University Microfilms International 300 N

INFORMATION TO USERS

This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.

1. The sign or "target” for pages apparently lacking from the document photographed is "Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.

2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame. If copyrighted materials were deleted you will find a target note listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photo graphed the photographer has followed a definite method in "sectioning” the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.

4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.

University Microfilms International 300 N. ZEEB HU., ANN ARBOR, Ml 48106 8214094

Gendler, Stephen Marshall

METABOLISM OF CITRATE TO OXALATE IN THE RAT AND THE ANALYSIS OF OXALATE IN URINE

The Ohio State University Ph.D. 1982

University Microfilms International300 N. Zeeb Road, Ann Arbor, MI 48106

OF OXALATE IN URINE

DISSERTATION

Presented in Partial fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Stephen Marshall Gendler, B.S.

*****

The Ohio State University

1982

Reading Committee: Approved By

Howard A. I . Newman, Ph.D.

John S. Rieske, Ph.D.

Keith E. Richardson, Ph.D. C y/\ A d v iser Department of Physiological C hem istry ACKNOWLEDGMENTS

I would like to express my deepest gratitude to my

research advisor, Dr. Keith Richardson. Without his

scientific and spiritual guidance, this document would have

been impossible.

I would like to thank the other members of the reading

com m ittee, Dr. John R ieske and Dr. Howard Newman, fo r t h e i r

suggestions and help in the writing of this dissertation.

As my academic advisor, Dr. Newman has been especially helpful in my clinical chemistry career. I am also thankful

for the technical assistance given me by Dr. Dennis Pollack, who also served as Graduate School Representative.

There is one person who stands above all others in the support and advice given me. I would have quit long ago had

it not been for my wife, Kathy. I will have this degree because of her love and understanding, and I thank God for her presence. VITA

November 19, 19 51 ...... Born, Detroit, Michigan

1973 ...... B.S., Michigan State University, East Lansing M ichigan

1974 ...... Registered Medical Technologist, American Society of Clinical P athology

1982...... Assistant Director, Clinical Biochemistry, Michael Reese Hospital, Chicago, Illinois

PUBLICATIONS

Gendler, S.M., M.P. Farinelli, K.E. Richardson, and D.W. Fry, Two enzymes with potential use in spectrophotometric oxalate determination, Clinical Chemistry, 2<>, 1032 (1981) (Abstract).

FIELDS OF STUDY

Major Field: Clinical Chemistry

Studies in Oxalate Metabolism and Oxalate Analysis. Professor Keith E. Richarson

iii TABLE OF CONTENTS

Page ACKNOWLEDGMENTS...... i i

VITA...... i i i

LIST OF TABLES...... v

LIST OF FIGURES ...... vi

INTRODUCTION...... 1

C h ap ter

I . THE METABOLISM OF CITRATE TO OXALATE 4

Literature Review ...... 4 H y p o th e s is...... 20 M ethods...... 21 Results and Discussion...... 26 Conclusions...... 30

I I . A POTENTIAL ASSAY FOR OXALATE UTILIZING COUPLED ENZYMATIC REACTIONS...... 32

Literature Review ...... 32 H y p o th e sis...... 59 M ethods...... 60 Results and Discussion...... 68 Conclusions...... 81

I I I . AN HPLC METHOD FOR URINARY OXALATE 84

Literature Review ...... 84 H y p o th e sis...... 91 M ethods...... 92 Results and Discussion...... 97 Conclusions...... 105

REFERENCES 106 LIST OF TABLES

T able Page

1. Results of the Study Showing the Metabolism of Citrate to Oxalate in the Male Wistar R a t...... 28

2. Influence of Media on Oxalate Decarboxylase Production in Aspergillus phoenicis at 30°C... 69

3. P u r if i c a t i o n of O x alate D ecarb o x y lase...... 73

4. Retention Time of Organic Acids of the TCA Cycle and Oxalate Metabolism ...... 98

v LIST OF FIGURES

Figure Page

1. Pathways of Oxalate Synthesis...... 6

2. Influence of Manganese on Oxalate Decarb oxylase Production in Aspergillus p h o e n ic is...... 71

3. Influence of Growth Temperature on Oxalate Decarboxylase Production in Aspergillus p h o e n ic is...... *...... 7 2

4. Influence of pH on the Activity of Oxalate Decarboxylase and Formate Dehydrogenase...... 7 5

5. Reaction Progression of the Coupled R e a c tio n s...... 80

6. Separation of Radiolabelled Acids ...... 99

7. Chromatogram of Blank...... 101

8. OD Treated Urine ...... 103

9. Chromatogram of Urine...... 104

vi INTRODUCTION

Problem Statement

Oxalate, a normal constituent of human urine, is

excreted at a rate of 20 to 40 mg per 24 hours. The source

of this oxalate is dietary and endogenous synthesis.

Endogenous synthesis accounts for 89 to 97 percent of the

total oxalate excreted by the kidney ( Richardson and

Farinelli, 1981 ). However, this excretion is exceeded in

several diseases in man. Primary hyperoxaluria, a rare

genetic disease, is characterized by recurrent calcium

oxalate nephrolithiasis, chronic renal failure, and early

death from uremia ( Williams and Smith, 1978 ). Hyper

oxaluria is also seen in patients who have undergone ileal

resection or have inflammatory ileal disease ( Earnest et

a l., 1974 ). In addition, there are 40 deaths per year

resulting from ethylene glycol poisoning, where formation

and deposition of oxalic acid occurs in renal, cardiopulmonary and central nervous system tissues ( Berman et al., 1957 ). Oxalic acid is present in 73 percent of all kidney stones and is the major component in over 70 percent of a series of 10,000 urinary calculi analyzed from across

the United States ( Herring, 1962 ). A kidney stone will develop in one to five percent of the population ( Boyce et a l . , 1956 ).

It is important to understand the metabolism of oxalic acid as it is a component of many major diseases. Citrate

is a intermediary metabolite present in all tissues.

Citrate is reported to be metabolized to oxalate in the rat

( Hodgkinson, 1978 ). Should a metabolite as ubiquitous as citrate be converted to oxalate, this conversion could be a major contributor to oxalate depositon in the kidney and in extrarenal tissues. Once the metabolism is mastered it may be possible to control oxalate formation in disease.

Therefore, the metabolism of citrate, a central intermediary metabolite, to oxalate was investigated.

Also, as oxalic acid is a major constituent in kidney stones, it is standard practice to analyze oxalate in urine as part of the diagnosis of urolithiasis ( Smith, 1979 ).

Oxalate levels in urine are also crucial for the diagnosis of primary hyperoxaluria ( Williams and Smith, 1978 ).

Evidenced by the many published assays for oxalate, a simple, precise, accurate, rapid method does not now exist.

Therefore, oxalate decarboxylase from the fungus

Aspergillus phoenicis and formate dehydrogenase from the soybean in the analysis of oxalate were investigated for their use in a coupled enzymatic assay for oxalate. The analysis of oxalate and its precursors using ion exclusion high performance liquid chromatography was also explored. CHAPTER I

The Metabolism of Citrate to Oxalate

Literature Review: Elements of Oxalate Metabolism

Exogenous Sources

It is usually stated that oxalate is poorly absorbed

from the intestinal tract. However, studies in man show absorption ranging from about two percent ( Zarembski and

Hodgkinson, 1969 ) to 100 percent ( Elder and Wyngaarden,

1960; Hodgkinson and Wilkinson, 1974 ). Studies showing the degree of absorption in the range of 10 to 15 percent also exist ( Williams, 1978; Earnest et al. , 1974; Chadwick et a l., 1973 ). Results are dependent on the analytical methodology and whether isotopes are used. Most experts feel that studies point to poor absorption ( Richardson and

Farinelli, 1981; Hagler and Herman, 1973; Hodgkinson,

1977 ). Surgical intervention has been shown to influence oxalate absorption. Ileal resection can lead to recurring oxalate stones due to increased intestinal oxalate absorption. Inflammatory bowel disease also increases oxalate absorption ( Earnest et al., 1974 ). The mechanism of absorption in man is non-energy dependent diffusion that 5 is probably mediated by an oxalate binding protein ( Pinto and Paternain, 1978 ). This passive mechanism occurs in the rat also, with the greatest absorption occuring in the jejunum and the least in the ileum ( Madorsky and Finlayson,

1977 ).

Oxalate content in the diet varies depending on the geographical location of the investigation and the socio-economic status of the population studied. Daily intakes range from 70 to 150 mg per kg day ( Zarembski and

Hodgkinson, 1962 ) in the English diet, and are as high as

2045 mg per kg day for an Indian on a seasonal rural diet

( Singh et al. , 1972 ).

Endogenous Sources

There are three immediate precursors of oxalate: ascorbate, glycolate, and glyoxylate. The metabolism of oxalate will be disscussed in relation to these precursors.

The pathways of oxalate biosynthesis are shown in figure 1.

A sco rb ate

Ascorbate is an important precursor of oxalate since from 10 to 40 percent of total urinary oxalate comes from this vitamin ( Crawhall et al., 1959; Atkins et al., 1964; 6

D-glycerate Lrglycerote

COM LON hydroxypyruvate COg

glycolaldehyde serin* CO* t ►

glycine

CO. Hydroxyprolme CAO CLDH) GLYCOLATE Tryplophon LDH GDOXYLATE CAO C ilrote? CAO XO LOH OXALATE NON-EHZYMC

ASCORBATE GLYOXYLATE

Figure 1: Pathways of oxalate synthesis. (Bichardson and Parinelli, 1980) Baker et al., 1966 ). In the rat, over eight percent of intraperitoneally injected [ 1 - 1 ]L-ascorbic acid appears in the urine as l^C-oxalic acid within 24 hours

( Hodgkinson, 1978 ). The pathway for the conversion of ascorbate to oxalate is probably non-enzymatic. Ascorbate is known to be unstable in solution, decomposing to form oxalate from carbons 1 and 2 ( Baker et al. , 1966 ).

Although an enzymatic conversion of ascorbate to oxalate has been proposed by Baker et a l. ( 1966 ), Gambardella and

Richardson ( 1977 ) believe the oxalate is formed as a result of the inate instability of the molecule. Their conclusion is based on the absence of any labelled metabolite of the injected [ l-l^C ]L-ascorbate being found in the rat urine other than labelled oxalate and the unchanged labelled ascorbate. This controversy has not been r e s o lv e d .

G ly c o la te

Glycolate is the second immediate precursor of oxalate.

Dietary glycolate is 100 percent absorbed in the rat and three percent of the orally administered labelled glycolate appears in the urine in 48 hours as labelled oxalate

( Harris and Richardson, 1980 ). Approximately 1.6 percent of intraperitoneally injected [ 1-l^C ]glycolic acid is in the urine as labelled oxalate within 24 hours ( Hodgkinson,

1978 ). 8

A dehydrogenase specific for the oxidation of glycolate to oxalate was isolated and characterized by Fry and

Richardson ( 1979 ). The enzyme is found only in the liver in the rat and in man, and is present in concentrations adequate to account for the entire endogenous synthesis of oxalate in man. The presence of the glycolic acid dehydrogenase ( GAD } explains several experiments that are not consistant with glyoxylate and ascorbate being the sole immediate precursors of oxalate. For example, male rats that are vitamin Bg deficient increased their metabolism of glycolate, ethylene glycol, and ethanolamine to oxalate by 18, 10, and 14 fold, respectively, over controls, whereas glyoxylate and glycine only increased 1.2 and 1.4 times, repectively ( Runyan and Gershoff, 1965 ). Should glyoxylate be the exclusive antecedant of oxalate besides ascorbate, the discrepancy between the metabolism of the precursors would not exist. That is, if all these precursors of oxalate joined at a common pathway, they would all be affected identically by the Bg deficiency. These data are consistent with the oxidation of ethylene glycol and ethanolamine to oxalate via glycolate. Furthermore, in the isolated perfused rat liver, hydroxypyruvate has been shown to completely inhibit the formation of labelled oxalate from I 1-14C 1glyoxylate, even though total oxalate synthesis increased. In contrast, hydroxypyruvate increased the oxidation of labelled glycolate to labelled oxalate ( Liao and Richardson, 1978 ). The isolated perfused

rat liver also converts [ 1-*4C ] glycolate and

[ U-14C Jethylene glycol to l4C-oxalate more efficiently

than [ U-14C ]glyoxylate, further evidence supporting

direct metabolism through glycolate. These data all point

to the presence of active GAD in vivo and in vitro. Oxalate

precursors that have metabolic paths via glycolate include

glycoaldehyde, tyrosine, and phenylalanine ( Gambardella and

Richardson, 1977 ).

G ly o x y late

The third and major immediate precursor of oxalate is

glyoxylate. Weinhouse and Freidman ( 1951 ) report that

over 27 percent of intraperitoneally injeceted J-4C-

glyoxylic acid is excreted in a 24 hour urine as labelled

oxalic acid. Hodgkinson ( 1978 ) repeated this experiment

and found only 15.7 percent of the given label in the urine

as l4C-oxalic acid. This discrepancy is probably due to

the different analytical techniques employed.

Hydroxyproline ( Dekker and Maitra, 1962 ), tryptophan

( Gambardella and Richardson, 1977 ), and glycine ( Liao and

Richardson, 1972 ) are metabolized to oxalate via glyoxylate. 4-Hydroxy-2-oxoglutarate, a metabolite of

hydroxyproline, is an immediate precursor of glyoxylate. 10

This pathway, however, makes minor contributions to oxalate metabolism ( Williams and Smith, 1978 ). The metabolism of

tryptophan to indolepyruvate and then oxalic acid was

proposed by Faber et a l. ( 1963 ). Gambardella and

Richardson ( 1977 ) show glyoxylate to be the immediate precursor to oxalate in the tryptophan pathway. Thus,

indolepyruvate is converted to glyoxylate and is not directed metabolized to oxalate. The contribution of tryptophan and other aromatic amino acids to oxalate excretion is three percent or less ( Cook and Henderson,

1969 ). The importance of glycine as an precursor of oxalate is far less in the rat than in man. Less than one percent of [ l-^C ]glycine is metabolized to oxalate in the intact rat ( Hodgkinson, 1978 ) or rat liver ( Liao and

Richardson, 1972 ). In man, 30 to 50 percent of total oxalate synthesis is from glycine ( Crawhall et al.,

1959 ),

Glycolic acid can also be metabolized through glyoxylate to oxalic acid with the enzyme glycolic acid oxidase ( GAO ) catalyzing both steps ( Richardson and

Tolbert, 1961 ). Lactate dehydrogenase ( LDH ) can also catalyze the oxidation of glyoxylate to glycolate and glycolate to oxalate. LDH is discussed in more detail below. Serine is metabolized to oxalate via either glyoxylate or glycolate ( Richardson and Farinelli, 1981 ). 11

Enzymes of glyoxylate

The oxidation of glyoxylate to oxalate is catalyzed by three enzymes: xanthine oxidase ( XO ), GAO and LDH.

The in vivo contribution of XO to oxalate synthesis is minor. This is suggested by Gibbs and Watts ( 1973 ), who found 1) that allopurinol, an inhibitor of XO, had no effect on oxalate excretion, and 2) two patients with xanthinuria, a genetic deficiency of XO, had normal oxalate excretion.

LDH has been shown to use both glycolate and glyoxylate as substrates in vitro. The oxidation of glycolate to glyoxylate by LDH is one-hundredth the oxidation of glyoxylate to oxalate ( Sawaki and Yamada, 1966 ). The reduction of glyoxylate to glycolate by beef heart LDH was reported by Meister ( 1952 ). This reduction was confirmed in rabbit muscle LDH ( Sawaki and Yamada, 1966 ). The reaction products were demonstrated to be glycolic acid and

NAD ( Sawaki et a l., 1966a ). The pH optimum is 7.0

( Romano and C e rra , 1971 ).

The oxidation of glyoxylate by LDH was observed by

Sawaki et al. ( 1966b ) and oxalic acid was identified as a reaction product by Warren ( 1970 ) using crystallized pig heart LDH. The pH optimum of the oxidation of glyoxylate 12

with rabbit muscle LDH is 9.5 to 10.0 ( Sawaki et a l, 1966b;

Romano and C e rra , 1971 ). The Km f o r both g ly o x y la te and

lactate at pH 10.0 is 1.5 x 10~3 m ( Sawaki et al. ,

1967 ).

The third enzyme catalyzing the oxidation of glyoxylate

to oxalate is GAO. This enzyme requires the coenzyme flavin

mononucleotide for catalytic activity, has a pH optimum

between 8.2 to 8.8, depending on the substrate, and has a

broad specificity towards alpha-hydroxy acids. Glycolate is

the most effective substrate with a Km of 3.3 x 10“4 M.

The Kra for glyoxylate is 3.5 x 10"3 m and catalysis

occurs at a rate one-sixth that of glycolate ( Pry and

Richardson, 1979b ).

The relative contribution of GAO and LDH to oxalate

metabolism is controversial. Gibbs suggested that LDH is

the major enzyme that oxidizes glyoxylate to oxalate in the

liver. She chromatographically separated the liver enzymes

with " glyoxylate oxidizing activity " and found the NAD

dependent " glyoxylate oxidizing activity " of liver LDH to

be 10,000 tim es g r e a t e r th an th e non-NAD dependent a c t i v i t y of GAO and XO ( Gibbs, 1971; Gibbs and Watts, 1973 ). Williams and Smith ( 1971 ) further implicated LDH in oxalate metabolism from studies of hydroxypyruvate and its stimulation of LDH oxidation of glyoxylate. Hydroxypyruvate is postulated to accumulate in L-glyceric aciduria due to a genetic deficiency of D-glycerate dehydrogenase.

D-Glycerate dehydrogenase catalyzes the conversion of hydroxypyruvate to D-glycerate, as shown below. In the absence of this enzyme, hydroxypyruvate accumulates since the normal means of its removal is absent. Hydroxypyruvate is , therefore, excreted in abnormally high amounts in the urine in this genetic disease. LDH reduces hydroxypyruvate to L-glycerate in the presence of NADH. These investigators hypothesize a mechanism of L-glyceric aciduria as shown below:

glyoxylate^ ,NA ^L-glycerate murine

NADH hydroxypyruvate—//->D-glycerate

u rin e D-Glycerate Dehydrogenase

They believe that this mechanism accounts for the high concentrations of L-glycerate and oxalate in the urine of patients with primary hyperoxaluria type II ( L-glyceric aciduria). The L-glycerate originates from the hydroxypyruvate which has accumulated due to the deficiency of D-glycerate dehydrogenase. The conversion of hydroxypyruvate to L-glycerate is catalyzed by LDH. The large amount of NAD which results from this reaction can

then be used in the LDH mediated oxidation of glyoxylate to

oxalate. However, other investigators have shown that the

stimulation of LDH by hydroxypyruvate is an artifact.

Raghaven and Richardson ( 1981 ) have shown that

hydroxypyruvate decarboxylates radiolabelled glyoxylate

non-enzymatically. The method of estimation of

14C-oxalate used by Williams and Smith to evaluate the

ability of LDH to oxidize labelled glyoxylate requires the

collection of 1 4CC> 2 subsequent to the enzymatic decarboxylation of the labelled oxalate. Since, with the

addition of hydroxypyruvate, more 14CC> 2 is evolved, they make the false interpretation that the glyoxylate oxidizing activity of LDH is being stimulated. In fact, it is the non-enzymatic decarboxylation of glyoxylate catalyzed by hydroxypyruvate. Should the appropriate blank be run, the

w stimulation M of LDH by hydroxypyruvate is not seen. This explains earlier results in the isolated perfused rat liver, in which hydroxypyruvate inhibited the metabolism of glyoxylate to oxalate. An alternative explanation for the oxaluria observed in primary hyperoxaluria type II comes from these perfusion studies and data that show glycolate inhibiting the conversion of glyoxylate to oxalate. In addition, there is evidence that hydroxypyruvate is a precursor of oxalate. These data are consistant with the accumulated hydroxypyruvate being metabolized 15 to glycolate and then oxidized directly to oxalate via GAD

( Liao and Richardson, 1978 ). There is considerable evidence that GAO is the major glyoxyate oxidizing enzyme, at least where high substrate levels are involved. The proposal has been well reviewed by Richardson and Farinelli

(1981). Their salient points are:

1) There is a correlation between the degree of

oxaluria and the concentration of liver GAO in the

male rat, a relationship not seen with liver LDH

( Richardson, 1964 ).

2) Ethylene glycol is much less toxic in the partially

hepatectomized rat. Recall that GAO is restricted

to the liver, whereas LDH is ubiquitous in the

body ( Richardson, 1973 ).

3) DL-phenyllactate, a specific inhibitor of GAO,

completely inhibits the metabolism of ethylene

glycol, glycolate and glyoxylate to oxalate in the

isolated perfused rat liver ( Liao and Richardson,

1973). Further, glycolate and glyoxylate metabolism

to oxalate is reduced when DL-phenyllactate is

administered to isolated rat hepatocytes ( Rofe and

Edwards, 1978 ).

4) The concentration of GAO in the liver is adequate

to account for all oxalate synthesis in man and rat

( Fry and Richardson, 1979b ). 5) Ratios of oxalate to glycolate produced by the rat

given radiolabelled glyoxylate show that glycolate

metabolism is dependent on LDH in the totally

hepatectomized rat, whereas in the normal rat, GAO

is the major enzyme. Should LDH be the major

enzyme, glyoxylate would be oxidized to oxalate,

with the NADH produced used to reduce glyoxylate

to glycolate. The glycolate/oxalate ratio would

then be 1.0. The control rats given glyoxylate had

a glycolate/oxalate ratio of 7:1, showing major

involvement of GAO. The totally hepatectomized

animal had a ratio of 1.4:1, demonstrating LDH as

the primary enzyme ( Farinelli, 1981 ).

Also, investigators who previously disputed the role of GAO in oxalate metabolism have since conceded the significance of GAO, at least in the peroxisomes obtained from rat liver

( Gibbs et al., 1977 ).

One enzymatic defect in oxalate metabolism, L-glyceric aciduria, has already been discussed. There is a second documented error in oxalate metabolism. It, too, is a defect in the glyoxylate pathway. Patients with glycolic aciduria ( primary hyperoxaluria type I ) have a deficiency in the enzyme 2-oxoglutarate:glyoxylate carboligase in the cytosol ( Koch et al. , 1967 ). This enzyme catalyzes the synergistic decarboxylation of alpha-ketoglutarate and

glyoxylate to alpha-keto-beta-hydroxyadipate and CO2 . In the absence of this enzyme, glyoxylate accumulates, with the subsequent formation and accumulation of glycolate and oxalate. High levels of glyoxylate, glycolate and oxalate are excreted in the urine and are diagnostic of primary hyperoxaluria type I ( Williams and Smith, 1978 ).

Citrate Metabolism to Glyoxylate

The fo rm atio n o f g ly o x y la te from c i t r a t e is w ell known in plants and bacteria. This formation is part of the glyoxylate shunt, a metabolic cycle which allows the net synthesis of glucose from acetyl-CoA. In theory, this synthesis is does not occur in animals. The shunt is as f o llo w s :

acetyl-C oA c i t r a t e

cis-aconitate

oxaloacetat

i s o c i t r a t e

isocit. lyas •succinate

g ly o x y la te

malate synthase 18

The two enzymes supposedly unique to the plant and bacterial

kingdoms are isocitrate lyase and malate synthase. These

enzymes provide a bypass around the steps in the

tricarboxylic acid ( TCA ) cycle which evolve CO2 . These

CO2 evolving steps prohibit the net synthesis of glucose

from acetyl-CoA in the animal kingdom. The overall equation

for the glyoxylate shunt is:

2 A cetyl-CoA + NAD+ + 2 H2 O -----^ Succinate + 2 CoA +

NADH + H+

The succinate can then enter steps for gluconeogenesis

( Lehninger, 197 5; Tolbert, 1981 ).

Glyoxylate shunt activity has been demonstrated in

various animal tissues. A cyclic oxidation of glyoxylate

was proposed to occur in rat liver mitochondria by Payes and

Laties ( 1963 ). They demonstrated the conversion of gamma-

hydroxy-alpha-ketoglutarate to malate and postulated this

step as an intermediate in the cylic oxidation of glyoxylate. Brown and Box ( 1968 ) document part of the

glyoxylate shunt in the skin of the rat. They add

isocitrate to rat skin homogenate and establish a proportionality between the isocitrate added and the glyoxylate produced. Malate synthase activity is not shown.

Malate synthase activity is demonstrated in rat liver homogenate. I C ]Glyoxylate is incorporated into

malate, oxaloacetate, and citrate ( Liang and Ou, 1971 ).

Isocitrate lyase and malate synthase are demonstrated

together in the homogenate of toad urinary bladder

epithelial cell layer ( Goodman et al., 1980 ) and in the

homogenate of the fetal guinea pig liver ( Jones, 1980 ).

Citrate Metabolism to Oxalate

Hodgkinson ( 1978 ) demonstrated the conversion of

[ 1,5-l^c ]citric acid to 1 4 C-oxalate. He gave 5 uCi of the labelled citric acid to male Wistar rats intra- peri toneally. He found that 2.24 percent of the administered dose appeared in the urine as oxalic acid within 24 hours. The efficiency of citric acid as a precursor to oxalate was exceeded only by [ U-^4C ] glyoxylate ( 15.7 percent of the administered dose ) and

[ 1-14C JL-ascorbate ( 8.48 percent of administered dose ). Given that the TCA cycle is a ubiquitous and very active enzyme system ( Krebs and Johnson, 1937 ), 2.24 percent conversion of citric acid to oxalate would make citrate a major precursor of oxalate. 20

H ypothesis

It was decided to determine the pathway of citrate to oxalate. Hodgkinson hypothesized the path to be: citrate to isocitrate to glyoxylate to oxalate. The TCA cycle enzyme aconitase and the glyoxylate shunt enzyme isocitrate lyase would operate in sequence. Conversion to oxalate would occur via normal glyoxylate oxidizing enzyme catalysis.

Should this be the case, high doses of unlabelled glyoxylate after intraperitoneal administration of labelled citrate would trap the labelled glyoxylate in the large, cold glyoxylate pool. This would cause as an increase in radiolabelled glyoxylate and a decrease in radiolabelled oxalate when compared to a control animal not given the unlabelled glyoxylate. A novel method for the determination o f l^C -labelled giyoxyate was designed. Methods used to enhance the excretion of oxalate in the experiments included the intraperitoneal injection of unlabelled citrate and the injection of unlabelled malonic acid, a specific competitive inhibitor of succinate dehydrogenase ( Krebs and Johnson,

1937 ). 21

Methods

Experimental

The purpose of the study was to elucidate the pathway

of citrate to oxalate. Radiolabelled citric acid was given

intraperitoneally with and without a large intravenous dose

of unlabelled glyoxylate. If the metabolic pathway of

citrate to oxalate includes glyoxylate, labelling would be

trapped in the glyoxylate pool.

Male Wistar rats were fasted 18 to 24 hours with water

given aJ libitum. The rats were intraperitoneally injected

with 5 uCi of [ 1,5-l^C Jcitrate in the sodium form

( Amersham Corp.; Arlington Heights, IL 60005 ) in 1.0 mL

sterile saline. The specific activity of the citric acid

was 98 mCi per mmole. The radiochemical purity was 99

percent by thin layer chromatography done by Amersham and 96

percent by HPLC. There was no detectable radiolabelled

oxalate in the labelled sodium citrate. Sodium glyoxylate

( Sigma Chemical Co., St. Louis, MO 63178 ) in saline ( pH

5.2 ) was intravenously injected to seven rats at a dose of

1.25 mmole per kg rat weight with the volume never exceeding

1.0 mL. Each of the rats was placed for six hours in a glass metabolic cage where urine was collected and

radiolabelled * 4 C0 2 was trapped in 10 M sodium hydroxide ( Mallincrodt, Inc., Paris, KY 40361 ). One drop of

concentrated hydrochloric acid ( Fisher Scientific Co., Fair

Lawn, NJ 07410 ) was added to the urine collection container

as a preservative. After the six hours in the glass cage, each rat was placed in a stainless steel metabolic cage for

an additional 18 hour urine collection with water available

ad libitum. Seven control rats with no intravenous glyoxylate were also done.

Rats weighing more than 300 g did not survive the intravenous dose of glyoxylate. The average weight of the glyoxylate injected group was 277 g . The control group had an average weight of 296 g.

In order to slow down the metabolism of citrate and allow TCA cycle intermediates to build up, malonic acid

( Fisher Scientific Co. ) was given to two non-fasted rats.

A dose of 3.0 mmole per kg rat weight was given intra peritoneally simultaneously with the labelled citrate.

Unlabelled sodium citrate monohydrate ( Sigma Chemical

Co. ) was also given in hopes of enhancing citrate metabolism to oxalate. The citrate was given at a dose of

2 . 0 mmole per kg rat weight to two non-fasted rats simultaneously with the labelled citrate. Inadvertantly, a lethal dose of citrate ( 6.4 mmole per kg rat weight ) was 23 given to one animal. The effect of feeding was checked on one rat given an intraperitoneal injection of only

radiolabelled citrate.

Determination of 1 4 C~Oxalate

Tho amount of 4 ^C-oxalate in the urine was estimated using a highly specific oxalate decarboxylase from Collybia velutipes ( Sigma Chemical Co. ). The reaction is:

> H14COOH + 14CO

To each of three 2.0 mL aliquots of a 24 hour urine in 12 x

100 mm test tubes was added 0.5 mL 3 M hydrochloric acid.

To one of the aliquots was added a known spike of

1 4 C-oxalic acid ( New England Nuclear, Boston, MA 02118 ).

Each aliquot was heated to 60°C for 30 minutes to dissolve any oxalate crystals that may have precipitated out

( Drewes, 1974; Hodgkinson, 1981 ). The pH of the urine samples were adjusted to 3.0 and the aliquots were transfered to 25 mL erlenmeyer flasks. Two ml of 1.0 N citrate buffer, pH 3.0, was added to each flask. To the spiked specimen and to one of the other two aliquots, 0.2 U of oxalate decarboxylase ( 2 U per mL ) was added. The third aliquot was run as a blank. Each flask was immediately sealed with a stopper using an attached

suspended plastic well. This well contained 0.2 mL CO 2 trapping agent, a mixture of phenethylamine ( New England

Nuclear ), methanol ( Mallincrodt ), and toluene ( Lehigh

Valley Chemical Co., Easton, PA 18042 ) in the proportion

2:1:1, by volume, respectively. The well also contained a 1 x 3 cm piece of Whatman No. 1 filter paper which increased the surface area of the phenethylamine mixture. This reaction was run at 37°C to completion. The wells were then placed in scintillation counting vials with 4 mL

Thrift-Solve scintillation cocktail ( Kew Scientific, Inc.,

Columbus, OH 43227 ). The vials were counted in a Packard

Tri-Carb scintilation spetrophotometer ( Packard Instrument

Co., Inc., Downers Grove, IL 60515 ). With counting efficiency and the blank taken into account, the recovery was calculated for each urine analyzed. Using the l^C-oxalate internal standard, recoveries of 83 to 97 percent were obtained.

Determination of l4 C-glyoxylate

A novel method for the analysis of radiolabelled glyoxylate using crystallized LDH from rabbit muscle was designed. The reaction at pH 10 is: 25

> 1 4 C00H + NADH + H+

The oxalate formed was then analyzed as shown in the previous section using the oxalate decarboxylase of

Collybia velutipes.

Three 2.0 mL aliquots of a 24 hour urine were adjusted to pH 10 and tranfered to 25 mL erlenmeyer flasks. Two mL of a 1.0 M glycine-sodium chloride buffer, pH 10.0 was added to each flask. One flask was then spiked with a known radioactivity of [ U-14C ]glyoxylate ( New England

Nuclear ). To each flask was added 0.25 mL of a 0.079 M solution of NAD ( Boehringer Mannheim, Darmstadt, Germany ) and 0.1 mL of a 4 mg per mL solution of LDH in 0.1 M phosphate buffer, pH 7.0 ( about 100 U per assay ), was added to the spiked aliquot and one other aliquot. The third was run as a blank. The reaction was run to completion at 37°C. Next 0.1 mL of a 0.05 g per mL solution of DL-phenyllactate was added and the pH adjusted to 3.0. The DL-phenyllactate prevented the interference of radiolabelled glycolate in the assay. The mixture was then

analyzed for ^4 C-oxalate activity as described above.

Endogenous ^4 C-oxalate activity was subtracted from the final calculated dpm. About 55,000 dpm of radiolabelled glycolate with a specific activity of 5.5 mCi per mmole 26 interfered less than 0.4 percent with the designed assay.

The high specificity of the oxalate decarboxylase used would prohibit the interference of any other labelled compound.

Determination of Respiratory1 4 C0 2

Respiratory ■L^C02 was trapped in 10 M sodium hydroxide while the rat was in the glass metabolic cage for six hours. Aliquots of 0.1 mL were counted in triplicate using 4 ml Neutralizer neutralizing counting cocktail

( Research Products International, Corp., Elk Grove Village,

IL 60007 ) with 0.3 mL water added to prevent phase separation.

Total 14C Urine Activity

Aliquots of 0.1 mL urine were counted in duplicate in

Thrift-Solve ( Kew Scientific, Inc. ) with 0.3 mL water added to prevent phase separation.

Results and Discussion

As can be seen in Table 1, there is no significant

* 4 C-oxalate or l4 C-glyoxylate activity recovered in the urine of rats that is formed from [ l,5-^-4C ]citrate. 27

In the fasting rat, whether or not IV glyoxylate is

given, there is no oxalate formed from citrate. The extent

to which the citrate is metabolized to CO2 proves that the

animal received the dose and that the citrate was

incorporated into the metabloic pathways of the rat.

In comparing Table 1 to the results reported by

Hodgkinson ( 1978 ), there was also significantly less total

radioactivity excreted in the urine in this study, 4.76

percent in this study versus 2 0 . 2 percent of the

administered dose in Hodgkinson's work. The description of conditions used in Hodgkinson's experiment was not clear.

For example, the specific activity of the administered

[ 1,5-14c Jcitrate, as well as whether or not the rats were fasted prior to the experiments, were not given.

Therefore, a non-fasted rat was studied to see if fasting had an effect. Feeding might decrease the oxidation of labelled citrate via the TCA cycle by having the animal in a higher energy state. However, the results showed no

significant difference in urine ^ 4 C-oxalate and respired

1 4 C0 2 between the fasted and non-fasted rats.

Administering cold citrate along with the radiolabelled citrate might enhance its metabolism to oxalate. The increased availability of citrate in the animal might enhance non-TCA cycle reactions to occur. As shown in 28

Table 1. Results of the Study Showing the Metabolism of Citrate to Oxalate in the Male Wistar Rat

percent of administered dose

6 h 6 h 1 4C 24 h 14C 24 h 14c 24 h 14c

1 4 c o2 UR EXC UR EXC UR OX UR GLX

24 h f a s tin g 62.85 4.30 4.76 0 . 1 0 ND 14c-citrate s « l l . 51 s= 4.70 s= 4 .97 s= 0.16

24 h fasting 69.43 2.31 3.01 0 . 0 2 ND 14c -citrate s = l l .44 s = l .15 s = l .25 s= 0.06 + 1.25 mmol/kg g ly o x y la te fed 69.60 7.20 7.25 ND ------14c-citrate ____* fed 32.14 2 1 . 0 0 ND 14c-citrate + 6.4 mmol/kg c i t r a t e fed 27.35 41.45 45.05 ND 14c-citrate s= 0.92 s= 0 .35 s = l .34

+ 2 . 0 mmol/kg c i t r a t e fed 2 1 . 0 0 30.20 31.85 ND 14c-citrate s - 0 . 28 s=18.60 s=20.29 + 3.0 mmol/kg m alonate * = rat died after five hours ND =» none detected = test not done s = standard deviation Table 1, administration of cold citrate, even at a lethal

dose ( 6 . 2 mmole/kg ), did not enhance excretion of oxalate.

In these two groups the decrease in respiratory 1 ^ 0 0 2 can be predicted by the decrease in the specific activity of the citrate entering the TCA cycle. The increase in total activity excreted in the urine can be predicted from the literature. Alkalosis or administration of citric acid cycle intermediates enhance urinary citrate loss by inhibiting renal citrate reabsorption ( Baruch et al.,

1975 ).

Increasing the citrate in the animal may not have increased the oxalate formed because of the velocity with which citrate is oxidized. An inhibitor of an enzyme catalyzing an intermediate reaction after citrate in the TCA cycle would therefore slow down the removal of citrate.

Malonate, a specific inhibitor of succinate dehydrogenase, would be expected to enhance oxalate formation from the citrate that would accumulate from TCA cycle inhibition. As

can be seen by the decreased respiratory ^^C 0 2 in the malonate group, the inhibitor successfully slowed the metabolism of citrate. Despite this, the label was not found in urinary oxalate. Increased total l^C activity was seen in the urine, probably due to the inhibition of renal reabsorption. The duplication of Hodgkinson's experiment was done with many variables taken into account. Hodgkinson, in the analysis for l4 C-oxalate, boiled and filtered the urine. This procedure is not recommended due to the formation of oxalate from non-oxalate substances

( Hodgkinson and Zarembski, 1961 ). Exact duplication of

Hodgkinson's method showed no detectable oxalate in the urine. Hodgkinson stated that 2.24 percent of a 5 uCi dose of citric acid was metabolized to oxalate. This would mean

that 266,400 dpm of l 4 C-oxalate was excreted in a 24 hour period. Given an average 24 hour urine volume of 20.9 mL, the amount of *4C activity as oxalate would be 12,746 dpm per mL or 25,492 dpm in each assay, well within the sensitivity of the assay. The recovery for the oxalate determination is high and is calculated for each specimen.

This eliminated the assay procedure as a factor in my inability to duplicate Hodgkinson's results.

Conclusions

1. There is no significant conversion of IP injected

[ 1,5-14C ]citrate to urinary l 4 C-oxalate, as has been previously reported.

2. Feeding does not enhance the reported metabolism of citrate to oxalate.

3. Increasing the amount of administered unlabelled citrate did not enhance the reported metabolism of citrate to oxalate. 31

4. Partial inhibition of the TCA cycle by malonate did not result in the conversion of citrate to oxalate. C h apter I I

A Potential Assay for Oxalate Utilizing Coupled Enzymatic

R ea ctio n s

Literature Review: The Assay of Urinary Oxalate

The Use of Urinary Oxalate Estimation

The analysis of oxalate in urine is an elementary part of any evaluation of kidney stone disease ( Smith, 1979 ).

The motive for the analysis of oxalate lies in the frequency with which this substance occurs in kidney stones. Over 73 percent of all kidney stones have oxalate as a constituent of the stone. In more than 70 percent of these oxalate stones, over half the stone is composed of oxalate

( Herring, 1962 ). Oxalate stones are common because calcium oxalate is, by far, the most insoluble of the salts found in the urine ( Robertson and Rutherford, 1980 ).

Recurrent calcium oxalate stone formers excrete more calcium oxalate crystals in their urine than do normal subjects ( Robertson et al. , 1969 ). When recurrent stone formers are compared to normals, the level of spontaneously precipitating calcium oxalate is clearly different. The

32 33

saturation level is higher in stone formers than controls,

with 92 percent of the values exceeding the level of

saturation required for spontaneous precipitation of calcium

oxalate. Only 23 percent of the controls exceed this level.

Further, the degree of oversaturation is directly propor

tional to the amount of oxalate crystalluria ( Robertson et

al. , 1976 ).

Although it was thought that calcium concentration was

the primary cause of calcium oxalate crystallization, a

small increase in oxalate is much more efficient than an

increase in calcium in causing urinary crystalization

( Robertson and Nordin, 1976 ). Furthermore, an individual with only mild hyperoxaluria will more likely form stones than a person with only hypercalciuria ( Robertson et al.,

1978 ).

The use of the analysis of oxalate in urine in the diagnosis of stones is debatable despite the above correlations between crystallization and oxalate. Some investigators do not observe any relationship between the magnitude of oxalate excretion and the occurrence of calcium oxalate stones ( Dempsey et al., 1960; Mayer et al. , 1963;

Butz and Kohlbecker, 1980 ). Rapado et a l. ( 1979 ) report that only 0.4 percent of 3,158 patients with renal lithiasis have persistent hyperoxaluria, where the oxalate exceeded the upper limit of normal in at least two samples. In

contrast, Robertson and coworkers ( 1979 ) report a high

incidence of mild hyperoxaluria in idiopathic calcium stone

formers, particularly those with recurring stones. When

mild hyperoxaluria is defined as urine oxalate excretion

greater than 41.4 mg per day, 13 percent of single stone

formers and 23 percent of recurrent stone formers have this

condition after referal from the urologist. However,

oxalate excretion when measured more immediate to the stone

episode, before referral from the urologist to Robertson's

group, show a higher incidence of mild hyperoxaluria, 27

percent in single stone formers and 42 percent in recurrent

formers. Butz and Kohlbecker ( 1980 ) do report a

significant difference in urine oxalate concentration in the

first morning specimen. This difference is seen between male stone formers and male normals but is not seen in

females. Thomas et a l. ( 1979 ) find the mean level of

urine oxalate in 2 0 0 active stone formers to be significantly higher ( p < 0.001 ) than 98 non-active stone

formers, 41.5 mg per day and 33.5 mg per day, respectively.

At least three different measurements in daily oxalate excretion are taken on each patient. In non-active stone

formers only 18 percent of the cases are hyperoxaluric

( above 38 mg per 24 hours ), but in active stone formners,

63 percent of the group is hyperoxaluric. In another study,

Thomas et al. ( 1973 ) find a large difference between controls and stone formers in daily oxalate excretion, 30 mg and 50 mg per day, respectively. They advise analysis of a series of eight to ten 24 hour collections, since the hyperoxalauria can be intermittant in stone formers. Aiken et al. ( 1981 ) confirmed the utility of this approach, but improve on the series method by following oxalate concentration after a calcium load. Using the series of oxalate determinations, 46 percent of oxalate stone patients have hyperoxalauria. However, 77 percent of the recurrent oxalate stone formers can be discriminated from the non-oxalate stone formers by following oxalate concentrations at two and four hours after the oral calcium administration. The study is preliminary and merits a larger, more rigorous evaluation.

Hyperoxaluria in genetic, toxicological and surgically induced states has already been discussed in Chapter I.

The controversy around the effectiveness of oxalate determinations will probably continue until a quick, easy, specific and precise method for urine oxalate is found.

What follows is a review of this methodology so that the nature of the problem can be appreciated. 36

Separation of Oxalate

The analysis of oxalate usually requires the separation

of oxalate from interfering substances present in the urine.

Newer enzymatic methods do not always require this

separation due to the specificity of the enzymes utilized,

although inhibitors in urine become a factor here.

Precipitation: The oldest method of oxalate separation is

by precipitation, usually as the calcium salt, but cerium,

europium, lead and thorium have been used.

The* calcium salt precipitation is the classical method of separation. Although the salt is considered to be

insoluble, its solubility in water is actually 6 to 7 mg per

L water at room temperature. Furthermore, the rate of crystalization is affected by many substances in the urine.

Magnesium and, to a lesser extent, sulfate, sodium and potassium increase calcium oxalate solubility ( Elliott and

Eusebio, 1965 ). Urea, citrate and lactate also increase solubility ( Miller et al., 1958 )* In addition, inorganic pyrophosphate ( Fleisch and Besaz, 1964 ) and polyelectrolytes such as heparin ( Crawford et al. , 1968 ) inhibit the precipitation of calcium oxalate from urine.

There are substances which co-precipitate with the calcium oxalate including phosphate and sulfate, uric acid ( Lonsdale, 1968 ) and citric acid ( Powers and Levatin,

1944 ), which in some cases can interfere with analysis. A widely used method is described by Archer et a l. ( 1957 ).

Urines are adjusted to pH 5 and the oxalate is precipitated for 16 hours at room temperature. The suspension is then centrifuged, the supernatant decanted and the precipitate washed with dilute ammonia solution. Reported recoveries range from 90 to 93 percent using the permanganate titration. Koch and Strong ( 1969 ) show about a 20 percent loss of oxalate using the ammonia wash, when compared to washing the precipitate with calcium oxalate saturated water. Thus, the wash step is a source of loss in precipitation methods.

Recognizing the numerous influences on calcium precipitation, modifications have been proposed to compensate for the variations in recovery. Fraser and

Campbell ( 1972 ) add known quantities of stable oxalate as an internal standard. Recoveries are 80 to 115 percent when analyzing the calcium not precipitated by oxalate with atomic absorption spectroscopy. Koch and Strong ( 1969 ) use radioactive oxalate to compensate for losses; recoveries are 83 to 100 percent.

Losses during precipitation are also addressed by

Zarembski and Hodgkinson ( 1965a ), when they propose 38

precipitation of calcium oxalate by precipitating calcium

sulfate at the same time. The precipitation of oxalate is

further enhanced by the addition of ethanol. They find

recovery to be complete employing fluorimetric analysis.

However, in the hands of others, recovery ranges from 7 5 to

108 percent ( Husdan et al., 1976 ).

The calcium precipitation of oxalate has been used to

quantitate the oxalate in urine. The excess calcium in the

supernatant after precipitation is measured using an KDTA

complexometric method. Recovery is adequate, 8 8 to 113

percent, but the normal range is low, 0 to 36 mg per 24

hours (Giterson et al. , 1970 ). The excess calcium can also

be measured using atomic absorption spectroscopy ( Menache,

1974 ). Phosphorus does not interfere because of the pH at which the precipitation of oxalate is carried out.

Recoveries are 85 to 102 percent; however, no internal standard is used. The normal range is 0 to 40 mg per day.

Another variation is the measurement of the calcium content of the precipitated oxalate with atomic absorbtion ( Fraser and Campbell, 1972 ). With an internal standard, recoveries range from 80 to 115 percent, with a coefficient of

variation of 8 . 6 percent. A normal range of 123 to 36 mg per

24 hours is reported on a limited number of urines. 39

Kamiya et a l. ( 1937 ) use cerium to precipitate

oxalate from blood with subsequent iodometric analysis. The

normal values he reports are very high compared to others.

The technique has not been used in urine.

Europium has been used as a precipitator by Vittu and

Lemahieu ( 1965 ), since the europium salt of oxalate is

less soluble the the calcium salt, i.e., 1 mg per L versus 6

to 7 mg per L, respectively ( Seidel, 1965 ). Inorganic

phosphate is first precipitated using magnesium chloride

since the phosphate interferes with oxalate precipitation.

However, magnesium ion also interferes with europium oxalate

precipitation, so excess addition of magnesium is avoided by quantitating the phosphate in the sample and adding

appropriate amounts of magnesium. Recoveries range from8 8 to 113 percent using a polarigraphic estimation.

Pernet and Pernet ( 1965 ) analyze serum oxalate as a lead salt. The normal values given are in line with accepted values of serum oxalate, but as with cerium, it has not been applied to urine.

Thorium oxalate, whose solubility is 0.7 mg per L, is the analyzed salt in the technique of Glenister and Alvi

( 1969 ). This is a gravimetric analysis of the Th0 2 after ignition of the oxalate salt. 40

Solvent Extraction: Separation of oxalate by solvents has traditionally involved diethyl ether. Powers and Levatin

( 1944 ) devised a method that requires six hours for extraction and results in a 10 percent loss of oxalate. The non-specificity of the extraction is " fortunate " since the permanganate quantitation used is non-specific. That is,

when oxalate is extracted from urine, the 1 0 percent loss of oxalate is compensated by the co-extraction of citric acid, which reacts in the permanganate titration.

Modifications of this extraction use peroxide free ether ( Yarbro and Simpson, 1956 ) and an improved continuous extraction apparatus ( Hodgkinson and Zarembski,

1961 ). With these improvements, recoveries are brought to

98+2 percent ( Hodgkinson and Zaremski, 1961 ).

A disadvantage of ether is that the partition coefficient for oxalate is about 0.1. Furthermore, ether is slightly soluble in water, i.e., 13 mL ether per L water.

This prompted Zarembski and Hodgkinson ( 1965a ) to find an alternative solvent, tri-n-butyl phosphate. Although inorganic phosphate and other electrolytes are co-extracted, quantitative extraction of microgram quantites of oxalate is possible in a single five minute extraction at room temperature. Ion Exchange Chromatography: The use of ion exchange for

oxalate purification was first used by Pik and Kerckhoffs

( 1963 ). They use Dowex 1 x 8 , a strongly basic anion

polystyrene resin, and although "satisfactory" ( no data

given ), the method is time consuming, since urine has to be

degassed before application to the column. Also, a high

salt concentration in some urines displaces the oxalate from

the column too early. An ether extraction was pursued

instead. Chalmers and Watts ( 1972 ) employ DEAE-sephadex

in their pretreatment. Recovery is quantitative using 15 mL

of 1.5 M pyridinium acetate buffer. Dowex 2 x 8 , a strongly

basic anion exchanger, has been utilized. Recoveries from

82 to 107 percent result with 100 mL 0.05 M HC1 ( Johansson and Tabova, 1974 ). Kruger Dagneaux et a l. ( 1976 ) use a weakly acidic anion exchange resin , Baker CGA 315, to treat plasma. The method is adapted to urine using another weak acid exchanger, Lewatit MP 7080. Application of six mL of

2.5 M H 2 SO4 gives recoveries of 98 + 2 percent ( Olthius et al., 1977 ). Invariably, in spite of recoveries being satisfactory, the ion exchange methods are time consuming and not conducive to routine use. 42

Quantitation of Oxalate

Once separated, oxalate may be reduced to glyoxylate or glycolate and those acids measured, or the oxalate measured directly. Earliest methods involved direct measurement.

Gravimetry; These are of historical interest only, due to their laborious nature. Lehmann ( 1851 ) evaporates urine to dryness, extracts the residue with ethanol, washes the

extract with ether, and weighs the amount of CaC0 3 p re s e n t in the ignited residue. Salkowski( 1900 ) estimates oxalic acid by extracting acidified urine with ether, precipitating oxalate as the calcium salt, and weighing the CaO formed by i g n i t i o n .

Titrimetry: Titration with potassium permanganate is another early method used by Dakin ( 1907 ). The oxalate is precipitated as calcium oxalate, extracted with ether, and then reprecipitated as calcium oxalate. The precipitate is then titrated with permanganate. The reaction is :

5 C 2 0 4 2- + 2 MnC>4 “ + 16 H+ ------>

10 C02 + 2 Mn2+ + 8 H2 O

A persistant pink color of KMn0 4 marks the end point.

Great care must be taken to standardize the temperature since at temperatures below 70°C, chloride interferes

( Baxter and Zanetti, 190 5 ) and at temperatures that are too high, oxalate decomposes ( McBride, 1912 ). Interfer

ences, as discussed above in the technique of Powers and

Levatin, are also a disadvantage. Archer et al. ( 1957 )

propose a simplified pretreatment. Recoveries are about 92

percent. The Pik and Kerckhoffs modification ( 1963 )

improves the extraction step. Although a reviewer contends

that titrim etric methods give falsely low results, with

normals from 3 to 28 mg per day ( Hodgkinson, 1978 ), at

least one study contradicts this. Gratzlova and Revusova

( 1971 ), using the method of Pik and Kerckhoffs, find

excellent recoveries with this method, 98.6 + 4.9 percent.

They show the reproducibility to be good, with the

coefficient of variation of 3.6 percent. Their observed

normal range is 15 to 44 mg per 24 hours for adults, a range

comparable to isotope dilution methods.

Adaptations of the titration technique are manifold.

One of the earliest is the gasometric technique of Van Slyke

and Sendroy ( 1929 ). The Van Slyke manometric apparatus is

used to measure the CO2 evolved from permanganate that is

added to excess.

Another variation on the titration theme is the use of cerate for the oxidation of oxalate:

C2°42” + 2 Ce4+ ------> 2 CO2 + 2 Ce3+

Koch and S trong ( 1969 ) d e s c rib e th e use o f c e r r i c ammonium 44

nitrate with a nitroferroin indicator

Electrophoresis: Direct quantitation of oxalate by

electrophoresis had been semi-quantitative until isotacho-

phoresis was applied to the problem. "Iso"-"tacho" implies

equal velocity, so ions travel with equal speed within an

electric field applied to a solution in a capillary tube.

Separation is dependent on the size of the molecule and the

net charge in the buffer system. The amount of a given ion

is directly related to the length of the zone to which it

migrates. Detection within the tube utilizes either UV

absorbance of the ions or the heat given off by the ions as

they migrate. There is no sample pretreatment needed on the

urine. The coefficient of variation of the method is 3.2

percent. Normal values ranged from 17 to 73 mg per 24 hours

( Tschope and Ritz, 1981 ). Related compounds, such as oxaluric acid, glyoxylic acid, and ascorbic acid were shown not to interfere ( Schmidt et al., 1980 ). Ferric ions, which originate from the needle of the sampling syringe, have been shown to interfere with analysis. This interference can be overcome by loading the sample through a teflon or glass cannula, or the addition of EDTA to decompose the iron oxalate complex ( Fredriksson, 1980 ).

The instrument needed for this type of detection is not in a typical clinical laboratory. It is questionable whether the purchase of an expensive research instrument to be employed 45

solely for the analysis of oxalate in urine is justified.

The time need for sample throughput was not addressed.

Photochemistry: The use of ferric alum and light in an acid

medium to quantitate oxalate was first proposed by Rao and

Aravamudan ( 1955 ). Ferrous iron is formed, which is then

estimated with sodium vanadate. Utilization of the decarb

oxylation of oxalate by iron and light for the analysis of

oxalate is reported by Riggs and Bricker ( 1966 ). Cooley

and Kratochvil ( 1978 ) automated this principle using

Techniccn Autoanalyzer components and a flow through CO2

electrode. They eliminate some interferences, such as mannitol, but interferences from acids such as citrate, malate, pyruvate and malonate render the system inapplicable to u rin e .

Gas Chromatography: The direct analysis of oxalate by gas chromatography ( GC ) requires derivatization once the oxalate is separated. Ch

Chromosorb 102. A flame ionization detector is used.

Recovery from the tri-n-butyl phosphate extraction is 60 to

80 percent. Replicates of an 11.5 mg per day urine exhibit

a 1.4 mg per day standard deviation. Trimethyl silyl derivatization has also been used.

After elution from a cation exchange resin, the eluent is

evaporated to dryness and derivatized. A flame ionization

detector is used to detect peaks of the Chromosorb G loaded

with 5 percent SE 30 ( von Nicholai and Zilliken, 1974 ).

Use of an electron capture detector does not increase

sensitivity but does double the range of linearity. The oxalate is precipitated from urine with calcium sulfate and e s t e r i f i e d w ith 2 - c h lo r o e th a n o l. R e te n tio n is on OV-17 on

GasChrom Q. The range of calibration, 5 to 40 ug, has a coefficient of variation of 7 percent at each point. The normal range is 15.1 to 51.8 mg per 24 hours ( Tocco et al. ,

1979 ).

One group proposes the deletion of the extraction step.

The urine is evaporated, the residue dissolved in ethanol and sulfuric acid added to catalyze the ethylation. The column used is 15 percent diethylene glycol succinate on

Chromosorb W. The method is not different from a method employing an ether extraction, with no decrease in column life. Normal urinary excretion runs from 10 to 45 mg per 47

Gas capillary GC of the dimethyl ester of oxalate is done on urine. However, backflushing techniques using complicated valving are required to shorten analysis time from two hours to eight minutes ( Dosch, 1981 ).

A mass spectrometer as a GC detector for oxalate has been suggested. Tocco et al. ( 1979 ) used it as a reference method and correlation with their GC/electron capture method is good, with r = 0.9918. Automated metabolic profiling for organic aciduria has been descibed

( Liebich et al. , 1980 ) employing methyl ester derivatization after an ethyl acetate/diethyl ether extraction. The analysis required over two hours to run on the GC/NS and requires a dedicated computer for data management. Duggan et al. , ( 1979 ) use a GC/MS to estimate pharmacokinetic parameters of oxalate in man. They used

[ 1 , 2 - 13C2 loxalate as an internal standard and make di-n-propyl esters of oxalate. Biosynthetic rates calculated from the product of the measured pool size and turnover rates are in excellent agreement with urinary excretion rates, confirming that urinary oxalate is a quantitative index of biosynthesis. Regardless of what detector is used in GC analysis, sample preparation time is considerable, with only one sample at a time being analyzed.

However, automated instruments help minimize the later disadvantage. 48

Colorimetry: Direct colorimetric determination of oxalate has been done utilizing the inhibition by oxalate on the

formation of a red uranium IV/4-( 2-pyridylazo ) rescorcinol complex. The reaction is adapted to a Technicon

Autoanalyzer, sampling the oxalate precipitated by calcium.

Normal daily excretion averages 31.5 mg per day, which is comparable to other methods ( Baadenhuysen and Jansen,

1975 ).

A direct determination of oxalate using its ability to accelerate the oxidation of ferroin by chromium is reported by Dutt and Mottola in 1974. A solvent extraction and calcium precipitation is required since citric acid, manganese and lead also accelerate the reaction. Urine normals average about 16 mg per day, which is low by accepted standards.

Some investigators prefer not to measure the oxalate directly, but rather reduce the oxalate to glyoxylate or glycolate, and then measure these colorimetrically or fluorimetrically. The oxalate must first be separated. The conditions for reduction to glyoxylate are critical using zinc and HC1. Some zinc samples are unable to reduce the oxalic acid. Also, the zinc has to be "activated" in nitric acid and the reduction step conducted at 0°C. Failure to a d h ere to th e se c o n d itio n s r e s u l t s in as much as 20 to 30 49 percent of the oxalate being reduced to glycolate

( Zarembski and Hodgkinson, 196 5a ). The glyoxylate is converted to a colored phenylhydrazone by reacting with phenylhydrazine followed by oxidation with either ferricyanide ( Pernet and Pernet, 1962 ) or hydrogen peroxide ( Nishi and Shimizu, 1978 ).

The most commonly used colorimetric reactions use reduction of oxalate to glycolate with zinc and sulfuric acid. Dempsey et al. ( 1960 ) first adapted the use of chromotropic acid ( 1,8-dihydroxynaphthalene-3,6-disulfonic acid ) to the analysis of oxalate in urine. After the oxalate is reduced to glycolate by powdered zinc and sulfuric acid, the solution is heated in the presence of chromotropic acid and sulfuric acid. The hot sulfuric acid converts the glycolate to formaldehyde which specifically forms the violet complex with chromotropic acid ( MacFayden,

1945 ). Hodgkinson and Zarembski optimized the reaction conditions in 1961; in 1972 Hodgkinson and Williams published a modification that eliminated the earlier six hour ether extraction by using precipitation of calcium oxalate with calcium sulfate and ethanol. The normal range is 17 to 43 mg per day, with a coefficient of variation of

5.2 percent. Recoveries range from 75 to 98 percent. The only interference is a 35 percent positive interference froa glucose at the level of 3 g per 100 mL, a level that may be achieved in severe diabetics. Twelve samples could be analyzed in a day. Investigators using a cumbersome ion exchange pretreatment report similar normals, 11.2 to 45.4 mg per day ( Olthius et al., 1977 ).

Fluorimetry: The only fluorimetric analytical procedure for urine is that of Zarembski and Hodgkinson ( 1965a ). It requires the less reliable reduction to glyoxylate with subsequent addition of resorcinol. The yellow complex absorbs at 490 nm and emits at 530 nm. Sensitivity is 1 ug per mLf and a coefficient of variation of 5 percent is observed with aqueous standards. Normal range is given as

6.3 to 38.7 mg per day, which is a slight underestimation.

This may be due to the incomplete extraction of oxalate or its incomplete reduction.

Isotope Dilution: Considered to be the most accurate determination, isotope dilution is the reference method in the view of most investigators. The methods all depend on the equation: stable oxalate in urine *» total radioactivity added measured specific activity

Hockaday et al. ( 1965 ) adapted the isotope dilution procedure of Archer et al. ( 1957 ). Oxalate is precipitated as the calcium salt after addition of 51 i^c-oxalate, and then reduced to glycolate with zinc dust and sulfuric acid. The glycolate is isolated using Dowex 1 x 8 and is estimated colorimetrically with

2,7-dihydroxynaphthalene. The absorbance of the purple colored complex is read at 530 nm. Recoveries range from 99 to 103 percent. Normals are 18.0 to 47.0 mg per day. Other isotope dilution techniques utilize alternate means of estimating specific activity. Gibbs and Watts ( 1969 ) extract the oxalate with tri-n-butyl phosphate, precipitate the oxalate with calcium, and then use chromotropic acid.

Normal range is 24.0 to 49.0 mg per 24 hours, similar to

Hockaday et a l. Johansson and Tabova ( 1974 ) use a technique akin to Hockaday, but separate the oxalate on

Dowex 2 x 8 and concentrate the oxalate as a calcium oxalate precipitation. Recoveries are 82 to 107 percent, with normal values being 8.4 to 49.8 mg per day. The values correlate well with the values reported of Zarembski and

Hodgkinson ( 1965b ).

Enzymatic; Oxalate Oxidase ( EC 1.2.3.4 ); The oxalate oxidase from barley was the first oxidase to be used in oxalate analysis. Kohlbecker et al. ( 1979 ) use the oxidase from barley to catalyze the reaction:

( COOH ) 2 + A 0 2 ------> 2 CO2 + H2 O2 2

The reaction is carried out in a closed vessel in 52

succinate/EDTA buffer at pH 3.8. The CO2 released is collected in a bicarbonate buffer, pH 10. The pH of the bicarbonate buffer decreases in proportion to the oxalate in

the urine. No sample preparation is required. However, the reaction has to be run for 16 hours. Recoveries run from9 2 to 109 percent, with a coefficient of variation of 5.6 percent. Butz and Kohlbecker ( 1980 ) later coupled the oxidase with catalase ( EC 1.11.1.6 ) and aldehyde dehydrogenase ( EC 1.2.1.3 ) ( ADH ) to colorimetrically determine the amount of peroxide evolved in the oxidase r e a c tio n :

H20 2 + CH3 CH2OH c a t a l a s e > CH3CH0 + 2 h20

CH3 CHO + NADP + H2 0— — >CH3COOH + NADPH + H+

This coupled assay can be used to analyze oxalate in both serum and urine. Urine normals are low, from 7.4 to 32.8 mg per day. Recovery data is not given. It may be that partial inhibition of the enzyme by the raw urine accounts for the low normal range.

The use of oxalate oxidase from moss has been reported.

In this case, the indicator reaction utilizes peroxidase

( EC 1.11.1.7 ) to oxidatively synthesize a purple indamine

chromogen from 3 -m e th y l-2 -benzothazolinone hydrazone,

N,N-dimethylanaline, and hydrogen peroxide. A cation exchange column is utilized to remove oxidase inhibitors, 53 such as sodium chloride and divalent metals, which are found

in urine. Recoveries are reported as 95 to 109 percent. A normal range on four individuals shows results from 28 to 43 mg per 24 hours. The oxidase from barley has been adapted to the Technicon Autoanalyzer by immobilizing the oxidase on a nylon tube. This scheme has not yet been applied to urine

( Bais et al. , 1980 ).

Enzymatic: Oxalate Decarboxylase ( EC 4.1.1.2 ): The reaction catalyzed by oxalate decarboxylase is:

( COOH )2 ------> HCOOH + C02

First reported by Shimazono in 19 51, the enzyme was partially purified in a survey of wood destroying fungi

( Shimazono, 1955 ). The enzyme, from Collybia velutipes, is most active at pH 3.0 and most stable at pH 4.5. It does not require ATP, coenzyme A, thiamine pyrophosphate, acetate or magnesium, which is in contrast to decarboxylases from plants and bacteria ( Shimazono and Hayaishi, 1957 ).

Maximal synthesis of the enzyme requires addition of oxalate after 25 days growth and harvest three days later. The enzyme is highly specific. Another fungi, Aspergillus phoenicis, has been shown to synthesize oxalate decarboxylase. The enzyme has a higher pH optimum, about

5.2. Also the enzyme has the advantage of not requiring induction of the fungus with oxalate, and requiring only three days growth for enzyme synthesis. It, too, has no 54

cofactor requirements and is highly specific, but the enzyme

does require oxygen as a co-catalyst, i.e., oxygen is not

consumed in the reaction. But there is no decarboxylation

under anaerobic conditions. The enzyme has been shown to

be highly specific. The only inhibitors shown thus far are

strong reducing substances such as hydroxylamine, sulfite,

and dithionite ( Emiliani and Bekes, 1964; Emiliani and

Riera, 1968 ).

Recently, an oxalate decarboxylase has been reported in guinea pig liver ( Murthy et a l., 1981 ), but the enzymes

from fungus are the only decarboxylases used for oxalate analysis at the present time.

The oxalate decarboxylase from C. velutipes has been more extensively utilized for oxalate analysis, probably because the enzyme was the first to be available commercially. The enzyme is used in the radioenzymatic mode and coupled to a color reaction that detects either the

formate or the CO2 products. The measurement of CO2 v ia

pH change, a CO2 electrode or a conductivity electrode has also been used.

The radioenzymatic method for the analysis of oxalate in urine was first published by Bennett et al. in 1978. The

oxalate is precipitated overnight with CaCl 2 / acid ether extracted, and evaporated to dryness. Recovery is 98.4 percent. The oxalate is decarboxylated in an enclosed test tube in the presence of 22.5 nCi 14C-oxalic acid.

*4C 0 2 is collected in a plastic well containing a CO2 absorbing solution. The standard curve, plotted as oxalate

concentration versus the reciprocal of the cpm of •L4C0 2 trapped, is linear from 0.25 to 10 micrograms per mL.

Interassay coefficient of variation is 10.9 percent.

Phosphate and sulfate, which are known inhibitors of the decarboxylase, are removed by this sample preparation, as shown by a line parallel to the standard curve when oxalate is determined over an eight-fold dilution of extracted u rin e . The 24 hour e x c r e tio n was n o rm ally 5.1 to 46.1 mg, which is comparable to the isotope dilution methods.

The first application of oxalate decarboxylase to the

analysis of urine followed CO2 evolution using the Warburg technique. Mayer et al. ( 1963 ) describes this procedure, which requires calcium precipitation, uses stable oxalate internal standardization and has no solvent extraction.

The reaction is run to completion, usually about one hour.

Note that 30 mg phosphate, an inhibitor of the enzyme which co-precipitates with oxalate, extends that period to one and one half hours. This level of phosphate exceeds that normally entrained in the precipitate. They show the addition of EDTA to increase recovery from about 7 5 percent 56

to an average recovery of 95 percent. The normal range is

4.7 to 36.3 mg per day, low perhaps due to enzyme

inhibition. Ribiero and Elliot ( 1964 ) eliminate the need

to separate oxalate by adding the enzyme in excess. Urine

pretreatment involved deproteinization by ultrafiltration

and then flash evaporation, a four hour process. Recoveries

range from 81 to 117 percent. Hallson and Rose ( 1974 )

also add enzyme in excess to minimize inhibition by phophate

and sulfate. The amount of enzyme needed to overcome this

inhibition is 0.15 units per assay. They detect CC> 2 using

the pH decrease in alkaline bicarbonate buffer in the 16

hour incubation time. The normal range is 12.9 to 41.3 mg

per day.

The detection of CO2 using this enzyme was automated

by Knowles and Hodgkinson in 1972. The method is designed

for serum and though very sensitive, is quite inconvenient.

Sample pretreatment is extensive, inhibitors have to be

compensated by their "representative" addition to the

aqueous standard, and the entire Autoanalyzer has to be

operated in a glovebox to supply a C0 2 -free environment.

CO2 from the enzyme reaction has also been measured

' utilizing a conductimetric change in the CO2 absorber,

Sr(OH)2* The method was applied to rat urine, but not

human urine ( Sallis et al. , 1977 ). A CO2 electrode has

also been joined with the enzyme reaction, but this system 57

was not applied to urine ( Yao et al. , 1975 ).

The decarboxylase has also been immobilized on a CO2

electrode membrane by both entrapment and covalent

attachment. Covalent attachment may be promising, judging

by the lack of any inhibition by either phosphate or

sulfate, and by the 95 to 100 percent recovery of oxalate

seen in urine ( Kobos and Ramsey, 1980 ).

An indicator enzyme reaction has been used to d e t e c t

the formate produced during the decarboxylase reaction.

Jakoby ( 1974 ) uses formyl tetrahydrofolate synthetase ( EC

6.3.4.3 ) from the bacterium Clostridium cylindrosporum to detect as little as 5.4 micrograms oxalate per mL of urine.

The reaction is:

formate + tetrahydrofolate + ATP ------> Nl0_fOrmyltetrahydrofolate + ADP + Pf

The optimal pH for the reaction is 8.0. In acid, the

N^°-formyltetrahydrofolate is converted to

5,10-methylenetetrahydrofolate, which absorbs at maximally

350 nm. Costello et al. ( 1976 ) describe a two step enzymatic assay using formate dehydrogenase ( EC 1.2.1.2 ) to detect formate:

form ate + NAD+ ------> CO2 + NADH + H+

NADH absorbs at 340 nm. The formate dehydrogenase is isolated from Psuedomonas oxalaticus, and has a pH optimum 58 respective pH, the first step taking one-half hour, and the second step being continuously monitored to endpoint, six to ten minutes. NADH formation is proportional to added oxalate but not stoichiometric. Therefore, the extinction coefficient has to be determined wwith each batch of formate dehydrogenase. Citrate extraction requires overnight treatment, yielding recoveries from 96 to 103 percent.

Normal values range from 14.0 to 37.6 mg per day.

A two step enzyme method employing the C. velutipes enzyme and the formate dehydrogenase from yeast has been adapted to the Technicon Autoanalyzer ( Yriberri and Posen,

1980 ). The urine must be pretreated using a calcium precipitation and extraction of the precipitate into citrate buffer. Recoveries are poor and imprecise, namely41 percent with a standard deviation of19 percent. Recovery of stable oxalate is94 t o 98 using a radiotracer to compensate for extraction efficiency. Normal daily excretion is reported to be18.0 t o 4 6 .0 mg. Two and one-half days are required for the analysis of30 s p e c im e n s , but other duties can be performed during that time.

The use of these same enzymes in a technique with no extraction was reported by Chalmers in 1979. The two step procedure has a 20 minutes total incubation time, and had recoveries from 98 to 100 percent. Seven healthy adults 59 have values from 7.1 to 34.3 mg per day.

The assay of oxalate in urine using the enzyme from A. phoenicis has not been reported. The assay of oxalate in aqueous standards has been shown to involve a single step requiring a 30 minute incubation at a single pH. However, large quantities of enzyme, 10 U of oxalate decarboxylase and 15 U of formate dehydrogenase, are required. The sensitivity of the reaction, 270 micrograms per mL, is not good enough for analysis of oxalate in unconcentrated urine

( Beuttler et al., 1980 ).

H y p o th esis

The need for a fast, simple, precise, sensitive, specific assay for urinary oxalate that is efficient in sample throughput is obvious. The absence of such a method in the literature is equally obvious.

The oxalate decarboxylase ( OD ) from A. phoenicis reported by Emiliani and Bekes ( 1964 ) has a pH optimum of

5.2. Also a formate dehydrogenase ( FDH ) from soybeans with a pH optimum curve flat from pH 5.8 to 7.5 has been isolated and characterized in our laboratory. The pH optimum is considerably lower than that of FDH from any other source ( Farinelli, 1981 ). The potential for these 60

enzymes to be used for the analysis of oxalate in urine was

therefore investigated. The coupled reaction would be:

o x a la te-----— ------> formate + CO2

formate + NAD-----— > CO2 + NADH + H+

To carry out this coupled reaction quickly at a single pH with minimal sample preparation would be of significant aid

to the clinical chemist and biochemist.

The problem had several facets to it. First,

conditions conducive to enzyme synthesis had to be found.

The fungus was used previously in our laboratory and at some point, all cultures of the fungus died. When reordered from the American Type Culture Collection ( ATCC ) the same

fungus no longer had oxalate activity when grown under the

"same" conditions. Second, the enzyme had to be isolated and purified to be used for the assay. Third, a buffer system had to be established that would have the pH, concentration of cofactors, and concentration of enzymes necessary for the assay of oxalate. Fourth, the assay would have to be adapted for the analysis of urinary o x a la te . 61

Methods

Experimental

The first part of the problem was to develop a maintenance protocol for the fungus, Aspergillus phoenicis, that would promote maximal oxalate decarboxylase activity.

Spores were purchased from the ATCC ( Rockville, MD 20852 ),

ATCC number 12847. The lyophilized spores were handled per

ATCC instructions.

Raper and Fennel ( 196 5 ) in the text The Genus

Aspergillis recommend the maintenance of cultures on agar slants. The "reference” medium, the medium used by most investigators to maximize and standardize sporulation, is

Czapek Solution Agar. This was the media found to be most satisfactory as far as quantity of spores produced and survival of spores at -20°C. Freezing eliminated subculturing after the spores were inoculated on the slants.

The fungus, maintained on agar slants frozen at

-20°C, was then inoculated on several liquid media for the purpose of growing the fungus for subsequent enzyme isolation. Liquid medium was used since it could be readily washed off the mycelia. Several media were tried. The original medium, as reported by Emiliani and Bekes ( 1964 ) 62 was inoculated. The carbon source was changed from sucrose to glucose, and this medium planted. As there is a correlation between the amount of citric acid produced and the amount of decarboxylase synthesized ( Emiliani and

Bekes, 1964 ), several attempts were made to maximize citric acid production. Four trace metals are required for optimal citric acid production. These were added in concentrations

reported as optimal by Tomlinson et al. ( 1951 ). ZnSC>4

( 500 ug ), FeCl 3 ( 110 ug ), CUSO4 ( 40 ug ) and

MnCl2 ( 3 ug ) were all added to a liter of media. Media containing these additions of metals were that of Emiliani and Bekes ( 1964 ) and that reported by Fry ( 1978 ).

Induction of the enzyme by the addition of two percent methanol to the medium was attempted, as this was reported to increase citric acid production ( Moyer, 1953 ). The alcohol was added to the media reported by Tomlinson

( 1951 ) and that of Fry ( 1978 ). Given that the decarboxylase is induced by the addition of oxalate in the wood r o t fu n g u s, C. v e l u t i p e s , t h i s mode was a ls o t r i e d on the media of Emiliani and Bekes and that of Fry. The fungus was also planted on the media of Emiliani and Bekes modified so that oxalic acid was the sole carbon source. On the hypothesis that perhaps the oxalic acid was not being metabolized or transported into the fungus, induction was attempted with precursors to oxalate. There are two immediate precursors of oxalate in microorganisms. Hayaishi 63

et al. in 1956 demonstrated the hydrolytic cleavage of

oxaloacetate by a soluble enzyme preparation from

Aspergillis niger. The products of the reaction are oxalate

and acetate. The enzyme catalyzing this reaction,

oxaloacetate acetylhydrolase ( EC 3.7.1.1 ), required Mn for

maximum a c t i v i t y . G ly o x y late is a p re c u rs o r to o x a la te v ia

the enzyme isocitrate lyase, and glycolate is a known

precursor of glyoxylate ( Nord and Vittucci, 1947 ). These

three acids, oxaloacetate, glycolate, and glyoxylate were

each added to Fry's medium to a final concentration of 10

mmole/L. Also, because assay of the Bacto-peptone used in

Fry's medium lacks manganese ( Difco, 1953 ), this metal was

added to Fry's medium. The most satisfactory medium

( see results ) was found to be Fry's medium with added

manganese. The formula was 20 g Bacto-peptone ( Difco

Laboratories, Detroit, MI 48201 ), 40 g glucose ( Fisher

Scientific ), 0.0006 g MnCl2 + 4 H 2 O ( Fisher

Scientific ) and 1.0 L distilled-deionized water. With this medium, growth temperature was then optimized by the testing

the growth of the fungus at 30°C, 35°C, 40°C and

45°C.

Once enzyme was being produced by the fungus, the second step was to begin solubilization and purification studies. The influence of several methods of homogenization

( mortar and pestle, sonication, high speed blender ) was investigated with maximal solubilization achieved with the

VirTis "45" homogenizer ( VirTis Co., Gardiner, NY ) and alumina, 200-400 mesh, neutral grade I ( Mallincrodt ).

Solubilization with detergents ( Triton X-100, Zwittergent, taurodeoxycholate ) was investigated. Triton X-100

( Sigma ), 1 mL per 100 mL buffer offered a two- to twenty-fold increase in enzyme solubilized. Methods of concentrating the enzyme from the buffer with protein precipitation methods were also investigated ( methanol, e th a n o l, b u ta n o l, a c e to n e , ammonium s u l f a t e ) w ith m ethanol

( Mallincrodt ) being the solvent of choice.

Thirdly, various constituents were added to the citrate buffer used to try to stabilize both FDH and OD enzyme activities and keep the protein in solution.

Methods

Fungus growth and maintenance

Aspergillus phoenicis lyophilized spores obtained from

ATCC were handled per enclosed directions, and inoculated on

Czapek solution agar slants ( Difco, 1953 ). The slants were allowed to grow in the dark for two weeks. They were then placed in a plastic bag and frozen at -20°C until needed. The medium was made in batches and dispensed in 50 mL aliquots into 250 mL erlenmeyer flasks, which were then

plugged with cotton and autoclaved at 121°C for 15

minutes. The medium was cooled to room temperature before

inoculation. A spore suspension using a sterile 1:10,000

solution of sodium dodecyl sulfate ( Sigma ) was used to

inoculate the Fry's medium plus manganese. To make this

suspension, 10 mL of sterile solution was pipetted onto the

slant and mixed with a sterile pipet. Each inoculation was

0.5 mL of spore solution. The fungus was grown at 35°C

for five days before harvesting the mycelia in cheesecloth.

The fungus was washed with tap water for one hour, squeezed

gently, and frozen at -20°C until needed.

Enzyme solubilization and purification

Twenty grams of the frozen fungus were ground in a

mortar and pestle with an equal weight of alumina. The

buffer used was 0.1 M citrate buffer, pH 5.6, with Triton

X-100 added to a concentration of 1 mL per 100 mL buffer.

The paste was then blended in a VirTis high speed blender and stirred for two hours. The homogenate was next centrifuged for ten minutes at 20,000 x g and resuspended in

fresh buffer to be again stirred for two hours and centrifuged for ten minutes at 20,000 x g. The two supernaces were retained and pooled. The pooled supernate was then cut with an equal volume of cold methanol, centrifuged for ten minutes at 20,000 x g, and the

precipitate resuspended in buffer. The methanol cut was

repeated, the suspension centrifuged, and the pellet

retained. The pellet was dissolved in 0.02 M acetate

buffer, pH 5.6 and applied to a 2.5 x 50 cm column of

DEAE-Sephacel ( Pharmacia, Uppsala, Sweden ). A linear gradient was run from 0.02 M to 0.50 M acetate buffer, pH

5.6. The flow rate was 0.8 mL per minute. Fractions of

8 mL each were collected. Tubes with OD activity were pooled and concentrated by ultrafiltration. Fractions 59 through 66 had maximal activity. All procedures above were carried out at 4°C.

Oxalate Decarboxylase Assay

If fungus was to be assayed, 0.2 g was weighed after gently squeezing the thawed fungus. The fungus was then teased into small pieces to facilitate contact with the substrate. The fungus was added to 3 mL 0.1 M citrate buffer, pH 5.2, contained in a 25 mL erlenmeyer flask.

Should a liquid be assayed, the liquid was pipetted into a similar flask. The flask was placed in a 37°C water bath fitted with a reciprocal shaker. The flask was warmed in this bath for 15 minutes. Then 0.1 mL of a solution that contained 10 micromoles sodium oxalate and 0.25 uCi

[ U-*4C ]oxalate ( New England Nuclear ) was added to the flask. The flask was imraediatedly stoppered with a double

seal rubber stopper that had suspended from it a 1.0 mL

plastic well. In that well was a 1 x 3 cm piece of folded

filter paper and 0.2 mL of a 2:1:1 ( by volume ) mixture of phenethylamine, methanol and toluene. The flask was

returned to the water bath and incubated for 30 minutes, at which time the reaction was stopped with 1 mL of 12 percent

sulfuric acid being injected through the stopper with a 1 mL syringe and hypodermic needle. The acid also drove off the

14C02 from the reaction mixture. The 14CC> 2 was allowed to collect in the well for one hour after stopping the reaction. The well was then removed from the flask and placed in a mini-liquid scintillation vial. Four mL

Thrift-Solve was then added to the vial. The vials were cooled in a Packard Tri-Carb scintillation spectrometer

overnight to allow the 14CC> 2 to diffuse into the cocktail. Counting efficiency was taken into account. A unit of OD was that amount of enzyme that decarboxylated 1 umole of oxalate in one minute at 37°C.

P r o te in

Protein was assayed using a Cooomasie Blue dye binding method ( Bio-Rad Laboratories, Richmond, CA 94804 ). 68

Formate Dehydrogenase

Formate dehydrogenase ( FDH ) was isolated, purified

and assayed according to the method of Farinelli ( 1981 ).

The assay of FDH activity was done utilizing 0.07 mL of a

solution containing 5 umole per mL NAD ( Boehringer

Mannheim ), and following the production of NADH by its

absorbance at 340 nm using a Gilford Model 8540 recording

spectrophotometer ( Gilford Instrument, Oberlin, OH 44074 ).

The buffer used was 0.1 M citrate buffer, pH 5.5. Total

volume of the assay was 1 mL.

Results and Discussion

Table 2 is the results of media designed to optimize

the production of oxalate decarboxylase by A. phoenicis.

The table represents inoculating the spores and determining

OD activity in the fungus each day for ten days starting at day three after inoculation. The peak activity observed during the ten days is given for each medium.

The OD was not inducible by oxalate or any of its

immediate precursors. Also, when inoculated on a medium with oxalate as the sole carbon source, no growth ensued.

Using glucose, a more easily metabolizable carbon source than sucrose, did not improve OD yield. Four metals, which Table 2. Influence of Media on Oxalate Decarboxylase Production in Aspergillus phoenicis at 30°C.

Maximun U n its/1 0 0 m y celia

Media of Emiliani and Bekes ( 1964 ) LT 0.1

modification: + oxalate LT 0.1

+ 2 % methanol LT 0.1

glucose as carbon source LT 0.1

+ o x a la te LT 0.1

+ Zn, Cu, Fe, and Mn LT 0.1

+ Zn, Cu, Fe, Mn and o x a la te LT 0.1

oxalate as carbon source no growth

Media of Tomlinson ( 1951 ) LT 0.1

modification: + 2 % methanol LT 0.1

Media of Fry ( 1979 ) LT 0.1

modification: + oxalate LT 0.1

+ 2 % methanol LT 0.1

+ glycolate and glyoxylate LT 0.1

+ oxaloacetate LT 0.1

+ MnCl2 5 .0 -8 .0

LT = less than 70

included Mn, necessary for citric acid production, did not

increase enzyme synthesis. However, the spores used in this experiment were from an older culture, so the absence of OD may be because the fungus had a limited OD producing lifespan. When new spores were ordered, some, but not all of the media experiments were repeated. " Poisoning " the culture with methanol, which is known to increase citric acid production, did not induce OD activity.

The addition of manganese to the media of Fry was highly successful. The addition of the trace metal increased enzyme synthesis 50 to 80 fold. The essential nature of manganese for OD production was proven by fungus grown on Fry's media without manganese. When manganese was not added OD activity was absent, as shown in figure 2.

Although the mechanism of the induction was not established, it may be speculated that the added manganese was serving as a cofactor for the oxaloacetate acetylhydrolase. Thus, although exogenous oxalate may not traverse the cell wall, endogenous oxalate produced by the acetylhydrolase may induce the OD within the organism.

Growth temperature also had a marked effect on OD yield. Figure 3 presents data that showed maximal OD synthesis to be at growth temperatures between 35°C and

45°C. The figure represents one culture analyzed in 71

m m cmiemi he a im s mu the influence of. manganese

100 ■

80 -• RELATIVE bO -■ ACTIVITY AO

20 -• 0 L W ITH O U T Mn 3 4 5 6 7 8 9 10. DAYS AFTER INOCULATION

Figure 2. Influence of Manganese on Oxalate Decarboxylase Production in Aspergillus phoenicis. 100 *45* C * 40*0 ■435*0 -• 30*0

UNITS/ 100 g MYCELIA

2 5-

DAYS AFTER INOCULATION

Figure 3. Influence of Growth Temperature on Oxalate Decarboxylase Production in Aspergillus phoenicis. to 73

Table 3. Purification of Oxalate Decarboxylase.

T o ta l T o tal U nits/m g Fold % U n its mL P ro te in In c re a se Y ield

Fungus ( 200 g ) 54.9

1st Extract 20.4 250 0.0070 2nd Extract 4.3 160 0.0044 45.0

1st Methanol pptn 8.0 18 0 0.2235 31.9 14.6

2nd Methanol pptn 3.4 20 0.5120 73.1 6.2

DEAE-Sephacel Column after Concentration 1.1 5 1.310 187 2. 0 74

duplicate. The temperature chosen was 35°C, since this

is the lowest temperature yielding maximal activity.

Table 3 shows the purification of OD. The final

concentrate was not homogeneous on SDS disc electrophoresis,

showing at least five bands. The specific activity

obtained, 1.31 units per mg protein, was poor in comparison

to that of Emiliani and Riera ( 1968 ), who report a

specific activity of 80 units per mg protein. Their assay

conditions are different than used in the present study.

They use the Warburg apparatus to detect CO2 manometrical-

ly instead of the radiotracer OD assay as used here. In

addition, they use o-phenylenediamine in their OD assay to

activate the OD. Phenylenediamine was omitted in the

radiotracer assay since the activator interfered with the

spectrophotometric assay of oxalate. The absorbance at 340

nm increased when phenylenediamine was added to a cuvette

containing OD, FDH, NAD, and buffer, but no oxalate.

Therefore, the omission of phenylenediamine from the

radiotracer assay of OD was a more accurate estimation of

enzyme that could be utilized in the photometric enzyme

assay of oxalate. The OD isolated in this study was shown

to be a decarboxylase, since the OD assay taken to

completion had a recovery of 48 + 4 percent. Should the enzyme be an oxalate oxidase, both carbons in oxalate would

be converted to CO2 , which would have resulted in 100 100 T FDH 80 RELATIVE^ ACTIVITY

4-8 5.0 5.2 5-4 5-6 5-8 6-0 6-2 6-4 p H

F ig u re 4. Influence of pH on the Activity of Oxalate Decarboxylase and Formate Dehydrogenase. 76

percent recovery. With an OD, one of the carbons is

converted to CO2 and the other carbon is converted to

fo rm ate, pro d u cin g a maximum re c o v e ry o f 50 p e rc e n t.

The FDH was isolated according to the method of

Farinelli ( 1981 ). The optima of FDH and OD are shown in figure 4. This demonstrated the potential use of the enzymes in the coupled enzymatic kinetic assay for oxalate.

However, the coupling was not straightforward.

Initially the FDH did not have any activity at pH 5.5. One of the major influences on FDH activity is the order of reagent addition in its photometric assay. If the NAD was added tD the cuvette first and the reaction initiated by the addition of FDH, the reaction rate was more than double than if the reaction was initiated with NAD. This may be because the presence of the cofactor in its binding site somehow stabilizes the enzyme.

The addition of various known enzyme stabilizers was also explored. In the presence of sucrose at a concentration of 20 g per 100 mL buffer, FDH retained 8 5 percent of its activity at pH 5.5 as it had at pH 6.4. FDH i s a t i t s maximum a c t i v i t y a t pH 6 .4 . E th y len e g ly c o l, 20 percent by volume, is a slightly better stabilizer, since at pH 5.5 98 percent of its activity at pH 6.4 is retained. 77

Glycerol, when used at a 20 percent by volume concentration,

stabilized the enzyme best, with FDH having 123 percent of

its activity at pH 5.5 as at its optimum pH.

Knowing that each enzyme in the coupled series must be

in first order kinetics to give a linear reaction curve, and

knowing the second or indicating enzyme must be in excess to

accomplish this ( Bergmeyer, 1974 ), concentrating the FDH

beyond published values was attempted. Concentration of FDH

greater than 2.0 units per mL resulted in precipitation of

the enzyme. The presence of the stabilizing agent sucrose

did not inhibit the precipitation.

The analog of NAD, 3-acetyl-pyridine adenine

dinucleotide ( APAD ), was investigated for enhancement of

FDH activity. APAD has a more positive redox potential than

NAD, - 248 mv for APAD versus - 320 mv for NAD. Thus, the equilibrium of the FDH reaction should be farther to the right using this analog. Unfortunately, no rate enhancement was observed. The maximum velocity of the reaction with

APAD was 22 percent of the reaction when NAD was the cofactor. This may be because the analog did not bind to the enzyme as well as NAD, although this is speculation. During the course of the project, Boehringer Mannheim

marketed the OD from Aspegillus phoenicis in high specific

activity, about 20 units per mg protein, and in high

concentration, 40 units per mL. However, when this enzyme

preparation was used, even as little as 0.010 mL in a three

mL total volume, FDH precipitated. Several detergents at

concentrations of 0.1 percent and 1.0 percent were used.

Triton X-100, Zwittergent 3-08, Zwittergent 3-10,

Zwittergent 3-16, octyl-beta-D-glucopyranoside, sodium

taurocholate, and SDS all either failed to inhibit the

precipitation or if they inhibited the precipitation, the

activity of the enyme was completely inhibited. Dialysis of

the Boehringer Mannheim OD to 0.1 M citrate buffer, pH 5.5,

20 percent sucrose, also did not prevent precipitation of

th e FDH.

Precipitation could also be instigated by concentrations of NAD greater than 5 mM.

To determine if there was an impurity in the prepared

OD, the possibility of an enzyme that acted on formate was investigated. In the initial extraction of the homogenized fungus, it was found that there was an enzyme that used formate as its substrate. Using an oximeter, when 0.1 mL of

1.0 M formate was added to the two mL cuvette containing the first extraction of the fungus and 0.1 M citrate buffer, pH 79

5.2, it was found that while 39 units of OD was extracted

from the fungus, 1257 units of a formate oxidase-type activity was simultaneously extracted. However, when

formate was added to the Boehringer Mannheim OD, no oxygen was consumed. To the author's knowledge, there is no formate oxidase reported in the literature, although purified FDH

from Psuedomonas oxalaticus is known to have some formate oxidase activity ( Hopner and Knoppe, 1974 ).

As shown in figure 5, conditions were found in which the two enzymes were coupled. The cuvette contained 0.212 units of OD ( 0.20 mL ), 0.1 units of FDH ( 0.20 mL ), 0.35 micromole of NAD ( 0.07 mL ), and 0.28 mL of 0.1 M citrate buffer with 20 g sucrose per 100 mL buffer added. The sample was 0.25 mL of an aqueous sodium oxalate standard, 90 micrograms per mL. This standard is about 20 times greater than the lower limit of oxalate seen in a normal urine. Yet the change in absorbance was only 0.4 in an 80 minute period. At this pH, there was no precipitation of FDH, and the assay was shown to be feasible, but again, sensitivity was inadequate.

To further delineate the problem, the FDH activity was investigated at pH 5.5. This would allow the FDH to still retain all its activity and allow the pH to be closer to the optimum of OD. To evaluate FDH activity at pH 5.5, a sodium 80

0.4 ABSORBANCE 0.3 AT 3A0 nm 0.2

20 40 6 0 TIME (MINUTES)

Figure 5. Reaction Progession of the Coupled Reactions. Conditions: 90 ug/ml sodium oxalate ( 0.25 mL ); OD 0.212 units ( 0.20 mL ); FDH 0.10 units ( 0.20 mL ); NAD 0.35 umol ( 0.07 mL ); Citrate Buffer, 0.1 M, pH 5.9, with 20 g Sucrose per 100 mL buffer added (0.28 mL ). 81 formate standard was made that was the molar equivalent of

10 micrograms of oxalate per mL. This is approximately the low normal concentration of oxalate in urine. At this concentration of formate, there was no observable change in absorbance at 340 nm despite the use of almost one unit of

FDH in the cuvette and the enzyme stabilizer glycerol.

Thus, FDH was inadequate for the analysis of oxalate at concentrations seen in urine for several reasons. The formate standard in the above experiment was added as a bolus to evaluate the FDH. This does not reflect the situation in a coupled assay, since in coupled reactions the intermediates build up over time. The coupled reaction would be less favorable than the addition of formate as a bolus. The enzyme could not be more concentrated, since this resulted in precipitation. The enzyme could not be made more active with an NAD analog, and the FDH had all of the activity it possessed at its pH optimum. Therefore, although the enzymes might have been usable to analyze oxalate in aqueous systems, the OD and FDH could not be coupled for the assay of oxalate in urine.

Conclusions

1. Manganese is a requirement for the synthesis of OD i-n Aspergillus phoenicis. 82

2. The grow th te m p e ra tu re y ie ld maximum enzyme

synthesis was 35°C.

3. The fungus could be maintained in the freezer on

Czapek solution agar slants until used, thus eliminating the

need for subculturing more than once.

4. OD from A. phoenicis and PDH from soybeans could be

coupled to produce an increase in the concentration of NADH

when an aqueous standard of 90 micrograms oxalate per mL is

u t i l i z e d .

5. FDH from soybeans has several usable stabilizers

that preserve activity at pH 5.5, namely, sucrose, ethylene glycol, and glycerol. Glycerol was shown to stabilize the enzyme best at this pH. Adding the NAD to the cuvette before the FDH also preserved activity at pH 5.5. The FDH

from soybeans has a lower pH optimum than FDH from any other s o u rc e .

6. The FDH from soybeans could be precipitated under several conditions, including 1) when the enzyme was concentrated greater than 2 units per mL, 2) when OD made by

Boehringer Mannheim was added, and 3) when concentrations of

NAD greater than 5 mM were added.

7. Activity of FDH could not be enhanced by APAD, a cofactor analog that is more easily reduced.

8. The FDH from soybeans at maximal concentrations was unable to produce a change in absorbance at 340 nm when concentrations of formate used were reflective of the amount of formate that would be produced from urinary oxalate by

the action of OD. Thus the enzyme was not usable for the analysis of oxalate in urine.

9. A formate oxidase-type activity has been observed

in extracts of A. phoenicis in large quantity. An isolated

formate oxidase has not been reported. This enzyme requires

further investigation. CHAPTER I I I

An HPLC Method for Urinary Oxalate

Literature Review: Oxalate by HPLC

The lack of importance of high pressure liquid chromatoraphy ( HPLC ) in urine oxalate analysis is reflected in the absence of this method in two of the major reviews on urine oxalate methods ( Robertson and Rutherford,

1980; Hodgkinson, 1970 ) and the minimal attention given in a third review ( Hodgkinson, 1977 ). However, given the proper type of detector, given a column allowing fast chromatography, and given the non-destructive nature of

HPLC, the technique holds much promise for the future.

The use of HPLC on oxalate has been primarily analytical rather than biochemical. That is, separation of oxalate in urine by HPLC is rare. One of the earliest chromatographic separations of oxalate is reported by

Isherwood ( 1946 ). He separates oxalate from organic acids using water as the stationary phase on a silica gel column.

The column eluate is mixed with a dilute solution of a pH indicator to detect the acids. Bulen et al. ( 1952 ) load

0.5 N sulfuric acid on the silica gel and elute with a butyl alcohol/chloroform system. Detection is by collecting 25 mL

84 fractions and titrating each fraction with 0.01 N NaOH.

Detection lim its are 2 to 10 microequivalents and column runs are nine hours long. The sulfuric acid stationary phase is used by others, but with different mobile phases.

Kinnory et a l. ( 1955 ) use a benzene/ethyl ether solvent mixture. Scott ( 19 55 ) employs a gradient of

4-methyl-2-pentanone in methylene chloride, while Resnick

( 19 55 ) uses tert. butanol/chloroform. Kesner and

Muntwylker ( 1966 ) use a tert. amyl alcohol gradient in chloroform. Their detection utilizes o-nitrophenol as both indicator and titrator. The column eluent is mixed with this indicator and free acid changes absorbance at 350 nm.

The eluent is continuously monitored, and has the advantage of shortening column runs to five hours and lowering the sensitivity to 0.2 to 0.5 microequivalents.

Celite columns are also used to hold the sulfuric acid stationary phase. Titrating the isotopically labelled peaks with 0.1 N NaOH, micromolar concentrations are recovered with 90 to 95 percent efficiency. Also, in order to separate oxalate it is necessary to rechromatograph the oxalate peak using ether instead of chloroform/butanol as in the first run ( Phares et a l., 1952 ). 86

Organic acids are also resolved on reverse phase

chromatography. Hyakutake and Hanai ( 1975 ) use a

octyldecylsilane bonded phase packing with detection

utilizing absorbance at 220 nm. Using a hexane/tetrahydro-

furan/tert. butanol solvent system, sulfuric acid is added

to supress ionization of the weak acid. Thus, retention is

based on the hydrophobic interaction between acid and bonded

phase, rather than on the ionic interaction. Detection

limits were in the 2 to 5 microgram range.

A modification of this technique is called ion-pair chromatography. Here, the solvent system has a positively charged counterion, for example, tetrabutyl ammonium ion.

The ionic end of the counterion binds with the ionic end of the acid. The hydrophobic end of the counterion allows the

ion-pair to partition into the reverse phase bonded to the column packing. Mayer et a l. ( 1979 ) use a strong cation exchange packing in their ion-pair chromatography of oxalate for reasons not obvious to the author. It may be that the reverse phase bleeds or otherwise compromises their detection system, although this is not discussed in their paper. The detection system is electrochemical, permitting a detection of 0.1 mg per L. . The linearity of the detector is 1 to 1000 mg per L. The detector is somewhat selective in that nothing with an oxidation potential greater than the

1.25 volts used is detected. Yet, there is major interference from uric acid. Therfore, a calcium precipitation of oxalate from urine is required. Recovery is 98 percent with a standard deviation of 2.5 percent.

This is one of the few HPLC methods applied to oxalate analysis in urine. Ion-pair chromatography is applied to oxalate in plant tissues utilizing UV absorbance at 220 nm for detection. Though the sensitivity is 1 mg per L, the technique may be difficult to use on urine due to the complicated nature of the chromatograms. Oxalate is major constituent of rhubarb, the sample analyzed, which is not the case in urine ( Libert, 1981 ).

Anion exchange is extensively used in chromatography of oxalic acid. Kunin and Meyers ( 1947 ) use Amberlite IR-4B and titration of the eluent for detection. Sargent and

Rieman ( 1958 ) use Dowex 1 x 8 and identify peaks by measuring the absorbance of the chromate ion that is added to each fraction collected. They use 0.1 N HC1 to elute the acids. Detection of acids using refractive index coupled to a similar column, Aminex A-14, is also reported ( Kaiser,

1973 ).

Metals can also be used to displace the acids in this form of ion exchange. Lee and Samuelson ( 1967 ) find that magnesium ion is the most convenient cation to use in anion exchange. Complexes formed when copper is used are adequate as long as the acids eluted are not easily oxidized.

Therefore, when uronic acid is to be separated, alternate

metals are employed. Zinc is usable in this case, but then

oxalate cannot be analyzed because of the low solubility of

zinc oxalate. Magnesium acetate is most convenient since

the chromic acid oxidation is used with a Technicon

Autoanalyzer following the absorbance of the Cr III

com plexes.

Anion exchange HPLC is coupled to electrochemical

detection in an oxalate method for urine reported by Asper

and Schmucki ( 1979 ). A two buffer elution is required,

taking about 20 minutes to complete. No pretreatment of

urine is required, and sensitivity is 1 mg per L. Recovery

and specificity are untested.

Chromatography of organic acids utilizing a cation exchange resin, Dowex 50 x 8, is first reported by Wheaton and Bauman in 1953. They call it ion exclusion and view it as an industrial scale operation in chemical manufacturing processes. Cost would be minimized using a chromatographic process where water is the mobile phase. The principle is as follows:

The resin bed may be divided into three entities, the solid carbonaceous skeleton of the polystyrene bead, the 89 volume of the liquid within the bead, and the volume of liquid outside the bead between the resin particles.

Although water is the only liquid, the character of the volume within the bead is different from the liquid outside the bead since the resin is highly charged from the 303- groups covalently attached to the polystyrene resin. This charged resin serves as a barrier to highly ionized acids, excluding acids from the interior of the bead. Thus a highly charged acid should emerge from the column in the void volume of the column ( Wheaton and Bauman, 1953a;

Wheaton and Bauman, 1953b ). The pKa of the acid being resolved is the crucial, although not the only determinant of retention. As the mobile phase is made more acidic, the ionization of acids is suppresed, and the acids are retained on the column longer. However, the correlation between retention time and acid strength ( pKa ) is, although strong, not absolutely linear. Also, there is a strong correlation between the number of carbons in a homologous series of acids and retention time. There is, therefore, a reverse phase component to the retention of acids in ion exclusion chromatography ( Jupille et al. , 1981 ). It should be pointed out that retention also varies with ionic form, type and crosslinkage of a resin ( Bauman et al. ,

1956 ). Ion exclusion was first applied to the analysis of

organic acids by Harlow and Morman ( 1964 ). They describe an automatic titrator for detection. The chromatogram is

actually the amount of of 0.1 N NaOH automatically delivered by motor driven syringe to maintain the column eluent at pH

8.5. Fifty nine organic acids, including oxalate, were used

to characterize this column. A column run is about an hour, depending on the sample. Concentrations used are 0.05 to 1 milliequivalent per sample. Sample size is 0.025 to 0.10 mL. Tanaka et a l. ( 1979 ) have coupled coulometric

detection to ion exclusion. The reaction between FeCl3 and organic acids is followed at 440 nm in an automated post-column reaction reported by Linder and Messori in 1979.

Ion exclusion has been coupled to ion chromatography for de tection ( Rich et al. , 1980 ). Here the eluent is concentrated onto a second anion exchange column and acids are eluted off this " detection column " with a different mobile phase. Detection off the detection column is conductimetric. This method has been applied to acids in plasma and to urine vanillylmandelic acid, but not to urine o x a la te .

The use of dilute acid as mobile phase permits the separation of strong acids that cannot be separated by water alone. Dilute acid mobile phase in ion exclusion was first reported by Patel et al. ( 1967 ). Their three hour 91

45 minute separation can be shortened to 10 to 20 minutes using higher flow rates and decreasing the acid concentra tion in the eluent from 0.1 N to 0.001 N HC1 ( Richards,

1975 ). Turkelson and Richards ( 1978 ) apply the mobile phase of 0.01 N to 0.001 N HC1 to Aminex 50W x 4 resin and get separation of citric acid cycle acids. Detection is based on UV absorbance at 210 nm.

In August, 1979 Bio-Rad Laboratories introduced an organic acids analysis column. They modified the column used by Turkelson and Richards by using an 8 percent crosslinked, 9 micrometer diameter bead instead of the 4 percent crosslinked, 35 micrometer diameter bead Turkelson and Richards used. The smaller diameter, more rigid resin bead increases resolution and reduces analysis time. It was this Bio-Rad column that was investigated for use in the analysis of oxalate in urine.

H y p o th esis

Separation of several organic acids, including oxalate, glyoxylate, glycolate, glycerate, and citrate was demonstrated to be separated by the column marketed by

Bio-Rad ( 1979 ). This column was investigated for use in a study of the hypothesized metabolism of citrate to oxalate via glyoxylate. Therefore, the separation of the 92 radiolabelled acids viewed important in this study, namely citrate, glyoxylate, oxalate and glycolate was studied.

Also, the possibility of using this column in a routine analysis of urine was evaluated.

Methods

Experimental

The HPLC used was made by Altex ( Berkeley, CA

94710 ), a model 420 microprocessor-controller with model

110A pumps. Detection was at 210 nm using an Altex/Hitachi model 155-10 continuously variable UV monitor. An aluminum block column heater maintained temperature at 70°C ( Jones

Chromatography, Columbus, OH 43214 ). The analysis column used was an Organic Acid Analysis column ( Bio-Rad ), a 1 x

30 cm column packed with the cation exchange resin, HPX-87.

To protect the analysis column, a small 1 x 5 cm guard column was used, which was thumb packed with Aminex Q-15S, a resin similar to the HPX-87, but a larger diameter bead.

The column was first characterized by separation of aqueous standards ( Sigma ). Several concentrations in the mobile phase between 0.01 M and 0.001 M sulfuric acid were used. Radiolabelled acids were obtained from New England

N u c le a r. Samples had to be pretreated to remove all positively

charged substances. This is necessary because resolution is

dependent on the resin having freely ionizable sulfonic acid

groups. Should any cation bind irreversibly to the resin in

large quantities, these sulfonic acid groups would be masked

and resolution would deteriorate. For similar r e a s o n s ,

concentrated salts and metals must also be eliminated by the

pretreatment. Second, if possible, the sample preparation should result in concentration, since the quantity of

radiolabelled oxalate in the urine would not be large.

Third, the preparation must have a high recovery. In light of the well known lability of glyoxylate ( Johansson and

Tabova, 1974 ), this problem would be the most difficult to s o lv e .

Since glyoxylate is the most labile of the acids to be analyzed, [ 14c ]glyoxylate ( New England Nuclear ) was used to judge recovery in sample technique.

Chromatography

The mobile phase was 0.01 M sulfuric acid made from concentrated sulfuric acid, reagent grade ( Fisher

Scientific ) and distilled-deionized water. The concentrated acid was weighed ( 0.981 g per L ) rather than 94

the pH adjusted since it was felt that weighing was a more

precise measurement than the pH meter. The dilute sulfuric

acid was then filtered using a 0.4 micrometer pore cellulose

filter apparatus ( Millipore, Corp., Bedford, MA 01730 ).

This functioned not only to filter but also to degas the

mobile phase.

Flow rate of the mobile phase was 0.6 mL per minute.

Pressure was 200 to 900 psi. The detector wavelength was

210 nm with the detector set to give 0.2 absorbance units

full scale on the recorder. Chart speed was 1.0 cm per

minute. Sample size was 0.020 mL using a sample loop which

was overfilled before each injection to insure precision

sampling. When fraction collections were done, collection

was directly into liquid scintilation mini-vials ( Kew

Scientific ). Counting cocktail was Thrift-Solve ( Kew

Scietific ).

Solvent extraction

The extraction procedure using either ether, ethyl

a c e t a t e , o r ch lo ro fo rm was tak en from Hammond and Goodman

( 1970 ). A 2.0 mL aliquot of sample ( standard or urine ) was pipeted into a 15 mL glass stoppered centrifuge tube.

The pH was adjusted to less than 1.0 by the dropwise addition of 2 M HC1 saturated with NaCl. The sample was

extracted with 5 mL of solvent by shaking for 10 minutes.

The extraction was repeated for a total of three

extractions, each time pooling the organic layer. The

pooled solvent was then dried with annhydrous sodium

sulfate, about 1 g, and filtered through glass wool. The

filtered solvent was evaporated at 60°C under nitrogen.

The residue was dissolved in 0.2 mL of water, and applied to

th e HPLC.

The extraction procedure using tri-n-butyl phosphate

( TBP ) was taken from Zarembski and Hodgkinson ( 1965a ).

However, the back extraction utilizing 2 M NaOH had to be

eliminated due to precautions against applying strong base

to the column. The TBP was not applied directly due to

precautions against applying solvents other than

acetonitrile to the column.

Ion Exchange

The ion exchange treatment is modified from personal

communication with Bio-Rad Laboratories. Two disposable mini-columns ( Bio-Rad Laboratories ) were used in tandem.

In the upper column was 1.0 g, dry weight, of AG 50W x 8,

100-200 mesh cation exchange resin in the H+ form

( Bio-Rad Laboratories ). The column was slurry packed with 96

5.0 mL water and blown dry with nitrogen. In the lower

column was 1.0 g dry weight of AG 1 x 8, 100-200 mesh anion

exchange resin in the OH” form ( Bio-Rad Laboratories ).

The column was slurry packed with 5.0 mL water. This column

was then equilibrated with 10 mL 0.1 M phosphate buffer, pH

6.0 and blown dry with nitrogen. With the two columns

connected, 5 mL urine was run through the columns and washed with 10 mL water. The lower column was removed and blown dry. The application of 2 M sulfuric acid containing 25

percent acetonitrile ( MCB Reagents, Gibbstown, NJ 08027 ) eluted oxalate and other acids fron the anion exchange mini-column.

Specificity of the HPLC

A 3.0 mL aliquot of urine was spiked with a known quantity of l 4C-oxalic acid, and pipeted into a 25 mL erlenmeyer flask. The pH was adjusted to 3 with 3.0 mL

1.0 M citrate buffer, pH 3.0. OD, 0.4 units ( Sigma ) was then added and the flask stoppered with a rubber stopper-plastic well assembly. In the plastic well was 0.2 ml of a 2:1:1 mixture of phenethylamine/toluene/methanol, a

CC> 2 absorbing solution. The flask was incubated overnight and the extent of the decarboxylation judged by

the recovery of the ^4C0 2 in the plastic well. 97

Results and Discussion

The characterization of the column is presented in

Table 4. The capacity factor ( k' ) given in the parentheses is a second, though smaller, peak observed during chromatography of the aqueous standards. When aqueous standards of the radiolabelled acids of interest were chromatographed, there was baseline resolution of these acids, not only by UV absorbance, but, as shown in figure 6, baseline resolution of the radioactive peaks.

Sample preparation proved to be most difficult due to the lability of glyoxylate. Between ether, chloroform and ethyl acetate, ethyl acetate had the best recovery ( 1.0 percent, 10 percent and 15 percent, respectively ). TBP extracted 88 percent of the glyoxylate in a single extraction, but the glyoxylate could not be back extracted into the aqueous phase with less than 2 M NaOH, which could not be applied to the column. The TBP extraction was therefore eliminated.

The use of the ethyl acetate extraction was also eliminated. When a sample of urine was spiked with radiolabelled glyoxylate, the radioactive peak was not observed upon chromatography, although the extraction procedure had been shown to recover 15 percent of the 98

Table 4. Retention Times of Organic Acids of the TCA Cycle and Oxalate Metabolism Conditions: Column, Bio-Rad HPX-87 Organic Acid Analysis Column; Mobile phase, 0.01 M Sulfuric Acid; Flow Rate, 0.6 mL/min; Pressure 200-900 psi; Temperature 70°C; Detector Wavelength, 210 nm.

S u b stan ce k_|_

o x a la te 0.145 cis-aconitate 0.339 ( 0.823 )* i s o c i t r a t e 0.435 ( 0.836 ) c i t r a t e 0.437 alpha-ketoglutarate 0.532 g ly o x y la te 0. 710 m alate 0.710 ( 1.435 ) oxaloacetate 0.758 ( 0.403 ) s u c c in a te 1.065 g ly c o la te 1.161 form ate 1.462 fum arate 1.468

* Capacity factors ( k* ) in parentheses are minor peaks present when the standards were used 25,000-1 47940 dpm

20,000 -

15,000 dpm

10,000

9 ,0 0 0 -

TIME (MINUTES) Figure 6. Separation of Radiolabelled Acids Conditions: Column, Bio-Rad HPX-87 Organic Acid Analysis Column; Mobile phase, 0.01 M Sulfuric Acid; Flow Rate, 0. mL/min; Pressure 200-900 psi; Temperature 70°C; Detector Wavelength, 210 nm; Fractions, 0.3 min each. Acid identified: l«=oxalate; 2=citrate; 3»glyoxylate; 4=glycolate; 5«formate. 100

radiolabel using the radioactivity to gauge the recovery.

That is, the recovery experiment did not have the

chromatography step as a part of it. Apparently the

combination of the heat and acid of the HPLC in combination

with the acidic conditions of the extraction decarboxylated

the glyoxylate. Since resolution on the column depended to

some degree on the column being heated, and acid had to be

in the mobile phase, the use of this column was excluded for

evaluating the metabolism of citrate to oxalate via

glyoxylate in rats.

The ion exchange method was not utilizable for the

analysis of glyoxylate, since the strong acid employed

decarboxylated the glyoxylate. The ion exchange

pretreatment was evaluated for the columns use as a routine

method for oxalate using the technique of standard

additions. Using this pretreatment, a peak eluted at the

same time as oxalate when a blank was taken through the ion

exchange preparation, as shown in figure 7. This peak was

consistant in height from run to run, and may have been a result of the sulfuric acid/acetonitrile interacting with

the anion exchange resin.

This method proved to be specific. A urine that had all the oxalate specifically removed by the enzyme oxalate decarboxylase was chromatographed. The peak observed was 101

210

0-2

01-

2 4 6 8 TIME (MINUTES)

Figure 7. Chromatogram of Blank. Conditions: Column, Bio-Rad HPX-87 Organic Acid Analysis Column; Mobile phase, 0.01 M Sulfuric Acid; Flow Rate, 0.6 mL/min; Pressure 200-900 psi; Temperature 70°C; Detector Wavelength, 210 nm. 102

the same height as the peak seen in the blank preparation,

as shown in figure 8. The urine before enzyme treatment is

shown in figure 9.

Using the standard addition method with two additions,

.020 mg and .040 mg, or three samples total, the

chromatography time alone, without sample preparation was

240 minutes. The correlation between the samples was good, with r = 0.992, but the time used in analysis was prohibitive. The 80 minute chromatography time for each

urine sample to have all peaks off the column could not be s h o rte n e d , s in c e th e maximum recommended te m p e ra tu re was being used, and flow rates above 0.6 ml per minute raised the pressure on the column above acceptable limits.

Therefore, though sensitivity was adequate, and specificity was proven, the HPLC technique here described was too time consuming for routine use. 0 5

0-4

0-3

02-

TIME (MINUTES) Figure 8. OD Treated Urine. Conditions as in Figure 7. 0-5n

0-4-

A 210

2 4 6 8 10 TIME (MINUTES)

Figure 9. Chromatogram of Urine. Conditions as in Figure 7 105

Conclusions

1. The ion exclusion column for organic acid analysis marketed by Bio-Rad produced baseline resolution of five acids of interest in the evaluation of metabolism of oxalate in rats. Those acids were oxalate, glycolate, glyoxyate, citrate and formate. This resolution was seen only in aqueous standards.

2. Glyoxylate was shown to be so labile as to preclude its analysis by this method.

3. The use of ion exchange pretreatment of urine was adequate for oxalate in urine. The HPLC method was shown to be specific and sensitive enough for normal urine oxalate.

The standard addition used had good linearity ( r = 0.992 ), but the time needed for analysis was prohibitive, 240 minutes without pretreatment taken into account. Thus it was shown to be too cumbersome for routine use. List of References

Aiken, P., Dosch, W., Joost, J. and Rosenmeier, A., in: Urinary Calculus, J. G. Brockis, B. Finlayson, eds., PSG Publishing, Littleton, MA, p. 387, (1981).

Archer, H. E., Dormer, a . E., Scowen, E. F. and Watts, R. W. E., Clin. Sci., 16, 405, (1957).

Asper, R. and Schmucki, O., in: Oxalate in Human Biochemistry and Clinical Pathology, G. A. Rose, W. G. Robertson and R. W. E. Watts, edsT, Proceedings, London, p. 76, (1979).

A tk in s, G. L . , Dean, B. M., Griffin, W. J., Scowen, E. F. and Watts, R. W. E., J. Biol. Chem., 239, 2975, (1964).

Baadenhuijsen, H. and Jansen, A. P., Clin. Chim. Acta, 62, 315, (1975).

Baker, E. M., Saari, J. C. and Tolbert, B. M., Amer. J. Clin. Nutr., 19, 371, (1966).

Bais, R., Potezny, N., Edwards, J. B., Rofe, A. M. and Conyers, R. A. J., Anal. Chem., 52, 508, (1980).

Baruch, S. B., Burich, R. L., Eun, C. K. and King, V. F., Med. Clin. North Amer., 59, 569, (1975).

Bauman, W. C ., Wheaton, R. M. and Simpson, D. W., in : Ion Exchange Technology, F. C. Nachod, J. Schubert, eds., Academic Press, N. Y., p. 182, (1956).

Baxter, G. P. and Zanetti, J. E., Amer. Chem. J ., 33, 500, (1905).

Bengtsson, L. and Samuelson, 0., Anal. Chim. Acta, 37, 359, (1967).

B e n n e tt, D. J ., Cole, F. E., Frohlich, E. D. and Erwin, D. T., J. Lab. Clin. Med., 91, 822, (1978).

Bergmeyer, H. U., in: Methods of Enzymatic Analysis, Vol. 1, 2n<3 English ed., H. U. Bergmeyer, ed., Academic Press, N.Y., p. 109, 123, (1974).

106 107

Berman, L. B., Schreiner, G. E. and Feys, J., Ann. Int. Med., 4b, 611, (1957).

Beutter, H.-O., Becker, J ., Michal, G. and Walter, E., Fresenius Z. Anal. Chem., 301, 186, (1980).

Bio Rad Laboratories, The Liquid Chromatographer, August, 1979.

Boyce, W. H., Garvey, F. K. and Strawcutter, H. E., J. Amer. Med. A sso c ., 161, 1437, (1956).

Brown, V. K. H. and Box, V. L., Brit. J. Dermot. , 80_, 740, (1968).

Bulen, W. A., Varner, J. E. and Burrell, R. C., Anal. Chem., 24, 187, (1952).

Butz, M. and Kohlbecker, G., Urol. Int., 3j>, 303, (1980).

Chadwick, V. S., Modha, K. and Dowling, R. H., New Engl. J. Med. , 2_89, 172, (1973).

Chalmers, R. A., in: Oxalate in Human Biochemistry and Clinical Pathology, G. A. Rose, W. G. Robertson and R. W. E. Watts, eds., Proceedings, London, p. 61, (1979).

Chalmers, R. A. and Watts, R. W. E., Analyst, ST7, 958, (1972).

Charransol, G. and Desgrez, P., J. Chromatog., £8, 530, (1970).

Colburn, S. P. and Packett, L. V., J. Nutr., 7£, 385, (1962).

Cook, D. A. and Henderson, L. M., Biochim. Biophys. Acta, 184, 404, (1969).

Cooley, J. M. and Kratochvil, B., Can. J. Chem., 56, 2452, (1978).

Costello, J., Hatch, M. and Bourke, E., J. Lab. Clin. Med., 87, 903, (1976).

Crawford, J. E., Crematz, E. P. and Alexander, A. E., Austral. J. Chem., 2JL, 1067, (1968).

Crawhall, J. C., Scowen, E. F. and Watts, R. W. E., Lancet, 2, 806, (1959). 108

Dakin, H. D., J. Biol. Chem., 3, 57, (1907).

Dekker, E. E. and Maitra, U., J. Biol. Chem., 237,2218, (1962).

Dempsey, E. F., Forbes, A. P., Melick, R. A. and Henneman, P. H., Metabolism, 9, 52, (1960).

Difco Laboratories, Difco Manual, Ninth Edition, Detroit, MI, p. 245, 265, (1953).

Dosch, W., in: Urolithiasis: Clinical and Basic Research, L. H. Smith, W. G. Robertson and B. Finlayson, eds., Plenum Press, N.Y., p. 977, (1981).

Drewes, P. A., in: Clinical Chemistry: Principles and Techniques, 2nd edition, R. J. Henry, D. C. Cannon and J. W. Winkelman, eds., Harper and Row, Hagerstown, MD, p. 1349, (1974).

Duggan, D. E., Walker, R. W., Noll, R. M. and VandenHeuvel, W. J. A., Anal. Biochem., 94, 477, (1979).

Dutt, V. V. S. E. and Mottola, H. A., Biochem. Med., S>, 148, (1974).

Earnest, D. L., Johnson, G., Williams, H. E. and Admirand, W. H., Gastroenterology, 66, 1114, (1974).

Elder, T. D. and Wyngarden, J. B., J. Clin. Invest., 39, 1337, (1960).

Elliott, J. S. and Eusebio, E., Invest. Urol., 3, 72, (1965).

Emiliani, E. and Bekes, P., Arch. Bioch. Biophys., 105, 488, (1964).

Emiliani, E. and Riera, B., Biochim. Biophys. Acta, 167, 414, (1968).

Faber, S. R., Feitler, W. W., Bleiler, R. E., Ohlson, M. A. and Hodges, R. E., Amer. J. Clin. Nutr., 12, 406, (1963).

Farinelli, M. P., Dissertation, The Ohio State University, (1981).

Fleisch, H. and Biroz, S., Experientia, 20, 276, (1964).

Fraser, J. and Campbell, D. J., Clin. Biochem., J5, 99, (1972). 109

Fredriksson, S ., J. Chromatog., 188, 266, (1980).

Fry, D. W. and Richardson, K. E., Biochim. Biophys. Acta, 567, 482, (1979a).

Fry, D. W. and Richardson, K. E., Biochim. Biophys. Acta, 568, 135, (1979b).

Gambardella, R. L. and Richardson, K. E., Biochim. Biophys. Acta, 499, 156, (1977).

Gambardella, R. L. and Richardson, K. E., Biochim. Biophys. Acta, 544, 315, (1978).

Gibbs, D. A., Clin. Sci., j41, 3P, (1971).

Gibbs, D. A., Hauschildt, S. and Watts, R. W. E., J. Biochem., 82, 221, (1977).

Gibbs, D. A. and Watts, R. W. E., Clin. Sci., 3JL, 285, (1966).

Gibbs, D. A. and Watts, R. W. E., J. Lab. Clin. Med., 73, 901, (1969).

Gibbs, D. A. and Watts, R. W. E., Clin. Sci., 44, 227, (1973).

Giterson, A. L., Sloof, P. A. M. and Schouten, H., Clin. Chim. Acta, J29, 342, (1970).

Glenister, P. R. and Alavi, Z. I., Amer. Soc. Brew. Chem.proceedings, p. 175, (1969).

Goodman, D. B. P., Davis, W. L. and Jones, R. G., Proc. Natl. Acad. Sci. U. S. A., 77, 1521, (1980).

Gratzlova, J. and Revusova, V., Lek. Obz., 2£, 661, (1971).

Hagler, L. and Herman, R. H., Amer. J. Clin. Nutr., 26, 758, (1973).

HalIson, P. C. and Rose, A. G., Clin. Chim. Acta, 55, 29, (1974).

Hammond, K. B. and Goodman, S. T . , C lin . Chem. 16, 212, (1970).

Harlow, G. A. and Morman, D. H., Anal. Chem., 36, 2438, (1964). 110

Harris, K. S. and Richardson, D. E., Invest. Urol., ]J3, 106, (1980).

Hayaishi, O., Shimazono, H., Katagiri, M. and Saito, Y., J. Amer. Chem. S o c ., TB.r 5126, (1956).

Herring, I. C. , J. Urol., 8i8, 545, (1962).

Hockaday, T. D. R., Frederick, E. W. and Clayton, J. E., J. Lab. Clin. Med., 65, 677, (1965).

Hodgkinson, A., Clin. Chem., 1(5, 547, (1970).

Hodgkinson, A., Oxalic Acid in Biology and Man. Academic Press, NY, 325 pp., (1977).

Hodgkinson, A., J. Nutr., 108, 1155, (1978).

Hodgkinson, A., Clin. Chim. Acta, 109, 239, (1981).

Hodgkinson, A. and Wilkinson, R. , Clin. Sci., 4J5, 61, (1974).

Hodgkinson, A. and Williams, A., Clin. Chim. Acta, 36^t 127, (1972).

Hodgkinson, A. and Zarembski, P. M., Analyst, 86, 16, (1961).

Hopner, T. and Knoppe, J ., in: Methods of Enzymatic Analysis, Vol. 1, 2n<^ English ed., H. U. Bergmeyer, ed., Academic Press, N.Y., p. 1551, (1974).

Husdan, H., Leung, M., Oreopoulos, D. and Rapoport, A., C lin . Chem. 22, 1538, (1976).

Hyakutake, H. and Hanai, T ., J. Chromatog., 108, 385, (1975).

Isherwood, F. A., Biochem. J ., £0, 688, (1946).

Jakoby, W. B., in: Methods of Enzymatic Analysis, Vol. 1, 2 nd English ed., H. U. Bergmeyer, ed., Academic Press, N.Y., p. 1542, (1974).

Johansson, S. and Tabova, R., Biochem. Med., ^1, 1, (1974).

Jones, C. T ., Biochem. Biophys. Res. Commun., 9j>, 849, (1980). I l l

Jupille, T. , Gray, M., Black, B. and Gould, M., Amer. Lab., 13, 80, (1981).

Kamiya, S ., Noye, Y. and S a to , H ., Ja p . J . Med. S c i . , 2[, 317, (1937).

Kesner, L. and Muntwyler, E., Anal. Chem., 2®/ 1164, (1966).

Kinnory, D. S., Takeda, Y. and Greenberg, D. M., J. Biol. Chem., 212, 379, (1955).

Knowles, C. F. and Hodgkinson, A., Analyst, JT7, 474, (1972).

Kobos, R. K. and Ramsey, T. A., Anal. Chim. Acta, 121, 111, (1980).

Koch, J., Stokstad, E. L. R., Williams, H. E. and Smith, L. H., Jr., Proc. Natl. Acad. Sci. U. S. A., 57, 1123, (1967).

Koch, G. H. and Strong, F. M., Anal. Biochem., 27, 162, (1969).

Kohlbecker, G., Richter, L. and Butz, M., J. Clin. Chem. Clin. Biochem., 1/7, 309, (1979).

Krebs, H. A. and Johnson, W. A., Enzymologica, 4, 148, (1937).

Krugers Dagneaux, P. G. L. C., Klein Elhorst, J. T. and Olthius, F. M. F. G., Clin. Chim. Acta, 71, 319, (1976).

Kunin, R. and Myers, R. J., J. Amer. Chem. Soc., £9, 2874, (1947).

Laker, M. F., Hofmann, A. F. and Meerise, B. J. D., Clin. Chem., 26, 827, (1980).

Lehmann, C. G., Physiological Chemistry, 2nd edition, Cavendish Society, London, p.4l, (1851).

Lehninger, A. L., Biochemistry, 2nd ed., Worth Publishers, inc., N. Y., pp. 4^"W ,"-(lt75).

Lee, J. C. and Timasheff, S. N., J. Biol. Chem., 256, 7193, (1981).

Liang, C.-C. and Ou, L.-C., Biochem. J ., 121, 447, (1971).

Liao, L. L. and Richardson, K. E., Arch Biochem. Biophys., 153, 438, (1972). 112

Liao, L. L. and Richardson, K. E., Arch. Biochem. Biophys., 154, 68, (1973).

Liao, L. L. and Richardson, K. E., Biochim. Biophys. Acta, 538, 76, (1978).

Libert, B., J. Chromatog., 210, 540, (1981).

Liebich, H. M., Pickert, A., Stierle, U. and Woell, J., J. Chromatog., 199, 181, (1980).

Lonsdale, K., Nature, 217, 56, (1968).

Lunder, T. L. and Messori, F., Chromatographia, 12, 716, (1979).

MacFayden, D. A., J. Biol. Chem., 158, 107, (1945).

Madorsky, M. L. and Finlayson, B., Invest. Urol., 14.' 274, (1977).

Mayer, W. J ., McCarthy, J. P. and Greenberg, M. S., J. Chromatog. Sci., 12, 656, (1979).

M cBride, R. S ., J . Amer. Chem. S o c ., 34^, 393, (1912).

Meister, A. J., J. Biol. Chem., 197, 309, (1952).

Menache, R. , Clin. Chem., 20^, 1444, (1974).

M iller, G. H., Vermeulen, C. W. and Moore, J. D., J. Urol., 79, 607, (1958).

Moyer, A. J., Appl. Microbiol., jL' 1, 7, (1953).

Moyer, G. G. , Markow, D. and Karp, F., Clin. Chem., 9, 334, (1963).

Murthy, M. S. R., Talwar, H. S., Nath, R. and Thind, S. K., I. R. C. S. Medical Science, j), 683, (1981).

Nishi, Y. and Shimizu, T., Japan. Kokai 78 48,796, in Chem. A b s tra c ts , 8j), a b s t r a c t 125690u, (1978).

Nord, F. F. and Vittucci, J. C., Arch. Biochem., 14, 229, (1947).

O'Keefe, C. M., Cies, L. and Smith, L. H., Jr., Biochem. Med., 7, 299, (1973). 113

Olthius, F. M. F. G., Marslag, A. M. G., Klein Elhorst, J. T. and Kruger Dagneaux, P. G. L. C., Clin. Chim. Acta, 25, 123, (1977).

Osswald, H. and Hautmann, R. , Urol. Int., 34.* 440, (1979).

Patel, D. J ., Bhatt, R. A. and Bafna, S. L., Chem. Ind., (London), p. 2110, (1967).

Payes, B. and Laties, G. G., Biochem. Biophys. Res. Commun., 1J3, 179, (1963).

Pernet, J. L. and Pernet, A., Ann. Biol. Clin., 20, 737, (1962).

Pernet, J. L. and Pernet, A., Ann. Biol. Clin., 22^, 1189, (1965).

Phares, E. F., Mosbach, E. H., Denison, F. W. and Carson, S. F., Anal. Chem., 2j4, 660, (1952).

Pik, C. and Kerckhoffs, H. P. M., Clin. Chim. Acta, 2/ 300, (1963).

Pinto, B. and Paternain, J. L. , Invest. Urol., 1_5, 502, (1978).

Powers, H. H. and Levatin, P., J. Biol. Chem., 154, 207, (1944).

Raghaven, K. G. and Richardson, K. E., Manuscript in preparation, (1981).

Rao, G. G. and Aravamudan, G., Anal. Chim. Acta, JL3.r 415, (1955).

Rapado, A., Traba, M. L., Castrillo, J. M. and Delatte, L. C., in: Oxalate in Human Biochemistry and Clinical Pathology, G. A. Rose, W. G. Robertson and R. W. E. Watts, eds., Proceedings, London, p. 168, (1979).

Raper, K. B. and Fennel, D. I., The Genus Aspergillus, Williams and Wilkins, Baltimore, (196 5).

Resnik, F. E., Lee, L. A. and Powell, W. A., Anal. Chem., 27, 928, (1955).

Ribeiro, M. E. and Elliot, J. S., Invest. Urol., 2, 78, (1964). 114

Rich, W., Johnson, E., Lois, L., Kabra, P., Stafford, B. and Morton, L., Clin. Chem., 26, 1492, (1980).

Richards, M., J. Chromatog., 115, 259, (1975).

Richardson, K. E., Endocrinol., 21, 128, (1964).

Richardson, K. E., Toxicol. Appl. Pharmacol., 24, 530, (1973).

Richardson, K. E. and Farinelli, M. P., in: Urolithiasis: Clinical and Basic Research, L. H. Smith, W. G. Robertson and B. Finlayson, eds., Plenum Publishing Corp, N. Y., p. 855, (1981).

Richardson, K. E. and Tolbert, N. E., J. Biol. Chem., 236, 1280, (1961).

Riggs, W. M. and Bricker, C. E. , Anal. Chem., 21(3, 897, (1966).

R o b ertso n , W. G. and Norden, B. E. C ., in : S c i e n t i f i c Foundation of Urology, D. I. Williams and G. D.Chishholm, eds., Heinemann, London, p. 254, (1976).

Robertson, W. G., Peacock, M., Heyburn, P. J., Marshall, R. W., Williams, R. E. and Clark, P. B.,in: Oxalate in Human Biochemistry and Clinical Pathology, G. A. Rose, W. G. Robertson and R. W. E. Watts, eds., Proceedings, London, p. 1731 (1979).

Robertson, W. G., Peacock, M., Heyburn, P. J., Marshall, D. H., and Clark, P. B., Brit, J. Urol., 50, 449, (1978).

Robertson, W. G., Peacock, M. and Norden, B. E. C., Lancet, 2, 21, (1969).

Robertson, W. G. and Rutherford, A., Scand. J. Nephrol. (Suppl.), J53, 85, (1980).

Robertson, W. G., Varnavides, C. K., Marshall, R. W., Peacock, M. and Norden, B. E. C., in: Colloquium on Renal Lithiasis, B. Finlayson and W. C. Thomas, Jr., eds., The University Presses of Florida, Gainesville, FL, p. 101, (1976).

Rofe, A. M. and Edwards, J. B., Bioch. Med., 2j), 323, (1978).

Romano, M. and Cerra, M., Ital. J. Biochem., 2JD, 129, (1971). 115

Runyan, T. J. and Gershoff, S. N. , J. Biol. Chem., 240, 1889, (1965).

Salkowski, E., Z. Physiol. Chem., 29, 437, (1900).

Sallis, J. D., Lumley, M, F. and Jordan, J. E., Biochem. Med. , 18, 371, (1977).

Sargent, R. and Rieman, W. Ill, Anal. Chim. Acta, lj), 203, (1958).

Sawaki, S., Hattori, N. and Yamada, K., J. Vitaminol., 12, 210, (1966a).

Sawaki, S., Hattori, N. and Yamada, K., J. Vitaminol., 12, 303, (1966b).

Sawaki, S. and Yamada, K., Nature, 210, 91, (1966).

Sawaki, S., Hattori, H. and Yamada, K., J. Biochem., 62, 263, (1967).

Schmidt, K., Hagmaier, V., Bruchelt, G. and Rutishauser, G., Urol. Fes., 8, 177, (1980).

Scott, R. W., Anal. Chem., 27, 367, (1955).

Seidell, A., in: Solubilities of the Inorganic and Metal Organic Compounds, 4th ed., W. F. Linke, ed., American Chemical Society, Washington, D. C., (1965).

Shimazono, H., J. Biochem., 4j2, 321, (1955).

Shimazono, H. and Hayaishi, O., J. Biol. Chem., 227, 151, (1957).

Singh, P. P., Kothari, L. K., Sharma, D. C. and Saxena, S. N., Amer. J. Clin. Nutr., 25, 1147, (1972).

Smith, L. H., Jr., in: Strauss and Welt's Diseases of the Kidney, 3rd ed., L. E. Earley and C. W. Gottschalk, eds., Little, Brown and Co-., Boston, p. 893, (1979).

Smith, L. H., Jr., Bauer, R. L. and Williams, H. E., J. Lab. Clin. Med., 78, 245, (1971).

Tanaka, K., Ishizuka, T., and Sunahara, H., J. Chromatog., 174, 153, (1979). 116

Thomas, J., Champagnac, A., Thomas, E., Landier, J. and Steg, A., in: Oxalate in Human Biochemistry and Clinical Pathology, G. A. Rose, W. G. Robertson and R. W. E. Watts, eds., Proceedings, London, p. 186, (1979).

Thomas, J ., Melon, J. M., Thomas, E., Duburque, M. T ., Pierre, R. , Charransol, G. and Desgrez, P., in: Urinari Calculi, L. C. Delatte, A. Rapado and A. Hodgkinson, eds., B. Karger, N. Y., p. 57, (1973).

Tocco, D. J ., Duncan, A. E. W., Noll, K. M. and Duncan, D. E., Anal. Biochem., £4, 470, (1979).

Tolbert, N. E., Ann. Rev. Bioch., J50, 133, (1981).

Tomlinson, N., Campbell, J. J. R. and Trussell, P. C., J. Bact., 61, 17, (1951).

Tschope, W. and Ritz, E., in: Urolithiasis: Clinical and Basic Research, L. H. Smith, W. G. Robertson and B. Finlayson, eds., Plenum Publishing Corp, N. Y., p. 951, (1981).

VanReem, R., Indacochea, N. and Hess, W. C., J. N utr., 69, 397, (1959).

VanSlyke, D. D. and Sendroy, J., Jr., J. Biol. Chem., 84, 217, (1929).

Vittu, C. and Lemahieu, J. C., Ann. Biol. Clin., 23^, 913, (1965).

Von Nicholai, H. and Zilliken, F., J. Chromatog., 92, 431, (1974).

Warren, W. A., J. Biol. Chem., 2^5, 1673, (1970).

Weinhou3e, A. and Friedman, B., J. Biol. Chem., 191, 707, (1951).

Wheaton, R. M. and Bauman, W. C., Ind. Eng. Chem., 45, 228, (1953a).

Wheaton, R. M. and Bauman, W. C., Ann. N. Y. Acad. Sci., 57, 159, (1953b).

Williams, H. E. , Kidney Int., _13, 410, (1978).

Williams, H. E. and Smith, L. H., Jr., Science, 171, 390, (1971). 117

Williams, H. E. and Smith, L. H., Jr., in: The Metabolic Basis of Inherited Disease, 4th ed., J. B. Stanbury, J. B. Wyngaarden and D. S. Prederickson, eds., McGraw-Hill Book Co., N.Y., p. 182, (1978).

Yao, S. J ., Wolfson, S. K. Jr. and Tokarsky, J. M., Bioelectrochem. BiOeng., 2, 348, (1975).

Yarbro, C. L. and Simpson, R. E., J. Clin. Lab. Med., 48, 304, (1956).

Yriberri, J. and Posen, S., Clin. Chem., £6, 881, (1980).

Zarembski, P. M. and Hodgkinson, A., J. Nutr., lj>, 627, (1962).

Zarembski, P. M. and Hodgkinson, A., Biochem. J., £6, 218, (1 9 6 5 a).

Zarembski, P. M. and Hodgkinson, A., Biochem. J. , 9(5, 717, (1965b).

Zarembski, P. M. and Hodgkinson, A., Clin. Chim. Acta, 25, 1, (1969).