A disposable microbial based biosensor for quality control in milk

Neelam Verma *, Minni Singh

Department of Biotechnology, Punjabi University, Patiala 147002, PB, India

Received 8 August 2001; received in revised form 30 October 2002; accepted 6 November 2002

Abstract

The food industry needs suitable analytical methods for quality control, that is, methods that are rapid, reliable, specific and cost-

effective as current wet chemistries and analytical practices are time consuming and may require highly skilled labor and expensive

equipment. The need arises from heightened consumer concern about food composition and safety. The present study was carried

out keeping in view the recently emerging concern of the presence of urea in milk, called ‘‘synthetic milk’’. The biocomponent part of

the urea biosensor is an immobilized urease yielding bacterial cell biomass isolated from soil and is coupled to the ammonium ion

selective electrode of a potentiometric transducer. The membrane potential of all types of potentiometric cell based probes is related

to the activity of electrochemically detected product, and thus to the activity of the substrate by a form of the Nernst equation.

Samples of milk were collected and analyzed for the presence of urea by the developed biosensor with a response time as low as 2

min. The results were in good correlation with the pure enzyme system.

# 2003 Elsevier Science B.V. All rights reserved.

Keywords: Biosensor; Potentiometer; Quality control; Milk

1. Introduction

Reliable and cost effective analytical methods are

increasingly needed in the food industry for determina-

tion of specific chemical compounds in foods and food

products. The need arises from increased regulatory

action and heightened consumer concern about food

composition and safety (Luong et al., 1997). The food

and drink industries need rapid and affordable techni-

ques for quality control as current wet chemistries and

analytical practices are time consuming and may require

highly skilled labor and expensive equipment (Luong et

al., 1991). The increasing demand for on-line measure-

ment of milk composition directs science and industry to

search for practical solutions, and biosensors may be a

possibility (Eshkenazi et al., 2000).

The specific objective of this work was to develop a

disposable microbial biosensor, in which the biocompo-

nent part is disposable, based on a potentiometric

transducer for monitoring the presence of urea in

milk. Urea is hydrolyzed by urease according to the

reaction (Guilbault and Kauffmann, 1987):

(NH2)2CO�2H2O�H� 0Urease

2NH�4 �2HCO�

3

The source of urease was a microbe Bacillus sp.

isolated from soil. The microbe was immobilized and

placed in intimate contact with a potentiometric trans-

ducer. The presence of urea in the media induces the

production of urease, which is absolutely specific

(Sumner, 1951). Cell based probes allow simple and

rapid determination which previously could not be

determined by electrochemical methods (Corcoran andRechnitz, 1985). The limited stability of isolated en-

zymes and the fact that some enzymes are expensive or

even unavailable in pure state has promoted the use of

cellular materials (plant tissues, bacterial cells etc.) as a

source for enzymatic activity (Rechnitz, 1981), e.g.

banana tissue, which is rich in polyphenoloxidase, can

be incorporated by mixing within the carbon paste

matrix to yield a fast responding and sensitive dopaminesensor. These biocatalytic electrodes function in a

manner similar to that of conventional enzyme electro-

des.

* Corresponding author. Tel.: �/91-175-282-461; fax: �/91-175-282-

881.

E-mail address: nverma@pbi.ernet.in (N. Verma).

Biosensors and Bioelectronics 18 (2003) 1219�/1224

www.elsevier.com/locate/bios

0956-5663/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.

doi:10.1016/S0956-5663(03)00085-X

Milk has been referred to as a human’s most nearly

perfect food from a nutritional standpoint. Its whole-

someness and acceptability depend on the strictest

sanitary control, and the sanitary practices employed

by the dairy industry (Potter and Hotchkiss, 1995). The

average percentage composition of milk is water

(82.4%), protein (4.7%), fat (7.4%), lactose (4.6%) and

ash (0.78%). The inorganic constituents of ash are Ca

(131), Mg (13), Na (54), K (143) and P(100) mg/100 g

(Kirk and Sawyer, 1991). The present study was carried

out keeping in view the recently emerging concern of the

presence of urea in milk. Urea is not a natural

constituent of milk and is present as an adulterant. Its

presence in milk of dairy cattle could be attributed to

excessive N uptake (Broderick and Clayton, 1997)

and the rumen efflux of crude protein intake (Hof et

al., 1997). The above factors can account for the

presence of urea in the range of 1.66�/2.66 mM (low

milk urea nitrogen (MUN)) and 2.83�/4.16 mM (high

MUN) (Melendez et al., 2000). A presence of high

MUN reduces fertility rates in dairy cattle (Butler et

al., 1996).

The fact that the developed biosensor has been

applied only to the analysis of urea in milk is because

the composition of the contents is known. Milk can be

considered as containing three basic components namely

water, fat and solid-not-fat (SNF). The organic matter

in the SNF consists mainly of casein and whey proteins

together with lactose and lactic and citric acids. The

average protein content of cow, human and buffalo milk

are 3.3, 2.0 and 4.7%, respectively. The amino acids have

different % compositions in these proteins, like trypto-

phan present is 26.39/4.8 mg/1.7 g protein.

We studied the interference caused by tryptophan.

Even at 65-fold concentration there is a negligible

increase in release of NH4� of 0.67 mM.

The amino acid degradation occurs in the mitochon-

drial matrix for which it would have to permeate

through cell wall, cell membrane, cytosol and mitochon-

drial membrane. Thereafter, the glutamate transaminase

would catalyze the transfer of the amino group to alpha-

ketoglutarate to form glutamate which then undergoes

oxidative deamination, catalyzed by glutamate dehy-

drogenase to release ammonium ion which would

require a greater response time in comparison to 2 min

of response time of the biosensor developed by us.

Regarding the interference caused by other nitrogenous

compounds like creatine and creatinine, the culture is

not positive for deaminases as per the biochemical

report of identification. Also, these compounds are

prevelant in clinical samples. By further improvement

of the present sensor its applications can be broadened,

for now it has been confined to analysis of milk samples

only.

2. Experimental

2.1. Reagents

All chemicals were of analytical grade. The source of

urease was a microorganism isolated from soil and of

pure urease from water melon seeds (CDH, Mumbai,

India).

2.2. Biocomponent-culture and immobilization of

microorganism

The microbial culture was isolated from soil (Stanier

et al., 1987) and cultivated in a media containing urea

(2.5%), beef extract (1.0%), peptone (1.0%) and sodium

chloride (0.5%) with pH 7.0 under aerobic conditions at25 8C for 24 h. It was identified by MTCC and Gene

Bank, Institute of Microbial Technology, Chandigarh as

Bacillus sp. (shown in Fig. 1). For immobilization, the

bacterial culture was centrifuged at 5000 rpm for 20 min

at 4 8C and the pellet was retained. The optical density

of the cell biomass, measured at 600 nm was set to 1.000

using PBS (pH 7.5). Aliquots of 1.5 ml cell biomass were

filtered off on Whatman No.1 filter paper; the paper wasdried and coupled to the body of the electrode with an

‘O’ ring to form the biocomponent of the biosensor

(shown in Fig. 2).

The growth profile of the microbe was studied and

since it is a microbial system the minor variations in the

activities of the probes was also studied. The stability of

the microbial probe was compared with that of pure

enzyme probe.

2.3. Transducer

The transducer was a potentiometer (Cyberscan-2500)in conjunction with a NH4

� Ion Selective Electrode

(ISE-Code No. EC-NH4-03) that detects the electrode

potential developed across the membrane of the elec-

Fig. 1. Urease yielding gram �/ve microorganism isolated from soil

and identified by MTCC and Gene Bank, Institute of Microbial

Technology, Chandigarh as Bacillus sp. (500�/).

N. Verma, M. Singh / Biosensors and Bioelectronics 18 (2003) 1219�/12241220

trode when it comes in contact with ammonium ions

released as a result of urea hydrolysis, forming a secondgeneration biosensor. The membrane potential devel-

oped is related to the concentration of the free ammo-

nium ions in solution by a form of Nernst equation

(Corcoran and Rechnitz, 1985):

E�E��RT=nF ln a

where E is the measured electrode potential; E8,reference electrode potential (constant); RT/nF, slope

of the electrode (mV per decade); and a is the

concentration of NH4� ions in solution.

The biocomponent together with the transducer form

a portable system giving a continuous real time analysis

as is clear from the diagrammatic sketch (Fig. 3).The Fig. 3 represents the real time analysis of the

developed biosensor. The biocomponent with the im-

mobilized biomass is coupled to the body of the

electrode with an ‘O’ ring and is dipped into the

substrate solution. As the hydrolysis begins the ammo-nium ions released are sensed by the Ion Selective

Electrode and displayed on the potentiometer display.

As the hydrolysis proceeds there is a change in

potential, which is continuously displayed.

2.4. Calibration of NH4� ISE

Prepared a 0.55�/10�11 M ammonium standard

stock solution and used serial dilution for preparing

NH4� standards with concentrations varying 10-fold, till

0.55�/10�6 M NH4� standard was prepared. The ionic

strength of all standards and samples was adjusted with

Ionic Strength Adjuster (ISA �/5 M NaCI). 2 ml of ISAwas added to every 100 ml of sample and standard

solutions to maintain a background ionic strength of 0.1

M. For preparing the calibration curve, placed the

Fig. 2. Assembly of NH4� selective electrode with microbial biomass.

Fig. 3. Diagrammatic representation of the biosensor showing its continuous real time analysis.

N. Verma, M. Singh / Biosensors and Bioelectronics 18 (2003) 1219�/1224 1221

beaker containing 25 ml of low value standard with ISA

on the magnetic stirrer and began stirring at a constant

rate. The electrode tip was dipped into the solution after

assuring that the meter is in the mV mode. When the

reading became stable, recorded the mV reading. Similar

measurements were done with increasing value of

ammonium standard and a slope check for the electrodewas done.

2.5. Application of the developed microbial biosensor

The kinetics of the microbial urease, Km and Vmax

were first studied. To apply the developed biosensor tomilk the biocomponent was coupled to the ISE. The

ionic strength of all standards and samples was adjusted

by ISA. Sample volumes of all standards and samples

was 25 ml and the hydrolysis was run at 25 8C with

constant stirring. Between every sample the ISE was

given washing in distilled water to maintain a base

potential. An initialization time of 30 s was optimized

after dipping the biosensor into the solutions for thepotential reading to stabilize followed by a response

time of 2 min. The interference caused by Na� and K�,

if any, was nullified by subtracting the value obtained in

an unhydrolyzed sample i.e. sample without the bio-

component in it. The reliability of the developed

microbial biosensor was checked by comparing with

pure enzyme system (Jenkins et al., 2000) (0.05 g urease

was dissolved in 10 ml PBS and filtered and aliquots of 1ml filtrate were used for hydrolysis) and spectrophoto-

metric determinations of NH4� ions (Vogel, 1978). It

was found to be in good agreement with both systems.

3. Results and discussion

3.1. Growth curve

The growth profile of the bacterial culture was studied

to establish the start of stationary phase at which the

culture could be used as an enzyme source. As seen in

Fig. 4 the culture attained a stationary phase at 24 h of

growth.

3.2. Data on variability of bioprobe preparations

We prepared five bioprobes in a day under the same

experimental conditions. Their activities were as follows:

Table 3.1: Data on variability of Bioprobe prepara-

tions

Probe DmV Enzyme activity (Units)

Probe 1 17.7 6.87Probe 2 17.6 6.80

Probe 3 17.6 6.80

Probe 4 17.7 6.87

Probe 5 17.5 6.77

The Bacillus sp. isolated contains urease which is

intracellular. This inherent immobilization of urease

within the cell prolongs the stability of the biocompo-

nent as shown in the comparative figures (Fig. 5).Thus, the enzyme bioprobe can be stored only for 10

days whereas, because of the inherent immobilization of

the microbial urease within the cell it can be stored after

Fig. 4. Growth profile of Bacillus sp.

Fig. 5. (a) Stability of the enzyme bioprobe, (b) stability of the microbial bioprobe.

Fig. 6. Calibration curve of NH4� ISE.

N. Verma, M. Singh / Biosensors and Bioelectronics 18 (2003) 1219�/12241222

air drying, for more than 1 month, at 4 8C. After usage,

the biocomponent can be washed with buffer, dried and

reused.

3.3. Calibration curve

The ammonium ion electrode was calibrated by using

standards of NH4Cl solutions in the concentration rangeof 0.55�/10�6 to 0.55�/10�11 M NH4

�. The calibra-

tion curve is a shown in Fig. 6.

With minor changes in environmental conditions

there is a variation in slope value, although the Nernst

equation slope is 59 mV at 25 8C.

Slope values 51.6, 52.0, 58.8, and 57.9

Mean�/55.07 S.D.�/9/3.80

3.4. Application of the developed microbial biosensor

3.4.1. Kinetics of the microbial urease

The Km and Vmax of the microbial enzyme have been

studied. As seen in Fig. 7 there is saturation at high urea

concentration. The curve is Hyperbolic following zero

order kinetics at high concentration.

The Km of the microbial urease is 2.08 mM at pH 7.5/

25 8C in phosphate buffer, with a Vmax of 6.25 mmol/min.

From literature study the Km of various ureases are

(Musculus, 1876):

Source of urease Km

Jack Beans 3.0 mM at pH 7.0/25 8C in

maleate buffer

Jack Beans 4.0 mM at pH 8.0/21 8C inTHAM buffer

Bacillus pasteurii 100 mM at pH 5.7/25 8C in

phosphate buffer

Corynebacterium re-

nale

30 mM at pH 7.5 in phosphate

buffer

Thus, it is clear that the microbial urease used in our

study has better kinetic characteristics than other best

known sources, having a lower Km thus higher affinityfor the substrate.

The precision and reliability of the developed system

is evident from the correlation between the spectro-

photometric and potentiometric determination (micro-

bial and pure enzyme) (Fig. 8).

4. Conclusions

We have developed a microbial based biosensor to

determine the presence of urea in milk. A good

correlation of results was obtained with pure enzyme

system and the spectrophotometric methods. However,

it is worth the mention that since milk is a complex

Fig. 7. Rate of the reaction.

Fig. 8. Data of comparison of potentiometric and spectrophotometric determination with fresh samples. (a) Spectrophotometric determination by

Nessler’s test. Standard curve for NH4� (Nessler’s test). (b) Potentiometric determination by NH4

�ISE. Standard curve for NH4� (ISE).

N. Verma, M. Singh / Biosensors and Bioelectronics 18 (2003) 1219�/1224 1223

system it contains many interferents which makes

conventional methods less reliable. We conclude that

the present biosensor has the advantages of specificity,

portability, simplicity, reliability and continuous realtime analysis.

References

Broderick, G.A., Clayton, M.K., 1997. A statistical evaluation of

animal and nutritional factors influencing concentrations of milk

urea nitrogen. J. Dairy Sci. 80 (11), 2964�/2971.

Butler, W.R., Calaman, J.J., Beam, S.W., 1996. Plasma and milk urea

nitrogen in relation to pregnancy rate in lactating dairy cattle. J.

Anim. Sci. 74 (4), 858�/865.

Corcoran, C.A., Rechnitz, G.A., 1985. Cell-based biosensors. Trends

Biotechnol. 3 (4), 92�/96.

Eshkenazi, I., Maltz, E., Zio, B., Rishpon, J., 2000. A three-cascaded-

enzymes biosensor to determine lactose concentration in raw milk.

J. Dairy Sci. 83 (9), 1939�/1945.

Guilbault, G.G., Kauffmann, J.M., 1987. Enzyme based electrodes as

analytical tools. Biotechnol. Appl. Biochem. 9, 95�/113.

Hof, G., Vervoorn, M.D., Lenaers, P.J., Tamminga, S., 1997. Milk

urea nitrogen as a tool to monitor the protein nutrition of dairy

cows. J. Dairy Sci. 80 (12), 3333�/3340.

Jenkins, D.M., Delwiche, M.J., Depeters, E.J., Bon Durant, R.H.,

2000. Refinement of the pressure assay for milk urea nitrogen. J.

Dairy Sci. 83 (9), 2042�/2048.

Kirk, R.S., Sawyer, R., 1991. Dairy products-I. In: Pearson’s

Composition and Analysis of Foods. Addison-Wesley, Longman,

England, pp. 530�/574.

Luong, J.H.T., Groom, C.A., Male, K.B., 1991. The potential role of

biosensors in the food and drink industries. Biosens. Bioelectron. 6,

547�/554.

Luong, J.H.T., Bouvrette, P., Male, K.B., 1997. Development and

applications of biosensors for food analysis. Trends Biotechnol. 15,

369�/377.

Melendez, P., Donovan, A., Hernandez, J., 2000. Milk urea nitrogen

and infertility in Florida Holstein cows. J. Dairy Sci. 83 (3), 459�/

463.

Musculus, M., 1876. Comput. Rend. Acad. Sci. 78(1) 132. Cited from:

Varner, J.E. (1960). Urease. In: P.B. Boyer, H. Lardy, K. Myrback

(eds.), The Enzymes. Academic Press, New York, pp. 247�/251.

Potter, N.N., Hotchkiss, J.H., 1995. Milk and milk products. In: Food

science. Chapman and Hall, New York, pp. 279�/315.

Rechnitz, G.A., 1981. Bioselective membrane electrode probes. Science

214 (4518), 287�/291.

Stanier, R.E., Ingraham, J.L., Wheelis, M.L., Painter, P.R., 1987.

Gram positive Eubacteria: unicellular endosporeformers. In: Gen-

eral Microbiology, fifth ed. Macmillan Press, London, pp. 475�/

494.

Sumner, J.B., 1951. The Enzymes, first ed., vol. I, p. 886. Cited from:

Varner, J.E. (1960). Urease. In: P.B. Boyer, H. Lardy, K. Myrback

(eds.), The Enzymes. Academic Press, New York, pp. 247�/251.

Vogel, A.I., 1978. Colorimetry and spectrophotometry. In: Textbook

of Quantitative Inorganic Analysis, fourth ed. Longman, New

York, pp. 730�/731.

N. Verma, M. Singh / Biosensors and Bioelectronics 18 (2003) 1219�/12241224