Meat Science 65 (2003) 765–769

www.elsevier.com/locate/meatsci

Isotachophoretic determination of added phosphate

in meat products
Martin Dušek*, František Kvasnička, Lenka Lukášková, Jana Krátká
Department of Food Preservation and Meat Technology, Institute of Chemical Technology,
Technická 1905, CZ-166 28 Prague 6, Czech Republic

Received 31 January 2002; received in revised form 9 October 2002; accepted 9 October 2002

Abstract
From the phosphate content in meat and meat protein a new ratio of free (non-protein) phosphate/protein (16
2 mg/g) was
calculated. From the level of soluble phosphate determined by capillary isotachophoresis (CITP), protein content determined by
Kjeldahl method and with the use of the earlier-mentioned ratio the amount of added phosphates could be calculated. The method
was compared with the spectrophotometric dry-ashing reference method on 34 samples and good agreement was found.
# 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Phosphates; Protein; Capillary isotachophoresis; Meat products

1. Introduction reason the use of a mixture of polyphosphates with dif-

ferent chain lengths is more effective.
Phosphates are important components of meat. They The amount of added phosphates is limited to 5000
are an integral part of some proteins and cell walls and mg kg1 (expressed as P2O5) because they may chelate
are part of the energy delivery process (e.g. ATP, ADP important metals ions (calcium and magnesium). The
and phosphocreatine). amount of added phosphate is determined by the dif-
Polyphosphates are widely used as additives in meat ference between the amount of total phosphate and
products. Phosphates function by the sequestration of protein bound phosphate. A spectrophotometric dry-
metal ions and the dissociation of the actomyosin com- ashing method (AOAC, 1984, 1990; Cattaneo & Bal-
plex bringing about an increase in water-holding capa- zaretti, 1997; Pulliainen & Wallin, 1996) is the most
city (WHC). Another important reason for using frequently used technique for the determination of
phosphates is their ability to increase meat pH and to phosphate in foods. The amount of bound protein
slow discoloration by stabilizing vitamin C. Several phosphate is calculated from the concentration of
polyphosphates are used in the meat industry and they nitrogen (Kjeldahl method) and from the known ratio
differ from one another in their properties. Those with a (0.0106 g P/g protein) of phosphate to protein (K•os-
low degree of polymerization (mono and diphosphates) sovska, 1998).
increase water-holding capacity in meat but their use is The soluble forms of free phosphates (mono-, di-, . . .,
limited by their low and slow solubility, especially in poly-) can be determined by anion-exchange HPLC and
cold brine. In contrast, triphosphate is soluble and dis- post-column molybdate colorimetric detection (Shamsi
solves rapidly but its effect on water-holding capacity in & Danielson, 1993), ion chromatography (Baluyot &
meat is slow because of the need for conversion (enzy- Hartford, 1996; He Cui, Fa Cai & Qiang Cu, 2000;
mic and acid hydrolysis) to pyrophosphate. For this Sekiguchi, Matsunaga, Yamamoto & Inoue, 2000;
Stover, Bulmahn & Gard, 1994 Svoboda & Schmidt,
1997), capillary electrophoresis (Stover, 1999; Stover &
* Corresponding author. Tel.: +420-2-2435-3003; fax: +420-2-
Keffer, 1993; Wang & Li, 1996) or by capillary iso-
3333-6284. tachophoresis (CITP; Kvasnička, 2000; Motooka, Nar-
E-mail address: dusekm@vscht.cz (M. Dušek). iai & Nakazaki, 1983; Suhaj, Ježek & Lacko, 2001).
0309-1740/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0309-1740(02)00279-6
766 M. Dusˇek et al. / Meat Science 65 (2003) 765–769

The aim of this study was to compare a new iso- with 50 ml of distilled water in an ultrasonic bath for
tachophoretic method with a standard method for the approximately 30 min. The mixture was cooled and
determination of phosphates. This new method is able made up to volume with distilled water and then filtered
to determine individual polyphosphates directly in a (FILTRAK, 388, Germany). The filtrate was refiltered
water extract, thus avoiding the need for dry-ashing. through a membrane filter (0.45 mm, MACHEREY-
Thirty-four samples of meat products and six samples of NAGEL, Düren, Germany) prior to CITP analysis.
different kinds of meat were used. A ZKI 02 isotachophoresis analyser (Labeco, Spišská
Nová Ves, Slovak Republic) with column coupling was
used. The separation was performed in a FEP (fluori-
2. Materials and methods nated ethylene–propylene copolymer) pre-separation
capillary (160  0.8 mm i.d.), which was coupled to a
2.1. Samples FEP analytical capillary (160 0.8 mm i.d.). Zones were
detected by conductivity. The isotachophoretic data
Meat and meat product samples were obtained from were collected and evaluated by a software package
different market sources and were minced and wrapped ITPWin 2.20 (KasComp, Slovak Republic) running on
in sealed PE bags and stored in a freezer at 15  C until a personal computer. Anionic analysis of phosphates
analysis. They included 14 samples of ham, 10 samples was performed with a leading electrolyte comprising of
of frankfurters, four samples of chopped and one sam- 10 mM HCl+20 mM glycylglycine+0.05% hydro-
ple of whole mortadella, five samples of pate, and six xypropylmethylcelullose (pH=3.0) and a terminating
samples of meat from four species. electrolyte of 5 mM citric acid. The currents applied to
the pre-separation and the analytical capillary were 250
2.2. Reagents and solutions and 50 mA, respectively.
Calibration curves were constructed using solutions of
Sodium phosphate and glycylglycine were purchased monophosphate and di-phosphate at concentrations of
from Sigma-Adrich (Czech Republic), ascorbic acid, 10, 50, 100, 200 and 500 mmol l1.
sodium, molybdate, zinc oxide, sodium diphosphate,
citric acid and hydrochloric acid were purchased from 2.3.4. Spectrophotometric determination of phosphorous
Lachema (Brno, Czech Republic). The remaining che- as molybdenum blue after dry-ashing
micals were supplied by PENTA (Czech Republic). All Phosphorous was determined, after dry-ashing, by
reagents were analytical grade. spectrophotometry at l=823 nm using the molybdate–
ascorbic acid reaction (Pulliainen & Wallin, 1996) and a
2.3. Analytical methods standard curve constructed using standard solutions (0,
0.01, 0.02, 0.03, 0.04, 0.05 and 0.06 mg P). Spectro-
2.3.1. Protein photometric measurements were carried out with the
Protein (N  6.25) was determined as total nitrogen Spectronic1 model 20 GENESISTM VIS spectro-
by the Kjeldahl method modified by the addition of a photometer (Spectronic Instruments, Rochester NY,
mixture of K2SO4 and CuSO4 as catalyst. Samples (1.5 USA). All determinations were repeated at least once.
g) were digested with a mixture of sulphuric acid and Results were calculated as an average of replicates.
hydrogen peroxide and the content of nitrogen was
determined using a KJELTEC AUTO 1030 Analyser
(Tecator, Sweden). 3. Results and discussion

2.3.2. Net protein The regression equations of CITP (concentration C in

A 2-g sample (all samples were weighed to an accu- mmol.l1 vs. zone length L in seconds) for mono- and di-
racy on three decimal places) of minced meat was phosphate were C=5.43103L8.66103 (corre-
homogenized in 75 ml water and the mixture boiled for lation coefficient 0.9999) and C=3.94103
30 min on a hot plate. After cooling, 10 ml of 10% tri- L0.24.103 (correlation coefficient 0.9996), respec-
chloracetic acid was added. The precipitate was filtered tively. Calibration curves were linear up to 1000 mmol
through filter paper (FILTRAK, 388, Germany) and l1. The limit of detection of phosphate (LOD) was 2
used for the determination of nitrogen (N6.25=net mmol l1 and limit of quatification (LOQ) was 5 mmol
protein) or mixed with ZnO for the determination of l1 for both analytes.
phosphate. All data were calculated from parallel measurements,
which were accepted under the condition that the RSD
2.3.3. Isotachophoresis of the parallel measurements is lower than the precision
One gram of minced product or 2 g of meat was of the method used. The precision of both methods,
placed in a 100-ml volumetric flask and homogenized calculated from the RSD observed for one sample mea-
 M. Dusˇek et al. / Meat Science 65 (2003) 765–769 767

Table 1
Results of chemical analyses of different meat species

Pork Beef hind-quarter Chicken leg Chicken breast Turkey leg Turkey breast

Protein (%) 20.2
0.3 21.9
0.2 20.8
0.2 24.1
0.6 22.64
0.05 26.3
0.2
Net protein (%) 20.8
1.0 19.6
0.2 16.8
0.4 20.0
1.2 19.7
0.5 20.4
1.5
Protein-bound P2O5 (mg/kg)a 513
1 770
8 981
39 825
1 494
36 567
40

Total P2O5 (RM) (mg/kg)b 4937
124 4441
105 4513
83 5377
63 4120
64 5445
122

Free P2O5 (CITP) (mg/kg)c 3549
2 3089
5 3295
21 4403
10 3620
24 3931
31
Free P2O5 (RM) (mg/kg)d 4424
125 3671
114 3531
198 4552
54 3626
308 4878
361

Ratio of Total P2O5 /Protein (mg/g) 24.5
0.7 20.4
0.5 21.8
0.5 22.2
0.7 22
1 20.8
0.5
Ratio of Free P2O5 /Protein (mg/g)e 17.6
0.3 14.0
0.1 15.8
0.2 18.2
0.5 15.9
0.1 14.9
0.2
a
Amount of P2O5 in net protein determined by reference method.
b
Amount of P2O5 determined by reference method.
c
Amount of free (non-protein) P2O5 determined by isotachophoresis.
d
Amount of free (non-protein) P2O5 calculated from the difference between total amount of P2O5 in meat and amount P2O5 found in net proteins
determined by reference method.
e
Amount of P2O5 determined by isotachophoresis.

sured six-times, was 1.6 and 4.3% for the reference Table 2
method and isotachophoretic methods, respectively. Results obtained by determination of phosphates in 34 meat products,
using the isotachophoresis (CITP) and photometric reference methods
(RM)
3.1. Ratio of free phosphates/proteins in meat
Sample Source Protein (%) Phosphates (mg P2O5/kg)
Table 1 shows the results of the determination of
RM CITP
protein, net protein, phosphate (total and protein-
bound) in the different meats. The total phosphate was 1F Frankfurters 11.46
0.08 560
15 639
19
obtained by the spectrophotometric reference method, 2F 13.5
0.3 445
9 34.6
0.5
3F 13.71
0.03 399
8 417
3
and free phosphate by CITP. Phosphate content was 4F 12.6
0.1 1870
52 1907
16
expressed as phosphorus pentoxide (P2O5=2.29P). 5F 12.54
0.05 627
3 831
10
According to the reference method the added phosphate 6F 12.59
0.08 540
5 590
17
content is calculated from the difference between the 7F 14.26
0.05 1390
15 423
5
total phosphorous (determined by the spectro- 8F 10.16
0.02 81
2 88.9
0.7
9F 16.6
0.3 553
14 472
5
photometric method) and phosphorous associated with 10F 9.8
0-2 960
30 676
5
the protein assuming 0.0106 expresses the amount 15M Mortadella 10.37
0.05 2150
22 1610
25
phosphorous in the meat proteins (g P/g protein). 11H Hams 19.2
0.1 3730
29 3573
45
Most of the phosphorous (85–90%) in meat is water- 12H 15.28
0.01 3330
55 2548
28
soluble (free) phosphates, which are determined by iso- 13H 23.58
0.07 2060
37 2044
15
14H 16.96
0.07 2660
89 1644
14
tachophoresis. Because of this it is necessary to change 16H 11.89
0.02 3830
85 3697
50
the method of calculation. The added phosphates were 17H 11.0
0.2 2000
44 1881
14
calculated from the difference between the sum of total 18H 14.09
0.01 3110
45 3150
22
free phosphorous and phosphorous associated with the 19H 19.80
0.05 1710
15 1934
14
proteins. Phosphate as phosphorous pentoxide in the 20H 11.68
0.08 1960
35 1227
11
21H 15.8
0.2 2030
46 1847
48
sample is calculated as follows: 22H 14.4
0.8 3220
187 2428
20
23H 9.41
0.05 3380
35 3221
46
Phosphate ðmg P2 O5 =kgÞ ¼ F  10ðPr  RÞ 24H 14.94
0.09 1550
24 1544
25
25H 13.5
0.8 2290
142 2950
23
where F=sum of soluble phosphates (mg/kg) deter- 26P Pate 14.5
0.2 330
5 469
5
27P 4.55
0.07 980
16 1032
12
mined by isotachophoresis; Pr=amount of protein (g/ 31P 11.2
0.1 1720
23 872
9
100 g); R is the ratio of free protein phosphates to the 33P 7.5
0.1 2300
47 986
8
protein content. R was defined as the ratio of the 34P 10.1
0.3 258
9 904
7
amount of free phosphate (sum of all soluble forms) 28S Chopped 11.3
0.4 710
27 114.4
0.9
determined by CITP [Table 1, line Ratio of Free P2O5 / 29S 11.0
0.1 650
11 585
7
30S 11.59
0.05 960
29 993
7
Protein] and protein content. The average R value was 32S 12.9
0.3 1070
27 480
5
16
2 mg/g.
768 M. Dusˇek et al. / Meat Science 65 (2003) 765–769

Because there are no added phosphates in fresh more variable than the total phosphate levels in meat
meat, the isotachophoretically determined phosphate in due to post-mortem changes (Witowska, 1981) as phos-
a water-extract of meat ‘‘corresponds’’ to the total phate is released from compounds such as proteins,
phosphate less the protein-bound phosphate deter- nucleotides, phospholipides, phosphocreatine.
mined by the reference method. The values of free
phosphates (non-protein) determined by CITP [see 3.2. Comparison of methods for the determination of
Table 1, line Free P2O5 (CITP)] are compared with added phosphates
those [Table 1, line Free P2O5 (RM)] calculated from
the diference between amount of total phosphate and Meat products (34) were divided into two groups,
amount of protein-bound phosphate determined from hams and other meat products (Fig. 1). Table 2 shows
the net protein [Table 1, line 4 and 3, respectively] by the concentration of added phosphates determined by
the reference method. Comparison of the free phos- the RM and CITP methods. It is clear that for the
phate levels in the meat samples, determined by CITP cheaper products (7P, 8P, 14H, 22H, 31P, 33P, 28S and
and RM shows small differences (Table 1). The lower 32S) the concentrations of added phosphates deter-
value of free phosphate determined by CITP, com- mined by CITP and RM were different. These differ-
pared with the reference method, could be because ences are probably due to added proteins (egg, milk,
there is some non-protein phosphates in meat, for plant protein extracts or isolates) in these products. The
example, nucleotides, phospholipids, phosphocreatine, different ratio of total phosphate/free phosphate in non-
etc. meat additives (Alli & Baker, 1981; Brooks & Morr,
From the data summarized in Table 1 the ratio of 1984) will decrease the value of added phosphates
total P2O5/protein in mg/g was calculated. The obtained determined by the CITP method. The negative results
value of 21
1 (it corresponds to (9.2
0.4)103 were obtained for products in which the content of
expressed as g P/g protein) is in good accord with the nitrogen compounds (proteins) was increased by the
value previously obtained (K•ossovska, 1998). The ratio addition of material with a lower ratio of total phos-
of Total P2O5/protein (mg/g) ranges from 22 to 24.5 phate/protein than found in meat.
(RSD=6.9%) and is likely dependent on the meat spe- The results of CITP determination of added phos-
cies. The ratio of Free P2O5/protein (mg/g) obtained by phates were compared with those obtained by the spec-
CITP was wider, from 14.0 to 18.2 (RSD=9.7%) and trophotometric reference method (Fig. 1). It can be
varied more than the ratio of Total P2O5/protein concluded from the regression coefficients (intercept
obtained by RM. This wider range could be because the 104
122, slope 0.904
0.063) than the CITP method
CITP method determines only soluble free phosphate, gives lower results than the reference method. This is
i.e., a part of the total phosphates determined by RM. due to fact that each method (CITP and reference)
The ratio of free soluble phosphate/total phosphate is determines different forms of phosphate.

Fig. 1. The relationship between added phosphates (* hams,  other meat products) calculated from the results determined by the isotachophoresis
(CITP) and by the reference method (RM).
 M. Dusˇek et al. / Meat Science 65 (2003) 765–769 769

Despite the differences found for some samples the Cattaneo, P., & Balzaretti, C. (1997). Fosforo totale nei prodotti
results show that the CITP method could be used as an carnei e metodi per la determinazione. Ingegneria Alimentare, 1, 7–
13.
alternative method for the determination of added
He, Cui, Fa, Cai, & Qiang, Xu. (2000). Determination of triphosphate
phosphates in meat products. Furthermore, the CITP in frozen cod and scallop adductor by ion chromatography. Journal
method is simple, quick and allows the determination of of Chromatography A, 884, 89–92.
polyphosphates of different chain lengths. The main Klossovska, B. (1998). Oznaczanie fosforu dodanego w produktach
advantage of CITP is the possibility to determine total miêsnych. Gospodarka miêsna, 6, 48–50.
soluble phosphates. Because free soluble phosphate, Kvasnička, F. (2000). Application of isotachophoresis in food analy-
sis. Electrophoresis, 21, 2780–2787.
coming not only from added phosphates but also from Motooka, I., Nariai, H., & Nakazaki, K. (1983). Determination of
naturally occurring phosphate, may deplete nutritionally average chain length of linear polyphosphates by isotachophoresis.
important metals, we suggest replacing RM with CITP. Journal of Chromatography A, 260, 3773–3782.
Instead of the acceptable maximum level of added phos- Pulliainen, T. K., & Wallin, H. C. (1994). Determination of total
phate (5000 mg kg1) in meat products the maximum phosphorous in food by colorimetry: summary of NMKL Colla-
borative Study. Journal of AOAC International, 79(6), 1408–1411.
level of water-soluble phosphate should by assessed. Sekiguchi, Y., Matsunaga, A., Yamamoto, A., & Inoue, Y. (2000).
Analysis of condensed phosphates in food products by ion chroma-
tography with an on-line hydroxide eluent generator. Journal of
Chromatography A, 884, 639–644.
Shahab, A.Shamsi, & Danielson, N. D. (1993). Ion chromatography
Acknowledgements
of polyphosphates and polycarbonates using a naphthalene-
trisulfonate eluent with indirect photometric and conductivity
The authors are grateful for financial support pro- detection. Journal of Chromatography A, 653, 153–160.
vided from the Grant Agency of the Czech Republic Stover, F. S. (1999). Capillary electrophoresis of oxo anions. Journal
(Project 525/99/1279) and from the National Grant of Chromatography A, 834, 243–256.
Agency for Agricultural Research of the Czech Repub- Stover, F. S., Bulmahn, J. A., & Gard, J. K. (1994). Polyphosphate
separations and chain length characterization using minibore ion
lic (Project EP9179). chromatography with conductivity detection. Journal of Chromato-
graphy A, 688, 89–95.
Stover, F. S., & Keffer, S. S. (1993). Separation of condensed phos-
References phates using capillary zone electrophoresis with indirect UV detec-
tion. Journal of Chromatography A, 657, 450–454.
AOAC. (1984). Phosphorous (total) in meat. 24.015. In Official meth- Suhaj, M., Ježek. J., & Lacko, L’. (2001). Možnosti využitia kapilárnej
ods of analysis (14th ed.). Arlington: AOAC Inc. izotachoforézy na stanovenie rôznych foriem fosfátov v potravinách.
AOAC. (1990). Phosphorous in meat. Automated method 972.22. In In Laboralim 2001 (pp. 236–240). Bratislava: Vydavatel’stvo STU.
Official methods of analysis (15th ed.). Arlington: AOAC Inc. Svoboda, L., & Schmidt, U. (1997). Single-column ion chromato-
Alli, I., & Baker, B. E. (1981). Constitution of leguminons Seeds. A note graphy of linear polyphosphates. Journal of Chromatography A,
on protein-phytic acid interactions during isolation of acid-soluble 767, 107–113.
protein from Phaseolus Beans. J. Sci. Food Agric, 32, 588–592. Wang, T., & Li, Sam F. Y. (1996). Separation of polyphosphates and
Baluyot, E. S., & Hartford, C. G. (1996). Comparison of polyphos- polycarboxylates by capillary electrophoresis in a carrier electrolyte
phate analysis by ion chromatography and by modified end-group containing adenosine 50 -triphosphate and cetyltrimethylammonium
titration. Journal of Chromatography A, 739, 217–222. bromide with indirect UV detection. Journal of Chromatography A,
Brooks, J. R., & Morr, Ch.V. (1984). Phosphorous and phytate con- 723, 197–205.
tent of soybean protein components. Journal of Agricultural and Witkovska, A. (1981). Post mortem increase of creatinine in pork
Food Chemistry, 32, 672–674. meat. Acta Alimentaria Polonica, 7(1-2), 51–58.