Vol. 179, No. 3, 1991 September
BIOCHEMICAL
AND BIOPHYSICAL
30, 1991
Nonreducible
M.W.
RESEARCH COMMUNICATIONS Pages 1582-1586
Crosslink Formation in Tibia1 Dyschondroplastic Cartilage From Broiler Chicks Fed Homocysteine
Orthl,
D.A. Martinez3,
M.E. Cook2,
Growth Plate
and A.C. Vailas
Departments of 1 Nutritional and 2Poultry Sciencesand the3Biodynamics Laboratory,University of Wisconsin-Madison, Madison, Wi. 53706 Received
August
20,
1991
Summary: In the study of tibia1 dyschondroplasia, scientists have for a long time thought that an altered extracellular matrix might be involved in the etiology of the disease. The results presented in this paper show that the collagen content was increased in the dyschondroplastic cartilage when compared to normal growth plate and day-old hypertrophic cartilage. Furthermore, nom-educible crosslinks were found only in dyschondroplastic cartilage, with the greatest amounts occurring in the distal region of the lesion, approximately lo-fold higher than that found in the dyschondroplastic growth plate. Thus, intermolecular collagen bonding is altered in the extracellular matrix of dyschondroplastic cartilage. Possible models for the etiology of the disease are discussed. Q 1991 Academic
press.
Inc.
Tibial dyschondroplasia (TD) is a disease found in fast growing birds, such as broiler chickens and turkeys, in which the epiphyseal growth plate cartilage accumulates in the metaphyseal region of the tibiotarsus. Apparently, the balance between cartilage formation and cartilage resorption with subsequent bone formation occurring in endochondral ossification has been perturbed. One theory proposed to explain the disease is that defective chondrocytes produce an abnormal cartilage matrix which is not resorbed (1). Research has shown that the percentage of proteoglycan and collagen in the dyschondroplastic cartilage are comparable to those found in normal growth plate cartilage (2). However, the scientists did not study how these macromolecules are distributed in the extracellular matrix. Hydroxylysylpyridinoline and lysylpyridinoline (HP,LP) are nom-educible, acidstable, intermolecular collagen crosslinks that are used as indices of connective tissue maturity. Collagen crosslinking will alter the spatial distribution of collagen fibrils leading to new biochemical and biomechanical purpose of our study was to determine the degree of collagen in epiphyseal growth plate cartilage, tibia1 dyschondroplastic 0006-291X/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1582
properties (3). The crosslinking found cartilage, and day-
Vol.
179,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
old hypertrophic cartilage from broiler chicks. Day-old hypertrophic cartilage was chosen because it is morphologically similar to the hypertrophic zone of normal growth plate cartilage (4). MAT-
AND METHODS
ale collectioa; Indian River broiler chicks were fed a standard cornsoybean based diet with or without 0.48% DL-homocystine. Homocystine at this level has been shown to be a potent inducer of TD (5). At the end of four weeks, the birds were killed by cervical dislocation and the tibiotarsus was removed and stored at 4OC in phosphate-buffered saline for several hours. Normal growth plate cartilage, free of bone, was collected from the proximal end of the tibiotarsus. In bones with white, opaque, dyschondroplastic cartilage protruding into the metaphyseal region, samples were taken from the growth plate, from a layer of cartilage right underneath the growth plate, and from a layer of cartilage just above the chondro-osseous junction. Newly hatched broiler chicks were killed by carbon dioxide asphyxiation and the hypertrophic cone from the tibiotarsus was removed. All samples were weighed and then stored at -55OC until further analysis. Before biochemical analysis, samples were lyophilized to determine the percentage dry weight. le preparatioll; Small samples of cartilage (l-10 mg) for HP and LP analysis were first extracted with 4M Guanidine HCl, 50 mM EDTA, 50 mM sodium acetate, pH 5.9, at 4% for 48 hours. The samples were then washed three times with distilled water over a four hour period. This was done to partially purify the collagen by removing the proteoglycans (6). All samples were hydrolyzed in 6M HCl for 24 hours at 1lOQ.Z in sealed ampules (Wheaton, Millville, NJ). Hydrolyzed samples were dried in a Savant Speed- Vat (Savant, Farmingdale, NY). HP an.d JIP analvsis; Hydrolyzed samples were resuspended in 1.0 ml of 1% HFBA (n-heptafluorobutyric acid, Pierce Chemical Co., Rockford, IL), filtered through a nylon 66 membrane 0.45 pm pore size (Gelman Sciences, Inc., Ann Arbor, MI), and centrifuged for 5 minutes through a 1.5 ml microfilterfuge tube, 0.22 pm pore size (Ranin Instrument Co., Woburn, MA). The filtrate was then transferred to amber glass vials and applied to the chromatography system. Samples were analyzed using a Waters Model 45 system, with a Waters WISP 712 autoinjector, and a Waters Model 470 spectrofluorometer, containing a xenon lamp source and 16 ~1 quartz flow cell (18 nm bandwidth). HP and LP crosslinks were monitored at 2901 (excitation) and 3951 (emission) with 2pmole sensitivity. Crosslink analysis was performed using a modification of the HPLC method developed by Eyre, et al. (7). The samples (loo-2OOpg of cartilage) were eluted through an Altex Ultrasphere ODS column (5 mm; 25cmx4.6mm) protected by a Brownlee guard cartridge (RP-18 spheri-5). Acetonitrile at 19% with O.OlM HFBA at a flow rate of 1.0 ml/min was used to elute HP and LP from the column. The chromatograms were stored and integrated on a CompuAdd 216 AT clone using Maxima 820 software (Dynamics Solutions, Milford, MA). Pyridoxamine 2HCl (Sigma, St. Louis, MO) was used as a standard since it has a similar fluorescence spectra as the HP crosslink with a molar fluorescence yield 4.62 times that of HP in our HPLC system. The collagen crosslinks were then expressed as mole crosslink/mole of collagen. droxvnrolineysis; Analysis of hydroxyprolme was accomplished by using the derivatizing agent, phenylisothiocyanate (PITC), according to the method of Dunphy, et al. (8). Hydrolyzed samples were resuspended in 1.0 ml of water, filtered as above, and applied to a solid-phase extraction medium (Sepak Light, Waters, Milford, MA) with 500-550~1 of methanol: water (1:l v/v) and eluted with 1.5 ml of O.lN HCl to purify the samples prior to derivatizing. The HPLC 1583
Vol.
BIOCHEMICAL
179, No. 3, 1991
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
system was the same as above except that a lOcmx4.6mm Cl8 PICO-TAG column (Waters) and a Waters Model 440 absorbance detector at 2545 were used. Hydroxyproline chromatograms were quantified relative to known amount of derivatized hydroxyproline standard (Sigma). Collagen concentrations were then determined. . based on the fact that 14% of collagen by weight is hydroxyproline. Statlstlcal Data were analyzed by one-way analysis of variance and least square differences using the general linear model (SAS Institute Inc., State College, Pa.) with the level of significance for rejecting the null hypothesis set at the 5% probability level. RESULTS
AND DISCUSSION
The dyschondroplastic cartilage lesion had a decrease in the percentage of dry weight when compared to growth plate cartilage (Table 11, which agrees with published results (4). Unlike previous findings (21, the collagen content increases (38% over that in the growth plate) ii-om the dyschondroplastic growth plate to the distal region of the lesion. The discrepancy could be due to the different methods used for measuring hydroxyproline. The previous work, using a calorimetric method, did show a tendency toward an increased percentage of collagen in the lesion (2). Both normal growth plate and day-old hypertrophic cartilage had no detectable HP crosslinks (Figure 1). For the normal growth plate, it is not surprising since the turnover time for chondrocytes in young broiler chicks is probably less than 48 hours (9). A more rapid collagen turnover time has been shown to impede crosslink formation (10). A very low degree of crosslinking would be functionally beneficial in the hypertrophic cartilage since it needs to be quickly resorbed and vascularized. Dyschondroplastic cartilage contained similar amounts of HP crosslinks in the growth plate and proximal lesion. The distal region of the lesion had a greater than ten-fold increase in HP over that found in the proximal region
Table 1. The relationship between cartilage type and collagen content and crosslinking Y0Di-y w&t Normal growth plate@) 12.95a TD growth plate (6) 12.27a Proximal TD lesion(8) 10.8gb Distal TD lesion (7) 10.76b tvoe (n1
Day-old hypertrophic (8) -Pooled SEM
0.37
Collagen content2 82.gc 82.1c 103.0b ll3.5a 88.3c 2.5
HP & ND4 .033b .030b .354a ND .016
1 Mean values within a column not followed by a common letter differ significantly (p
BIOCHEMICAL
AND BIOPHYSICAL
30, 1991
Nonreducible
M.W.
RESEARCH COMMUNICATIONS Pages 1582-1586
Crosslink Formation in Tibia1 Dyschondroplastic Cartilage From Broiler Chicks Fed Homocysteine
Orthl,
D.A. Martinez3,
M.E. Cook2,
Growth Plate
and A.C. Vailas
Departments of 1 Nutritional and 2Poultry Sciencesand the3Biodynamics Laboratory,University of Wisconsin-Madison, Madison, Wi. 53706 Received
August
20,
1991
Summary: In the study of tibia1 dyschondroplasia, scientists have for a long time thought that an altered extracellular matrix might be involved in the etiology of the disease. The results presented in this paper show that the collagen content was increased in the dyschondroplastic cartilage when compared to normal growth plate and day-old hypertrophic cartilage. Furthermore, nom-educible crosslinks were found only in dyschondroplastic cartilage, with the greatest amounts occurring in the distal region of the lesion, approximately lo-fold higher than that found in the dyschondroplastic growth plate. Thus, intermolecular collagen bonding is altered in the extracellular matrix of dyschondroplastic cartilage. Possible models for the etiology of the disease are discussed. Q 1991 Academic
press.
Inc.
Tibial dyschondroplasia (TD) is a disease found in fast growing birds, such as broiler chickens and turkeys, in which the epiphyseal growth plate cartilage accumulates in the metaphyseal region of the tibiotarsus. Apparently, the balance between cartilage formation and cartilage resorption with subsequent bone formation occurring in endochondral ossification has been perturbed. One theory proposed to explain the disease is that defective chondrocytes produce an abnormal cartilage matrix which is not resorbed (1). Research has shown that the percentage of proteoglycan and collagen in the dyschondroplastic cartilage are comparable to those found in normal growth plate cartilage (2). However, the scientists did not study how these macromolecules are distributed in the extracellular matrix. Hydroxylysylpyridinoline and lysylpyridinoline (HP,LP) are nom-educible, acidstable, intermolecular collagen crosslinks that are used as indices of connective tissue maturity. Collagen crosslinking will alter the spatial distribution of collagen fibrils leading to new biochemical and biomechanical purpose of our study was to determine the degree of collagen in epiphyseal growth plate cartilage, tibia1 dyschondroplastic 0006-291X/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
1582
properties (3). The crosslinking found cartilage, and day-
Vol.
179,
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
old hypertrophic cartilage from broiler chicks. Day-old hypertrophic cartilage was chosen because it is morphologically similar to the hypertrophic zone of normal growth plate cartilage (4). MAT-
AND METHODS
ale collectioa; Indian River broiler chicks were fed a standard cornsoybean based diet with or without 0.48% DL-homocystine. Homocystine at this level has been shown to be a potent inducer of TD (5). At the end of four weeks, the birds were killed by cervical dislocation and the tibiotarsus was removed and stored at 4OC in phosphate-buffered saline for several hours. Normal growth plate cartilage, free of bone, was collected from the proximal end of the tibiotarsus. In bones with white, opaque, dyschondroplastic cartilage protruding into the metaphyseal region, samples were taken from the growth plate, from a layer of cartilage right underneath the growth plate, and from a layer of cartilage just above the chondro-osseous junction. Newly hatched broiler chicks were killed by carbon dioxide asphyxiation and the hypertrophic cone from the tibiotarsus was removed. All samples were weighed and then stored at -55OC until further analysis. Before biochemical analysis, samples were lyophilized to determine the percentage dry weight. le preparatioll; Small samples of cartilage (l-10 mg) for HP and LP analysis were first extracted with 4M Guanidine HCl, 50 mM EDTA, 50 mM sodium acetate, pH 5.9, at 4% for 48 hours. The samples were then washed three times with distilled water over a four hour period. This was done to partially purify the collagen by removing the proteoglycans (6). All samples were hydrolyzed in 6M HCl for 24 hours at 1lOQ.Z in sealed ampules (Wheaton, Millville, NJ). Hydrolyzed samples were dried in a Savant Speed- Vat (Savant, Farmingdale, NY). HP an.d JIP analvsis; Hydrolyzed samples were resuspended in 1.0 ml of 1% HFBA (n-heptafluorobutyric acid, Pierce Chemical Co., Rockford, IL), filtered through a nylon 66 membrane 0.45 pm pore size (Gelman Sciences, Inc., Ann Arbor, MI), and centrifuged for 5 minutes through a 1.5 ml microfilterfuge tube, 0.22 pm pore size (Ranin Instrument Co., Woburn, MA). The filtrate was then transferred to amber glass vials and applied to the chromatography system. Samples were analyzed using a Waters Model 45 system, with a Waters WISP 712 autoinjector, and a Waters Model 470 spectrofluorometer, containing a xenon lamp source and 16 ~1 quartz flow cell (18 nm bandwidth). HP and LP crosslinks were monitored at 2901 (excitation) and 3951 (emission) with 2pmole sensitivity. Crosslink analysis was performed using a modification of the HPLC method developed by Eyre, et al. (7). The samples (loo-2OOpg of cartilage) were eluted through an Altex Ultrasphere ODS column (5 mm; 25cmx4.6mm) protected by a Brownlee guard cartridge (RP-18 spheri-5). Acetonitrile at 19% with O.OlM HFBA at a flow rate of 1.0 ml/min was used to elute HP and LP from the column. The chromatograms were stored and integrated on a CompuAdd 216 AT clone using Maxima 820 software (Dynamics Solutions, Milford, MA). Pyridoxamine 2HCl (Sigma, St. Louis, MO) was used as a standard since it has a similar fluorescence spectra as the HP crosslink with a molar fluorescence yield 4.62 times that of HP in our HPLC system. The collagen crosslinks were then expressed as mole crosslink/mole of collagen. droxvnrolineysis; Analysis of hydroxyprolme was accomplished by using the derivatizing agent, phenylisothiocyanate (PITC), according to the method of Dunphy, et al. (8). Hydrolyzed samples were resuspended in 1.0 ml of water, filtered as above, and applied to a solid-phase extraction medium (Sepak Light, Waters, Milford, MA) with 500-550~1 of methanol: water (1:l v/v) and eluted with 1.5 ml of O.lN HCl to purify the samples prior to derivatizing. The HPLC 1583
Vol.
BIOCHEMICAL
179, No. 3, 1991
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
system was the same as above except that a lOcmx4.6mm Cl8 PICO-TAG column (Waters) and a Waters Model 440 absorbance detector at 2545 were used. Hydroxyproline chromatograms were quantified relative to known amount of derivatized hydroxyproline standard (Sigma). Collagen concentrations were then determined. . based on the fact that 14% of collagen by weight is hydroxyproline. Statlstlcal Data were analyzed by one-way analysis of variance and least square differences using the general linear model (SAS Institute Inc., State College, Pa.) with the level of significance for rejecting the null hypothesis set at the 5% probability level. RESULTS
AND DISCUSSION
The dyschondroplastic cartilage lesion had a decrease in the percentage of dry weight when compared to growth plate cartilage (Table 11, which agrees with published results (4). Unlike previous findings (21, the collagen content increases (38% over that in the growth plate) ii-om the dyschondroplastic growth plate to the distal region of the lesion. The discrepancy could be due to the different methods used for measuring hydroxyproline. The previous work, using a calorimetric method, did show a tendency toward an increased percentage of collagen in the lesion (2). Both normal growth plate and day-old hypertrophic cartilage had no detectable HP crosslinks (Figure 1). For the normal growth plate, it is not surprising since the turnover time for chondrocytes in young broiler chicks is probably less than 48 hours (9). A more rapid collagen turnover time has been shown to impede crosslink formation (10). A very low degree of crosslinking would be functionally beneficial in the hypertrophic cartilage since it needs to be quickly resorbed and vascularized. Dyschondroplastic cartilage contained similar amounts of HP crosslinks in the growth plate and proximal lesion. The distal region of the lesion had a greater than ten-fold increase in HP over that found in the proximal region
Table 1. The relationship between cartilage type and collagen content and crosslinking Y0Di-y w&t Normal growth plate@) 12.95a TD growth plate (6) 12.27a Proximal TD lesion(8) 10.8gb Distal TD lesion (7) 10.76b tvoe (n1
Day-old hypertrophic (8) -Pooled SEM
0.37
Collagen content2 82.gc 82.1c 103.0b ll3.5a 88.3c 2.5
HP & ND4 .033b .030b .354a ND .016
1 Mean values within a column not followed by a common letter differ significantly (p
