Food Protein Analysis
Food Protein Analysis
Food Protein Analysis
EDITORIAL BOARD
Senior Editors
Flavor chemistry and sensory analysis John H.Thorngate III University of California-
Davis
There is no book dealing with food protein analysis exclusively, that is, with the analysis
of proteins in the food system. This books attempts to fill this niche. Protein analysis
comes in two forms: 1) Quantitative analysis, and 2) fractionation and characterization.
The first activity is described here. This publication provides a reference for planning,
performing and interpreting assays for food proteins. Many approved methods derive
from the late-19th century, but they have undergone rigorous testing and modernization.
This book does not focus on reviewing the latest research methods for protein analysis.
With the exceptions of Chapters 6 and 7, each of the 14 self-contained chapters describes
one protein assay—principles, practices, and expected results.
This book describes the effect of food processing on protein assay results with the
emphasis on how to analyze proteins in real foods. A number of “Methods” sections
provide instructions for specific tests. Sample pretreatment and clean-up procedures are
described. General pretreatment strategies help in the avoidance of interference. More
specific clean-up methods apply to particular protein assays and are described along with
these. Example results, performance characteristics, case reports, and practical problems
and solutions related to a wide range of foods are detailed in numerous figures, tables,
and references.
Food protein analysis is a hugely important activity performed by thousands
worldwide. The book should appeal to professionals interested in food proteins and
anyone working in the food system formerly called the food chain. This includes
researchers and workers in agricultural production, food processing, and wholesale and/or
retail marketing. It provides information for the grain or dairy farmer, extension worker,
agricultural scientist, food scientists and technologists, or college professor. Some
techniques described in this book were first used by clinicians, nutritionists, and
veterinary scientists. The book may also be of interest to those in small businesses,
private or government laboratories, research institutes, colleges, and universities. It will
be useful to undergraduate, postgraduate, or postdoctoral students. Sections dealing with
mechanisms assume graduate level chemistry and/or analytical biochemistry.
Any shortcomings of this project are wholly my responsibility. I thank all those
colleagues worldwide whose research is reported here. My thanks to Anna Dolezal, Mr.
DeSouza and Professor Arthur Finch for teaching me to think for myself. I am grateful to
my past students: Drs. Yetunde Folawiyo, Despina Galani, Michael Anaydiegwu,
Kiattisak Duangmal, Pitaya Adulyatham, Kwanele Mdluli, Halima Omar and Sripaarna
Banerjee for raising my awareness of protein assay issues and for reading parts of the
manuscript. Thanks to Dr. Bob Roberts (The Pennsylvania State University) for his
advice on combustion methods. I am grateful to Dr. S. Khokhar and Marcel Dekker, Inc.,
for their commitment. I am also grateful to my family for their support.
R.K.Owusu-Apenten
Contents
Preface xi
Chapter 8. Immunological Assay: General Principles and the Agar Diffusion 196
Assay
Chapter 9. Speciation of Meat Proteins by Enzyme-Linked Immunosorbent 218
Assay
Chapter 10. Speciation of Soya Protein by Enzyme-Linked Immunoassay 248
Chapter 12.Biological and Chemical Tests for Protein Nutrient Value 299
Index 389
1
Kjeldahl Method, Quantitative Amino Acid
Analysis and Combustion Analysis
Protein analysis is a subject of enormous economic and social interest. The market value
of the major agricultural commodities (cereal grains, legumes, flour, oilseeds, milk,
livestock feeds) is determined partly by their protein content. Protein quantitative analysis
is necessary for quality control and is a prerequisite for accurate food labeling. Proteins
from different sources have varying aesthetic appeal to the consumer. Compliance with
religious dietary restrictions means excluding certain protein (sources) from the diet. The
variety of protein consumed is also extremely important in relation to food allergy.
Detecting undeclared protein additives and substitutions is a growing problem. Proteins
show differing nutritional quality or ability to support dietary needs. In summary, protein
analysis has legal, nutritional, health, safety, and economic implications for the food
industry (1).
The estimated global food production total for 1988 was 4 billion metric tons.
Allowing an average of 10% protein in foodstuffs yields 400 million metric tons of
protein annually (2). Nonetheless, sensitivity is a major consideration for protein analysts.
Some immunological methods can detect nanomole (10−9 mole) amounts of protein.
Other important considerations when choosing a method for food protein analysis include
TABLE 1 Approximate Chronology for Methods
for Food Protein Analysis
Date Technique
1831 Dumasa
1843 Nessler’s reagenta
1849 Biuret method
1859 Alkali-phenol reagent or Bethelot’s methoda
1883 Kjeldahla
1927 Folin-Ciocalteau
1944 Dye bindinga
1951 Lowry
1960 Direct alkaline distillation
1960 Near-infrared reflectance (NIR)a
Food protein analysis 2
high sample throughput, simplicity, and low capital costs. Some of the most significant
methods (Dumas, Kjeldahl, and biuret assays) date from the late 1800s (Table 1).
Techniques for food protein analysis are described in this book. I will focus on the
techniques that feature most often in the food science literature. Infrared analysis of food
proteins is not discussed here.
especially for cereal proteins (5). Procedures involving copper-based reagents (Lowry
and bicinchoninic acid assays) continue to be important. Finally, a range of empirical
(viscosity, refractive index, specific gravity) measurements are used for protein
quantitation within industry.
where a is the gradient and b is the intercept for the calibration graph. For each Xi result
we can determine the calculated % Kjeldahl protein value (Ycalc) via Eq. (2).
Ycalc=aXi+b
(2)
Food protein analysis 4
Values for Yi and Ycalc can be compared in order to evaluate the test
method(see later). Some investigators choose to plot the Kjeldahl results
on the X-axis. Therefore, rather than Eq. (1) we get
(3)
where Xi* is % Kjeldahl protein and Yi* is the test result. To compare Eq. (1) and Eq. (3),
notice that a′=1/a and b′=Yi−(Xi / a).
For sample calibration, the assay technique is assumed to be valid. We analyze a set of
(standard) samples containing known amounts of protein. In Eq. (1), Xi now represents a
range of known protein concentrations and Yi are the corresponding instrument responses.
Calibration factors (a, b, etc.) can be determined from simple algebra or statistical
analysis of paired (Xi, Yi) results. From the principles of least-squares analysis,
(4)
and
b=Ym−aXm
(5)
where Xm and Ym are the mean values for all Xi and Yi observations.
Agreement between the reference and test results is measured by the correlation
coefficient (R); R≈1 shows excellent agreement. When Yi and Ycalc observations are
poorly correlated, R≈0. The squared correlation coefficient (R2) can be calculated from
Eq. (6). Most handheld calculators can perform this operation automatically.
(6)
Precision is another measure of the (dis)agreement between Yi and Ycalc values. This can
be expressed as the standard deviation (SD) or coefficient of variation (CV). High-
precision methods produce low values for the SD and CV.
(7)
CV=(SD/Ym)×100
(8)
We can also measure precision (commonly called error) from n-replicate (Yi)
measurements on a single test sample. Thereafter, the numerator in Eq. (7) becomes
(Yi−Ym)2, which is the square of the differences between individual observations and the
Kjeldahl method, quantitative amino acid analysis and combustion analysis 5
mean for all observations. A low CV implies good agreement between successive test
results.
(9)
Usually Yo and b are both set to zero when the analyst sets the instrument baseline
response to zero. Consequently, Eq. (9) becomes
LLD=2.326SDo/a
(10)
This relation shows that LLD decreases with increasing assay sensitivity and with
increasing baseline quality (see decrease in the value for SDo). In order to ensure high
sensitivity, it is important to obtain a stable instrumental baseline.
Williams et al. (7) calibrated beer protein analyses using quantitative sodium dodecyl
sulfate polyacrylamide gel electrophoresis (QSDS-PAGE). A range of test methods were
investigated including biuret, bicinchoninic acid (BCA), Bradford, Kjeldahl, Lowry, and
pyrogallol-red molybdate (PRM) assays. QSDS-PAGE revealed that beer has between
0.5 and 1mgmL−1 protein. Only the Bradford and PRM assays gave accurate results (Fig.
1). The main sources of error were low-molecular-weight interferences. Beer contains
plant pigments, starch, sugars, alcohol, and natural dyes of barley origin. Both Kjeldahl
and combustion analyses were subject to interferences by nonprotein nitrogenous (NPN)
compounds. Dialysis did not improve accuracy for BCA, Lowry, and biuret assays, which
were affected by high-molecular-weight Cu- reducing agents such as pectin and starch.
Calibration issues are discussed in two articles by Pomeranz and co-workers (8,9).
They considered the reliability of several test methods (biuret, dye binding, infrared
reflectance, alkaline distillation method) for analyzing proteins in hard red winter wheat
varieties from the American Great Plains. The test methods were highly correlated with
the Kjeldahl assay (R=0.976− 0.992). The order of precision was Kjeldahl > biuret > dye
binding > infrared analysis. Pomeranz and More (9) also considered the reliability of four
“rapid” methods for barley or malt protein analysis.* A summary of assay performance
statistics is given in Table 4. For barley samples, the precision and sensitivity of analysis
were highest for the Kjeldahl and infrared analyses. The use of Kjeldhal analysis to
calibrate protein assays for dairy products was discussed by Luithi-Pent and Puhan (10)
and also Lynch and Barbano (11).
2. KJELDAHL ANALYSIS
Johan Kjeldahl was born on August 16, 1849 in the town of Jaegerpris in Denmark. In
1876 he was employed by the Carlsberg brewery to develop an improved assay for grain
protein. The Kjeldahl method was published in 1883. The original technique has been
extensively modified. Key steps for the assay are (a) sample digestion, (b) neutralization,
(c) distillation and trapping of ammonia, and (d) titration with standard acid. An
exhaustive
*For the purposes of calibration, 44 samples of barley and 49 samples of malt were analyzed with
biuret, dye binding, infrared, alkaline distillation, and Kjeldahl tests. Such results were the basis for
deriving calibration relations between Kjeldahl and each test method. Then a further 76 samples of
barley and 72 samples of malt were analyzed using only the rapid test methods. Each Xi result gave
rise to a corresponding Kjeldahl protein (Ycalc) value.
Kjeldahl method, quantitative amino acid analysis and combustion analysis 7
Dye bindingc 15
Infrared 0.5–1.0 Y=1.060 cP −1.03 R=0.96 0.838–1.980
a
cP, crude protein determined by Kjeldhal method (N×6.25); Y, response from the test method.
b
Assay standard error calculated from
c
No information given.
Source: Ref. 8.
Food protein analysis 8
account of the Kjeldahl method can be found in the monograph by Bradstreet (12). The
book is divided into five chapters: 1, introduction to the Kjeldahl method; 2, the Kjeldahl
digestion; 3, digestion procedure (for fertilizers, leather, cereals, foods and proteins, coal
and fuels); 4, the distillation and detection of ammonia. Chapter 5 is an extensive
bibliography. A standardized Kjeldahl procedure appears in the International Standard
ISO-1871 (13). Further descriptions are given by Gaspar (14) and Osborne (15).
Initially, only sulfuric acid was used for sample digestion. Then solid potassium
permanganate was added to facilitate sample oxidation. Mercuric oxide was introduced as
a catalyst in 1885. During the acid digestion phase, the food sample is heated with
concentrated sulfuric acid, which causes dehydration and charring. Above a sample
decomposition temperature, carbon, sulfur, hydrogen, and nitrogen are converted to
carbon dioxide, sulfur dioxide, water, and ammonium sulfate [Eq. (11)].
NH2(CH2)p COOH+(q+1) H2SO4→(p+1)
CO2+q(SO2)+4p(H2O)+NH4HSO4 (11)
Digestion is complete when the mixture turns clear (light green color), usually after 20–
30 minutes of heating. A further (after-boiling) period of heating is necessary to ensure
quantitative recovery of nitrogen. Data from McKenzie and Wallace (cited in Ref. 14)
show that adding X (mg) of potassium sulfate per mL of sulfuric acid increases its boiling
point according to the relation Y (°C)=55.8X+ 331.2. A maximum boiling point elevation
of 130°C is achivable by adding 2mg (potassium sulfate) per mL acid. A high boiling
point reduces the sample digestion time. Sample digestion can also be facilitated by using
a catalyst; the order of effectiveness for metal oxide catalysts is Hg > Se > Te > Ti > Mo
> Fe > Cu > V > W > Ag (16). A proprietary brand of Kjeldahl catalyst (Kjeltabs from
Foss Electric Ltd.) comes as tablets. Each tablet contains 0.25g of mercuric oxide and 5g
of potassium sulfate. A working selenium catalyst can be formulated with potassium
sulfate (32 g), mercuric sulfate (5g), and selenium powder (1 g). Chemical oxidants
(hydrogen peroxide, perchloric acid, or chromic acid) can be added to the sulfuric acid to
speed up sample digestion.
Ammonium sulfate is first neutralized with alkali to form ammonia. This is then
distilled and trapped using 4% boric acid. Ammonium borate is then titrated with
standard acid in the presence of a suitable indicator. Low-cost Quick-fit glassware is
readily available for distillation and titration. Sophisticated semiautomatic distillation
systems are also available. The processes of neutralization, distillation, and titrimetric
analyses are summarized as follows.
(12)
(13)
(14)
Suitable titration indicators include methyl orange, methyl red, Congo red,and Tashiro
indicator (a 1:1 mixture of 0.2% methyl red solution and 0.1%methylene blue).
The Kjeldahl technique measures sample nitrogen (SN) as ammonia. The value for SN is
later converted to crude protein (cP) by multiplying by a Kjeldahl factor, FK.
cP(%)=SNFK
(15)
The units for SN are g-N 100g−1 (g-nitrogen released per 100g of sample). The Fk (g-
proteing−1 N) is the amount of protein that produces a gram of nitrogen. Fk is also called
the nitrogen-to-protein conversion factor. AOAC-recommended values for FK for meat
and other food are summarized by Benedict (17). Frequently, FK is given a default value
of 6.25 or 5.7 for animal and plant proteins, which are assumed to have an average N
content of 16% and 17.5%, respectively. In fact, most proteins deviate significantly from
these averages (18). FK is also affected by the presence of NPN (e.g., adenine, ammonia,
choline, betaine, guanidine, nucleic acid, urea, free amino acids). Soya beans have 3–10%
NPN, which increases to about 30% for immature seeds. The amount of NPN also
changes with growth conditions as well as with geographic factors. There is generally no
correlation between NPN and protein content (19). No single FK value applies to all food
types. Ideally, FK should be determined for each individual food type (Table 5).
FK can be calculated from amino acid data (18,20–24). Table 5 lists the 20 naturally
occurring amino acids along with their formula weight, number of nitrogen atoms,
percent nitrogen, and the value for FK. For arginine, FK is 3.11 (=100/32.15). An
idealized protein having all 20 amino acids in equal numbers has a nitrogen content of
14.73%. The FK value is therefore 6.79 (100g/14.73). Evaluating FK from amino acid
data (for
TABLE 5 Determination of the Kjeldahl Factor for
Milk Protein Using Amino Acid Composition Dataa
1 Amino 2 N ato 3 4 5 Am 6 AA AA- 7 8 AA 9 Mole 10 biX
Acid Formula ms % N ino compo N (moles/g frac
weight Acid sition (mg/g N) tion (Xi)
(bi) FK (mg/g N)
N)
Arginine 174.2 4 32.15 3.11 234.0 75.2 0.001343 0.027082 4.71761
Histidine 155.2 3 27.06 3.70 188.0 50.9 0.001211 0.024422 3.79022
Asparagine 132.1 2 21.20 4.72 292 61.9 0.00221 0.044564 5.88694
Glutamine 146.1 2 19.16 5.22 783 150.1 0.005359 0.108048 15.7859
Lysine 146.2 2 19.15 5.22 487.0 93.3 0.003331 0.067157 9.81828
Glycine 75.1 18.64 5.36 128.0 23.9 0.001704 0.034362 2.58058
Alanine 89.1 15.71 6.36 219.0 34.4 0.002458 0.049553 4.4152
Tryptophan 204.2 2 13.71 7.29 90.0 12.3 0.000441 0.008886 1.81447
Serine 105.1 13.32 7.51 338.0 45.0 0.003216 0.064837 6.81433
Proline 115.1 12.16 8.22 571.0 69.5 0.004961 0.100016 11.5118
Food protein analysis 10
skimmed milk) can be achieved in the following steps: (a) express each amino acid as mg
per gram of total nitrogen (see column 6 of Table 5) and (b) calculate the mass of
nitrogen derived from each amino acid (column 7 in Table 5). FK is the weight of amino
acids divided by the weight of amino acid nitrogen (AA-N).
(16)
Some typical values for FK are listed in Table 6 for a range of foods. The use of FK values
for quantitative amino acid analysis is discussed in Sec. 4.
TABLE 6 Nitrogen-Protein Conversion (Fk)
Factors for Selected Food Protein Sources
Food product Fk Food product Fk
Dairy products and egg Roots and tuber
Casein 6.15 Carrot 5.80
Milk 6.02 Beet 5.27
Cheese 6.13 Potato 5.18
Egg 5.73 Potato protein 5.94
Kjeldahl method, quantitative amino acid analysis and combustion analysis 11
B. Potatoes
Mohyuddin and Mazza (26) analyzed proteins from 14 potato cultivars. Potato tubers
were peeled, sliced, diced, and dried in a vacuum oven at 70°C and 48.8 mm Hg pressure.
Each sample was milled and sieved through a 40-mesh sieve. Potato flour (100mg) was
added to each 100-mL Kjeldahl flask, followed by concentrated sulfuric acid (3mL),
hydrogen peroxide (30% solution; 1.5mL), and commercial catalyst (500 mg; 10:0.7 w/w
Food protein analysis 12
ratio of potassium sulfate and mercuric oxide). Heating for 45 minutes digested the
samples. After cooling to room temperature, the contents of Kjeldahl flasks were diluted
to 75 mL and then subjected to colorimetric analysis to determine ammonia.
D. Beer Protein
Concentrated sulfuric acid (2mL) was added to 50 mL of beer (bitter, lager, or stout) and
the mixture was heated until nearly dry (7). Kjeldahl catalyst (10g) and more sulfuric acid
(20mL) were added, followed by further heating for 25 minutes. After cooling for 2
hours, water (250mL) was added and the Kjeldahl flask was connected to a condenser
with one end immersed in a 2% boric acid solution (200mL). Bromocresol green was
used as indicator. Sodium hydroxide (10M, 70mL) was added, followed by heating until
the distillate tested neutral. The borate solution was later titrated with 0.1N HCl. The
nitrogen content in beer was calculated from the relation
%N=14Va/Wd
(17)
where Va (mL) is the volume of HCl required to neutralize ammonia and Wd (g) is the dry
weight of beer. Table 7 summarizes characteristics of the Kjeldahl method used in the
brewing and allied industries (28).
TABLE 7 Macro-Kjeldahl Procedures Currently in
Use in the Brewing and Allied Industries
Parameter Commenta
Instrumentationb Kjel-Tec 1030 (Auto), Kjel-Tec 1007/ 1002*; Kjel-Foss 16120†
Sample weight 0.92 ± 0.23 g(barley or malt) range 0.5–1.5 g, 14 ± 8.4 mL (beer) range
5–25 mL
Catalyst (weight range)c K2SO4+CuSO4+TiO2; 3.5–15.8 g
Volume of conc. H2SO4 16 ± 4.3 mL, range 10–25 mL
d
Digestion temperature 411±14°C, range 380–430°C
Digestion timee 75 min or 9.6 min
End-point detection Mainly colorimetric
Kjeldahl method, quantitative amino acid analysis and combustion analysis 13
a
Average values unless otherwise stated.
b
Local suppliers * Tectator Ltd. and Perstop Ltd., both of Bristol, UK, † Foss Electric (UK) Ltd.,
The Chantry, Bishopthorpe, York, UK.
C
A large amount of a single catalyst or a smaller quantity of a combination catalysts was used. HgO
was used in 2 of the 25 laboratories.
d
Digestion temperatures were not reported for seven laboratories using a manual Kjeldahl
technique.
e
Digestion time plus after-boiling time. The digestion time is 9.6 min when H2O2 is used as a
prooxidant.
Source: Summarized from Ref. 28.
Foss instrument, five used the official AOAC method, and eleven used a block digester
with steam
higher than normal for batch digestion. A superheated layer of acid forms, which
facilitates sample digestion (44). Later tests showed that the recovery of nitrogen from
refractory materials (arginine, creatine, or nicotinic acid) was only 70%. Davidson, et al.
(45) concluded that the AutoAnalyzer digestion module was not reliable if an accuracy of
1% was desired for Kjeldahl analysis. Over 70 different animal feeds (corn grain, wheat,
barley, rice, alfalfa, mixed feeds, feed concentrates) were analyzed using the
AutoAnalyzer digestion module. The recovery of nitrogen was 88–90% (46). In contrast,
using a Technicon block digest or followed by AutoAnalyzer Sampler II led to 100%
recovery of nitrogen from cattle supplement, swine ration, pig starter, and poultry ration
(47). Suitable catalysts include copper sulfate and oxides of mercury, selenium, or
titanium. Ammonia was detected using alkaline phenol reagent (Sec. 3.1). Quantitative
recovery of nitrogen was also demonstrated by Kaul and Sharma (25), who used a
Tectator® heating block to digest 23 assorted strains of rice and 15 other cereal-legume
mixtures. The electrically heated block digests 40 samples per hour under controlled
temperature conditions. Samples were then transferred to the AutoAnalyzer Sampler II
for ammonia detection using the alkaline phenol reagent.
Food protein analysis 16
Colorimetric analysis simplifies Kjeldahl analysis and increases the sample throughput.
Other benefits include increased sensitivity and a greater potential for automation.
Reagents for colorimetric Kjeldahl-N analyses include (a) alkali-phenol reagent (APR),
also called indophenol reagent; (b) ninhydrin (indanetrione hydrate) reagent; (c) Nessler’s
reagent; and (d) acetylacetone formaldehyde reagent. These colorimetric techniques are
reviewed next.
Alternatively, ammonia may first react with phenol to form p-aminophenol, which is then
oxidized by hypochlorite to form (I). Irrespective of how (I) forms, it reacts with 1 mole
of phenol to form indophenol (II). The yellow compound ionizes under strongly alkaline
conditions (pKa 13.4) giving a blue anion
Substituted phenols (o-chlorophenol, m-cresol, and guaiacol) undergo the indophenol
reaction. However, para-substituted phenols and some meta-derivatives do not react.
Color intensity and the rate of color formation increase in the presence of manganese (+2)
ions or acetone. Sodium nitroprusside (10–40 mg L−1) is another catalyst.
A simple APR assay suitable for detecting ≤3 ppm ammonia is described in Ref. 48
(Fig. 3). Indophenol formation is pH and temperature dependent. The linear dynamic
range for ammonia was 0.3–3 ppm with a sensitivity of 0.3284 (absorbance units/1 ppm
NH3). The assay precision (for 1 ppm NH3) was ±3%. Although performed with boric
acid as the background medium, the simple APR assay is probably not suitable for
Kjeldahl method, quantitative amino acid analysis and combustion analysis 17
Method 1
Analysis of ammonia using the APR assay (48,49).
Reagents
1. Phenol (crystalline, >85% pure)
2. Sodium hydroxide solution (5 N)
3. Sodium hypochlorite solution (or commercial bleach)
4. Ammonium chloride solid
5. EDTA (6% w/v)
6. Acetone
Procedure
Preparation of alkali-phenol reagent. Place 62.5 g of solid phenol in a 500-mL beaker
and add 135mL of sodium hydroxide (5 N) slowly with stirring. Caution: Use an ice bath
Food protein analysis 18
to avoid excessive heat buildup. Add 12mL of acetone and make up the volume to 500
mL with deionized water.
Sodium hypochlorite (1% w/v available chloride). Prepare by diluting commercially
available bleach.
Ammonium chloride standards (1000 ppm NH3 and 100 ppm NH3). Dissolve 314.1 mg
of solid NH4Cl in 100 mL of water and then dilute 10-fold. Prepare a working standard
solution (0.5 ppm NH3) daily.
The APR assay sequence. Place 1 mL of sample (or standard) in a test tube. Add
EDTA solution (100 µL) with gentle shaking. Next, add APR (1 mL) and hypochlorite
(0.5mL) in quick succession, mixing after each addition. Finally, add 2.5mL of water and
incubate at 25°C for 60 minutes. Take A625 readings for samples. Prepare a reagent blank
as described next.
Reagent blank (reverse addition method). First, mix hypochlorite (0.5mL) and APR (1
mL) solutions and allow to react for 5–10 minutes. Next add EDTA (100 µL) followed
by 3.4mL of water (or the designated blank solution).
When reverse addition is used, hypochlorite reacts with phenol first. Traces of NH3
present in the blank are not detected (48,49). Reverse addition is useful where ammonia-
free water is not available for sample preparation. After optimization, the linear dynamic
range for ammonia analysis was 50–500 ppb. Assay sensitivity was 200% greater than
the results shown in Fig. 2. Color formation with 50–800 ppb NH3 was virtually complete
after 15 minutes at 14–30°C. Temperature variations had little effect on the reaction.
Thermostating at 25°C for 60 minutes improved the precision.
Addition of acetone to APR increased the response to ammonia by 10-fold. The color
yield with 500 ppb ammonia declined by 2.65%, 4.8%, and 6.8% for 4.5-hour-, 1-day- or
5-day-old APR. Addition of EDTA prevented interference from 100 ppb copper. The
intervals between addition of various reagents must not exceed 1 minute to ensure
optimum precision. Comparing results for normal and reverse addition provides a means
for detecting very small amounts of NH3 in samples—such as water. Otherwise,
ammonia-free water is needed for preparing reagents and blanks. The calibration curve
for the APR assay is described by the equation A625=0.7120X, where X is the
concentration of nitrogen (ppm). The analytical sensitivity was 0.7120 (absorbance units)
for 1 ppm ammonia. The SD for the reagent blank was ±0.0005 ppm. These values lead
to an expected LLD for ammonia of 1.6
TABLE 9 The Comparative Costs of Manual APR
Assay and Other Techniquesa
Technique Capital costb Running cost per Analysis per Cost per
yearc year analysisd
Manual APR 6000 (1) 5200 (1) 8000 0.72 (1)
method
Micro-Kjeldahl 12000 (2) 7000 (1.3) 2000 4.1 (5.4)
AutoAnalyzer 81,000 (13.5) 17,000 (3.3) 32,000 0.78 (1)
Kjeldahl method, quantitative amino acid analysis and combustion analysis 19
APR test
a
Costs are given in deutsche marks. At current exchange rate 2.8 DM=$1.4=£1. All methods
employ a digestion unit costing DM6000.
b
Capital costs for the micro-Kjeldahl method include the cost of a distillation unit and an
autotitrator.
c
The running cost includes DM5000 for miscellaneous chemicals.
d
Calculated for a 10-year period as capital cost/10+running cost)/no. of samples. Ratios of costs are
given in parentheses for each column.
ppb. Assuming a default FK of 6.25, the LLD for protein is 10 ppb. The APR assay is
widely used in conjunction with the AutoAnalyzer.The composition of the APR used in
CFA is pretty much the same as described earlier (45,52).
Kaul and Sharma (25) describe a rare attempt to deploy a manual Kjeldahl-APR assay
for protein analysis. They used a Tectator heating block for micro-Kjeldahl digestion of
grain. Sample nitrogen was then analyzed by the APR assay. The analytical performance
was similar to results obtained with the AutoAnalyzer-APR assay or the conventional
micro-Kjeldahl analysis. From Table 9, the capital cost for the manual Kjeldahl-APR
assay was 2 times lower than for the micro-Kjeldahl and 13.5 times lower than for the
AutoAnalyzer method. Running costs were also lowest for the manual APR assay.
For laboratories handling 40 or more analyses per day, it may be worth investing in an
automated technique. The manual Kjeldahl-APR analysis was advantageous for small
laboratories lacking the wherewithal to purchase an AutoAnalyzer. Mohyuddin and
Mazza (53) used the manual Kjeldahl-APR assay to analyze potatoes (see Sec. II.B.2).
The mean protein content for 14 potato cultivars was 10.65 (±1.23)% by the manual APR
assay and 10.53 (±1.13)% using the AutoAnalyzer.
Hach et al. (54) developed a commercial Nesslerization reagent for use in Kjeldahl
analysis. A sulfuric acid-digested sample (0.4 mL) is diluted with 24.6mL of 0.01%
(w/w) polyvinyl alcohol (PVA) solution. One ml of Nessler’s reagent is added and the
sample is agitated mechanically before absorbance measurements are recorded at 430 nm.
As the product of Nesslerization is colloidal in nature, spectrophotometric analysis is
sensitive to the degree of sample agitation. PVA acts as a colloidal stabilizer and
improves the precision of the Nessler method.
formaldehyde in the presence of sodium acetate. The yellow product (3, 5-diacety1–1, 4-
dihydrolutidine) is measured at 410 nm (∆ε=1.4×103 Lmol−1cm−1). The color-forming
reaction is shown in Eq. (19).
(19)
Acetylacetone-formaldehyde reagent was used for the analysis of medicinal agents such
as paracetamol, sulfanilamide, and chloropramide. The potential for colorimetric Kjeldahl
analysis is obvious.
Method 2
Determination of ammonia using acetylacetone-formaldehyde reagent (55).
Reagents
Acetylacetone-formaldehyde reagent. Place 15mL of formaldehyde (37% w/v) and
acetylacetone (7.8 mL) into a 100-mL flask. Make up to 100mL with distilled water.
Sodium acetate (2M). Dissolve sodium acetate (82 g) in 1 L of distilled water.
Procedure
Add prediluted Kjeldahl digest (<2mL; 25–100µgN) to a 25-mL conical flask followed
by sodium acetate solution (3mL) and acetylacetone-formaldehyde reagent (3mL).
Incubate the mixture at 97.8°C for 15 minutes and cool to room temperature.
Bring the total volume to 25 mL and record A412 values using a 1-cm cuvette.
The linear dynamic range for the preceding assay was 0.5–6.0 µg N (per final reaction
mixture). The calibration graph was described by
where X is the amount of
nitrogen present in the final (25-mL) reaction mixture. The new method shows levels of
accuracy and precision equal to those of the micro-Kjeldahl method.
Kjeldahl method, quantitative amino acid analysis and combustion analysis 21
The following steps are involved in quantitative amino acid analysis: (a) hydrolyze a
sample of food using concentrated hydrochloric acid, (b) determine the amino acid
profile, (c) calculate the concentration of each amino acid in the sample, and (d) calculate
the weight of each amino acid.
Quantitative amino acid analysis is reportedly one of the most reliable methods for
protein quantitation (59–61).
Food protein analysis 24
(20)
However, amino acid profiles are reported in terms of mole fraction of each amino acid
(Xi):
Xi=Ci/Cnet
(21)
(22)
The term ∑(biXi) is called the mean residue weight, W (g mole−1). This is the average
formula weight for all amino acids in the sample adjusted for their frequency.*
(23)
and
cP=WCnet or cP=FCnet
(24)
In Eq. (24), F is the mean residue weight. Usually, W (g mole−1) is adjusted to take into
account two routine errors in amino acid analysis: (a) many colorimetric reagents for
amino acids do not react with proline and (b) tryptophan is destroyed during acid
hydrolysis of proteins. Proline and tryptophan are usually determined by separate
experiments. After correcting for such errors, one gets the conversion factor, F (g
mole−1):
(25)
where XPro and XTrp are the mole fractions of proline and tryptophan. For a
range of meat products, F (g mole−1) was 10–20% larger than W [Eq. (23)]. Calculations
of F (g mole−1) appear in last three columns of Table 5. Typical values for F (g mole−1)
are given in Table 10. Values for F range from 100 to 125 g mole−1 for most
proteins.Most protein sources are now routinely analyzed for amino acid scores during
nutritional evaluations (Chapter 14). This yields all the information necessary for total
protein estimation. Zarkadas and co-workers in Canada are strong exponents of
quantitative amino acid analysis (Table 11).
(26)
quantitative amino acid analysis (62). The results were poorly correlated (R=0.4) with
Kjeldahl results (22.4% protein per dry weight basis). A more recent analysis of freeze-
dried mushroom powder led to estimates of 7.0% protein by dry weight (63). In the later
study, mushroom powder was extracted with 0.5M NaOH and precipitating with
trichloroacetic acid (TCA) before analysis. This removed large amounts of TCA-soluble
NPN associated with mushrooms. Of the total NaOH-soluble nitrogen extracted from
mushrooms, 20% was protein, 60% was urea or ammonia, and 20% was free amino acids.
Food samples are now routinely extracted with organic solvents to remove NPN before
quantitative amino acid analysis (Table 11). Advocates for quantitative amino acid
analysis point to its compatibility with plant proteins. There is no interference from
phenols, tannins, and lignin. By contrast, the Kjeldahl method is unsuitable for plant
tissues regardless of the conversion factor used (73).
The Dumas assay predates Kjeldahl analysis by 50 years (Table 1). The former technique
was invented by Jean Baptiste Dumas. Early applications include the analysis of plant
materials (74,75), meat (76), casein, whole powdered milk, soybeans, and maize flour
(77). The first-generation instruments for the Dumas method were not user friendly. The
volume of nitrogen gas produced by combustion was determined with a manometer. The
advent of easy-to-use and highly accurate combustion nitrogen analyzers (CNAs)
rekindled interest in the Dumas method.
CNAs from various manufacturers work on the same principle. The sample is dropped
into a 950–1050°C furnace, purged free of atmospheric gas, and filled with pure (99+ %)
oxygen. Complete sample combustion leads to CO2, water, SO2, NO2, and N2. The
product gases are cooled and a portion is passed through tubing packed with hot lead
chromate, copper, sodium hydroxide (solid), or phosphorus pentoxide to remove SO2, O2,
CO2, and water, respectively. The NO2 is then reduced to N2 and measured with a thermal
conductivity detector (TCD). Sample protein content is calculated by taking into account
the mass of sample injected, the proportion of the combustion gases analyzed, and the
nitrogen-protein conversion factor (FK). The calculations are now automated.
1994 and 1996, respectively (91–93). Trials for the combustion method usually follow
guidelines described by Youden and Sleiner (105):
Kjeldahl method, quantitative amino acid analysis and combustion analysis 29
1. The number of laboratories ranges from 7 to 12. Studies involving as few as three
laboratories have been reported.
2. All studies compare CNAs with Kjeldahl analysis.
3. Interlaboratory studies focus on a single food group. Therefore, CNAs tend to receive
approval for one food group at a time (Table 12).
4. Trials usually test a “generic combustion method” and are independent of the choice of
instruments.
Minimum performance guidelines for CNA instruments include (a) a furnace temperature
of 950°C, (b) a separation system for trapping CO2 and water, (c) a thermal conductivity
detector for nitrogen, (d) sufficient accuracy to produce results within +0.15% of the
mean (% nitrogen) results for 10 successive measurements using a standard compound,
and (e) sufficient precision to produce a relative standard deviation of 0.01%. The LECO
FP-428 analyzer was used by about 80% of the laboratories involved in collaborative
trials. The Foss-Heraue Macro-N analyzer, Carlo Erba NA-5000, and Perkin Elmer
PE2410 also feature. The LECO FP-2000 combustion analyzer appears in the latest trials.
Other suppliesa Kjeldahl and Erlenmeyer flasks, burettes, Tinfoil squares, brushes, tin capsules,
acid, alkali and water dispensers, stirring combustion, reduction and absorption
equipment, large containers for acid, etc. tubes, cotton wool, steel wool, particle
filters, tubing
Food protein analysis 30
Ancillary Ductwork for corrosive fumes, acid- Ductwork for warm airb
equipment resistant fans, fume washer, fans, etc.
size for the LECO-F228 analyzer from 150 mg to 1g by palletizing before loading into
the instrument port. Adding a few drops of polyethylene (2% w/w in ethyl acetate
solvent) prevented flaking of the pellets. Table 15 shows nitrogen and protein data for
feeds determined using the CNA (78,79). Results have been averaged for samples of
sizes 0.15–1.0 g. Protein values are calculated as %N×6.25. The results agree favorably
with Kjeldahl analysis (Fig. 6). There appeared to be significant positive bias for
feedstuffs having <2% nitrogen (Fig. 7). The bias was ascribed to plant-derived materials
containing high levels of nitrate. The Kjeldahl method achieves low recoveries of
nitrogen from refractory compounds (75) with N—O or N—N bonds (nitrite; nitrate;
oximes; azo, nitro, nitroso-compounds).
Sachen and Thiex (81) found that CNAs showed a ≤1.38% (protein) bias for hay
samples. They attributed such results to “atmospheric error” arising from air being
trapped in the interstices of the (fluffy) hay samples. Compressing samples to remove
trapped air led to agreement between the Kjeldahl and CNA results (81). The LECO FP-
2000 nitrogen analyzer was not subject to an atmospheric blank because of improvements
in instrument design and efficient purging of atmospheric gases before sample
combustion. Sachen and Thiex examined a range of pelleting equipment and procedures
for eliminating the atmospheric blank for the LECO FP-428 instrument. They proposed
that powdered cellulose could be analyzed to check for an atmospheric error (81). Not all
investigators agree about the nature of the atmospheric error.
TABLE 15 Nitrogen and Protein in Feedstuffs
Determined by Combustion Method
Sample N (g kg−1) Protein (%)
Straw 0.620 3.88
Corn silage 1.175 7.34
Porc soup 1.185 7.41
Hay 1.280 8.00
Corn grain 1.375 8.59
Barley 1.515 9.47
Grass silage 1.835 11.47
Oats 1.760 11.00
Triticale 1.835 11.47
Wheat 1.985 12.41
Dried grass 2.240 14.00
Cow premix 2.755 17.22
Alfafa pellets 2.770 17.31
Hog feed 3.390 21.19
Broiler finisher 3.420 21.38
Food protein analysis 32
CNAs were approved for beer analysis in Europe in 1999 (84,85). In the collaborative
trial organized by the IOB and EBC, five samples of beer and malt were analyzed in
duplicate by 18 laboratories from the brewing and allied industries. The collaborators
used the following CNAs: LECO FP-428 instrument (11 laboratories), LECO FP-2000 (3
laboratories), and Macro-N analyzer (3 laboratories). Glycine, Tris, or EDTA was used as
the calibrant. Samples of beer were found to contain 362–1159 mg (nitrogen) L−1 and
malt samples had 0.534–0.706% nitrogen. Repeatability and reproducibility statistics for
beer analysis were deemed satisfactory, leading to method approval.
5.7. Meat
King-Brink and Sebranek (100) described a 12-laboratory trial to evaluate CNAs for meat
product analysis. Participants in the trial used the LECO FP-428 instrument (9
laboratories), the Foss Heraeus Macro-N analyzer (2 laboratories), or the Perkin Elmer
PE2410 analyzer. In all, 15 pairs of meat products, purchased from 30 different
manufacturers, were analyzed. All participants used CNAs satisfactorily judging from (a)
the low number of data outliers and (b) the high precision of results for standard
compounds. The CNA results were slightly higher than Kjeldahl figures: 15.75% versus
15.59% (w/w). Estimates for repeatability and reproducibility were comparable. A
recommendation that the Dumas method should be adopted as a reference test for meat
Food protein analysis 36
proteins was approved by the AOAC. A 14-laboratory trial for analysis of meat and meat
products was reported in
TABLE 16 Analysis of Protein in Five Categories
of Baby Foods (97)
Samplea Declared Kjeldahlb Dumasb
protein
1. Formula milk (7) 15.55 14.94 15.36
2. Cereal-based products (6) Cream of rice, semolina+honey, 10.28 10.12 10.35
wheat flour+milk+oats, milk soup+cereal+fruit, milk
soup+cereal+apples
3. Biscuits (2) 9.50 9.40 14.70
4. Lyophilized products (2) Veal, ham, and eggs 54.20 52.15 53.20
5. Homogenized products (6) Beef, beef+ham, chicken, 9.52 9.62 9.72
veal+brain, turkey, chicken+carrot+potatoes
a
Numbers in parentheses represent number of different foods in the category analyzed. Protein
values are averaged for each food category
b
Protein values were calculated with a conversion factor (Fk)=6.38 for milk formula or 6.25 for all
other categories of baby food.
Germany. This trial, too, concluded that the performance of the Dumas method was
comparable to that of the Kjeldahl assay but that the former method was quicker and
more environment friendly (101).
REFERENCES
1. WH Tallent. USA current developments in protein food regulations—labeling. J Oil Chem Soc
56(3):239, 1979.
2. DK Salunkhe, SS Deshpande. Foods of Plant Origin. Production, Technology and Human
Nutrition. New York: Van Nostrand Reinhold, 1991.
3. H Frankel-Conrat, M Cooper. The use of dyes for the determination of acid and basic groups in
proteins. J Biol Chem 154:239–340, 1944.
4. DC Udy. Estimation of protein in wheat and flour by ion-binding. Cereal Chem 33:190–197,
1956.
5. AJ Pinckney. Wheat protein and the biuret reaction. Cereal Chem 26:423–439, 1949.
6. Z Duda, M Szot. A comparison of several methods of protein determination in pig blood plasma.
Proceedings of the European Meeting of Meat Research Workers 32, Vol II 9, 1986, pp 447–
450.
7. KM Williams, P Fox, T Marshall. A comparison of protein assays for the determination of the
protein concentration of beer. J Inst Brew 101:365–369, 1995.
8. Y Pomeranz, RB More, FS Lai. Reliability of five methods for protein determination in barley
and malt. J Am Soc Brew Chem 35:86–93, 1975.
9. Y Pomeranz, RB More. Reliability of several methods for protein determination in wheat. Bakers
Digest February:44 -58, 1975.
Kjeldahl method, quantitative amino acid analysis and combustion analysis 37
10. QQ Luthi-Peng, Z Puhan. The 4th derivative UV spectroscopic method for the rapid
determination of protein and casein in milk. Milchwissenschaft 54:74–77, 1999.
11. JM Lynch, DM Barbano. Kjeldahl nitrogen analysis as a reference methood for protein
determination in dairy products. J Assoc Off Anal Chem Int 82:1389–1398, 1999.
12. RB Bradstreet. The Kjeldahl Method for Organic Nitrogen. London: Academic Press, 1965.
13. International Standard ISO-1871 (1975). Agricultural food products—general directions for the
determination of nitrogen by the Kjeldahl method.
14. L Gaspar. General laboratory methods. In I Kerese, ed. Methods of Protein Analysis.
Chichester, UK: Ellis Horwood, 1984, pp 30–86.
15. BG Osborne. The determination of protein in cereals. In BJF Hudson, ed. Developments in
Food Proteins—4. London: Elsevier Applied Science, 1986, pp 247–290.
16. RA Osborne, JB Wilkie. A study of the Kjeldahl method IV. Metallic catalysts and metallic
interferences. J Assoc Off Anal Chem 18:604–609, 1935.
17. RC Benedict. Determination of nitrogen and protein content of meat and meat products. J Assoc
Off Anal Chem 70:69–76, 1987.
18. R Tkachuk. Nitrogen-to-protein conversion factors for cereal and oilseed meals. Cereal Chem
46:419–423, 1969.
19. OA Krober, SJ Gibbons. Nonprotein nitrogen in soybeans. J Agric Food Chem 10:57–58, 1962.
20. R Tkachuk. Note on the nitrogen-to-protein conversion factor for wheat flour. Cereal Chem
43:223, 1966.
21. JAD Ewart. Amino acid analyses of cereal flour proteins. J Sci Food Agric 18:558, 1967.
22. S Boisen, S Bech-Andersen, B Eggum. A critical view on the conversion factor 6.25 from total
nitrogen to protein. Acta Agric Scand 37:299–304, 1987.
23. MAJS van Boekel, B Ribadeau-Dumas. Addendum to the evaluation of the Kjeldahl factor for
conversion of the nitrogen content of milk and milk products to protein content. Neth Milk
Dairy J 41:281–284, 1987.
24. FW Sosulski, GI Imafidon. Amino acid composition and nitrogen-to-protein conversion factors
for animal and plant foods. J Agric Food Chem 38:1351–1356, 1990.
25. AK Kaul, TR Sharma. Rapid determination of nitrogen content in grain-meal samples with
alkaline-phenol reaction, manually and with an AutoAnalyzer. Z Anal Chem 280:133–138,
1976.
26. M Mohyuddin, G Mazza. Determination of potato protein by alkali-phenol, dye-binding and
other methods. Am Potato J 55:621–626, 1978.
27. BG Venter, HP Sheepers, J Floor, JH Snyman. Semi-micro Kjeldhal procedures for protein
determination of dairy products. S Afr J Dairy Technol 17(3):107–111, 1985.
28. GK Buckee. Determination of total nitrogen in barley, malt and beer by Kjeldahl procedures
and the Dumas combustion method—collaborative trial. J. Inst Brew 100:54–64, 1994.
29. MJ Brennan. Automated Kjeldahl yields poetry. Food Eng 46(5): 102–103, 1974.
30. JE Trevis. Seven automated instruments. Cereal Sci Today 19(5): 182–189, 1974.
31. R Oberrieth, NH Mermelstein. Instrument automates, accelerates nitrogen determinations. Food
Technol 28(6):40–41, 43, 1974.
32. PC Williams, KH Norris, RL Johnsen, K Standing, R Fricioni, D MacAffrey, R Mercier.
Comparison of physicochemical methods for measuring total nitrogen in wheat. Cereal Foods
World 23(9):544–547, 1978.
33. OC Bjarno. Kjel-Foss automatic analysis using an antimony-based catalyst: collaborative study.
J Assoc Off Anal Chem 63:657–663, 1980.
34. DL McGill. Comparison of automated method and improved AOAC Kjeldahl method for
determination of protein in meat and meat products. J Assoc Off Anal Chem 64:29–31, 1980.
35. FB Suhre, PA Corrao, A Glover, AJ Melanoski. Comparison of three methods for
determination of crude protein in meat: collaborative study. J Assoc Off Anal Chem 65:1339–
1345, 1982.
Food protein analysis 38
36. JF Marten, G Catanzaro. Fundamental studies in automatic nitrogen digestion. Analyst 91:42–
47, 1966.
37. JA Varley. Automatic methods for the determination of nitrogen, phosphorus and potassium in
plant material. Analyst 91:119–126, 1966.
38. LL Wall, CW Gehrke. An automated total protein nitrogen method. J Assoc Off Anal Chem
58:1221–1226, 1975.
39. DE Uhl, EB Lancaster. Automation of nitrogen analysis of grain and grain products. Anal
Chem 43:990, 1971.
40. KR Vincent, WF Shipe. The effect of calibration procedures on accuracy and precision of
automated Kjeldahl nitrogen analysis in some formulated foods. J Food Sci 41:157–162, 1976.
41. WM Gantenbein. Collaborative study of the automated determination of nitrogen in meat
products. J Assoc Off Anal Chem 56:31–35, 1973.
42. HG Lento, CE Daugherty. Automated determination of protein nitrogen in foods. Food Prod
Dev 5:86, 1971.
43. HG Lento, CE Daugherty. The automated protein-nitrogen analysis of foods. Adv Auto Anal
Technicon Int Congr 2:75–80, 1970.
44. JF Marten, G Catanzaro. Fundamental studies in automatic nitrogen digestion. Analyst 91:42–
47, 1966.
45. J Davidson, J Mathieson, AW Boyne. The use of automation in determining nitrogen by the
Kjeldahl method, with final calculation by computer. Analyst 95:181–193, 1970.
46. CW Gehrke, LL Wall Sr, JS Absheer. Automated nitrogen method for feeds. J Assoc Off Anal
Chem 56:1096–1105, 1973.
47. LL Wall, CW Gehrke. Feeds: an automated total protein nitrogen method. J Assoc Off Anal
Chem 58:1221–1226, 1975.
48. WT Bolleter, CJ Bushman, PW Tidwell. Spectrophotometric determination of ammonia as
indophenol. Anal Chem 33:592–594, 1961.
49. JA Tettlow, AL Wilson. An absorptiometric method for determining ammonia in boiler feed-
water. Analyst 89:453–465, 1964.
50. CW Gherke, FE Kaiser, JP Ussary. Automated spectrophotometric method for nitrogen in
fertilizers. J Assoc Anal Chem 51:200–211, 1968.
51. PL Searle. The Berthelot or indophenol reaction and its use in the analytical chemistry of
nitrogen. Analyst 109:549–568, 1984.
52. JE McNeal, A Karasz, E Gorge Jr. Automation of methods for meat and meat products. 1.
Determination of protein. J Assoc Anal Chem 53:907–910, 1970.
53. G Mohyuddin, G Mazza. Determination of potato protein by alkali-phenol, dye-binding and
other methods. Am Potato J 55:621–626, 1978.
54. CC Hach, SV Brayton, AB Kopelove. A powerful Kjeldahl nitrogen method using
peroxymonosulfuric acid. J Agric Food Chem 33:1117–1123, 1985.
55. MB Devani, CJ Shishoo, SA Shah, BN Suhagia. Spectrophotometric method for micro-
determination of nitrogen in Kjeldahl digest. J Assoc Off Anal Chem 72:953–956, 1989.
56. S Jacobs. The determination of nitrogen in organic compounds by the indanetrione hydrate
method. Analyst 85:257–264, 1960.
57. S Jacobs. The effect of temperature on the Kjeldahl digestion process. Analyst 89:489–494,
1964.
58. JR Quinn, JGA Boisvert, I Wood. Semi-automated ninhydrin assay of Kjeldahl nitrogen. Anal
Biochem 58:609–614, 1974.
59. ND Heidelbaugh, CS Huber, JF Bednarczyk, MC Smith, PC Rambaut, HO Wheeler.
Comparison of three methods for calculating protein content of foods. J Agric Food Chem
23:611–612, 1975.
60. H-J Horstmann. A precise method for the quantitation of proteins taking into account their
amino acid composition. Anal Biochem 96:130–138, 1979.
61. GL Peterson. Determination of total protein. Methods Enzymol 91:95–119, 1983.
Kjeldahl method, quantitative amino acid analysis and combustion analysis 39
62. JC Weaver, M Kroger, LR Kneebone. Comparative protein studies (Kjeldahl, dye binding,
amino acid analysis) of nine strains of Agaricus bisporus (Lange) imback mushrooms. J Food
Sci 42:364–366, 1977.
63. A Braaksma, DJ Schaap. Protein analysis of the common mushroom Agaricus bisporus.
Postharvest Biol Technol 7:119–127, 1996.
64. CG Zarkadas, EA Meighen, GC Zarkadas, CN Karatzas, AD Khalili, JA Rochemont, M
Berthelet. Determination of the myofibrillar and connective tissue proteins in the bovine
diaphragm. J Agric Food Chem 36:1105–1108, 1988.
65. CG Zarkadas, N Karatzas, AD Khalili, S Khanizadeh, G Morin. Quantitative determination of
the myofibrillar proteins and connective tissue content in selected porcine skeletal muscles. J
Agric Food Chem 36:1131–1146, 1988.
66. CN Karatzas, CG Zakardas. Determination of the myofibrillar and connective tissue protein
contents and amino acid composition of selected composite meat products. J Agric Food Chem
36:1109–1121, 1988.
67. CG Zarkadas, NJ Drouliscos, CN Karatzas. Comparison of the total protein, nitrogen and amino
acid composition of selected additives and ingredients used in composite meat products. J Agric
Food Chem 36:1121–1131, 1988.
68. Q Nguyen, MA Fanous, LH Kamma, AD Khalili, PH Schuepp, CG Zarkadas. Comparison of
the amino acid composition of two commercial porcine skins (rind). J Agric Food Chem
34:565–572, 1989.
69. CN Karatzas, CG Zarkadas. Comparison of the amino acid composition of the intracellular and
extracellular matrix protein fractions isolated from avian skeletal muscles. Poult Sci 68:811–
824, 1989.
70. S Khanizadeh, D Buszar, CG Zarkadas. Comparison of three methods for calculating protein
content in developing apple flower buds. J Assoc Off Anal Chem 75:734–737, 1992.
71. CG Zarkadas, Y Ziran, GC Zarkadas, A Minero-Amador. Assessment of the protein quality of
beefstock bone isolates for use as an ingredient in meat and poultry products. J Agric Food
Chem 43:77–83, 1995.
72. CG Zarkadas, HD Voldeng, YI Ziran, Keijin-Shang, PL Pattsion, Comparison of the protein
quality of five new northern adapted natto soybean cultivars by amino acid analysis. J Agric
Food Chem 45:2013–2019, 1997.
73. S Khanizadeh, D Buszard, CG Zarkadas. Misuse of the Kjeldahl method for estimating protein
content in plant tissue. HortScience 30:1341–1342, 1995.
74. R Fiedler, G Proksch, A Koepf. The determination of total nitrogen in plant materials with an
automatic nitrogen analyzer. Anal Chim Acta 63:435–443, 1973.
75. DW Nelson, LL Sommers. Total nitrogen analysis of soil and plant tissue. J Assoc Off Anal
Chem 63:770–778, 1980.
76. YS Lee, CE Damon, JP Crisler. A micro method for the determination of protein and screening
of added water in meat and meat products. Mikrochim Acta 4:477–483, 1972.
77. TL Lunder. Determination of total nitrogen in foodstuffs according to Dumas, by means of the
Micro-Rapid N automatic analyzer. Lab Pract 23(4): 170–172, 1974.
78. RA Sweeney, PR Rexroad. Comparison of LECO FP-228 “Nitrogen Determinator” with
AOAC copper catalyst Kjeldahl method for crude protein. J Assoc Off Anal Chem 70:1028–
1030, 1987.
79. BM Schmitter, R Rihs. Evaluation of a macrocombustion method for total nitrogen
determination in feedstuffs. J Agric Food Chem 37:992–994, 1989.
80. RA Sweeney. Generic combustion method for determination of crude protein in feeds:
collaborative study. J Assoc Off Anal Chem 72:770–774, 1989.
81. RW Sachen, NJ Thiex. Effect of sample introduction and atmospheric blank on determination
of nitrogen (crude protein) by combustion. J Assoc Off Anal Chem Int 80:14–19, 1997.
82. DF Tate. Determination of nitrogen in fertilizer by combustion: collaborative study. J Assoc
Off Anal Chem 77:829–839, 1994.
Food protein analysis 40
83. American Society of Brewing Chemists. Report of Sub-Committee on Nitrogen in wort and
beer by combustion method. J Am Soc Brew Chem 51(4):183–185, 1993.
84. BA Johnson, CG Johansson. Determination of total soluble nitrogen content of malt and beer by
the Dumas combustion method: collaborative trial. J Inst Brew 105:360–364, 1999.
85. BA Johnson, CG Johansson. Determination of total soluble nitrogen of malt and beer by the
Dumas combustion method: collaborative trial. Monatsschr Brauwiss 53(3/4):50–53, 2000.
86. American Society of Brewing Chemists. Report of Sub-Committee on Total Nitrogen in
brewing grains by combustion method. J Am Soc Brew Chem 50:147–148, 1992.
87. GK Buckee. Review of methods for the measurement of total nitrogen. Ferment 8:357–361,
1995.
88. L Krotz, L Ragalia, F Adreolini. Nitrogen/protein determination in beer, wort and malt by
combustion method as the alternative to Kjeldahl method. Current status and future trends.
Proceedings of EURO FOOD CHEM VIII, Vol 3, Vienna, September 18–20, 1995, p 752.
89. CG Johansson. Determination of total nitrogen in barley and malt by combustion method.
Collaborative trial. Monatsschr Brauwiss 49(11/12):326–328, 1996.
90. SAGF Angelino, HPM van Laarhoven, JJM van Westerop, BM Broekhuijse, HCM Mocking.
Total nitrogen content in single kernel malting barley samples. J Inst Brew 103(l):41–46, 1997.
91. RC Bicsak. Comparison of Kjeldahl method for determination of crude protein in cereal grains
and oil seeds with generic combustion method: collaborative study. J Assoc Off Anal Chem
76:780–786, 1993.
92. R Bicsak. FGIS method of measuring protein. Bull Assoc Oper Millers August:6603–6606,
1995.
93. P Williams, D Sobering, J Antoniszyn. Protein testing methods at the Canadian Grain
Commission. Proceedings of the Wheat Protein Symposium, Saskatoon, Saskatchewan, March 9
and 10, 1998 [conference paper online], October 19, 1998. World Web Web citation. Available
from:http://%20www.cgc.ca/Pubs/confpaper/Williams/ProteinOct98/proteinl-e.htm.
94. PG Wiles, IK Gray. A collaborative trial for the establishment of a skim milk powder reference
protein standard. Aust J Dairy Technol 51:17–21, 1996.
95. PG Wiles, IK Gray, RC Kissling. Routine analysis of proteins by Kjeldahl and Dumas methods:
review and interlaboratory study using dairy products. J Assoc Off Anal Chem Int 81:620–632,
1998.
96. EH Simonne, AH Simonne, RR Eitenmeiller, HA Mills, CP Creswell. Could the Dumas method
replace the Kjeldahl digestion for nitrogen and crude protein determinations in food? IFT
Annual Meeting, 1995, p 241.
97. EH Simonne, AH Simonne, RR Eitenmeiller, HA Mills, CP Cresman III. Could the Dumas
method replace the Kjeldahl digestion for nitrogen and crude protein determinations in foods? J
Sci Food Agric 73:39–45, 1997.
98. CJ Huang, R McDonald, RS Lyon, E Elkins. Comparison of Kjeldahl method for crude protein
determination in fruit and vegetable products with combustion method. IFT Annual Meeting,
1995, p 198.
99. G Bellemonte, A Costantini, S Giammorioli. Comparison of modified automatic Dumas method
and the traditional Kjeldahl method for nitrogen determination in infant food. J Assoc Off Anal
Chem 70:227–229, 1987.
100. M King-Brink, JG Sebranek. Combustion method for determination of crude protein in meat
and meat products: collaborative study. J Assoc Off Anal Chem 76:787–793, 1993.
101. R Buschmann, K Westphal. Determination of the nitrogen content of meat and meat products
with the Dumas method. Report from the Paragraph 35 LMBG working group
‘Fleischerzeugnisse’ of the BgVV. Fleischwritschaft 81:82–84,2001.
102. JK Duan, DR DeClercq. Comparison of combustion and Kjeldahl methods for determination
of nitrogen in oilseeds. J Am Oil Chem Soc 71:1069–1072, 1994.
Kjeldahl method, quantitative amino acid analysis and combustion analysis 41
103. DL Berner, J Brown. Protein nitrogen combustion method collaborative study. I. Comparison
with Smalley total Kjeldahl nitrogen and combustion results. J Am Oil Chem Soc 71:1291–
1293, 1994.
104. MW Young, DKL Mackerron, HV Davies. Calibration of near infrared reflectance
spectroscopy to estimate nitrogen concentration in potato tissues. Potato Res 40:215–220, 1997.
105. WJ Youden, EH Sliener. Statistical Manual of the AOAC. Arlington, VA:AOAC, 1975.
106. Anonymous. Protein: the Dumas technique. Food Rev 21(4):39, 1994.
107. PR Wilson, A new instrumental method in food protein analysis. Food Tech Int Eur 181–183,
1994.
108. JRN Taylor. Out with the old and in with the older. Food Ind 48(5):35, 37, 1995.
109. R Marsili. Don’t wait—automate: revamping labs yields speedy results. Food Prod Design
7(3):71–72, 74, 79–80, 83–84, 1997.
110. ME Ebeling. The Dumas method for nitrogen in feeds. J Assoc Off Anal Chem 51:766–770,
1968.
111. R Winker, S Botterbrodt, E Rabe, MG Linhauer. Nitrogen/protein determination in wheat and
wheat products (flour and meal) with the Dumas method. Getreide Mehl Brot 54(2):86–91,
2000.
112. MG D’Egidio, C Cecchini, G Novembre. Comparison of Dumas and Kjeldahl methods for
determination of crude protein in cereal. Tec Molitoria 50:1189–1195.
References
87. http://www.fda.gov/oc/opacom/hottopics/bse.html#3anchor.
were both specific for Gly m Bd 30K. Sandwich ELISA for Gly
m Bd 30K employed
with 1% w/v SDS and 20mM 2-ME). The extract was centrifuged
and SDS was precipitated with
Kori-dofu 64 +ve
Kinako 29 +ve
Abura-age 59 +ve
Yuba 66 +ve
Miso np b
Shoyu np
Natto np
Meatballs 17 +ve
Fish sausages np
Hamburger np
* Soy flour was extracted with cold Tris-HCl buffer (10 mM,
pH 8, with 1% w/v SDS and 10mM
and also Warner (85) deal with the clinical and public
health aspects of peanut allergy.
etc.) are not present. Johns and Jones (86) divided the
peanut globulins into two groups,
26.0, 0
36 20–30
and Esen (91) found that treating peanut protein with SDS
and 2-ME had no effect on the
follows:
allergen from dark chocolate was only 2–3% when using the
manufacturer’s extraction
staple diet
nontropical sprue.
Sulfur rich
β-Gliadin
α-Gliadin
architectures. B. Gluten
electrophoresis.
Biopharm test. The LLD values of the two tests were 0.04
and 0.1% mg %, respectively. 4.6. Competitive ELISA for
Gluten Using pAb
Skerritt and Smith (133) and Skerritt (142) found two mAbs
(mAb-401/21 and mAb
Infirmary (Bristol).
and rye prolamin was higher than obtained for wheat. The
response toward avenin was
The Skerritt and Hill (156) home test kit for gluten
performed well with ordinary citizens,
gluten, and 18% recovery from wheat flour with 11% gluten.
Home testing for gluten involved five simple steps: (a) add
0.8mL of reaction buffer
(30 µL) of food extract and mix gently for 30 seconds, (c)
add three drops of HRP-mAb
Gluten-negative foods
Tropical cereals
Millet, sorghum, rice, jawar Negative
Legumes
Miscellaneous
beet
Cooked foods a
Uncooked products
Wheat-free products
Cooked products
Uncooked products
Flours
Buckwheat b 4
Sorghum b 13
Maize b 46
Rice b 47–57
Oats b 100–240
Barley 400
Rye 580