Pharmacokinetics of sodium dichloroacetate


Material Information

Pharmacokinetics of sodium dichloroacetate
Physical Description:
vii, 190 leaves : ill. ; 29 cm.
Chu, Pei-I, 1959-
Publication Date:


Subjects / Keywords:
Sodium -- pharmacokinetics   ( mesh )
Acetic Acids -- pharmacokinetics   ( mesh )
Dissertations, Academic -- Pharmaceutics -- UF   ( mesh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1987.
Includes bibliographical references (leaves 182-189).
Statement of Responsibility:
by Pei-I Chu.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000957149
notis - AER9837
oclc - 19891843
System ID:

This item is only available as the following downloads:

Full Text








I would like to thank my major professor, Dr. Stephen

H.Curry, for his excellent guidance and support. His

wonderful personality has also been appreciated. I would

also like to thank the committee members, Dr. Hartmut

Derendorf, Dr. Margaret O. James, Dr. Stephen G. Schulman,

and Dr. Peter Stacpoole, for their advice throughout my

graduate career. I would also like to thank Dr. George

Henderson for reviewing the manuscript for me. Finally, I

would like to thank my parents and my husband for their

constant encouragement.



ACKNOWLEDGEMENTS ..................................... ii

ABSTRACT ............................................. vi


I. INTRODUCTION ...................................... 1

Dichloroacetic Acid ........................... 1
Lactic Acidosis ............................ .. 2
Lactate Formation .......................... 2
Formation of Glucose from Lactate .......... 3
Lactic Acidosis .......................... 4
Pharmacological Activities of Sodium
Dichloroacetate ............................ 6
DCA and Diabetes .......................... 11
DCA and Lactic Acidosis .................... 13
Metabolism ................................. 13
Kinetics of DCA in Humans .................. 14
Analysis of DCA ............................ 16
Oxalic Acid .................................. 16
Oxalate Content of Foods and Nutrition ..... 18
Endogenous Synthesis ....................... 18
Assay for Oxalate in Biological Fluids ........ 20
Precipitation Methods ..................... 20
Colorimetric Methods ...................... 21
Fluorimetric Methods ...................... 22
Enzymatic Methods ......................... 23
Chromatographic Methods .................... 24
Radioisotope Methods ...................... 25
The Real and Apparent Plasma Oxalate ....... 26
Stability .. .................................. 26
Accelerated Stability Study ................ 27
Effects of pH on Stability ................. 28

II. OBJECTIVES ...................................... 31

III. MATERIALS AND METHODS .......................... 33

Chemicals ..................................... 33
Analytical Methods ........................... 33
Analysis of DCA Powder ..................... 33
Preparation of DCA Injection Solutions ..... 34
Preparation of DCA Oral Dosage Form ........ 34

Preparation of DCA Oral Dosage Form ........ 34
Preparation of Enzyme Inhibitors ........... 34
Assay of DCA ............................... 35
Studies of the Effects of DCA Metabolite
(Oxalate, Glyoxalate, Glycolic Acid) to
DCA Assay ................................. 36
Study of DCA Protein Binding Using a Dianorm
Equilibrium Dialyser ....................... 36
Red Blood Cell Localization study and Partition
Coefficient ................................ 38
Stability of DCA .............................. 39
Assay of Oxalic Acid .......................... 40
Urine Oxalic Acid by Atomic Absorption ..... 40
Urinary Oxalic Acid by Gas Chromatography .. 40
Assay of Plasma Oxalate .................... 41
Oxalate in Plasma .......................... 42
Identifications of Oxalic Acid Derivatives
Using GC-MS ............................. 43
Pharmacokinetic Studies of DCA ................ 44
Selection of Subjects ...................... 44
Five-Dose DCA Study ....................... 44
Cross Over Study ........................... 46
Bioavailability Study ...................... 53
Multiple Dose DCA Study .................... 54
Computer Modeling .......................... 56

IV. RESULTS ......................................... 57
Identification of DCA ......................... 57
Analysis of DCA in Plasma and Urine ........... 57
Analysis of Oxalate in Plasma and Urine ....... 64
Structual Verification of Chloroethylated
Esters of Oxalic Acid and Malonic Acid
Formed in Derivatized Standards and
Plasma by GC-MS ............................ 64
Analysis of Oxalate in Plasma .............. 64
Analysis of Oxalate in Urine ............... 73
Comparison of Urinary Oxalate by Atomic
Absorption and GC Method ................ 76
Stability of DCA ............................. 76
Extent of Protein Binding Determined by Using
Equilibrium Dialysis ....................... 77
Red Cell Localization and Erythrocyte
Partitioning ............................ 80
In Vivo Pharmacokinetics ..................... 86
Five Dose Study ........................... 86
Crossover Study ............................ 91
DCA Plasma Concentration and Fitting of
Pharmacokinetic Data ................... 97
Bioavailability of DCA .................... 115
Male vs. Female ........................... 116
DCA and Oxalate Excretion in Urine ........ 116
Plasma Oxalate and Urinary Oxalate
Excretion ......... ................... 144
Multiple Dose DCA Study ................... 146

Digit Symbol Substitution Test .............. 146

V. DISCUSSION ....................................... 150

Analytical Methods ......................... 151
Analysis of DCA in Plasma and Urine ...... 151
Analysis of Oxalate in Urine ............. 152
Analysis of Oxalate in Plasma ............ 153
Stability of DCA in Various Buffer Solution 154
Protein Binding ......................... 154
Red Cell Localization and Partition
Coefficient .............................. 155
Pharmacokinetic Studies ..................... 157
Five Dose Intravenous Infusion ........... 157
Crossover Study .......................... 158
Evaluation and Fitting of Pharmacokinetic
Data ................................. 160
Apparent Volume of Distribution .......... 162
Renal Clearance of DCA ................... 163
Estimation of Elimination Rate Constant
Using Urinary Data .................... 165
Bioavailability of DCA ...................... 167
The Percent Unabsorbed-Time Plots by
Wagner-Nelson Method .................. 167
Plasma Oxalate and Urinary Oxalate
Excretion after IV or Oral DCA
Administration ........................ 170
Half Life Change ........................ 170

VI. CONCLUSIONS ..................................... 173

APPENDIX ............................................ 179

REFERENCES .......................................... 182

BIOGRAPHICAL SKETCH ................................ 190

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Pei-I Chu

December, 1987

Chairman: Dr. Stephen H. Curry
Major Department: Pharmaceutics

The objective of this study was to determine the

pharmacokinetic properties of sodium dichloroacetate (DCA) in

healthy volunteers. DCA is a possible treatment for lactic

acidosis. Plasma concentrations and urinary excretion of DCA

were monitored by gas chromatography using electron capture

detection. An assay method for one of the metabolite of DCA,

namely oxalate was developed by gas chromatography using

electron capture detection.

The stability of IV dosage forms of DCA was studied. The

stability study affected the sterilization method used in the

preparation of injectable DCA solutions.

The kinetics of plasma protein binding and red blood

cell/plasam distribution were determined for DCA. The average

fraction of protein binding was 0.23. The average partition

coefficient between the isotonic buffer solution and the red


blood cells was 0.44.

The distribution and elimination of DCA followed a one

compartment or two compartment model with zero or first order

absorption and first order elimination. The mean terminal

half life was 2.07 + 0.91 hr. The average volume of

distribution was 18.86 + 4.04 liters. The average renal

clearance was 42.9 23.9 ml/hr. Renal clearance was

independent of urinary pH and urinary flow rate. There was no

evidence that the bioavailability of DCA was less than unity.

There was a small increase in plasma oxalate following

intravenous DCA treatment. There was not significant

difference in the area under curve (from 0 to 16 hrs) of

oxalate between the IV (27.59 7.66 and oral DCA

(25.7 1.51 treatment. There was no significant

difference in the percent of 24 hour urinary oxalate

excretion/DCA dose between IV and oral administration.

The elimination half lives changed after repetitive DCA

dosing. The volume of distribution did not change after

repetitive DCA dosing. The observation disagreed with earlier

observations in patients. This may due to that there was a

positive inotropic effect of DCA on patients but not on healthy


The order of elimination remained unchanged despite the

half life changes. A Michaelis-Menten type mechanism was used

to explain this phenomenon. Because the half life change

phenomenon of DCA is still unresolved, it is too early to

predict if DCA can be used in a large scale on patients.



Dichloroacetic Acid

This dissertation is concerned with the pharmacology

of sodium dichloroacetate. This chemical has recently

become important in the treatment of lactic acidosis.

Dichloroacetic acid was first synthesized over 100

years ago. It has been known for many years as an

escharotic, topical keratolytic and topical astringent

compound (Windholz et al., 1976). It is a liquid at room

temperature (melting point 13.5"). Its boiling point is

194'C. It is soluble in water, alcohol and ether.

Dichloroacetic acid is a relatively strong organic acid.

Its dissociation constant is 5140 x 10(-5) (pKa = 1.29).

This constant lies between that of monochloroacetic acid

[155 x 10(-5)] and trichloroacetic acid [12100 x 10(-5)].

Chemically, dichloroacetic acid can be obtained through

the reaction of chloral hydrate and aqueous sodium cyanide

(Rosenblum et al., 1960):

boil I
C13C-CHOH + NaCN ----> C1-C-COOH (Eq. 1.1)


Dichloroacetic acid can be used as a starting

material to form substituted carboxylic acids. For

example, because it is unstable, boiling dichloroacetic

acid in ten volumes of water for 30 minutes to one hour

can produce glyoxylic acid (Karrer, 1938). Dichloroacetic

acid is also unstable in alkali. Heating it in alkaline

solution will give oxalic and acetic acids (Molinari,


Lactic Acidosis

Lactate Formation

Tissues need energy to do work. Adenosine

triphosphate (ATP) provides each tissue with the energy it

requires. It is produced in the metabolism of acetyl

Coenzyme A in the citric acid cycle.

Cardiac and skeletal muscles require large amounts of

ATP during exercise. In this condition, even maximal

rates of mitochondrial oxidation are inadequate. During

exercise, skeletal muscle and to a lesser extent cardiac

muscle obtain the necessary large increment in ATP from

glycolysis, the sequence of reactions leading from glucose

to lactate.

The conversion from glucose to lactate in the cytosol

can proceed in the complete absence of oxygen, which is at

very low concentrations in exercising muscle and is

essential for ATP production. The overall process can be

summarized by the following reaction.

Glucose + 2ADP + 2Pi ----> 2 lactate + 2 ATP (Eq. 1.2)

The buildup of lactic acid in working muscles and in the

blood stream is the manifestation of what is known as an

"oxygen debt".

Formation of Glucose from Lactate

The replenishment of carbohydrate from non-

carbohydrate precursors cannot proceed by simple reversal

of glycolysis. For the overall process of conversion of

lactate to glucose, under standard conditions, the Gibb's

free energy is equal to 48 Kcal per mole. Lacate is

converted to pyruvate in the cytoplasm by lactate

dehydrogenase, with production of the reduced form of

nicotinamide adenine dinucleotide (NADH). Pyruvate thus

formed enters the mitochondria and gives rise, by the

pyruvate carboxylase reaction, to oxaloacetate. This, by

transamination with glutamate, gives a-ketoglutarate and

aspartate. The last two compounds then leave the

mitochondria and in the cytosol undergo transamination to

give glutamate and oxaloacetate.

Glutamate generated in the cytosol renters the

mitochondria for continuation of the mitochondrial

transamination, whereas cytosolic oxaloacetate is

converted to phosphoenolpyruvate (PEP), by the pyruvate

carboxykinase, and PEP is converted to glucose. The

overall reaction can be described by the following


2 Lactate + 6 ATP ----> glucose + 6ADP + 6Pi (Eq. 1.3)

Lactic Acidosis

Lactic acidosis (L.A.) is characterized by elevated

blood lactate levels ( 5 mM/1), along with a decrease in

both blood pH and bicarbonate concentration. In this

condition, the [lactate]:[pyruvate] ratio is also raised

(normal is approximately 10). This is the most commonly

encountered form of metabolic acidosis. It can be caused

by the overproduction of lactate, underutilization of

lactate, or both. Lactate production is normally balanced

by lactate utilization, with the result that in a normal

body the blood concentration is not greater than 1.2 mM.

L.A. can be divided into two groups. In the A type, there

is accelerated glycolysis which can be due to shock

(tissue hypoperfusion and thus tissue hypoxia). Type B is

not associated with clinical shock.

Examples of type A L.A. include cardiogenic (e.g.,

post myocardial infarction) or endotoxin shock, left

ventricular failure and general anesthesia induced

depressed cardiopulmonary function. The clinical signs of

shock (e.g. cyanosis) usually follow rather than precede

the rise in blood lactate. In type A, mortality seems to

be related to the blood lactate concentration: lactate

level > 4 mmol/L has a 70-100% mortality. The cause of

lactate accumulation in type A probably is a combination

of both underutilization and overproduction of lactate by

liver and peripheral tissues. When the blood pH is below

7, the liver switches from being the principal organ of

lactate removal to a major site of lactate synthesis.

Type B L.A. occurs in association with diseases such

as uremia, bacterial infection, alcoholism, starvation and

diabetes mellitus. It is well established that the first

and most important step in treating lactic acidosis is to

attempt to correct the underlying cause (e.g. infection

with antibiotics, and heart failure with pressor drugs).

Also, an attempt to control the acidosis associated with

lactic acid accumulation is made. Vigorous administration

of sodium bicarbonate, sometimes supplemented with

dialysis or potent diuretics to minimize the risk of fluid

overload, is also used. Raising the pH (2 7.0) will in

theory switch the liver from a lactate producer to a

lactate consumer (Cohen and Woods, 1976). Recently, some

studies have indicated that alkalinization of body fluids

stimulates lactate production (Arieff et al., 1982, and

Graf et al., 1985). Even massive doses of alkali may not

succeed in substantially increasing the plasma bicarbonate

level in malignancy-associated chronic lactic acidosis

(Fraley et al., 1980, and Fields et al., 1981). When

bicarbonate is administered, frequent monitoring of the

acid-base status is essential so that alkalemia will not

occur. The swift transition from severe acidemia to

alkalemia can produce tetany, altered mental status,

generalized convulsions, and cardiac arrhythmias. In

conclusion, treatment of L.A. with NaHCO3 needs to be

reevaluated (Stacpoole, 1986).

Pharmacological Activities of Sodium Dichloroacetate

Sodium dichloroacetate (DCA) is the sodium salt of

dichloroacetic acid. At room temperature it is a white

crystalline solid. Like most salts, it dissolves in water

readily (> 100 mg/ml).

The pharmacological effects of DCA were not studied

extensively until diisopropylammonium dichloroacetate

(DIPADCA) was introduced as a hypoglycemic reagent (Lorini

and Ciman, 1962). Lorini & Ciman found that DIPADCA

exhibited a constant hypoglycemic action in alloxan

diabetes, but it did not affect blood glucose in normal

animals. It was concluded that in diabetic animals

DIPADCA improves the peripheral utilization of glucose

when the animal is impaired. Stacpoole & Felts (1970)

carried out an in vitro study on DIPADCA and DCA and found

that both DIPADCA and DCA stimulated glucose-U-14C

oxidation to 14CO2 in isolated hemidiaphragms from

diabetic but not from nondiabetic rats, whereas

diisopropylammonium hydrochloride (DIPA) alone was not

effective in promoting glucose oxidation in tissue either

from diabetic or non-diabetic rats. They concluded that

the effect of DIPADCA in vivo is due to its acid moiety.

Later, the effects of DCA on various aspects of

intermediary metabolism were studied in several models.

DCA has been shown to increase glucose oxidation

(Stacpoole & Felts 1970) in skeletal muscle of diabetic

rats, and to increase glucose, pyruvate, and lactate

extraction, but to decrease fatty acid oxidation in the

perfused dog heart (McAllister et al., 1973).

Whitehouse and Randle (1973) studied the activity of

pyruvate dehydrogenase in extracts of rat hearts in vitro

perfused with media containing glucose and insulin +

acetate + dichloroacetate. Dichloroacetate (100 AM, 1 mM

or 10 mM) increased the activity of pyruvate dehydrogenase

in perfusion with glucose or glucose and acetate.

Whitehouse and Randle (1973) concluded that DCA may

facilitate the conversion of pyruvate dehydrogenase from

an inactive (phosphorylated) form into an active

(dephosphorylated) form. Whitehouse et al.(1974) also

found that DCA is an inhibitor of pig heart pyruvate

dehydrogenase kinase, and thus increases the pyruvate

dehydrogenase activity. The concentration for 50%

inhibition was approximately 100 MM. Dichloroacetate

cannot increase the catalytic activity of purified pig

heart pyruvate dehydrogenase. Injection of DCA into rats

starved overnight led, within 60 min, to activation of

pyruvate dehydrogenase in extracts from heart, psoas

muscle, adipose tissue, kidney and liver. The blood

lactate started to fall within 15 mins and reached a

minimum after 60 mins. The blood concentration of glucose

fell after 90 mins and reached a minimum after 120 mins.

It was concluded that inhibition of pyruvate dehydrogenase

kinase by DCA might account for the activation of pyruvate

dehydrogenase and pyruvate oxidation. Blackshea and

Holloway (1974) studied the metabolic effects of DCA in

starved rats. A dose of 300 mg/kg/h DCA was infused into

rats for 4 hours. Blood glucose decreased significantly

and plasma lactate and pyruvate decreased 50% and 30%

respectively. The livers were removed and assayed at the

end of 4 hours' infusion. There were significant

decreases in hepatic glucose, glucose-6-phosphate, 2-

phosphoglycerate, the lactate/pyruvate ratio, citrate and

malate and also alanine, glutamate and glutamine

suggesting a diminished supply of gluconeogenic

substrates. Animals subjected to a functional hepatectomy

at the end of infusions lasting two hours showed no

difference in blood-glucose disappearance but a highly

significant decrease in the rate of accumulation of

lactate, pyruvate, and alanine compared with control

animals. It was concluded that DCA caused hypoglycemia by

decreasing the net release of gluconeogenic precursors

from extrahepatic tissues. This finding is in agreement

with the theory of activation of pyruvate dehydrogenase in

rat muscle described by Whitehouse and Randle (1973).

This effect suggested that a combination of insulin and

DCA might be useful in the therapy of diabetes.

Crabb et al.(1976) examined the effects of DCA on

liver metabolism with isolated hepatocytes and found that

DCA has a small but significant effect on the starved rat

liver It does not increase glucose utilization but

slightly inhibits gluconeogenesis from lactate and

stimulates gluconeogenesis from alanine. With hepatocytes

prepared from fed rats, DCA was found to activate pyruvate

dehydrogenase, to increase the utilization of lactate and

pyruvate without effecting an increase in the net

utilization of glucose. Crabb et al. (1976) concluded

that the hypoglycemic action of DCA is a result of the

peripheral effects.

The results of animal studies on DCA are also

encouraging. Loubatieres et al.(1976) produced

experimental hyperlactataemia in dogs and found that

administration of 30 mg/kg/h produced a reduction in the

hyperlactataemia produced by phenformin (from 50.6 10.2

mg per 100 ml to 14.3 2.5 mg per 100 ml, n = 13), by

intense muscular work (from 42.3 10.3 mg/ 100 ml to 9.0

1.5 mg/ 100 ml, n = 3), or by adrenaline (from 61.3 13

mg/100 ml to 29 2.3 mg/100 ml, n = 3). Dichloroacetate

did not reduce the hyperlactataemia produced by hypoxia

(84.7 12.1 mg/100 ml v.s. 100.8 8.7 ml, n = 3).

Dichloroacetate also slightly reduced the increase in

lactataemia induced by alloxan (50 mg/kg IV)


Dichloroacetate can also reverse hyperlactataemic,

hyperpyruvataemic and hyperalaninaemic effects of

phenformin in both normal and diabetic rats (Holloway &

Alberti, 1975). In functional hepatectomized dogs treated

with DCA, the arterial lactate decreased by 50% and both

muscle and liver lactate decreased to 80% of control

values (Park et al., 1979). Similar results were obtained

in diabetic dogs treated with DCA, after functional


The metabolic effects of DCA alone and with

phenformin were studied by Man and Alberti (1976) for up

to 28 days. Both blood glucose and lactate were decreased

gradually for the first five days. Phenformin caused a

decrease in hepatic glycogen and increases in alanine,

pyruvate, lactate, phosphoenolpyruvate, citrate, malate

and aspartate, all of which were reversed by DCA. Man and

Alberti (1976) concluded that DCA prevented the

development of phenformin-induced metabolic abnormalities

without attenuating the beneficial hypoglycemic effects,

and the combination could be of use in the treatment of

mild diabetes.

DCA and Diabetes

Irsigler et al. (1977) first tried DCA in a 76-year-

old diabetic woman who was admitted after buformin

treatment, and with severe lactic acidosis (pH = 6.9,

lactate 24 mmol/l). DCA was given as a 4 g IV bolus and

then 12 g over 1 hr; and subsequently 7 g by IV infusion

over the next hour. Unfortunately, the lady died. This

showed that severe biguanide-induced lactic acidosis was

not reversed by DCA.

Long-term treatment (50 mg/kg/day, twice daily)

lactic acidosis was investigated by Coude et al. (1978).

Lactate levels were lowered, but there was no clinical

improvement during the six months treatment. Growth and

psychomotor retardation remained unchanged. There was no

sedation or other adverse clinical effect. Coude et al.

(1978) concluded that the return to normal of lactate

levels in their 18-month-old patient was probably

secondary to an activation of the muscular pyruvate

decarboxylase and subsequent stimulation of pyruvate and

lactate oxidation by DCA. The long term effects of DCA

therapy, particularly on growth and psychomotor

retardation, needed further investigation.

Stacpoole et al. (1976) treated human diabetes

mellitutus with dichloroacetate. This study suggested

that DCA inhibits peripheral release of 3-carbon fragments

and of free fatty acid, thereby reducing plasma glucose

and lipids without affecting insulin secretion.

Standl et al. (1977) examined the effects of DCA in

ten buformin-treated maturity-onset diabetic patients, and

found that DCA did not affect blood lactate or ketone

bodies under normal circumstances. During exercise, the

DCA-treated patients showed smaller increases in lactate

and less ketone body utilization than patients given

buformin alone.

The metabolic effects of DCA were evaluated for six

to seven days in patients with diabetes mellitus and

hyperlipoproteinemia (Stacpoole et al. 1978). DCA

decreased fasting hyperglycemia and induced falls in

plasma lactate and alanine levels. Plasma insulin, free

fatty acid and glycerol levels were not affected. Some

patients experienced mild sedation in the period of

testing. Serum uric acid rose, whereas excretion and

renal clearance fell. A twenty-one-year-old man was

treated with DCA for severe receptor-negative homozygous

familial hypercholesterolemia (Stacpoole et al., 1979).

After 16 consecutive weeks on an oral regimen at 50 mg/kg,

the patient developed symptoms of polyneuropathy. The

neuropathy improved upon cessation of DCA therapy.

Diabetes tends to cause neuropathies which could be

exacerbated by DCA and go unrecognized. However, because

of the potential risks of oral DCA to the central nervous

system, long-term treatment of diabetes mellitus with DCA

has not been further investigated.

DCA and Lactic Acidosis

Irsigler et al. (1979) used DCA to treat human

phenformin-related lactic acidosis. Their first two

patients were severely acidotic (pH < 6.95) and did not

survive. Their third patient was only slightly acidotic

(pH = 7.3), and administration of DCA, fluids, glucose,

insulin, furosemide, tris-buffer, and oxygen brought about

marked clinical improvement.

Stacpoole et al. (1983) administered DCA to thirteen

patients with lactic acidosis of various causes. Their

acidemia had resisted treatment with sodium bicarbonate.

The metabolic effects of DCA were evaluated in 11 of the

patients. In seven patients, DCA significantly reduced

arterial lactate and raised the levels of bicarbonate and

pH. In six of these seven, the acidemia resolved

completely with therapy. Despite improvement in their

lactic acidemia, all patients but one died of their

underlying disease, although no serious drug-related

toxicity occurred. It was concluded that DCA is a safe

and effective adjunct in the treatment of lactic acidosis



Demaugre et al. (1978) studied the inhibition of

gluconeogenesis by DCA in isolated rat hepatocytes. They

found that hepatocytes and liver mitochondria metabolized

[14C]DCA to oxalate.

Harris et al. (1978a) also postulated that DCA might

be hydrolyzed to glyoxylate, then oxidized to oxalate.

Harris et al. (1978b) studied the regulation of leucine

catabolism by DCA. They found that in isolated liver

cells, [2-14C] DCA is converted to 14C02, [14C] glycine

and [14C] oxalate. By assuming that glyoxylate

aminotransferase is involved in the transformation from

glyoxylate to glycine, they found that DCA is converted by

liver cells at about 80 nmol/min/g wet wt to glyoxylate

which then can be transaminated to yield glycine or

oxidized to 14C02 or [14C] oxalate. They also found their

commercial preparation of DCA was contaminated with 0.015%

glyoxylate. Heating DCA solutions at pH 12 can also

produce glyoxylate. Rats injected intraperitoneally with

DCA also showed a five-fold increased of urinary oxalate

excretion compared to control (Crabb & Harris, 1979).

These authors mentioned that free glyoxylate is difficult

to measure because it is continuously metabolized to

glycine and oxalate. The formation of glyoxylate and

oxalate from DCA in vivo can affect metabolic and

physiological processes. It also complicated

interpretation of the effects of DCA considerably.

Kinetics of DCA in Humans

Wells et al. (1980) studied the effects and

pharmacokinetics of DCA on 16 healthy subjects. DCA (1-50

mg/kg) was infused into the healthy volunteers over 30

min. It was found that peak plasma DCA concentrations

were linearly related to dose up to 30 mg/kg of DCA. At

doses of 35 and 50 mg/kg, four out of seven subjects

exhibited peak plasma drug concentrations

disproportionately higher than predicted by the linear

relation seen at lower doses. The plasma DCA

concentrations fell in a convex fashion with respect to

time indicating Michaelis Menten kinetics. The terminal

phase of elimination half life was 32 11 min. Plasma

drug clearance was lower for higher doses. Urinary

recovery of DCA over 12 hours was less than 1% of the

administered dose in three patients. Within 2 hours of

administration of 35 mg/kg DCA, plasma lactate

concentrations fell 75% below baseline and alanine fell

50% below baseline, while blood glucose was unaffected.

Lukas et al. (1980) infused 10 mg/kg or 20 mg/kg into four

normal subjects. Plasma DCA decay followed first order

kinetics. The average half life of the 10 mg/kg DCA dose

was 0.34 hr whereas that of the 20 mg/kg was 0.51 hr. The

volume of distribution of the 10 mg/kg dose (337 ml/kg)

was also different from that of the 20 mg/kg dose (190

ml/kg). Lukas et al. (1980) concluded that the DCA

elimination rate in humans was does dependent although

this study was limited by the number of volunteers


Analysis of DCA

Analysis of any drug is very important to the study

of its pharmacokinetic properties. DCA is conventially

assayed by gas chromatography, with detection using

electron capture. The stationary phase of the column is

Chromosorb 101, which consists of porous particles.

Separation is based on particle size of each compound.

The effluent from the column passes over a P emitter

(nickel 63). An electron from the emitter causes

ionization of the carrier gas and the production of a

burst of electrons. The electron capturing species

eluting from the column reacts with the electrons to form

ions, which are swept from the cell. The net result is a

reduction in the number of electrons found at steady state

and a drop in the standing current.

Quantitative analysis is based upon a comparison of

the height of the chromatographic peak of the analyte with

the standards. Trichloroacetic acid (TCA) is used

internal standard. The ratio of DCA to TCA peak height

serves as the analytical parameter. The calibration curve

is linear up to at least 200 Ag/ml and the coefficient of

variation varies from 2.5 to 4.03%.

Oxalic Acid

Oxalic acid (dihydrate) occurs as odorless monoclinic

crystals. Its melting point is 101.5"C. Its solubility

at room temperature is 8.9 g/100 g H20. The dihydrate

readily loses its water of crystallization when heated.

Anhydrous oxalic acid is odorless, hydroscopic and white

in color. It sublimes readily at 125. Anhydrous oxalic

acid exists in two crystalline forms, an a(orthorhombic)

form and a P(monoclinic) form. The a form melts at 187"

and the P form 182", with decomposition to formic acid,

carbon monoxide, carbon dioxide and water. Alpha

anhydrous oxalic acid is more stable at room temperature

than is th p form.

Oxalic acid is a relatively strong acid, the first

dissociation constant (pKa) is 1.23. The strong acidity

results from the electron-withdrawing inductive effect of

the second carboxyl group. The second pKa is weaker

(3.83) because the second proton has to be removed from a

negatively charged species containing an electron donating

group (CO2-).

Oxalic acid has a wide variety of industrial

applications in addition to its use as an analytical

reagent. It is a constituent of cleaning solutions for

removing paint, varnish, rust and ink stains. It is also

used for removing carbonaceous deposits from steel plates

and in dips for cleaning and phosphatizing steel. In the

chemical industry, oxalic acid is used as an intermediate

in the manufacture of dyes and as a stabilizing agent of

anhydrous hydrogen cyanide, to purify methanol and to

decolorize crude glycerol.

Oxalate Content of Foods and Nutrition

Oxalate exists in food (Table 1.1)(Hodgkinson 1977).

Among the leafy vegetables, spinach contains a high amount

of oxalate (Hodgkinson, 1977). Tea might be the largest

single source of oxalate in the English diet (Hodgkinson,

1977). Archer et al. (1957) analyzed the diets of six

healthy male adults and found an average oxalic acid

content of 920 mg/day. Zarembski and Hodgkinson (1962)

observed a lower oxalic acid value ranging from 70 mg to

150 mg daily. The reasons for this discrepancy are not

known. A high oxalate intake reduces the intestinal

absorption of calcium, magnesium, iron and a number of

trace metals because of the formation of insoluble salts

with oxalate.

Endogenous Synthesis

Feeding experiments and radioisotope studies have

indicated that a considerable number of compounds are

precursors of oxalic acid in animals and man, for example,

glycine, glyoxylic acid, glycolic acid, ethylene glycol,

ascorbic acid and tryptophan. However, only two pathways

appear to be quantitatively important at present: (1) the

oxidative metabolism of ascorbic acid; and (2) the

oxidation of glyoxylic acid. A number of compounds such

as glycine, glycolic acid and ethylene glycol form oxalic

acid because of the intermediate formation of glyoxylate

which is subsequently oxidized to oxalate. In the case of

ascorbic acid, the conversion to oxalate does not involve

the intermediate formation of glyoxalate. Ascorbic acid

is a precursor of oxalic acid in the guinea pig (Burns et

al., 1951; Banay and Dimant, 1962), and man (Hellman and

Burns, 1958; Lamden and Chrystowski, 1954). Studies with

Table 1.1. Oxalic acid content of some foods.

Food Method of Oxalic acid
preparation (mg/100 g of
fresh material)


Banana, ripe

Bacon, streaky

Beer, mild
Coca Cola









[l-14C] ascorbic acid have shown that the main excretory

products of vitamin C in man are oxalate, ascorbic acid

and dehydroascorbic acid; 14% to 40% of administered [1-

14C] ascorbic acid was excreted as [14C] oxalic acid

(Atkins et al., 1964 ; Baker et al., 1966).

Assay for Oxalate in Biological Fluid

Precipitation Methods

There is considerable uncertainty regarding the

normal concentration of oxalic acid in biological fluids.

Conventionally, oxalate can be separated by precipitation

with calcium chloride and then the titration with

permanganate (Merz and Maugeri, 1931). However,

biological fluids contain many substances that affect the

solubility or rate of crystallization of calcium oxalate

or co-precipitate with the salt (e.g. magnesium, phosphate

and sulphate) (Elliot and Eusebio, 1965; Fleisch and

Bisaz, 1964; Powers and Levatin, 1944).

Merz and Maugeri (1931) concluded that the normal

plasma oxalate ranges from 2 mg/100 ml to 4 mg/100 ml.

Thomsen (1935) pointed out that because of the

contamination of calcium oxalate, the real oxalate

concentration in normal blood is probably less than 1

mg/100 ml.

Various modifications have been introduced in an

attempt to minimize or correct for losses, for example,

addition of a known amount of sodium oxalate to the test

solution to achieve a more complete and reproducible

precipitation, (Koch and Strong, 1965,1969; Fraser and

Cambell, 1972), and the use of a radioisotope-determined

factor to correct for incomplete precipitation (Koch and

Strong, 1965). However, the potential source of error at

low oxalate concentration is the need to measure

accurately small changes in a relatively high

concentration of calcium.

Colorimetric Methods

Pernet and Pernet (1965) precipitated oxalic acid

from plasma as lead oxalate, after adding a known amount

of oxalate. The precipitated oxalate was reduced to

glyoxylate by zinc and HC1. The glyoxylate was then

determined spectrophotometrically by the reaction with

phenylhydrazine and subsequent oxidation with hydrogen

peroxide to give a red-coloured formazan. A mean value of

288 gg of anhydrous oxalic acid/100 ml was reported for

normal human blood.

Methods based upon reduction of oxalic acid to

glycolic acid, followed by reaction with 2,7-

dihydroxynaphthalene or 1,8-dihydroxynaphlene-3,6-

disulphonic acid (chromotropic acid) were also reported by

Calkins (1943), Dempsey et al., (1960), Hodgkinson and

Zarembski (1961), and Hodgkinson and Williams (1972). The

oxalate in urine determined in this method is between 17.2

mg and 46.8 mg/24 hr. Other methods include the ability

of oxalate to increase the rate of oxidation of tris (1,

10 phenanthroline)iron II complex (Ferroin) by chromium

VI ( Eswara Dutt and Mottola, 1974) and the effect of

oxalate ion on the absorbance of a red complex uranium IV-

4(-2-pyridylazo) resorcinol at 515nm (Neas and Guyon,

1972; Baadenhuijsen and Jansen, 1975). This method gives

results which agree closely with the recent isotope

dilution methods of urinary excretion at 31.5 mg/24h

(Baadenhuijsen and Jansen, 1975).

Fluorimetric Methods

Zarembski and Hodgkinson (1965) determined oxalate in

blood and urine by a fluorimetric method. Oxalic acid was

first extracted from plasma with tris-n-butyl phosphate

and then precipitated with calcium sulphate. The

precipitated oxalic acid was reduced to glyoxylic with

zinc and HC1 and coupled with resorcinol to yield a

colored fluorescence. They reported a normal plasma

oxalate range from 1.00 Ag/ml to 2.35 Ag/ml.

Britton and Guyon (1969) also measured plasma oxalate

by a fluorimetric method. Their method was based on the

quenching by oxalate of the fluorescence of a 1:1

zirconium-flavonol chelate in the dilute sulphuric acid

solution. Fluorescence was measured at 460 nm and there

was a linear relationship between the quenching effect and

oxalate concentration when it was between 0 and 10 pg/ml.

Enzymatic Methods

Oxalate decarboxylase (EC can catalyse the

decomposition of oxalate to yield CO2 and formic acid.

This enzyme can be isolated from a wood rotting fungus.

It does not require ATP, coenzyme A, acetate or magnesium

ion, nor is it oxygen dependent. Crawhall and Watts

(1961) used this enzyme to measure oxalic acid in human

plasma by monitoring the amount of CO2 released. However,

the limit of detection of this method was too high (8

Ag/ml) to measure normal plasma levels. Mayer et al.

(1963) used the enzymatic method to measure urinary

oxalate. They precipitated oxalate as the calcium salt

and then dissolved the salt in potassium citrate buffer,

pH 3.2. EDTA was added to the reaction mixture to

suppress an enzyme inhibitor which existed in the urine.

The mean daily oxalate excretion by normal adults was 20.5

mg. Similar results were obtained by other people using a

modified enzymatic method (Ribeiro and Elliot, 1964;

Hallson and Rose 1974).

Methods involving the use of a second enzyme to

determine the formic acid released by the action of

oxalate decarboxylase were described by Jakoby (1974), and

Costello et al. (1976). A limitation to the enzymatic

methods is the relatively large amount of enzyme required

for each determination and the consequent high cost. The

existence of enzyme inhibitors such as phosphate and

sulphate in biological fluids is also a problem.

Chromatoqraphic Methods

Gas and liquid chromatographic methods can be applied

in measuring oxalate in biological fluids. Basically

these methods involve two steps : (1) separation from

impurities, (2) derivatization. Separation can be made by

extracting the sample with HC1-ethanol, ether ethanol or

tri-n-butyl phosphate (Duburque et al., 1970; Charransol

and Desgrez, 1970; Mee and Stanley, 1973). Separation can

also be carried out with an ion exchange column (Chalmers

and Watts, 1972, Krugers et al., 1976). The separated

solvent extract can be esterified with methanol, ethanol,

isopropanol or BSTFA-TMS. The esterified product can be

injected into a gas chromatograph and detected with a

flame ionization detector. Murray et al. (1982) measured

urinary oxalic acid by derivatization coupled with liquid

chromatography. Oxalic acid in urine was reacted with o-

phenylenediamine to form the strongly UV-absorbing

compound 2,3-dihydroxyquinoxaline. Isolation and

quantitation of this derivative were accomplished using a

reverse phase C8 column. The mobile phase was 5% methanol

in 0.1 M ammonium acetate buffer (pH 6.6). Urinary

oxalate ranged from 3.8 to 24.0 ig/ml.

Radioisotope Methods

Williams et al. (1971) measured the renal clearance

of 14C oxalic acid and concluded that the average

concentration of oxalic acid in normal human plasma is

16.5 gg/100ml. Hodgkinson and Wilkinson (1974) did a

similar study and suggested that normal human plasma

values range from 11.8 Ag/100 ml to 14.3 Mg/100 ml.

Constable et al. (1978) also measured plasma oxalate

radioisotopically and found that normal plasma oxalate

ranges from 0.75 to 1.40 Amol/l (6.75 Mg/100 ml to 102.6


The radio isotope methods were carried out by giving

a small dose of [14C] oxalate to each volunteer. The

[14C] oxalate radioactivity was counted in plasma and

urine and total urinary oxalate excretion was measured

either colorimetrically or enzymatically. To calculate

plasma oxalate, it was assumed that the specific

radioactivities of oxalate in plasma and urine were

constant and equal. That is,

plasma radioactivity urine radioactivity
= (Eq. 1.4)
plasma oxalate urine oxalate

Hence plasma oxalate could be calculated. This assumption

made in measuring plasma oxalate radioisotopically could

not be tested. At present, the tenfold difference between

plasma oxalate concentrations obtained by radioisotope

methods and those obtained by chemical or enzymatic

methods or other methods remains as an unsolved problem.

It is believed that the real plasma oxalate value lies in

between those obtained by the two methods.

The Real and Apparent Plasma Oxalate

Akay and Rose (1979) did an enzymatic assay on human

plasma oxalate concentration and found that the apparent

concentration of 1.1-16 Amol/l was too high. They

speculated that this is due to the possible conversion of

glyoxylate to oxalate. Glyoxylate is oxidized to oxalate

by three enzymes, namely xanthine oxidase, lactic

dehydrogenase and glycollate oxidase. Akay and Rose

decided to add inhibitors of those enzymes to plasma.

After the addition of enzyme inhibitors allopurinoll,

borate and phenyllactate) apparent plasma oxalate fell

considerably. It was then concluded that the normal blood

spontaneously generates oxalate on standing and the higher

values obtained by other in vitro methods are fallacious.


The stability of a drug is always of concern to

everyone connected with pharmaceutics. The drug that is

on the market must be sufficiently stable that it can be

stored for a reasonable length of time without changing to

an inactive or toxic form. Generally speaking, light,

temperature, and ionic strength of a solution and solvent

will affect the stability of a pharmaceutical product. In

the past, pharmaceutical companies evaluated the stability

of pharmaceutical preparations by observing them for a

year or more corresponding to the normal time that they

would remain in stock and in use.

Accelerated Stability Study

The method of accelerated testing of pharmaceutical

products based on the principles of chemical kinetics was

first demonstrated by Garrett and Carper (1955). In their

study, Garrett and Carper used the Arrhenius equation to

quantitatively measure the change of rate constant with

temperature. The Arrhenius equation can be described as


k = A-e-Ea/RT (Eq. 1.5)

Ea 1
or log k = log A --- --- (Eq. 1.6)
2.303 RT

in which k is the specific reaction rate, A is a constant

known as the frequency factor, E is the energy of

activation, R is the gas constant 1.987 calories/deg mole,

and T is the absolute temperature.

The constants A and Ea may be evaluated by determine

k at several temperatures and plotting 1/T against log k,

and the slope of the plot is -Ea/2.303R and the intercept

on the vertical axis is log A. So the reaction rate

constant at lower (room) temperature can be obtained by

measuring the reaction rate constant at elevated

temperature and once the activation energy (Ea) and

frequency factor (A) are determined, the rate constant at

lower temperature can be calculated.

The advantage of the elevated temperature

decomposition method is that the experiment time can be

significantly reduced (i.e. from 15 months to 1 month).

Thus the cost for this study is considerably reduced.

Effect of PH on Stability

The rates of reaction in aqueous solutions are often

markedly dependent on the solution pH, usually as a

consequence of catalytic processes. The study of the pH

dependence of reaction rates can give insight into the

mechanism of the catalysis, and yields very practical

information about drug stability at various pH.

The data in the pH effect study consist of rate

constant as a function of pH, with other factors

(temperature, ionic strength, solvent composition) held

constant throughout the series of measurements. Sometimes

it may be necessary to take into account the possible

catalytic effects by the buffer components.

The data presented in plots either of k against pH or

of log k against pH are called pH-rate profiles. For

example, let k be a pseudo first order rate constant

studied as a function of pH. For a system composed of a

nonionizable reactant (substrate S) that is subject to

uncatalyzed, specific acid catalyzed, and specific base

catalyzed reactions, a general hypothetical rate equation

can be expressed:

rate = ki[S][H+]n + k2[S] + k3[S][OH-]m (Eq. 1.7)

The experimental rate equation is

rate = k[S] (Eq. 1.8)

thus k = kl[H+]n + k2 + k3[OH-]m (Eq. 1.9)

or k = kl[H+]n + k2 + k3km/[H+]m (Eq. 1.10)

where kw is the dissociation rate constant of water. This

is equal to 1.00x10-14 at room temperature. The rate

constants kl, k2, k3 are respectively the specific acid

catalyzed, specific acid uncatalyzed, and specific base

rate constants. When the pH is very low, the above

equation can be simplified to

k = kl[H+]n. (Eq. 1.11)

Taking the logarithm of eq. 1.11

logk = logkl pH (Eq. 1.12)

That is, a plot of log k against pH at low pH should have

a slope of -n; in this way the order with respect to

hydrogen ion concentration is determined. Similar results

can be applied to the high pH region. The slope of the

log k pH profile at high pH region will be +m.

Once m and n are determined, k1 can be found from the

data at low pH by applying equation (1.10); k3 can be

found at high pH region. Finally k2 can be found from

data in the intermediate pH range by using the known

values of m, n, k1 and k3.

The pH of maximum stability can be found by

differentiating the rate equation and setting the

derivative equal to zero. This is another advantage of

determine the rate equation. For weak electrolyte drugs,

the pH profile would be more complex. The pH rate profile

depends on the ionization of reactant and the reactivity

of the ionized species as well as the unionized species.

Very complex equations will be needed to describe the



The overall objective of this study was to determine

the pharmacokinetic properties of DCA with special

reference to (1) changes in half-life and volume of

distribution observed in patients and (2) bioavailability

of oral doses.

The stability of IV dosage forms of DCA was studied.

The extent of protein binding of DCA to plasma protein was

studied using equilibrium dialysis. The time to dialysis

equilibrium, concentration dependency and the effects of

temperature on protein binding were also investigated.

The partitioning of DCA between isotonic buffer and

erythrocytes as well as between plasma and erythrocytes

was studied to determine how much drug existed in the free

form and was available for pharmacologic activities.

The pharmacokinetics of DCA were studied in healthy

human volunteers (1) to determine whether changes in half

lives and volume of distribution (Vd) occurred after

repetitive administration of the drug in healthy

volunteers as in patients, (2) to determine the

bioavailability of DCA, (3) to determine the effect of

thiamine on DCA kinetics, (4) to determine how much

oxalate is formed from DCA in the body, (5) to determine

whether there is a sex difference, (6) to determine the



dose dependency of kinetics of DCA, and (7) to evaluate

the sedation potential.

The strategy of the study was (1) to administer the

drug intravenously into healthy volunteers at 10, 25 and

50 mg/kg doses, (2) to design a crossover study on

volunteers receiving intravenous DCA, oral DCA and oral

DCA plus thiamine alternatively, (3) to administer a

single intravenous or oral DCA dosage on each volunteer

(male or female), and (4) to determine the time needed for

each volunteer to recover from the previous treatment by

giving a second dose at various intervals up to eight

weeks after the first dose and measuring plasma half




Dichloroacetic acid was as the sodium salt purchased

from Tokyo Kasei Kogyo Company, Japan. Boric Acid,

trichloroacetic acid, methyl dichloroacetate, 10% boron

trifluoride-methanol, sodium glyoxylate, glycolic acid,

allopurinol DL-f-phenyllactate were obtained from Sigma

Chemical Company, St. Louis, Missouri. Sodium acetate,

benzene, potassium monobasic phosphate, sodium

bicarbonate, sodium chloride, sodium phosphate dibasic and

sodium carbonate were all reagent grade obtained from

Mallinckrodt Chemical Inc. St. Louis, Missouri.

Analytical Methods

Analysis of DCA Powder

The DCA sample purchased was analyzed by NMR

spectroscopoy, IR spectroscopy, and elemental analysis.

The sample was dissolved in CD3OD and run on an Varian

TM3-60 NMR. IR spectrum was run using KBr disc method.

Elemental analysis was performed by sending the sample to

Atlantic Microlab, Inc., Atlanta, Georgia.

Preparation of DCA Injection Solution

Two and one half grams of DCA were dissolved in 19 ml

istonic saline and 6 ml sodium phosphate injection

solution (276mg of monobasic sodium phosphate per

mililiter of H20) to give a final concentration of 100

mg/ml. Early batches were sterilized by autoclaving at

121' for 30 minutes after passage through an 0.22 Posidyne

Nylon 66 microfilter (Pall) (Baumgartner, et al. 1986).

However, when the stability of DCA was tested at high

temperature, it was found that DCA has a decomposition

half life of 41 hrs at 100" and pH 7.4. Later sample

preparation was made by first weighing DCA powder

aseptically. The DCA was then aseptically transferred

with on-line 0.22 p filtration and then added to a

pyrogen-free container. Thioglycolate and Blood Agar

tests were performed to test for bacteria and fungi. A

lymulus lysate test was also performed for pyrogens.

Samples of the injection solution were assayed by gas


Preparation of DCA Oral Dosage Form

DCA samples (500 mg, 250 mg, 100 mg, 50mg) were

weighed and put into gelation capsules to ensure accurate

DCA dosage.

Preparation of Enzyme Inhibitor

Following the method of Ackay and Rose (1980), the

enzyme inhibitor solution was prepared and added to

Vacutainer tubes prior to withdrawing blood, to prevent

conversion of glyoxylate to oxalate. Glyoxylate is

oxidized to oxalate by three enzymes, namely xanthine

oxidase, lactic dehydrogenase, and glycolate oxidase. The

inhibitors for those enzymes are boric acid, allopurinol,

and phenyllactic acid respectively. The inhibitor

solution contained 2.22g boric acid, 120 mg allopurinol,

and 2.25 g phenyllactic acid in 100 ml of H20. The final

pH of the solution was adjusted to 7.4 with 1N NaOH

solution; the volume added to each Vacutainer tube was 0.4

ml (10% of the blood to be added into the tube); this

necessitated corrections to assay values.

Assay of DCA

A sample (0.1 ml) of plasma (unknown or standard) was

placed in a Hypo-vial sealablee thick glass vial). If the

plasma concentration was too high, 0.1 ml of diluted

plasma was used. Then 50 Al of a 100 ig/ml TCA solution

and 1 ml of 14% boron trifluoride in methanol were added.

The vial was closed with a rubber septum and then the

septum was covered with an aluminum cap and agitated

briefly and heated in a boiling water bath for ten

minutes. The vial was then cooled to room temperature.

One milliliter of water and two milliliters of benzene

were then added to the vial through the rubber septum.

The vial was then shaken on a vibrator for ten minutes and

opened. The contents of the vial were then poured into a

test tube and centrifuged at 2000 rpm for two minutes.

Samples (2 gl) of the supernatant (benzene layer) were

then injected into the gas chromatography. The Chromosorb

101 column was 6ft x 2mm i.d. at a temperature of 170.

The carrier gas was nitrogen at a flow rate of 60 ml/min.

The Ni-63 electron capture detector was set at 200.

Studies of the Effect of DCA Metabolites
(Oxalate. Glvoxylate. Glycolic Acid) in Relation
to DCA Assay

Solutions were prepared containing 100 Mg/ml DCA,

with 1000, 100, and 10 Ag/ml of oxalate, glyoxylate and

glycolic acid individually. Each sample was assayed in

duplicate by gas chromatography following the procedure

described in the DCA assay.

Study of DCA Binding Using a Dianorm Equilibrium

The Dianorm equilibrium dialyser consists of a set of

four racks of cell holders, each containing five dialysis

cells located between spring-loaded metal spacers to avoid

leakage. The cells are made of Teflon to minimize

absorption of the drug into the material of the dialyser.

The cells are divided into two halves (half cells A and B,

respectively) which are separated by a semi-permeable

membrane. The solutions under study are injected into a

half cell. The dialysis membranes used were previously

soaked in deionized water for 30 minutes. Each half cell

was filled with 1.5 ml of plasma or pH 7.4 isotonic buffer

solution. For consistency and convenience, plasma was

always put into the thick half cell and buffer into the

thin half cell. The racks of the cells were rotated at 4

revolutions per minute.

To study the time to equilibrium, 100 Ag/ml DCA was

put on the buffer side of the cell and the cell was

rotated. Samples were withdrawn after 1, 2, 4, 8 and 16

hours of dialysis. Both buffer and plasma were assayed

for DCA content. To study whether DCA equilibrium was

affected by adding it to either the buffer side or the

plasma side, a solution, (100 ig/ml) of DCA was also put

into the plasma side initially and dialysed for 1, 2, 4, 8

and 16 hours. After dialysis, the DCA concentrations on

both sides were measured.

To study the effect of temperature on binding,

dialysis was performed at 37 and 25 in a water bath. A

solution (100 Ag/ml) of DCA was added to the buffer side

and dialysed for one hour. The DCA concentrations on both

sides were measured. To study the effect of DCA

concentration on protein binding, DCA concentrations of

100, 150, 200, 250 and 300 Mg/ml were dialysed at 37" for

one hour. The DCA on both sides were assayed.

Red Blood Cell Localization Study and Partition

Eighteen capped test tubes, each with 10 ml of whole

blood inside containing DCA at 100 gg/ml, were put on a

tube rotator. Samples were withdrawn at 5, 10, 30, 60,

90, and 120 minutes to study the time to reach

equilibrium. Three tubes were used at each time point.

Centrifugation was at 2000 r.p.m. for five minutes after

rotation. Plasma was separated from the red blood cells

and the plasma content of DCA was measured.

To study the effect of DCA concentration on red blood

cell localization, whole blood concentrations of 50, 100,

300, 600, and 1000 ug/ml DCA were prepared. These samples

were rotated for two hours to make sure equilibrium had

been reached. After centrifugation at 2000 r.p.m. for 5

minutes, and separation of plasma from the red blood

cells, the plasma concentration of DCA was assayed. The

partition coefficients were determined in the following

way. Fifteen capped test tubes, each with 5 ml whole

blood inside, were taken and washed with pH 7.4 isotonic

buffer solution three times in order to remove all the

plasma protein. The washed blood cells were reconstituted

by adding pH 7.4 isotonic buffer solution to give the

original blood volume. This formed "pseudoblood." DCA

was added to the pseudoblood, resulting, in triplicate, in

whole "blood" concentrations of 50, 100, 300, 600, and

1000 Ag/ml. After centrifuging at 2000 r.p.m., the DCA

content in the buffer was assayed.

Stability of DCA

The stability study was conducted at 4, 25', 100

and 121' in constant ionic (1=0.1) buffer solution. The

composition of each buffer solution is listed on Table

3.1. Four milliliters of 100 gg/ml of DCA of various pH

were placed in a hypovial and sealed with a neoprene seal

and aluminum cap. Then each vial was stored at different

temperature. The decomposition at 121"C was effected by

Table 3.1. The composition of buffer solution(I=0.1) at
various pH. Buffer was prepared by adding
composition A to composition B and addig
distilled water to a final volume of 1 liter.

pH Composition A Composition B

1.1 0.1N HC1 500ml 0.1N HC1 500ml
1.4 0.2M HC1 257ml 0.2M KC1 243ml
1.8 0.2M HC1 102ml 0.2M KC1 398ml
3.6 0.2M HAC 463ml 0.2M NaOAC 37ml
5.0 IN HAC 100ml IN NaOAC 44.7ml
7.4 IN NaH2PO4 8.95ml IN Na2HPO4 30.4ml
10.0 IN NaHCO3 24.98ml IN Na2CO3 25.01ml
12.9 0.1N NaOH 500ml 0.1N NaOH 500ml

autoclaving in the sterilization department in the Shands

hospital of the University of Florida. Sampling was done

at appropriate intervals so that it was possible to

determine the degradation rate constant at various


Assay of Oxalic Acid

Urine Oxalic Acid by Atomic Absoption

Twenty four hour urine was collected and adjusted with

6N HC1 to a pH value less than 3, and mixed throughly. A

60 ml aliquot was submitted to the Clinical Chemistry

Laboratory in the Shands Hospital of the J. Hillis Miller

Heath Center, University of Florida. Analysis was done by

atomic absorption spectrophotometry (Henry, et al. 1974).

Urinary Oxalic Acid by Gas Chromatography

Aliquots (100 Ml) of urine (when the concentration

was low, 200 Al was used) or working standard solutions of

oxalic acid (5 Ag to 30 Ag per ml) were pipetted into

capped test tube. Aliquots (50 Al) of a deionized water

solution of malonic Acid (1.23 g / 100 ml), the internal

standard, was added to each tube. Each sample was dried

under a stream of dry N2 at room temperature. A sample (1

ml) of 10% BCl3 in 2-chloroethanol was then added to each

tube. The tubes were loosely capped, vortexed for 20 sec

and heated in an oven at 70' for 30 min. After the tubes

were cooled to room temperature, 2 ml of deionized water

were added to each tube followed by the addition of 4 ml

of ethyl acetate / isopropyl ether (1:3, v/v). The tubes

was capped tightly, shaken vigorously by hand and

centrifuged in a centrifuge for 3 min. The organic phase

was then taken out. A 1 Ml sample of the organic phase was

injected into a Varian 3700 Gas chromatograph with Ni-63

electron capture detector. The gas chromatographic

conditions were as follows:

Packing : 4% SE30-6% OV210, 80/100 mesh, on Gas Chrom.Q.

Temperature: inlet 180', column 160', detector 200'

Carrier Gas: N2 at 50 ml/min.

Assay of Plasma Oxalate

In order to find out an efficient way to precipitate

plasma protein in an oxalate sample, radio labelled [14C]

oxalate was used. A sample containing 4.44 x 104 d.p.m.

of radioactive oxalic acid in ethanol plus 20 gl of 1

mg/ml oxalic acid was pipetted into a capped test tube and

one mililiter of plasma was added. Deproteinization was

made with

(1) 0.1 ml trichloroacetic acid (70% W/V)

(2) 3 ml acetonitrile

(3) 0.1 ml of 70% W/V perchloric acid.

The precipitated sample was than centrifuged and the

supernatant and precipitate were separated and transferred

to liquid scintillation vials to measure radioactivity.

The radioactivity was measured on a Packard Tri-Carb 460

CD Liquid Scintillation System after the addition of 10 ml

of aquasil cocktail. A quench curve was made by using

[14C] toluene as standard. The result of precipitation

was tabulated on Table 3.2.

Table 3.2. Total radioactivity of the supernatant and the
precipitant after mixing 1 ml of plasma with
various deproteination reagent.

0.lml TCA 3ml Aceto- 0.lml Perchloric
(70% W/V) nitrile Acid (70% W/V)
natant 2.99 x 104 4.37 x 104 3.46 x 104
(d.p.m) + 738 + 5990 + 4290

Preci- 1.49 x 104 3580 1.27 x 104
pitant + 1470 + 1330 + 1090

Because acetonitrile gave the most recovery (93%) of the

total radioactivity, later experiments were performed by

using acetonitrile as a deproteinization regeant.

Oxalate in Plasma

One ml of plasma sample (or standard) was pipetted

into a test tube and 150 fl of 1.23 g/100 ml of malonic

acid were added as internal standard. Acetonitrile (3 ml)

was then added. The tube was vortexed for 30 sec to

ensure maximum mixing. The tubes were then centrifuged at

2000 r.p.m. for 10 min. The supernatant was then

transferred to another test tube and evaporated under a

stream of nitrogen at 25' until dry. The dried sample was

then reacted with 1.5 ml boron trichloride in 2-

chloroethanol solution. The assay then followed the same

procedure as mentioned on the urine sample treatment.

Identifications of Oxalic Acid Derivatives Usinq GC-MS

The derivatives of oxalic acid formed either in urine

or in plasma was identified by GC-MS verification. The

samples were assayed at the Pesticide Research Lab of the

department of Food Science and Human Nutrition in the

University of Florida. A Finnigan 4021 gas chromatograph-

mass spectrometer was used. Electron-impact ionization

(EI) and chemical ionization (CI) (positive and negative

modes) were examined.

Table 3.3. Demographic data of volunteers in the five
dose DCA study.

Volun- Sex Age Wt Hemato- Dose
teer (kg) crit(%) (mg/kg)

1 M 25 74.8 44.4 10
2 M 22 76.0 49.0 10
3a M 42 72.4 39.2 10
4 F 23 61.4 41.3 15
5 M 24 80.6 45.5 25
6 M 24 67.8 42.6 25
7 M 33 73.3 42.8 25
8 F 37 56.8 39.9 25
3b M 42 72.4 39.2 25
9 M 57 101.4 47.5 50
10 F 45 57.6 41.5 50
11 F 27 65.0 44.1 50
12 F 27 58.0 40.9 50

Pharmacokinetic Studies of DCA

Selection of Subjects

Healthy volunteers conforming to the desired height-

weight range were chosen. Persons having a history of

hypersensitivity or adverse reaction to any drug, chronic

abuse of alcohol or drugs, any disease diagnosed or under

investigation, and receiving any medication, were

excluded. Volunteers were evaluated by physical

examination, medical history, complete blood count and

routine chemical analysis of renal, hepatic and

coagulation function. All the examinations and

evaluations were done at the beginning and at the end of

the study. After the nature of the study was explained to

the volunteers, they gave their informed consent to

participate. The study was approved by the local institu-

tional review board (Appendix). An Initial New Drug

certificate for the study of DCA had been issued by the

FDA to Dr. P.W. Stacpoole, the physician in charge of the


Five-Dose DCA Study

Seven male and five female healthy volunteers

participated (Table 3.3). One male was studied twice,

each time at a different drug dose. Subjects were

admitted as inpatients on the morning of study to the

Clinical Research Center of Shands Hospital, University of

Florida. The subjects were fasted overnight. A small

diameter catheter was inserted into a large superficial

vein in each forearm. A heparin lock was attached to one

catheter for blood withdrawal, while the other catheter

was kept open with a slow saline infusion between DCA


After the catheter placement, the first DCA infusion

was begun. Dichloroacetate in phosphate-buffered saline

(pH 5.2) was diluted to a total volume of 50 ml and

administered over 30 min using a Harvard infusion pump.

On each occasion, the subject received five identical drug

doses, spaced two hours apart. Repeated doses of 10 mg/kg

DCA were administered in the first three investigations.

One subject received 15 mg/kg doses. The last four

investigations were at the 50 mg/kg dose level. Blood

samples were obtained at 30-minute intervals for the first

8.5-9 hours, and subsequently at 3-4 hours up to 24 hours

after starting the first infusion. Two ml of blood were

collected in heparin-containing Vacutainer tubes for DCA

assay. The blood was stored in an ice bucket until

centrifuged. After centrifugation, the supernatant was

frozen at -20" until assayed. Immediately before the

first drug infusion subjects voided and then commenced a

24 hr urine collection for oxalate determination by atomic

absorption. The 24 hour urine was also analyzed for its

creatinine content by the clinical chemistry laboratory in

Shands Hospital. Creatinine concentration was determined

by the red color development on reacting with picric acid

(Israel and Henry, 1974).

Crossover Study

Once the volunteers were chosen, they were given a

controlled diet prepared by a dietician to control oxalate

intake from food one week before drug treatment. The

nutritive components are listed in Table 3.4. The

percentages by calorie of protein, carbohydrate and fat

are 13%, 56%, and 31% respectively. A typical one-day

food intake is shown in Table 3.5. A twenty-four-hour

urine sample was collected one day before the study to

measure urinary oxalate excretion without treatment.

After an overnight fast for 10 hours, the drug study

began. Each subject received three treatments, namely (A)

50mg/kg intraveneous DCA, (B) 50mg/kg oral DCA, and (C)

50mg/kg oral DCA plus 50mg thiamine. The orders of

treatment for 6 subjects are listed in Table (3.6).

Treatments were 4 days apart. Blood was withdrawn from a

forearm vein by inserting a plastic intravenous butterfly

cannula (heparin lock). Heparinized (100 unit/ml) normal

saline solution was used to prevent the blood from

clotting in the cannula. Care was taken to ensure that

blood samples were not diluted by the heparinized saline

and to prevent heparin from being injected into the blood

stream. Samples (8 ml) of blood were withdrawn at

predetermined times up to twelve hours after drug

administration (Table 3.7). Two heparin-containing green

top Vacutainer tubes were used. Each tube contained 0.4

ml of inhibitor plus 4 ml of whole blood. The blood was

immediately shaken gently after withdrawal and centrifuged

at 2000 r.p.m. for seven minutes to separate plasma and

red blood cells. The plasma sample was then stored at

-20 until assayed.

Table 3.4. Typical nutritive components of a low oxalate
diet for a healthy subject. The nutritive
components varies slightly for different
subject because body weight and metabolic
rate varies for each individual.

1 Calories 1969.3c 11 Oxalic 49.9mg
2 Protein 65.4g 12 Carbohydrate 274.9g
3 Calcium 357mg 13 Potassium 2456.4mg
4 Phosphorus 829.4mg 14 Sodium 2605.8mg
5 Iron 13.9mg 15 Fat 68.7mg
6 Vitamin A 7826U 16 Mg 0.2g
7 Riboflavin 0.74mg 17 C1 2.9g
8 Thiamin 0.74mg 18 Sulf 0.5g
9 Niacin 11.6mg 19 Weight 1922g
10 Vitamin C 53mg 20 Water 897g

Table 3.5. Typical diet of a healthy volunteer
in one day to maintain a low oxalate intake.

Food Name Weight (Grams)

Whole Eggs 50
Whole Wheat 115
Grape Jelly 20
Regular Mayonnaise 10
Celery Sticks 30
Carrot Sticks 30
CRC Peaches 50
7 UP 720
Top Round Ground Beef 100
Dry Mashed Potatoes 20
CRC Green Beens 100
Vegetable Salad Oil 7
Cider Vinegar 7
Iceberg Lettuce 80
Fresh Tomato 80
CRC Pineapple 50
CRC Canned Apple Juice 177
Table Salt 3
Regular Margarine 40
Kraft Party Mints 28
CRC Asparagus 100
Ross' Polycose Powder 30
Brach's Sour Balls 14
Sanka 6
Cooked Baked Ham 55

Table 3.6. Crossover study dosing schedule.

Day 1 Day 5 Day 9
Number Regimens

1 A B C
2 B A C
3 B C A
4 C A B
5 A C B
6 C B A

Urine was collected at various intervals up to 48

hoursafter drug administration (Table 3.7). Urinary pH

and volume were measured immediately after urine was

collected. Urine samples were frozen at -20 until

assayed. Vital signs (blood pressure, pulse rate), and

also a digit symbol substitution test (Fig. 3.2)(Curry and

Whelpton, 1983), were recorded before drug administration

and at various times after dosing (Table 3.7). The digit

symbol substitution test examined the alertness of each

subject. The subject was required to identify each number

in sequence (Fig 3.2), note the corresponding symbol from

the code, and enter the symbol in the appropriate space.

The score was the number of correct substitutions in 60












0 *r



Table 3.7 Procedures used in the crossover study
and bioavailability study.

Procedure Time(hr)

Venous 0,0.125,0.25,0.375,0.5,0.625,0.75,
Blood 0.875,1,1.25,1.5,1.75,2,2.5,3,3.5,
Sample 4,4.5,5,5.5,6,7,8,9,10,11,12.

Urine -24-0,0-1,1-2,2-3,3-4,4-5,5-6,6-8,
Sample 8-10,10-12,12-24,24-36,36-48.

Vital 0,1,2,3,4,5,6,7,8,9,10,11,12.

Digital 0,0,1,2,5,12.

Table 3.8 Demographic data of volunteers
participating in the
bioavailability study.

Subject Age Wt. Sex

1 31 61.6 M
2 38 81.6 M
3 26 74.5 M
4 40 67.5 M
5 38 67.9 M
6 21 82.0 M
7 26 78.2 M
8 33 79.6 M
9 40 49.9 F
10 23 51.2 F
11 31 65.8 F
12 26 49.8 F

Bioavailabilitv Study

Eight healthy male and four healthy female volunteers

were chosen to be involved in this study (Table 3.8). The

criteria for choosing the subjects were the same as in the

cross over study. The oxalate intake of each volunteer

was also controlled for one week before drug treatment.

Twenty four hour urine samples were collected prior to the

study to measure basic oxalate excretion. After an

overnight fast for 10 hours, the drug study began. Four

male subjects received 50 mg/kg intravenous doses of DCA.

The other subjects, both male and female, received

50mg/kg oral doses of DCA. Blood was withdrawn from a

forearm vein by inserting a plastic intravenous butterfly

cannula. Eight ml of blood were withdrawn at each

predetermined time up to twelve hours after drug

administration (Table 3.7). Two heparin-containing green

top Vacutainer tubes were used to preserve the blood.

Each tube contained 0.4 ml of inhibitor plus 4 ml of whole

blood. The blood was shaken gently immediately after

blood withdrawal and centrifuged at 2000 r.p.m. for seven

minutes to separate plasma and red blood cells. The

plasma sample was stored at -20" until assayed.

Urine was collected at various intervals up to 48

hours after drug administration (Table 3.7). Urinary pH

and volume were measured immediately after urine

collection. Urine samples were frozen at -20" until


Vital signs(blood pressure, pulse rate) and

performance on the digit symbol substitution test were

recorded before drug administration and at various time

after dosing (Table 3.7).

Multiple Dose DCA Study

The purpose of this study was to find out the minimum

amount of time needed for healthy subjects to "recover"

from previous DCA treatment. It had been found that when

a healthy subject was given a second DCA dose two hours to

three days after an earlier treatment, the rate of

elimination of the second treatment was significantly

slower than that of the first one. The elimination half-

life increased. This might be due to a metabolite of DCA

inhibiting the metabolism of DCA (product inhibition) or

to some other causes. It was necessary to determine the

time needed for a subject to revert to the rate of

metabolism following the initial treatment.

Twelve healthy subjects participated in this study.

Subjects were divided into three groups, each group

consisted of 4 subjects (Table 3.9). Each volunteer

received at least two treatments. The second treatment

was three to eight weeks after the first one (Table 3.9).

Each volunteer fasted for at least 8 hours before the

study. The next morning, a 50 mg/kg of oral DCA was given

to each subject. The subjects were not allowed to eat

until two hours after the drug was taken. Two ml of blood

Table 3.9. Demographic data of volunteers receiving
multiple DCA treatment.

Subject Age Weight(kg) Sex 2nd Treatment
(Weeks Later)

1 27 67.1 F 3
2 30 62.2 F 3
3 31 70.3 M 3
4 32 65.5 M 3
5 24 52.7 F 4
6 35 53.5 F 4
7 28 65.0 M 4
8 38 65.8 M 4
9 28 81.6 M 8
10 25 52.2 F 8
11 28 45.4 F 8
12 27 45.4 F 8

were withdrawn at 2, 3, 4, and 5 hours for the first

study. Blood was withdrawn at 2, 3, 4, 6, and 7 hours for

the second study. The blood sample was centrifuged

immediately and frozen at -20 until assayed. A third

dose six weeks after the second one was given to subjects

10 and 12. The fourth treatment was given to subject 10

two weeks after the previous one. A third dose was given

to subjects 3 and 7, and two weeks after that a fourth

dose was given. The 95% confidence interval of the

elimination half life of each subject was calculated

according to Huston (1958). Statistical comparison of the

half life was evaluated using the calculated confidence

interval. If the second half life falls outside the

confidence interval of the first one, it is considered

that the half life changed significantly.

Computer Modeling

The pharmacokinetic parameters for plasma

concentrations-time profiles were first estimated using R

strip program (MacroMath, Inc.) on an IBM XT personal

computer. The goodness of fit is mainly determined by the

Model Selection Criterian (MSC). The MSC places a burden

on the model with more parameters to not only have a

better coefficient of determination, but quantifies how

much better it must be for the model to be more

appropriate. The initial estimations of those parameters

were then fitted in the PCNONLIN program (Statistical

Consultants, Inc., Lexington, Kentucky) to obtain the best

estimations of parameters. The best fitted concentration-

time data points were then calulated from those best

fitted parameters using Lotus 1-2-3 program (Weber System

Inc.). The data file was then translated into Sigma-Plot

program (Graphic Software Systems, Inc.) data file and

drawn on a Hewlett Packard 7470A plotter.


Identification of DCA

The DCA sample was first shown to be of required

identity and purity by NMR spectroscopy. The sample was

dissolved in CD30D and the zero reference was

tetramethylsilane(TMS). The spectrum showed the CH bond

of the CHCl2 grouping (Fig. 4.1). Similarly, the IR

spectrum showed bands consistent with C=O, COOH and C-Cl

bonds (Fig. 4.2). Elemental analysis gave a content which

agreed with the putative formula (Table 4.1).

Analysis of DCA in Plasma and Urine

A gas chromatography trace for DCA is shown in Fig.

4.3. The first peak is the methylester of DCA obtained by

reaction of DCA with BF3/methanol. The identity of this

peak was proved by comparison with an injection of an

authentic sample of the ester. The second peak was the

methyl ester of TCA. Injection solutions typically

assayed at 99.5% of the stated concentration using the

specific gas-chromatographic method and both the DCA

sample and the authentic ester sample as standards. A

typical calibration curve of peak height ratio versus





*-4 0

0 3

o a

0 *


O 01

fr1 *r1 01


-.0 0



S Iu

0 0
'--- -




4 1

0 0


1 o:

So *Qo
u LI e .

-. .,. 1
o3 It

Eu *'-1 I





\4 u

.1 ..
\ oE

o .0
\ ,.

4 0 .1


(00 0 4


\ 0 66 6

\OfCco CD 41 4
\ c51 a

\1101 ;tbaH Z 1

Table 4.1. Identity of sodium dichloroacetate by
elemental analysis.

Element Found Theory

C 15.71 15.8
H 1.39 1.3
0 21.1
C1 46.35 46.7
Na 15.1

Table 4.2 The within-day and between day coefficients of
variation (c.v.) of the gas chromatographic
estimation of DCA in water and in plasma.

Concentration number of samples c.v.%

5 6 6.5
25 6 4.0
50 6 3.3
100 6 3.2
200 6 3.2

DCA concentration in plasma as well as in urine is

presented in Fig. 4.4. The square of the correlation

coefficient for the relationship was 0.992. The limits of

quantification were 5 Ag/ml and 200 Ag/ml. The within

day coefficients of variation were between 6.5% (5 ig/ml)

and 3.2% (200 Ag/ml) (Table 4.2). The between day

coefficients of variation were the same as within day

coefficient of variation.

It has been known for some years that DCA is

metabolized to glyoxylic acid which is then converted to

oxalic acid. There was no evidence in the literature to

prove that these putative DCA metabolites did not

interfere with the DCA assay. By adding various amounts

(10, 100, 1000 Ig/ml) of oxalic, glyoxylic and glycolic

acid separately to a 100 ig/ml DCA sample, it was shown

that there was no relationship between contaminant and

found DCA concentration (Table 4.3). Besides, there were

no interfering gas chromatography peaks.

Table 4.3. Determination of the interference of DCA
metabolite in DCA assay. 100 ig/ml DCA was
initially added to three different
concentrations of DCA metabolite.
Results were expressed as gg/ml of DCA.

DCA Plus: 1000 ug/ml 100 ag/ml 10 Ig/ml
metabolite metabolite metabolite

Oxalate 103.6 99.9 102.3
104.8 99.9 102.3

Glycolate 97.5 105.9 101.1
101.1 97.5 98.9

Glyoxylate 110.8 103.6 104.8
97.5 107.2 105.9
no significant difference between columns and rows.

Analysis of Oxalate in Plasma and Urine

Structural Verification of Chloroethvlated Esters of
Oxalic Acid and Malonic Acid Formed in Derivatized
Standards and Plasma by GC-MS

The major ions detected after El mass spectrometry of

the derivatized oxalic acid were m/e 63 (100%) (Figure

4.5), representing CH2CH235Cl+ and m/e 65 (30%),

representing CH2CH237C1+. The ions detected after CI

(positive mode) mass spectrometry were m/e 215 (50%), m/e

179 (33%), and m/e 107 (100%) (Figure 4.6). The ions

presented after CI (negative mode) mass spectrometry were

151 (100%) and m/e 153 (30%) (Figure 4.7). For the malonic

acid derivative, the ions found after El were m/e 149

(100%), m/e 63 (95%), and m/e 65 (29%) (Figure 4.8). The

ions found after CI (positive mode) of the malonic

derivative were m/e 149 (100%), m/e 193 (40%), and m/e 229

(38%)(Figure 4.9). The derivatized plasma sample

corresponded in retention time to standard preparations of

bis-2-chloroethyl esters of oxalic and malonic acids and

their identities were confirmed by CI (negative mode)

spectra (Figure 4.10).

Analysis of Oxalate in Plasma

Fig. 4.11 shows a typical gas chromatogram of oxalate

in plasma, with its internal standard. Fig. 4.12 shows a

typical calibration curve of added oxalate in plasma. The

calibration curve can be described by the equation : y =

0.573 + 0.237x, where y is the peak height



U >
U -

; 0




?-- ^ Hp
I ----- .-----------__________ 3 W
ol o 1

S u








U .



o -o





r-- C


15 "it




6 8:r



$Z i
~:; o
i~ ~io


V -a

U2 (a

a >


10 r-o
F Fc


Hi 4

= 0
r 0),4
'" = hlJ







.a 4

r- 0 U

'C )

s. -

SC, 0

<= U
s =





,, u




". "




0 a
>1 C

U) -4


0 '

. 1-C

- U0

I *

0 >
4J -4

- U

UG )

"a>S ^ 4 ^
a-'S2? 10
i-S-~a 1
^Oo : (D
S r '' "i ^ *

5^ =: G|
S2^2-r 2^ o
&i~i~a- '^

aieuo I- ?



0 -

0 -


0 'D

oa r,
rn-e g

0 C
0 0
00 to
c d

O!Ddy 4q6!aH *>ald

4 )
a4 .)
10 in,
,- 4 .

o a-r
4J *

U) cm
m r0 (

>, 'o a
a > o
(0 -i-I
C Ca C


.CM 0*
a r- C 'aC

(a)o a)
i N 4J (a w
C0 a)
o + om
*H C C
4Jm m CU
a r a x --
)- Le) 4J 0 U
.0 c *-
-H C -H 44 w
1 0 Q)
i i a o
>i ro o
N- (0 4 c
a) 40

Cn .-4 cC
-M C O
*- a)C (U 0
a 0 0
a U U




T- E








oxalate/internal standard ratio, and x is the

concentration of oxalate added to the plasma. The reason

why the intercept is different from zero is that oxalate

exists endogenously in plasma. Table 4.4 shows that the

within day coefficient of variation was between 15.8% and

27.7%. The average endogenous plasma oxalate

concentration measured with this method was 1.71 0.97

Ag/ml (n=10, range from 0.67 to 3.41). Between day

coefficient of variation values varied over a wide range,

so daily standardization with spiked plasma was performed

throughout the study.

Analysis of Oxlate in Urine

Fig. 4.13 shows the gas chromatogram of blank urine

and a two hours post dosage urine sample from a volunteer

receiving a 50 mg/kg intravenous infusion of DCA. The

calibration curve of oxalate in water is shown in Fig.


Table 4.4. The within-day coefficients of variation of
the gas chromatographic estimation of oxalate
added to plasma.

concentration number of samples c.v.

0 4 15.8
0.4 4 26.5
1.2 4 27.7
1.6 4 18.1







4J o
a m


0 3

x C
O *3

S4 0C
4-1 >

) 4-
4D C (

M4 01
*C 0
W 4J 4n

0: 0

( C0

.4 0 4 O

X ) 41J

O o OE 10

( 0
\N 0 a m

a a

.. o E c
\ 0 0 -I

0 > u 0

0ON 0
O 0O
0 o >
C ) Of

X (0 fn .-. 4J

z C 0 -X

II 0

)I I I 0 4-0 40 a*
O Bo! H a Ed
\ *I *

011D0y 14f6l3H *>Dgd

The calibration curve is described by the equation y

= -0.043 + 0.214x, where y is the peak height ratio and x

is the concentration of oxalate in water. The correlation

coeficient was 0.9976. However, the coefficients of

variation of this assay method varied between 10.0% and

16.6% (Table 4.5). To ensure best results, daily

standardization of oxalate in water was performed

throughout the study and each sample was assayed twice.

Comparison of Urinary Oxalate by Atomic Absorption and GC

Table 4.6 compares the results of the atomic

absorption and gas chromatographic procedures. The atomic

absorption method measured the total amount of 24 hour

urinary oxalate excretion whereas the G.C. method measured

urinary oxalate at 0-1, 1-2, 2-3, 3-4, 4-6, 6-12, and 12-

24 hrs. The total amount was then summed and compared.

The results using these two methods agreed with each


Stability of DCA

When injectable DCA solutions were originally

prepared prior to this work and in the initial studies

described in this dissertation, they were made by

autoclaving at 250F for 15 to 20 minutes. When the

solutions prepared in this way were administered to

volunteers, some of the volunteers complained about pain

at the site of infusion. It was then speculated that the

solution was not stable and maybe some acid was produced

during sterilization. The pH of the sterilized solution

was then measured and it was found that the pH of the

injectable solution was low (between 2 to 4) although the

assayed DCA content of the injectable solution remained at

virtually 100% of the labelled concentration. It was

decided to study the stability of DCA solution.

The stability of DCA was studied at 4'C, 25*C, 100C,

and 121C. A constant ionic strength (I = 0.1) was

maintained in this study. The degradation of DCA at 121'C

and 100'C was significant (Table 4.7). DCA was stable at

4'C for up to 6 months except in pH 1.1 buffer solutions

(Table 4.8). At room temperature, DCA degradation was

significant only in pH 1.1 solutions (Table 4.9).

Extent of Protein Binding Determined by Using Equilibrium

Table 4.10 shows that the time to reach protein

binding equilibrium was one hour. The effect of

temperature on protein binding was also studied. There

was no significant difference between 37"C and 25*C after

dialysis for one hour (Table 4.11). The DCA was added to

both plasma and buffer sides of the membrane separately,

and there was no significant difference in the results of

the two experiments (Table 4.10 & 4.12).

These results showed that protein binding of DCA was

reversible. The protein binding of DCA was constant up to

300 ug/ml DCA (Table 4.13).

Table 4.5 The within-day coefficients of
gas chromatographic estimation

variation of the
of oxalate in

concentration number of samples c.v.%

5 4 10.0
10 4 16.6
15 4 13.7
20 4 15.7

Table 4.6 Comparison of urinary oxalic acid measured by
gas chromatographic and atomic absorption

Sample 24 hours urinary oxalate excretion
GC(;g/ml) Atomic A. (Ag/ml)

1 76.2 64
2 89.3 82
3 36.5 24
4 90.0 99
5 45.7 51
6 71.7 96
7 89.0 87
8 38.0 51

Table 4.7 Decomposition of DCA at 121*C and 100*C.
Results were expressed as percentage of initial
concentrations. The results were average of four

0.5 1 1.5 2.0 4.0 6.0 8.0 10.0

121 78.0 68.0 63.1 59.0

100 94.2 91.6 90.6 89.1 80.0

Table 4.8 Stability study of DCA in various buffer
solution at 4C for 6 months.

pH PHR of Standard PHR of Sample Statistics

5.0 0.95 + 0.06 0.88 + 0.03 N.S.
7.4 0.90 + 0.05 0.85 + 0.04 N.S.
10.0 0.88 + 0.02 0.94 + 0.06 N.S.
12.9 0.97 + 0.08 0.98 + 0.06 N.S.

n = 4, student t test
N.S.: Non-significant

Table 4.9 Stability study of DCA in various buffer
solution at 25C for various time up to
3 months.

DCA Concentration (jg/ml)

Months 1.1 5.0 7.4 10.0 12.9

0.53 16.4 85.1 91.7 96.3 91.6
1.0 4.0 83.9 81.3 84.7 82.0
2.0 + 84.1 70.0 90.9 88.4
3.0 + 89.3 61.7 86.4 86.6

Red Cell Localization and Ervthocvte Partitioning

Equilibrium distribution between red blood cells and

plasma was reached within 5 minutes (Table 4.14). Since 5

minutes was the earliest data point, the equilibrium might

well have been reached earlier.

The average partition coefficient was found to be 0.44

+ 0.03. The partition coefficient between red blood cells

and isotonic buffer was constant at 100, 200, 300, and 1000

Ag/ml of DCA (Table 4.15). The partition coefficient at 600

Ag/ml of initial whole blood concentration was slightly but

significant higher than the other concentration. This is

probably a "false positive" type of result, since it was not

observed at 1000 ig/ml. No conclusion was drawn as to any

lack of linearity of erythrocyte partitioning.

Table 4.10 Determination of time to reach dialysis
equilibrium using Dianorm Equilibrium Dialyser.
Dichloroacetate 100 Mg/ml was initiately added
to the plasma side. Concentrations of DCA on
both sides were assayed.

Time (hr)
1 2 4 8 16

Plasm Side 55.1 49.9 54.2 54.6 58.1
DCA (Mg/ml) 53.4 50.8 62.9 57.9 53.6

Buffer Side 35.8 33.4 38.4 36.0 31.6
DCA (Mg/ml) 36.2 36.2 33.7 36.2 31.6

Non-significant difference between each other at p=0.05.

Table 4.11 Studies of the effect of temperature on protein
binding. Dichloroacetate (100 gg/ml) was
initially added to buffer side. Both buffer
and plasma side were assayed after dialysis.
"f" stands for fraction of protein binding.

37'C 25'C

Plasma Buffer Plasma Buffer
Side Side Side Side
Cone. Cone. Conc. Conc.
(jg/ml) (Ag/ml) f (Ag/ml) (Mg/ml) f

Dialysis 57.1 45.4 0.20 59.5 43.9 0.26
After 57.8 42.4 0.27 58.9 45.1 0.23
One 55.8 41.4 0.26 53.3 45.5 0.15

AVE 56.9 43.1 0.24 57.2 44.8 0.21

SE 0.57 1.21 0.02 1.96 0.46 0.03

* Non-significant difference between "f" at 25C and 37C


Table 4.12 Determination of time to reach dialysis
equilibrium. Dichloroacetate (100 Ag/ml) was
initially added to the buffer side.
Concentrations of DCA on both sides were

Time (hr)
1 2 3 4 5

Plasma 58.2 59.4 52.6 52.9 53.4
47.4 52.2 46.3 55.0 47.1

Buffer 35.5 39.2 35.6 33.8 37.4
34.0 33.5 33.8 39.8 30.9

Table 4.13 Studies of protein binding of DCA at different
concentrations. Sodium dichloroacetate
was assayed on both buffer (b) and plasma
side (p). Y = 0.59x 2.82, R = 0.996,
Y stands for average plasma side concentration
after dialysis. X stands for initial DCA
concentration. Y' = 2.54 + 0.38X, R = 0.998,
Y' stands for average buffer side concentration
after dialysis. "f" is the calculated fraction
of protein binding at each concentration.

Original DCA
added to
buffer side



p b f p b f p b f

55.3 37.6 0.32 78.7 66.0 0.16 95.1 61.6 0.35
56.6 38.3 0.32 85.6 61.9 0.28 126.1 84.8 0.33
53.0 38.3 0.28 68.2 58.5 0.14 141.9 80.0 0.44
56.5 40.1 0.29 91.2 66.4 0.27 110.3 76.5 0.30
109.8 80.0 0.27

Mean 55.4 38.6 0.30 81.0 63.0 0.21 116.5 76.6 0.33

SE 0.09 0.55 0.01 4.3 1.9 0.03 8.05 3.90 0.03

Original DCA
added to
buffer side 250 300

p b f p b f

259.9 112.6 0.25 160.2 108.1 0.33
117.8 98.5 0.45 164.3 116.4 0.29
128.4 97.3 0.24 173.5 107.0 0.38
140.1 95.5 0.30 178.1 116.8 0.34

Mean 149.3 100.9 0.31 169.4 112.1 0.34

SE 10.5 3.9 0.07 4.1 2.6 0.02

Table 4.14 Determination of time to reach equilibrium
for cell localization. Dichloroacetate (100
Ag/ml) was initially added to the whole blood.
Plasma concentration was assayed. Blood
was obtained from volunteer 3a. P/C stood
for the ratio of DCA in plasma to red cell.

Time (min)

5 10 30 60 90 120

140.7 130.6 131.6 131.6 130.1 133.9

135.2 139.9 125.9 120.9 127.0 120.5

concen- 143.0 132.0 136.1 134.0 127.9 136.2
(Ag/ml) 130.8 129.1 145.3 127.9 124.2 115.3

140.7 126.4 140.7 129.4 130.3 146.6

138.8 135.3 135.9 123.4 135.5 152.7

Mean 138.2 132.2 135.9 127.9 129.2 134.2

SE 1.8 2.0 2.8 2.0 1.6 5.9

P/C 2.6 2.2 2.4 1.9 2.0 2.3

D* 0.55 0.65 0.59 0.74 0.71 0.62

*Non-significant difference between each time studied.
D* = C/P-f, where f was the fraction of free DCA in plasma,
f = 0.7 for subject 3a.

Table 4.15 Partition coefficient (D) of DCA between
red blood cells and isotonic buffer up to
1000 Ag/ml whole pseudoblood concentration.
Blood was obtained from subject 3a.

DCA concentration Mg/ml
50 100 300 600 1000

0.19 0.43 0.46 0.22 0.39

0.32 0.39 0.35 0.58 0.44

D 0.47 0.46 0.39 0.73 0.45

0.41 0.33 0.48 0.79 0.41

0.23 0.32 0.52 0.45 0.42

0.47 0.52 0.36 0.71 0.42

Mean 0.35 0.41 0.42 0.58* 0.42

SE 0.05 0.03 0.03 0.09 0.01

"D" measured at

* Statistically different from the other
different DCA concentration.

In Vivo Pharmacokinetics

The buffered (pH = 7.4) DCA solution, labeled

concentration 100 mg/ml, was assayed before administration.

The measured concentration was always within 5 % of its

labelled concentration.

Five Dose Study

The average plasma DCA concentration-time profile in

the volunteers is shown in Fig (4.15). The area under the

curve for each set of data from zero to the 24 hour point

using the trapezoidal rule was also calculated. The area

under the curve from 24 hour to time infinity was estimated

by dividing the plasma concentration at 24 hour by the

estimated elimination rate constant of this subject. The

correlation coefficient for a linear relationship between

area under curve to infinity and dose was 0.72 and of

statistical significance (p < 0.05) (Fig. 4.16). An

alternative approach to interpretation of this data is to

ignore the 10 mg/kg data because of the large individual

variation involved (Table 4.16); the average area under

curve for the 50 mg/kg dose was 8.3 + 1.8 mg-hr/ml which is

more than double the average area under curve for the 25

mg/kg dose (2.5 + 1.8 mg-hr/ml) (Table 4.17). The renal

clearance at steady state DCA concentration was estimated by

assuming that steady state was reached after the fourth drug

infusion (6 hours following initial drug administration at a

2 hour dosing interval) because the elimination half life is



m 0

I o
/ -a vo

I 41

-a 0 so

EE i C
-4 W 0o

I I o Ig

o *e o ..0
~< >

d) 0 U5 C >

OO 0

'o i -PU (
-'3 g E g a

Is >

I/ "ouo *o 7
(a 4.J
'--------- r~- B ^ d
<4 in 1 3 m

o. o -m

0--- -- o

(|LU/B7Y) sU03 BO-)

< U


0 0
L ( *4-



Sr ul

\ ****



O )
S> 0 U


\ me

\ U C
-> 03
I ------- 1 -------- t ------- ------ -- oQ"-

* 0




D C )

Table 4.16 Average plasma DCA concentration in 3 healthy
volunteers after five 10mg/kg DCA infusion
for 30 min. with 2 hour interval each.

Time 1 1.5 2 2.5 3 3.5 4 4.5

Conc. 26.4 22.6 22.5 50.1 52.5 39.0 28.1 47.0

S.E. 16.1 19.2 18.6 8.4 57.6 29.3 16.2 11.2

Time 5 5.5 6 6.5 7 7.5 8 8.5

Conc. 67.3 59.7 55.5 46.2 72.8 59.8 64.9 46.6

S.E. 84.5 40.3 44.2 5.4 44.7 38.0 44.3 11.0

Time 12 16 20 24

Conc. 68.7 53.9 46.1 40.9

S.E. 48.5 49.5 46.1 40.9

two hours. The total body clearance at steady state was

estimated by the following equation:

Dose Dose
Clss = = (Eq. 4.1)
AUCss (Cmin + Cmax)/2*r

where T is the dosing interval (2 hours). The Clss values

at 25 mg/kg and 50 mg/kg are listed in Table 4.17. It was

found that there was no significance between the average of

the two doses.

Twenty four hour urinary oxalate excretion was plotted

against various doses of DCA (Fig. 4.17). It was shown that

there was a linear relationship (correlation coefficient =

0.67) between oxalate excretion and DCA doses.

Crossover Study

Figures (4.18-4.21) show the plasma DCA concentration-

time profile on each volunteer in the crossover study. It

was clear that the rate of elimination of DCA decreased with

order of treatment irrespective of the type of the

treatment. The average kel values for the first, second and

third treatments were 0.44 + 0.11, 0.19+ 0.09 and 0.07 +

0.06 (1/hr). There was no significant change in the volume

of distribution for each treatment (20.7 1.0, 22.0 + 0.8,

22.3 + 1.0). The pharmacokinetic parameters are summarized

in Table 4.18. Under such circumstances, a crossover design

is a poor method of comparing IV and oral doses of DCA.

(lU/657) ouo03



-4 + C 0
1 4 4* J

40 40
40 (a 1w

*4 co

.. *41 Mi
40 .
4.1 40)(
C 0 0 C0
0) *O 0)44

o a)
OC P 4.

0 C IS

34 (fV (0 0
G tl O0
L4. C X004

0 0 0
o 4 2


*a CO

4J C-t


0- -4

C0 ri (
0 41 a.4

4 >W 0


C3) 44

(D M 0-i
.C 0 4
Um C
u .0 *C-

Cn 4J

00 40)4J


4 s O

(Iw/571) 03uo3

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E1H69R4WO_EDPH6U INGEST_TIME 2012-09-24T14:12:08Z PACKAGE AA00011845_00001