Relationship between the content of lysyl oxidase-dependent cross-links in skin collagen, nonenzymatic glycosylation, and long-term complications in type I diabetes mellitus.

Many abnormalities in collagen have been reported in insulin-dependent diabetes mellitus, some or all of which have been attributed to increased cross-linking. Although recent work has focused on the role of glucose-derived collagen cross-links in the pathogenesis of diabetic complications, relatively few studies have investigated the role of lysyl oxidase-dependent (LOX) cross-links. In the present study, LOX cross-links and nonenzymatic glycosylation were quantified in skin collagen from diabetic subjects. There was an increase in the difunctional cross-link dihydroxylysinonorleucine (DHLNL) as well as in one of its trifunctional maturation products, hydroxypyridinium. All other LOX crosslinks were normal. Nonenzymatic glycosylation was increased in diabetic skin collagen, and this increase was correlated with increases in DHLNL (P less than 0.001). The biochemical results were examined for correlations with clinical data from the same subjects. Increases in DHLNL content were associated with duration of diabetes (P less than 0.003), glycohemoglobin levels (P less than 0.001), hand contractures (P less than 0.05), skin changes (P less than 0.005), and microalbuminuria (P less than 0.01). In nondiabetic subjects age was not correlated with collagen cross-link content with the exception that his-HLNL increased with age (r = 0.79, P less than 0.02). In diabetic subjects, PA levels decreased with age (r = 0.51, P less than 0.02). With increased duration of diabetes, DHLNL content was increased (r = 0.55, P less than 0.003) and OHP was increased (r = 0.59, P less than 0.01), whereas PA levels were decreased (r = -0.48, P less than 0.04). Nonenzymatic glycosylation of collagen was also increased with increased duration of diabetes (hex-lys, r = 0.47, P less than 0.02; hex-hyl, r = 0.39, P less than 0.05). We conclude that: (a) lysyl oxidase-dependent cross-linking is increased in skin collagen in diabetes and (b) that these changes in skin collagen are correlated with duration of diabetes, glycemic control, and long-term complications.

Modulation by beta-aminopropionitrile of vessel luminal narrowing and structural abnormalities in arterial wall collagen in a rabbit model of conventional balloon angioplasty versus laser balloon angioplasty.

This study was designed to assess the potential relationship between the late loss of angiographic luminal diameter and biochemical abnormalities of arterial wall collagen in rabbits subjected to angioplasty, and to test the hypothesis that beta-aminopropionitrile (beta APN), an inhibitor of lysyl oxidase, would inhibit such changes when administered orally for 1 mo after angioplasty. Endovascular injury was induced in rabbit iliac arteries by ipsilateral balloon angioplasty (BA) and by contralateral balloon angioplasty accompanied by exposure to continuous wave neodymium: yttrium aluminum garnet laser radiation (LBA). Computer measurement of angiographic luminal diameter demonstrated significant vessel narrowing at 1 and 6 mo after both procedures. By quantitative histology, the majority of the 1-mo loss in angiographic diameter could not be attributed to neointimal thickening. Analysis of collagen cross-linking by HPLC in collagen obtained from the LBA-injured segments of the arteries 1 mo after angioplasty revealed a significant increase, relative to values from uninjured arteries (P < 0.05), in the difunctional cross-link dihydroxylysinonorleucine (DHLNL). 6 mo after angioplasty, the content of hydroxypyridinium, the trifunctional maturational product of DHLNL, was significantly elevated in both BA- and LBA-treated arteries compared with values from uninjured arteries (P < 0.05). In animals administered beta APN, luminal narrowing at 1 mo, compared with controls, was attenuated (P < 0.01) and DHLNL content was decreased (P < 0.05) in arteries subjected to LBA, but not in arteries subjected to BA. The results suggest that lathyrogenic agents may be efficacious in favorably modulating LBA-induced alterations in vessel diameter and mural connective tissue.

Biosynthesis of collagen crosslinks. II. In vivo labelling and stability of lung collagen in rats.

Rat lung collagen was labelled in vivo by a single intraperitoneal injection of [3H]lysine at several key timepoints in lung development: days 11 (alveolar proliferation), 26 (start of equilibrated growth), 42 (end of equilibrated growth), and 100 (adult lung structure present). The rates of deposition of labelled hydroxylysine and the difunctional, Schiff base-derived crosslinks hydroxylysinonorleucine (HLNL) and dihydroxylysinonorleucine (DHLNL) were quantified. We also measured total lung content of the trifunctional, mature crosslink hydroxypyridinium (OHP) in these same animals. While the relative rates of accumulation of labelled collagen [3H]hydroxylysine differed by a factor of about 6 at the different times of injection of labelled precursor, quantitative and qualitative patterns of collagen crosslinking were very similar at all of the lung developmental stages studied. Furthermore, there was little or no breakdown of the lung collagen pool as defined by the presence of labelled crosslinks; changes in lung DHLNL content could be completely accounted for by its maturation to OHP, regardless of the age of the rats when injected with the radioactive precursor. We conclude that mature, crosslinked collagen in the lungs of rats, which is obligatorily an extracellular pool, is not being degraded at a measurable rate. Therefore, studies of others that have shown apparent high rates of breakdown of newly synthesized collagen in lungs of whole animals using different methods are probably not reflective of the metabolic fate of total lung collagen, and may indicate that degradation of normal lung collagen occurs predominantly or exclusively intracellularly.

Lysyl oxidase-mediated crosslinking in granulation tissue collagen in two models of hyperglycemia.

Enzymatically mediated crosslinks and nonenzymatic glycation were quantified in granulation tissue collagen in two models of hyperglycemia, diabetes and galactosemia, that have opposite effects on collagen solubility. The effects of castration, which alters collagen solubility, was also investigated. Collagen from both diabetic and galactosemic rats had significantly increased levels of dihydroxylysinonorleucine (DHLNL), a difunctional reducible crosslink. Galactosemic rats had significantly decreased levels of hydroxypyridinium, a trifunctional product of DHLNL and hydroxylysine, relative to control values, while diabetic rats had normal levels. Values for all other detectable crosslinks in collagen from hyperglycemic rats were indistinguishable from control values. Nonenzymatic glycation was increased in both groups of hyperglycemic rats. In diabetic rats, but not in galactosemic rats, nonenzymatic glycation was strongly correlated with DHLNL content. Castration had no effect on crosslink content of collagen from diabetic or galactosemic rats. This study demonstrates that (1) collagen crosslinking is abnormal in granulation tissue collagen in both experimental diabetes and galactosemia, (2) these changes are similar to those observed in skin collagen from insulin-dependent diabetic subjects and (3) the crosslinking abnormalities are not correlated with alterations in collagen solubility. We conclude that hyperglycemia-associated increases in immature crosslinks cannot account for altered collagen solubility, although impaired maturation of such crosslinks may be partially responsible for the lathyrogenic effect of galactosemia.

Lung collagen cross-links in rats with experimentally induced pulmonary fibrosis.

Rats were intratracheally instilled with bleomycin or with silica (quartz) dust to induce lung fibrosis. Several weeks later, purified collagen chains (or collagen digests) were isolated from the lungs of these animals and from age-matched controls instilled intratracheally with saline solution, and the ratios of hydroxylysine to lysine and of the dysfunctional cross-links DHLNL to HLNL were quantified. Collagen from fibrotic lungs had significantly higher ratios of DHLNL:HLNL than did control lungs, 15.5 +/- 4.8 and 17.1 +/- 4.8 vs. 2.3 +/- 0.5 for the silica-instilled and the bleomycin-instilled animals, respectively. The hydroxylysine:lysine ratio was significantly increased for the alpha 1(I) chain, to a value 170% of that of lung collagen from control animals, and for several of its constituent CNBr peptides. Lung tissue was exhaustively digested with collagenase and specific cross-linked peptides were isolated and characterized. The cross-linked alpha 1(I) x alpha 1(I) peptide linked by the residues 87 x 16C, with a ratio of DHLNL:HLNL of 17:1, demonstrated that the increased hydroxylation of the dysfunctional cross-links in fibrotic lung collagen could be accounted for in part by increased hydroxylation of the lysine residue at position 16C of the C-terminal telopeptide of the collagen alpha 1(I) chain. It proved impossible to locate the corresponding N-terminal cross-linked fragment from alpha 1(I) x alpha 1(I) chains, 9N x 930, possibly due to further reactions of this material to form the material referred to as poly(CB6). Isolated poly (CB6) accounted for more than half of the total alpha 1(I)CB6 peptide expected in lung collagen, and had a hydroxylysine:lysine content 2.8 times greater in bleomycin-treated animals than in their age-matched controls. Evidence was also found for a cross-linked alpha 1(III) x alpha 1(I) peptide linking residue 87 from the alpha 1(III) chain with residue 16C from the alpha 1(I) chain; it also had an increased ratio of DHLNL:HLNL. We conclude that the increased hydroxylation of lysine observed in two different animal models of lung fibrosis occurs preferentially at the N- and C-terminal nonhelical extension peptides of the alpha 1(I) collagen chains, and that this apparent specificity of overhydroxylation of fibrotic collagen may have important structural and pathological consequences.

Analysis of age-associated changes in collagen crosslinking in the skin and lung in monkeys and rats.

The present study was designed to address a specific question: can we define collagen aging in vivo in terms of alterations in collagen crosslinking? In order to assess the complete spectrum of change throughout life, tissues from rats, monkeys and (where available) humans were examined at ages ranging from fetal to old. Skin and lung were selected in order to include all of the crosslinks derived from lysyl oxidase-generated aldehydes that have been identified thus far, both reducible and nonreducible. Crosslinks analyzed included hydroxylysinonorleucine, dihydroxylysinorleucine, histidinohydroxymerodesmosine, hydroxypyridinium, lysyl pyridinium, and a deoxy analogue of hydroxypyridinium found in skin that differs structurally from lysyl pyridinium. Tissues from both a short-lived species (rats) and a long-lived species (monkeys) were analyzed to test further the hypothesis that changes in crosslinking are linked predominantly to biological age of the animal, rather than temporal aging. We found that biological aging seems to regulate certain predictable changes during the first part of the lifespan: the disappearance postnatally of dihydroxylysinonorleucine in skin, the rapid decrease in difunctional crosslink content in lung and skin during early growth and development, and the gradual rise in hydroxypyridinium and lysyl pyridinium in lung tissue. Changes in crosslinking were far less predictable during the second half of the lifespan. Although hydroxypridinium content continued to rise or reached a plateau in rat and monkey lungs, respectively, it showed a decrease in human lungs. The analogous trifunctional crosslink in skin, the so-called 'pyridinoline analogue', decreased dramatically in both rats and monkeys in later life. Our data suggest that caution must be taken in drawing inferences about human connective tissue aging from experiments performed in short-lived species such as rodents. Furthermore, the finding that there may be fewer total lysyl oxidase-derived crosslinks per collagen molecule in very old animals as compared with young animals suggests that we may need to expand our concepts of collagen crosslinking.

Discordant effects of guanidines on renal structure and function and on regional vascular dysfunction and collagen changes in diabetic rats.

We examined the effects of aminoguanidine and methylguanidine on vascular dysfunction, glomerular structural changes, and indexes of early and late nonenzymatic glycation in 7-month streptozotocin-induced diabetic rats. Kidney weight, glomerular volume, fractional mesangial volume, glomerular capillary basement membrane width, and urinary albumin excretion were increased in diabetic rats. Diabetes also 1) increased vascular albumin permeation twofold in retina, sciatic nerve, aorta, skin, and kidney; 2) decreased renal collagenase-soluble collagen; 3) increased collagen-associated fluorescence in kidney and skin but not in aorta; and 4) increased glycated hemoglobin levels and aortic pentosidine levels. Aminoguanidine reduced albuminuria by 70% after 4 months, and both guanidines 1) normalized aortic pentosidine levels and renal collagenase-soluble collagen, 2) had no effect on glycated hemoglobin levels or collagen-associated fluorescence (in aorta, kidney, or skin), and 3) had little or no effect on regional albumin permeation. These discordant effects of aminoguanidine on diabetes-induced vascular changes versus parameters of nonenzymatic glycation are consistent with a multifactorial pathogenesis of diabetic complications, including roles for metabolic imbalances independent of nonenzymatic glycation. To the extent that glomerular matrix accumulation and increased regional albumin permeation in chronically diabetic rats are sequelae of nonenzymatic glycation, these findings point to an important role for early glycation reactions and products.

Effect of aminoguanidine on the frequency of neuroaxonal dystrophy in the superior mesenteric sympathetic autonomic ganglia of rats with streptozocin-induced diabetes.

Aminoguanidine, which prevents formation of advanced glycation end products and is a relatively selective potent inhibitor of the inducible (versus constitutive) isoform(s) of nitric oxide synthase, has been reported to ameliorate structural and functional abnormalities in peripheral somatic nerves in rats with streptozocin (STZ)-induced diabetes. In the present studies, the effects of aminoguanidine treatment on ultrastructural changes in the autonomic nervous system of rats with STZ-induced diabetes were examined. The frequency of neuroaxonal dystrophy, the neuropathological hallmark of sympathetic autonomic neuropathy in diabetic rats, increased 9- to 11-fold in the superior mesenteric ganglia of 7- and 10-month STZ-diabetic rats compared with that in age-matched controls. Administration of aminoguanidine continuously from the time of induction of diabetes at a dose equal to or in excess of that providing a salutary effect in the diabetic somatic peripheral nervous system did not alter the severity of diabetes as assessed by plasma glucose level, 24-h urine volume, and levels of glycated hemoglobin. Chronic aminoguanidine therapy did not diminish the frequency or affect the ultrastructural appearance of neuroaxonal dystrophy in diabetic or age-matched control rat sympathetic ganglia after 7 or 10 months of continuous administration. Our findings (under these experimental conditions) do not support a role for aminoguanidine-sensitive processes in the development of sympathetic neuroaxonal dystrophy in diabetic rats. Glycation-linked aminoguanidine-insensitive processes, however, such as the formation of early glucose adducts (Schiff bases and Amadori products) with intracellular and/or extracellular proteins and amine-containing lipids, superoxide anion generation during subsequent autoxidation of these glucose adducts, and non-glycative processes, remain potential pathogenetic mechanisms for diabetic autonomic neuropathy.

Effects of aspirin or basic amino acids on collagen cross-links and complications in NIDDM.

OBJECTIVE: To determine if long-term therapy with aspirin or basic amino acids for subjects with NIDDM reduces the severity of clinical complications and/or reduces tissue levels of markers of glycooxidative damage. RESEARCH DESIGN AND METHODS: Subjects with NIDDM were administered either aspirin (100 mg/day) or a combination of basic amino acids consisting of L-arginine (2 g/day) plus L-lysine (0.5 g/day) for 1 year. The study was double-blind and placebo-controlled. The presence and severity of retinopathy, nephropathy, and neuropathy were assessed in all subjects at 4-month intervals, as were serum blood glucose, glycohemoglobin levels, and presence of albuminuria. Collagen cross-linking and collagen glycation were measured in skin collagen obtained by biopsy at the beginning and the end of the study. Skin biopsies were also obtained from age-matched control subjects. RESULTS: Skin samples obtained from NIDDM subjects at the beginning of the study had significantly increased levels of glucitolyllysine, pentosidine, and hydroxypyridinium, as compared with age-matched control subjects. Pentosidine levels were significantly correlated with severity of retinopathy and neuropathy, but not nephropathy. Subjects receiving aspirin, but not amino acids or placebo, had significantly decreased levels of skin pentosidine after 1 year of therapy. CONCLUSIONS: It is concluded that 1) low-dose aspirin may reduce glycooxidative damage in people with NIDDM, and 2) treatment may need to continue for more than 1 year before clinical status improves.

Collagen biosynthesis.

Collagen is the major structural protein of the lung. At least five genetically distinct collagen types have been identified in lung tissue. However, the precise role of collagen in nonrespiratory lung function is not well understood, in part because of the difficulties inherent in studying lung collagen, regardless of the type of assay used. A major problem is the insolubility of lung collagen; generally less than 20% of total lung collagen can be solubilized as intact chains, even with harsh extraction procedures. Since such collagen may not be representative of total lung collagen, errors in quantitating collagen types, for example, may arise from using such material. Measurement of total lung collagen content may also pose problems, unless appropriate parameters of normalization are chosen. Biopsy dry weight, protein content, and DNA content, for example, may all change in certain disease states. Despite these difficulties, a number of changes in lung collagen have been documented in experimental pulmonary fibrosis, including increased collagen content, increased collagen synthesis rates, and changes in collagen type ratios. Many questions remain. For example, why do diverse toxic substances appear to cause essentially the same fibrotic response, even though initial sites of damage may vary? Conversely, why do similar toxic substances, such as ozone and NO2, cause diverse responses (fibrosis and emphysema, respectively)? Much work remains to be done to elucidate the mechanisms underlying the lung's choice of response.

Correlation of biochemical and morphologic manifestations of acute pulmonary fibrosis in rats administered paraquat.

Rats were administered 24 mg/kg of paraquat intraperitoneally. One, two, three, six and seven days after this injection, the rats were evaluated by several toxicologic, biochemical, and morphologic criteria. These included determinations of weight loss, visible lung hemorrhage, pulmonary edema, in vitro protein and collagen biosynthesis rates of lung tissue, and staining properties of lung sections. Morphologic evidence of mild edema was seen at day 1, while more severe edema was observed at days 2 through 7. Weight loss was apparent from day 1 throughout the experiment. In vitro protein biosynthesis rates by tissue were elevated above control values from day 3 onwards, while collagen biosynthesis rates were elevated from day 2 onwards. Increased staining with the Gomori reagent was observed from day 1, while increased staining with van Gieson's or Masson's trichrome appeared by day 2 or 3.

Source of dietary carbohydrate affects life span of Fischer 344 rats independent of caloric restriction.

Previous investigations suggest that increased life span of calorie-restricted rodents is a function of caloric intake rather than the macro- or micronutrient composition of the diet. However, the dietary source of carbohydrate has not been widely investigated. We hypothesized that the dietary carbohydrate source may affect the life span of rats independent of caloric restriction. This hypothesis was tested in male Fischer 344 rats fed ad libitum or restricted to 60% of ad libitum, an isocaloric diet containing 14% protein, 10% fat, and 66% sucrose or cornstarch. Body weights of the ad libitum- and restricted-fed sucrose rats were consistently greater throughout the experimental period compared to diet-matched animals. Food intake did not differ significantly. The survival curves of ad libitum starch- vs sucrose-fed rats were significantly different. That is, the mean, median and upper 10th percentile survival were significantly greater in the ad libitum starch- vs sucrose-fed rats (mean life span: cornstarch-fed, 720 +/- 23 days; sucrose-fed, 659 +/- 19 days). Calorie-restricted starch-fed rats had poorer early life survival, and no significant increase in mean life span compared to ad libitum cornstarch-fed animals (726 vs 720 days). These animals did, however, have the greatest upper 10th percentile survival of all four experimental groups. Mean life span of calorie-restricted sucrose-fed rats was significantly greater than that of all other groups (890 +/- 18 days). The differences in survival rates between sucrose- and cornstarch-fed animals could not be attributed to the effects of carbohydrate source on body weight, energy absorption, or on the timing and severity of the pathological lesions normally associated with aging and/or caloric restriction in this species. These data support the hypothesis that the dietary source of carbohydrate, i.e., sucrose vs cornstarch, can significantly affect life span independently of caloric intake.

Silicosis and fibrogenesis: fact and artifact.

Although the pulmonary and extrapulmonary manifestations of silicosis in humans have been extensively documented, the mechanisms by which the fibrogenic effects of silica are manifested remain obscure. In this review, both in vitro and in vivo models of silicosis are discussed, with emphasis on the potential methodological pitfalls of each. In animal models, for example, species variability, silica type and route of administration all effect the results obtained. Tissue culture work has provided evidence that the fibroblast-macrophage interaction is a key event in fibrogenesis. However, critical variables in experimental design make it difficult to compare the often conflicting results of different workers. Experimental conditions that directly affect collagen chain biosynthesis and subsequent hydroxylation of proline appear to be of particular importance. It is concluded that, in part because of methodological difficulties, there are insufficient data to draw firm conclusions regarding the effect of silica-exposed macrophages on collagen biosynthesis by fibroblasts in vitro; there are few, if any, data concerning the role of the macrophage that has ingested silica in human or animal models of silicosis.

Ozone exposure, food restriction and protein deficiency: changes in collagen and elastin in rodent lung.

Two groups of weanling or young adult rats were fed ad lib casein-based diets containing 4 or 16% protein. Food was restricted in a third group (fed the 16% protein diet) to the amount consumed daily by rats (adult or weanlings) fed the 4% diet. After 3 weeks (weanlings) or 1, 3 or 5 weeks (adults), one-half of the rats in each group were exposed to 0.64 ppm (1.28 mg/m3) of ozone for 7 days (23.5 h each day). Several parameters were then evaluated related to lung connective tissue metabolism including: (1) total lung hydroxyproline, (2) total lung elastin, (3) apparent rates for lung collagen synthesis and elastin accumulation and (4) lung and body weights. In general, the response to protein deficiency and food restriction was more pronounced than to ozone exposure. Protein deficiency and food restriction resulted in decreased lung size and collagen content. However, the ability of lung to respond to ozone (in relative terms) was not altered by changes in diet as assessed by changes in lung weight or the collagen synthetic rate.

Lung collagen and elastin after ozone exposure in vitamin B-6-deficient rats.

The effects of vitamin B-6 deficiency and ozone exposure on selected features of connective tissue metabolism in lung were investigated in groups of weanling male rats fed one of three diets: B-6-supplemented, fed ad lib; B-6-deficient, fed ad lib; or B-6-supplemented, restricted to the food intake of deficient rats for 5 weeks. Also, perinatal rat pups were studied that were nursed from dams fed one of the 3 diets from parturition to day 15 of lactation. During the final week of each experiment, half of the rats in each of the groups were exposed to 0.64 ppm of ozone (23.5 h per day). The collagen and elastin content, collagen synthesis rate, total protein synthesis rate, and lysyloxidase activity of lungs were measured. Perinatal pups rendered vitamin B-6-deficient were particularly sensitive to ozone exposure (65% died as compared to fewer than 5% of the ad lib or food-restricted controls). When L-proline incorporation into collagen and total protein was investigated using lung minces, food restriction and B-6-deficiency resulted in about one-half the incorporation normally observed. Total lung lysyl oxidase activity was also decreased in B-6-deficient and food-restricted rats compared to B-6-supplemented rats fed ad lib. Exposure to ozone resulted in increased lysyl oxidase activity and collagen synthesis in lungs from B-6-supplemented rats, but such responses were not observed in B-6-deficient or food-restricted (FR) rats exposed to ozone.

Nonenzymatic glycation of collagen in aging and diabetes.

Considerable progress has been made in our understanding of nonenzymatic glycation of collagen, and the relationship between glycation of collagen and changes in connective tissue associated with aging and diabetes. Recent studies surveyed in this review suggest the following conclusions: 1. Collagen content of early glycation products does not appear to increase throughout the life span in normal human subjects, although small increases may occur that are linked to glycemic changes. These products are increased, relative to age-matched controls, in experimental diabetes and in diabetes mellitus in collagen from virtually all tissues analyzed. 2. Collagen content of browning products increases with aging and appears to be higher in diabetic subjects than in age-matched controls. Rates of accumulation may be accelerated in subpopulations of diabetic subjects at high risk for developing complications. 3. Increases in early glycation products do not appear to be associated with alterations in collagen solubility, thermal rupture time, or mechanical strength, nor is there an association with most diabetic complications. Alterations in these products may, however, affect conformation, ligand binding, lysyl oxidase-mediated cross-linking, and interactions between collagen and other macromolecules in the extracellular matrix. 4. Increased content of browning products is associated with many physicochemical changes in collagen as well as with long-term complications in diabetes mellitus. 5. Regulatory mechanisms have been identified in vivo that may serve to control or limit the formation of glycation products. 7. Pharmacologic agents have been identified that may be able to reduce collagen content of late glycation products. Despite the progress that has been made in this field, many areas of uncertainty and controversy exist. For example, there is not yet a consensus that the browning products associated with collagen exclusively comprise advanced Maillard products derived from nonenzymatically glycated residues. There is evidence that oxidative reactions involving lipids also play a role in generating fluorophores and chromophores that may alter properties of collagen. Thus, in the extracellular matrix collagen may be continuously modified by at least three very different processes: Maillard reactions, interactions with oxidizing lipids, and enzymatically mediated cross-linking. The interrelationships between these and possibly other posttranslational modifications remain a poorly understood area of great complexity.

Influence of age and long-term dietary restriction on enzymatically mediated crosslinks and nonenzymatic glycation of collagen in mice.

This study was designed to investigate the effects of lifetime diet restriction on collagen crosslinking in skin, tail tendon, aorta, and lung in mice. Difunctional enzymatic crosslinks decreased with age in all tissues except skin, while mature crosslinks showed almost no change with age. Collagen-associated fluorescence, assayed in skin and tail tendon, increased with age, as did pentosidine, a specific advanced glycation product, in aorta. There was no change in glucitolyllysine content with age. Difunctional crosslinks, glucitolyllysine, and collagen-associated fluorescence were decreased in diet-restricted animals relative to ad libitum fed animals in some tissues at some time points; however, correlations were not observed among these different effects, or between different tissues. Diet restriction did not affect nonreducible "mature" crosslinks. These studies suggest that: (1) lifetime diet restriction is associated with decreased collagen-associated fluorescence, suggestive of advanced glycation products, in older animals; (2) age-related increases in collagen stiffening and its decrease by dietary restriction cannot be explained solely by alterations in lysyl oxidase-mediated crosslinking, the levels of which are tissue dependent; (3) lysyl oxidase-mediated crosslinking and nonenzymatic glycation of collagen are independently influenced by dietary restriction and aging.