CLIN.CHEM.

24/12, 2176-2178(1978)

Radial DiffusionAssay of NAD- and NADP-Dependent

Oxidoreductases
Jason C. H. Shih and E. P. Teulings

With a radial diffusion

dehydrogenase

of six NAD- and NADP-dependent

(NAD(P)H)
(EC 1.6.4.2); Pseudomonas
testosteroni
(ATCC
11996) hydroxysteroid
dehydrogenase (a mixture of the a- and
13-enzymes); yeast glucose-B-phosphate
dehydrogenase
(EC
1.1.1.49); chicken-heart
lactate dehydrogenase
(EC 1.1.1.27);
and pig-heart malate dehydrogenase (EC 1.1.1.37). If the
enzyme was in crystalline
or lyophilized
form, it was first
dissolved
in the least-necessary
volume of an appropriate
buffer. The solution was then dialyzed at 5 #{176}C
overnight
vs.
the buffer, through which nitrogen was bubbled. If an enzyme
was in the form of a suspension in ammonium
sulfate solution,
it was first dialyzed against the buffer, to remove the salt. The
protein concentrations
of the dialyzed enzymes were determined (8) and, before assay, the enzyme solution was diluted
to give 1.0 mg of protein per milliliter.
A series of twofold
dilutions
were made from this solution.
The livers from three-week-old
chicks were homogenized
with fourfold their volume of potassium
phosphate
buffer (0.1
mol/liter,
pH 7.0) containing,
per liter, 1 mmol of ethylenediaminetetraacetate
(EDTA)
and 5 g of Triton X-100 surfactant,
by a procedure
described
previously
(7). This homogenate
was used as a source of enzymes for the assay.
Ge! plates.
Table 1 summarizes
the substrates
and the
buffers used in the gels for different
enzyme assays. The
substrate
was incorporated
into 100 ml of a molten 10 g/liter
agarose (low electroendosmosis,
Sigma Chemical Co.) solution
at 50-60 #{176}C;
then the gel was poured onto a 20 X 20 cm glass
plate and allowed to solidify. Wells, punched
with 7.0-mm
(o.d.) glass tubing, were filled with 50 il of the enzyme preparation to be analyzed.
A more detailed
procedure
has been
described
previously
(7). Testosterone,
which is water insoluble, was first dissolved in dimethylformamide,
mixed rapidly
into the molten agarose solution, and this mixture
was immediately
poured onto the glass plate.
Enzyme
assay. After the enzyme or the tissue homogenate
was placed into the small wells the gel plate was kept in a dark,
moist chamber at room temperature
(20 #{176}C)
for 24 h to allow
the enzyme to diffuse radially.
At the end of the diffusion
period, the small wells were just filled with molten agarose
solution (10 g/liter). When the filling gel had solidified,
a solution of pyridine nucleotide coenzyme was spread evenly over
the gel. The coenzyme applied could be either the oxidized or
reduced form of NAD or NADP, depending
upon the direction
of the assay reaction and the enzyme to be determined.
After
a certain time (Table 1), the plate was placed under a longwavelength
(360 nm) ultraviolet light (Ultra-Violet
Products,
San Gabriel, Calif. 91778; Black-ray
lamp, Model XX-15)
in a dark room. The change of fluorescence
against the background, indicating
either the enzymic reduction
or oxidation
of the coenzyme,
could be seen. With a caliper we measured
the diameter
ofthe radial diffusion zone that had been made

assay we measured the activity

oxidoreductases:
alcohol, glucose-6-phosphate,
hydroxysteroid,
lactate and
malate dehydrogenases, and giutathione reductase. The

enzyme was allowed to diffuse for 24 h in an agarose gel

in which the substrate was incorporated, then reacted with
the pyridine nucleotide coenzyme. The size of an enzyme
diffusion zone could be made visible by the change of the

fluorescence of the coenzyme against the background

when the coenzyme was either oxidized or reduced. The
procedure for each enzyme is reported. The results indicate that thisnew technique may be applicable to all NADand NADP-dependent enzymes. Because of itssimplicity
and potentiality for screening many samples, we think this
method has applications in the clinical laboratory and in
nutrition
studies.

AddItIonal Keyphrases:

enzymes
pyridine nucleotide
gel
detection from fluores

coenzymes
agarose
cence
screening
.

The radial

diffusion

assay (RDA)

technique

for hydrolytic

enzymes has been known for years. An enzyme diffuses in an

agar gel containing
the substrate and can be quantitated by
measuring
the size of the zone of hydrolysis
(1-6). Recently
this technique
was extended
to assay of an oxidoreductase,
lipoamide dehydrogenase
(NADH) (EC 1.6.4.3), based on the
quenching
of NADH fluorescence
when it is enzymically
oxidized (7). The diffusion
zone of the enzyme was measured
under long-wavelength
ultraviolet
light. Use of the assay of
lipoamide dehydrogenase to assess riboflavin deficiency in
chicks was also demonstrated.
In addition to lipoamide dehydrogenase,
we have applied
this technique to measure
the activity of six NAD- and
NADP-dependent enzymes. Our results indicate that the new

method may have a broad application to all pyridine nucleotide-coenzyme oxidoreductases. Here, we report the individual
procedures and working conditions for assay of each enzyme.

Materials and Methods

Enzyme
coenzymes

Louis,

preparations.
were

Mo. 63178.

The

purchased

The

enzymes,

substrates,

and
Co., St.
were yeast alcohol

from Sigma Chemical

enzymes

studied

Laboratory
of Applied Biochemistry,
Department
of Poultry
Science, North Carolina State University, Raleigh, NC 27650.
Paper number 5642 of the Journal Series of the North Carolina
Agricultural Experiment Station, Raleigh, NC.
Received Aug. 11, 1978; accepted Sept. 25, 1978.

2176

CLINICALCHEMISTRY,Vol. 24, No. 12, 1978

(EC 1.1.1.1); yeast

glutathione

reductase

Table 1. ConditIons for Assay of Various Enzymes

Buffer concn.
Enzyme8

Reaction

(potassium phosphate)

pH

time, mm

7.5
0.1mol/liter,
1 mmol/liter
EDTA
0.1 mol/llter, 1 mmol/liter EDTA
8.0
0.1mol/Iiter
9.0
G6P-DH 0.1 mol/liter TrisHClc
7.5
WH
0.1 mol/liter, 1 mmol/liter/3MEC
7.0
MDH
0.1 mol/liter
7.5
LipDH
0.05 mol/liter, 1 mmol/liter EDTA 6.6
8 For enzyme abbreviations and EC numbers,see legend to Fig.
ADH
GSR
HSDH

15

Substrate

Coenzyme

Ethanol, 50 g/liter

Detection b

NAD, 1 mmol/Iiter

F
Ox. glutathione, 1.25 mmol/liter NADPH, 0.5 mmol/liter
Q
F
Testosterone, 5 mmol/liter
NAD, 1 mmol/liter
10 Glucose-6-P,
2 mmol/liter
NADP, 0.5mmol/liter F
30 Na pyruvate,10 mmol/liter
NADH, 0.5mmol/liter
Q
10
Na oxalacetate,
10 mmol/liter NADH, 0.5mmol/liter
0
NADH, 0.5 mmol/liter
120 Lipoate, 5 mmol/liter
Q
2. b Mode of detection, F: zone of fluorescence, 0:zone of quenched fluorescence.
Note buffer
30
30

other than phosphate used. j3-Mercaptoethanol.

visible in this way. Table 1 summarizes the various reaction

durations, the different forms of pyridine nucleotide coenzymes applied, and the mode of detection of fluorescence or
quenching of fluorescence for the different enzymes.

Results and Discussions

The radial diffusion assay was applied to detect the activities of six NAD- and NADP-dependent
enzymes, in addition to lipoamide dehydrogenase
already described (7). The
activity
oftheseenzymes was measured by measuring
the size

ADH

Fig. 1. Radial diffusion

zones of a series of twofold dilutions of

of their diffusion zones, made visible by the change of fluorescence caused by either the enzymic oxidation of the reduced
pyridine nucleotide coenzyme or the reduction of the oxidized
coenzyme. Table 1 lists optimum conditions we settled on,
such as the concentrations
of various substrates, the forms and
concentrations
of coenzymes, the reaction times, the appropriate
kinds of buffer with additives (i.e., EDTA or 13-mercaptoethanol),
and themodes of detection. This method, with
proper

modifications,

may well be applicable

to all NAD- and

NADP-dependent
oxidoreductases.
We assayed a series of twofold dilutions of each enzyme
(starting with 1.0 g/liter). Figure 1 depicts one such gel plate
for alcohol dehydrogenase. The fluorescence
against a dark
background contrasts to the detection of quenching zones
against a fluorescent background as was the case in our diffusion assay for lipoamide dehydrogenase (7).
As was done for lipoamide dehydrogenase, we established
the linearity of the relationship between diameter of the diffusion zone and logarithm of enzyme concentration for each
enzyme tested (Figure 2). Each point on Figure 2 is an average
value for three determinations,
and the standard
deviation
among the three was less than 5%. The results were consistently reproducible
for each enzyme during many trials. Some
nonlinearity
at the ends of a few curves, malate dehydrogenase, for example, is observed. This may be attributed
to the

yeast alcohol dehydrogenase, starting with 1.0 g of protein per

liter

Table 2. Enzymes Detected in the Gel with

Different Substrate-Coenzyme Systems8

The numbers are dilution factors

Coenzyme

NAD+

NADP+

NAOH

NADPH

ADHb
Tissue
IISDH
ADH
Tissue
(ADH)

(ADH)
(Tissue)
(Tissue)

NT

NT

NT

NT

NT

NT

NT

G6P-DH
(Tissue)
NT

GSR

Pyruvate

NT

NT

Oxalacetate

NT

NT

Fig. 2. Semilogarithmic relation between the diameters of the

Lipoate

NT

NT

diffusion zones and the concentrations

of enzymes
MDH, malate dehydrogenase(EC 1.1.1.37);ADH, alcohol dehydrogenase(EC
1.1.1.1); G6P-DH, glucose-6-phosphate dehydrogenase (EC 1.1.1.49); GSR,
glutathione reductase (1.6.4.2); HSOH,hydroxysterold dehydrogenase (a mixtixe
of EC 1.1.1.51 and EC 1.1.1.50); and LDH, lactate dehydrogenase (EC

LDH
Tissue
MDH
LDH
Tissue
LipDH
Tissue

0. none of the preparations reactive; NT, not tested.

For enzyme abbreviations and EC numbers, see legend to Fig. 2.

Substrate

Ethanol
Testosterone

C.)

w
I-

Glucose-6phosphate
Ox. glutathione

LU

Tissue

0.Oi

PROTEIN

01

1
(mg/mi)

Preparations listed for being reactive; listed In parentheses,

LDH
Tissue
(MDII)
(LDH)
Tissue
0

weakly reactive;

1.1.1.27)

CLINICALCHEMISTRY,Vol. 24, No. 12, 1978 2177

This work is partly supported by the NHLI, NIH, Grant HL21912

denaturation of enzymes in very dilute solutions. The slope

of each curve seems to be a function of relative molecular mass
and other factors influencing the mobility of an enzyme.
To test the substrate and coenzyme specificities, we applied
allthe enzymes toeachdifferent gel plate containing different
substrateand reacted them with different coenzymes. These
resultsare summarized in Table 2, where only the reactive
preparations
are listed
undereach assaycondition.
In general,
fairly
specific
enzyme-substrate--coenzyme
relationships
were
observed,as expected.Some cross reactions
may be the result
of nonspecificity
ofan enzyme,suchas alcoholdehydrogenase,
or of significant
impurityin the enzyme preparation
(e.g.,
lactatedehydrogenase contamination in the malate dehydrogenase preparation; analytical information report from
Sigma).From the tissue homogenate, different enzyme activities can be detected according to the substrate-coenzyme
system used in the gel.
Our method
is about 50-fold less sensitive
than the conventional
spectrophotometric
analyses for these, but this can
be an advantage
because preliminary
sample dilutions can be
obviated.
The application
of this technique
to the enzymic
diagnosis of riboflavin
deficiency
in chicks has recently been
demonstrated
(7). The technique
is simple, inexpensive,
and
seems well suited to use in screening.
Other clinical and
nutritional
applications
of this method are anticipated.

(to J.C.H.S.).

References
1. Sandholm, M., Smith, R. R., Shih, J. C. H., and Scott, M. L., Determination of antitrypsin activity on agar plates: Relationship between antitrypsin and biological value of soybean for trout. J. Nutr.
106, 791 (1976).
2. Fossum, K., Proteolytic enzymes and biological inhibitors. Acta
Pathol. Microbiol. Scand., Sec. B., 78, 350 (1970).
3. Fossum, K., and Whitaker, J. R., Simple method for detecting
amylase inhibitors in biological materials. J. Nutr. 104,930 (1974).
4. Ceska, M., A new approach for quantitative and semi-quantitative
determinations
of enzymatic
activities with simple laboratory
equipment.
Detection of a-amylase.
Clin. Chim. Acta 33, 135
(1971).
5. Goldberg, J. M., and Pagast, P., Evaluation of lipase activity in
serum by radial enzyme diffusion. Clin. Chem. 22,633 (1976).
6. Montenecourt,
B. S., and Eveleigh, D. E., Semiquantitative
plate
assay for determination
of cellulase production by Trichoderma
viride. Appi. Environ. Microbiol. 33, 178 (1977).
7. Shih, J. C. H., Radial diffusion assay of lipoamide dehydrogenase
and its use to assess riboflavin deficiency. Anal. Biochem. 89, 103
(1978).
8. Lowry, 0. B., Rosebrough, N. J., Farr, A. L., and Randall, F. J.,
Protein measurement with the Folin phenol reagent. J. Biol. Chem.
193, 265 (1951).

CLIN. CHEM.24/12, 2178-2179 (1978)

Non-Chromatographic Screening Test for Hyperprolinemia

Yoshihisa Yamaguchi

This procedure for determining serum proline in patients

with hyperprolinemia
involves protein precipitation
(Folin-Wu

method),

color development

with isatin, ex-

traction of the color with methylene chloride, and measurement of its absorbance at 600 nm. The specificity and
analytical recovery show the method to be suitable for this
use.

AddItIonal Keyphrases: isatin chromogen

colorimetry

The reaction between a-amino acids and ninhydrin is one

of the most commonly used methods for detecting and estimating amino acids. Recently, fluorophotometric
assay with
use of o-phthalaldehyde (1) or fluorescamine (2) has become
widely used for sensitive determination of amino acids.
In hyperprolinemia,
an inherited disorder reflecting abnormal metabolism of imino acid, high serum proline conCentral Laboratory for Clinical Investigation, Osaka University
Hospital, 1-1-50 Fukushima-ku,
Osaka, Japan.
Received Aug. 2, 1978; accepted Sept. 11, 1978.
2178

CLINICAL CHEMISTRY,

Vol. 24, No. 12, 1978

centrations are observed, but the methods mentioned are not

suitable for determination of proline. A disadvantage of nmhydrin is the ease with which the relatively weak color produced by proline is masked by that of other amino acids. The
ninhydrin method has been modified for determination of
proline,but ornithine and hydroxyproline produce similar
color(3, 4). The fluorometric assay isalmost insensitive
to
proline (1,2).
Determination of proline by use of isatin was first reported
in 1950 (5); the reagent is highly specific for proline in paper
chromatography.
Here I describe a method for the high concentrations of
proline in the serum of patients with hyperprolinemia,
although it is not suitable for normal concentrations of proline.

Materials and Methods

All reagents were of analytical
Color reagent.
Dissolve 1.5 g
ethanol. After cooling, add 4 ml
Standard
solution.
Prepare

grade.
of isatin

in 100 ml of warm
of acetic acid.
a 100 mg/liter solution of