Effect of fasting on free and esterified carnitine levels in human serum and urine: Correlation with serum levels of free fatty acids and β-hydroxybutyrate

Effect of Fasting on Free and Esterified Carnitine Levels in Human Serum and Urine: Correlation With Serum Levels of Free Fatty Acids and ,&Hydroxyhutyrate Jiri Frohlich, David W. Seccombe,

Peter Hahn, Peter Dodek,

Serum levels of free L-carnitine, acylcarnitines, creatinine, @hydroxybutyrate, free fatty acids, cholesterol, triglycerides, and glucose were determined in healthy volunteers during a 24-36hr fast. The effect of oral administration of free L-carnitine (1 g/person) on these parameters was studied. Urinary excretion of carnitine and creatinine was monitored throughout. Serum and urine levels of free carnitine and its renal clearance decreased during the fast. However, the serum concentration and urinary excretion of acylcarnitines increased during the same interval. Fol-

and Ivo Hynie

lowing the ingestion of free L-carnitine, both serum and urinary levels of free l-carnitine rose. Within 6 hr of ingestion, 10% of the administered dose could be accounted for by urinary excretion. No signiflcant effect on the other serum constituents under study was seen following the oral L-carnitine dose. A significant negative correlation was found between serum levels of free Gcarnitine and &hydroxybutyrate and free fatty acids (r = -0.567, p < 0.001 and I = -0.607, p < 0.001, respectively) during the fast.

TARVATION, DIABETES, AND PROLONGED EXERCISE are, in part, characterized by increased rates of lipolysis with concomitant increases in circulating levels of free fatty acids (FFA) and ketone bodies. L-carnitine (y-trimethylamino+%hydroxybutyrate) is essential for the transport of long-chain fatty acids into mitochondria and thus plays an important role in the oxidation of FFA and the formation of ketone bodies under these conditions.‘v2 The regulation of serum concentrations of L-carnitine and its excretion by the kidney is poorly understood. Serum levels of free carnitine have been found to differ in men and women.3 They are apparently unrelated to body weight, body cell mass, or muscle carnitine concentration.4 A reduction in serum concentrations of free L-carnitine has been reported in malnutrition, malabsorption, Crohn’s disease, and anorexia nervosa.5 Increased excretion of L-carnitine has been reported in muscle diseases,6l7 after ACTH injection,* and in hyperthyroidism.9 Studies of serum concentration and renal handling of carnitine in man may assist in our understanding of its role in those conditions, both normal and pathologic, which are marked by increased rates of FFA turnover. This study examines the effect of a 24-36-hr fast on serum and urinary free

S

From the Division of Clinical Chemistry, Vancouver General Hospital, Centre for Developmental Medicine and Departments of Pathology and Paediatrics. University of British Columbia, Vancouver, B.C.. and Health and Welfare, Ottawa, Ont., Canada. Received for publication June 23, 1977. Supported by a grant from the Vancouver Foundation and a contract from Health and Welfare. Canada. Address reprint requests to Dr. J. Frohlich, Division of Clinical Chemistry, Vancouver General Hospital, 855 West I2th Avenue, Vancouver. B.C. VSZ 1 M9. Canada. @ 1978 by Grune di Stratton, Inc. 00260495/78/2705~0009$01.00/0 Mefobolism, Vol. 27, No. 5 (May), 1978

555

FROHLICH

556

ET At.

carnitine concentrations and the effects of oral administration of L-carnitine. In addition, the relationship between serum free carnitine levels and FFA and ,&hydroxybutyrate (BOH) was determined. The effects of fasting on urinary and serum levels of acylcarnitines were also studied. MATERIALS

AND METHODS

Methods Cholesterol,

triglycerides,

glucose,

and

creatinine

were determined

by standard

methods.‘0-‘3

FFA determined according to NovakI and BOH according to Persson.” Creatinine (cre) and carnitine (car) clearances (C) as well as the percentage tubular reabsorption of carnitine (% TRC) were calculated from the serum (S) and urinary (U) levels (I’, urinary volume) according to the following formulae:

%TRC=

I-+ (

xl00 cre)

Free carnitine was measured according to Seccombe et al. I6 in experiments I and 2 and according to McGarry and Foster ” in experiment 3. The two methods showed good correlation on repeated analyses of quality control sera in both the middle and lower range of serum carnitine concentration. In the McGarry and Foster method, a HEPES buffer was substituted for the TRIS buffer since it has been reported that TRIS buffer will undergo acetylation reactions in the presence of acetylcoenzyme A and carnitine acetyltransferase.‘* Acylcarnitines were determined by subjecting an aliquot of the sample to mild alkaline hydrolysis at 37°C for I hr prior to the determination of carnitine. A set of free L-carnitine standards was routinely subjected to the hydrolysis procedure, from which a standard curve was generated. This curve was used for calculating the carnitine content of hydrolyzed specimens. Serum and urine were either analyzed immediately or stored for a maximum of 2 wk at -4O’C. No changes in free L-carnitine concentration were observed in specimens stored under these conditions for up to I6 mo. Statistical analysis (Student’s t test, paired t test, and calculation of correlation coefficient r) was performed using a Gemsac computer programmed according to Barnett.”

Materials Chemicals were of the highest grade available and were purchased from the following companies: Grand Island Biological Company, Grand Island, N.Y. (L-carnitine); Sigma Chemicals, St. Louis, MO. (acetylcoenzyme A, Dowex-I x IO - 400 in chloride form, NAD, P-hydroxybuTRIS, 5,5’ dithiobis-2-nitrobenzoic acid tyrate dehydrogenase, carnitine acetyltransferase, (DTNB); Calbiochem, La Jolla, Calif. (HEPES); K & K Laboratories, Cleveland, Ohio (sodium tetrathionate); Amicon Corp., Lexington, Mass. (Centriflo ultrafiltration membrane cones); Abbott and Fisher Scientific, Vancouver, B.C. Laboratories (Diagnostic Division) Pasadena, Calif., (barium hydroxide, zinc sulphate, and all extraction solvents). Fluorometric determinations of BOH were made on a G.K. Turner fluorometer, model I I I, Palo Alto, Calif.. and all scintillation counting was performed on a Beckman scintillation counter model LS-230 using New England Nuclear 950-A scintillant.

Experimental Design Three experiments involving 6. I I, and 6 conducted. All the subjects were laboratory ment 1 involved 6 individuals (5 males and stop eating at 9:00 p.m. on day 1 (time 0) and of the following morning. This was designated

apparently healthy volunteers, respectively, were workers fully informed about the study. Experi1 female, ages 21-54 yr). They were instructed to to collect their urine from 9:00 p.m. to 9:00 a.m as the first urinary collection. Blood was collected

CARNITINE

LEVELS

AND

557

FASTING

at 9:00 a.m. and 9:00 p.m. on day 2 and at 9:00 a.m. on day 3. Urinary collections were also made during these 12-hr periods. L-carnitine (1 g in water neutralized to pH 7.0 by NaOH) was administered orally at 36 hr and further blood specimens were obtained 1 and 4 hr later. Urinary collections were made during these intervals and also over the next 6 hr. Experiment 2 involved I I individuals (7 males and 4 females, ages 18-54 yr). Blood was collected at 9:00 a.m. on day I, I hr after breakfast. Another blood specimen was obtained after 12 and 24 hr of fasting. Then neutralized L-carnitine was administered to 7 individuals (5 males and 2 females) and placebo (a 0.9% NaCl solution) to 4 individuals (2 males and 2 females). Serum and urine were collected I, 4, and 5: hr later. Experiment 3 involved 6 volunteers (5 males and 1 female, ages 22-55 yr) who fasted from 9:00 p.m. of day 1 to 12 p.m. on day 3, a total of 39 hr. Blood specimens were obtained after 12, 36, and 39 hr of fasting. Urine was collected every 12 hr and over the last 3 hr of the experiment. The usual amount of activity was allowed during the experiments. In the first experiment, tea and coffee (without sugar and milk) was consumed. Only water was allowed in the second and third experiments. Blood was collected into vacutainer tubes and allowed to clot at room temperature before analysis or storage. RESULTS

The results are summarized in Table ence in the data obtained in experiments together. Both the serum and urinary level of with fasting. The urinary concentration gestion of 1 g of free L-carnitine in both Table 1. Serum and Urinary in 2 Grows

I. Since there was no significant differ1 and 2, the findings will be discussed free L-carnitine decreased significantly increased markedly 1 and 4 hr after inexperiments.

Carnitine

of Volunteers

and Serum BGH and FFA levels (Exaeriments

I and 2)

Cornitine Serum

Experiment

Serum

Urine

(pM/liter)

WVW

FFA

Serum

(meq/liter)

EOH

(mM/liter)

1

12.hr

fast

42.4

zk 11.6

5.2

zt 2.5

0.44

f

0.88

0.795

ZIZ 0.420

24-hr

fast

37.5

f

4.8

+ 3.3

1.44

i

0.46*

0.760

zt 0.225

36-hr

fast

30.4

zt 5.7*

1.2

zt 1.5*

1.13

+ 0.44*

1.475

+ o.a75*

z!z 0.67

1.52

f

0.32”

1.810

zt l.OlO*

1.79

*

0.47*

1.985

ZIZ 0.845*

10.1

1 hr after

carnitine

23.8

+ 5.1*

9.2

4 hr after

carnitine

33.6

zt 10.7

14.6

breakfast

Experiment

i

2.2*?

2 57.9

*

a.3

8.9

& 4.5

0.43

*

0.15

0.115

*

0.130

12-hr

fast

46.1

f

10.6

3.2

zt 3.3

0.80

f

0.24

0.520

f

0.365

24-hr

fast

40.7

zt 13.a*

0.4

+ 0.4*

1.22

*

0.49*

1.060

+ 0.845*

1 hr after

1 hr after

carnitine

47.3

*

13.8

3.3

;t 2.3*

1.59

*

0.47*

0.960

f

1 hr after

placebo

42.6

f

12.7

3.5

+ 3.7

1.73

i

0.39*

1.100

+0.140*

4 hr after

carnitine

54.0

+

17.8

16.6

f

1.54

f

0.54*

1.540

?z 0.340*

4 hr after

placebo

35.4

zt 15.0*

1.8

zt 2.6*

1.67

i

0.50*

1.705

f

40.4

& 14.8

4.9

f

1.81

zto.51*

1.750

f. 0.375*

23.9

& 16.5*

0.2

*0.1*

1.80

+z 0.408*

2.060

f

12.0t

0.235*

0.260*

5 ‘/? hr after carnitine 5%

placebo Two

separate

subjects. terials

5.9

hr after

and

Methods.

*Significantly 1 hr after

experiments

The volunteers

breakfast

tSignificantly

Means

different

were

received f (p

Experiment

carnitine

or

placebo

1 involved after

6 subjects;

experiment

24 or 36 hr of fasting.

2 involved

For details

11

see Ma-

SD are shown. < 0.01)

in experiment

different

performed.

either

0.450*

from

from

the

base

value

(by

paired

t test)

(12-hr

fast

2). placebo

group

(p

< 0.01)

or before

carnitine

administration.

in experiment

1 or

FROHLICH

558

ET AL.

2200 -

; ,”

1600-

1 :! LL 5

1200-

6

1400-

IOOO-

Fig. serum

1. Relationship between FFA and free Gcamitine. Serum values for these constituents at 12, 24, and 36 hr of fasting are shown. l, experiment 1; o, experiment 2. The regression equation (shown by the straight line) is 6 = -24.05. A + 2038, r = -0.607,

800600 -

4002000

IO

20

30

40

50

ScramFras t-Carniline

60

70

80

p < 0.001.

(PM/I)

The differences in the decrease in the serum L-carnitine levels between the two experiments may be due to the different lengths of the fast. In addition, carnitine esters were not determined in these two experiments and it is likely that their levels rose, particularly after the ingestion of carnitine in experiment 1, in which fasting was prolonged. Neither the serum concentration of FFA and BOH nor the levels of glucose, cholesterol, or triglycerides were affected by the carnitine ingestion. A significant negative correlation existed between serum free carnitine and FFA and BOH levels in both experiments (Figs. 1 and 2). The clearance of creatinine and free carnitine decreased over the 24 and 36 hr of fasting. Combined data on clearances and the tubular reabsorption of free L-carnitine are given in Table 2. Only the decrease in L-carnitine clearance was statistically significant. In the third experiment, the effect of fasting on the levels of acylcarnitines in serum and urine was determined. The carnitine esters were quantified by subtracting free carnitine from total carnitine values, the latter being obtained .

.

I

00 0

0’

Cl .

Fig. 2. Relationship between serum @-hydroxybutyrate and free l-carnitine. The same sp=imens as those shown in Fig. 1 were analyzed. The regression equation is g = -0.021 .A + 1.495, r = -0.567, p < 0.001.

. ,\ , ‘X .

0..

o

.O

ON

0%

0

‘0

IO

20

30 Serum

: 40

o

.

me

ii

0

o

50

Free L-Cornihe

.

60 (PM/I)

0

. 70

80

CARNITINE

tEvEts

AND

Table 2. Changes

559

FASTING

in the Creatinine Reabsorption

and Free Carnitine

of Free L-Carnitine

Clearances

and in the Tubular

With Fasting

Clearance (ml/min) Hours of Fasting

Tubular Reabsorption

Free Carnitine

Creotinine

of Free Cornitine (%)

93.1 & 13.2

1.21 kO.34

98.7 + 1.4

12 (17)

94.3

f

1.22 zt 0.39

98.7

f

1.5

24 (17)

89.1

zt 17.3

99.0

f

1.4

82.2

i

99.2

zt 0.8

l(11)’

36 (6) Values

are means

f SD. The data

14.3

.a9 f 0.34t

19.1 from

.66 l 0.217 experiments

1 and

2 have been combined

(12 and

24 hr of fast-

ing). ‘Number

of subjects is given in parentheses.

tsignificant

(p < 0.01)

by paired

t test.

Table 3. Effect of Fasting on Serum levels and Renal Handling

of Acylcarnitines

Acvlcornitines Serum Concentration Hours of Fasting

This

Tubular (%)

12 (6)*

13.5 i

1.1 ho.6

90.8

k 8.5

28.9

zt 10.87

2.9 f 2.0

96.5

f 2.7

39 (6)

26.6

i 8.2t

2.9 zt 1 .1x

96.4

i

table

is based

3 described

in Material

and Methods.

Values

are means

2.7

& SD.

of subjects is given in parentheses.

< 0.001,

$p < 0.05,

an experiment

11.1

Reabsorption

36 (6)

*Number tp

CleoWKe (ml/min)

(pM/liter)

Versus 12 hr of fast (paired

t test).

versus 12 hr of fast.

following mild alkaline hydrolysis. The results are shown in Table 3 and Fig. 3. During the fast, total serum carnitine tended to remain constant or increase slightly. However, the concentration of carnitine esters and the ratio of serum carnitine esters to free carnitine increased markedly. The clearance of the carnitine esters also increased over the 39 hr of fasting.

Ratio of acylcarnitines to free LFig. 3. carnitine concentration in sera of six fasting subjects. Significant increases ( p < 0.01) in the ratio occurred after 36 and 39-hr.

12

36 Hours

of Fasting

39

560

No significant and triglycerides

FROHLICH

changes were observed during the fast.

in serum

levels of glucose,

ET AL.

cholesterol,

DISCUSSION

We found a decrease in serum free L-carnitine with fasting. This finding could be explained by interruption of exogenous supply and/or an increased uptake of carnitine by muscle and other tissues when the rate of oxidation of FFA and the production of ketone bodies is increased. Other possibilities include increased gastrointestinal losses or decreased synthesis from lysine*O during the fasting state. However, our data indicate that at least a portion of the decrease can be accounted for by the formation of acylcarnitines. The fact that a significant negative correlation exists between the levels of free carnitine and that of BOH and FFA may reflect an increase in L-carnitine utilization for the formation of acetyl-, acetoacetyl-, ,&hydroxybutyryland/or other acylcarnitines under conditions of augmented lipolysis. It is of interest that two patients with L-carnitine deficiency syndrome had abnormally high serum ketone bodies2’s22 and one of them had increased serum FFA.2’ In the present work, urinary free carnitine excretion declined with fasting, as did its rate of clearance. In two volunteers, one male and one female, who fasted for five days, free carnitine excretion markedly decreased by 36 hr of fasting and remained at virtually undetectable levels for the remainder of the fast. These results seem contrary to the findings of Maebashi et aLs who described an increase in urinary carnitine excretion in fasting subjects. However, it is not clear whether their method measured free or total carnitine. In experiment 3 we found that urinary acylcarnitine excretion increased with fasting. The overall excretion of “total carnitine” (the sum of free plus esters) increased in all subjects in experiment 3 during the 39 hr of fasting. It is evident that both free and total carnitine levels must be measured before one attempts to interpret the changes in serum and urinary concentrations during fasting. The renal handling of free carnitine is characteristic of a threshold substance. The decreasing serum L-carnitine level correlated well with decreased urinary excretion. After L-carnitine ingestion at the end of the fasting period, approximately 10% of ingested carnitine was excreted into the urine within the first 3-6 hr. This finding is similar to that of others. 8,23It is interesting to note that in each experiment following carnitine administration, free L-carnitine excretion rose far out of proportion to the rise in serum levels, also suggesting an increased clearance and a tubular maximum for reabsorption. The quantity of free L-carnitine excreted into the urine after an oral dose may be related to the length of the fast, perhaps reflecting increased carnitine utilization. The differences in urinary excretion of free carnitine between experimental groups may be attributed in part to the larger proportion of females in experiment 2 and, in part, to the great variability of free carnitine excretion in normal subjects.23 REFERENCES 1. Fritz IB: Carnitine and its role in fatty acid metabolism, in Paoletti R, Kritchevsky D (eds): Advances in Lipid Research, vol 1. New York, Academic 1963, pp 285-334 2. McGarry JD, Robles-Valdes C. Foster

DW: Role of carnitine in hepatic ketogenesis. Proc Nat1 Acad Sci USA 72:4385-4389, 1975 3. Cederblad G: Plasma carnitine and body composition. Clin Chim Acta 67:207-212, 1976 4. Cederblad G, Lindstedt S, Lundholm K:

CARNITINE

LEVELS

561

AND FASTING

Concentration of carnitine in human muscle tissue. Clin Chim Acta 53:31 l-321, 1974 5. Bohmer T, Rydning A, Solberg HE: Carnitine levels in human serum in health and disease. Clin Chim Acta 57:55-61, 1974 6. Dimauro S, Scow C, Penn AS, et al: Serum carnitine: An index of muscle destruction in man. Arch Neural 28:186-190. 1973 7. Maebashi M, Kawamura N, Yoshinaga K: Urinary excretion of carnitine in progressive muscular dystrophy. Nature 249:1733174, 1974 8. Maebashi M, Kawamura N. Sato M, et al: Urinary excretion of carnitine in man. J Lab Clin Med 87:760-766, I976 9. Maebashi M, Kawamura N. Sato M, et al: Urinary excretion of carnitine in patients with hyperthyroidism and hypothyroidism: Augmentation by thyroid hormone. Metabolism 26:35 l356, 1977 10. Allain CC, Poon L, Chan SG, et al: Enzymatic determination of total serum cholesterol. Clin Chem 20:470-474, 1974 I I. Buccolo G, David H: Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 19:476-481, 1973 12. Magar M, Farese G: What is true blood glucose? A comparison of three methods. Am J Clin Pathol44:104-109. 1965 13. Technicon Method No. SF 4-0011 Tarrytown, NY, Technicon Corp 14. Novak

M:

Calorimetric

FH4.

ultramicro-

method for the determination of free fatty acids. J Lipid Res 6:431-433, 1965 15. Persson B: Determination of plasma acetoacetate and D-beta-hydroxybutyrate in newborn infants by an enzymatic fluorometric micromethod. Stand J Clin Lab Invest 25:9- 18, 1970 16. Seccombe DW, Dodek P, Frohlich J, et al: Automated method for L-carnitine determination. Clin Chem 22:1589-1592, 1976 17. McGarry JD. Foster DW: An improved and simplified radioisotopic assay for the determination of free and esterified carnitine. J Lipid Res 17:277-28 I, 1976 18. Christiansen RZ, Bremer J: Active transport of butyrobetaine and carnitine into isolated liver cells. Biochem Biophys Acta 448: 5622577, 1976 19. Barnett RN: Clinical Laboratory Statistics. Boston, Little, Brown, 1971 20. Tanphaichitr V, Horne DV, Broquist HP: Lysine, a precursor of carnitine in the rat. J Biol Chem 246:6364, 1971 21. Vandyke DH, Griggs RC. Markesberg W, et al: Hereditary carnitine deficiency of muscle. Neurology 25:154-159, 1975 22. Angelini C, Pierobon S, Lucke S, et al: Carnitine deficiency: Report of a treated ease. Neurology 25:374, 1975 23. Cederblad G. Lindstedt S: Excretion of L-carnitine in man. Clin Chim Acta 33: Il7- 123. 1971