuming labeling techniques and cell preparation
[10,14,15,28,29]. Specific interactions of the target ana-
lytes with the receptor can cause an apparent surface
stress and nanomechanical bending that may be conven-
iently monitored based on the beam bending technique
commonly used in atomic force microscopy. The static
bending (tip deflection, zmax) of the MC varies in selec-
tivity and sensitivity due to preferential binding of ana-
lyte molecules on the functionalized, active MC surface
and is governed by Sto ney ’s equation [30]
2
max 2
3(1 )
Et
lv
z

 (1)
where v and E are, respectively, the Poisson ratio and
Young’s modulus for the cantilever, t is the thickness of
the MC, l is the cantilever effective length, and  is
analyte-induced differential surface stress (active side -
passive side).
We demonstrate that detection and screening for phar-
maceuticals and endocrine disrupting chemicals (EDCs)
can be accomplished with functionalized MCs. These
sensors provide real-time measurements of surface stress
changes resulting from ligand interaction with immobi-
lized proteins in the low-to-sub-nanomolar range [20].
Sensitivity is critical in biosensors due to the ultra-trace
concentrations of many xenobiotics that can impact bio-
logical systems. Nanostructured MC biosensors allow
detection without labels and in a detection range that is
applicable for real biological systems [31]. PXR immobi-
lized on a nanostructured MC surface provides sensitive
and reversible detection of various pharmaceuticals and
environmental contaminants or EDCs. To our knowledge,
this is the first time the orphan nuclear receptor PXR has
been immobilized on a MC surface. Due to PXR’s u niq ue
nature we saw interesting results in our concentration
studies with rifampicin as well as surprising responses
when utilizing differently tagged PXR-receptors.
2. Materials and Methods
2.1. Reagents
Experiments were performed using commercially avail-
able silicon arrays of MCs having dimensions 400µm
length, 100µm wid th, and approximately 1µm thick (Mi-
kro Masch Co., Sunnyvale, CA). Chromium, gold, and
silver metals deposited on the MCs were obtained from
Kurt J. Lesker, Gatewest, and Alfa Aesar Co., respec-
tively, at 99.9% purity. 2-aminoethanethiolhydrochloride
(AET), glutaraldehyde (GA), the salts employed for the
preparation of buffer solutions, and all other reagents
were purchased from Sigma-Aldrich Chemical Co.(St.
Louis, MO) or Fisher Scientific at highest available pu-
rity and used as received. The test ligands rifampicin,
pregnenolone-16-carbonitrile (PCN), 3-methyl-cholanthr-
ene (3-MC), phthalic acid, nonylphenol and bisphenol A
were also obtained from Sigma-Aldrich. Glutathione-
s-transferase (GST)-tagged human PXR, alexa fluor 633
anti-IgG, and estrogen receptor  (ER-) were purchased
from Invitrogen (Carlsbad, California). 6-histidine (6-
HIS)-tagged human PXR was generously provided by
Astra Zeneca. Ovalbumin and FITC-anti-IgG was pur-
chased from Sigma-Aldrich. Water used to prepare solu-
tions was obtained from a Branstead E-pure water filtra-
tion system.
2.2. Cantilever Modification
The process of preparing and creating nanostructured
surfaces on MCs is described in detail elsewhere [32].
The MCs were placed into a physical vapor
deposition chamber (Cooke Vacuum Products, Model
CVE 30 1, S outh Norwalk, CT) to be coated on on e side
with the appropriate metallic films using thermal
deposition. To create a nanostructured MC, a thin film
(~ 5 n m) of c hr omiu m was app lied to th e surface to act
as an adhesion layer followed by a thin film of gold
(~1 5 nm). N ex t, a f il m c ons i st in g of g ol d an d si l ve r wa s
co-depo-sited. Subsequently, the silver was chemically
C
opyright © 2010 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 136 Receptor-Ligand Interactions
removed via oxidation from the film (“dealloying”)
using an aqueous solution of 5 mg/mL HAuCl4 leaving
a gold surface with nanosized, colloid-like features.
The thick-ness of the dealloyed gold layer was ~100 or
~150 nm in these studies.
Nanostructured MCs were chemically modified for
protein immobilization by the process detailed elsewhere
[17]. MCs were immersed in 1 mM aqueous solution of
AET for on e hour producing a self -assembled monolayer
on the cantilever surface. Following thorough rinsing in
deionized water, the amino groups were derivatized with
the cross linker by immersing the cantilever in a 1% (w/v)
solution of GA in 10 mM phosphate buffered saline
(PBS), pH 8.0 for three hours [33,34]. Subsequently,
immobilization of both the PXR LBD nuclear receptor
and the ovalbumin was achieved in random orientation
by dipping the functionalized cantilevers into 100 mg/L
solutions of protein in 10 mM PBS, pH 7.0 for four hours
at 4℃. Although we used an array of MCs, in this study
we chemically treated all the cantilevers the same where
both PXR and ovalbumin were separately immobilized
on the functionalized surfaces of different cantilevers
from separate arrays and a single randomly chosen MC
within an array response was recorded. For the two pro-
tein array, the immobilization of both the PXR and the
ER- was achieved in random orientation utilizing a ca-
pillary coating method described in detail elsewhere [35].
Each capillary contained 100 mg/L solution of receptor
in 10 mM PBS, pH 7.0, which was placed over each lev-
er for 1 hour at room temperature.
2.3. Instrumentation
The MC deflection measurements were carried out using
the optical beam-deflection technique described else-
where [16,17]. The apparatus similar to that depicted in
Figure 1 included a 5 mW 632 nm diode laser (Coherent
Laser Corp., Auburn, CA), a spatial filtering and focus-
ing system, and an in-house built position sensitive opti-
cal detector. The output of the detector was displayed
and recorded using a SRS 850 DSP lock-in amplifier as a
multichannel digital recorder (Stanford Research Sys-
tems, Sunnyvale, CA). The signal output is recorded as
volts (approximately 1 nm zmax per mV output). Data was
collected at 1 Hz and then a moving averaging algorithm
covering 180 data points was used to generate the figures
presented herein. This smoothing did not alter the shape
of the true response curves [19].
The cantilever system was mounted inside a ~5µL vo-
lume flow cell built in house previously described else-
where [16]. A Watec CCD camera (Edmund Indust- rial
Optics, Barrington, NJ) was used to image the MC chip
in the flow cell and facilitate in aligning the focused laser
Figure 1. Schematic of laser array setup with a single PSD.
Chip array within flow cell is shown. The lower left shows
bending of a single MC due to adsorption-induced differen-
tial stress. Note that deformation of the MC is characterized
by the radius of curvature, R, and tip deflection, z, ac-
cording to Equation (1).
beam to reflect off the cantilever tip. Analyte solutions
were delivered to the flow cell via a system of vessels
connected to three-way valves allowing for switching
between diffe r e nt s ol uti o ns. The gravity-dri ven fl ow was
generally adjusted to 100 L/minute by adjusting vessel
height.
Many of the pharmaceuticals and EDCs are sparingly
soluble in water. Thus, 10 mM stock solutions of all
EDCs and some pharmaceuticals were prepared in pure
methanol and then diluted with 10 mM PBS, pH 7.0 to
make the desired concentration of each analyte [ Ca u t i o n :
because of their potential harmful effects, care must be
taken in the handling and disposing of EDC and
pharmaceutical solutions]. PCN pharmaceutical stock
solution was prepar ed in acetone then dil uted with PBS.
PBS was also used as a background solution. MCs
mounted in the flow cell were initially allowed to
equilibrate in PBS until the signal was stable. For our
purposes, tensile and compressive responses involve
contraction and expansion of the active MC surface,
respectively.
MC deflection measurements of single protein func-
tionalized chips using a single diode laser or multiple
protein arrays can be monitored using an array of vertical
cavity surface emitting lasers (VCSELs) which are fo-
cused onto the tip of each MC (Figure 1). The reflected
beam is captured and monitored by a single position sen-
sitive detector as depicted in the figure. Although in the
work reported herein employed a single laser setup, we
often use a VCSEL setup [36-38].
3. Results and Discussion
Our studies focus on developing MC systems utilizing
the nuclear receptor PXR as the immobilized bio receptor
C
opyright © 2011 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 137
Receptor-Ligand Interactions
phase. We demonstrate that our immobilization process
does not appreciably denature the PXR. The magnitude
based response order for PXR activators is maintained
when compared to well-established assays. With these
PXR active analytes, PXR selectivity is demonstrated
when compared to a similarly sized immobilized protein.
The assay magnitude based response order is maintained
when PXR is exposed to EDCs. Reproducibility and sen-
sitivity is illustrated using a potent PXR activ ator, rifam-
picin. A two protein array is developed to compare PXR
and ER- interaction with rifampicin.
Sensitivity enhancement in biosensing is key due to
the small concentrations of ligands that can activate nu-
clear receptors. By nanostructuring the active MC sur-
face, we are able to substantially improve the sensitivity,
which has in many cases surpassed the increase in sur-
face area [18,32,39]. The initial “dealloying” of the MC
surface provides a greater surface area for nuclear recep-
tor immobilization, which leads to an enhancement in
sensitivity. Our developing PXR MC biosensor, in sup-
port of fundamental nuclear receptor studies, may prove
to be a quick, less expensive method for EDC screening
and possible drug-drug interaction predictions.
3.1. Detection of Known PXR Activators and
EDCs Using PXR Modified MCs
Figure 2 compares the responses of a PXR functional-
ized MC when exposed to three analytes in PBS. This
figure illustrates that 1.0 M rifampicin is the most po-
tent PXR activator in this group. Our experiments in the
detection of rifampicin using the nuclear receptor, PXR,
showed good measurement reproducibility in the same
day tested via three replicate consecutive injections of
1.0 M solution (see Figure 2 inset). Coefficients of
variation (CVs) for intra-day measurements using a given
system of MC and molecular-recognition phase are gen-
erally 10% or better [16,18-20,23]. Although for MCs
that are nanostructured and biofunctionalized on different
days the CVs can be considerably larger therefore cali-
bration is necessary for each individual system. The rela-
tively slow response kinetics is comparable to prior bio-
sensor MC experiments [16,20,23]. This provides evi-
dence that the ligands interact with the LBD causing rel-
atively slow conformational changes in the immobilized
PXR, which translates into a large apparent surface stress
on the cantilever. The high binding affinity of human
PXR for rifampicin followed by PCN has been observed
by other researchers [7,9,15,28,40,41]. Although, 3-MC
is predicted to have a low binding energy to the LBD of
human PXR [42], it has been shown to be activated by
members of the cytochrome P450 family of enzymes,
which include activating CYP1A1/2 enzymes [42-45].
Figure 2. Comparison of nanomechanical responses of PXR
functionalized MC on exposure to 1.0 M of rifampicin,
PCN, and 3-MC in PBS. Inset shows data for triplicate se-
quential injections of 1.0 M rifampicin wi th a CV of 6.32.
The magnitude response order of rifampicin > PCN >
3-MC is what we predicted from the literature. In our
previous studies, reversible responses like those seen in
Figure 2 are also observed for other bioreceptor func-
tionalized dealloyed surfaces [20,23].
Figure 3 compares the response of specific protein
(PXR) functionalized MC to nonspecific protein (oval-
bumin) functionalized MC (blank) on exposure to the
same concentration (1.0 M) of rifampicin in 10 mM
PBS, pH 7.0. A large compressive response was ob-
served due to the binding of rifampicin with a MC modi-
fied with PXR receptor whereas no response was ob-
served when the same analyte was exposed to the non-
binding protein (ovalbumin) immobilized MC. It is im-
portant to note that our system does not show a nonspe-
cific blank response. This indicates that the surface im-
mobilization procedure does not significantly alter the
PXR’s LBD function and that the observed responses
represent specific interactions. However, it can not be
assumed that the surface immobilized receptors will re-
tain the same ligand binding affinity as observed in free
form. Figure 3 further compares the specific and non-
specific responses to 1.0 M rifampicin, PCN, and 3-MC
to immobilized PXR and ovalb umin (blank).
EDC exposure can be adverse even at very small con-
centrations [46]. EDCs can alter or inhibit the function of
the endocrine system by binding to estrogen receptors,
which are part of the nuclear receptor superfamily [47-
49]. Studies have shown that the nuclear receptor PXR
binds to certain EDCs with different affinities [50,51].
Figure 4 shows the nanomechanical response magnitude
of PXR functionalized MCs on exposure to 1.0 M
EDCs, phthalic acid, nonylphenol, and bisphenol A (in
10 mM PBS, pH 7.0). The response magnitude order of
C
opyright © 2010 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 138 Receptor-Ligand Interactions
Figure 3. Table compares the nanomechanical responses of
PXR and ovalbumin functionalized MCs exposed to 1.0 M
rifampicin, PCN, and 3-MC in PBS. Representative traces
of ovalbumin and PXR functionalized MCs on exposure to
1.0 M rifampicin are shown.
PXR with these three EDCs is similar to prior work with
phthalic acid > nonylphenol and with relatively no re-
sponse or a very low response to bisphenol A [50]. After
MC exposure to EDCs, we injected 1.0 M rifampicin, a
potent PXR activator, for comparison. As predicted, the
Figure 4. Comparison of response magnitudes in 8.5 min-
utes of PXR functionalized MC on exposure to 1.0 M ana-
lytes in PBS, which include three EDCs. The inset shows
representative responses of PXR functionalized MC ex-
posed to 1.0 M nonylphenol and rifampicin.
rifampicin caused a large compressive response when
compared to the EDCs (see inset).
3.2. Calibration and Sequential Exposure Studies
In Figure 5 the kinetic response of the PXR functional-
ized cantilever at 7 minutes of expo sure is plotted ag ainst
different concentrations of rifampicin in the range of 0.1
nM to 1 M where the response increases with increas-
ing concentration. The response magnitude does not in-
crease as significantly as predicted with increasing con-
centration at relatively high concentrations. This may be
due in part to the structural flexibility of the LBD. The
expansion of the LBD size to accommodate for rifam-
picin’s large size could have an effect on the interaction’s
translation to MC surface stress. Rifampicin is one of the
largest known ligands for PXR at 823 Da [14]. As illus-
trated in the figure, the response increased predictably,
but reached a saturation plateau by 100 nM. The inset in
Figure 5 shows a linear dynamic range for two orders of
magnitude (the first data point corresponds to the lowest
concentration in Figure 5 of 0. 1 nM ) i n c on cent rat i o n.
The renaturing of the protein after ligand exposure
may influence receptor calibration experiments, thus se-
quential ligand exposure was studied. Immobilized LBD
PXR was exposed to sequential injections of 1.0 M ri-
fampicin in 10 mM PBS, pH 7.0 with varying exposure
and recovery time. The injection times and recovery
times were approximately the same. As injection time
increased the response maximum increased as well,
which is likely a result of increased exposure time. This
longer interaction time may provide larger kinetic re-
sponses, which is the nanomechanical measured parame-
ter. All injection volumes and times were considerably
less than saturation and were reversible, although base-
line was not always reached with shorter recovery times.
Exposure and recovery times of LBD PXR were also
Figure 5. Responses to rifampicin at 6.7 minutes for PXR
are plotted over a concentration range from 0.1 nM to 1.0
M, respectively.
C
opyright © 2011 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 139
Receptor-Ligand Interactions
studied with two sequential three minute rifampicin in-
jections then a 60 minute recovery time (background
flow) then two more consecutive injections (three min-
utes and nine minutes). The first two injections had simi-
lar response maximums, 35.8 mV versus 33.4 mV, al-
though between injections baseline was not reached dur-
ing the short recovery window. After a 60 minute recov-
ery time or background flow, the three minute response
to rifampicin increased to 58.3 mV. This was subse-
quently followed by a nine minute rifampicin injection
where the response maximum almost doubled. This study
may indicate that recovery time, exposure time, and ana-
lyte volume play a role in kinetic response maximums for
the LBD PXR- rifampicin system.
3.3. Effect of Different Molecular Tags on
Measuring Surface Stress
In our current functionalization procedure, the protein is
immobilized on the MC surface with GA without appre-
ciable denaturing [52]. This is accomplished by the al-
dehyde groups in the GA binding to the lysine residues in
the receptor [53]. This makes the number and location of
the lysine residues that make up the protein or receptor
important to our random immobilization process. The
tags on the LBD PXR allows for the purification of the
receptor. The amino acids present in the tag and their
location could be beneficial or detrimental to the nano-
mechanical responses of PXR functionalized MCs.
GST-tagged human LBD PXR from Invitrogen has a
calculated molecular mass of approximately 65 kDa.
Human LBD PXR receptor is approximately 37 kDa,
which leads to the calculated molecular mass for the
GST-tag to be approximately 27 kDa. The GST-tag and
the LBD PXR have similar molecular masses in this case.
The 6-HIS-tag on the Astra Zeneca LBD PXR is ap-
proximately 2524 Da and contains no lysine residues.
Conversely, the GST-tag contains an estimated 21 lysine
residues, depending on which version of the GST was
used in the preparation. Thus, the 6-HIS-tagged LBD
PXR was not bound to the surface by the tag, but by the
nuclear receptor itself, where as the much larger GST-tag
could have been the surface bound component.
Figure 6 compares the responses of 6-HIS-tagged
LBD PXR and GST-tagged LBD PXR upon exposure to
1.0 M rifampicin in 10 mM PBS, pH 7.0. The
6-HIS-tagged LBD PXR gives a large compressive re-
sponse whereas the GST-tagged LBD PXR shows no
response. This indicates that the GST-tag may preferen-
tially bind to the surface leaving the PXR too far from
the surface to produce nanomechanical surface stress. In
prior work, random versus orientated antibody immobi-
lization methods were studied, which demonstrated that
Figure 6. Comparison of 6-HIS-tagged PXR LBD and GST-
tagged PXR LBD on exposure to 1.0 M rifampicin in PBS.
Inset shows a schematic depiction of 6-HIS tagged PXR and
GST tagged PXR immobilized on DA MC surface, which
highlights the spatial arrangement between the DA and the
PXR LBD depending on the tag.
orientated functionalization did not provide higher re-
sponses as expected. Orientated anti-IgG and IgG immo-
bilization was accomplished by using protein A as an
orientation linker. This linker may have caused the anti-
body-antigen interaction not to be as readily translated
into MC surface stress as the randomly orientated anti-
body [17]. As with the GST-tagged PXR case this may
not allow the conformational change upon ligand binding
to translate efficiently to measureable surface stress. The
very small, lysine deficient 6-HIS-tag requires the recep-
tor to be surface bound and could easily transfer confor-
mational change to the MC surface although only some
ligand binding sites may be accessible. This is illustrated
through a schematic in Figure 6. Although in some in-
stances the immobilized orientation of LBD PXR may
block the ligand binding pocket thereby hindering ligand
binding.
3.4. LBD PXR and ER-  Array Study
A single MC array functionalized with multiple bio-
molecular recognition phases will mimic a living system
and it’s interaction with contaminates and/or pharmaceu-
ticals. EDCs or pharmaceuticals can stimulate changes in
multiple receptor proteins. Therefore, having a MC array
platform functionalized with a multiple receptor protein
system (Figure 7) can provide information on total bio-
logical systems. MC arrays functionalized with different
receptor proteins can simultaneously probe these living
systems, which make them highly desirable analytical
tools. LBD PXR and ER- were differentially immobi-
lized onto separate levers of one MC chip creating an
array. Figure 7(a) shows a fluorescent microscope image
of immobilized fluorescently labeled antibodies on sepa-
C
opyright © 2010 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 140 Receptor-Ligand Interactions
rate levers prepared in the same manner as the LBD PXR
and ER- array. The receptor functionalized levers were
exposed to 1.0 M rifampicin and the bending response
was recorded. First, the LBD PXR functionalized lever
was monitored with a five minute rifampicin injection,
subsequently the ER-
 immobilized lever was monitored
in the same manner.
Figure 7(b) shows the response profiles of the LBD
PXR and the ER-
 when exposed to rifampicin in 10
mM PBS, pH 7.0. ER-
 demonstrating a larger response
than LBD PXR may be due to the structure of the recep-
tors that are present. Immobilized ER- contains both the
(a)
(b)
Figure 7. (a) Fluorescence microscopy images of differentially func-
tionalized MCs: (a.) and (b.) are the same lever set where lever 1 is
functionalized with Alexa Fluor 633 anti-IgG, lever 2 is not func-
tionalized with antibody, lever 3 is functionalized with FITC-
anti-IgG. (a.) FITC-anti-IgG on lever 3 is excited by 488 nm laser.
(b.) Alexa Fluor anti-IgG on lever 1 is excited by 633 nm laser. (b)
MC array of PXR and ER- exposed to 1.0 M rifampicin.
DNA binding domain and the LBD, whereas the immbi-
lized PXR only contains the LBD and the small 6-HIStag.
In previous studies with ER proteins, immobilization at
the DNA binding sites which change configuration in
response to conformational LBD changes may have been
a factor in producing large surface stress responses [16].
In this case the ER- interacts to some degree with ri-
fampicin causing configuration change of the LBD,
which may be translated into a conformational chang e in
the DNA binding domain. This conformational change is
transferred to surface stress on the MC. Another possibil-
ity for smaller PXR responses may be that it does not
have time to fully unfold to accommodate for the large
size of rifampicin, so the interaction between the receptor
and analyte is limited, which hinders the response mag-
nitude. In any event a synergistic system of receptors is
demonstrated on a single MC array and there is evidence
of endocrine disrupting behavior of rifampicin.
4. Conclusions
In summary, a sensitive, selective b iosensor for the study
of PXR activating chemicals has been developed using
nanostructur ed MCs. Our r esu lts indicate that the in terac-
tion of LBD PXR with different ligands produced dif-
ferent cantilever responses showing the maximum re-
sponse for the antibiotic rifampicin. PXR functionalized
MCs showed responses to rifampicin in concentrations
down to 0.1 nM. Dissimilarity in tag size and residue
makeup also indicates differences in receptor-ligand
binding translation into MC surface stress. The nuclear
receptor array provides information about the interaction
of rifampicin with both human PXR and ER-.
5. Acknowledgements
This research has been supported in part by a grant from
the U.S. Environmental Protection Agency’s Science to
Achieve Results (STAR) program (EPA-83274001). The
authors extend their gratitude to Dr. Melanie Eldridge of
the Center of Biotechnology at the University of Ten-
nessee, Knoxville for her useful discussions and to Dr.
Roderic Cole for his assistance in obtaining PXR.
REFERENCES
[1] D. R. Nelson, L. Koymans, T. Kamataki, J. J. Stegeman, R.
Feyereisen, D. J. Waxman, M. R. Waterman, O. Gotoh, M. J.
Coon, R. W. Estabrook, I. C. Gunsalus and D. W. Nebert, “P450
Superfamily: Update on New Sequences, Gene Mapping, Ac-
cession Numbers and Nomenclature,” Pharmacogenetics, Vol. 6,
1996, pp. 1-42. doi:10.1097/00008571-199602000-00002
[2] S. A. Kliewer, J. M. Lehmann an d T. M. Wil lson, “Orphan Nu-
clear Receptors: Shifting Endocrinology into Reverse,” Science,
Vol. 284, 1999, pp. 757-760. doi:10.1126/scien ce.284.5415.757
C
opyright © 2011 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 141
Receptor-Ligand Interactions
[3] E. L. Michalets, “Update: Clinically Significant Cyto-Chrome
P-450 Drug Interactions,” Pharmacotherapy, Vol. 18, 1998,
pp. 84-112.
[4] P. R. Ortiz de Montellano, “Cytochrome P450: Structure,
Mechanism and Biochemistry,” Plenum Publishers, New York,
2005, pp. 377-425.
[5] J. Zhang, P. Kuehl, E. D. Green, J. W. Touchman, P. B. Wat-
kins, A. Daly, S. D. Hall, P. Maurel, M. Relling, C. Brimer, K.
Yasuda, S. A. Wrighton, M. Hancock , R . B . K i m, S . Str o m , K .
Thummel, C. G. Russell, J. R. Hudson, Jr., E. G. Schuetz and
M. S. Boguski, “The Human Pregnane X Receptor: Genomic
Structure and Identification and Functional Characterization of
Natural Allelic Variants,” Pharmacogenetics, Vol. 11, 2001,
pp. 555-572. doi:10.1097/00008571-200110000-00003
[6] S. A. Kliewer, J. T. Moore, L. Wade, J. L. Staudinger, M. A.
Watson, S. A. Jones, D. D. McKee, B. B. Oliver, T. M. Will-
son, R. H. Zetterst röm, T. Perlmann and J. M. Lehmann, “An
Orphan Nuclear Receptor Activated by Pregnanes Defines a
Novel Steroid Signaling Pathway,” Cell, Vol. 92, 1998, pp.
73-82. doi:10.1016/S0092-8674(00)80900-9
[7] B. Blumberg, W. Sabbagh, Jr., H. Juguilon, J. Bolado, Jr., C.
M. van Meter, E. S. Ong and R. M. Evans, “SXR, a Novel
Steroid and Xenobiotic-Sensing Nuclear Receptor ,” Genes and
Development, Vol. 12, 1998, pp . 3195-3205.
doi:10.1101/gad.12.20.3195
[8] G. Bertilsson, J. Heidrich, K. Svensson, M. Åsman, L. Jende-
berg, M. Sydow-Bäckman, R. Ohlsson, H. Postlind, P. Blom-
quist and A. Berkenstam, “Identification of a Human Nuclear
Receptor Defines a New Signaling Pathway for Cyp3a Induc-
tion,” Proceedings of the National Academy of Sciences, Vol.
95, 1998, pp. 12208-12213. doi:10.1073/pnas.95.21.1 2208
[9] R . E. Watkins, G. B. Wisel y, L. B. Moore, J. L. Colli ns, M. H.
Lambert, S. P. Williams, T. M. Willson, S. A. Kliewer and M.
R. Redinbo, “The Human Nuclear Xenobiotic Receptor PXR:
Structural Determinants of Directed Promiscuity,” Science,
Vol. 292, 2001, pp. 2329-2333 . doi:10.1126/science.1060762
[10] L. B. Moore, J. M. Maglic h, D. D. McKee, B. Wisely, T. M.
Willson, S. A. Kliewer, M. H. Lambert and J. T. Moore,
“Pregnane X Receptor (PXR), Constitutive Androstane Re-
ceptor (CAR) and Benzoate X Receptor (BXR) Define Three
Pharmacologically Distinct Classes of Nuclear Receptors,”
Molecular Endocrinology, Vol. 16, 2002, pp. 977-986 .
doi:10.1210/me.16.5.977
[11] C. Zhou, S. Verma and B. Blumberg, “The Steroid and Xeno-
biotic Receptor (Sxr), Beyond Xenobiotic Metabolism,” Nu-
clear Receptor Signaling, Vol. 7, 2 009, pp. 1-21.
[12] S. A. Kliewer, B. Goodwin and T. M. Willson, “The Nuclear
Pregnane X Receptor: A Key Regulator of Xenobiotic Me-
tabolism,” Endocrine Reviews, Vol. 23, 2002, pp. 687-702.
doi:10.1210/er.2001-0038
[13] J. E. Chrencik, J. Orans, L. B. Moore, Y. Xue, L. Peng, J. L.
Collins, G. B. Wisely, M. H . Lambert, S. A. Kliewer and M. R.
Redinbo, “Structural Disorder in the Complex of Human
Pregnane X Receptor and the Macrolide Antibiotic Rifam-
picin,” Molecular Endocrinology, Vol. 19, 2005, pp. 1125-
1134. doi:10.1210 /me.2004-0346
[14] J. T. Moore and S. A. Kliewer, ‘Use of the Nuclear Recaptor
Pxr to Predict Drug Interactions,” Toxicology, Vol. 153, 2000,
pp. 1-10. doi:10.1016/S0300-483X(00)00300-0
[15] P. Dutta, K. Hill, P. G. Datskos and M. J. Sepaniak, “Devel-
opment of a Nanomechanical Biosensor for Analysis of Endo-
crine Disrupting Chemicals,” Lab Chip, Vol. 7, 2007, pp.
1184-1191. doi:1 0.1039/b704723a
[16] K. Hill, P. Dutta, A. Zareba, M. L. Eldridge and M. J. Sepaniak,
“Morphological and Chemical Optimization of Microcantile-
ver Surfaces for Thyroid System Sensing and Beyond,” Ana-
lytica Chimica Ac ta, Vol. 625, 2008, pp . 55-62.
doi:10.1016/j.aca.2008.07.005
[17] C. A. Tipple, N. V. Lavrik, M. Culha, J. Headrick, P. Datskos
and M. J. Sepaniak, “Nanostructured Microcantilevers with
Functionalized Cyclodextrin Receptor Phases: Self Assembled
Monolayers and Vapor Deposited Films,” Analytical Chemis-
try, Vol. 74, 2002, pp. 3118-3126. doi:10.1021 /ac020074o
[18] P. Dutta, P. J. Chapman, P. G. Datskos and M. J. Sepaniak,
“Characterization of Ligand-Functionalized Microcantilevers
for Metal Ion Sensing,” Analytical Chemistry, Vol. 77, 2005,
pp. 6601-6608. doi:10.1021/ac051082i
[19] P. Dutta, J. Sanseverino, P. G. Datskos and M. J. Sepaniak,
“Detection of Cytokine Using a Microcantilever Biosensor,”
NanoBiotechnology, 1, 2005, pp . 237-244.
doi:10.1385/NBT:1:3:237
[20] R. Berger, E. Delamarche, H. P. Lang, C. Gerber, J. K. Gim-
zewski, E. Meyer and H. J. Guntherodt, “Surface Stress in the
Self-Assembly of Alkanethiols on Gold,” Scie nc e, Vol. 276,
1997, pp. 2021-2024. doi:10.1126/science.276.5321 .2021
[21] T. Thundat, E. A. Wachter, S. L. Sharp and R. J. Warmack,
“Detection of Mercury Vapor Using Resonating Microcanti-
levers,” Applied Physics Letters, Vol. 66 , 1995, pp. 1695-1697.
doi:10.1063/1.113896
[22] P. Dutta, C. A. Tipple, N. V. Lavrik, P. G. Datskos, O. Hof-
stetter and M. J. Sepaniak, “Enantioselective Sensors Based on
Antibody Mediated Nanomechanics,” Analytical Chemistry,
Vol. 75, 2003, pp. 2342-2348. doi:1 0.1021/ac034031z
[23] G. Wu, R. H. Datar, K. M. Hansen, T. Thundat, R. J. Cote and
A. Majumdar, “Bioassay of Prostate-Specific Antigen (Psa)
Using Microcantilevers,” National Biotechnology, Vol. 19,
2001, pp. 856-860. doi:10.1038/nbt0901-856
[24] C. Grogan, R. Raiteri, G. M. O’Connor, T. J. Glynn, V. Cun-
ningham, M. Kane, M. Charlton and D. Leech, “Characteriza-
tion of an Antibody Coated Microcantilever as a Potential
Immuno-Based Biosensor,” Biosensor Bioelectronics, Vol. 17,
2002, pp. 201-207. doi:10.1016/S0956-5663(01)00276-7
[25] J. Fritz, M. K. Baller, H. P. L ang, H. Rothuizen, P. Vettiger, E.
Meyer, H. Guntherodt, C. Gerber and J. K. Gimzewski,
“Translating Biomolecular Recognition into Nanomechanics,”
Science, Vol. 288, 2000, pp. 316-31 8.
doi:10.1126/science.288.5464.316
[26] G. Shekhawat, S. H. Tark and V. P. Dravid, “MOS-
FET-Embedded Microcantilevers for Measuring Deflection in
Biomolecular Sensors,” Science, Vol. 311, 2006, pp. 1592-
1595. doi:10.1126/science.1122588
C
opyright © 2010 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X 142 Receptor-Ligand Interactions
[27] S. A. Jones, L. B. Moore, J. L. Shenk, G. B. Wisely, G. A.
Hamilton, .D. D. McKee, N. C. O. Tomkinson, E. L. LeCluyse,
M. H. Lambert, T. M. Willson, S. A. Kliewer and J. T. Moore,
“The Pregnane X Receptor: A Promis-Cuous Xenobiotic Re-
ceptor That Has Diverged during Evolution,” Molecular En-
docrinology, Vol. 14, 2000, pp. 27-3 9. doi:10.1210/me.14.1.27
[28] S. J. Shukla, D. Nguyen, R. MacArthur, A. Simeonov, W. J.
Frazee, T. M. Hallis, B. D. Marks, U. Singh, H. C. Eliason, J.
Printen, C. P. Austin, J. Inglese and D. S. Auld, “Identification
of Pregnane X Receptor Ligands Using Time-Resolved Fluo-
rescence Resonance Energy Transfer and Quantitative High-
Throughput Screening,” Assay and Drug Development Tech-
nologies, Vol. 7, 2009, pp. 143-169.
doi:10.1089/adt.2009.193
[29] G. G. Stoney, “The Tension of Metallic Films Deposited by
Electrolysis,” Proceeding of the Royal Society of London, Se-
rial A, Vol. 82, 1909, pp. 172-177 .
[30] N. P. Mahalik, “Micromanufacturing and Nanotechnology,”
Springer, Berlin, 2006, pp. 299-321.
[31] N. V. Lavrik, C. A. Tipple, M. J. Sepaniak and P. G. Datskos,
“Enhanced Chemi-M echanical Tran sduction at Nanos tructured
Interfaces,” Chemical Physics Letters, Vol. 336, 2001, pp.
371-376. doi:10.1016/S0009-2614(01)00155-5
[32] J. R. Premkumar, O. Lev, R. S. Marks, B. Polyak, R. Rosen
and S. Belkin, “Antibody-Based Immobilization of Biolumi-
nescent Bacterial Sensor Cells,” Talanta, Vol. 55, 2001, pp.
1029-1038. doi:10.1016/S0039-9140(01)00533-1
[33] S. W. Park, Y. I. Kim, K. H. C hung, S. I. Hong and S. W. Kim,
“Covalent Immobilization of Gl-7-Aca Acylase on Silica Gel
Through Silanization,” Reactive Functional Polymers, Vol. 51,
2002, pp. 79-92. doi:10.1016/S1381-5148(02)00028-7
[34] P. J . C ha pm an , Z. Long, P. G. Datskos, R. Archib ald and M. J.
Sepaniak, “Ligand-Functionalized Microcantilever Arrays for
Metal Ion Sensing,” Analytical Chemistry, Vol. 79, 2007, pp.
7062-7068. doi:10.1021/ac070754x
[35] L. R. Senesac, P. Dutta, P. G. Datskos and M. J. Sepaniak,
“Analyte Identification Using Differentially-Functionalized
Microcantilever Arrays and Artificial Neural Nets,” Analytical
Chemistry Acta, Vol. 558 , 2006, pp. 94-101.
doi:10.1016/j.aca.2005.11.024
[36] P. J. Chapman, Z. Long, P.G. Datskos, R. Archbald and M. J.
Sepaniak, “Ligand-Functionalized Microcantilever Arrays for
Metal Ion Sensing,” Analytical Chemistry, Vol. 79, 2007, pp.
7062-7068. doi:10.1021/ac070754x
[37] Z. Long, J. Story, S. Lewis and M. J. Sepaniak “Landfill Si-
loxane Gas Sensing Using Differentiating, Responsive Phase
Coated Microcantilev er Arrays” An alytical Chemistry, Vol. 81,
2009, pp. 2575-2580. doi:10.1021/ac802494v
[38] M. J. Sepaniak, P. G. Datskos, N. V. Lavrik and C. Tipple,
“Microcantilever Transducers: A New Approach in Sensor
Technology,” Analytical Chemistry, Vol. 74, 2002, pp. 568A-
575A.
[39] J. L. Staudinger, B. Goodwin, S. A. Jones, D. Hawkins-Brown,
K. I. MacKenzie, A. LaTour, Y. Liu, C. D. Klaassen, K. K.
Brown, J. Reinhard, T. M. Willson, B. H. Koller and S. A.
Kliewer, “The Nuclear Receptor PXR is a Lithocholic Acid
Sensor That Protects against Liver Toxicity,” Processings of
the National Academy Sciences , Vol. 98, 2001, pp. 33 69-3374.
doi:10.1073/pnas.051551698
[40] E. Y. H. Yeung, T. Sueyoshi, M. Negishi and T. K. H. Chang,
“Identification of Ginkgo Biloba as a Novel Activator of
Pregnane X Receptor,” Drug Metabolism and Dispo sition, Vol.
36, 2008, pp. 2270-2276. doi:10.1124/dmd.108.0234 99
[41] S. Kortagere, D. Chekmarev, W. J. Welsh and S. Ekins, “Hy-
brid Scoring and Classification Approaches to Predict Human
Pregnane X Receptor Activators,” Pharmaceutical Research,
Vol. 26, 2009, pp. 1001-1011.
doi:10.1007/s11095-008-9809-7
[42] P. Li, “Evaluation of Luciferin-Isopropyl Acetal as a CYP3A4
Substrate for Human Hepatocytes: Effects of Organics Sol-
vents, Cytochrome P450 (P450) Inhibitors and P450 Induc-
ers,” Drug Metabolism and Disposition, Vol. 37, 2009, pp.
1598-1603. doi:10.1124/dmd.109.027268
[43] M. Abdelrahim, E. Ariazi, K. Kim, S. Khan, R. Barhoumi, R.
Burghardt, S. Liu, D. Hill, R. Finnell, B. Wlodarczyk, V. C.
Jordan and S. Safe, “3-Methylcholanthrene and Other Aryl
Hydrocarbon Receptor Agonists Directly Activate Estrogen
Receptor,” Cancer Research, Vol. 66, 2006, pp. 2459-2467.
doi:10.1158/0008-5472.CAN-05-3132
[44] J. P. Jackson, K. K. Kabirov, I. M. Kapetanovic and A. Lyu-
bimov, “In Vitro Assessment of p450 Induction Potential of
Novel Chemopreventive Agents sr13668, 9-cis-uab30 and
Pentamethychromanol In Primary Cultures of Human Hepa-
tocytes,” Chemical-Biology Interactive, Vol. 179, 2009, pp.
263-272. doi:10.1016/j.cbi.2008.12.005
[45] H. Kuramitz, J. Natsui, K. Sugawara, S. Itoh and S. Tanaka,
“Electrochemical Evaluation of the Interaction between Endo-
crine Disruptor Chemicals and Estrogen Receptor Using
17-Estradiol Labe led with Daunomycin,” Analytical Chemistry,
Vol. 74, 2002, pp. 533-538. doi:10 .1021/ac010426b
[46] G. G. J. M. Kuiper, J. G. L emmen , B. Carlson, J. C. Cor ton, S .
H. Safe, P. T. Van Der Saag, B. Van Der Burg a nd J. Gusta fs-
son, “Interaction of Estrogenic Chemicals and Phytoestrogens
with Estrogen Receptor,” Endocrinology, Vol. 139, 1998, pp.
4252-4263. doi:1 0.1210/en.139.10.4252
[47] S. Stoica, B. S. Katzenellenbogen and M. B. Martin, “Activa-
tion of Estrogen Receptor—by the Heavy Metal Cadmium,”
Molecular Endocrinology, Vol. 14, 2000, pp. 545-553 .
doi:10.1210/me.14.4.545
[48] D. J. Mangelsdorf, C. Thummel, M. Besto, P. Herrlich, G.
Schütz, K. Umesono, B. Blumberg, P. Kastner, M. Mark, P.
Chambon and R. M. Evans, “The Nuclear Receptor Super-
family: The Second Decade,” Cel l, Vol. 83, 1995, pp. 835-839.
doi:10.1016/0092-8674(95)90199-X
[49] H. Masuyama, Y. Hiramatsu, M. Kunitomi, T. Kudo and P. N .
MacDonald, “Endocrine Disrupting Chemicals, Phthalic Acid
and Nonylphenol, Activate Pregnane X Receptor-Mediated
Transcription,” Molecular Endocrinology, Vol. 14, 2000, pp.
421-428. doi:10.1210/me.14.3.421
[50] H. Masuyama, N. Suwaki, Y. Tateishi, H. Nakatsukasa, T.
Seqawa and Y. Hiramatsu, “The Pregnane X Receptor Regu-
lates Gene Expression in a Ligand and Promoter-Selective
C
opyright © 2011 SciRes. JBNB

Microcantilever-Based Nanomechanical Studies of the Orphan Nuclear Receptor Pregnane X
Receptor-Ligand Interactions
143
Fashion,” Molecular Endocrinology, Vol. 19, 2005, pp. 1170-
1180. doi:10.1210/me.2004-0434
[51] D. Hopwood, C. R. Allen and M. McCabe, “The Reactions
between Glutaraldehyde and Various Protein. An Investigation
of Their Kinetics,” Histochemical Journal, Vol. 2, 19 70, pp.
137-150. doi:10.1 007/BF01003541
[52] D. R. Walt and V. I. Agayn, “The Chemistry of Enzy me and
Protein Immobilization with Glutaraldehyde,” Trends Analyti-
cal Chemistry, Vol. 13, 1994, pp . 425-430.
doi:10.1016/0165-9936(94)85023-2