Age Effect on Human Aortic Valvular Glycoproteins

Archives of Medical Research 38 (2007) 495e502

ORIGINAL ARTICLE

Age Effect on Human Aortic Valvular Glycoproteins

Ma1gorzata Przyby1o,a Ewa Stxpie�n,b Roman Pfitzner,b Anna Lity�nska,a and Jerzy Sadowskib

aDepartment of Glycoconjugate Biochemistry, Institute of Zoology, Jagiellonian University, Krakow, PolandbDepartment of Heart and Cardiovascular Surgery, Institute of Cardiology, Medical College, Jagiellonian University, Krakow, Poland

Received for publication October 5, 2006; accepted February 2, 2007 (ARCMED-D-06-00433).

Background. The aortic valve has been the subject of many hemodynamic studies but, toour knowledge, posttranslational modification of human valve proteins has not yet beenstudied. Thus, the aim of this study was to determine whether any age-related changes inthe protein composition of normal human aortic valves and their glycosylation patterncould be observed.

Methods. Aortic valves harvested from male cadaveric donors free of cardiovascular dis-eases were divided into four age groups: I, mean age 21 years; II, 30 years; III, 41 years;IV, 51 years. Proteins were separated by SDS-PAGE and transferred to PVDF mem-branes. Identification of monosaccharide moieties or oligosaccharide units was performedwith the use of eight lectins of narrow specificity: Galantus nivalis agglutinin, Sambucusnigra agglutinin, Maackia amurensis agglutinin, Datura stramonium agglutinin, Aleuriaaurantia agglutinin, Arachis hypogeae agglutinin, Phaseolus vulgaris agglutinin, andLycopersicon esculentum agglutinin.

Results. Isolated proteins showed no age-related changes in SDS-PAGE protein profile,contrary to their glycosylation. Protein sialylation, number of tri/tetraantennary complexglycans, proteins having terminal galactose and polylactosaminyl units increased withage, whereas protein fucosylation showed the opposite relationship. Moreover, groupsIII and IV possessed a larger number of proteins bearing high-mannose and/or hybrid-type glycans, and the quantity of these structures seemed to change, in particular proteins,with the age of donors.

Conclusions. Our results clearly demonstrate that glycosylation profile in human aorticproteins is associated with the age of the donor. � 2007 IMSS. Published by ElsevierInc.

Key Words: Age-associated changes, Aortic valve, Glycoproteins, Heart, Human, Lectins.

Introduction

Aortic valve leaflets are complex, multilayered arrays ofcells and extracellular components of connective tissueand are covered by a continuous monolayer of endothelialcells (1,2). Cell�cell and/or cell�matrix interactions areessential in the physiological function of valves; therefore,it is likely that interstitial cells influence cardiac valve dy-namic properties and function (3). In the cardiovascular

Address reprint requests to: Ma1gorzata Przyby1o, PhD, Department of

Glycoconjugate Biochemistry, Institute of Zoology, Jagiellonian Univer-

sity, Ingardena 6, 30-060 Krakow, Poland; E-mail: kloc@zuk.iz.uj.edu.pl

0188-4409/07 $esee front matter. Copyright � 2007 IMSS. Published by Eldoi: 10.1016/j.arcmed.2007.02.001

system the progressive shift from young age to senescenceis characterized by structural and functional changes in thecardiac extracellular matrix, which supports and alignsmyocardial cells and blood vessels and maintains myocar-dial mass, structure and function (4). It has also been shownthat advancing age was associated with greater collagen,fibronectin, a1b1 and a5b1 integrin content in rat heart,suggesting that these matrix proteins and their receptors un-dergo coordinated regulation in the aging heart (5,6). More-over, in human valve leaflets, the number of myofibroblastsdecreases with age and is accompanied by a degenerationof collagen fibers (7). To our knowledge, the structuralcharacterization of human valve glycoproteins has not been

sevier Inc.

496 Przyby1o et al./ Archives of Medical Research 38 (2007) 495e502

carried out and it is not clear whether aging mechanismscan change the glycoprotein pattern of valves. Thus, thepresent study was undertaken to determine whether anyage-related changes in the protein composition of normalhuman aortic valves as well as their glycosylation patterncould be observed.

Materials and Methods

Chemicals

Digoxigenin (DIG)-labeled lectins Galanthus nivalis agglu-tinin (GNA), Sambucus nigra agglutinin (SNA), Maackiaamurensis agglutinin (MAA), Datura stramonium aggluti-nin (DSA), Aleuria aurantia agglutinin (AAA) and Arachishypogeae (peanut) agglutinin (PNA) were purchased fromRoche Biochemicals (Indianapolis, IN), as were anti-DIG-alkaline phosphate conjugate, 4-nitroblue-tetrazolium saltand 5-bromo-4-chloro-3-indolylphosphate solution. Biotin-ylated Phaseolus vulgaris agglutinin (PHA-L) andLycopersicon esculentum (tomato) agglutinin (TL), avidinalkaline phosphatase conjugate, proteinases inhibitorcocktail (P 2714), N-acetyl-D-glucosamine, methyl-a-D-mannopyranoside, fucose and high molecular weightstandards were from Sigma Chemical Co. (St. Louis,MO). Biotinylated-succinylated Lycopersicon esculentum(tomato) agglutinin (TL) was from Vector Laboratories(Burlingame, CA). Polyvinylidene difluoride (PVDF) mem-branes were from Millipore (Bedford, MA). N-acetylneura-minic acid and galactose were from Serva (Heidelberg,Germany). All remaining chemicals were analytical grade.

Aortic Valve Specimens

Twenty aortic valves, together with ascending aorta andsurrounding tissues, were harvested at the Department ofForensic Medicine, Jagiellonian University, during autop-sies of selected, previously healthy males who had sufferedtraumatic death. All donors were without cardiovasculardiseases and without severe metabolic disturbances suchas diabetes and renal failure. All specimens were preparedby the Valve and Tissue Bank, Department of Cardiovascu-lar Surgery and Transplantology, Jagiellonian Universityfollowing the protocol of cardiovascular tissue banking es-tablished by European Homograft Bank, c/o Military Hos-pital Queen Astrid, Brussels. Valves were not used forimplantation because their parameters did not meet the cri-teria. After preparation, homografts were sterilized inParker Medium 199 supplemented with 25% calf serum,to which the following antibiotics were added: cefuroximesodium salt 0.24 mg/mL, colistin 1000 IU/mL, amphoteri-cin B 0.025 mg/mL, vancomycin 0.05 mg/mL, doxycyclinehydrochlorate 0.2 mg/mL. This was followed by conserva-tion in Parker medium 199 containing 40% calf serum and10% DMSO to prevent cell damage due to water crystalli-

zation. Then valves were placed into plastic packages, fro-zen and stored at the temperature of liquid nitrogen vapor(�140�C) until testing as biovital specimens. Valves weredivided into four groups depending on the age of donors:group I, mean age 21 years (range 20e21 years); groupII, mean age 30 years (range 30e31 years); group III, meanage 41 years (range 39e42 years); and group IV, mean age51 years (range 47e54 years).

Tissue Preparation

Protein samples were prepared from the aortic valve wholeleaflets (for each valve separately). To extract proteins, eachfrozen leaflet was shattered into pieces and then sonicatedon ice for 15 sec in 50 mM Tris-HCl buffer, pH 7.5, con-taining 5 mM MgCl2, 0.2 mM EDTA, 1 mM PMSF andprotease inhibitors (Sigma) (8). Homogenate was centri-fuged at 80,000 � g for 1 hr at 4�C, and the supernatantwas collected and stored at �70�C.

Gel Electrophoresis and Transfer

Bradford (9) protein assay was performed on the superna-tant to determine the total protein concentration of the ho-mogenates. For electrophoresis, 50 mg of protein mixedwith sample buffer and heated were separated on 12.6%SDS-polyacrylamide gels in reducing condition accordingto the method of Laemmli (10). Following separation, theproteins were transferred to a PVDF membrane using a wet-blotter in transfer buffer (25 mM Tris and 192 mM glycinein 20% methanol, pH 8.4) for 60 min at 100 V with cooling.Alternatively, gels were stained with Coomassie BrilliantBlue R-250. Gels were calibrated for molecular weightdetermination using the Sigma protein standard kit forelectrophoresis in SDS.

Glycan Chain Analysis

Glycan chain analysis of proteins was performed with theuse of a Glycan Differentiation Kit as previously describedin detail (11). Glycan chain analysis results were electron-ically captured with the use of an imaging device and thendensitometrically traced (UVIMAP v. 99 Software; SyngenBiotech, Wroclaw, Poland).

Statistics

Significance level for differences between mean values wascomputed using Duncan’s new multiple range test; p !0.05was considered statistically significant.

Results

Under electrophoretic conditions performed in this investi-gation, proteins from aortic valve extracts were separatedinto bands ranging from 205 to 35 kDa with two additional

497Age Effect on Human Aortic Valvular Glycoproteins

Figure 1. Densitometric measurement of the intensity of Coomassie Brilliant Blue R-250 staining of human aortic valve proteins separated by SDS-PAGE on

12.6% gel from different age groups: group I (mean age 21 years), group II (mean age 30 years), group III (mean age 41 years), group IV (mean age 51 years).

X-axis, proteins arranged according to their molecular weight (MW), Y-axis, volume: sum of all intensities included under a peak. Values are expressed as

mean � standard deviation (SD) of the results obtained for each valve. The differences in Coomassie Brilliant Blue staining of a particular MW among age

groups are not statistically significant (Duncan’s new multiple range test). Arrow indicates lack of 35 kDa protein in group I.

bands of 8 kDa and 5 kDa as revealed by Coomassie Bril-liant Blue staining (not shown). In all examined age groups,protein patterns were almost similar, judging from the pro-files of protein composition in each experimental groupgenerated by the densitometric scanning technique(Figure 1), except for the 35-kDa protein that was notstained in group I.

In the present study, eight lectins of well-known andnarrow specificity were used to gain information on glyco-sylation of aortic valve proteins in relation to the age ofdonors (Figures 2 and 3). GNA is a lectin specific forterminal-bound mannose a1e2, a1e3 and a1�6 to man-nose, and GNA-reactive glycoproteins were detected inmultiple bands distributed between an apparent molecularmass ranging from 205 to 36 kDa. The binding resultsshowed that groups III and IV possessed larger numbersof proteins bearing high-mannose and/or hybrid-type gly-cans, and the quantity of these structures seemed to change,in particular, proteins, with the age of donors. In the case ofsialylated oligosaccharides as detected by MAA and SNAlectins, which interact with oligosaccharides terminatingwith Neu5Ac a2e3 Gal and Neu5Ac a2e6 Gal groups, re-spectively, a2e6-sialylation predominated over a2e3-sialylation in all age groups. SNA showed a positivereaction with numerous glycoproteins with molecularmasses ranging from 142.5 to 42 kDa (Figures 2 and 3),but only three proteins of 73, 60 and 49.5 kDa had glycanswith a2e3-linked sialic acids (not shown). Moreover, thepattern of reaction of aortic valve proteins with MAA wasalmost identical in all age groups. On the contrary, the num-ber of proteins bearing a2e6-linked sialic acids (positivereaction with SNA) as well as intensity of staining ofa few proteins seemed to increase with the age of donors.Analysis of AAA lectin staining (a lectin specific for fu-

cose) showed that the number of AAA-reactive proteinsas well as the magnitude of this reaction reached the max-imum in groups II and III, and declined thereafter (Figures2 and 3). The opposite relationship was observed for TLlectin specific for polylactosaminyl units, where the numberof reactive proteins and the intensity of staining weregreater in older donors (Figures 2 and 3). Similar phenom-ena were observed for PHA-L lectin, which recognizeda1e6 branches in tri-/tetraantennary complex-type glycans(Figures 2 and 3). As shown in Figures 2 and 3, all agegroups showed DSA-reactive proteins with molecularmasses ranging from 94 to 39 kDa. DSA is a lectin specificfor galactose a1e4 linked to N-acetylglucosamine. Thepattern of reaction was almost identical in all age groups,except for 76- and 39-kDa proteins, which were visible onlyin groups III and IV. No reaction with PNA, a lectin specificfor the core disaccharide galactose a1e3 N-acetylgalactos-amine of O-glycans was observed (not shown).

Discussion

The topic of our study was to define the glycosylation pro-file of human aortic valve proteins in relation to age, whichhas not been previously elucidated. We therefore investi-gated the oligosaccharide structures of the glycoproteinswith the use of plant lectins with specific affinities for sugarmoieties. Our results clearly demonstrated that glycosyla-tion profile in human aortic valve proteins is associatedwith the age of donors, particularly at the level of proteinsialylation, number of tri-/tetraantennary complex glycans,proteins having terminal galactose and polylactosaminylunits, which have increased with age, as well as proteinfucosylation, which has shown an opposite relationship.

498 Przyby1o et al./ Archives of Medical Research 38 (2007) 495e502

Figure 2. Representative blots showing reactivity of specific lectins with human aortic valve proteins from different age groups. After electrophoresis, pro-

teins were transferred onto a PVDF membrane and stained with lectins. GNA, SNA, DSA, AAA, PHA-L, TL, acronyms for lectins, see Chemicals under

Materials and Methods. Lane S, molecular weight standards; Lane I, group I; Lane II, group II; Lane III, group III; Lane IV, group IV.

Moreover, we observed quantitative and qualitative changesin high-mannose/hybrid-type oligosaccharides. It should beemphasized that, to our knowledge, studies on glycosyla-tion profile of human aortic valves with respect to age havenot been previously studied, even in animal models. There-fore, our study is the first to highlight the differencesbetween human aortic valves at different ages with respectto glycosylation profile.

Several studies have demonstrated that physiological ag-ing is accompanied by significant cardiovascular modifica-tions, both structural and functional and that the net effectof aging in the cardiovascular system is a loss of elasticityand distensibility (7,12e14). Among structural alternationsin heart, the following changes are usually observed: a lossof myocytes with subsequent hypertrophy of the remainingcells, left ventricular hypertrophy mostly due to an increase

499Age Effect on Human Aortic Valvular Glycoproteins

Figure 3. Densitometric measurement of the intensity of lectin binding of human aortic valve proteins from different age groups. GNA, SNA, DSA, AAA,

PHA-L, TL, acronyms for lectins; see Chemicals under Materials and Methods. X-axis, alternatively: age groups (I, II, III, and IV), or molecular weight of

proteins (MW in kDa) reacting with a particular lectin. Y-axis, volume: sum of all intensities included under a peak. Values are expressed as mean � standard

deviation (SD) of the results obtained for each valve. Statistical analysis was carried out using Duncan’s new multiple range test. In each case, SD bars do not

overlap, and the differences in lectin staining of a particular molecular weight among age groups are statistically significant.

in average myocyte size, an increase in size of the leftatrium, increased amounts of collagen, fibrous tissue andlipofuscin, a change in the physical properties of collagenpurportedly due to nonenzymatic cross-linking. Moreover,the circumference of all four cardiac valves increases, withthe greatest change observed in the aortic valve, and thevalve cusps undergo thickening and become fibrotic along

their appositional surface, whereas annuli are the sites ofcollagen degradation, lipid accumulation, and calcification(14,15). Reduction of elastic properties may in turn pro-mote the development of degenerative disease of the trileaf-let aortic valve. Simulation of age-related aortic wallstiffening in rabbits has resulted in changes in their micro-structure (16). As revealed by immunoconfocal microscopy,

500 Przyby1o et al./ Archives of Medical Research 38 (2007) 495e502

Figure 3. (Continued)

increase of the length and area of the leaflet, as well as re-duction of the collagen density, were observed. Moreover,upregulation of matrix metalloproteinases (MMPs) andangiotensin-converting enzyme (ACE) levels in rabbit andmice aged valves might indicate constant dynamic turnoverof extracellular matrix component (16,17).

In our studies, the most remarkable changes in glycosyl-ation of the human aortic valve proteins seemed to appearat the age of 40 years, as the lectin-staining profiles ob-

tained for groups I and II were in general alike and quitedifferent from the lectin-staining profiles for groups IIIand IV. In line with our findings are also other reports thathave confirmed that in humans some alternations in cardio-vascular system are closely associated with age 40 years ormore. Lin and co-workers (18) have recently demonstratedthat the prevalence of the thickened aortic valves increasesprogressively with age, but it was not observed in humans!40 years old. In addition, Christie and Barratt-Boyes

501Age Effect on Human Aortic Valvular Glycoproteins

(19) studying age-dependent changes in the radial stretch ofhuman aortic valve leaflets have reported that the stretch,which is important for maintenance of adequate cooptationarea, stays approximately constant until the age of 40 yearsand then undergoes a steady decline. Moreover, studies onhuman thoracic aortas performed by Cattell and co-workers(20) have demonstrated that the concentration of both col-lagen and elastin increase significantly with age of donorsand, in both cases, the increase occurs substantially afterthe age of 45 years. What is most important for us is thework of Allalouf et al. (21) on sialic acid content and sia-lyltransferase activity in human lymphocytes that clearlyconfirmed that sialyltransferase activity enhanced in thegroup aged 41e60 years as compared to the group aged20e40 years.

The structure of the oligosaccharides of glycoproteins isapt to be modified by changes in the physiological condi-tion of cells. Our data suggest that changes in valve proteinglycosylation with age are related to changes in the glyco-sylation machinery but, currently, data are too limited todraw conclusions about the possible molecular basis ofthe observed phenomena. Imbalances in glycosylation canemerge for different reasons: alteration in the biosynthesisprocess due to the enhancement and/or decline of the activ-ities of the respective glycosyltransferases, alteration inglycosyltransferase gene expression, alteration in the com-position of membranes that influence the activities ofmembrane-bound enzymes, changes in monosaccharidetransfer, influence of hormones (22), inflammatory cyto-kines (23), toxins (24), dietary aspects in regard to supplyof the organism with special carbohydrates (25,26), stress(27) or even mechanical forces (28). Therefore, indicationof factor(s) responsible for observed changes requiresfurther investigations.

The physiological roles of elevated/decreased levels ofsome monosaccharide or oligosaccharide structures for aor-tic valve function remain unclear, although changed glyco-sylation profile may function to bacterial colonization byserving as binding determinants. Interestingly, it has beenreported that Staphylococcus aureus binding to cardiacvalve endothelial cells was partly mediated by a 130-kDamannose-containing glycoprotein, because the additionof Concanavalin A, which is a lectin specific for high-mannose type, hybrid type and diantennary non-bisectedcomplex-type oligosaccharides, inhibited the specific bac-terial binding by 30e40% (29). Thus, it could be hypothe-sized that surface membrane glycoproteins might playa role in bacterial adherence in the initial stages of nativevalve endocarditis.

There is also another possibility that changes in glyco-sylation profile of human aortic valve proteins could besigns of aging of the organism, similar to glycation, whichhas been associated with cellular functional defects that oc-cur with age. Indeed, recently it was found that significantchanges in N-glycosylation were associated with replicative

senescence (30,31). Changes in glycosylation could beimportant in the aging process just as changes in N-linkedoligosaccharides have been shown to play essential rolesin the development and growth of mammals (32,33). Al-though analysis of the changes in glycosylation of a proteincan be useful for diagnosis or monitoring of metabolic al-ternations, such as those that occur in disease, thesechanges do not provide insight into the function that theglycan confers on the protein. The question is whetheralternation in the aortic valve glycoproteins with age mightlead to problems associated with the functioning of thisvalve. Further studies are required to examine this problem.

AcknowledgmentsThis work was supported by the state Committee for Scientific Re-search (KBN), Warsaw, Poland, grant no. 6 P05A 06921. The au-thors thank Mrs. Marzena Jaskier from the Bank of HomogenicValves, Department of Cardiovascular Surgery and Transplantol-ogy, Jagiellonian University, Krakow, for technical help in valvepreparation.

References1. Roy A, Brand NJ, Yacoub MH. Molecular characterization of intersti-

tial cells isolated from human heart valves. J Heart Valve Dis 2000;9:

461e465.

2. Taylor PM, Allen SP, Yacoub MH. Phenotypic and functional charac-

terization of interstitial cells from human heart valves, pericardium

and skin. J Heart Valve Dis 2000;9:150e157.

3. Taylor PM, Batten P, Brand N, Thomas PS, Yacoub MH. The cardiac

valve interstitial cells. Int J Biochem Cell Biol 2003;35:113e118.

4. Burgas M, McCrea JC, Hedrick HL. Age-associated changes in car-

diac matrix and integrins. Mech Age Dev 2001;122:1739e1756.

5. Terracio L, Rubin K, Gullberg D, Balog E, Carver W, Jyring R, et al.

Expression of collagen binding integrins during cardiac development

and hypertrophy. Circ Res 1991;68:734e744.

6. Nawata J, Ohno I, Isoyama S, Suzuki J, Miura S, Ikeda J, et al. Dif-

ferential expression of alpha 1, alpha 3 and alpha 5 integrin subunits

in acute and chronic stages of myocardial infarction in rats. Cardio-

vasc Res 1999;43:371e381.

7. Pugh KG, Wei JY. Clinical implication of physiological changes in the

aging heart. Drugs Aging 2001;18:263e276.

8. Amoresano A, Amedeo S, D’andrea G, Siciliano R, Gagna G,

Castagnaro M, et al. N-linked glycans of proteins from mitral valves

of normal pigs and pigs affected by endocarditis. Eur J Biochem

2000;267:1299e1309.

9. Bradford MM. A rapid and sensitive method for quantitation of micro-

gram quantities of protein utilizing the principle of protein-dye bind-

ing. Anal Biochem 1976;72:248e254.

10. Laemmli UK. Cleavage of structural proteins during the assembly of

the head of bacteriophage T4. Nature 1970;227:680e685.

11. Lity�nska A, Przyby1o M, Pochec E, Hoja-qukowicz D, Cio1czyk D,

Laidler P, et al. Comparison of the lectin-binding pattern in different

human melanoma cell lines. Melanoma Res 2001;11:1e8.

12. Ferrari AU. Modifications of cardiovascular system with aging. Am J

Geriatr Cardiol 2002;11:30e33.

13. Lakatta EG, Sollott SJ. Perspectives on mammalian cardiovascular

aging: humans to molecules. Comp Biochem Biophys 2002;132:

699e721.

502 Przyby1o et al./ Archives of Medical Research 38 (2007) 495e502

14. Cheitlin MD. Cardiovascular physiology changes with aging. Am J

Geriatr Cardiol 2003;12:9e13.

15. McDonald PC, Wilson JE, McNeill S, Gao M, Spinelli JJ,

Rosenberg F, et al. The challenge of defining normality for human mi-

tral and aortic valve geometrical and compositional analysis. Cardio-

vasc Pathol 2002;11:193e209.

16. Fokin AA, Robicsek F, Cook JW, Thubrikar MJ, Schaper J. Morfolog-

ical changes of the aortic valve leaflets in non-compliant aortic roots:

in-vivo experiments. J Heart Valve Dis 2004;13:444e451.

17. Lindsey ML, Goshorn DK, Squires CE, Escobar GP, Hendrick JW,

Mingoia JT, et al. Age-dependent changes in myocardial matrix

metalloproteinase/tissue inhibitor of metalloproteinase profiles and

fibroblast function. Cardiovasc Res 2005;66:410e419.

18. Lin SL, Lin CP, Young ST, Lin M, Chiou CW. Age-related changes in

aortic valve with emphasis on the relation between pressure loading

and thickened leaflets of the aortic valves. Int J Cardiol 2005;103:

272e279.

19. Christie GW, Barratt-Boyes BG. Age-dependent changes in the radial

stretch of human aortic valve leaflets determined by biaxial testing.

Ann Thorac Surg 1995;60(suppl 2):S156eS159.

20. Cattell MA, Anderson JC, Hasleton PS. Age-related changes in

amounts and concentration of collagen and elastin in normotensive

human thoracic aorta. Clin Chim Acta 1996;245:73e84.

21. Allalouf FD, Komlos L, Notmann J, Halbrech TI, Levinsky H. Sialic

acid content and sialyltransferases activity in human lymphocytes with

advancing age. Mech Age Dev 1988;44:45e50.

22. Medvedova L, Farkas R. Hormonal control of protein glycosylation:

role of steroids and related lipophilic ligands. Endocr Regul 2004;

38:65e79.

23. Garcia-Vallejo JJ, van Dijk W, van Die I, Gringhuis SI. Tumor necro-

sis factor-alpha up-regulates the expression of beta1,4-galactosyltrans-

ferase I in primary human endothelial cells by mRNA stabilization. J

Biol Chem 2005;280:2676e2682.

24. Sillanaukee P, Ponnio M, Seppa K. Sialic acid: new potential marker

of alcohol abuse. Alcohol Clin Exp Res 1999;23:1039e1043.

25. Brand Miller J, McVeagh P. Human milk oligosaccharides: 130

reasons to breast-feed. Br J Nutr 1999;82:333e335.

26. Engfer MB, Stahl B, Finke B, Sawatzki G, Daniel H. Human milk

oligosaccharides are resistant to enzymatic hydrolysis in the upper

gastrointestinal tract. Am J Clin Nutr 2000;71:1589e1596.

27. Dabelic S, Flogel M, Maravic G, Lauc G. Stress causes tissue-specific

changes in the sialyltransferase activity. Z Naturforsch [C] 2004;59:

276e280.

28. Sarpie TG. Anionic surface properties of aortic and mitral valve endothe-

lium from New Zealand white rabbits. Am J Anat 1985;174:145e160.

29. Johnson CM. Staphylococcus aureus binding to cardiac endothelial

cells is partly mediated by a 130 kilodalton glycoprotein. J Lab Clin

Med 1993;121:675e682.

30. Kuwahara I, Ikebuchi K, Hamada H, Niitsu Y, Miyazaki K, Okyashiki K,

et al. Changes in N-glycosylation of human stromal cells by telomerase

expression. Biochem Biophys Res Commun 2003;101:532e540.

31. Kawado T, Hayashi O, Sato T, Ito H, Hayakawa S, Takayama E, et al.

Rapid cell senescence-associated changes in galactosylation of N-

linked oligosaccharides in human lung adenocarcinoma A549 cells.

Arch Biochim Biophys 2004;426:306e313.

32. Asano M, Furukawa K, Kido M, Matsumoto S, Umesaki Y, Kochibe N,

et al. Growth retardation and early death of beta-1,4-galactosyltrans-

ferase knockout mice with augmented proliferation and abnormal dif-

ferentiation of epithelial cells. EMBO J 1997;16:1850e1857.

33. Lu Q, Hasty P, Shur BD. Targeted mutation in beta1,4-galactosyltrans-

ferase leads to pituitary insufficiency and neonatal lethality. Dev Biol

1997;181:257e267.