Purine nucleotide catabolism in rat liver. Certain preliminary aspects of uricase reaction

Article abstract of DOI:10.1080/15257770902865381

We investigated the mechanism of action of uricase, which oxidizes uric acid to allantoin, in the rat. Allantoin may decompose chemically to urea and hydantoin, containing the carbons in positions 2 and 8 of the purine ring, respectively. These carbons ar

Full text of DOI:10.1080/15257770902865381

Nucleosides, Nucleotides and Nucleic Acids, 28:193–203, 2009
Copyright ꢀC Taylor & Francis Group, LLC
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770902865381
PURINE NUCLEOTIDE CATABOLISM IN RAT LIVER. CERTAIN
PRELIMINARY ASPECTS OF URICASE REACTION
Lucia Terzuoli,1 Brunetta Porcelli,1 Fabio Ponticelli,2 and Enrico Marinello1
1Dipartimento di Medicina Interna, Scienze Endocrino-Metaboliche e Biochimica,
Sezione di Biochimica, University of Siena, 53100, Siena, Italy
2Dipartimento di Chimica, University of Siena, 53100, Siena, Italy
We investigated the mechanism of action of uricase, which oxidizes uric acid to allantoin, in the
2
rat. Allantoin may decompose chemically to urea and hydantoin, containing the carbons in positions
2 and 8 of the purine ring, respectively. These carbons are derived by formylation, catalyzed by
formyltransferase, in two reactions of de novo synthesis. Since uric acid and allantoin are represented
in equivalent amounts in the liver, we expected to find identical incorporation of radioactivity in
C2 and C8 of both compounds after administration of 14C-formate. In the case of 14C-allantoin, this
was true, but not for 14C-uric acid extracted from rat liver. We interpret these results through a series
of experiments and considerations.
Keywords Allantoin; uric acid; uricase
INTRODUCTION
The catabolism and synthesis of purine nucleotides has been exten-
sively studied by many authors.[1–5] Through a series of very well known
steps (dephosphorylation, deamination and phosphorolytic splitting), free
bases, adenine and guanine, are formed. Deamination of adenine leads to
hypoxanthine.[6,7] Hypoxanthine is the key compound for the final steps
of nucleotide breakdown, which occurs through the sequence: hypoxan-
thine → xanthine → uric acid → allantoin. Xanthine oxidase regulates
the sequence hypoxanthine → uric acid (UA) and uricase the oxidative
reaction leading to allantoin (ALL). ALL is the terminal compound in most
mammals, whereas humans, arthropod apes and some monkeys lack this
Received 28 February 2008; accepted 4 March 2009.
Address correspondence to Lucia Terzuoli, Dipartimento di Medicina Interna, Scienze Endocrino-
Metaboliche e Biochimica, Sezione di Biochimica, Universita` di Siena, Via A. Moro 2, 53100, Siena, Italy.
E-mail: terzuoli@unisi.it
193

194
L. Terzuoli et al.
enzyme, and as a consequence excrete UA. The purine metabolic pathway
is commonly followed using labeled precursors as markers.[8]
We considered specifically the incorporation of 14C-formate. Since UA
and ALL are represented in equivalent amounts in the liver, we expected
to find similar incorporation of radioactivity in both compounds after
administration of 14C-formate. On the contrary, we found that labeling of
ALL was much higher at short times after administration, and that the
phenomenon was specific to liver.[9]
In the frame work of this research, we investigated the mechanism
of action of the peroxisomal enzyme uricase (E.C. 1.7.3.3.). Uricase
shows different properties in certain species. The mechanism of the urate
oxidase reaction has not been clearly established, but involvement of
radical intermediates has been hypothesized. The most widely accepted
mechanism[10–12] show that the reaction occurs through formation of an
intermediate and that C2 of UA is distributed equally among both hydantoin
and the ureido moiety of ALL, when the urate is degraded enzymatically
in the absence of borate ions. This hypothesis is belied by Bongaerts and
Vogels[13] that propose the maintenance of a high degree of positional and
stereochemical specificity, in the enzymatic oxidation of UA, in all steps.
Kahn and Tripton[14] observed two discrete enzyme-bound intermediates
and we carried out a study using EPR spectroscopy and spin trapping
of radical intermediates and for the first time, the presence of a radical
intermediate in the uricase reaction was proven experimentally.[15] Gabison
et al.[16] characterize the active site of urate oxidase from Aspergillus flavus,
analyzing the x-ray structure. The urate oxidase illustrates an unexpected
plasticity of the active site; it can accommodate not only the well known
cyclic purine NH-CO-NH-CO motif, but also open ring compounds like
ALL. In this complex, all the N and O atoms are involved in strong polar
interactions toward either the protein and/or water relays. The authors
conclude, with two hypotheses, that the uricase reaction lead to an optically
active [S]-allantoin. Now, it can be concluded that the catalytic mechanism
of UA oxidation by urate oxidase is complex. Many studies have been done
and are still on going to understand how this oxidase works without any
cofactor or metal ion.
For a better comprension of this mechanism, in the present study
we investigated certain aspects of the uricase reaction: labeling of C2
and C8 and tautomerization. We followed the formation of labeled urea
and hydantoin from labeled ALL which was purified from liver of rats
previously treated with 14C-formate or 14C8-urate. ALL was also produced
from labeled UA treated in vitro with uricase. In Scheme 1 is represented the
intermediate compound A, which undergoes tautomerization to form two
differently 14C-labeled ALL, which can be split chemically, not enzymatically,
to urea and hydantoin. From the same scheme, it is evident that both labeled
carbonyl groups of urea and hydantoin may derived, respectively, from C2

Uricase Reaction
195
and C8 of the intermediate A. Really, in intermediate A the C of our interest
there is not in eight position, but to indicate that this C derived from C8-UA,
∗
we mark as C8 . In conclusion, the labeling of urea and hydantoin should
∗
reflect the labeling of C2 and C8 of ALL, due to the original labeling of UA
and to the efficiency of tautomerization process.
Our results provide interesting indications on the formation of the
intermediate compound reported in Scheme 1, under different conditions,
and, thus on the mechanism of action of uricase in vitro and in vivo.
MATERIALS AND METHODS
Materials
Allantoin, uric acid, trichloroacetic acid, hydantoin, urea, red phos-
phorus, hydroiodic acid, and uricase from Bacillus fastidious (15 units/mg
protein) were purchased from Sigma Aldrich Corporation (St. Louis, MO,
USA). Urease was obtained from Sclavo (Italy). Anhydrous acetic acid
was obtained from Farmitalia Carlo Erba (Milan, Italy). Methanol (HPLC
grade) was obtained from Baker (Phillisburg, NJ, USA), 14C-formate (54.8
mCi/mmole) from Amersham Pharmacia Biotech Europe (Freiburg, Ger-
many) and 14C8-uric acid (50 mCi/mmole) from American Radiolabeled
Chemicals Inc. (St Louis, MO, USA).
Animals
Male Wistar rats, weighing 250 g, previously kept on a standard diet, were
used. All animals were fasted for 12 hours before sacrifice. They were housed
three to four animals per clear plastic cages in a room kept at a constant
temperature (24 1◦C) and humidity (45% 5%) under the illumination
of a 12:12-hour light/dark cycle. Food and water were available ad libitum.
Treatment with 14C-formate or 14C8-UA
14C-formate or 14C8-uric acid precursor were injected intraperitoneally
at a dose of 10 µCi/100 g b.w. Rats were killed 15ꢁ after administration. The
livers were rapidly excised, washed in normal saline, homogenized in 5%
trichloacetic acid (TCA) and centrifuged to obtain the supernatant.
Purification of UA and ALL
14C–UA and 14C-ALL were purified from supernatant by our previously
reported procedure.[17] UA was determined by the enzymic procedure

196
L. Terzuoli et al.
uric acid
O
H
N
HN
O
8
2
N
H
O
N
H
uricase
CO2H
CO2H
H
N
H
N
H
N
H
N
O
2
O
O
O
2
8*
8*
N
H
N
N
N
H
intermediate compound A
O
O
HN
NH
NH2
NH2
2
8*
8*
O
O
2
N
H
N
N
O
O
N
H
H
H
allantoin
(chemical reaction)
O
O
NH2
8*
NH2
HN
NH
2
2
8*
O
O
H2N
O
O
NH2
N
N
H
H
SEQUENCE 2
SEQUENCE 1
SCHEME 1 Uric acid treatment with uricase, formation of allantoin and subsequent chemical reaction
to obtain urea and hydantoin.

Uricase Reaction
197
of Praetorious and Poulsen.[18] ALL was assayed by the Rimini-Schryver
reaction.[19] The procedure is summarized in Scheme 2.
5% TCA extract (1 g tissue + 5 ml 5% TCA)
acid hydrolysis of nucleotides
(1 M H2SO4, 1 h at 100 °C)
Hg-acetate precipitation of free bases, UA and ALL
Supernatant
Removal of Hg as Hgs
Dowex Ag 50W-X8 column
free bases (adenine,
guanine, xanthine,
(cationic exchange chromatography)
hypoxanthine)
Dowex 1-X8
(anionic exchange chromatography)
uricase#
*UA
*ALL
Hydroiodic acid
and red phosphorous
Mixture: ALL+ hydantoin + urea
Separation by HPLC
*Urea
*Hydantoin
#UA was transformed into ALL by treatment:
-
-
with uricase from Bacillus Fastidiosus
with uricase from rat liver peroxysomes
SCHEME 2 Purification of uric acid and allantoin obtained of rat liver and chemical degradation of
allantoin.
Preparation of Rat Liver Peroxysomes
We obtained rat liver uricase from crude tissue preparation. The rat
liver was excised and minced in homogenization buffer (5 mM Mops pH
7.4, 250 mM sucrose, 1 mM EDTA, 0.1% ethanol). Five grams of liver was
homogenized for 2 minutes at 1000 rpm in 20 ml of ice-cold buffer, hold on
ice, using a Potter-Elvehejm homogenizer. The homogenate was centrifuged

198
L. Terzuoli et al.
at 2000 g for 10 minutes and the precipitate resuspended with phosphate
buffer pH 7.5. After dialysis against double-distilled water at 0◦C, the final
suspension (suspension A), containing the peroxysomes, was used for UA
treatment.
Treatment of UA with Uricase from Bacillus fastidiosus or Rat
Liver Peroxisomes
To a solution containing 0.05 µmol/ml UA we added: 0.1 ml Bacillus
fastidiosus uricase and 0.02 ml catalase; and 1 ml fresh suspension A
containing liver peroxysomes.
The mixture was incubated at 37◦C for 1 hour. TCA 12% was added up
to pH 2.0 and the mixture centrifuged at 10000 g for 15 minutes. ALL was
separated from UA by anionic exchange chromatography (Dowex 1-X8).
Chemical Degradation of ALL to Hydantoin and Urea
The chemical degradation of ALL to urea and hydantoin was carried out
as previously described[20] and shown in Scheme 2. Urea was determined by
a colorimetric enzyme method[21] and hydantoin by HPLC.
Mass Spectrum of Allantoin
ALL obtained from rat liver or from UA transformed into ALL by Bacil-
lus fastidiosus or rat liver peroxysomes was purified by HPLC and derivatized
with N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide (MTBSTFA) in
pyridine at 90◦C for 24 hours.[22] The compound was then analyzed by
ion trap mass spectrometry (ITMS, Finnigan, St. Jose, CA, USA). Collision
experiments were performed selecting parent ions by the two-step proce-
dure, at a q2 of 0.3. A supplementary AC “tickle” voltage applied to the two
endcaps was used to excite the ions. Typical tickle voltage and tickle time
values were in the range 150–200 mV and 15–35 msec, respectively.
Determination of Radioactivity
This was carried out on a Nuclear Chicago Delta Scintillation Counter,
using Instagel as liquid scintillation cocktail.
RESULTS
a) In a first set of experiments we focused attention on 14C8-uric acid
obtained from a commercial source, treating it first with uricase from
Bacillus fastidiosus, and then chemically degrading it. The procedure

Uricase Reaction
199
TABLE 1 Specific radioactivity of urea and hydantoin obtained by commercial 14C8-uric acid
trasformed in 14C8-allantoin by uricase of Bacillus Fastidiosus or rat liver peroxisomes
Specific radioactivity (dpm/µmoles)
Experiments
Uric acid
Allantoin
Urea
% C2
Hydantoin
% C8
Bacillus Fastidiosus uricase
1
2
3
4
13444954
17555823
15168738
21706900
13527536
17767376
15222935
20170929
3000004
7102012
6283540
6799924
22
40
41
37
10869375
10479931
9067449
78
60
59
63
13076621
Rat peroxisomal uricase
1
2
10324472
13235831
10169134
13052459
16222
11584
0.1
0.1
10191500
12901255
99.9
99.9
showed that urea and hydantoin were both labeled, though hydantoin
accumulated more 14C than urea. When 14C8-uric acid was first degraded
by uricase from peroxysomes, and then underwent further chemical
degradation, only hydantoin was strongly labeled, as predictable from
Sequence 1 Scheme 1 (Table 1). Experiments in which 14C8-uric acid
obtained from rat liver after injection of commercial 14C8-uric acid was
treated with uricase of Bacillus fastidiosus or rat liver peroxysomes after
injection of commercial 14C8-uric acid, are reported in Table 2. It was
evident that hydantoin was more strongly labeled than urea, indeed
about 100% after treatment with peroxysomal enzyme.
b) Treatment of 14C-UA obtained from rat liver after administration of
14C-formate are reported in Table 3. 14C-UA was then transformed
into 14C-ALL by uricase of Bacillus fastidiosus or rat liver peroxysomes.
Different results were obtained in the two experiments: in the case of
Bacillus fastidiosus enzyme high variability was evident, though labeling
TABLE 2 Specific radioactivity of urea and hydantoin obtained by liver 14C8-uric acida trasformed
in 14C8-allantoin by uricase of Bacillus Fastidiosus or rat liver peroxisomes
Specific radioactivity (dpm/µmoles)
Experiments
Uric acid
Allantoin
Urea
% C2
Hydantoin
% C8
Bacillus fastidiosus uricase
1
2
227441
335339
228833
329005
52016
58940
23
18
174142
268508
77
82
Rat peroxisomal uricase
1
2
301566
279561
300798
280016
20999
17587
7
6
278652
264338
93
94
a14C8-uric acid was obtained from rat treated with commercial 14C8-uric acid.

200
L. Terzuoli et al.
TABLE 3 Specific radioactivity of urea and hydantoin obtained by liver 14C-uric acida trasformed in
14C-allantoin by uricase of Bacillus Fastidiosus or rat liver peroxisomes
Specific radioactivity (dpm/µmoles)
Experiments
Uric acid
Allantoin
Urea
% C2
Hydantoin
% C8
Bacillus Fastidiosus uricase
1
2
3
4
5
6
7
3300
3520
3146
2820
7,026
10100
2180
3471
3543
2981
2701
7,052
10463
2237
2875
2760
2017
915
4296
4977
449
78
73
67
34
61
48
20
806
1011
993
1800
2683
5462
1745
22
27
33
66
39
52
80
Rat liver peroxisomal uricase
1
2
10600
15425
9580
15932
8381
12694
85
79
1479
3375
15
21
a14C-uric acid was obtained from rat treated with 14C-formate.
was prevalently of urea. In the case of treatment with rat peroxysomal
enzyme, urea was labeled 80%.
c) When we considered the formation of 14C-urea and 14C-hydantoin from
14C-ALL, purified from the livers of rats treated with 14C-formate, the
results were in perfect agreement from they and showed that rat liver
∗
allantoin, obtained in this way, contained equally labeled C2 and C8
(Table 4).
In determining whether ALL molecules were identical, irrespective of the
source of uricase (from Bacillus fastidiosus or rat liver), we purified ALL by
HPLC and derivatized it for analysis by mass spectrometry. In all cases we
confirmed that the molecules of ALL obtained are identical.
TABLE 4 Specific radioactivity of urea and hydantoin obtained by liver
14C-allantoina
Specific radioactivity (dpm/µmoles)
Experiments
Allantoin
Urea
% C2
Hydantoin
% C8
1
2
3
4
5
23000
24293
17000
15250
28999
10250
12416
8030
7062
14876
49
50
49
48
51
10750
12525
8406
7620
14174
51
50
51
52
49
a14C-allantoin was obtained from rat treated with 14C-formate.

Uricase Reaction
201
DISCUSSION
a) Experiment In Vitro Using 14C8-Uric Acid From a Commercial
Source or Obtained After Injection In Vivo
In this case, when UA was treated with uricase from Bacillus fastidiosus,
urea and hydantoin were both differently labeled. This may be due to
variable tautomerization degree of the intermediate compound probably
occurring in the medium not to the enzyme. When we use the peroxysomal
enzyme, the result was cleared: most radioactivity appeared in hydantoin. In
this case, the rate of tautomerization is slower than the rate of formation of
ALL, leading to only 14C-labeled hydantoin.
The conclusion is that the bacterial enzyme can only be used to obtain
∗
approximate indications, leading to mixing up of C2/C8 labeling, due to
an efficient tautomerization (Scheme 1, Sequence 1 and 2), whereas that
of peroxysomes is very specific and acts as shown in Scheme 1, Sequence 1,
∗
according to which hydantoin is derived from C8 .
b) Experiment in Which of 14C-formate was Administered
and 14C-UA was Obtained From Rat Liver
In this case when UA is treated with uricase, a clearly prevalent labeling
of urea is obtained, especially when peroxysomal enzyme is used. In line
with the above conclusion, that the mechanism of action of peroxysomal
enzyme in according to Sequence 1 of Scheme 1, we may conclude that in
Experiment B UA was labeled at C2, which is the source of urea.
c) Experiment in Which of 14C-formate was Administered
and 14C-ALL was Obtained From Rat Liver
ALL obtained in vivo after 14C-formate injection was equally labeled at
∗
C2 and C8 , which can be explained by assuming that it is derived from
∗
UA containing equally C2 and C8 labeled, or that it is formed by perfectly
symmetrical tautomerization (50%).
The results of Experiments B and C seem in disagreement, because ALL
is derived from UA, which is formed during de novo nucleotide synthesis
and catabolism. However, during de novo synthesis of the purine ring, from
which UA and ALL are derived, formylation of the two carbons in the well
known reactions of formation of 5ꢁ-phosphoribosyl-N-formylglycinammide
and 5ꢁ-phosphoribosyl-4-carboxamide-5-formamidoimidazole, catalyzed by
formyltransferase, produces two carbon atoms which could be labeled at
the same extent. If we consider our results for 14C-ALL, this seems true, but
it is not true for 14C-UA extracted from rat liver: unlike ALL, all UA is not
equally labeled at carbons 2 and 8∗.

202
L. Terzuoli et al.
Presumably, there are two possible explanations for this. A reasonable
interpretation is that not all UA formed in vivo, in different cells and
different cell compartments is equally labeled at carbons 2 and 8. This can
only occur in the presence of saturating pools of formyltetrahydrofolate,
which are not necessarily distributed in all tissues, and can be found only
in some cells and cell districts, possibly only in liver cells. Only this uric
acid is transformed into allantoin. It does not accumulate, but is attacked
very rapidly by liver peroxysomes, producing allantoin, equally labeled at
Carbons 2 and 8∗.
Another pool of UA, present in liver, or derived from other tissues,
is labeled at C2, it is not attacked by liver peroxysomal uricase, and is
eliminated as such in urine.
Another interpretation of our results could be that in vivo most UA
is labeled prevalently at Carbon 2, but the action of peroxysomal uricase,
is different from that in our experimental preparations. The structure,
linkages and organization of the enzyme in cells are different from those of
free detached peroxysomes and have a different mechanism of action. They
trigger a very fast tautomerization, producing both tautomers in equivalent
amounts. This mean that labeling of Carbon 2 and Carbon 8 in the ALL
formed.
To confirm our hypotheses additional experiments should be per-
formed, also including the use of rat recombinant enzyme.
REFERENCES
1. Carlucci, F.; Rosi, F.; Di Pietro, M.C.; Marinello, E.; Pizzichini, M.; Tabucchi, A. Purine nucleotide
metabolism: specific aspects in chronic lymphocytic leukemia lymphocytes. Biochim. Biophys. Acta
1997, 1360, 203–210.
2. Carlucci, F.; Tabucchi, A.; Pagani, R.; Marinello, E. Synthesis of adenine and guanine nucleotides at
the ‘‘inosinic branch point’’ in lymphocytes of leukemia patients. Biochim. Biophys. Acta 1999, 1454,
106–114.
3. Christjanson, L.J.; Middlemiss, P.J.; Rathbone, M.P. Stimulation of astrocyte proliferation by purine
and pyrimidine nucleotides and nucleosides. Glia 1993, 7, 176–182.
4. Hershfield, M.S.; Mitchell, B.S. Immunodeficiency diseases caused by adenosine deaminase de-
ficiency and purine nucleoside phosphorylase deficiency, in The Metabolic and Molecular Basis of
Inherited Disease, eds. C.R. Scriver, A.L. Beauded, W.S. Sly, D. Valle, McGraw-Hill Inc, New York. 1995,
pp. 1725–1768.
5. Moriwaki, Y.; Yamamoto, T.; Higashino, K. Enzymes involved in purine metabolism—A review
of histochemical localization and functional implications. Histology and Histopathology 1999, 14,
1321–1340.
6. Zimmermann, H. 5ꢁ-nucleotidase: molecular structure and functional aspects. Biochem. J. 1992, 285,
345–365.
7. Zimmermann, H. Biochemistry, localization and functional roles of ecto-nucleotidases in the
nervous system. Prog. Neurobiol. 1996, 49, 589–618.
8. Zoref-Shani, E.; Kessler-Iceksonm, G.; Wassermanm, L.; Sperling, O. Characterization of purine
nucleotide metabolism in primary rat cardiomyocyte cultures. Biochim. Biophys. Acta 1984, 804,
161–168.
9. Pizzichini, M.; Pandolfi, M.L.; Arezzini, L.; Terzuoli, L.; Fe`, L.; Bontemps, F.; Van Den Berghe, G.;
Marinello, E. Labelling of uric acid and allantoin in different purine organs and urine of rat. Life
Sci. 1996, 59, 893–899.

Uricase Reaction
203
10. Canellakis, E.S.; Cohen, P.P. The end-products and intermediates of uric acid oxidation by uricase.
J. Biol. Chem. 1955, 213, 385–395.
11. Mahler, H.R. Uricase, in The enzymes eds. P. Boyer, H. Lardy, and K. Myrback, Academic Press, New
York, 1963. 3rd ed., vol. 8, pp. 285–296.
12. Pitts, O.M.; Priest, D.G. Uricase reaction intermediate. Mechanism of borate and hydroxide ion
catalysis. Biochemistry 1973, 12, 1358–1363.
13. Bongaerts, G.P.A.; Vogels, G.D. Mechanism of uricase action. Biochim. Biophys. Acta 1979, 567,
295–308.
14. Kahn, K.; Tripton, P.A. Spectroscopic characterization of intermediates in urate oxidase reaction.
Biochemistry 1998, 37, 11651–11659.
15. Busi, E.; Sinicropi, A.; Terzuoli, L.; Marinello, E.; Basosi, R. Identification and structural characteri-
zation of a transient radical species in the uricase reaction mechanism. Appl. Magn. Reson. 2007, 31,
471–482.
16. Gabison, L.; Chiadmi, M.; Colloc’h, N.; Castro, B.; El Hajji, M.; Prage´, T. Recapture of [S]-allantoin,
the product of the two-step degradation of uric acid, by urate oxidase. FEBS Lett. 2006, 580,
2087–2091.
17. Di Stefano, A.; Pizzichini, M.; Leoncini, R.; Vannoni, D.; Pagani, R.; Marinello, E. Quantitative
separation of uric acid and allantoin from rat liver tissue. Biochim. Biophys. Acta 1992, 1117, 1–6.
18. Praetorius, E.; Poulsen, H. Enzymatic determination of uric acid. Scand. J. Clin. Lab. Invest., 1953, 5,
273–280.
19. Young, E.G.; Conway, C.F. On the estimation of allantoin by the Rimini-Schryver reaction. J. Biol.
Chem. 1942, 142, 839.
20. Terzuoli, L.; Pizzichini, M.; Arezzini, L.; Pandolfi, M.L.; Marinello, E.; Pagani, R. Separation and
determination of allantoin, uric acid, hydantoin and urea. J. Chromatogr. A 1994, 684, 350–353.
21. Fawcett, J.K.; Scott, J.E. A rapid and precise method for the determination of urea. J. Clin. Pathol.
1960, 13, 156–159.
22. Hagiwara, T.; Watanabe, Y. Determination of allantoin in Discoreae Rhizoma by capillary gas chro-
matography as tert-butyldimethyl silyl derivatives. Annual Reports Tokyo Metr Research Laboratory 1992,
43, 51–55.