A new enzymatic assay of urinary guanidinoacetic acid

Clinica Chimica Acta, 202 (1991) 227-236 0 1991 Elsevier Science Publishers B.V. All rights reserved 0009-8981/91/$03.50

221

CCA 05106

A new enzymatic assay of urinary guanidinoacetic acid Yoshio Shirokane, Moto-o Nakajima and Kiyoshi Mizusawa Research & Development Division of Kikkoman Corporation, Noda-shi, Chiba-ken (Japan)

(Received 5 April 1991; revision received 12 July 1991; accepted 15 July 1991) Key wordr: Guanidinoacetic acid; Guanidinoacetate

kinase; Enzymatic assay; Urine

Summary We describe a new enzymatic determination of urinary guanidinoacetic acid (GAA) with guanidinoacetate kinase (ATP: guanidinoacetate N-phosphotransferase, EC 2.7.3.11, which does not require a blank to correct for endogenous constituents (ADP and pyruvate). In the first step, pyruvate kinase (ATP: pyruvate 20phosphotransferase, EC 2.7.1.40) and lactate dehydrogenase (L-lactate: NAD+ oxidoreductase, EC 1.1.1.27) were used to eliminate endogenous constituents (ADP and pyruvate) in the presence of phosphoenolpyruvate and NADH. In the second step, urinary GAA was phosphorylated in the presence of ATP by guanidinoacetate kinase to form phosphoguanidinoacetate and ADP. The resultant ADP was sequentially measured at 340 nm in a coupled reaction catalyzed by pyruvate kinase and lactate dehydrogenase. The standard curve was linear up to 20 mg/dl for standard solutions of GAA. Analytical recovery of GAA added to normal urines ranged from 97.0 to 103.2% (mean 100.7%). The within-run and between-run studies gave CV values of I 3.6% and I 4.8%, respectively. No significant interferences by endogenous urinary compounds were observed with the proposed method under this study. The results obtained by the present method correlated well with those obtained by a high-performance liquid chromatographic method. This method is accurate and simple, and less time-consuming than those previously reported. We determined the concentrations of GAA in 24-h urine samples by the proposed method, and observed that the urinary excretion of GAA decreased markedly in patients with renal failure.

Correspondence 278, Japan.

to: Y. Shirokane, R & D Division, Kikkoman Corp., 399 Noda, Noda City, Chiba Pref.

228

Guanidinoacetic acid (GAA) is produced from L-arginine (LArg) in the presence of glycine (Gly) by L-Arg : Gly amidinotransferase (transamidinase, EC 2.1.4.1) mainly in kidney and then transported to the liver, where it is methylated to yield creatine. Localization of this transamidinase in kidney has been demonstrated immunohist~hemically in the proximal tubule [l], and the transamidinase activity was recently recognized only in the first and second portions of the proximal tubule [2]. Although GAA is physiologically excreted into urine, it significantly decreases in patients with renal dysfunction compared with healthy controls [3]. In recent years, the determination of GAA in urine has been reported to be useful as a sensitive index of renal tubular damage [2,4-73. The urinary GAA has been determined by automated ‘high-perfo~ance’ liquid chromatography (HPLC) [f&9],but this method is time-consuming and inappropriate for use in a large number of analyses. Recently, in our laboratory, a new enzymatic method available for the determination of urinary GAA was first developed using guanidinoacetate (GAA) amidinohydrolase (EC 3.5.3.2) and urease (EC 3.5.1.5) IlO,ll]. However, this enzymatic method was found to be somewhat complicated, because it required a sample blank to correct for endogenous color-producing constituents in urine. In order to establish a convenient determination of urinary GAA, we purified guanidinoacetate (GAA) kinase from a polychaete, Perinereis sp., because it was strictly specific to GAA [12]. This paper presents a new enzymatic endpoint assay of urinary GAA using the enzyme. In our newly proposed ho-reagent system, endogenous constituents (ADP and pyruvate) in urine are eliminated before determination, thus a quantity of GAA can be measured using the following reactions. There are two reactions in the ‘first step’ and three reactions in the ‘second step’. First step (elimination of ADP and pyruvate): pyruvate kinase

ADP + phosphoenolpyruvate

Pyruvate + NADH + H+

lactate dehydrogenase

Second step (determination GAA

+ ATp

GAA kinase ------+

ADP + phosphoenolp~uvate Pyruvate -t NADH + H*

+

ATP + pyruvate t lactate + NADf

of GAA):

phosphoguanidinoacetate pyruvate kinase

lactate dehydrogenase

)

+ ADP

ATP + pyruvate t lactate + NAD+

229

The decrease of NADH in the second step, as measured by the change in extinction at 340 nm, is proportional to the amount of GAA present. Materials

and methods

Urine specimens

Random urine specimens were obtained from healthy volunteers to assess the reliability of the proposed enzymatic method and to compare it with an HPLC method. Samples of 24-h urine were also collected from 34 out-patients (16 men and 18 women, ages 28-74) who went to Kikkoman General Hospital to take the tests of renal function. These samples were stored at - 20°C until analysis. Apparatus

Throughout this study all absorbances were measured using a Hitachi Model 557 Double Wavelength Double Beam Spectrophotometer (Hitachi, Ltd., Tokyo, Japan) equipped with a cuvette holder keeping constant temperature. To compare with the enzymatic method, an HPLC apparatus assembled according to Hiraga et al. [8] was also used for the separation and fluorometric determination of GAA. Enzymes and chemicals

Guanidinoacetate

(GAA) kinase was extracted and purified from a polychaete, involving ammonium sulfate fractionation and chromatographies on Butyl-Toyopearl 650 C, Sephadex G-200 and DEAESephacel. The specific activity of the enzyme was 43-46 U/mg protein at 25°C and adenosineJ’-triphosphatase activity was < 0.002% of GAA kinase activity. GAA kinase was strictly specific to GAA, and showed almost no susceptibility (< 1% of the rate obtained with GAA) toward other guanidino compounds in urine, such as creatinine, creatine, L-arginine, taurocyamine, 3_guanidinopropionate, 4-guanidinobutyrate, guanidine, guanidinosuccinic acid and methylguanidine [12]. GAA kinase was stored at - 20 o C in Tris-acetate buffer (20 mmol/l), pH 8.1, containing 5% glycerol and dithiothreitol (I mmol/l) until use. Pyruvate kinase (PK, from rabbit muscle) and lactate dehydrogenase (LDH, from rabbit muscle) were both purchased from Sigma Chemical Co., St. Louis, MO 63178, USA. GAA, ATP, NADH and phosphoenolpyruvate (PEP) were also obtained from Sigma Chemical fo. All other chemicals were of analytical grade.

Perinereis sp. by successive procedures

Reagents

GAA standard solution was freshly prepared by dissolving 10.0 mg of GAA in 100 ml of distilled water. Reagent A was prepared to contain 10 kU of PK, 10 kU of LDH, 2.2 mmol of ATP, 0.84 mmol of PEP, 5.0 mmol of MgSO,, 10 mmol of KC1 and 0.2 mmol of NADH/l in Tris-HCI buffer (100 mmol/l), pH 7.5. Reagent

230 TABLE I Procedure far determination

of GAA in urine Microcuvette no. I

III

II

1. Pipette into the microcuvettes: Urine Standard solution Distilled water Reagent A

50 PI SOPI 1.3 ml

5Op1 1.3 ml

1.3 ml

2. Mix and measure the absorbances at 340 nm against distilled water after incubation at 37°C for 5 min A 340 nm

ES0

EstO

EbO

0.1 ml

0.1 ml

0.1 ml

3. Add into the microcuvettes: Reagent B

4. Mix and measure the absorbances at 340 nm against distilled water after reaction at 37°C for 10 min A l‘m“rn

Es1

Estl

Ebl

B was prepared to contain 150 kU of GAA kinase, 1.0 mmol of dithiothreitol 5% glycerol/l in Tris-HCI buffer (100 mmol/l), pH 7.5.

and

GAA in urine was assayed by the procedure as described in Table I with use of microcuvettes. The amount of GAA in urine was calculated by the following formula.

(~OXF-Esl)-(E~XF-Ebl) GAA( mg/dl)

=

(EstOxF-Estl)-(EbOxF-Ebl)

F = 1.35/1.45 (a factor for correction

x1O

of the reaction volume).

Results Study of assay conditions

Optimal assay conditions to include the elimination of endogenous constituents (ADP and pyruvate) were studied by measuring the decrease of NADH at 340 nm as described in ‘Materials and Methods’. Relatively large amounts of three coupling enzymes (GAA kinase, PK and LDH) were used for GAA assay system, so that the reactions in the two steps proceeded sufficient rapidly. The GAA assay was carried out at pH 7.5 because all three enzymes were more active around pH 7-8.

231

.I

0

2

4 Time

6

8

10

12

t min,

Fig. 1. The reaction curves of GAA standards (10 and 20 mg/dl) and two urine samples (Ul and U2) in the second step.

The optimal incubation time for eliminating endogenous constituents in 50 ~1 of urine was investigated with use of urine samples of various GAA values (1.13 to 16.75 mg/dl) and only 5 min was required for complete elimination of constituents (ADP and pyruvate) at 37°C in the first step. The ,optimal reaction time for determination of GAA was also examined with use of G;?A standard solutions (10 and 20 mg/dl) and two urine samples. Figure 1 shows the reaction curves obtained after the addition of reagent B in the second step. The reaction with use of GAA standard solutions went to completion within about 6 min, while about 8 min were required for complete reaction with use of urine samples (Ul and U2). This difference in time to reach equilibrium between standard and sample indicated that some unwon substances in urine slowed the reaction of GAA measurement. However, we confirmed that the reactions in the second step proceeded completely within 10 min with use of about 40 urine samples (GAA values, 1.13-16.75 mg/dl). Thus, the final procedure for determination of urinary GAA was established as outlined in ‘Materials and Methods’.

The calibration curve constructed from aqueous GAA standard solutions was found to be linear in the range of O-20 mg/dl, and passed through the origin. In routine use, the calibration curve was not constructed and urinary GAA values were obtained by comparison with a single standard solution analyzed in duplicate. The analytical recovery was investigated with normal urine samples of three different GAA values (2.47-8.93 mg/dl) by adding 3.0, 6.0, or 9.0 mg of GAA per deciliter (Table 10. The average recovery of GAA added was 100.7% (range 97.0-103.2%).

232 TABLE II Analytical recovery of GAA added to urines Samples

Added (mg/dI)

Found (mg/dl)

Urine A

0.0 3.0 6.0 9.0

2.47 5.54 8.51 11.53

3.07 6.04 9.06

102.3 100.7 100.7

0.0 3.0 6.0 9.0

5.74 8.76 11.84 15.03

3.02 6.10 9.29

100.8 101.7 103.2

0.0 3.0 6.0 9.0

8.93 11.91 14.75 17.96

2.98 5.82 9.03

99.3 97.0 100.3

Urine B

Urine C

Recovery (mg/dl)

(%)

_

_

Mean

100.7

To determine within-run and between-run precision (CV values), normal urine samples of various GAA values were assayed 10 times and 6 times, respectively. The within-run precision (CV) for urinary GAA was < 3.6% and the between-run precision (CV) for daily GAA analyses of urines was less than 4.8% (Table III). The effect of several substances, listed in Table IV, on the determination of urinary GAA was examined. Although GAA standard solution (10 mg/dl) was assayed with and without addition of possible interfering substances at the concentrations shown for each substance, these did not affect significantly the present method.

TABLE III Within-run and between-run precision Samples Within-run (n = 10)

Cv(%o)

GAA (mg/dII Mean

SD

Urine A B C D

3.08 6.85 10.71 14.18

0.11 0.18 0.22 0.20

3.57 2.63 2.05 1.41 2.42

E F G H

2.52 6.24 7.06 13.86

0.12 0.19 0.16 0.30

4.76 3.04 2.27 2.16 3.06

Mean Between-run (n = 6)

Mean

233 TABLE Effect

IV of substances

added

to GAA

Substances

None Urea Creatinine Creatine Glucose Albumin EDTA-Na, Citrate Oxalate Ascorbate Glutathione

standard

solution

on the GAA assay

Cone (mg/dl)

Apparent f%)

3.000 1,000 200 5,000 5,000 500 500 100 500 200

100.0 96.3 99.3 98.3 98.4 101.2 95.8 98.7 97.7 100.4 103.4

GAA

The enzymatic method (x) was compared with an HPLC method (y) with use of 39 urine samples for the determination of urinary GAA. A good correlation (r = 0.996) was obtained between the two methods and the regression equation for these data was y = 0.976x + 0.236 mg/dl. GAA excretion in 24-h urine samples We determined the GAA concentrations in 24-h urine samples from 34 outpatients with suspected or proven renal insufficiency by the present method. Figures 2 and 3 show the relationship between the urinary GAA excretion (U-GAA) and the level of creatinine clearance (Ccr), blood urea nitrogen (BUN)

150

Y-0.871x+34.5 r=0.703.Pc0.001 n=34

.

l

. 0.0

.

01

L

0

30 U-GAA

Fig. 2. Relationship

between

urinary

60

90

121

(w/day)

GAA excretion

W-GAA)

and creatinine

clearance

(Ccr).

234 60 .

t 40 = P E

5

Y--0.166x+27.5 r=-0.469.P~0.01 n=34

t

I l

4 . g

#a

3-

e ;

2 3 20 m

: *.a

c;,

L 0

'\* 30

80

U-GAA

Fig. 3. Relationship

tmQ/W')

90

l

2-

1-t

0

Y=-0.0166x+1.81 r=-0.515. P c 0.01 n= 34

0

0

.

'.." . . . l '7..

30 U-GAA

.

. .

.

60

90

121

(me/day)

between urinary GAA excretion (U-GAA) and blood urea nitrogen (BUN) or serum creatinine (S-0).

or serum creatinine (S-Cr). The urinary GAA excretion indicated a significant positive correlation with Ccr (r = 0.703, P < O.OOl), while there were significant negative correlations between the urinary GAA excretion and BUN (r = - 0.469, P < 0.011, and between the urinary GAA excretion and S-Cr (r = - 0.515, P < 0.01). However, no correlation was observed between the urinary GAA excretion and the urinary excretion of protein (mg/day) (data not shown). These results revealed that the urinary excretion of GAA was markedly lowered in proportion to the fall of Ccr below the normal level (70-130 ml/min), the rise of BUN over the normal level (7-20 mg/dl) or the rise of S-Cr over the normal level (0.5-1.4 mg/dl). Discussion

Although determination of GAA in urine has been done by HPLC [8,91 and our enzymatic method previously mentioned [lO,llI for the diagnosis of renal failure, these methods are somewhat time-consuming and complicated in clinical practice. To obtain an economical and specific method of GAA analysis, we developed a new enzymatic endpoint assay including the two steps. Endogenous constituents (ADP and pyruvate) were successfully eliminated in the first step, thus only GAA was quantitatively determined by the described method, because GAA kinase was strictly specific to GAA [12]. The present enzymatic method for urinary GAA had satisfactory accuracy, as shown by the linearity of the calibration curve and by the good analytical recovery of GAA. The precision of this method was similar to those of the HPLC method and our previous enzymatic method, and there was good correlation (I = 0.996) between the results obtained by the present enzymatic and HPLC methods. Similarly, the proposed enzymatic method was compared with the previous method [ill with use

235

of 36 urine samples, and a good correlation (r = 0.985) was revealed between the two methods (data not shown). Although we examined about eighty urine samples from healthy persons and outpatients with suspected or proven renal insufficiency, none was found to contain interfering substances on the GAA assay system as described in Table I. The enzymatic method proposed here fulfilled the need for an accurate, specific and rapid assay of GAA, and was simple as compared to the HPLC method and our previous enzymatic method. In addition, the present enzymatic method can be adapted to automatic analyzers presently used in many laboratories. To confirm the relation between the decrease of GAA in urine and renal failure, we determined the GAA concentrations of 24-h urine samples from outpatients with suspected or proven renal insufficiency by the present method. A significant positive correlation between the urinary GAA excretion and creatinine clearance (Ccr) has been reported in patients with chronic glomerulonephritis and diabetes mellitus [13], essential hypertension [4] and renal disorder owing to medication with antibiotics [61. The decrease of urinary GAA excretion was also demonstrated in patients with chronic renal failure [13,14] and patients receiving renal transplants 1151 as serum creatinine (S-Cr) and/or blood urea nitrogen (BUN) also rose over the normal level. Our results agreed with previous findings which showed a correlation between renal dysfunction and the decrease of urinary GAA excretion and it is obvious that the urinary GAA excretion as well as the levels of Ccr, S-Cr and BUN are useful to detect renal dysfunction. In recent years measurements of P,-microglobulin @,-MG) and/or N-acetyl-PD-glucosaminidase (NAG) have been widely used as indicators for detecting renal tubular damage [16-181. On the other hand, the decrease of urinary GAA excretion was observed without any alteration of urinary &-MG and NAG in patients with essential hypertension [19] and renal disorder owing to medication with antibiotics 161, suggesting that the urinary excretion of GAA may be a more useful marker of renal tubular damage than that of &-MG or NAG. In conclusion, it is likely that the measurement of urinary GAA by the method proposed here is valuable and helpful for the early diagnosis of renal dysfunction especially renal tubular metabolic dysfunction. Acknowledgements

We thank Drs. S. Sugiyama, N. Saito and H. Sugimoto for support and encouragement, Mr. M. Utsushikawa for collection of urine samples and Ms. K. Yajitate for technical assistance. References 1 McGuire DM, Gross MD, Elde RP, Van Pilsum JF. Localization of L-arginine-glycine amidinotransferase protein in rat tissues by immunofluorescence microscopy. J Histochem Cystochem 1986;34:429-435.

236 2 Takeda M, Kiyatake I, Yaguchi Y, et al. Intrarenal biosynthesis of guanidinoacetic acid and its application to gentamicin nephrotoxicity. Igaku No Ayumi 1990;153:653-654. 3 Sasaki M, Takahara K, Natelson S. Urinary guanidinoacetate/guanidinosuccinate ratio: an indicator of kidney dysfunction. Clin Chem 1973;19:315-321. 4 Takano Y. Estimation of urinary excretion rate of guanidinoacetic acid in essential hypertension. Jpn J Nephrol 1989:31:1187-1196. 5 Kuwagaki Y, Sudo J. Interrelation of urinary and plasma levels of guanidinoacetic acid with alteration in renal activity of glycine amidinotransferase in acute renal failure rats. Chem Pharm Bull 1989;37:781-784. 6 Nakayama S. Kiyatake 1, Shirokane Y. Koide H. Effect of antibiotic administration on urinary guanidinoacetic acid excretion in renal disease. In: Mori A, Cohen BD, Koide H. eds. Guanidines 2. New York: Plenum Press, 1989:313-322. 7 lsoda K, Mitarai T, Imamura N, et al. The significance of serum and urinary guanidinoacetic acid level for the restoration of renal metabolic function in patients with kidney transplantation. Ibid.. 337-344. 8 Hiraga Y, Kinoshita T. Post-column derivatization of guanidino compounds in high-performance liquid chromatography using ninhydrin. J Chromatogr 1981;226:43-51. 9 Hung Y, Kai M, Nohta H, Ohkura Y. High-performance liquid chromatographic analysis for guanidino compounds using benzoin as a fluorogenic reagent. J Chromatogr 1984;305:281-294. 10 Shirokane Y, Utsushikawa M, Nakajima M. A new enzymic determination of guanidinoacetic acid in urine. Clin Chem 1987;33:394-397. 11 Shirokane Y, Nakajima M, Mizusawa K. Easier enzymatic determination of guanidinoacetic acid in urine. Clin Chem 1991;37:478-479, 12 Shirokane Y, Nakajima M, Mizusawa K. Purification and properties of guanidinoacetate kinase from a polychaete, Perinereis sp. Agric Biol Chem 1991;55(9):in press. 13 Tsubakihara Y, Iida N, Yuasa S, et al. Guanidinoacetic acid (GAA) in patients with chronic renal failure (CRF) and diabetes mellitus (DM). In: Mori A. Cohen BD. Lowenthal A, eds. Guanidines. New York: Plenum Press, 1985;309-316. I4 Tsubakihara Y, Yamato E. Yokoyama K, et al. Short-term protein load in assessment of guanidinoacetic acid synthesis in patients with chronic renal failure. In: Mori A, Cohen BD. Koide H, eds. Guanidines 2. New York: Plenum Press. 1989:331-336. 15 Ishizaki M, Kitamura H, Takahashi H, et al. Evaluation of the efficacy of anti-rejection therapy using the quantitative analysis of guanidinoacetic acid (GAA) urinary excretion as a guide. In: Mori A, Cohen BD, Lowenthal A, eds. Guanidines. New York: Plenum Press 1985;353-363. 16 Shimojo N. Naka K. /3,-Microglobulin; Determination and significance of urinary low molecular weight proteins. Rinsho Kensa 1988;32:857-862. activities in I7 Wellwood JM, Ellis BG, Price RG. et al. Urinary N-acetyl-beta-D-glucosaminidase patients with renal disease. Br Med J 1975;3:408-411. I8 Kinoshita S, Saeki S, Shimojo N, Naka K, Okuda K, Kohno M. Urinary and serum N-acetyl-P-Dglucosaminidase (NAG) activities in clinical significance. Rinsho Byohri 1986:34:687-693. 19 Takano Y, Gejyo F, Shirokane Y, Nakajima M. Arakawa M. Urinary excretion rate of guanidinoacetic acid in essential hypertension. Nephron 1988:48:167,168.