www.elsevier.com/locate/clinchim

Clinica Chimica Acta 348 (2004) 49–55

Effect of cigarette smoking on serum a-tocopherol and the lipid

profile in a Portuguese population

Paula A. Lopesa, Maria C. Santosb, Luıs Vicentea, Ana M. Viegas-Crespoa,*

aCentro de Biologia Ambiental and Departamento de Biologia Animal, Faculdade de Ciencias, Universidade de Lisboa, Campo Grande,

Bloco C2, 3o piso, 1749-016 Lisboa, PortugalbCentro de Quımica e Bioquımica and Departamento de Quımica e Bioquımica, Faculdade de Ciencias, Universidade de Lisboa,

1749-016 Lisboa, Portugal

Received 28 January 2004; received in revised form 23 April 2004; accepted 29 April 2004

Available online

Abstract

Background: The present study was conducted to determine the effects of cigarette smoking on the levels of serum a-

tocopherol and on the lipid profile in humans from the Lisbon population. Methods: Serum a-tocopherol was evaluated by a

reverse-phase HPLC method with UV detection. Enzymatic tests were used for the evaluation of the lipid profile. Results:

Smokers in general and female smokers in particular had decreased a-tocopherol levels when compared with nonsmokers.

Smokers had also lower HDL cholesterol (HDL-C) contents, but this difference was statistically significant only for females.

Regardless of sex, in smokers, there was a positive correlation between a-tocopherol and triglyceride (TG) levels. Cluster

analysis showed a sex-independent separation between smokers and nonsmokers. Conclusion: These results suggest a different

interaction of these blood parameters in smokers versus nonsmokers that should be further investigated.

D 2004 Elsevier B.V. All rights reserved.

Keywords: a-Tocopherol; Lipid profile; Cigarette smoking; Humans

1. Introduction vitamin E is the most abundant tocopherol in human

Vitamin E is the major free radical chain terminator

in the lipophilic environment. All forms of vitamin E

are taken up by intestinal cells and released to the

blood circulation into chylomicrons. In the liver, the

a-tocopherol transfer protein (a-TTP) selectively sorts

out a-tocopherol from all incoming tocopherols for

posterior incorporation into VLDL [1]. This form of

0009-8981/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.cccn.2004.04.033

* Corresponding author. Tel.: +351-217500000; fax: +351-

217500028.

E-mail address: amcrespo@fc.ul.pt (A.M. Viegas-Crespo).

serum [1].

Most investigators have studied the antioxidant

potential of vitamin E and, based on this property,

have suggested several explanations for the vitamin’s

various actions. Accordingly, it is widely believed that

vitamin E is helpful in preventing or in modulating

diseases that are associated with oxidative stress [1].

There are evidences that oxidized low lipoproteins

(LDL) may be of central importance in the in vivo

atherogenic process [2,3]. However, supplements of

dietary vitamin E in clinical trials have not prevented

consistently cardiac attacks in humans with estab-

P.A. Lopes et al. / Clinica Chimica Acta 348 (2004) 49–5550

lished coronary heart disease [3,4]. Such inconsistent

results have questioned the role of vitamin E as a

protective agent in human atherosclerosis.

Antioxidants may behave differently depending

on the environmental conditions as it occurs when

vitamin E is incorporated into lipoproteins. In this

case, the tocopherol radical that is generated in

consequence of this vitamin’s chain breaking role

might itself become an oxidant agent that promotes

lipid peroxidation in LDL [4]. This putative double

action of vitamin E raises doubts about this vita-

min’s ability to increase in vivo antioxidant mecha-

nisms [1,4]. Probably, the biological functions of

vitamin E were underestimated when considered

solely as an antioxidant defense. Recently, specific

new functions of a-tocopherol have been found,

which certainly will become an important subject

for future research [1]. These functions may be

related to the role of vitamin E in the regulation

of cellular signaling, gene transcription, and induc-

tion of apoptosis [1].

Chronic cigarette smoking is believed to be asso-

ciated with coronary and peripheral vascular disease,

and represents a risk factor for the development of

atherosclerosis [5–7]. Cigarette smoke contains oxi-

dants that can promote oxidative modifications in

LDL and in other critical biomolecules. The direct

exposure to cigarette smoke represents only part of the

total oxidative stress experienced eventually by smok-

ers, since cigarette smoke also contributes to addi-

tional endogenous oxidant formation, through its

effects on the inflammatory-immune response [7].

Additionally, tobacco consumption can increase the

concentration of triglycerides (TGs) and lowers the

concentration of HDL cholesterol (HDL-C), which

may contribute to the atherogenic potential of tobacco

consumption [8]. Both dietary polyunsaturated fats

and vitamin E appear to be associated with reduced

atherosclerosis, but in smokers, they may contribute to

lipid oxidation [6]. Moreover, a-tocopherol in smok-

ers appears to have a particular kinetics compared

with nonsmoking individuals [5].

This work is part of a large project that investigates

blood parameters associated with antioxidant systems

in humans from different areas of Portugal. The

purpose of the present study was to evaluate the effect

of cigarette smoking on serum levels of vitamin E (a-

tocopherol), and to determine its relationship with the

lipid profile in healthy Portuguese subjects who were

born and inhabit the city of Lisbon.

2. Material and methods

2.1. Subjects

The protocol of this study was approved by the

Human Ethics Committee of the National Health Insti-

tute Dr. Ricardo Jorge. A total of 182 volunteer blood

donors (133 females and 49 males) were recruited in

the Lisbon area. The subjects belonged to the middle

class and had urban dietary habits. Before blood was

drawn informed consent forms were obtained from

each donor. All subjects were screened to determine

eligibility, which was based on a clinical report where

the subjects stated all relevant information. Therefore,

information regarding general clinical state, namely

drug intake and smoking habits was obtained. Healthy

reference subjects were considered eligible if they had

no acute or chronic illness. No subjects were, under

vitamin E supplementation or under drug therapy,

known to affect serum lipid levels. The subjects were

divided according to sex and tobacco consumption.

The smokers consumed about 15 cigarettes per day.

Throughout this paper, whenever we refer to tobacco

consumption, we are describing cigarette smoking.

2.2. Blood collection

The blood collection took place at the National

Health Institute Dr. Ricardo Jorge in the morning and

after 12 h of fasting conditions. Peripheral blood

samples were obtained by venipuncture and placed

into polypropylene tubes, and serum was removed

after centrifugation at 1500� g for 10 min at 4 jC.Serum aliquots were drawn into eppendorf tubes,

frozen, and stored at � 20 jC for further a-tocopherol

determination. The remaining serum was divided into

aliquots for lipoprotein separation and lipid profile

analysis.

2.3. Analytical procedure

Serum a-tocopherol was evaluated by a reverse-

phase HPLC method using UV detection, according to

Julianto et al. [9]. Separation of HDL lipoproteins was

Table 1

Serum lipid parameters for subjects (females and males) from the

Lisbon population, according to smoking habits

Nonsmokers Smokers

Female Male Female Male

Age (years) (n= 112)

42F 15

(n= 35)

38F 16

(n= 21)

38F 13

(n= 14)

34F 12

Parameter

TC (mg/dl) 199

(126–334)

196

(127–279)

182

(129–258)

192

(105–259)

HDL-C (mg/dl) 65

(34–119)a,b53

(30–121)b51

(24–82)a44

(25–71)

LDL-C (mg/dl) 114

(46–206)

117

(75–190)

105

(61–187)

114

(60–182)

TG (mg/dl) 78

(25–283)

85

(34–242)

79

(35–235)

95

(37–155)

Regarding age, values represent the meanF standard deviation.

Data for the lipid parameters represent the median and the 95%

confidence limits of the median are shown in parentheses. Student’s

t test was applied for TC and HDL-C, and Mann–Whitney U test

was applied for LDL-C and TGs. Values in each row with the same

superscript are significantly different, p< 0.05.

P.A. Lopes et al. / Clinica Chimi

done by adding polyethylene glycol to the fresh

samples to precipitate other lipoproteins [10]. The

direct quantitative determinations of total cholesterol

(TC) and HDL-C in the sera were based on enzymatic

assays using the CHOD-PAPR and the HDL-C PlusRkits, respectively (Roche Diagnostics, Mannheim,

Germany). The concentration of LDL cholesterol

(LDL-C) was calculated by using the Friedewald

formula [11]. The determination of serum triglycer-

ides was based on an enzymatic test using the GPO-

PAPR analysis kit (Roche Diagnostics).

Table 2

Serum a-tocopherol concentrations and respective ratios to TC and to T

according to tobacco consumption

a-Tocopherol (Amol/l)

Female Nonsmokers 20.8 (4.3–62.6)a

(n= 112)

Smokers 16.2 (8.8–40.2)a

(n= 21)

Male Nonsmokers 17.0 (8.2–56.3)

(n= 35)

Smokers 15.6 (6.3–36.1)

(n= 14)

Data represent the median and the 95% confidence limits of the median

superscript are significantly different, Mann–Whitney U test, p< 0.05.

2.4. Statistical analysis

The software used for statistical analysis was Stat-

graphics Plus for Windows 4.0R (Statistical Graphics,

USA). Normality was tested through coefficients of

skewness and kurtosis. With the exception of total

cholesterol and HDL-C, the measured parameters

were not normally distributed. Results were expressed

as median and 95% confidence limits of the median

for all parameters, including those that were normally

distributed. The median confidence limits were calcu-

lated through bootstrapping. Mean differences be-

tween the groups with normal distributions were

analysed with Student’s t test for independent sam-

ples; the Mann–Whitney U test was used for the other

groups comparisons. The a level was set at 0.05.

Associations among parameters were investigated

using the Spearman’s rank correlation coefficient. A

cluster analysis (multivariate analysis) was performed

to establish the distance among groups, taking into

account possible associations between variables.

ca Acta 348 (2004) 49–55 51

3. Results

3.1. Lipid profile and a-tocopherol levels according tosmoking habits

In each sex, lipid profile values were similar for

both groups, except for HDL cholesterol contents that

were lower in female smokers (Table 1). For male

smokers, a tendency for a decrease in this parameter

appeared to occur as compared with nonsmokers.

Gs for subjects (females and males) from the Lisbon population,

a-Tocopherol/TC (Amol/g) a-Tocopherol/TG (Amol/g)

9.7 (2.8–21.7) 23.7 (4.9–58.4)b

(n= 106) (n= 103)

8.1 (5.3–21.6) 18.3 (12.3–37.3)

(n= 21) (n= 21)

9.6 (4.7–19.7) 21.1 (6.0–41.4)b

(n= 34) (n= 33)

10.8 (3.2–19.3) 18.4 (10.1–42.2)

(n= 14) (n= 14)

are shown in parentheses. Values in each column with the same

P.A. Lopes et al. / Clinica Chimica Acta 348 (2004) 49–5552

Furthermore, significant lower HDL cholesterol levels

were found for nonsmoking males relatively to non-

smoking females (Table 1).

Female smokers had lower levels of a-tocopherol

when compared with nonsmokers (Table 2). Similar

tendency was observed for males, although the differ-

ence was not statistically significant (Table 2). Also,

when both sexes were combined, a slight decrease in

a-tocopherol levels was observed in smokers when

compared with nonsmokers (smokers: median = 15.6

Amol/l, 95% confidence limits of the median: 6.3–

40.2; nonsmokers: median = 19.7 Amol/l, 95% confi-

dence limits of the median: 4.3–62.6; Mann–Whit-

ney U test, p>0.05). In addition, nonsmoking males

had a lower a-tocopherol to triglycerides ratio than

nonsmoking females (Table 2).

3.2. Regression analysis

With respect to the relationship between a-tocoph-

erol and the lipid parameters for female smokers, a

statistically highly significant positive correlation

( p < 0.001, r= 0.67) was observed between a-tocoph-

erol and triglycerides (Fig. 1). For male smokers, a

Fig. 1. Relationship between serum a-tocopherol and triglyc

statistically significant association ( p < 0.05, r = 0.57)

was also observed between a-tocopherol and trigly-

cerides (Fig. 2). In these two cases, the linear regres-

sion model was the one that best fitted the data. The

two regression lines were not statistically different (F

ratio for intercept = 0.01, p>0.05 and F ratio for

slope = 0.06, p>0.05). Regardless sex, a positive cor-

relation between serum a-tocopherol and total cho-

lesterol was observed when all subjects were

considered (r = 0.29, p < 0.05).

3.3. Multivariate analysis

A cluster analysis considering a-tocopherol and

triglycerides, showed a clear proximity between non-

smoker groups of females and males (FNS and MNS,

respectively), as compared to smokers (FSK and

MSK, respectively) (Fig. 3).

4. Discussion

In this study, the concentrations of serum a-tocoph-

erol are generally in the same range of those reported

erides for female smokers from the Lisbon population.

Fig. 3. Cluster representation taking into account a-tocopherol and triglycerides as parameters and dividing subjects according to sex and

smoking habits. FSK= Female smokers; MSK=Male smokers; FNS= Female nonsmokers; MNS=Male nonsmokers.

Fig. 2. Relationship between serum a-tocopherol and triglycerides for male smokers from the Lisbon population.

P.A. Lopes et al. / Clinica Chimica Acta 348 (2004) 49–55 53

P.A. Lopes et al. / Clinica Chimica Acta 348 (2004) 49–5554

in the literature for healthy people [1,8,9,12,13]. How-

ever, the a-tocopherol serum concentrations measured

in this sample of the Portuguese population was

slightly lower than the values reported for other

European populations (about 27 Amol/l), including

that for the Spanish population (28 Amol/l) [14]. No

significant sex-based differences in serum a-tocoph-

erol levels were observed in our sample which is in

agreement with data reported by Winklhofer-Roob et

al. [15], but differs from results found for two Asian

populations [13,16]. The intake of vitamin E through

the diet appears to be the main influencing factor on its

physiological levels [7], but metabolic and genetic

factors, as well as lifestyles, may interfere with the

serum levels of this lipophilic antioxidant [1,7]. The

similarity between sexes shown in this study, might

reflect identical dietary habits between females and

males of the Lisbon population, contrasting with what

happens in other populations [13,16], but this discrep-

ancy must be further investigated.

Smoking habits led to lower levels of serum HDL

cholesterol, and this effect was clearer in females,

which is essentially in accordance with previous

reports [8,17]. In fact, Imamura et al. [17] observed

decreased levels of HDL cholesterol, HDL2 choles-

terol, total cholesterol, and a lower level of activity for

lecithin/cholesterol acyltransferase (LCAT) in smok-

ing young women. Although the HDL cholesterol

contents are affected by a variety of factors, the more

evidenced effect in female smokers might be related to

possible interactions with hormonal metabolism, in-

cluding the conceivable estrogens’ modulator effect

on lipoprotein metabolism [18,19]. Anyway, these

results may emphasize the adverse effects of cigarette

smoking, in particular its putative role in atherogen-

esis, since HDL cholesterol may be negatively asso-

ciated with the occurrence of coronary artery disease

[20].

The present findings suggest that smoking habits

promote lower levels of serum a-tocopherol mainly in

females, which is consistent with the hypothesis that

smoking leads to a faster disappearance of serum a-

tocopherol due to an increased oxidative stress in

smokers [5]. Generally, our results are very similar

to others reported in the literature for levels of

circulating antioxidant micronutrients in smokers [7].

Ratios of a-tocopherol to total cholesterol and to

triglycerides were also considered because either of

these two indices is thought to be a more reliable

indicator of vitamin E status than the concentration of

a-tocopherol by itself [8]. In particular, a higher value

of a-tocopherol to cholesterol ratio is also thought to

express some degree of protection against cardiovas-

cular disease [8,13]. In a study by Kharb and Singh

[8], patients with acute myocardial infarction or sta-

bilized individuals who had survived to cardiovascu-

lar disease had lower values of these ratios, mainly

due to an increase of serum cholesterol and triglycer-

ides. The similarity of those ratios for smokers and

nonsmokers shown in the present study is certainly

related to the moderate variations of the involved

parameters and can reflect the apparent healthy con-

dition of the studied subjects [21]. However, in

nonsmoking subjects, the a-tocopherol to triglycer-

ides ratio was higher in females than in males prob-

ably due to an increase of the statistical power that

results from the combination of a-tocopherol (slightly

higher in females) with triglycerides levels (slightly

increased in males).

A positive correlation between serum a-tocopherol

and total cholesterol was found for all subjects, which

is in accordance with data reported by other authors

[13,15]. A positive correlation was also found be-

tween a-tocopherol and triglycerides in smoking

subjects of both sexes. Furthermore, a separation

through cluster analysis between smokers and non-

smokers by these two parameters was detected.

According to Traber et al. [5], human smokers under

a diet supplemented with deuterated a-tocopherol

acetate had similar concentrations of unlabeled plasma

a-tocopherol but exhibited lower plasma concentra-

tions of deuterated a-tocopherol than nonsmokers,

suggesting different vitamin E kinetics for smokers

and nonsmokers. In turn, Zaratin et al. [22] defend

that smoking habits may increase the lipolysis of

triglyceride-rich lipoproteins. Based on the hypothesis

that vitamin E transport from the liver via VLDL and

out of VLDL after lipolysis is controlled by the

metabolism of triglycerides [23], then one could

expect an alteration in the smokers’ ability for transfer

a-tocopherol. Similarly to the findings of Traber et al.

[5], our results may point out a different interaction

between serum a-tocopherol and triglycerides in

smoking versus nonsmoking subjects. Therefore, it

is important to clarify the mechanisms involved in the

transport and remotion of a-tocopherol under oxidant

P.A. Lopes et al. / Clinica Chimica Acta 348 (2004) 49–55 55

conditions in order to explain any changes in serum

a-tocopherol levels due to cigarette smoking.

Acknowledgements

This work is part of the project ‘‘Blood parameters

associated with antioxidant function in human pop-

ulations from Portuguese regions’’ that was supported

by Fundac�ao para a Ciencia e a Tecnologia (PRAXIS

XXI/PSAU/66/96) and Centro de Biologia Ambiental.

Paula Alexandra Lopes is grateful for a PhD fellow-

ship (FCT/PRAXIS XXI/BD/21444/99). Authors are

thankful to National Health Institute Dr. Ricardo Jorge

for blood collection and particularly to Maria Odete

Rodrigues for helping in the assessment of serum lipid

profile. We thank Jose Pedro do Amaral for revising

the manuscript.

References

[1] Brigelius-Flohe R, Kelly FJ, Salonen JT, Neuzil J, Zingg JM,

Azzi A. The European perspective on vitamin E: current know-

ledge and future research. Am J Clin Nutr 2002;76:703–16.

[2] Meydani M. Vitamin E and atherosclerosis: beyond preven-

tion of LDL oxidation. J Nutr 2001;131:366S–8S.

[3] Heinecke JW. Is the emperor wearing clothes? Clinical trials

of vitamin E and the LDL oxidation hypothesis. Arterioscler

Thromb Vasc Biol 2001;21:1261–4.

[4] Heinecke JW. Clinical trials of vitamin E in coronary artery

disease: is it time to reconsider the low-density lipoprotein

oxidation hypothesis? Curr Atheroscler Rep 2003;5:83–7.

[5] Traber MG, Winklhofer-Roob BM, Roob JM, Khoschsorur G,

Aig Cross C, Ramakrishnan R, et al. Vitamin E kinetics in

smokers and nonsmokers. Free Radic Biol Med 2001;31:

1368–74.

[6] Weinberg RB, VanderWerken BS, Anderson RA, Stegner JE,

Tho MJ. Pro-oxidant effect of vitamin E in cigarette smokers

consuming high polyunsaturated fat diet. Arterioscler Thromb

Vasc Biol 2001;21:1029–33.

[7] Alberg A. The influence of cigarette smoking on circulating

concentrations of antioxidant micronutrients. Toxicology

2002;180:121–37.

[8] Kharb S, Singh GP. Effect of smoking on lipid profile, lipid

peroxidation and antioxidant status in normal subjects and in

patients during and after acute myocardial infarction. Clin

Chim Acta 2000;302:213–9.

[9] Julianto T, Yuen KH, Noor M. Simple HPLC method for

determination of a-tocopherol in human plasma. J Chroma-

togr B Biomed Sci Appl 1999;732:227–31.

[10] Ballantyne FC, Clarck RS, Simpson HS, Ballantyne D. HDL

and LDL subfractions in myocardial infarction in control sub-

jects. Metabolism 1982;31:433–7.

[11] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the

concentration of low-density lipoprotein cholesterol in plasma,

without use of the preparative ultracentrifuge. Clin Chem

1972;18:499–502.

[12] Schwedhelm E, Maas R, Troost R, Boger R. Clinical pharmo-

cinetics of antioxidants and their impact on systemic oxidative

stress. Clin Pharmacokinet 2003;42:437–59.

[13] Abiaka C, Olusi S, Simbeye A. Serum concentrations of mi-

cronutrient antioxidants in an adult Arab population. Asia Pac

J Clin Nutr 2002;11:22–7.

[14] Oldemilla B, Granado F, Southon S, Wright AJA, Blanco I,

Gil-Martinez E, et al. A European multicentre, placebo-con-

trolled supplementation study with alpha-tocopherol, caro-

tene-rich palm oil, lutein or lycopene: analysis of serum

responses. Clin Sci 2002;102:447–56.

[15] Winklhofer-Roob BM, Van’t Hof MA, Shmerling DH. Refer-

ence values for plasma concentrations of vitamin E and A and

carotenoids in a Swiss population from infancy to adulthood,

adjusted for seasonal influences. Clin Chem 1997;43:146–53.

[16] Kim SY, Lee-Kim YC, Kim MK, Suh JY, Chung EJ, Cho SY,

et al. Serum levels of antioxidant vitamins in relation to cor-

onary artery disease: a case control study of Koreans. Biomed

Environ Sci 1996;9:229–35.

[17] Imamura H, Teshima K, Miyamoto N, Shirota T. Cigarette

smoking, high density lipoprotein cholesterol subfractions,

and lecithin:cholesterol acyltransferase in young women. Me-

tabolism 2002;51:1313–6.

[18] Knopp RH, Walden CE, Wahl PW, Hoover JJ. Effects of oral

contraceptives on lipoprotein triglyceride and cholesterol:

relationships to estrogen and progestin potency. Am J Obstet

Gynecol 1982;142:725–31.

[19] Srivastava N, Chowdhury PR, Averna M, Srivastava RA. Es-

trogen increases hepatic lipase levels in inbred strains of mice:

a possible mechanism for estrogen-dependent lowering of

high density lipoprotein. Mol Cell Biochem 2001;220:87–93.

[20] Shaul PW, Mineo C. HDL action on the vascular wall: is the

answer NO? J Clin Invest 2004;113:509–13.

[21] Expert Panel on Detection, Evaluation, and Treatment of High

Blood Cholesterol in Adults (Adult Treatment Panel III). Ex-

ecutive summary of the third report of the National Cholesterol

Education Program (NCEP) Expert Panel on Detection, Eval-

uation, and Treatment of High Blood Cholesterol in Adults

(Adult Treatment Panel III). JAMA 2001;285:2486–97.

[22] Zaratin ACM, Bertolami MC, Faludi AA, Rocha JC, Nunes

VS, Nakandakare ER, et al. Acute in vivo chylomicron

metabolism and postalimentary lipoprotein alterations in

normolipidemic male smokers. Clin Chim Acta 2001;305:

99–105.

[23] Parks EJ, Dare D, Frazier KB, Hellerstein MK, Neese RA,

Hughes E, et al. Dependence of plasma a-tocopherol flux on

very low-density triglyceride clearance in humans. Free Radic

Biol Med 2000;29:1151–9.