Analyses of lysine aldehyde cross-linking in collagen reveal that the mature cross-link histidinohydroxylysinonorleucine is an artifact.

Lysyl oxidase-generated intermolecular cross-links are essential for the tensile strength of collagen fibrils. Two cross-linking pathways can be defined, one based on telopeptide lysine aldehydes and another on telopeptide hydroxylysine aldehydes. Since the 1970s it has been accepted that the mature cross-linking structures on the lysine aldehyde pathway, which dominates in skin and cornea, incorporate histidine residues. Here, using a range of MS-based methods, we re-examined this conclusion and found that telopeptide aldol dimerization is the primary mechanism for stable cross-link formation. The C-telopeptide aldol dimers formed labile addition products with glucosylgalactosyl hydroxylysine at α1(I)K87 in adjacent collagen molecules that resisted borohydride reduction and after acid hydrolysis produced histidinohydroxylysinonorleucine (HHL), but only from species with a histidine in their α1(I) C-telopeptide sequence. Peptide MS analyses and the lack of HHL formation in rat and mouse skin, species that lack an α1(I) C-telopeptide histidine, revealed that HHL is a laboratory artifact rather than a natural cross-linking structure. Our experimental results also establish that histidinohydroxymerodesmosine is produced by borohydride reduction of N-telopeptide allysine aldol dimers in aldimine intermolecular linkage to nonglycosylated α1(I) K930. Borohydride reduction of the aldimine promotes an accompanying base-catalyzed Michael addition of α1(I) H932 imidazole to the α,ß-unsaturated aldol. These aldehydes are intramolecular at the N terminus but at the C terminus they can be both intramolecular and intermolecular according to present and earlier findings.

Focus on O-phosphohydroxylysine, O-phosphohydroxyproline, N 1-phosphotryptophan and S-phosphocysteine.

The synthesis and chemistry of the lesser-known phosphoamino acids, O-phosphohydroxylysine, O-phosphohydroxyproline, N 1-phosphotryptophan and S-phosphocysteine are described in detail. In addition, where anything at all is known, the biological synthesis, occurrence and functions of these phosphoamino acids are described. Of these phosphoamino acids, only N 1-phosphotryptophan has not been reported to occur in proteins; however, apart from the roles of S-phosphocysteine in the sugar transporter component (EII) and in catalysis by protein phosphotyrosine phosphatase, little is currently known about the biological roles of the phosphoamino acids when they occur as post-translational modifications.

Synthesis of an azido precursor to (2S,5R)-5-hydroxylysine using an asymmetric organocatalytic chlorination/reduction sequence.

An efficient, robust, and scalable synthesis of an azido precursor to the modified amino acid (2S,5R)-5-hydroxylysine was developed on the basis of the use of a highly stereoselective organocatalytic α-chlorination-reduction protocol. The final Fmoc-protected (2S,5R)-6-azido-5-hydroxylysine derivative can be used in solid-phase peptide synthesis, providing access to proteins that contain large quantities of post-translationally modified lysine (e.g., collagens).

The third activity for lysyl hydroxylase 3: galactosylation of hydroxylysyl residues in collagens in vitro.

Lysyl hydroxylase (LH, EC 1.14.11.4), galactosyltransferase (EC 2.4.1.50) and glucosyltransferase (EC 2.4.1.66) are enzymes involved in posttranslational modifications of collagens. They sequentially modify lysyl residues in specific positions to hydroxylysyl, galactosylhydroxylysyl and glucosylgalactosyl hydroxylysyl residues. These structures are unique to collagens and essential for their functional activity. Lysines and hydroxylysines form collagen cross-links. Hydroxylysine derived cross-links, usually as glycosylated forms, occur especially in weight-bearing and mineralized tissues. The detailed functions of the hydroxylysyl and hydroxylysyl linked carbohydrate structures are not known, however. Hydroxylysine linked carbohydrates are found mainly in collagens, but recent reports indicate that these structures are also present and probably have an important function in other proteins. Earlier we have shown that human LH3, but not isoforms LH1, LH2a and LH2b, possesses both LH and glucosyltransferase activity (J. Biol. Chem. 275 (2000) 36158). In this paper we demonstrate that galactosyltransferase activity is also associated with the same gene product, thus indicating that one gene product can catalyze all three consecutive steps in hydroxylysine linked carbohydrate formation. In vitro mutagenesis experiments indicate that Cys(144) and aspartates in positions 187-191 of LH3 are important for the galactosyltransferase activity. Our results suggest that manipulation of the gene for LH3 can be used to selectively alter the glycosylation and hydroxylation reactions, and provides a new tool to clarify the functions of the unique hydroxylysine linked carbohydrates in collagens and other proteins.

Biological variability in assessing the clinical value of biochemical markers of bone turnover.

Analytical and biological variability of three bone markers, deoxypyridinoline (DPD), CrossLaps (CTx) and galactosylhydroxylysine (GHYL) were compared. From 14 healthy subjects (six women, eight men; age 29-44 years) recruited from our laboratory staff, two sets of samples of early morning urine were obtained - four samples taken weekly for 4 weeks (all subjects) and three samples taken monthly for 3 months from five subjects. Data were expressed as the ratio to creatinine concentration. All the methods met the analytical goals (CV(A)< or =1/2CV(I(within-subject))) DPD 0.06, CTx 0.05 and GHYL 0.07 with CV(I(within-subject)) being 0.22, 0.19 and 0.38, respectively. The reference values were of limited usefulness particularly for CTx and GHYL, the index of individuality (II) being 0.50 and 0.48 respectively. As the index of heterogeneity (IH) was not significant, being 0.23 for CTx, 0.28 for DPD and 0.46 for GHYL, which are all <1.71 (1+2S.D.), within-subject variances can be used to calculate the reference change value (RCV): 0.58 for DPD, 0.54 for CTx and 1. 08 for GHYL. Moreover, we found constant variations in DPD and CTx, week to week and month to month. Our findings suggest that DPD and CTx provide more reliable results than GHYL, showing a lower within-subject variation, a lower and time-constant RCV allowing reliable monitoring without regard for timing.

A proposal for standardizing urine collections for bone resorption markers measurement.

Diurnal variations in the excretion of bone resorption markers were assessed in order to identify the type of urine collection which provides the most information on bone resorption rate and its relation to measuring bone dynamics in a postmenopausal population. Sixty women, ages 43-67 and without disease or treatment known to affect bone mineral density, were divided into two groups on the basis of femoral mineral density T-score: <1.5 (Group I), >1.5 (Group II). Bone formation was assessed by measuring bone alkaline phosphatase activity and osteocalcin concentration, bone resorption by urinary hydroxyproline, pyridinoline and deoxypiridinoline, N-telopeptide, galactosyl hydroxylysine, and CrossLaps. To identify the more appropriate collection times, urine samples were collected from 7 am to 3 pm; from 3 pm to 11 pm; from 11 pm to 7 am. Twenty-four hour urine collection and first morning void urine samples were also measured. The findings suggest that nocturnal collection and first morning void samples provide the most reliable data on the rate of bone degradation, possibly showing bone loss not only in osteopenic patients but also in women with a low T-score. Nocturnal and first morning samples should therefore be recommended in order to standardize sample collection, as they enable an accurate assessment of bone resorption markers and improved comparability to results from different studies, as well as a less cumbersome collection modality.

Ovariectomy in the rat induces a rapid increase in the urinary excretion of hydroxylysine glycosides and non-reducible crosslink residues.

The ovariectomized rat is the most commonly used animal model of human postmenopausal osteoporosis, exhibiting a high rate of bone turnover with resorption exceeding formation. At present, bone turnover is quantified directly by dynamic histomorphometry. The aim of the present study was to determine whether the measurement of the urinary output of some specific bone collagen catabolites--pyridinolines and hydroxylysine glycosides--could be used to indirectly monitor the initial phase of bone turnover increase in ovariectomized 90-day-old rats. Ninety-day-old female rats were randomly divided into three groups (n = 6): ovariectomized, sham-operated and non-treated controls. Urine samples (24 h) were collected 6 days before surgery and twice weekly for the 4 weeks following ovariectomy. Urinary excretion of pyridinoline (PYD), deoxypyridinoline (DPD), glucosyl-galactosyl-hydroxylysine (GGHYL) and galactosyl-hydroxylysine (GHYL) were measured. As expected, ovariectomy was associated with a significant decrease in bone mineral density in both the proximal tibial and distal femoral metaphysis. Compared with both sham-operated and control animals, ovariectomized rats showed significant increases in PYD, GGHYL, and GHYL urinary output 8 days after surgery and in DPD output after 15 days. These changes were maintained throughout the study. The results confirm that measurement of the urinary excretion of pyridinolines and hydroxylysine glycosides represents a powerful tool for detecting the onset of bone turnover in ovariectomized 90-day-old rats.

Longitudinal changes in bone mineral density and bone turnover in postmenopausal women with primary hyperparathyroidism.

The aims of this study were to determine 1) whether primary hyperparathyroidism (PHPT) is associated with accelerated bone loss in postmenopausal women, 2) whether bone mineral density (BMD) and bone turnover change to a similar extent with surgery and hormone replacement therapy (HRT) in these patients, and 3) whether biochemical markers of bone turnover measured at baseline can be used to predict the change in BMD in these patients after different therapies. We studied 33 postmenopausal women with PHPT; their ages at the time of study ranged from 48-80 yr (mean +/- SD, 63 +/- 10). Total body (TB), lumbar spine (LS), and femoral neck (FN) BMD and biochemical markers of bone turnover were measured at baseline and 10-30 months (19 +/- 5) after parathyroid surgery, HRT, or no treatment. BMD was measured in 33 age-matched healthy controls at baseline and at a mean of 24 months. Baseline biochemical markers of bone turnover were measured in controls. In PHPT at baseline, the mean z-score of BMD was -1.25 at TB (95% confidence interval, -1.64 to -0.86), -0.95 at LS (-1.37 to -0.53), and -1.30 at FN (-1.65 to -0.95), whereas the mean z score was 0.45 for serum carboxy-terminal propeptide of human type I procollagen (0.02-0.89), 1.05 for bone alkaline phosphatase (0.38-1.71), 2.38 for 24-h urinary excretion of cross-linked N-terminal telopeptide of type I collagen (NTx; 1.63-3.13), and 2.36 for 24-h urinary excretion of galactosyl hydroxylysine (1.97-2.74). After surgery and HRT, BMD increased and bone turnover decreased during the follow-up. In the untreated group, BMD decreased at TB and FN, and levels of bone alkaline phosphatase, NTx/creatinine, and galactosyl hydroxylysine/creatinine increased. When the rate of change in BMD (percentage per yr) was compared with that in the control group, bone gain was significant at all three skeletal sites after surgery and HRT, and bone loss was significant at TB and FN, but not at LS, in the untreated group. There was a weak, but significant, correlation between baseline urinary NTx and the change in femoral neck BMD in the untreated group (r = -0.36; P = 0.05). We conclude that untreated postmenopausal women with PHPT have low BMD resulting from accelerated bone loss at the TB and FN. Surgery and HRT both restore BMD and bone turnover toward normal in postmenopausal women with PHPT. A single measurement of bone turnover is insufficient to predict BMD changes in individual patients with PHPT.

Comparison between urinary pyridinium cross-links and hydroxylysine glycosides in monitoring the effects of ovariectomy and 17 beta-estradiol replacement in aged rats.

This study was undertaken to assess the sensitivity of hydroxylysylpyridinoline (HP), lysylpyridinoline (LP), galactosylhydroxylysine (GHyl) and glucosylgalactosylhydroxylysine (GGHyl) to monitor bone response to estrogen deficiency and replacement by comparing their excretory patterns in ovariectomized aged (11-14 months old) rats. The ovariectomized (OVX) rats were randomized into two groups: (1) OVX plus vehicle; (2) OVX plus 17 beta-estradiol (17-beta E, 10 micrograms/kg, s.c., 4 days/week). Treatment with 17-beta E started immediately after OVX and continued for 60 days. The collagen catabolites were measured in urine for 1 month before OVX and thereafter for 60 days. In temporal coincidence with urine collection, bone area and bone mineral density (BMD) of lumbar vertebrae, femoral diaphysis and distal metaphysis were measured by dual-energy X-ray absorptiometry. In the untreated rats, BMD of the femoral metaphysis and lumbar vertebrae decreased significantly and the urinary excretion of LP, HP, GHyl and GGHyl increased with different patterns. In the treated rats, 17-beta E replacement prevented the increment in LP excretion, partially prevented the increase in HP excretion, but had no effect on the excretion of GHyl and GGHyl. In conclusion pyridinolines and glycosides have different sensitivities to the bone response to OVX. Glycoside excretion after OVX also reflects metabolic processes not strictly related to bone loss and, in contrast with LP, is not sensitive to estrogen replacement.

Composition and posttranslational modification of individual collagen chains from osteosarcomas and osteofibrous dysplasias.

The composition of collagen was analyzed and the degree of lysyl hydroxylation of individual collagen chains was determined in four osteosarcomas and two osteofibrous dysplasias. In addition, the tumor proliferation (number of mitoses, proliferating-nuclear-antigen-positive cells, MIB) as well as the response to chemotherapy (morphological regression grade) were checked. All tumors contained a high proportion of collagen III and, in all but one osteosarcoma, pepsin-extracted collagens I and III were overmodified. Furthermore, the proportion of diglycosides in collagen I was about four times higher than in controls. The collagen composition and modification resembled those of bones at early stages of human development. One osteosarcoma and both osteofibrous dysplasias were in the normal range of lysyl hydroxylation. There was no correlation between the collagen properties and the histopathological marker of tumor proliferation.

Effect of bisphosphonate therapy and parathyroidectomy on the urinary excretion of galactosylhydroxylysine in primary hyperparathyroidism.

BACKGROUND: In patients with mild or asymptomatic primary hyperparathyroidism a reliable index of bone resorption might be useful for appropriate management. Hydroxyproline is the most commonly used marker of bone resorption but its low specificity and sensitivity are known. Galactosylhydroxylysine, an amino acid mainly represented in bone collagen, has been proposed as a more suitable index of bone resorption. In this study we evaluated the sensitivity of galactosylhydroxylysine and hydroxyproline assays in following the changes of their urinary levels in 12 patients with mild primary hyperparathyroidism before and after treatment with bisphosphonate and surgery. METHODS: Serum and fasting urine specimens were obtained from 12 women with mild primary hyperparathyroidism before and after bisphosphonate treatment (2.5 mg daily for 5 days, intravenously) and after a further 25 days; in 7 patients biochemical tests were also performed 1 and 6 days after parathyroidectomy. Galactosylhydroxylysine was assayed by an HPLC method and hydroxyproline by a RIA commercial kit. RESULTS: Baseline galactosylhydroxylysine urinary levels were far above the normal range in all the patients whilst in 8 of them baseline hydroxyproline levels were normal. Bisphosphonate treatment significantly decreased bone turnover as shown by a significant fall in serum calcium (from 2.9 to 2.6 mmol/l; P < 0.001) and in galactosylhydroxylysine and hydroxyproline (-55 and -31% respectively). Twenty-five days after the end of treatment, resorption increased again and serum calcium and galactosylhydroxylysine, but not hydroxyproline, rose significantly towards basal levels. One day after parathyroidectomy serum calcium, galactosylhydroxylysine and PTH showed reduction below normal ranges. PTH and galactosylhydroxylysine returned to normal values at day 6 after parathyroidectomy. No changes in hydroxyproline levels were seen. Galactosylhydroxylysine, but not hydroxyproline, correlated significantly with serum calcium and PTH. CONCLUSION: Galactosylhydroxylysine appears to be a sensitive index of bone resorption, useful in the clinical assessment of bone involvement and in the management of patients with mild primary hyperparathyroidism.

Biochemical markers for detecting bone metastases in patients with breast cancer.

A study was carried out to assess the best use of biochemical bone markers to exclude metastases in patients with breast cancer. Urinary galactosyl-hydroxylysine and serum alkaline phosphatase were used to monitor bone resorption and deposition, respectively. Hydroxyproline was also measured. In a selected population of patients, possibly affected by metastases on the basis of scintigraphic examination, which is highly sensitive but poorly specific, we assessed the efficiency of the markers by a double statistical analysis. In this group, the only marker able to predict metastases was galactosyl-hydroxylysine.

A study of osteoporosis as it relates to metabolic manifestations in edentulous women.

Among some patients, regardless of age, the jaw loses bone mass, leading to loosening and falling out of otherwise healthy teeth. This study seeks to establish whether this bone loss is associated with the metabolic manifestations of other forms of localized decalcifications, such as in Paget's disease, or with generalized osteoporosis. Sixteen women being fitted with dental implants to compensate for bone losses provided 24-hour urine samples for the quantitative determination of calcium and galactosyl hydroxylysine, a bone collagen metabolite. These patients provided demographic information, relevant medical, dental, and dietary history, a profile of their current medications, and the status of their smoking and exercise habits. Urinary excretion of galactosyl hydroxylysine, which is increased in the presence of progressive increased bone resorption, remained within normal values in the patients of this study. These results suggest that the thinning of the jaw bones and subsequent tooth loss of these subjects were osteoporotic processes too limited and too localized to produce measurable increases in urinary bone metabolites.

Amino acid analysis of collagen hydrolysates by reverse-phase high-performance liquid chromatography of 9-fluorenylmethyl chloroformate derivatives.

A recently described procedure for amino acid analyses has been modified and adapted for use in quantitating the unique mixture of products commonly found in hydrolysates of the collagens. The method involves precolumn derivatization of hydrolysates with 9-fluorenylmethyl chloroformate (FMOC-CL), chromatographic separation of the derivatives and excess reagent on a reverse-phase column, and quantitation based on the fluorescent properties of the derivatives. The method takes advantage of the ease with which stable derivatives are formed with the FMOC reagent. Using a ternary gradient system, a complete amino acid analysis with good resolution of all components can be performed within 35 min. The sensitivity of the method is comparable to levels attained by other derivatives and the fluorescence response of each derivative is linear over the total range of 1-800 pmol. Given these parameters, the method allows complete amino acid analyses to be performed on 100 ng of collagen corresponding to a single picomole of a collagen chain (Mr 100,000).

High predictivity of galactosyl-hydroxylysine in urine as an indicator of bone metastases from breast cancer.

We measured the urinary excretion of galactosyl-hydroxylysine (GH) and hydroxyproline in two groups of women with breast cancer, with (M+, n = 24) and without (Mo, n = 30) clinical, scintigraphic, or radiological evidence of bone metastases. Both these compounds are excreted in larger amounts in the M+ group than in the Mo patients. However, GH, which is a specific marker for bone collagen, provides better predictivity for bone metastases than does hydroxyproline: 92% sensitivity and 90% specificity vs 74% and 79%, respectively, for hydroxyproline.

Collagen metabolite excretion as a predictor of bone- and skin-related complications in spinal cord injury.

Immediately after the trauma, spinal cord injury patients have an increased rate of collagen synthesis and an even greater increase of collagen degradation. Collagen lost from bone is implicated in the etiology of osteoporosis and heterotopic ossification, and collagen lost from skin might lead to a propensity to develop pressure ulcers. The urinary excretion of two collagen metabolites has been monitored and their fluctuation related to the onset of skin or bone complications in these patients. One metabolite, glucosyl-galactosyl hydroxylysine, is more abundant in skin collagen; the other, galactosyl hydroxylysine, is more abundant in bone collagen. The excretion of both metabolites increases after injury, reaching a peak between three and six months after injury, and declines gradually, reaching control values about a year after injury. If a skin pressure ulcer develops, the urinary excretion of the diglycoside remains elevated instead of gradually decreasing. Similarly, if osteoporosis or heterotopic ossification is diagnosed, the monoglycoside excretion does not return to control values until the bone turnover stabilizes. Monitoring of the urinary excretion of both glycosides might prove helpful in prompting early examination to establish the presence of emerging skin and bone complications. Thus, aggressive preventive therapy could be given sooner.

Identification of the post-translational modifications of the core-specific lectin. The core-specific lectin contains hydroxyproline, hydroxylysine, and glucosylgalactosylhydroxylysine residues.

The core-specific lectin (CSL) synthesized and secreted by rat hepatocytes and the rat hepatoma H-4-II-E shows affinity for mannose and N-acetylglucosamine residues in the "core" region of asparagine-linked oligosaccharides. The CSL undergoes two stages of post-translational modification which result in an increase in its Mr from 24,000 to 26,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. We have determined that the lectin undergoes hydroxylation of proline and lysine and that the hydroxylysine is glycosylated to form glucosylgalactosylhydroxylysine (GlcGalHyLys). CSL metabolically labeled with [3H]lysine and [3H]proline contains hydroxylated forms of proline and lysine. The mature form of the lectin can also be metabolically labeled with [3H]galactose. alpha,alpha'-Dipyridyl, an inhibitor of collagen prolyl and lysyl hydroxylases, prevents the metabolic incorporation of [3H]galactose and the post-translational increases in the Mr of the CSL, indicating that both events are dependent upon hydroxylation of proline and lysine. Virtually all of the hydroxylysine present in the CSL is recovered as glucosylgalactosylhydroxylysine after alkaline hydrolysis. The post-translational modifications of the CSL place it in a select family of secreted proteins which contain collagen-like sequences, including the pulmonary surfactant proteins, complement component C1q, and the 18 S asymmetric form of acetylcholinesterase.

Characterization of collagen from normal human sclera.

Human scleral tissue contains approximately 50% collagen by weight, consisting predominantly of type I collagen. There is little or no evidence for the presence of substantial quantities of type II, type III or other collagen types. There appears to be no difference in either collagen content or genetic type in sclera between adult and juvenile tissues or between anterior and posterior segments of the sclera, although, with increased age there is a marked increase both in the extent of glycosylation of the collagen and its resistance to solubilization by treatment with pepsin.