Ascorbic Charles

acid and carnitine

biosynthesis1’2

J Rebouche It has

ABSTRACT scurvy

(fatigue

and

deficiency. Ascorbate quiring dioxygenase and -y-butyrobetaine biosynthesis.

been

Carnitine

and

may

early

be attributed

features

of

concentrations

renal

are variably

low

in some

pigs. Ascorbic acid deficiency in guinea activity of hepatic -y-butyrobetaine but

not

hepatic

-N-trimethyllysine

hy-

when exogenous substrates were provided. It remains whether vitamin C deficiency has a significant impact on the overall rate of carnitine synthesis from endogenous substrates. Nevertheless, results of studies of enzyme preparations and perfused liver in vitro and of scorbutic guinea pigs in droxylase unclear

vivo

provide

acid

in

compelling

carnitine

evidence

for participation

Am

biosynthesis.

J Clin

ferase

(EC

2.3. 1 .7) and

carnitine-acylcarnitine

translocase,

also

participates in removal of chain-shortened organic acids from mitochondria. This process presumably maintains a pool of nonesterified coenzyme A necessary for efficient functioning of the mitochondria (3-6). Thus, Ciman et al (7) and Hughes et al

to carnitine

is a cofactor for two a-ketoglutarate-rereactions (-N-trimethyllysine hydroxylase hydroxylase) in the pathway of carnitine

tissues ofscorbutic guinea pigs resulted in decreased hydroxylase

that

suggested

weakness)

(2) suggested

that

the lassitude

progression

ofscurvy

may

fatty

for energy

acids

and

be due

fatigability

seen

early

to an impaired

ability

secondary

to carnitine

production

in the

to utilize

defi-

ciency.

Human requirements for carnitine normally are met both by endogenous synthesis and from diet. Primary dietary sources of carnitine are meat, poultry, fish, and dairy products (8). Vegetables, fruits, and grains contain relatively little of this amino acid.

of ascorbic

Nutr

199 l;54:

Pathway

of carnitine

biosynthesis

I1475-525. KEY sine

WORDS

Ascorbic

acid,

hydroxylase,

-y-butyrobetaine

L-carnitine,

acids

hydroxylase

subject.

When

these

signs

appeared,

all physical

huslight diet findings

relating to scurvy were negative. The condition became more marked as the skin lesions and defects in wound healing appeared later in the progression ofthe disease. Early descriptions of the onset and features of scurvy, attributed by Hughes et al (2) to Woodall in 1639 and Lind in 1753, included a generall lazinesse and evil disposition of all the faculties shortnesse a difficultie ofbreathing especially when they moove themselves and “The first indication of the approach of this disease is a listlessness to action or an aversion to any sort of exercise [which] degenerates soon into a universal lassitude, with a breathlessness upon motion Explanations for these early signs in vitamin C deficiency were difficult to reconcile with the known role of ascorbate in collagen synthesis until it was shown that ascorbate is a cofactor, “.

.

.

.

.

.

.

.“

.

for two

.

.

.

.

.

enzymes

.

.

.

in carnitine

biosynthesis.

199l;54:l

l47S-52S.

Printed

in USA.

© 1991 American

Downloaded from https://academic.oup.com/ajcn/article-abstract/54/6/1147S/4715163 by guest on 19 February 2018

is synthesized

from the essential amino 1). The four-carbon chain three through six ofthe lysine

L-methionine

(Fig

arises

from

the

ep-

silon amino group of lysine (9, 10). The methyl groups of carnitine are derived from L-methionine (1 1, 12), via S-adenosylmethionine. In Neurospora crassa the epsilon amino group of lysine is sequentially methylated to form -N-trimethyllysine (13). This series of reactions is catalyzed by a 22-kDa protein, Sadenosylmethionine:-N-L-lysine methyltransferase (14). In mammals methylation apparently occurs only as a posttranslational modification of peptide-bound lysine residues and is catalyzed by one or more 16). S-Adenosyl-L-methionine

protein-lysine provides

these reactions. A for these enzymes -N-Trimethyllysine teolytic digestion to formation of first hydroxylated

of mammalian

19).

It then

variety (17).

is cleaved

is catalyzed

by

an

in an enzyme by an

From the Department

Medicine, Iowa City. 2 Address reprint requests University Society

oflowa

for Clinical

proteins

aldolase-type

and

glycine

that

appears

reaction

(20).

College Nutrition

of Pediatrics,

to Ci Rebouche, of Medicine,

reaction to

high-af-

of Iowa College

Department

Iowa City,

serine

is oxidized

specific,

University

by proleading acid is (18, to -y-tri-

This

identical

NAD-requiring,

(15, for

are substrates

(EC 2. I .2. 1) (1 8). The aldehyde

to -y-butyrobetaine

I

methyltransferases the methyl groups

released from peptide linkage undergoes a series of four reactions L-carnitine. The methylated amino to form /3-hydroxy--N-trimethyllysine

methylaminobutyraldehyde hydroxymethylase

L-Carnitine is a zwitterionic, quaternary amino acid with important biological functions in lipid metabolism. This compound, in association with carnitine palmitoyltransferase (EC 2.3. 1 .21) and carnitine-acylcarnitine translocase, provides a means by which long-chain fatty acids, as acylcarnitine esters, enter mitochondria, where they are oxidized to provide metabolic energy. L-Carnitine, in association with carnitine acetyltransAm J C/in Nutr

and

is provided by carbons and the amino group of carnitine

molecule,

Crandon et al (1), in their classic report ofexperimental man scurvy, noted “a feeling of easy fatigability and weakness” within 3 mo of initiation of the ascorbate-free

in vitro,

L-lysine

ofcarnitine

Introduction

in their

ultimately

L-Carmtine

-N-trimethylly-

of

of Pediatrics,

IA 52242.

I l47S

REBOUCHE

1 l48S

S-AdenosylL-Methionine

9 HN-CH2-CH2-CH2-CH-CH-

#{149}CH3

C-O-peptide

CH3

NH-peptide

L-Lysine

9

CH3-N-CH2-CH2-CH2-CH-CH-C-O-peptide

(peptide-linked)

NH-peptide

e-N-Trimethyllysine

(peptide-Iinked) ,

#{149}H3

CH31

#{149}CH3

0

CH2-CH2-CH2-

CH- CH-

CH3

OH

ProtsIn Hydrolysis

CH3-I1-

____________

NH3

CH2-CH2-CH2-

CH-CH-

CH3

c-N-Trlmilhylysins

CO#{174}

NH3O

Hydroxysi

3-Hydroxy-#{128}-N-trimethyIIysine

c-N-Trimethyllysine

Glycine

#{149}CH3

9

CH3-N-CH2-CH2-

#{149}CH3

______ ___________

CH2-C -H

dehydrogenase

tame

as well

dehydrogenase participates

pathway, reaction

as by

of lower

in the

second

a nonspecific, affinity

hydroxylation

yielding L-carnitine (22, in this four-step sequence,

of the tissues

pathway (24-26).

appears

to be

cytosolic

(2 1). -y-Butyrobereaction

of the

23). Except for the final activity of the enzymes

ubiquitous

in rat

and

human

L-Carnitine

biosynthesis

in mammals.

Sachan and Hoppel (30) and Stein and Englard (31) investigated hydroxylation of #{128}-N-trimethyllysine in rat-kidney homogenates and 70% ammonium sulfate fractions of rat kidney homogenates, respectively. The renal enzyme activity was stimulated slightly by catalase and dithiothreitol. Moreover, Ca2 but not Mg2, Mn2, or Zn2 produced a twofold stimulation ofactivity (30). The enzymatic reaction was inhibited by EDTA in the reaction

Ascorbate-dependent in carnitine biosynthesis The

two

biosynthesis ygenases. ducing ylases

reactions

are catalyzed These enzymes (ascorbic

for activity.

to the family

participate

pathway

of carnitine

by a-ketoglutarate-dependent require reduced iron (Fe2) acid)

resemblance

that

reactions

in the

in the

In this

regard,

of prolyl

synthesis

medium.

was parallel 1.3.99.1)

hydroxylation

agent

a striking

hydroxylation

and

dioxa re-

they

bear

and lysyl hydrox-

of collagen.

&N-Trimethyllysine

to the distribution and

citrate

synthase

Trimethy//ysine,2-oxog/utarate

Broquist

EC

(EC

dioxygenase

eimide,

iodoacetamide,

Stein and methyllysine

and colleagues

(27, 28) first demonstrated

is an Hoppel

in

trimethyllysine Subsequently,

intermediate and coworkers

droxy--N-trimethyllysine

as an

carnitine (19, 29)

intermediate

that

in the

Hulse

et al (1 8) demonstrated

hydroxylation

of

lysine

in rat-liver

The

reaction

ct-ketoglutarate, Dithiothreitol of ascorbate No enzymatic tions tivity

mitochondria.

Fe2, and

and ascorbate

citrate

were

but not NADH

almost

and a-ketoglutarate, activity was found

of liver extracts. Catalase in rat-liver mitochondria.

enzymatic totally

did

not

pathway.

-N-trimethylrequired

or NADPH.

ineffective

respectively, in microsomal stimulate

i-N-

biosynthesis. identified /3-hy-

in this reaction. or soluble fracenzymatic

ac-

Downloaded from https://academic.oup.com/ajcn/article-abstract/54/6/1147S/4715163 by guest on 19 February 2018

indicating

(EC a mito-

and

iodoacetate

Substrate,

24 imol/L,

(26).

cosubstrate,

and

respectively

(26).

Sachan

cofactor

of -N-triand skeletal

requirements

and

the same in all tissues. Apa-ketoglutarate and fer0.46-0.53 mmol/L, and 20and

Hoppel

(30)

reported

a Km for -N-trimethyllysine (presumably the L isomer) of 1.6 mmol/L. No Km has been reported for ascorbate but 2.5 mmol/ L is required for optimal activity in vitro in preparations from liver

(18)

were

reported

per

in place

4.1.3.7),

Englard (26) compared the properties hydroxylase in rat kidney, liver, heart,

for each were essentially parent Km for -N-trimethyl-DL-lysine, rous iron were 0.1-0.12 mmol/L,

(trimethy//ysine

activity

dehydrogenase

chondrial localization (30). The renal enzyme has a broad pH optimum of 6.5 to 7.5 at 37 #{176}C (26). It is inhibited by the sulfhydryl reagents p-chloromercuriphenyl sulfonate, N-ethylmal-

muscle.

1.14.11.8)

hydroxylase

ofsuccinate

Km values

hydroxy/ase;

C-00

OH

x asi

‘y-Butyrobetaine of carnitine

CH-CH--

CH3 H

FIG 1. Pathway

finity

CH341-CH2-

CH3

‘-ThmethyIaminobutyraIdehyde

aldehyde

#{149}CH3

CH3N_CH2CH2CH2_COe

r.T#{241}m.i.mino. b1yrald.hyd. D.hydras.

CH3

?

unit

droxylase However, the whole

and weight

kidney

(30).

Higher

to be inhibitory of tissue

(18).

or protein)

concentrations

Highest

of ascorbate

activity

(expressed

of -N-trimethyllysine

hy-

was found in kidney of rats (31) and humans (25). because of its large mass relative to other tissues of animal, skeletal muscle contains a majority of e-N-

trimethyllysine has not been

hydroxylase purified from

activity in the any source.

body.

This

enzyme

ASCORBIC TABLE

ACID

AND

CARNITINE

l149S

BIOSYNTHESIS

1

Kinetic

constants

for substrates

and cofactors

for ‘y-butyrobetaine

hydroxylase

from

various

sources*

H uman liver

Bovine liver

Rat liver

Pseudomonas

sp AK

I

Kmt

‘y-Butyrobetaine

(jzmol/L)

16-66

(38)

510

(40)

200 (39)

820

(40)

300 (39)

2400

(43)

7(44)

a-Ketoglutarate

50-290 500

(zmol/L)

Ascorbate (mmol/L) Fe2 (imol/L) C

Numbers

(35)

12

EC

Linneweh

dioxygenase

Lindstedt

ofcarnitine.

(22)

demonstrating

and

Bremer

droxylation preparation

Lindstedt

(33)

(23)

reported

demonstrated Isoascorbate

was

reductants,

in rats in vivo.

reported

catalyzed

studies

in vitro

by a soluble

protein

substitute

for ascorbate

as 2,6-dichlorophenolindophenol

but and

(35).

The

preventing

activity

bovine

serum

oxide

dismutase,

effective.

by

(37).

Rb.

from

lecular

weights

different

of the

Enzyme mildly

(38). hydroxylase

(38).

less glu-

catalase hy-

Pseudomonas of39

also

inhibits

-y-butyrobetaine

is a heterodimer

000

and

amino-terminus

37 000

amino

acid

hydroxylase from by chromatofocusing

5.7, and

5.8.

The

combinations

of two

subunits

(42 kDa). man liver

-y-Butyrobetaine also was separated

(41).

purified

three

forms

differing hydroxylase into three

rabbits, absent

in varying

and rat testis velopmentally not

and

(25,

mus-

of hamsters,

is present in human

in liver of brain (25)

hydroxylase activity is de(25, 47) and rat (48) liver

47).

administered

In all spe-

skeletal

in kidney

hydroxylase activity was demonstrated

kidney

by orally

and

but is totally or almost totally pigs, mice, and dogs (25, 45,

(24). -y-Butyrobetaine regulated in human

in human

increased

cardiac

abundance

cats, monkeys, and humans from kidney of rats, guinea

46). -y-Butyrobetaine all species studied

but

from

Hepatic

enzyme

L-thyroxine

activity

is

in rats (49).

of carnitine

biosynthesis

rat

had

but

biosynthesis (50).

is determined In humans

it is

not known if availability of -N-trimethyllysine or activity one of the enzymes in the pathway from -N-trimethyllysine

of to

-y-butyrobetaine

In

human

is rate-limiting

for

carnitine

biosynthesis.

adults

50-52).

purified

Evidence biosynthesis Effect

ofascorbic

carnitine

synthesis decrease

size

of huKinetic

Downloaded from https://academic.oup.com/ajcn/article-abstract/54/6/1147S/4715163 by guest on 19 February 2018

acid

for ascorbate pig

deficiency

on tissue

in carnitine

and

concentrations

The discovery of the in vitro requirement for ascorbate by two enzymes in the pathway ofcarnitine biosynthesis prompted several investigations into the role ofascorbate in carnitine bio-

points not

for a requirement in the guinea

serum

resolved

to be dimeric extracts (42).

rate ofcarnitine of &N-trimethyllysine

hy-

mo-

Highly

availability

(47,

from

kidney was (42). Isoelectric

by the

by

subunit

in crude isoenzymes

It is present

is absent

biosynthesis,

specific.

in-

subunits

appear

activity

is tissue

de-

two

in charge

dc.

this

activity

various

also

with

residues.

human

for the

and

is stimulated

The

reported

relative rates of dietary carnitine intake and unnary carnitine and #{128}-N-tnmethyllysine excretion indicate that at least half of available endogenous -N-trimethyllysine is convented to carnitine (5 1). -y-Butyrobetaine hydroxylase activity is not rate-limiting for carnitine biosynthesis in rats or humans

K

hydroxylation

partially

cies studied

In rats the normal

of -y-butyrobetaine

activity

was

cofactors

hydroxylase

Regulation

super-

ofreduced

efficient than -y-butyrobetaine

enzyme

-y-butyrobetaine

including bovine

but were somewhat

affinities

and

stimulated

human kidney (39) and purified to homogeneity (40) and Pseudomonas sp AK 1 (41). The enzyme is a homodimer with 46-kDa subunits (40). The

-y-butyrobetaine into three forms

5.6,

the to the

Phosphate

droxylase activity -y-Butyrobetaine

enzyme

and

for substrates

enzyme preparations are listed in Table 1. Unlike other enzymes in the pathway ofcarnitine

by

a purified

proteins,

in the presence

efficiency

reactions.

In

stimulates rat liver -y-butyrobetaine synthesis and a-ketoglutarate

binding

coupling

(35) and calfliver calfliver

Other

hemoglobin,

carnitine by increasing

carboxylation

liver from from

(36).

a-ketoglu-

1 , catalase

sp AK

less

activity

ascorbate.

for catalase

K

and

enzyme

shown to be ten times more partially purified rat liver

a-ketoglutarate

and

and

peroxidase

activity

the

Fe2

substituted

hydroxylase-catalyzed decarboxylation creases

oxygen

stimulates

human

Glutathione

droxylase

requires

---300-fold

albumin,

tathione was in protecting

were

by

were

Pseudomonas

from

enzyme

NH

also

Catalase

inactivation

preparation

and

enzyme

other

2-amino-

5,6-dimethyl-4-hydroxy-5,6,7,8-tetrahydropteridine,

as cosubstrates.

hy-

In partially purified preparations, they for Fe2 and ascorbate (34, 35).

an effective

such

isotope

relationship

subsequently

of -y-butyrobetaine

from rat liver. requirements

to dogs and noted years later Lindstedt

Thirty-two

the precursor-product and

constants

(-y-butyrobetaine

(32) in 1929 fed -y-butyrobetaine

Lindstedt

tarate

-

60 (43)

are references.

1.14.11.1)

excretion

effective

-

10 (39)

(40)

Km.

hydroxylase;

and

5. 1 (40)

-

‘y-Butyrobetaine,2-oxoglutarate

increased

450 (43)

(35)

100

in parentheses

t Apparent

(38)

in vivo

in cardiac in scorbutic guinea

decreased

carnitine

in guinea

pigs.

Ciman

et al (7) first

reported

a

muscle but not liver carnitine concentration pigs. Subsequently, in scorbutic guinea pigs concentrations

muscle (2, 53-55), cardiac muscle and kidney (54). In the same studies

were observed in skeletal (53, 54, 56), liver (54-57),

brain

and serum

(54), heart

11 50S

REBOUCHE

(55),

liver

and

(53),

(53, 55) carnitine

kidney

were not different in scorbutic animals In one study carnitine concentration ascorbic

acid-deficient

Sandor

guinea

et al (57)

pair-fed against (thus chronically tion

that

animals underfed

receiving because

by scorbutic

carnitine

pigs than

observed

animals)

concentrations

concentrations

compared in serum in control

animals

ascorbate-sufficient

reduced

compared

with

(53).

guinea

an ascorbate-deficient ofdecreased food

had

guinea

with controls. was higher in

hepatic

ad libitum-fed

the two restored

and

ascorbate before perfusion with results led the authors to conclude hydroxylase tivity

after

18-2

1 d ofthe

dietary

(58).

Analysis

concentration

was

restored

macologic

administration

carnitine

Jones

and

The

acid.

without

not

weight

(59) reported of animals

a carnitine

ceiving d, P

acid deficiency and carnitine on weight change and surviva/

and Hughes

24 d in the

the carnitine 0.01)