Enzyme Field Effect Transistor (ENFET) for Estimation of

Triglycerides using Magnetic Nanoparticles

A. Vijayalakshmi @, Y.Tarunashree@, B. Baruwati#,S.V. Manorama#,

B.L.Narayana@, R.E.C. Johnson@ and N.M. Rao@*

@ Centre for Cellular and Molecular Biology

# Indian Institute of Chemical Technology
Uppal Road
Hyderabad
India 500 007
 * Corresponding author
Phone # +91-40-27192514
Fax #+91-40-27160311
madhu@ccmb.res.in
Abstract:

Ion selective field effect transistor (ISFET) is a robust platform to develop

biosensors. A variety of methods are used, covalent or entrapment, to

associate enzymes or antibodies to the gate surface of a FET. We have

employed a novel method of retaining the enzyme molecules at the gate

surface by immobilizing the enzyme on magnetic ferrite nanoparticles and

applying a permanent magnet below the gate of the FET. By immobilizing a

stable, designed lipase on nanoparticles we are able to estimate the

triglyceride concentrations in the range of 0.1% to 1.5%. Tributyrin,

trioctanoate and triolein have provided similar results. The reaction volume

could be scaled down to 0.2 ml without a loss in slope or sensitivity. Ionic

strength (>150 mM of NaCl) has strong influence on the sensitivity of the

measurement. The advantage of this configuration of associating the

sensing enzyme at the gate is circumvention of mass transfer problems,

increasing the amount of enzyme at the gate surface besides providing an

opportunity to use a single device for multiple analyte detection.

Key words : Biosensor/ ISFET/ triglyceride/ lipase/ magnetic nanoparticles/

immobilization
Introduction:

Biosensors incorporate a biological molecule into a device to generate a

specific and measurable event. Biosensors are judged on the sensitivity,

response times, dynamic ranges, robustness and ease of fabrication

(Setford and Newman, 2005). Field effect transistors have become attractive

biosensors when the standard metal oxide FETs were redesigned by

replacing the gate electrode with a reference electrode dipped in solution to

make a contact with the gate oxide (Bergveld,2003). The surface potential at

the gate is influenced by the metal (SiO2, Si3N4, Al2O3 or Ta2O5) and the pH of

the solution (Yuqing et.al. 2003). Application of ISFET as a biosensor is

realized when the pH changes near the gate are mediated by specific

enzymes in the presence of their substrates in a dose dependent manner.

Enzyme based field effect transistors (ENFET) have been widely investigated

to many important clinically important small molecules including glucose,

urea, creatinine, organophosphates, penicillin and hydrogen peroxide

(Soldatkin et.al. 1993; Hamlaoui et.al.2002; Singh, et.al. 1999; Tang et.al.

2003; Hai et.al. 2006; Premanode and Toumazou,2007). In addition ENFETs

are used in detection of non-enzymatic proteins such as cell metabolism

(Castellarnau et.al., 2007).

One of the challenges in ENFET design is the method of generating the

enzyme layer, i.e., immobilization, near the gate surface. The methods

determine the sensitivity, response time and reuse of the ENFET (Jianrong

et.al., 2004). Broadly two protocols are followed wherein the enzyme is cross

linked to the surface of the gate or entrapped in a polymer coat on the gate.

Polymer entrapment leads to slower response times due to mass transfer of

the analyte. Cross linking the enzyme to the gate is widely followed with the
development of “Softer” immobilization methods (Hermanson, 1996). In

addition of use of polymer films and metal oxide particles have been useful

in increasing the enzyme layer at the gate (Luo et.al.,2004; Luo and Do,2004,

Yin et.al. 2001). However, the particles have to be retained on the surface by

entrapping them using a polymer membrane usually made with Nafion,

resulting in diffusion barriers (Gorchkov et.al. 1996, Luo et.al., 2004;

Soldatkin et.al. 1993; Gupta and Chaudhury,2007).

Magnetic nanoparticles have been extensively used in purification of cells,

proteins, gene delivery, contrast agents, therapeutic agents (Jianrong

et.al.,2004; Gupta and Chaudhury,2007). Simple methods of preparation,

manipulation with external magnets and easy chemistry to biofunctionalise

the magnetic particles have made them very popular (Neuberger et.al.2005).

Though few metal nanoparticles have been used in ISFET magnetic

nanoparticles have not been used so far (Luo et.al., 2004; Luo and Do,2004).

Enzyme immobilized magnetic NPs could be retained on the gate surface by

a magnet during the measurement. Since the particles in the range of 50 nm

behave as ferrofluids, simple lab magnets are sufficient to retain the

particles on the gate surface. The advantage is in preventing the loss of

particles during measurements in liquid due to washing and at the same

time providing large amount of enzyme at the gate surface compared to

direct immobilization, which is restricted by the available area. The particles,

after aging or loss of activity, could be simply removed and the same FET

device could be reused. Magnetic fields generated by the lab magnets have

no influence on the electrical characteristics of the FET device.

Serum triglyceride concentration is an important diagnostic marker for

coronary heart diseases (Avramoglu, Basciano and Adeli,2006). Elevation of

normal triglyceride levels (0.3-2.5 mM) in the serum, along with other serum

parameters, is considered unhealthy. Current methods of estimation of

triglycerides is by using a battery of enzymes to provide a colored product

(Dykstra and Lawrence,1976, Carter et.al. 1976, Fossati and Prencipe,1986).

Initially triglycerides are hydrolysed by a lipase producing glycerol and fatty

acids. Stoichiometrically produced glycerol is converted by glycerol kinase

into glycerol phosphate, which is further converted by either an oxidase or

dehydrogenase to produce a colored signal. Recently few reports were

made where triglycerides were detected by acidification signal produced

due to the action of lipases ( Pijanewska et.al. 2001; Nakako et.al,1986;

Reddy et.al. 2001;, Basua et.al. 2005). An electrolyte–insulator–

semiconductor capacitor (EISCAP) based or porous silicon- based sensors

that show shifts in C-V characteristics depending on the pH (Basua et.al.

2005) and microreactor based detection of triglycerides using ISFET

(Pijanewska et.al. 2001) were reported where the pH changes are brought

about by the presence of lipase. In this report we present data on detection

of three triglycerides by ISFET, where the enzyme, lipase, was immobilized

on magnetic nanoparticles and retained on the gate surface with a magnet.

Materials and Methods:

Materials: Lipase ( Bacillus subtilis) was purified after over-expression in E.

coli by methods described earlier (Acharya et.al. 2004). The lipase used in

this study was a laboratory evolved mutant lipase with improved

thermostability properties. ISFETs were purchased from LioniX BV,

Netherlands. Tributyrin, trioctanoate, triolein and gum arabic were

purchased from Sigma, USA. Maleimido-Butyryloxy-Succinimide ester

(GMBS) and mercaptopropyltrimethoxysilane (MPTS) were purchased from

Fluka, USA. p-nitrophenyloleate was synthesized by conventional means

from an equimolar solution of oleic acid and p-nitrophenol in

dichloromethane using an equivalent amount of N,N-methylcyclohexyl-

carbodiimide (DCC) and a catalytic amount of 4-N,N-dimethylaminopyridine

(DMAP). All solvents used were of analytical grade.

Methods:

Preparation of NiFe2O4 nanoparticles: NiFe2O4 naoparticles were

synthesized via a hydrothermal route. In a typical synthesis procedure,

stoichiometric amounts of iron (III) nitrate nonahydrate and nickel (II)

nitrate hexahydrate (2:1 molar ratio) were dissolved in 700 ml of deionized

water under magnetic stirring. The pH of the solution was adjusted to 8 by

the drop-wise addition of aqueous ammonia. The solution was then stirred

vigorously for 2 hrs and transferred to a stainless steel autoclave

(indigenously designed). The temperature of the autoclave was gradually

raised to 225 C over a period of 2 hrs and maintained for 2 hrs under auto-

generated pressure of 20 kg/cm2. After cooling, the precipitate was washed

several times with distilled water, filtered, and dried in an oven at 100 C

overnight. The sample was then ground to a fine powder using a mortar and

pestle and used for further analysis.

Complete physicochemical characterizations of the synthesized

nanoparticles’ phase, morphology have been carried out by X-ray diffraction

(XRD) and Transmission Electron Microscopy (TEM). Elemental composition

and magnetic properties have been evaluated by Atomic Absorption

Spectroscopy (AAS) and Vibrating Sample Magnetometer (VSM) studies. X-

Ray diffraction patterns of the powdered samples are recorded on a Rigaku

Miniflex Table Top diffractometer using CuK radiation (1.5406 Ao) source

over a 2 range of 2 to 80.

Transmission electron micrographs are recorded on a Phillips Tecnai G2 FEI

F12 electron microscope (Fig.1A). The powder samples are dispersed in

ethanol by ultrasonication before loading onto a carbon coated copper grid

and then allowed to dry at room temperature before recording the

micrographs. Lipase immobilized NPs were dried on the grid and allowed to

dry in air. Atomic absorption spectroscopy (AAS) studies were carried out on

a Perkin Elmer Analyst 300 atomic absorption spectrometer. 2.5 mg of each

sample was dissolved in 2.5 ml of concentrated H2SO4 and the volume was

adjusted to 25 ml. 5 ml of this solution was then further diluted to 25 ml.

This diluted sample was then used for the AAS analysis.

Magnetic studies are carried out on a Lakeshore 7000 Vibrating sample

magnetometer up to a field of 10000 Gauss at room temperature and the

saturation magnetization (Ms) and the coercivity (Hc) were calculated for

each sample.

Measurement of pH: A measuring circuit was built with a commercial

ISFET as the sensing element ( Support information Fig. A & B). A linear,

near-Nernst sensitivity of 57.6 mV/pH was achieved in the pH range 4 to 9.

The circuit design and the V-I characteristics were provided as support

material. A commercial reference Ag/AgCl electrode was used for the gate

connection.The slope of the FET was marginally marginally higher (59.9

mV /pH) with the magnet attached to the underside of the pit of the FET.

Immobilization of lipase NiFe2O4 particles: The method involves

derivatization of the NPs with MPTS followed by attachment of a

heterobifunctional agent, GMBS. This would reduce cross-linker self

condensation, prevents polymerization and results in reproducible

monolayer attachment of protein to these surfaces. Albumin was used to

block the unbound areas on the NPs. Briefly the procedure involves washing

the particles with toluene several times and to a 100mg/ml suspension of

particles 50 l of MPTS was added. The solution was stirred under nitrogen

atmosphere for 2 h at room temperature. MPTS was removed with several

washes of toluene and then particles were washed with absolute ethanol.

GMBS was added to particles in ethanol and stirred for 1 h at room

temperature under nitrogen atmosphere. Ethanol was removed with

washes in double distilled water and later with phosphate buffered saline

(PBS; 10 mM sodium phosphate buffer pH 7.2 in 150 mM NaCl). To the

particles washed in PBS lipase was added (0.22 mg /ml) and was left on ice

for one hr. The unbound lipase was removed with several washes with

bovine serum albumin solution (1% BSA in PBS) and finally with PBS.

Separation of nanoparticles during these procedures was done by placing a

small magnet at the bottom of the tube. Magnetic nanoparticles were

processed in identical fashion without the addition of lipase and were used

as control. Enzyme bound particles are stored at 4oC. The activity of bound

lipase was measured by assaying its hydrolytic activity on p-

nitrophenyloleate (PNPO). The activity was measured in 10 mM phosphate

buffer ( pH 8) containing 2 mM of PNPO solubilized in Triton X-100. Lipase

mediated hydrolysis of PNPO results in release of chomogenic p-

nitrophenol, whose absorption was monitored at 405 nm (Acharya et.al.

2004) Based on activity the amount of bound lipase was found to be 1.7

g/mg of nanoparticles.
Substrate preparation : Triglycerides-tributyrin, trioctonate and triolein -

were prepared as emulsions in 0.8% gum arabic in PBS. Emulsification was

achieved by sonication in a Branson sonifier. Fresh stocks (10X) were

prepared each day and discarded at the end of the day. The stocks were

stable emulsion during the experiments.

pH measurement: Complete hydrolysis of triglycerides by lipase results in

stoichiometric production of fatty acids and protons ( Fig.2A). One glycerol

molecule was produced after complete hydrolysis of one triglyceride

molecule. The setup for pH measurement is shown in Fig.2B. A simple

neodymium magnet was glued to the back of the gate of the ISFET.

Triglyceride hydrolysis was monitored either by dipping the ISFET along with

the strapped reference electrode and a magnet into a beaker containing the

triglyceride solution or by drop method where 200 l of triglyceride solution

was placed in the well of the gate. The reference electrode was a

commercial glass electrode of 3 mm diameter. The ISFET was washed

extensively with double distilled water (DDW) and nanoparticle solution was

placed on gate, which is followed by further washes with DDW. After dipping

into the buffer the device shows a drift towards lower pH values (0.2-0.5 pH

units) for few minutes before the signal reaches a stable value. The reaction

was initiated by the addition of triglyceride solution from a stock (10X). The

samples were manually shaken during the experiments since magnetic

stirrers could not be used. All measurements were made in PBS. Addition of

triglyceride in the absence of nanoparticles or in the presence of enzyme-

free nanoparticles did not cause any drift.

Results and Discussion:

Characterization of Nanoparticles: The crystallinity and the phase of the

as synthesized nanoparticles were identified and compared with the data

from Joint Committee on Powder Diffraction standards (JSPDS) card no

10-325. JSPDS card provides information about the standard “d” (inter planar

spacing) values, relative intensities of the corresponding diffraction peaks

and the crystal planes. It also provides information about the unit cell

dimensions. XRD confirmed the formation of cubic spinel NiFe2O4 (Support

information Fig.2). It is obvious from the TEM micrographs that the particles

are nearly monodisperse with average particle size of 15 nm that is

comparable with the crystallite size calculated from XRD data (Fig.2). Atomic

absorption spectroscopic studies reveal that the weight percentage ratio of

Fe to Ni in the nanoparticles was 2.06 and very close to the stochiometirc

value. From the hysterisis plot it is seen that the samples are slightly

ferromagnetic in nature at room temperature but the magnetization does

not saturate even up to the applied field of 8,000 Gauss. The coercivitiy value

calculated was 60.11 and the magnetization value observed at the applied

field of 8000 Gauss is 44.85emu/g, which was less than the saturation value

for bulk NiFe2O4 sample (55emu/g).

Estimation of Triglycerides with Lipase-FET:

Lipase mediated hydrolysis of the triglycerides leads to production of

stoichiometric amounts of proton and fatty acids in addition to glycerol

(Fig.3, Woolley and Petersen,1994). Initial experiments were performed with

tributyrin. When dipped in a fresh solution of buffer the NP containing FETs

showed a small shift towards lower pH, which stabilizes after 3-4 minutes.

The reaction was always started with the addition of triglyceride. The change

in the pH upon addition of triglyceride was rapid and usually 90% of the full
scale deflection occurred in less than 120 s (Fig.3). The final value of the pH

was stable for several minutes. The change in pH was proportional to the

amount of tributyrin present in the assay (0.01 to 1 %, Fig4A). The final pH

value was unaltered when the amount of NPs on the gate were varied by

four fold indicating the final value was independent of the amount of lipase

present ( data not presented). However, the initial rate in the change of the

pH was proportionally higher with higher amounts of lipase containing NPs.

To rule out the contribution of voltage changes arising from events other

than hydrolysis of triglycerides, following control experiments were done

and were found to have no contribution to the pH signal changes: a, Buffer

alone; B, gum arabic in PBS; C, NPs without the enzyme with tributyrin as

substrate and D, NPs with bovine serum albumin in the presence of

tributyrin as substrate.

Similar experiments were done by monitoring the pH changes with

trioctanoate, an 8-carbon triglyceride and triolein, 18-carbon triglyceride.

The change in pH was proportional to the amount of trioctanoate and

triolein present in the medium. The observed pH changes and the linearity

range obtained with trioctanoate and triolein were similar to tributyrin,

indicating the hydrolysis was independent of the chain length of the

triglyceride (Fig.4 B,C).The regression coefficient of  ph vs. [TG] was in the

range of 0.25-0.32 indicating that the hydrolysis is similar with the three

substrates. Lower values of slope with triolein could be due to lesser

preference of triolein as substrate. Lipase from Bacillus subtilis was

demonstrated to have preference of trioctanoate over triolein (Acharya et.al.

2004). The active site of the lipase fits a trioctanoate efficiently compared to

other triglycerides. The hydrolysis of triolein was also followed by adding

200 l of substrate on the well of the gate (Fig.5). All the control experiments

including substrate blanks and NPs without the lipase were found to cause

negligible changes in the ISFET response when tested at 0.2 ml volume. The

change in pH was much larger in small volumes compared to larger volumes

for a given percent of triolein (slope was 1.17 in 0.2 ml volume compared to

0.32 observed in 5ml reaction volume). Absence of stirring in low volume

measurements may be partly responsible for the observed differences.

The stability of the enzyme immobilized NPs was tested for two weeks. We

measured the activity both by using p-nitropheyloleate (PNPO) as substrate

and also the pH response to triglycerides on the FET sensor. We observed a

15% loss in activity (with PNPO) in the first week and 45 % loss of activity

after two weeks of preparation. However we did not observe significant

changes when the pH was measured using the FET device. This could be

due to the fact that pH ( end point of reaction) was insensitive to the small

variations in the amount of enzyme. Ionic strength altered the extent of

voltage signal but the slope of the voltage vs. pH was unaltered.

Consequently measurements made in the presence of high salt (> 0.2M

NaCl) resulted in reduced voltage signal ( Data not reported). Based on our

earlier studies we know that salt (NaCl) up to 0.5 M does not alter the activity

of soluble lipase (Acharya et.al. 2004). Hence the changes in the presence of

salt were due to altered response of ISFET to salt. Different buffer salts, Tris,

HEPES and phosphate (all at pH 7.2 and 10mM), were investigated with

triolein as substrate. We did not find any dependency of the pH signal on the

nature of the buffer salt.

Electron Microscopy of the NPs:

Identically processes NPs with and with out the immobilized lipase were

viewed by electronmicroscopy (Fig.1). The average size of the NP was

around 15 nm. Despite extensive sonication the NPs have clumped and the

clumping was less extensive in lipase treated NPs. Lipase immobilized NPs

were, more frequently, observed as beads on a string.

Compared to other clinically important analytes such as glucose, cholesterol

etc. fewer sensor designs for detection of triglycerides using lipase were

reported. Lipase was used as sensor for estimation of triglycerides using

porous silica (Reddy et.al. 2003), ISFET (Nakako et.al. 1986), microreactor

(Pijanewska et.al. 2001) and by electrolyte–insulator–semiconductor

capacitor (Basua et.al. 2005). Lipase immobilized on porous silica showed a

linear range of 5mM-30 mM for triglyceride detection with the measurement

time of 15 min. Similarly, using silicon dioxide as insulator and matrix for

immobilization of lipase tributyrin was detected in the concentration range

of 5 mM -50 mM (Reddy et.al. 2003). A minireactor containing lipase

immobilized silica particles was used to detect the pH change using an ISFET

(Pijanewska et.al. 2001). The sensitivity of the reactor was up to 5 mM for

tributyrin and showed very poor pH response with triolein. Device reported

here was has lower sensitivity for tributyrin but larger dynamic range ( 0.4-

1.0 %) compared to the minireactor system ( 0.01 -0.16 %), however the

minireactor was insensitive to triolein whereas our device has shown

sebsitivity and dynamic range equal to tributyrin. Lipase immobilized on

ISFET surface as membrane showed linear response in the range of 0.6 mM -

3 mM for tributyrin and 0.6 mM-3 mM for triolein (Nakako et.al. 1986). In all

these techniques the linear range for detection of tributyrin was up to 40

mM and the measurement times are 10-30 min. In addition the amount of
enzyme used for measurement was very high varying from 12 mg -30 mg. In

the presented work the response times were less than 5 min and the

amount of enzyme used for assay was 1.2 g, far less than those reported

for triglycerides. In comparison to the multiple enzyme based detection of

triglycerides using spectrophotometers, ISFET based detection of

triglyceride using only lipase could be a viable alternative.

Magnetic nanoparticles have been in use in biology especially for separation

of cells (Willner,Baron and Willner,2007;McCloskey,Chalmers and

Zborowski,2003). Magnetic nanoparticles have been used in drug delivery,

tumor ablation by hyperthermia, magnetic resonance imaging, delivery of

nucleic acids into cells, cell labeling and tissue repair (Gupta and Gupta,2005

and references therein). Recently, superconducting quantum interference

device (SQUID) using superparamagnetic nanoparticles and a ‘‘microscope’’

is used to detect biological targets rapidly (Chemla et.al.2000). Presence of

magnet or nickelferrite nanoparticles has not affected the sensitivity of the

FET and the slopes in all these conditions was 57.6 + .5 mV. Since the NPs are

magnetic, processing of NPs during the protocol was rapid. Magnetic

nanoparticles were not used so far in ISFET design to the best of our

knowledge. NPs made of manganese oxide and silicon oxide were used to

immobilize enzymes for sensing glucose (Luo et.al., 2004; Luo and Do,2004).

Immobilization of lipase is achieved in 3-4 hours time and the employed

chemistry using succinimidylesters is known to be a soft protocol. The lipase

NPs were stable for two weeks (tested period) when stored at 40C. The lipase

employed in the detection of triglycerides was developed in the laboratory

using in vitro evolution properties. This lipase variant was characterized to

be thermostable and has higher activity than the wild type lipase (Acharya

et.cl. 2004). Enzymes could be evolved or modified specifically to suit the

sensor design and also robustness.

Interfacing the enzyme with the transducer is considered to be critical for

the efficiency of sensing (Setford and Newman,2005). The popular methods

of immobilizing or entrapment or a combination of both are convenient but

still provide considerable challenge to overcome the diffusion limitations or

enhance the longevity of the enzyme (Gupta and Chudhury,2007; Luo and

Do,2004). Enzyme entrapment methods increase detection and recycling

times due to limitations of mass transfer. Immobilization of enzyme on the

surface of the sensor gate limits the amount of enzyme that could be

accommodated on the surface and also requires that the enzyme

withstands repeated washing cycles. Inactivation of immobilized enzyme

after repeated usage necessitates replacement of the entire device.

Immobilization of enzyme on magnetic nanoparticles and usage of magnet

allows interfacing larger quantities of enzyme to the gate. In addition the

same device could be repeatedly used by replacing the NPs with fresh NPs.

The method of immobilizing enzymes on magnetic NPs also allows the use

of same device for sensing of other analytes, that show changes in the pH

upon hydrolysis, viz. urea-urease, cholesterol-cholesterol oxidase,

creatinine-creatinase etc. The device could be made further miniaturized

using integrating a reference electrode to the device (Oelbner et.al.,2005)

Conclusion :

Methods of interfacing enzyme to the sensor surfaces are critical in the

function and utility of a biosensor. In this work we demonstrate that

enzymes immobilized on the magnetic nanoparticles could be conveniently

retained on the ISFET gate surface by providing a laboratory magnet

underneath the gate. The observed change in the pH was proportional to

the triglyceride concentration in the sample. The immobilized lipase could

detect triglycerides- tributyrin, trioctanoate and triolein- in the range of 5-30

mM . The detection of triglycerides was rapid (< 5 min) and the enzyme NPs

and the device could be used for several days without any loss in the slope

of  pH vs. triglyceride curves. In addition the measurements could also be

made in small volumes ( 200 microlitres) thus reducing the requirements of

the sample volume. Soft immobilization methods, easy methods of

synthesizing magnetic NPs, rapid measurements and use of device for

multiple analyte detection are attractive features of this ISFET design.

Acknowledgements:

The work was supported by a network grant by Council of Scientific and

Industrial Research, India (CMM 0011). The authors thank Drs V.K. Khanna

and V.K. Dwivedi of CEERI, Pilani for their advice and help. Authors thank Dr.

Shashi Singh for the electron microscopy of nanoparticles. BB is a research

scholar supported by CSIR.

Legends to the Figures:

1. Transmission electron microscopy images of NPs without (A) and with

(B) lipase.

2. (A) Lipase mediated hydrolysis of triglycerides by a lipase. Complete

hydrolysis of the triglycerides by lipase produces a mole of glycerol,

three moles of fatty acid and three moles of protons.(B) Schematic of

 the ISFET braced to a permanent magnet . The commercial ISFET

hooked to the measuring device ( see support information for circuit

and image) is used for pH measurement. A Neodymium magnet (1cm X

1.5 cm) was glued underneath the sensor pit of the FET.

3. Lipase ENFET: Time course of change in the pH in the presence of

tributyrin, measured by the FET with lipase immobilized NPs. Hydrolysis

of tributyrin at 0.01 %(),0.05% ();0.1% ( ); 0.5 % ( ) and 1% () was

monitored in the presence of 20 l of NPs in a 5 ml reaction volume.

The reaction was started with the addition of tributyrin from a 20X

stock.

4. Lipase ENFET: Changes pH with tributyrin ( A), Trioctanoate (B) and

triolein (C) were plotted in the presence of 20 l of NPs in a 5 ml

reaction medium. The three triglycerides were emulsified in the

presence of gum Arabic. Experiments were done on three preparations

of NPs and a representative data was presented. pH was measured as

a difference between the pH value at the time of addition of the

substrate to the pH value stabilized, usually within 10 min. The

coefficient of variation associated with the replicates ( n>3) was

conistently less than 10%with all the three triglycerides. The coefficient

of correlation and its errors associated with the three triglycerides is as

follows : tributyrin : slope= 0.361; R=0.97±0.03; trioctanoate : slope =

0.31; R= 0.96±0.032; triolein : slope = 0.303; R=0.99±0.004.

5. (A) Changes in pH with a Lipase ENFET of triolein in 0.2 ml reaction

volume. The position of the reference electrode was altered in these

measurements compared to large volume ( 5 ml) measurements. The

reference electrode was kept at an angle normal to the FET surface so

that the liquid junction and the bottom of the electrode were in contact

with the substrate solution. Rest of the procedures of measurement is

as given in the methods.

References:

Acharya,P., Rajakumara,E., Sankaranarayanan,R., Rao,N.M., 2004. J. Mol. Biol.

341, 1271–1281.

Avramoglu,R.K, Basciano,H., Adeli,K. 2006. Clinica Chimica Acta 368, 1 – 19.

Basua,I., Subramanian,R.V., Mathewa,A., Kayasthac, A.M. Chadha,A.

Bhattacharyaa,E. 2005. Sensors and Actuators B 107, 418–423.

Bergveld, P. 2003. Sensors and Actuators B 88, 1–20.

Carter,T., Martin,P.J., Wilding,P., Goldberg,D.M. 1976, Clin. Chem. 22, 520-

524.

Castellarnau,M., Zine,N., Bausells,J., Madrid,C., Ju´arez,A., Samitier,J.,

Errachid,A. 2007. Sensors and Actutators B 120,615-620.

Chemla Y.R., Grossman H.L., Poon Y., McDermott R., Stevens R., Alpert M.D.,
Clarke,J. 2000. Proc Natl Acad Sci U S A, 97, 14268– 72.

Dykstra,S.L., Lawrence,J.D. 1976. Clin.Chem. 22, 1896-1899.

Fossati,P., Prencipe,L., 1982. Clin. Chem. 28, 2077-2080.

Gorchkov,D.V., Soldatkin,AP, Maupa,H., Martelet, C., Jafferzic-Renault N.,

Analutica Chimica Acta 1996. 331 217-223.

Gupta,R. Chaudhury,N.K. 2007. Biosensors and Bioelectronics 22, 2387–

2399.

Gupta,A.K., Gupta,M. 2005. Biomaterials 26 3995–4021.

Hai,A., Ben-Haimb, D., Korbakov,N., Cohena,A., Shappir,J., Oren,R.,

Spira,M.E. Yitzchaik,S., 2006. Biosensors and Bioelectronics 22, 605–612.

Hamlaoui, M.L., Reybier, K., Marrakchi, M., Jaffrezic-Renault, N.,Martelet, C.,

Kherrat, R., Walcarius, A., 2002. Anal.Chim. Acta 466, 39–45.
Hermanson, G.T. 1996. Bioconjugate Techniques, San Diego, Academic
Press.

Jaffrezic-Renault,N., Wan,K., Senillou,A., Chovelon,J-M., Martelet,C.

ANALUSIS, 1999. 27, 578-586.

Jianrong, C., Yuqing,M., Nongyue,H., Xiaohua,W., Sijiao,L. 2004.

Biotechnology Advances 22, 505–518.

Luo,Y-C., Do,J-S., Biosensors and Bioelectronics 2004. 20, 15–23.

Luo,X-L, Xu,J-J, Zhao,W., Chen,H-Y 2004. Sensors and Actuators B 97 249–255.

McCloskey, K.E., Chalmers J.J., Zborowski, M., 2003. Anal Chem., 75,6868–74.

Merkoc¸A., 2007. FEBS Journal 274, 310–316.

Nakako,M., Hanazato,Y., Maeda, M. and Shiono,S., 1986. Anal. Chim. Acta

185, 179-185.

Neuberger,T., Schöpf, B., Hofmann,H., Hofmann,M., von Rechenberg,B.,

2005. J. Magnetism and Magnetic Materials, 293,483-496.

Oelßner,W., Zosel,J., Guth,U., Pechstein,T., Babel,W.,Connery,J.G., Demuth,C.,

Gansey,M.G.,Verburg,J.B.2005. Sensors and Actuators B 105, 104–117.

Pijanewska, D.G.P., Baraniecka,A., Wiater, R., Ginalska, G., Labareswski,J.,

Torbicz,W., 2001. Sensors and Actuators B 78, 263-266.

Premanode,B., Toumazoub,C., 2007. Sensors and Actuators B 120,732–735.

Reddy,R.R.K., Chadha, A., Bhattacharya,E., 2001. Biosensors and

Bioelectronics 16, 313–317.

Setford,S.J., Newman,J.D. in Barredo,J.L. (Ed.) 2005. Methods in

Biotechnology, Vol.17: Microbial Enzymes and Biotransformations, Humana
Press Inc., Totowa.
Singh,A.K., Flounders,A.W., Volponi,J.V., Ashley,C.S., Wally,K., Schoeniger, J.S.,
1999. Biosensors and Bioelectronics 14 703–713.

Soldatkin, A.P., El’skaya, A.V., Shul’ga, A.A., Nctchiporouk, L.I., Nyamsi Hendji,
A.M., Jaffrezic-Renault, N., Martelet, C., 1993. Anal. Chim. Acta 283, 695–701.

Willner,I., Baron,R., Willner,B., 2007. Biosensors and Bioelectronics 22,

1841–1852.

Woolley,P., Petersen,S.B. 1994. Lipases: Their Structure, Biochemistry and

Application, Cambridge University Press.

Yin,L-T., Chou,J-C., Chung,W-Y., Sun, T-P., Hsiung,K-P., Hsiung,S-K., 2001.

Sensors and Actutators B 76, 87-192.

Yuqing,M., Jianguo,G., Jianrong,C., 2003. Biotechnology Advances 21, 527–

534.
Fig.1

A B

50nm
nmn
m
 Fig.2

(A)
O

OCOR1 Lipase R1COO-

OH
+ R2COO
-
OCOR2 OH + H+
OCOR3 H2O R3COO-
OH

O
Triglyceride Glycerol Fatty acid
Proton

Reference
electrode
(B)

Magnetic NPs

Source Drain

Magnet
Fig.3
Fig.4
Fig.5
Support Information:

Fig.1 : (A) Layout of the measuring circuit used in the experiments.

(B) Photograph of the complete system:

Fig.2 Representative x-ray diffraction pattern of the synthesized NiFe2O4
samples