Age-related changes in the physical properties, cross-linking, and glycation of collagen from mouse tail tendon

Abstract

Collagen is a structural protein whose internal cross-linking critically determines the properties and functions of connective tissue. Knowing how the cross-linking of collagen changes with age is key to understanding why the mechanical properties of tissues change over a lifetime. The current scientific consensus is that collagen cross-linking increases with age and that this increase leads to tendon stiffening. Here, we show that this view should be reconsidered. Using MS-based analyses, we demonstrated that during aging of healthy C57BL/6 mice, the overall levels of collagen cross-linking in tail tendon decreased with age. However, the levels of lysine glycation in collagen, which is not considered a cross-link, increased dramatically with age. We found that in 16-week-old diabetic db/db mice, glycation reaches levels similar to those observed in 98-week-old C57BL/6 mice, while the other cross-links typical of tendon collagen either decreased or remained the same as those observed in 20-week-old WT mice. These results, combined with findings from mechanical testing of tendons from these mice, indicate that overall collagen cross-linking in mouse tendon decreases with age. Our findings also reveal that lysine glycation appears to be an important factor that contributes to tendon stiffening with age and in diabetes. Collagen is a structural protein whose internal cross-linking critically determines the properties and functions of connective tissue. Knowing how the cross-linking of collagen changes with age is key to understanding why the mechanical properties of tissues change over a lifetime. The current scientific consensus is that collagen cross-linking increases with age and that this increase leads to tendon stiffening. Here, we show that this view should be reconsidered. Using MS-based analyses, we demonstrated that during aging of healthy C57BL/6 mice, the overall levels of collagen cross-linking in tail tendon decreased with age. However, the levels of lysine glycation in collagen, which is not considered a cross-link, increased dramatically with age. We found that in 16-week-old diabetic db/db mice, glycation reaches levels similar to those observed in 98-week-old C57BL/6 mice, while the other cross-links typical of tendon collagen either decreased or remained the same as those observed in 20-week-old WT mice. These results, combined with findings from mechanical testing of tendons from these mice, indicate that overall collagen cross-linking in mouse tendon decreases with age. Our findings also reveal that lysine glycation appears to be an important factor that contributes to tendon stiffening with age and in diabetes. The literature surrounding the mechanical and chemical properties of collagen and changes which occur with age is extensive. The general consensus is that as collagen ages there is an increase in the stiffness with loss of elasticity and that this is due to an increase in covalent intermolecular cross-linking between collagen molecules which develops with age (1Fratzl, P., (ed) (2008) Collagen: Structure and Mechanics. Springer, New York, NY.Google Scholar). In this paper we challenge the simplicity of this conclusion. The collagen cross-links can be divided into two groups, those that are of enzymatic origin and those that form through purely chemical reactions with reactive molecules perfusing the tissues. The enzymatically derived cross-links that are first formed can be analyzed after reduction with sodium borohydride (2Bailey A.J. Peach C.M. Isolation and structural identification of a labile intermolecular crosslink in collagen.Biochem. Biophys. Res. Commun. 1968; 33 (5723342): 812-81910.1016/0006-291X(68)90233-7Crossref PubMed Scopus (110) Google Scholar, 3Tanzer M.L. Mechanic G. Gallop P.M. Isolation of hydroxylysinonorleucine and its lactone from reconstituted collagen fibrils.Biochim. Biophys. Acta. 1970; 207 (5452678): 548-55210.1016/S0005-2795(70)80017-4Crossref PubMed Google Scholar) and the reduced products measured as dihydroxy-lysino-norleucine (DHLNL), hydroxy-lysino-norleucine (HLNL), and lysino-norleucine (LNL) (structures shown in Fig. 1). Another collagen cross-link, histidine-hydroxymerodesmosine (HHMD), is commonly found during the analysis of reduced samples (4Tanzer M.L. Housley T. Berube L. Fairweather R. Franzblau C. Gallop P.M. Structure of two histidine-containing cross-links from collagen.J. Biol. Chem. 1973; 248 (4684687): 393-402Abstract Full Text PDF PubMed Google Scholar) by MS. While there has been disagreement (5Bernstein P.H. Mechanic G.L. A natural histidine-based imminium cross-link in collagen and its location.J. Biol. Chem. 1980; 255 (7430129): 10414-10422Abstract Full Text PDF PubMed Google Scholar, 6Eyre D.R. Weis M. Rai J. Analyses of lysine aldehyde cross-linking in collagen reveal that the mature cross-link histidinohydroxylysinonorleucine is an artifact.J. Biol. Chem. 2019; 294 (30733334): 6578-659010.1074/jbc.RA118.007202Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar) with respect to the exact chemical structure that the analyzed compound represents within collagen, for the purposes of this publication it can be considered an indicator of aldol cross-links present in collagen before analysis. These cross-links are early structures in the enzymatic cross-linking process and are often described as immature cross-links. The immature cross-links can go on to form irreversible cross-links (1Fratzl, P., (ed) (2008) Collagen: Structure and Mechanics. Springer, New York, NY.Google Scholar) (pyridinolines), often described in the literature as “mature” cross-links, through further reactions. It has been noted in numerous papers that the number of immature cross-links per mole of collagen decreases with age and that the increase in mature pyridinoline cross-links does not seem to match this decrease (7Robins S.P. Shimokomaki M. Bailey A.J. The chemistry of the collagen cross-links. Age-related changes in the reducible components of intact bovine collagen fibres.Biochem. J. 1973; 131 (4722452): 771-78010.1042/bj1310771Crossref PubMed Scopus (202) Google Scholar, 8Sricholpech M. Perdivara I. Yokoyama M. Nagaoka H. Terajima M. Tomer K.B. Yamauchi M. Lysyl hydroxylase 3-mediated glucosylation in type I collagen: molecular loci and biological significance.J. Biol. Chem. 2012; 287 (22573318): 22998-2300910.1074/jbc.M112.343954Abstract Full Text Full Text PDF PubMed Scopus (71) Google Scholar, 9Saito M. Marumo K. Collagen cross-links as a determinant of bone quality: a possible explanation for bone fragility in aging, osteoporosis, and diabetes mellitus.Osteoporos. Int. 2010; 21 (19760059): 195-21410.1007/s00198-009-1066-zCrossref PubMed Scopus (643) Google Scholar), implying that the loss of immature cross-links cannot be entirely explained by their conversion into mature cross-links. In papers describing mature cross-link formation, the emphasis is usually on the increase in mature cross-links, and this increase is correlated to an increase in stiffness of the tissue; however, this overlooks the fact that there is potentially an overall decrease in cross-linking with age from the loss of the immature cross-links. The paradox here is that if total cross-linking is decreasing with age, then how is it that the tissues get stiffer? Cross-links formed through nonenzymatic processes involving sugars are referred to as advanced glycation end products (AGEs), and in this paper, we include them when referring to mature cross-links. These are formed through the reaction of sugars or products of sugar metabolism with collagen, which then react further to create cross-links. The AGE cross-links generally considered to be most important are glucosepane and pentosidine (9Saito M. Marumo K. Collagen cross-links as a determinant of bone quality: a possible explanation for bone fragility in aging, osteoporosis, and diabetes mellitus.Osteoporos. Int. 2010; 21 (19760059): 195-21410.1007/s00198-009-1066-zCrossref PubMed Scopus (643) Google Scholar, 10Monnier V.M. Sun W. Sell D.R. Fan X. Nemet I. Genuth S. Glucosepane: a poorly understood advanced glycation end product of growing importance for diabetes and its complications.Clin. Chem. Lab. Med. 2014; 52 (23787467): 21-3210.1515/cclm-2013-0174Crossref PubMed Scopus (50) Google Scholar). A significant proportion of the literature has focused on increased AGE cross-linking, particularly glucosepane, as a potential cause of stiffening in aging tendon. This has been concluded through correlations of measured AGE cross-link levels with the mechanical testing of diabetic tissue and of collagen incubated in vitro in the presence of sugars. However, these studies measured the absolute glucosepane concentration not as a proportion of the total collagen content but usually as a relative increase in signal (11Nash A. Notou M. Lopez-Clavijo A.F. Bozec L. de Leeuw N.H. Birch H.L. Glucosepane is associated with changes to structural and physical properties of collagen fibrils.Matrix Biology Plus. 2019; 4: 10001310.1016/j.mbplus.2019.100013Crossref Scopus (10) Google Scholar) or as a proportion of the extracted insoluble collagen content (12Monnier V.M. Bautista O. Kenny D. Sell D.R. Fogarty J. Dahms W. Cleary P.A. Lachin J. Genuth S. Collagen glycation, glycoxidation, and crosslinking are lower in subjects with long-term intensive versus conventional therapy of type 1 diabetes.Diabetes. 1999; 48 (10102706): 870-88010.2337/diabetes.48.4.870Crossref PubMed Scopus (425) Google Scholar). Here we measured the absolute glucosepane content as a proportion of the total collagen content of tendon in addition to the pentosidine content and the glycation products of lysine and hydroxylysine. Although increases in glucosepane with age have been demonstrated, without knowing the absolute amount relative to the total collagen content it is impossible to assess if glucosepane cross-linking is an important factor in the stiffening of normal healthy tendon during aging and what the impact of other AGEs might be. In this study we chose to use mouse tail tendon because it is primarily collagen I and the impact of altered cross-linking on the mechanical properties can be readily tested. We used tendon from WT C57BL/6 mice because they have been shown not to develop diabetes (13Leiter E.H. Premdas F. Harrison D.E. Lipson L.G. Aging and glucose homeostasis in C57BL/6J male mice.FASEB J. 1988; 2: 2807-281110.1096/fasebj.2.12.3044905Crossref PubMed Scopus (68) Google Scholar) and so represent a model of healthy aging. We studied both the mechanical and chemical changes across a population of mice from 8 weeks to 100 weeks old. In order to remove as many population variables as possible and show what happens in the normal, healthy aging process, a cohort of C57BL/6 mice housed under highly controlled conditions of diet and environment were used. The aim of the study described here to the changes in cross-linking that occur in normal healthy mouse tail tendon with age and which cross-links are to to the altered physical properties with age. glucose levels were found to be controlled across age under and conditions a and The of collagen in tendon and weeks of age by for and then the of the into the collagen within the tissues. The of of found to be between and collagen lysine content per in tail change with age The for tail tendons from mice with a of ages were measured a is as is as are tendons for bone in the which from the of the tail to bone in of The tendons were from the for tendons were to the a of 1 and the measured The for the from were across the population those from from mice an by a before The of tendons from a with a and a lower with a age, a proportion of the tendons an increase in the of the that the stiffness of the tendons The in of ages to before into the tendons in the 20-week-old a before either tendons or tendons 1 old. The of changes here is the usually in the a of stiffness of tendon with age is The cross-links present and collagen content were measured by tissue and for this It of to to remove reactive from as which are to be highly reactive this we used a to the components from the We that the for of components not impact the of under study by the levels in tendon after without and after In addition to a for the analysis of the AGE cross-link glucosepane by the products of both the and glucosepane used to the before were Fig. changes in the cross-links found in mouse tail tendon with age. The most cross-link ages and the change with age in the of this A in to weeks when the decreasing from cross-links per collagen 8 weeks to cross-links per collagen levels to decrease after the to cross-links per collagen While we not the absolute amount of due to the of a we were to the relative change in the can be to increase to weeks and then decrease as the age and are formed from by the of The also be from the of and because we a in to we a between and be were two in the shown in the between and In the first between 8 and it can be that as In the from to it can be that both and with age. These that as the levels in early the aldehyde an aldol reaction and aldehyde or aldol In the of of aldol is then lower the of and so both the levels of and are to In an to the possible of we an that the change between the ages of weeks and weeks is into the measured change in on the this the of to per collagen weeks and to per collagen weeks conversion factor is to 1 in Fig. This of is with levels of the compound found in tendon and M. G. S. M. S. Yamauchi M. collagen cross-linking by lysine in the and of tendon type I collagen.J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar). mature cross-link type increased dramatically with however, as a mature cross-links to a increase in overall number of cross-links per collagen The increase in mature cross-links to an increase from 1 cross-link per 33 collagen molecules weeks to 1 cross-link in collagen molecules This increase in mature cross-links the in between weeks and weeks and The of mature and immature cross-links weeks cross-link per collagen the the of the there a increase in AGE however, in with the increase in lysine glycation a change The glycation of lysine is the addition of a to the of lysine and is not a further reaction with in the collagen The addition reaction can also on the of hydroxylysine. weeks of age, the glycation of tendon collagen lysine per collagen the glycation of lower per of The of lysine glycation appears to after weeks of age. levels of glycation to with decreased both were during the and the changes not be It has been that glycation the same as the cross-links M. K.B. D.R. of type I collagen the same lysine as Biol. Chem. Full Text Full Text PDF PubMed Google however, in that glycation of In we glycation of as the the changes in collagen lysine glycation in healthy WT C57BL/6 mice, we to the levels in diabetic mice. tendon samples from 16-week-old db/db mice on the were These diabetic mice have a in the and are used as an model for diabetes The tendon were to physical and chemical analysis as described and the are shown in Fig. and Fig. The and stiffness observed in the of tendon from 16-week-old db/db mice those in tendon from WT while the often that in tendon from 1 and in the of the db/db tendon can be in the and were in WT tendon. analysis that levels of of cross-link for the diabetic from were lower the levels found in both and WT and The not between 20-week-old WT and 16-week-old db/db tendon. The in 16-week-old db/db tendon found to be per collagen which is lower the per collagen found in 20-week-old WT tendon. The of glycation in the 16-week-old db/db tendon to that in WT mice and that in 20-week-old WT mice. The levels of irreversible cross-links in the db/db mice, both the mature and AGE were lower those in 20-week-old WT mice. In the literature on aging collagen, the is usually on the increase in mature cross-links formed through mature covalent as the for the properties we here that the is this and that both the loss of cross-links and increase in lysine glycation are important changes to in the overall of how tendon collagen The view that tendon with age due to an increase in cross-linking does not the shown and in of the db/db it appears that the mechanical changes not have to be to an increase in AGEs as glucosepane or the decrease in cross-linking in db/db mice, it might be that the tendons be those of WT however, the stiffness of the db/db tendons is similar to that in WT mice. The factor that to the observed stiffness of db/db tendon is the increased glycation of lysine which is similar to that found in tendon from 98-week-old mice. It is from the and in the db/db that there are in the tendon It is that these are due to the lower levels of cross-linking, why the diabetic tendon is often to structure under While tendon does generally with age, important change is the loss of the It from the here that the loss of the is due to a decrease in immature cross-links with an increase in mature cross-linking, while the stiffening is by both the glycation of lysine and mature cross-link The by which cross-links to these properties and a of the of lysine in tendon are the of an paper M. A. J. changes cross-linking and glycation levels in the collagen of mouse tail Biol. Chem. Scholar). In tendon stiffening and the loss of the in are important mechanical changes with age and diabetes observed in this the for healthy aging an understanding of the chemical changes that these mechanical Here we have shown that lysine glycation products are to a significant in these mechanical changes as as the altered of immature and mature enzymatic cross-links. were to the and by the and of C57BL/6 mice were the 1 of for the mice by as the of The glucose measured in the from the and from a of from the tail after The glucose measured glucose from of C57BL/6 mice were on mouse from for the mice were to a normal diet for a further and then of the from tendon of the tail under by the and the of the tail with the the the tail to tendon with it to the bone a the tail from the to the were from the tail with a and components were from tail tendon by in in a for by in for here is that that the chemistry of the immature cross-links is potentially that changes chemical environment before reduction the and impact the The of labile to the of Res. PubMed Scopus Google Scholar, The of on collagen Res. Scopus Google Scholar). when we incubated with a in for we found that after reduction there a in the of and a in Fig. the impact on tendon collagen of as commonly used M. G. S. M. S. Yamauchi M. collagen cross-linking by lysine in the and of tendon type I collagen.J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar) with that we used in this It can be that in is lower and the lysine is lower that in and similar to and in a similar tendon to in reduced by the addition of of in 1 for the tendon with and then by the addition of 1 the tendon with a further 1 before and internal and samples by with of were then under a of and in through a samples were before in were then with internal for the analysis of the of for analysis a 100 have been described in the analysis of S. S. A of the properties of and A. 2012; PubMed Scopus Google Scholar) and collagen cross-links R. G. M. D. G. for the of collagen and A. PubMed Scopus Google Scholar). The here is a of these A of to the of which are in The into an of a and the were with and with with is a 1 is a change to the with and with with is a 1 is a change to the in a glucosepane in a lysine and are to under the the and were and the were to the total lysine and total were from were as the internal for and and as the internal for and and were used for with as an internal The and internal were used to collagen which not been reduced for the The used to a as the internal we and for and pyridinoline with a and were as to to be The of the found to be the same as that of The of found to be a factor of that of the used as an internal for for and used for from by to C. T. total of PubMed Scopus Google Scholar) with The for glucosepane is to use of collagen with a of over weeks because glucosepane is not considered to be to are a number of with the is the it to the which does not to of and is that we as have (11Nash A. Notou M. Lopez-Clavijo A.F. Bozec L. de Leeuw N.H. Birch H.L. Glucosepane is associated with changes to structural and physical properties of collagen fibrils.Matrix Biology Plus. 2019; 4: 10001310.1016/j.mbplus.2019.100013Crossref Scopus (10) Google Scholar), that is with a which is and the of a to be with the of cross-links, because it is possible that this might the cross-links of We found that when glucosepane is in there is an addition of of to the structure by MS. the of a glucosepane before we found that glucosepane be measured after by of this were to the by both and to collagen samples to We have not how this is for the analysis of glucosepane in tissues other mouse and mouse tendon. on the were to for molecular in reduced tendon samples that to a of a typical product of of found which in the on the for a structure of the type by (4Tanzer M.L. Housley T. Berube L. Fairweather R. Franzblau C. Gallop P.M. Structure of two histidine-containing cross-links from collagen.J. Biol. Chem. 1973; 248 (4684687): 393-402Abstract Full Text PDF PubMed Google Scholar). of the a with the We then a of the and after reduction and the and after This to the of the in the The from this are shown in Fig. These when to the before show that in the there is a amount of from a and a of from aldehyde similar to the of the The of and are with the structure shown in the paper by (4Tanzer M.L. Housley T. Berube L. Fairweather R. Franzblau C. Gallop P.M. Structure of two histidine-containing cross-links from collagen.J. Biol. Chem. 1973; 248 (4684687): 393-402Abstract Full Text PDF PubMed Google Scholar). of the from reduction reduction on the and without which is with the structure by (4Tanzer M.L. Housley T. Berube L. Fairweather R. Franzblau C. Gallop P.M. Structure of two histidine-containing cross-links from collagen.J. Biol. Chem. 1973; 248 (4684687): 393-402Abstract Full Text PDF PubMed Google Scholar). We concluded that the here is with the and we on the that it to be the same compound that (4Tanzer M.L. Housley T. Berube L. Fairweather R. Franzblau C. Gallop P.M. Structure of two histidine-containing cross-links from collagen.J. Biol. Chem. 1973; 248 (4684687): 393-402Abstract Full Text PDF PubMed Google Scholar) analyzed and used were from and from were from and were either or were on and were on a in the of The an of lysine which the same as lysine by MS. It is that in this and this for or of the in the This a internal to be which in the 1 to glucose in The reaction for and then to borohydride then and the reaction to A 1 concentration of then and the to an and the to The a and with and the were with The then a number of which were analyzed by MS. The the of with the after of on a with a to then to with and The reaction under and for of the reaction by MS. The by the reaction through a and to of the The and that of glycation with a of are shown in Fig. a to and are shown in Fig. This for both and the the of the were and a of used. glucose 1 in and and the to borohydride in 1 sodium The reaction but the reaction found to be by of sodium borohydride and the reaction in by to the reaction and the over and under The product on with as the to were in with of and a highly of Scopus Google Scholar). 8 to and to A of in over and to react for a further 1 in over The reaction for and then over and to react for The reaction to to over and to for a further 1 and the product extracted into over and The product on to the without and were to the of in The reaction in the and for and then the product as the reaction to and for The and the in and with 1 sodium The over and under The product on with with as the by from to as a are shown in Fig. 1 in by addition of by The reaction under and by MS. the reaction through a to remove the and under reduced The product from to are shown in Fig. a to total and testing a with a and a to of the a of between the were in and to a of The of its and the used to measured as the to a of 1 were were used. were by were and were to the between as a of and are in the analysis are from the We the for their without which this not have been We also for and were by and of with with dihydroxy-lysino-norleucine glycation end product analysis of histidine-hydroxymerodesmosine hydroxy-lysino-norleucine internal lysino-norleucine pyridinoline

Topics & Concepts