Clin 8iochem, Vol. 25, pp. 109-114, 1992

0009-9120/92 $6.00 + .00 Copyright © 1992 The Canadian Society of Clinical Chemists.

Printed in the USA. All rights reserved.

Homogeneous Trinder-Coupled Assay for the Determination of Glucose-6-Phosphatase Activity in Tissue Extracts W A L I D G. Y A S M I N E H , 1 J A N E L L E

I. C A S P E R S , 2 a n d A T H A N A S I O S

THEOLOGIDES

2

1Department of Laboratory Medicine and Pathology, University Hospital, and 2Department Medicine, Hennepin County Medical Center, University of Minnesota Medical School, Minneapolis, MN 55455, USA We describe an automated, homogeneous, glucose oxidasecoupled method for the determination of glucose-6phosphatase activity in tissue extracts. The method is based on measurement of the rate of glucose formation by the Trinder reaction, in which the end product is a quinoneimine dye which absorbs maximally at 505 nm and has a molar extinction coefficient of 5700. The incubation mixture contains 20 p.L of tissue extract, 25 ILL of 0.5 M phosphate buffer, pH 7.0, 175 I~L of Trinder/glucose-6-phosphate reagent, and 30 p.L of distilled water. After a delay period of 15 min, to exhaust any glucose endogenously present in the extract, glucose production from glucose-6-phosphate is monitored at 505 nm for 5 min in a centrifugal analyzer. The Km was 13 mM over a 10-fold range in glucose-6-phosphate concentration and the reaction was linear up to about 250 U/L. Within-run CV of the assay at activities of 48 and 190 U/L ranged between 2.5-5.0%. The between-run CV at 190 U/L was 5.1%.

KEY WORDS: gluconeogenesis; glucose-6-phosphatase; Trinder reaction.

of

lowing incubation of an aliquot of the extract with glucose-6-phosphate (G6P). This is a tedious, heterogeneous assay which requires stopping the reaction by precipitating the protein with acid, and determining the amount of Pi in the supernatant fluid spectrophotometrically after forming a complex with ammonium molybdate. Belfield and Goldberg (4,8) described an assay in which the rate of glucose production is measured by coupling to a glucose oxidase system with guaiacol as a hydrogen donor. Endogenous glucose in the sample was exhausted by preincubation with the glucose oxidase prior to addition of the dye. We describe an automated homogeneous assay in which the rate of the reaction is monitored continuously by coupling with the Trinder reaction (9) to produce a quinoneimine dye that absorbs maximally at 505 nm:

Introduction G6Pase

lucose-6-phosphatase (G6Pase; EC 3.1.3.9) is one of the key enzymes of gluconeogenesis which regulates the overall conversion of lactate to glucose. The activity of the enzyme is elevated in the liver of fasted and tumor-bearing animals, and is therefore a good indicator of glucose homeostasis (13). G6Pase is normally undetectable in normal human serum and is elevated in liver disease, although as a diagnostic test it is of little value because of low sensitivity and specificity (4). It is also frequently used as a marker of the microsomal fraction in tissue fractionation procedures (5,6). The standard method for the determination of the activity of G6Pase in tissue extracts (7) is measurement of the inorganic phosphate (Pi) liberated fol-

G

G6P

~ glucose + Pi

glucose oxidase

Glucose

~ gluconic acid + H202

H202 + 4-aminoantipyrine + horseradish p-hydroxybenzene sulfonate

~ dye peroxidase

Endogenous glucose in the sample was exhausted rapidly during the first 15 rain, after which dye formation was derived solely from the action of the enzyme on G6P. Materials and methods REAGENTS

Correspondence: Walid Yasmineh, Ph.D., Department of Laboratory Medicine and Pathology,Box 198, University of Minnesota Hospital and Clinic, 420 Delaware Street SE, Minneapolis,M N 55455, USA. Manuscript receivedAugust 21, 1991; revisedNovember 26, 1991; acceptedDecember 5, 1991. CLINICALBIOCHEMISTRY,VOLUME25, APRIL1992

The following reagents were purchased from Sigma Chemical Co., St. Louis, M O , USA: E G T A , ethylene glycol-bis-(~-arninoethylether)NflVJV'N'tetraacetic acid; E D T A (ethylenedinitrilo-tetraacetic acid); ~-glycerophosphate; Tris, tr/s(hydroxy109

YASMINEH, CASPERS, A N D THEOLOGIDES

methyl)aminomethane; cacodylic acid; glucose-6phosphate; phenyl phosphate, disodium salt; acid phosphatase from h u m a n semen; alkaline phosphatase from bovine intestinal mucosa. Sucrose was purchased from Spectrum Chemical Manufacturing Co., Redondo Beach, CA, USA. Glycerol was purchased from Fisher Scientific Co., Fair Lawn, NJ, USA. Trinder glucose reagent was purchased from Sigma Diagnostics as the glucose (Trinder) kit, Catalogue No. 315. The reagent contains the following per liter of Tris-succinic acid at pH 7.0:0.5 mmol 4-aminoantipyrine; 20 mmol p-hydroxybenzene sulfonate; 15,000 U glucose oxidase (from Aspergillus niger); 10,000 U horseradish peroxidase.

Phosphorus reagent Reagent B of the DART phosphorus kit manufactured by Coulter Electronics, Inc., Hialeah, FL, USA, is used for the determination of serum Pi. It contains 0.84 mmol/L ammonium molybdate and 0.32 mmol/L sulfuric acid. Before the assay is run, 10 mL of the reagent are diluted to 14 mL with distilled water.

G6P reagent, 0.3 M A 10-mL solution is prepared by dissolving 0.847 g of G6P in distilled water. The solution is stable at 4 °C for 1 week.

Trinder-coupled assay (Figure 1). This medium extracts only about 25% of the G6Pase activity in liver while it extracts all of the endogenous glucose. All homogenates were then centrifuged at 30,000 x g for 30 min and the supernatant fluids were kept at - 20 °C until assayed, or diluted two-fold with glycerol, divided into small aliquots, and kept at - 20 °C for the precision studies. The supernatants were essentially post-mitochondrial and lysosome-free, comprising the microsomal and cytosolic fractions.

Determination of G6Pase by the standard Pi method To establish a correlation between the present method for measuring G6Pase activity with the standard Pi method, the latter was performed by an adaptation of the method of Daly and Ertingshausen (10). The incubation mixture at 30 °C contained 60 ~L of 0.1 M cacodylate buffer (pH 6.5), 10 ~L of 0.45 M G6P, 20 ~L of sample, and 60 ttL of distilled water. The sample consisted of 1-4 ~L of a rat liver homogenate diluted to 20 ~L with distilled water. It should be noted that the homogenizing m e d i u m was in this case Tris/sucrose/EGTA and did not contain inorganic phosphorus (see above). A reagent blank, in which water was substituted for the sample, was included. At zero time, the reaction was started by addition of the sample. After a 4-rain incubation period, the reaction was stopped by pipetting 75 ~L of the mixture into a 1-mL plastic centrifuge tube containing 575 }xL of phosphorous reagent. A small

Phosphate buffer, 0.5 M, pH 7.0 1.5

This is prepared by mixing 60 mL of 0.5 M Na2HPO4 with 40 mL of 0.5 M KH2PO4.

Trinder glucose/G6P/phosphate working reagent, pH 7.1 This is prepared by mixing 1 mL of 0.3 M G6P reagent, 1 mL of phosphate buffer, and 6 mL of the Trinder glucose reagent. This reagent is stable for only 1 h, and should be mixed immediately before each assay.

Preparation of tissue homogenate Fresh rat liver and kidney were homogenized in a Potter-Elvehjem homogenizer in 10 volumes of cold medium containing 0.2 M phosphate buffer, pH 7.4, 5 mM EGTA, 5 mM EDTA, and 10% (v/v) glycerol. The EGTA and EDTA were added as enzyme stabilizers and did not inhibit the activity of G6Pase at the concentrations used. For the determination of G6Pase activity by the Pi method, an alternate homogenizing medium not containing inorganic phosphorus had to be used and consisted of i0 rnM Tris buffer (pH 7.4), 0.25 M sucrose, and 1 rnM E G T A . Another m e d i u m containing Tris and E G T A only was also used to determine the combined effects of endogenous glucose and inorganic phosphate in the 110

1.3

i

D • • • • O

0.9 1.1 0.7 0.5

Liverextract LNer extract + Pi RgLblank Rgt.blank+ Pi Trindar

G6P(30raM)

0.3 0.1

-0.1

I

5

1 I0

I

15

I

20

Minutes

Figure 1 -- Typical absorbance vs. time curves of the G6Pase activityof a rat liver homogenate in the presence or virtual absence of inorganic phosphate. The reaction mixtures of 250 ~L final volume contained 10 ~L of the supernatant from a rat liver homogenate (62 U/L) in the presence of 66 m M (designated + Pi) or 0.12 m M inorganic phosphate. In the lattermixture, the small amount of inorganic phosphate was that present endogenously in the 10 btL of supernatant, and the inorganic phosphate w a s omitted from the Trinder reagent (see Table 1). The final reaction mixtures also contained 0.13 m M endogenous glucose. The regression lines were drawn using only the 15-20-rain interval to reflectthe activitydue solelyto G6Pase.

CLINICALBIOCHEMISTRY,VOLUME25, APRIL 1992

GLUCOSE-6-PHOSPHATASE ASSAY

amount of precipitate developed which was spun down at 7500 x g for 5 min. The supernatant fluid was siphoned into a 1-mL cuvette and the absorbance was read at 340 nm in a Beckman DU-7 spectrophotometer (Beckman Instruments Inc., Fullerton CA, USA). After subtracting the absorbance of the reagent blank, the concentration of Pi in the supernatant fluid was read from an absorbance v s . concentration curve made up of Pi standards containing from 30-180 ~mol of Pi per liter of incubation mixture. The G6Pase activity of the rat liver homogenate in U/L was then calculated as follows: G6Pase (U/L) =

}~mol Pi/L incubation mixture 4 min x dilution factor

where the dilution factor is the total volume of the reaction mixture (650 ~L) divided by the volume of homogenate in microliters.

Results The design of the Trinder-coupled G6Pase method is very simple and easily automated by using a centrifugal analyzer such as the Cobas Fara (Roche Diagnostics, Nutley, NJ, USA). As shown in Table 1, the reaction mixture has a total volume of 250 ~L consisting of 20 ~L of sample in homogenizing medium, 200 ~L working glucose Trinder/G6P/ phosphate reagent, and 30 ~LL distilledwater. The sample volume of 20 ~LL is a m a x i m u m volume designed to accommodate the analysis of any tissue homogenate. Routinely, rat liver and kidney homogenates are diluted 10-fold and 5-fold, respectively, before analysis. The concentration of G 6 P in the final reaction mixture is 30 m M . The water acts both as a sample diluent and wash. The phosphate buffer is added to minimize the nonspecific hydrolysis of G 6 P in the presence of the Trinder reagent (see below). Its final concentration in the reaction in 66 rnM, consisting of 16 m M phosphate in the homog-

TABLE 1

Components of the G6Pase Assay Reagent

Sample Homogenizing medium Distilled water Trinder/G6P/phosphate reagenta Total volume

Sample (~L)

Blank (~L)

20 -30 200 250

-20 30 200 250

"The 200 ~L of this reagent included 25 }zL 0.5 M phosphate buffer (pH 7.0), 25 ~L 0.3 M G6P, and 150 ~L Trinder reagent. The final reaction contained 66 mM inorganic phosphate, of which 16 mM were in the homogenizing medium and 50 mM in the Trinder/G6P/phosphate reagent (see Methods). CLINICAL BIOCHEMISTRY, VOLUME 25, APRIL 1992

enizing medium and 50 mM phosphate in the Trinder/G6P/phosphate reagent. A reagent blank which contains 20 ~L of homogenizing medium instead of sample is run in the first cuvette of the rotor. After initiation of the run at 30 °C, the absorbance at 505 um of the quinoneimine dye that is produced is monitored at 1-min intervals for 20 min. G6Pase activity is determined from the change in absorbance between 15-20 min, after any glucose endogenously present in the sample is exhausted. Figure 1 shows typical absorbance versus time curves for a rat liver homogenate and its reagent blank in the presence of inorganic phosphate (66 mmol/L of final reaction mixture) and its virtual absence (0.12 mmol/L). The homogenate was prepared in Tris-EGTA medium (see Methods section) and contained 62 U/L G6Pase activity, 3 mmol/L endogenous inorganic phosphate, and 3.3 mmol/L endogenous glucose. The concentrations of endogenous inorganic phosphate and glucose in the final reaction mixture of 250 ~L were 0.12 mM and 0.13 raM, respectively. In the virtual absence of inorganic phosphate (0.12 mM), the rate of reaction of the homogehate was fast for the first 10 rain, because of the presence of 0.13 mM endogenous glucose, and subsequently achieved a slower linear rate reflecting G6Pase activity. The amount of endogenous glucose varied in different tissues. Liver homogenates contained about 3 mmol of glucose per liter (or about half the concentration in serum), while kidney homogenates contained only 0.5 mmol/L. The reagent blank showed a linear rate representing about 70% of that of the homogenate. Similar curves were obtained in the presence of 66 mM inorganic phosphate except that the linear rates were appreciably lower and the linear rate of the reagent blank represented only 30% of that of the homogenate. After subtraction of the respective blanks, however, the net linear rates due solely to G6Pase activity were similar (Aabsorbance/min, 0.014). Figure 1 also indicates that neither the G6P nor the Trinder reagents generate any change in absorbance. This suggests that the relatively large rate of the blank was caused by gradual hydrolysis of the G6P by nonspecific phosphatase(s) present in the glucose oxidase and peroxidase preparations used in the Trinder reagent. Analysis for alkaline phosphatase in the Trinder reagent showed an activity of about 20 U/L, or the equivalent of 12 U/L in the final reaction mixture. In investigating this problem further, it was found that the relatively high reagent blank could be reduced significantly by the addition of various organic phosphates such as ~-glycerophosphate and phenyl phosphate, which are intended to divert the activity of nonspecific phosphatases, as suggested by Belfield and Goldberg (4,8). As shown in Table 2, the blank decreased by 52% at a concentration of 25 mM phenyl phosphate, and to a lesser extent (30%) at a concentration of 50 mM ~-glycerophosphate. Interestingly, inorganic salts also decreased the blank. For example, ammonium sulfate decreased the 111

YASMINEH, CASPERS, A N D THEOLOGIDES

TABLE 2 Percent Reduction of the Reagent Blank by Various Compounds mM

Conc. ~

6 12 20 25 40 50 66

~GP

PheP

Pib

AS

-7 -16 -30 --

20 36 -52 ----

--30 46 62 -78

--10 --20 --

aConcentration in the final reaction mixture. t~he concentrations of Pi shown include 16 mM Pi present in the homogenizing medium. Abbreviations: pGP, p-glycerophosphate; PheP, phenyl phosphate; Pi, inorganic phosphate; AS, ammonium sulfate. b l a n k slightly (20%) at a 50-mM concentration, while inorganic phosphate was as effective as phenyl phosphate, reducing the blank by 46% and 62% at 25 mM and 40 mM concentrations, respectively. At 66 mM concentration, inorganic phosphate decreased the blank by 78%. This became the concentration of inorganic phosphate routinely used in the assay to minimize the nonspecific hydrolysis of G6P. As shown in Figure 1, this concentration of inorganic phosphate does not inhibit G6Pase activity. This was confirmed in a not he r exper i m ent using a crude microsomal preparation of G6Pase from rabbit liver (Sigma Chemical Co., Catalogue No. 5758) in the presence of 1 0 - 7 0 mM inorganic phosphate in the final reaction mixture. Reagent blank determinations were also done in the presence or absence of alkaline phosphatase, acid phosphatase and phenyl phosphate, as shown in Table 3. The results indicate t h a t the r a t e of reaction of the r e a g e n t bl ank was essentially unchanged in the presence of relatively high activities of alkaline or acid phosphatase (138 and 125 U/L in the

final reaction mixture). When 25 mM phenyl phosphate was added, the rate of the reaction was decreased by about 50% in the presence or absence of either alkaline or acid phosphatase (see also Table 2). Despite these results the presence of other nonspecific phosphatases in the T ri nder reagen t cannot be excluded. Figure 2 shows the kinetics of the T ri nde r glucose reaction. T he final r e a c t i o n m i x t u r e s c o n t a i n e d 0.04-0.16 mM glucose, 20 t~L of homogenizing medium, 66 mM inorganic phosphate, and 200 i~L of Trinder reagent without G6P (see Table 1). Over 75% of the color was developed within the first 5 min, and color development was essentially complete after 15 min. The absorbance was proportional to the concentration of glucose at every time point on the reaction curves and yielded a molar extinction coefficient of 5700 after 15 min. This coefficient was used to obtain G6Pase activity in units per liter as follows: U/L = Aabsorbance/min x sample dilution factor x 1,000,000/5700 Although the extinction coefficient is r e m a r k a b l y constant when different batches of T ri nder r e a g e n t are used, it is advisable to check new batches periodically to ascertain t h a t no change has occurred. These results suggested t h a t any endogenous glucose present in biologic fluids/tissue homogenates should be exhausted within 15 min, following which G6Pase activity m ay t h e n be measured. This was tested furt her by determining the G6Pase activity of a r a t l i ver h o m o g e n a t e c o n t a i n i n g 150 U/L of G6Pase activity in the presence of known a m o u n ts of added glucose, as shown in Figure 3. The final reaction mixtures contained 4 ~L of r a t liver homogenate with the equivalent of 0.06 mM endogenous glucose. As can be seen, addition of 0.1 and 0.2 1.0

TABLE 3 Hydrolysis of G6P in the Reagent Blank By Alkaline Phosphatase (AP) and Acid Phosphatase (AcP) in the Presence and Absence of Phenyl Phosphate (PheP)

•

0.16 mM

•

0.12 mM

•

0.08 mM

e= 0.04 mM

AAbsorbance/min

0 ¢/) .0

Blank Blank Blank Blank Blank Blank

+ + + + +

Phel~ AP b AP + PheP AcP¢ AcP + PheP

0.0279 0.0149 0.0294 0.0145 0.0280 0.0148

~Twenty-five millimolar PheP in the final incubation mixture. bActivity of AP was 138 U/L in the final incubation mixture. CActivity of AcP was 127 U/L in the final incubation mixture. i12

0.2

0.0 0

5

10

15

20

25

30

Minutes

Figure 2 - - Kinetics of the glucose Trinder reaction. The incubation mixtures were similar to those described in Table 1 but, instead of homogenate, contained 20 ~L of 0.5, 1.0, 1.5, and 2.0 mM glucose in homogenizing medium, to achieve the final concentrations shown. Also, the G6P was omitted from the Trinder/G6P/phosphate reagent. CLINICAL BIOCHEMISTRY,VOLUME 25, APRIL 1992

GLUCOSE-6-PHOSPHATASE A S S A Y 1.5

0.12

1.2

P,

0.09

0.9

•

0.2 mM

m

0,1 mM

•

NO Gluccc, e

B

Blank

0.6

0.06 "~ 1,-

g ,¢~

A .-I

0.3 0.03 0.0 ~ 0

I

s 5

10

115

y . 0.051 + 0.668x

210 0.00 -0,08

Minutes

Figure 3 -- Effect of endogenous glucose on G6Pase activity in the Trinder-coupled assay. The incubation mixtures were similar to those described in Table 1, but the sample was composed of 4 I~L of the supernatant from a rat liver homogenate (150 U/L), 6 I~L of homogenizing medium, and 10 I~L of 2.5 or 5.0 mM glucose in homogenizing medium. The final concentrations of glucose added to the incubation mixture are indicated. mM glucose increased the initial rate of reaction but the glucose was exhausted after 14 min and reached a constant, slower rate t h a t was the same for each mixture. The assay is linear to a Aabsorbance of 0.015/min (Figure 4). For a normal rat liver homogenized in 10 volumes of medium, this is equivalent to an activity of 263 U/L of homogenate, or 2.6 U/g of wet liver tissue. Homogenates of normal rat kidney contain about 20-50% less activity. Figure 5 shows the L i n e w e a v e r - B u r k plot for G6Pase from rat liver. The apparent Km for G6P was 13 mM over a 10-fold range in substrate concentration. Since the concentration of G6P used in the G6Pase assay is 30 mM, or 2.3 times the Km, the 0.020 =

.

,

=

.

0.016

.C_ E ¢J e-

0.012

,.D f#}

0.008