Embed Size (px)
Citation preview
A
K
I
petfofhs
2
0d
Applied nutritional investigation
Serum selenoprotein-P levels in patients with inflammatory boweldisease
Akira Andoh, M.D.a,*, Masaki Hirashima, Ph.D.b, Hiroaki Maeda, Ph.D.b,Kazunori Hata, M.D.a, Osamu Inatomi, M.D.a, Tomoyuki Tsujikawa, M.D.a,
Masaya Sasaki, M.D.a, Kazuhiko Takahashi, Ph.D.c, and Yoshihide Fujiyama, M.D.a
a Department of Internal Medicine, Shiga University of Medical Science, Otsu, Japanb The Chemo-Sero-Therapeutic Research Institute, Kumamoto, Japan
c Department of Hygienic Chemistry, Graduate School of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan
Manuscript received May 5, 2004; accepted August 23, 2004.
bstract Objective: Selenoprotein-P is a selenium-rich serum protein that carries more than 50% of serumselenium. We evaluated changes in serum selenoprotein-P levels in patients with inflammatorybowel disease.Methods: Serum selenoprotein-P levels were measured by enzyme-linked immunosorbent assay.Twenty healthy individuals (controls), 34 patients with ulcerative colitis, and 37 patients withCrohn’s disease (CD) were studied.Results: A highly significant correlation was found between the serum selenium and selenopro-tein-P levels. There was no significant difference in serum selenoprotein-P levels between healthycontrols (average 3.4 � 0.8 �g/mL, n � 20) and patients with ulcerative colitis (3.0 � 1.0 �g/mL,n � 34). Serum selenoprotein-P levels were significantly lower in patients with CD (average 1.8 �0.5 �g/mL, n � 37). Serum selenoprotein-P levels were significantly lower in the elemental dietgroup of patients who had CD (average 1.4 � 0.4 �g/mL, n � 17) than in the non-elemental dietgroup of patients who had CD (average 2.1 � 0.3 �g/mL, n � 20).Conclusion: We found that the serum selenoprotein-P level is decreased in patients with CD. It maybe a useful marker to monitor the systemic selenium status in various disorders. © 2005 ElsevierInc. All rights reserved.
Nutrition 21 (2005) 574–579www.elsevier.com/locate/nut
eywords: Inflammatory bowel disease; Antioxidant; Selenium; Enzyme-linked immunosorbent assay
tttpegpoa
oop
ntroduction
Selenium (Se) is an essential micronutrient and is incor-orated into proteins as the Se-containing amino acids sel-nocysteine (Sec) and selenomethionine [1,2]. Selenopro-ein is a Sec-containing protein and can be distinguishedrom proteins that non-specifically incorporate selenomethi-nine [3,4]. More than 15 selenoproteins have been found:our types of glutathione peroxidase, three types of thyroidormone deiodinate, three types of thioredoxin reductase,elenophosphate synthetase, and selenoprotein-P [3,4].
* Corresponding author. Tel.: �81-77-548-2217; fax: �81-77-548-219.
hE-mail address: [email protected] (A. Andoh).
899-9007/05/$ – see front matter © 2005 Elsevier Inc. All rights reserved.oi:10.1016/j.nut.2004.08.025
Selenoprotein-P is the most peculiar selenoprotein. It con-ains 10 Sec residues per molecule [5,6] and is a plasma proteinhat carries more than 50% of plasma Se [3–6]. Selenopro-ein-P is as an important Se supplier from the liver to theeripheral tissues [3,7]. More importantly, several lines ofvidence suggest that selenoprotein-P is a free radical scaven-er for peroxynitrite [8,9]. Selenoprotein-P also reduces phos-holipid-hydroperoxide in the presence of glutathione or thi-redoxin [10]. Thus, selenoprotein-P is an important serumntioxidant factor and protects cells against oxidative stress.
In humans severe Se deficiency has been linked to the devel-pment of Keshan’s disease [11], a dilated congestive cardiomy-pathy that primarily occurs in children living in China. Someathologic conditions of Se deficiency, such as cancer, coronary
eart disease, and liver necrosis, are also thought to be associated
wceaaoesapIsw
pitpc
M
C
mcu(
(GMhUd
aawrwtptaop
([
ia
A
s
TB
MSADB5CT
ECA
T
C
d
575A. Andoh et al. / Nutrition 21 (2005) 574–579
ith a decrease in Se levels [12]. Several cases of myopathy andardiomyopathy in Se-depleted patients receiving long-term par-nteral nutrition have also been reported [13–17]. Because Se isbsorbed throughout the small intestine, mainly in the duodenumnd proximal jejunum [18], extremely low blood Se levels with nobvious clinical symptoms have been found in patients with dis-ases of the small intestine, such as Crohn’s disease (CD) andhort bowel syndrome after intestinal resection [19]. Due to mal-bsorption, decreased plasma Se levels also have been reported inatients with CD who did not receive intestinal resection [20–23].n Japan, long-term elemental diet (ED), which contains a verymall amount of Se (1.6 �g/1000 kcal), is one factor associatedith decreased plasma Se level in patients who have CD [24–27].
Based on these reports, we hypothesized that serum seleno-rotein-P levels might be decreased in patients who havenflammatory bowel disease. We assessed serum selenopro-ein-P levels as a marker of nutritional and antioxidant states inatients who had inflammatory bowel disease by using a re-ently developed enzyme-linked immunosorbent assay [28].
aterials and Methods
ontrols and patients
Twenty healthy control subjects (10 women and 10 men,edian age 33 y), 34 in- and outpatients with ulcerative
olitis (UC; 17 women and 17 men, median age 28 y; 5 hadndergone colectomy), and 37 in- and outpatients with CD
able 1aseline clinical characteristics*
Control (n � 20)
en/women 10/10mokers 2ge (y) 32.4 � 6.5isease duration (y) —MI (kg/m2) 23.8 � 2.8-ASA treatment —orticosteroid use —ype of CDIleal —Ileocolonic —Colonic —
D over 1 y —DAI � 150 —verage CDAIED groupNon-ED group
ype of UCProctitis —Left-side colitis —Pancolitis —
AI � 5 —
5-ASA, 5-aminosalicylic acid; BMI, body mass index; CAI, clinical acisease; CDAI, Crohn’s disease activity index according to Best et al. [29
* Data are presented as mean � standard deviation.
20 women and 17 men, median age 29 y) were studied r
Table 1). All patients were managed in the Division ofastroenterology at the Hospital of the Shiga University ofedical Science (Otsu, Japan). Baseline characteristics of
ealthy controls were matched to those of patients who hadC and those who had CD. All patients had an establishediagnosis by radiologic, histologic, and clinical criteria.
Eighteen patients with CD had high disease activity (CDctivity index � 150) [29], and 19 patients had low diseasectivity (CD activity index � 150). In this study, patientshose CD activity index score was above 150 points were
egarded as in active phase. Seventeen patients with CDere placed on an ED with Elental (Ajinomoto Pharmaceu-
ical Ltd., Tokyo, Japan; �1200 kcal/d for �1 y) as theirrimary therapy [23,24], and these patients were regarded ashe ED group. The formula of Elental contains a very smallmount of Se. Other patients with CD ingested normal foodr normal food plus less ED. No patients received Se sup-lementation.
Twelve patients who had UC had high disease activityclinical activity index � 5) as described by Lichtiger et al.30], and these patients were regarded as in active phase.
Blood was sampled with patients’ informed consent,mmediately separated to serum by centrifugation, and keptt �80°C until use.
nalytical methods
Se concentrations were determined by atomic absorptionpectroscopy. The coefficient of variation for Se in the
UC (n � 34) CD (n � 37)
17/17 17/204 4
34.4 � 14.5 31.4 � 6.58.7 � 7.5 8.5 � 6.8
22.3 � 1.8 21.3 � 2.831 3518 16
— 5— 26— 6— 17— 18
194 � 63163 � 26
9 —16 —9 —
12 —
ndex of ulcerative colitis according to Lichtiger et al. [30]; CD, Crohn’selemental diet; UC, ulcerative colitis
tivity i]; ED,
eference serum between series was 2.6%. Other laboratory
dt
vwpmbOtarwhLbaI1m
S
Ksr
R
e
f0
i(pspcdt�Ut�
FpSi
Fnd
576 A. Andoh et al. / Nutrition 21 (2005) 574–579
ata were measured by routine methods at the Hospital ofhe Shiga University of Medical Science.
Serum selenoprotein-P levels were determined by a pre-iously described enzyme-linked immunosorbent assayith some modifications [28]. Ninety-six–well microtiterlates were coated with rat anti-human selenoprotein-Ponoclonal antibody (BD1). Wells were washed and incu-
ated with Block Ace (Dainippon Pharmaceutical Co. Ltd.,saka, Japan) at 37°C for 1 h. After washing, selenopro-
ein-P standards and serum sample were added to each wellnd incubated at 37°C for 1 h. After washing, a biotinylatedat anti-human selenoprotein-P monoclonal antibody (AA3)as added and incubated at 37°C for 1 h. After washing,orseradish peroxidase conjugated to streptavidin (Vectoraboratories, Burlingame, CA, USA) was added and incu-ated for 1 h. After washing, color development waschieved with TMB Microwell Peroxidase Substrate (KPL,nc., Gaithersburg, MD, USA). The reaction was stopped by
N of sulfuric acid, and absorbances at 450 nm wereeasured.
tatistical analysis
Differences between groups were analyzed with theruskal-Wallis test. P � 0.05 was considered statistically
ignificant. Correlations were investigated with Spearman’sank correlation test.
esults
The relation between serum Se and selenoprotein-P lev-
ig. 1. Correlation between the serum levels of selenoprotein-P and sele-ium (n � 70). Serum selenoprotein-P levels were measured by a recentlyeveloped enzyme-linked immunosorbent assay [30].
ls is shown in Fig. 1. A highly significant correlation was d
ound between serum Se and selenoprotein-P levels (r �.791, P � 0.001, n � 70).
As shown in Fig. 2A, there was no significant differencen serum selenoprotein-P levels between healthy controlsmean � standard deviation 3.4 � 0.8 �g/mL, n � 20) andatients with UC (3.0 � 1.0 �g/mL, n � 34). However,erum selenoprotein-P levels were significantly lower inatients with CD (1.8 � 0.5 �g/mL, n � 37) than in healthyontrols or patients with UC. There were no significantifferences in serum selenoprotein-P levels between pa-ients with active UC and those with inactive UC (3.2 � 1.2g/mL for active UC, n � 12; 2.8 � 0.8 �g/mL for inactiveC, n � 22). Among patients with CD, serum selenopro-
ein-P levels were lower in those with active CD (1.7 � 0.5g/mL, n � 18) than in those with inactive CD (1.9 � 0.5
ig. 2. (A) Serum selenoprotein-P levels in healthy controls (n � 20) andatients with inflammatory bowel disease (UC, n � 34; CD n � 37). (B)erum selenium levels in healthy controls (n � 20) and patients with
nflammatory bowel disease (UC, n � 34; CD, n � 37). CD, Crohn’s
isease; N.S., not significant; UC, ulcerative colitis.
�(s1F
S(p�twc(
stwnF
eatgE
D
t
F(p
FCwaC
577A. Andoh et al. / Nutrition 21 (2005) 574–579
g/mL, n � 19), but there was no significant differenceFig. 3A). In contrast, serum selenoprotein-P levels wereignificantly lower in the ED group (1.4 � 0.4 �g/mL, n �7) than in the non-ED group (2.1 � 0.3 �g/mL, n � 20;ig. 4A).
Similar results were observed for serum Se levels. Serume concentrations are shown in Fig. 2B. In healthy controlsn � 20), serum Se concentration was 13.2 � 2.1 �g/dL. Inatients with UC (n � 34), serum Se concentration was 13.3
2.6 �g/dL, and there was no significant difference be-ween healthy controls and patients with UC. In all patientsith CD (n � 37), serum Se level was significantly de-
reased (9.2 � 2.3 �g/dL) compared with healthy controls
ig. 3. (A) Serum selenoprotein-P levels in patients with active and inactiveD (active, n � 18; inactive, n � 20). Patients with CD (CDAI � 150 points)ere regarded as in active phase. (B) Serum selenium levels in patients with
ctive and inactive CD (active, n � 18; inactive, n � 20). CD, Crohn’s disease;DAI, Crohn’s disease activity index; N.S., not significant.
P � 0.01) and patients with UC (P � 0.01). There were no E
ignificant differences between patients with active CD andhose with inactive CD (Fig. 3B). Serum Se concentrationas significantly lower in the ED group (7.8 � 2.0 �g/dL,� 17) than the non-ED group (10.5 � 1.8 �g/dL, n � 20;ig. 4B).
In patients who had CD, decreased selenoprotein-P lev-ls were not correlated with serum total protein levelsnd/or serum albumin levels (serum selenoprotein-P versusotal protein, r � 0.10 in ED group, r � 0.13 in non-EDroup; serum selenoprotein-P versus albumin, r � 0.14 inD group, r � 0.22 in non-ED group).
iscussion
Recognition of the importance of selenoprotein-P has ledo interest in this serum factor. However, changes in serum
ig. 4. (A) Serum selenoprotein-P levels in patients with Crohn’s diseaseED group, n � 17; non-ED group, n � 20). (B) Serum selenium levels inatients with Crohn’s disease (ED group, n � 17; non-ED group, n � 20).
D, elemental diet.
smpTsdr
s0osssls6tpt[adElstmmeta
wultmsssstilsaSrteeTCsi
twflAmSpo
R
[
[
[
[
[
[
[
578 A. Andoh et al. / Nutrition 21 (2005) 574–579
elenoprotein-P levels in various pathologic conditions re-ain unclear. Only one report has described plasma seleno-
rotein-P levels in patients with hypercholesterolemia [28].o our knowledge, this is the first study to report serumelenoprotein-P levels in patients with inflammatory bowelisease, in whom decreased plasma Se levels have beeneported by several clinical studies [19–23].
We found a statistically significant correlation betweenerum selenoprotein-P levels and plasma Se levels (r �.791, P � 0.001, n � 70). This means that a determinationf plasma selenoprotein-P levels can be used to assess Setatus. Previous studies have reported that determination oferum glutathione peroxidase activity is useful to assess Setatus [18,19,21]. However, Se incorporated into extracel-ular glutathione peroxidase represents less than 20% oferum Se [7]. In contrast, selenoprotein-P carries more than0% of plasma Se and is regarded as the major selenopro-ein in serum [7]. Selenoprotein-P contains 10 Sec residueser molecule, whereas other selenoproteins such as gluta-hione peroxidase contain one Sec residue per molecule5,6]. Further, selenoprotein-P exerts strong antioxidativectivity against water-insoluble oxidants such as lipid hy-roperoxide [10] and is stronger than glutathione or vitamin
[31]. Thus, a determination of serum selenoprotein-Pevels may more accurately reflect the serum Se status anderum antioxidant activity in the body. Moreover, in con-rast to Se measurement, another advantage is that environ-ental contamination is of no concern in selenoprotein-Peasurement by enzyme-linked immunosorbent assay. For
xample, determination of Se by atomic absorption spec-roscopy is sensitive and is affected by adhesion of smallmounts of Se on sampling tubes or laboratory instruments.
In the present study, serum selenoprotein-P and Se levelsere significantly decreased in patients with CD. In partic-lar, serum selenoprotein-P and Se levels were significantlyower in the ED group than the non-ED group. The liver ishe major organ of selenoprotein-P synthesis, but the preciseolecular mechanism that regulates hepatic selenoprotein-P
ynthesis is unknown. One possibility is that decreasedelenoprotein-P may be related to decreased protein synthe-is in the liver. However, we found no correlation betweenerum selenoprotein-P level and serum albumin level, andhis suggests that a decreased level of serum selenoprotein-Ps not associated with decreased protein synthesis in theiver. The more likely possibility is that the decrease inelenoprotein-P levels in patients with CD is associated withmalabsorption of Se in the small intestine and low serume levels, although it remains unclear how serum Se levelsegulate hepatic selenoprotein-P synthesis. The formula ofhe ED, which contains a very small amount of Se, mightmphasize the decrease in serum Se levels [27] and maynhance the decrease in the serum selenoprotein-P levels.hus, Se supplement is recommended to patients who haveD and use ED long term, and determination of serum
elenoprotein-P level might be a suitable marker of Se status
n such patients.In conclusion, we have demonstrated for the first timehat serum selenoprotein-P level is decreased in patientsith CD. Although in vitro studies have shown various roles
or selenoprotein-P in the body, the clinical importance ofow selenoprotein-P levels requires further investigation.n increasing number of reports has associated clinicalanifestations of Se deficiency with Se-free nutrition or ae-free diet. The accumulation of clinical data on seleno-rotein-P in such cases will clarify the clinical significancef determination of serum selenoprotein-P levels.
eferences
[1] Brown KM, Arthur JR. Selenium, selenoproteins and human health:a review. Public Health Nutr 2001;4:593–9.
[2] Kohrle J, Brigelius-Flohe R, Bock A, Gartner R, Meyer O, Flohe L.Selenium in biology: facts and medical perspectives. Biol Chem2000;381:849.
[3] Schomburg L, Schweizer U, Holtmann B, Flohe L, Sendtner M,Kohrle J. Gene disruption discloses role of selenoprotein P in sele-nium delivery to target tissues. Biochem J 2003;370(pt 2):397–402.
[4] Schweizer U, Michaelis M, Kohrle J, Schomburg L. Efficient sele-nium transfer from mother to offspring in selenoprotein-P–deficientmice enables dose-dependent rescue of phenotypes associated withselenium deficiency. Biochem J 2004;378(pt 1):21.
[5] Hill KE, Lloyd RS, Yang JG, Read R, Burk RF. The cDNA for ratselenoprotein P contains 10 TGA codons in the open reading frame.J Biol Chem 1991;266:10050.
[6] Hill KE, Lloyd RS, Burk RF. Conserved nucleotide sequences in theopen reading frame and 3= untranslated region of selenoprotein PmRNA. Proc Natl Acad Sci USA 1993;90:537.
[7] Saito Y, Takahashi K. Characterization of selenoprotein P as a sele-nium supply protein. Eur J Biochem 2002;269:5746.
[8] Arteel GE, Mostert V, Oubrahim H, Briviba K, Abel J, Sies H.Protection by selenoprotein P in human plasma against peroxynitrite-mediated oxidation and nitration. Biol Chem 1998;379:1201.
[9] Takebe G, Yarimizu J, Saito Y, Hayashi T, Nakamura H, Yodoi J, etal. A comparative study on the hydroperoxide and thiol specificity ofthe glutathione peroxidase family and selenoprotein P. J Biol Chem2002;277:41254.
10] Saito Y, Hayashi T, Tanaka A, Watanabe Y, Suzuki M, Saito E,Takahashi K. Selenoprotein P in human plasma as an extracellularphospholipid hydroperoxide glutathione peroxidase. Isolation andenzymatic characterization of human selenoprotein p. J Biol Chem1999;274:2866.
11] Keshan Disease research Group of the Chinese Academy of MedicalScience. Epidemiological studies in etiologic relationship of seleniumand keshan disease. Chin Med J 1979;92:472.
12] Allan CB, Lacourciere GM, Stadmann TC. Responsiveness of sel-enoproteins to dietary selenium. Annu Rev Nutr 1999;19:1.
13] Kien CL, Ganther HE. Manifestations of chronic selenium deficiencyin a child receiving total parenteral nutrition. Am J Clin Nutr 1983;37:319.
14] Watson RD, Cannon RA, Kurland GS, Cox KL, Frates RC. Seleniumresponsive myositis during prolonged home total parenteral nutritionin cystic fibrosis. JPEN 1985;9:58.
15] Brown MR, Cohen HJ, Lyons JM, Curtis TW, Thunberg B, CochranWJ, Klish WJ. Proximal muscle weakness and selenium deficiencyassociated with long term parenteral nutrition. Am J Clin Nutr 1986;43:549.
16] Johnson RA, Baker SS, Fallon JT, Maynard III, EP Ruskin JN, WenZ, et al. An occidental case of cardiomyopathy and selenium defi-
ciency. N Engl J Med 1981;304:1210.
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
579A. Andoh et al. / Nutrition 21 (2005) 574–579
17] Fleming CR, Lie JT, McCall JT, O’Brien JF, Baillie EE, Thistle JL.Selenium deficiency and fatal cardiomyopathy in a patient on homeparenteral nutrition. Gastroenterology 1982;83:689.
18] Scharrer E, Senn E, Wolffram S. Stimulation of mucosal uptake ofselenium from selenite by some thiols at various sites of rat intestine.Biol Trace Elem Res 1992;33:109.
19] Rannem T, Ladefoged K, Hylander E, Hegnhoj J, Staun M. Seleniumdepletion in patients with gastrointestinal diseases: are there anypredictive factors? Scand J Gastroenterol 1998;33:1057.
20] Thomas AG, Miller V, Shenkin A, Fell GS, Taylor F. Selenium andglutathione peroxidase status in paediatric health and gastrointestinaldisease. J Pediatr Gastroenterol Nutr 1994;19:213.
21] Rannem T, Ladefoged K, Hylander E, Hegnhoj J, Jarnum S. Seleniumstatus in patients with Crohn’s disease. Am J Clin Nutr 1992;56:933.
22] Hatanaka N, Nakaden H, Yamamoto Y, Matsuo S, Fujikawa T,Matsusue S. Selenium kinetics and changes in glutathione peroxidaseactivities in patients receiving long-term parenteral nutrition andeffects of supplementation with selenite. Nutrition 2000;16:22.
23] Ringstad J, Kildebo S, Thomassen Y. Serum selenium, copper, andzinc concentrations in Crohn’s disease and ulcerative colitis. Scand J
Gastroenterol 1993;28:605.24] Tsujikawa T, Andoh A, Fujiyama Y. Enteral and parenteral nutritiontherapy for Crohn’s disease. Curr Pharm Des 2003;9:323.
25] Ikeuchi H, Kusunoki M, Yanagi H, Yamamura T, Fukuda Y, Shi-moyama T. Effects of elemental diet (ED) on surgical treatment inCrohn’s disease. Hepatogastroenterology 2000;47:390.
26] Hiwatashi N. Enteral nutrition for Crohn’s disease in Japan. DisColon Rectum 1997;40(suppl):S48–53.
27] Kuroki F, Matsumoto T, Iida M. selenium is depleted in Crohn’sdisease on enteral nutrition. Dig Dis 2003;21:266.
28] Saito Y, Watanabe Y, Saito E, Honjoh T, Takahashi K. Productionand application of monoclonal antibodies to human selenoprotein P.J Health Sci 2001;47:346.
29] Best WR, Becktel JM, Singleton JW, Kern Jr F. Development of aCrohn’s disease activity index. National Cooperative Crohn’s DiseaseStudy. Gastroenterology 1976;70:439–44.
30] Lichtiger S, Present DH, Kornbluth A, Gelernt I, Bauer J, Galler G,et al. Cyclosporine in severe ulcerative colitis refractory to steroidtherapy. N Engl J Med 1994;330:1841.
31] Hirashima M, Naruse T, Maeda H, Nozaki C, Saito Y, Takahashi K.Identification of selenoprotein P fragments as a cell-death inhibitory
factor. Biol Pharm Bull 2003;26:794.
