- Email: [email protected]
The effects of Bordetella avium infection on elastin and collagen content of turkey trachea and aorta
The Effects of Bordetella avium Infection on Elastin and Collagen Content of Turkey Trachea and Aorta1 A. G. YERSIN,*,2 F. W. EDENS,*,3 and D. G. SIMMONS†
ABSTRACT Turkey poults were inoculated at hatch with the “W” isolate of Bordetella avium. At 17 d of age, serum copper levels and ceruloplasmin activities were determined. The trachea and aorta were analyzed for collagen and elastin content in an attempt to relate these structural proteins to the clinical observations of tracheal ring distortion and cardiac dysfunction associated with bordetellosis. Serum copper levels and ceruloplasmin activity were elevated in the B. avium-infected poults and indicated enzyme activity sufficient for elastin and collagen cross-link formation. In the infected poults, crude elastin content was increased significantly (0.67% infected vs 0.59% control) in the trachea but not in the aorta (13.12% infected vs 12.68% control). However, collagen content in infected poults (69.7 hydroxyproline residues per 1,000 amino acid residues) was decreased in
the trachea compared to the controls (97 hydroxyproline residues per 1,000 amino acid residues), whereas collagen and elastin cross-links (HLNL, hydroxylysinohydroxy-norleucine, moles per mole of collagen per 300 residues hydroxyproline) were increased in the trachea of infected poults (2.85 in infected vs 1.80 in control) and also increased (DHLNL, dihydroxylysinohydroxy-norleucine, moles/mole of collagen/300 residues hydroxyproline) in the aorta (0.49 in infected vs 0.39 in control) of infected poults. The differences in collagen and elastin content, in association with differences in the cross-linking, appeared to be the cause of tracheal collapse that is characteristic of B. avium infection and also may have an adverse influence on cardiovascular function.
(Key words: Bordetella avium, trachea, collagen, elastin, aorta) 1998 Poultry Science 77:1654–1660
INTRODUCTION Bordetella avium is the etiologic agent of turkey coryza or bordetellosis (Simmons et al., 1979; Simmons, 1984). Clinical signs of the upper respiratory disease in turkeys have been well documented and include an oculo-nasal discharge, rhinitis, dyspnea, and tracheal collapse. Reduced weight gain has been documented (Simmons et al., 1979; Simmons, 1984). The association between B. avium colonization in ciliated trachea epithelium and development of lesions and cellular pathology also are well documented (Gray et al., 1983; Arp and Cheville, 1984; Arp and Fagerland, 1987). However, it has been difficult to characterize the cause of tracheal collapse
Received for publication March 2, 1998. Accepted for publication June 24, 1998. 1The use of trade names in this publication does not imply endorsement of the product named nor criticism of similar products not mentioned. 2Current address: HY-VAC Laboratory Egg Co., 2147 HWY 6, PO Box 285, Adel, IA 50003. 3To whom correspondence should be addressed: fwedens @mindspring.com
and of the tracheal ring distortion commonly seen in infected birds. The articulated hyaline cartilage of tracheal rings (Hodges, 1974) and other connective tissues such as blood vessels are composed of matrix proteins that contribute to their structural and mechanical properties (Tinker and Rucker, 1985). Two predominant proteins in connective tissue are collagen, which provides mechanical and tensile properties, and elastin, which provides recoil properties (Tinker and Rucker, 1985). The structural integrity of the connective tissue is dependent on the quartinary structure of these proteins and the formation of extensive intermolecular covalent crosslinks (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). Alterations in cardiac function defined by electrocardiograms have been observed in association with B. avium infection (Yersin et al., 1991). Part of this cardiac disturbance can be attributed to increased sensitivity of
Abbreviation Key: DES = desmosine; DHNL = dihydroxylysinohydroxy-norleucine; HLNL = hydroxy-lysinohydroxy-norleucine; ISDES = isodesmosine.
1654
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
*North Carolina State University, Department of Poultry Science, Raleigh, North Carolina 27695-7635 and †American Veterinary Medical Association, 1931 North Meacham Road, Suite 100, Schaumburg, Illinois 60173-4360
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN
MATERIALS AND METHODS
Birds Seventy-six 1-d-old, Large White Nicholas turkey poults of both sexes were obtained from the North Carolina Agricultural Research Service turkey research unit. Poults were divided randomly into two groups consisting of 38 control and 38 infected birds and placed in metal growing batteries in separate environmentally controlled isolation rooms. The poults were maintained at an environmental temperature of 32 C, and ambient humidity varied from 45 to 65% over an experimental period of 17 d. The North Carolina Agricultural Research Service turkey starter diet and water were available for ad libitum consumption.
Infection The challenged poults were inoculated at 1 d of age, through the delivery of 0.05 mL of a 24 h brain heart infusion broth4 culture of the “W” isolate of B. avium (approximately 1 × 107 bacteria). The inoculum was administered by placing two drops on the external nares followed by forced closed-mouthed breathing by the poults.
Analyses At 17 d after B. avium challenge, poults from both the Control and infected groups were killed by cervical dislocation and the following tests performed. Histology. Tracheal ring sections were collected from six control and six infected turkeys (three poults per treatment by two replicates) and fixed in 10% neutral, buffered formalin. The formalin-fixed segments were embedded in paraffin, sectioned to a 5 mm thickness, and stained with hematoxylin and eosin, Masson’s Trichrome (collagen), and Verhoeffs van Geisen (elastin) for examination by light microscopy. Copper and Ceruloplasmin. Serum was obtained from 20 control and 20 infected poults (10 poults per treatment by two replicates). Copper concentrations were 4Difco
Laboratories, Detroit, MI 48232-7058.
determined by atomic absorption analysis after diluting the serum with an equal volume of deionized water (Association of Official Analytical Chemists, 1970). Ceruloplasmin activity was expressed in International Units and was reported as activity in micromoles of product formed from the oxidation of pphenylenediamine per hour per liter (Harris and DiSilvestro, 1981). Crude Elastin. Aorta and trachea segments were obtained from the same 20 control and 20 infected poults that had been bled for copper and ceruloplasmin analysis. A segment of the aorta was removed, cleaned of adhering tissue, blotted, and weighed (Hill et al., 1967). The segment included the common carotids to the point of bifurcation into the subclavian arteries and the aorta from the base of the heart to a point on the aortic arch equal in length to the carotids. A 1-cm segment of the trachea immediately below the larynx was removed, cleaned of adhering tissue, blotted, and weighed. Both aorta and tracheal tissues were cut into small pieces and extracted in hot 0.1 N NaOH for 90 min, washed in distilled water, and dried at 50 C overnight in a vacuum oven (Hill et al., 1967). The dried material was weighed and the amount of elastin was determined as a percentage of the wet weight (Starcher, 1976). Amino Acid and Cross-Link Analyses. Aorta and trachea segments as described above were obtained from the remaining 24 poults; 12 control and 12 infected (6 poults per treatment in each of two replicates). The tissues were weighed and frozen immediately by immersion in liquid nitrogen. Aorta and trachea segments were pulverized to a powder and washed three times with cold 0.02 M phosphate buffer (pH 7.4) followed by centrifugation (15,000 rpm/20 min). Residues obtained were dialyzed against cold distilled water (twice daily for 2 d) and lyophilized. Aliquots of the dried aorta and trachea were suspended in 0.15 M TES buffer, (pH 7.4) and reduced with standardized NaB3(H)4 using the method described by Yamauchi et al. (1986). The reduced samples were hydrolyzed in 6 M HCl in vacuo and an aliquot of the hydrolysate was subjected to amino acid analysis (Yamauchi et al., 1986). Relative content of collagen was expressed as hydroxyproline residues per 1,000 amino acid residues, and elastin content, as isodesmosine (ISDES) and desmosine (DES) residues, were expressed as leucine equivalent per 1,000 amino acids. The latter two cross-links are separated from other amino acids and cross-links (Yamauchi et al., 1988ab). Reducible cross-links of collagen were quantified in moles of cross-link per mole of collagen.
Statistical Analysis The data for serum copper, ceruloplasmin, and crude tissue elastin content were subjected to analysis of variance procedures (SAS Institute, 1991). Percentages were converted to angles (arc sine) and analyzed. Differences between treatment means for amino acid and
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
the heart and associated structures to increased levels of catecholamines. These changes may be due to increased tissue turnover of catecholamines and the possible upregulation of adrenergic receptors (Edens et al., 1987). The observed cardiac disturbances may be influenced also by changes in the connective tissue matrices of the vessels of the heart as well as in the cardiac muscle. The objective of this study was to examine the elastin and collagen content of trachea and aorta in B. aviuminfected vs noninfected control turkey poults and to relate the associated clinical signs with a pathophysiological effect on these tissues.
1655
1656
YERSIN ET AL.
TABLE 1. Serum copper levels and ceruloplasmin activity from poults infected with Bordetella avium Treatment
Copper
Ceruloplasmin
Control Infected
(mg/mL) 24.44 ± 1.27b 33.65 ± 1.68a
(IU/mL) 4.76 ± 0.26b 9.59 ± 0.48a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
RESULTS Serum copper concentrations for the control and infected poults are presented in Table 1. The B. avium infected poults had significantly greater copper concentrations (33.65 ± 1.68 mg/mL) than the control poults (25.44 ± 1.27 mg/mL; Table 1). Serum ceruloplasmin activity in infected poults was significantly greater (9.59 ± 0.48 IU) than the activity in control poults (4.76 ± 0.26 IU; Table 1). The crude elastin content of the aorta and trachea as determined by hot alkaline extraction is presented in Table 2. The trachea elastin content was significantly greater in the infected poults (0.67%) than in their control counterparts (0.59%). However, the crude elastin content of the aorta was not significantly different between the two groups. The relative concentrations of collagen and elastin based on amino acid analysis from the two treatment groups are presented in Tables 3 and 4, respectively. The trachea collagen content of the infected poults was significantly lower (69.7 hydroxyproline residues per 1,000 amino acids) than the control group (97.0 residues), but there was no difference between the two groups in aorta collagen content. The elastin content of both trachea and aorta from the treatment groups was expressed as the cross-links of ISDES and DES residues (leucine equivalent per 1,000 amino acids). In the trachea, both the ISDES and DES cross-links in the infected poults were increased significantly, but in the aorta, only the ISDES cross-links were decreased. However, the ratio of the tracheal and aorta cross-links (ISDES:DES) in control and infected poults were not significantly different (Table 4).
DISCUSSION The primary function of connective tissue is support, and this function is dependent on the physiochemical and biomechanical properties of the connective tissue matrix proteins (Yamauchi and Mechanic, 1988). Two of the most extensively studied proteins are collagen and elastin (Tinker and Rucker, 1985). The tensile strength and integrity of the connective tissue are dependent on the extensive formation of covalent intermolecular crosslinks. Alteration in cross-linking has been shown to be involved with various types of connective tissue disorders induced by genetic selection, disease, or dietary imbalances (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). The biosynthesis of collagen and elastin begins intracellularly with the transcription and translation of collagen- and elastin-specific mRNA (Tinker and Rucker, 1985). Hydroxylation reactions result in directed fibril formation occurring extracellularly. The crosslinking reactions are initiated, resulting in tissue specific cross-links comprised of specific amino acid residues.
TABLE 2. Crude elastin content of trachea and aorta from poults infected with Bordetella avium TABLE 3. Amino acid residue analysis: Collagen in trachea and aorta from turkey poults infected with Bordetella avium: hydroxyproline residues per 1,000 amino acid residues
Elastin Treatment Control Infected
Trachea 0.59 ± 0.03a 0.67 ± 0.02b
Aorta (% wet weight) 12.68 ± 0.75a 13.12 ± 0.82a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
Treatment
Trachea
Aorta
Control Infected
97.0 ± 4.9a 69.7 ± 2.9b
29 ± 1.5a 27 ± 1.4a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
cross-link analysis data were determined by Student’s t test. Statements of significance are based upon P ≤ 0.05.
The cross-links of collagen, expressed as moles per mole of collagen, are shown in Table 5. The cross-links associated with collagen are dihydroxylysinonorleucine (DHLNL) and hydroxylysinonorleucine (HLNL). There were no significant differences between the control and infected poults for DHLNL in trachea tissue. However, there was a significant difference between control and infected trachea cross links for HLNL and for DHLNL cross links in the aorta. Values for the HLNL from aorta were not obtained due to the presence of many unidentified radioactive peaks on the chromatogram. Representative histopathologies for trachea from infected and control poults are shown in Figures 1a and 1b and 2a and 2b, respectively. There was a loss of ciliated epithelium in the trachea of the infected poults, but not in the control poults (Figures 2a and 2b). The elastin fibers that are associated with the cartilage matrix can be observed along the proximal border of the hyaline ring. The fibers in the control trachea appeared uniform in distribution and comprise a network with overlapping arrangement. The fibers from the infected trachea appeared condensed, knotted, and gave an appearance of lost organization. Collagen fibers, which stain green with Masson’s Trichrome, were not easily discriminated with light microscopy.
1657
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN TABLE 4. Amino acid residue (isodesmosine1 and desmosine2) analysis: elastin in trachea and aorta from turkey poults infected with Bordetella avium Trachea
Aorta
Treatment
ISDES
DES
Ratio
ISDES
DES
Ratio
Control Infected
0.45 ± 0.02b 0.60 ± 0.03a
0.45 ± 0.02b 0.60 ± 0.03a
1 1
3.24 ± 0.16a 2.78 ± 0.14b
2.07 ± 0.11a 1.94 ± 0.09a
1.565a 1.433a
± SEM, in a column, with no common superscript differ significantly (P ≤ 0.05). = isodesmosine, residues/1,000 amino acid residues. 2DES = desmosine, residues/1,000 amino acid residues. a,bMeans 1ISDES
The results of this study indicated that both copper concentration and ceruloplasmin activity were elevated in the B. avium-infected poults. Other studies have shown that ceruloplasmin activity can be increased due to a general stress response. The induction of the enzyme has been accomplished by the administration of adrenocorticotropic hormone, hydrocortisone, infectious agents, and bacterial endotoxin (Starcher and Hill, 1965; Butler and Curtis, 1973; Freeman et al., 1973; Curtis and Butler, 1980). It is known that B. avium significantly elevates the corticosterone concentration during the period of infection (McCorkle et al., 1985), and therefore, it can be anticipated that ceruloplasmin would increase as well during the course of the bordetellosis. Additionally, it has been demonstrated that bordetellosis increases turnover of tissue catecholamines and there is evidence to suggest that adrenergic receptors that bind catecholamines may be up regulated during the disease (Edens et al., 1987). It is also known that catecholamines induce significant increases in plasma ceruloplasmin activity (Freeman et al., 1973), and it is known that reserpine depletes tissue stores of catecholamines by increasing secretion rates, and inhibiting reuptake caused a significant increase in plasma ceruloplasmin activity in chickens (Freeman et al., 1973). Therefore, in cases of bordetellosis characterized by elevated plasma levels of corticosterone, increased turnover of tissue catecholamines, increased copper availability, and increased ceruloplasmin activity, lysyl oxidase activity should have increased, with resultant increases in cross link activity. This hypothesis was supported by the data presented herein.
TABLE 5. Collagen cross-link analysis for trachea and aorta from turkey poults infected with Bordetella avium Trachea
Aorta
Treatment
DHLNL1
HLNL2
DHLNL
HLNL
Control Infected
1.49 ± 1.63 ±
1.80 ± 2.85 ±
0.39 ± 0.49 ±
NA3 NA
0.12a 0.13a
0.04b 0.06a
0.01b 0.02a
± SEM, in a column, with no common superscript differ significantly (P ≤ 0.05). = dihydroxy-lysinohydroxy-norleucine, moles/mole of collagen/300 residues hydroxyproline. 2HLNL = hydroxylysinonorleucine, moles/mole of collagen/300 residues hydroxyproline. 3NA = not obtained. a,bMeans
1DHLNL
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
The initiation of cross-linking for both collagen and elastin is dependent on the enzyme lysyl oxidase (Siegal et al., 1970; Tinker and Rucker, 1985). In collagen and elastin, specific e-amino groups of peptidyl lysine and hydroxylysine residues are oxidatively deaminated to yield aldehyde derivatives (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). The condensation of these aldehyde derivatives yields specific classes of cross-links that are used for identification of collagen and elastin, respectively. For collagen, the cross-links are HLNL, DHLNL, and pyridinoline (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). For elastin, the cross-links are stable DES and ISDES derivatives (Tinker and Rucker, 1985). The role that lysyl oxidase plays is extremely important, especially as the enzyme’s activity is dependent on copper, which acts as a co-factor in the various reactions (Martin et al., 1970; Harris et al., 1977). Copper deficiency, due to disease or diet, leads to impaired ability of the host to produce adequate tissue elastin (Hill et al., 1967; Harris et al., 1977; Rayton and Harris, 1979). Additionally, ceruloplasmin, an acute phase protein and antioxidant in blood, functions in a transport/delivery capacity for copper and is associated with the mineral’s metabolism (Harris and DiSilvestro, 1981). There are correlations among the activity of ceruloplasmin, copper concentrations, and activity of lysyl oxidase (Hill et al., 1967; Harris and DiSilvestro, 1981). Due to the enzyme’s short half-life, the functional capacity of the enzyme may be assessed by the activity of serum ceruloplasmin and serum copper concentration.
1658
YERSIN ET AL.
FIGURE 1. Photomicrographs (400× ) of tracheal epithelial borders from Bordetella avium-infected (top) and control (bottom) turkey poults. In the B. avium-infected trachea there is a total loss of cilia along the border of the tracheal epithelium and this is contrasted with the ciliated epithelium of the control (indicated by the arrow).
Crude elastin was measured in both aorta and trachea to assess the integrity of the connective tissue. In the aorta, there was no difference in elastin content between the infected and control groups, but tracheal elastin was increased significantly in the B. avium group (Table 2). The amino acid analysis (Table 4) revealed that there was actually a reduced amount of cross-link components (ISDES) in the aorta of the infected poults. The crosslinks establish a quantitative difference specifically for the elastin content. Therefore, the reduced amount of cross-links in the aorta from infected poults suggested a less resilient tissue. At necropsy, the aorta from infected poults was flaccid, distorted, and collapsed and appeared to have lost structural integrity as compared to that of controls. The loss of structural integrity in the aorta from infected poults may be related to the decreased elastin cross-links found in this study. The collagen content (hydroxyproline residues per 1,000 amino acid residues) of the aorta was not decreased in the infected poults (Table 3). The crosslinks analysis revealed, however, that there was actually
FIGURE 2. Photomicrographs (400×) of elastin within tracheas from Bordetella avium-infected (top) and control (bottom) turkey poults. The B. avium infection has altered the normal arrangement of the elastin fibers causing them to be nonuniform and to be concentrated in bunches rather than to be uniformly distributed (indicated by large arrow in each panel). Small arrow indicates ciliated epithelium in the control.
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
an increase in DHLNL, one of the immature derivatives (Table 5). The DHLNL cross-link analysis indicated that collagen had been synthesized recently in response to tissue injury or in response to decreased elastin crosslinks (Table 4). Therefore, at 17 d of age, it appeared that the aorta was under repair and was in the early stages of recovery from the disease. Bordetellosis-associated dyspnea is indicative of reduced oxygen availability due to tracheal occlusion with excessive mucus and/or trachea collapse. These factors cause an increase in cardiac work and hypertrophy of the structures and vessels of the heart. The small increases in the collagen cross-links may herald the development of cardiac hypertrophy in response to bordetellosis. The crude elastin content of the trachea was increased significantly in the infected poults. This finding paralleled the amino acid cross-link analysis for DES and ISDES in infected poults. These observations
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN
REFERENCES Arp, L. W., and N. F. Cheville, 1984. Tracheal lesions in young turkeys infected with Bordetella avium. Am. J. Vet. Res. 45: 2196–2200.
Arp, L. W., and J. A. Fagerland, 1987. Ultrastructural pathology of Bordetella avium infection in turkeys. Vet. Pathol. 24:411–418. Association of Official Analytical Chemists, 1970. Official Methods of Analysis. 11th ed. Association of Official Analytical Chemists, Washington, DC. Butler, E. J., and M. J. Curtis, 1973. The effects of Escherichia coli endotoxin and ACTH on plasma zinc concentration in the domestic fowl. Res. Vet. Sci. 15:363–367. Curtis, M. J., and E. J. Butler, 1980. Response of caeruloplasmin to Escherichia coli endotoxins and adrenal hormones in the domestic fowl. Res. Vet. Sci. 28:217–222. Edens, F. W., F. M. McCorkle, D. G. Simmons, and A. G. Yersin, 1987. Brain monoamine concentrations in turkey poults infected with Bordetella avium. Avian Dis. 31: 504–508. Freeman, B. M., A.C.C. Manning, and D. S. Pole, 1973. Factors affecting plasma ceruloplasmin activity in Gallus domesticus. Comp. Biochem. Physiol. 45A:689–698. Gentry-Weeks, C. R., B. T. Cookson, W. E. Goldman, R. B. Rimler, S. P. Porter, and R. Curtiss, III, 1988. Dermonecrotic toxin and tracheal cytotoxin, putative virulence factors of Bordetella avium. Infect. Immun. 56:1698–1707. Gray, J. G., J. F. Roberts, R. C. Dillman, and D. G. Simmons, 1983. Pathogenesis of change in the upper respiratory tracts of turkeys experimentally infected with an Alcaligenes faecalis isolate. Infect. Immun. 42:350–355. Harris, E. D., and R. A. DiSilvestro, 1981. Correlation of lysyl oxidase with p-phenylenediamine oxidase activity (ceruloplasmin) in serum (41102). Proc. Soc. Exp. Biol. Med. 166:528–531. Harris, E. D., J. K. Rayton, and T. E. Degroot, 1977. A critical role for copper in aortic elastin structure and synthesis. Pages 543–565 in: Elastin and Elastic Tissue. I. A. Sanberg, ed. L. B. Plenum Press, New York, NY. Hill, C. H., B. Starcher, and C. Kim, 1967. Role of copper in the formation of elastin. Fed. Proc. 26:129–133. Hodges, R. D., 1974. The respiratory system. Pages 113–139 in: Histology of the Fowl. Academic Press, London, UK. Krane, S. M., 1981. Connective tissue. Pages 755–798 in: Pathophysiology: The Biological Principles of Disease. L. H. Smith, and S. O. Thier, ed. W. B. Saunders Co., Philadelphia, PA. Martin, G. R., S. R. Pinnell, R. C. Siegal, and E. R. Goldstein, 1970. Lysyl oxidase: the enzymatic step in collagen and elastin synthesis. Pages 405–410 in: Chemistry and Molecular Biology of the Intracellular Matrix. Vol. 1. E. A. Balazs, ed. Academic Press, New York, NY. McCorkle, F. M., F. W. Edens, and D. G. Simmons, 1985. Alcaligenes faecalis infection in turkeys: effects on serum corticosterone and serum chemistry. Avian Dis. 29:80–89. Rayton, J. K., and E. D. Harris, 1979. Induction of lysyl oxidase with copper: properties of an in vitro system. J. Biol. Chem. 254:621–626. SAS Institute, 1991. A User’s Guide to the Statistical Analysis System: Statistics. SAS Institute Inc., Cary, NC. Siegal, R. C., S. R. Pinnell, and G. R. Martin, 1970. Crosslinking of collagen and elastin properties of lysyl oxidase. Biochemistry 23:4486–4492. Simmons, D. G., 1991. Turkey Coryza. Pages 251–256 in: Diseases of Poultry. 8th ed. M. S. Hofstad, ed. Iowa State University Press, Ames, IA. Simmons, D. G., J. G. Gray, L. P. Rose, R. C. Dillman, and S. E. Miller, 1979. Isolation of an etiological agent of acute
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
strongly suggest an increase in elastin synthesis during the course of infection. One might argue that the trachea ring distortion, normally observed with the infection, would be better explained by the opposite effect or a decrease in elastin resulting in loss of stability of the trachea ring and ultimate collapse. However, collagen is the primary matrix protein associated with the cartilage structure of the trachea (Tinker and Rucker, 1985). The current study showed that the tracheal collagen content was reduced significantly during Bordetella infection compared to the uninfected Control. Similar to the responses of collagen and elastin in the aorta, when primary connective tissue protein is decreased, the secondary protein appears to respond with increased synthesis. These responses of trachea collagen and elastin appeared to be in response to injury caused by the bacteria or bacterial toxin(s) (Gentry-Weeks et al., 1988). Histological observations of the trachea collagen and elastin show that elastin in the infected poults was not deposited in a regular pattern along the border of the tracheal cartilage. In the infected poults, the elastin fibers in the trachea sections appeared to be knotted and irregular. It is possible that the irregular pattern may be a reflection of increased cross-linkage. The elastin arrangement was similar to the formation of scar tissue in wound healing (Krane, 1981). Bordetella avium produces a tracheal-cytotoxin, but its mechanism of action, other than being toxic to the ciliated epithelium, has not been determined (GentryWeeks et al., 1988). It is possible that the bacteria or toxin(s) induce serum proteases or collagenases to actively degrade the connective tissue matrix, and this degeneration could be the mechanism for the decreased elastin and collagen content of the aorta and trachea, respectively. However, one can not discount the possibility that collagen and elastin content of the trachea and aorta may be antigenic targets of the host immune response to infection. With initiation of the recovery from inflammation associated with bordetellosis, synthesis of new cross-links would be required to re-establish structural integrity of both the aorta and trachea. Results presented herein suggest that in cases of bordetellosis, there is a decrease in tracheal collagen but increases in trachea elastin. The increased elastin content of the trachea represents mature cross-linking. However, there is an increase also in the number of collagen and elastin cross-links in the immature types as well. The disease-associated increases in trachea (HLNL) and aorta (DHLNL) elastin cross linking (Table 5) with decreased collagen in the trachea, and lesser amounts in the aorta (Table 3) suggests a mechanism that may explain disease-associated tracheal collapse and flaccid texture of the aorta.
1659
1660
YERSIN ET AL. Yamauchi, M., E. P. Katz, and G. L. Mechanic, 1986. Intermolecular Cross-linking and Stereospecific molecular packing in Type I collagen fibrils of the periodental ligament. Biochem. 25:4907–4913. Yamauchi, M., D. T. Woodley, and G. L. Mechanic, 1988a. Aging and cross-linking of skin collagen. Biochem. Biophys. Res. Commun. 152:898–903. Yamauchi, M., D. R. Young, G. S. Chandler, and G. L. Mechanic, 1988b. Cross-linking and new bone collagen synthesis in immobilized and recovering primate osteoporosis. Bone 9:415–418. Yersin, A. G., F. W. Edens, and D. G. Simmons, 1991. Bordetella avium-infection effects on electrocardiograms in turkey poults. Avian Dis. 35:668–673.
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
respiratory disease (rhinotracheitis) of turkey poults. Avian Dis. 23:194–203. Starcher, B. C., 1976. Determination of the elastin content of tissues by measuring desmosine and isodesmosine. Anal. Biochem. 79:11–15. Starcher, B. C., and C. H. Hill, 1965. Hormonal Induction of ceruloplasmin in chicken serum. Comp. Biochem. Physiol. 15:429–434. Tinker, D., and R. B. Rucker, 1985. Role of selected nutrients in synthesis, accumulation, and chemical modification of connective tissue proteins. Physiol. Rev. 65:607–657. Yamauchi, M., and G. L. Mechanic, 1988. Cross-linking of collagen. Pages 157–172 in: Collagen. Vol. 1. M. E. Nimni, ed. CRC Press, Boca Raton, FL.
ABSTRACT Turkey poults were inoculated at hatch with the “W” isolate of Bordetella avium. At 17 d of age, serum copper levels and ceruloplasmin activities were determined. The trachea and aorta were analyzed for collagen and elastin content in an attempt to relate these structural proteins to the clinical observations of tracheal ring distortion and cardiac dysfunction associated with bordetellosis. Serum copper levels and ceruloplasmin activity were elevated in the B. avium-infected poults and indicated enzyme activity sufficient for elastin and collagen cross-link formation. In the infected poults, crude elastin content was increased significantly (0.67% infected vs 0.59% control) in the trachea but not in the aorta (13.12% infected vs 12.68% control). However, collagen content in infected poults (69.7 hydroxyproline residues per 1,000 amino acid residues) was decreased in
the trachea compared to the controls (97 hydroxyproline residues per 1,000 amino acid residues), whereas collagen and elastin cross-links (HLNL, hydroxylysinohydroxy-norleucine, moles per mole of collagen per 300 residues hydroxyproline) were increased in the trachea of infected poults (2.85 in infected vs 1.80 in control) and also increased (DHLNL, dihydroxylysinohydroxy-norleucine, moles/mole of collagen/300 residues hydroxyproline) in the aorta (0.49 in infected vs 0.39 in control) of infected poults. The differences in collagen and elastin content, in association with differences in the cross-linking, appeared to be the cause of tracheal collapse that is characteristic of B. avium infection and also may have an adverse influence on cardiovascular function.
(Key words: Bordetella avium, trachea, collagen, elastin, aorta) 1998 Poultry Science 77:1654–1660
INTRODUCTION Bordetella avium is the etiologic agent of turkey coryza or bordetellosis (Simmons et al., 1979; Simmons, 1984). Clinical signs of the upper respiratory disease in turkeys have been well documented and include an oculo-nasal discharge, rhinitis, dyspnea, and tracheal collapse. Reduced weight gain has been documented (Simmons et al., 1979; Simmons, 1984). The association between B. avium colonization in ciliated trachea epithelium and development of lesions and cellular pathology also are well documented (Gray et al., 1983; Arp and Cheville, 1984; Arp and Fagerland, 1987). However, it has been difficult to characterize the cause of tracheal collapse
Received for publication March 2, 1998. Accepted for publication June 24, 1998. 1The use of trade names in this publication does not imply endorsement of the product named nor criticism of similar products not mentioned. 2Current address: HY-VAC Laboratory Egg Co., 2147 HWY 6, PO Box 285, Adel, IA 50003. 3To whom correspondence should be addressed: fwedens @mindspring.com
and of the tracheal ring distortion commonly seen in infected birds. The articulated hyaline cartilage of tracheal rings (Hodges, 1974) and other connective tissues such as blood vessels are composed of matrix proteins that contribute to their structural and mechanical properties (Tinker and Rucker, 1985). Two predominant proteins in connective tissue are collagen, which provides mechanical and tensile properties, and elastin, which provides recoil properties (Tinker and Rucker, 1985). The structural integrity of the connective tissue is dependent on the quartinary structure of these proteins and the formation of extensive intermolecular covalent crosslinks (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). Alterations in cardiac function defined by electrocardiograms have been observed in association with B. avium infection (Yersin et al., 1991). Part of this cardiac disturbance can be attributed to increased sensitivity of
Abbreviation Key: DES = desmosine; DHNL = dihydroxylysinohydroxy-norleucine; HLNL = hydroxy-lysinohydroxy-norleucine; ISDES = isodesmosine.
1654
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
*North Carolina State University, Department of Poultry Science, Raleigh, North Carolina 27695-7635 and †American Veterinary Medical Association, 1931 North Meacham Road, Suite 100, Schaumburg, Illinois 60173-4360
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN
MATERIALS AND METHODS
Birds Seventy-six 1-d-old, Large White Nicholas turkey poults of both sexes were obtained from the North Carolina Agricultural Research Service turkey research unit. Poults were divided randomly into two groups consisting of 38 control and 38 infected birds and placed in metal growing batteries in separate environmentally controlled isolation rooms. The poults were maintained at an environmental temperature of 32 C, and ambient humidity varied from 45 to 65% over an experimental period of 17 d. The North Carolina Agricultural Research Service turkey starter diet and water were available for ad libitum consumption.
Infection The challenged poults were inoculated at 1 d of age, through the delivery of 0.05 mL of a 24 h brain heart infusion broth4 culture of the “W” isolate of B. avium (approximately 1 × 107 bacteria). The inoculum was administered by placing two drops on the external nares followed by forced closed-mouthed breathing by the poults.
Analyses At 17 d after B. avium challenge, poults from both the Control and infected groups were killed by cervical dislocation and the following tests performed. Histology. Tracheal ring sections were collected from six control and six infected turkeys (three poults per treatment by two replicates) and fixed in 10% neutral, buffered formalin. The formalin-fixed segments were embedded in paraffin, sectioned to a 5 mm thickness, and stained with hematoxylin and eosin, Masson’s Trichrome (collagen), and Verhoeffs van Geisen (elastin) for examination by light microscopy. Copper and Ceruloplasmin. Serum was obtained from 20 control and 20 infected poults (10 poults per treatment by two replicates). Copper concentrations were 4Difco
Laboratories, Detroit, MI 48232-7058.
determined by atomic absorption analysis after diluting the serum with an equal volume of deionized water (Association of Official Analytical Chemists, 1970). Ceruloplasmin activity was expressed in International Units and was reported as activity in micromoles of product formed from the oxidation of pphenylenediamine per hour per liter (Harris and DiSilvestro, 1981). Crude Elastin. Aorta and trachea segments were obtained from the same 20 control and 20 infected poults that had been bled for copper and ceruloplasmin analysis. A segment of the aorta was removed, cleaned of adhering tissue, blotted, and weighed (Hill et al., 1967). The segment included the common carotids to the point of bifurcation into the subclavian arteries and the aorta from the base of the heart to a point on the aortic arch equal in length to the carotids. A 1-cm segment of the trachea immediately below the larynx was removed, cleaned of adhering tissue, blotted, and weighed. Both aorta and tracheal tissues were cut into small pieces and extracted in hot 0.1 N NaOH for 90 min, washed in distilled water, and dried at 50 C overnight in a vacuum oven (Hill et al., 1967). The dried material was weighed and the amount of elastin was determined as a percentage of the wet weight (Starcher, 1976). Amino Acid and Cross-Link Analyses. Aorta and trachea segments as described above were obtained from the remaining 24 poults; 12 control and 12 infected (6 poults per treatment in each of two replicates). The tissues were weighed and frozen immediately by immersion in liquid nitrogen. Aorta and trachea segments were pulverized to a powder and washed three times with cold 0.02 M phosphate buffer (pH 7.4) followed by centrifugation (15,000 rpm/20 min). Residues obtained were dialyzed against cold distilled water (twice daily for 2 d) and lyophilized. Aliquots of the dried aorta and trachea were suspended in 0.15 M TES buffer, (pH 7.4) and reduced with standardized NaB3(H)4 using the method described by Yamauchi et al. (1986). The reduced samples were hydrolyzed in 6 M HCl in vacuo and an aliquot of the hydrolysate was subjected to amino acid analysis (Yamauchi et al., 1986). Relative content of collagen was expressed as hydroxyproline residues per 1,000 amino acid residues, and elastin content, as isodesmosine (ISDES) and desmosine (DES) residues, were expressed as leucine equivalent per 1,000 amino acids. The latter two cross-links are separated from other amino acids and cross-links (Yamauchi et al., 1988ab). Reducible cross-links of collagen were quantified in moles of cross-link per mole of collagen.
Statistical Analysis The data for serum copper, ceruloplasmin, and crude tissue elastin content were subjected to analysis of variance procedures (SAS Institute, 1991). Percentages were converted to angles (arc sine) and analyzed. Differences between treatment means for amino acid and
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
the heart and associated structures to increased levels of catecholamines. These changes may be due to increased tissue turnover of catecholamines and the possible upregulation of adrenergic receptors (Edens et al., 1987). The observed cardiac disturbances may be influenced also by changes in the connective tissue matrices of the vessels of the heart as well as in the cardiac muscle. The objective of this study was to examine the elastin and collagen content of trachea and aorta in B. aviuminfected vs noninfected control turkey poults and to relate the associated clinical signs with a pathophysiological effect on these tissues.
1655
1656
YERSIN ET AL.
TABLE 1. Serum copper levels and ceruloplasmin activity from poults infected with Bordetella avium Treatment
Copper
Ceruloplasmin
Control Infected
(mg/mL) 24.44 ± 1.27b 33.65 ± 1.68a
(IU/mL) 4.76 ± 0.26b 9.59 ± 0.48a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
RESULTS Serum copper concentrations for the control and infected poults are presented in Table 1. The B. avium infected poults had significantly greater copper concentrations (33.65 ± 1.68 mg/mL) than the control poults (25.44 ± 1.27 mg/mL; Table 1). Serum ceruloplasmin activity in infected poults was significantly greater (9.59 ± 0.48 IU) than the activity in control poults (4.76 ± 0.26 IU; Table 1). The crude elastin content of the aorta and trachea as determined by hot alkaline extraction is presented in Table 2. The trachea elastin content was significantly greater in the infected poults (0.67%) than in their control counterparts (0.59%). However, the crude elastin content of the aorta was not significantly different between the two groups. The relative concentrations of collagen and elastin based on amino acid analysis from the two treatment groups are presented in Tables 3 and 4, respectively. The trachea collagen content of the infected poults was significantly lower (69.7 hydroxyproline residues per 1,000 amino acids) than the control group (97.0 residues), but there was no difference between the two groups in aorta collagen content. The elastin content of both trachea and aorta from the treatment groups was expressed as the cross-links of ISDES and DES residues (leucine equivalent per 1,000 amino acids). In the trachea, both the ISDES and DES cross-links in the infected poults were increased significantly, but in the aorta, only the ISDES cross-links were decreased. However, the ratio of the tracheal and aorta cross-links (ISDES:DES) in control and infected poults were not significantly different (Table 4).
DISCUSSION The primary function of connective tissue is support, and this function is dependent on the physiochemical and biomechanical properties of the connective tissue matrix proteins (Yamauchi and Mechanic, 1988). Two of the most extensively studied proteins are collagen and elastin (Tinker and Rucker, 1985). The tensile strength and integrity of the connective tissue are dependent on the extensive formation of covalent intermolecular crosslinks. Alteration in cross-linking has been shown to be involved with various types of connective tissue disorders induced by genetic selection, disease, or dietary imbalances (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). The biosynthesis of collagen and elastin begins intracellularly with the transcription and translation of collagen- and elastin-specific mRNA (Tinker and Rucker, 1985). Hydroxylation reactions result in directed fibril formation occurring extracellularly. The crosslinking reactions are initiated, resulting in tissue specific cross-links comprised of specific amino acid residues.
TABLE 2. Crude elastin content of trachea and aorta from poults infected with Bordetella avium TABLE 3. Amino acid residue analysis: Collagen in trachea and aorta from turkey poults infected with Bordetella avium: hydroxyproline residues per 1,000 amino acid residues
Elastin Treatment Control Infected
Trachea 0.59 ± 0.03a 0.67 ± 0.02b
Aorta (% wet weight) 12.68 ± 0.75a 13.12 ± 0.82a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
Treatment
Trachea
Aorta
Control Infected
97.0 ± 4.9a 69.7 ± 2.9b
29 ± 1.5a 27 ± 1.4a
a,bMeans ± SEM in a column with no common superscript differ significantly (P ≤ 0.05).
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
cross-link analysis data were determined by Student’s t test. Statements of significance are based upon P ≤ 0.05.
The cross-links of collagen, expressed as moles per mole of collagen, are shown in Table 5. The cross-links associated with collagen are dihydroxylysinonorleucine (DHLNL) and hydroxylysinonorleucine (HLNL). There were no significant differences between the control and infected poults for DHLNL in trachea tissue. However, there was a significant difference between control and infected trachea cross links for HLNL and for DHLNL cross links in the aorta. Values for the HLNL from aorta were not obtained due to the presence of many unidentified radioactive peaks on the chromatogram. Representative histopathologies for trachea from infected and control poults are shown in Figures 1a and 1b and 2a and 2b, respectively. There was a loss of ciliated epithelium in the trachea of the infected poults, but not in the control poults (Figures 2a and 2b). The elastin fibers that are associated with the cartilage matrix can be observed along the proximal border of the hyaline ring. The fibers in the control trachea appeared uniform in distribution and comprise a network with overlapping arrangement. The fibers from the infected trachea appeared condensed, knotted, and gave an appearance of lost organization. Collagen fibers, which stain green with Masson’s Trichrome, were not easily discriminated with light microscopy.
1657
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN TABLE 4. Amino acid residue (isodesmosine1 and desmosine2) analysis: elastin in trachea and aorta from turkey poults infected with Bordetella avium Trachea
Aorta
Treatment
ISDES
DES
Ratio
ISDES
DES
Ratio
Control Infected
0.45 ± 0.02b 0.60 ± 0.03a
0.45 ± 0.02b 0.60 ± 0.03a
1 1
3.24 ± 0.16a 2.78 ± 0.14b
2.07 ± 0.11a 1.94 ± 0.09a
1.565a 1.433a
± SEM, in a column, with no common superscript differ significantly (P ≤ 0.05). = isodesmosine, residues/1,000 amino acid residues. 2DES = desmosine, residues/1,000 amino acid residues. a,bMeans 1ISDES
The results of this study indicated that both copper concentration and ceruloplasmin activity were elevated in the B. avium-infected poults. Other studies have shown that ceruloplasmin activity can be increased due to a general stress response. The induction of the enzyme has been accomplished by the administration of adrenocorticotropic hormone, hydrocortisone, infectious agents, and bacterial endotoxin (Starcher and Hill, 1965; Butler and Curtis, 1973; Freeman et al., 1973; Curtis and Butler, 1980). It is known that B. avium significantly elevates the corticosterone concentration during the period of infection (McCorkle et al., 1985), and therefore, it can be anticipated that ceruloplasmin would increase as well during the course of the bordetellosis. Additionally, it has been demonstrated that bordetellosis increases turnover of tissue catecholamines and there is evidence to suggest that adrenergic receptors that bind catecholamines may be up regulated during the disease (Edens et al., 1987). It is also known that catecholamines induce significant increases in plasma ceruloplasmin activity (Freeman et al., 1973), and it is known that reserpine depletes tissue stores of catecholamines by increasing secretion rates, and inhibiting reuptake caused a significant increase in plasma ceruloplasmin activity in chickens (Freeman et al., 1973). Therefore, in cases of bordetellosis characterized by elevated plasma levels of corticosterone, increased turnover of tissue catecholamines, increased copper availability, and increased ceruloplasmin activity, lysyl oxidase activity should have increased, with resultant increases in cross link activity. This hypothesis was supported by the data presented herein.
TABLE 5. Collagen cross-link analysis for trachea and aorta from turkey poults infected with Bordetella avium Trachea
Aorta
Treatment
DHLNL1
HLNL2
DHLNL
HLNL
Control Infected
1.49 ± 1.63 ±
1.80 ± 2.85 ±
0.39 ± 0.49 ±
NA3 NA
0.12a 0.13a
0.04b 0.06a
0.01b 0.02a
± SEM, in a column, with no common superscript differ significantly (P ≤ 0.05). = dihydroxy-lysinohydroxy-norleucine, moles/mole of collagen/300 residues hydroxyproline. 2HLNL = hydroxylysinonorleucine, moles/mole of collagen/300 residues hydroxyproline. 3NA = not obtained. a,bMeans
1DHLNL
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
The initiation of cross-linking for both collagen and elastin is dependent on the enzyme lysyl oxidase (Siegal et al., 1970; Tinker and Rucker, 1985). In collagen and elastin, specific e-amino groups of peptidyl lysine and hydroxylysine residues are oxidatively deaminated to yield aldehyde derivatives (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). The condensation of these aldehyde derivatives yields specific classes of cross-links that are used for identification of collagen and elastin, respectively. For collagen, the cross-links are HLNL, DHLNL, and pyridinoline (Tinker and Rucker, 1985; Yamauchi and Mechanic, 1988). For elastin, the cross-links are stable DES and ISDES derivatives (Tinker and Rucker, 1985). The role that lysyl oxidase plays is extremely important, especially as the enzyme’s activity is dependent on copper, which acts as a co-factor in the various reactions (Martin et al., 1970; Harris et al., 1977). Copper deficiency, due to disease or diet, leads to impaired ability of the host to produce adequate tissue elastin (Hill et al., 1967; Harris et al., 1977; Rayton and Harris, 1979). Additionally, ceruloplasmin, an acute phase protein and antioxidant in blood, functions in a transport/delivery capacity for copper and is associated with the mineral’s metabolism (Harris and DiSilvestro, 1981). There are correlations among the activity of ceruloplasmin, copper concentrations, and activity of lysyl oxidase (Hill et al., 1967; Harris and DiSilvestro, 1981). Due to the enzyme’s short half-life, the functional capacity of the enzyme may be assessed by the activity of serum ceruloplasmin and serum copper concentration.
1658
YERSIN ET AL.
FIGURE 1. Photomicrographs (400× ) of tracheal epithelial borders from Bordetella avium-infected (top) and control (bottom) turkey poults. In the B. avium-infected trachea there is a total loss of cilia along the border of the tracheal epithelium and this is contrasted with the ciliated epithelium of the control (indicated by the arrow).
Crude elastin was measured in both aorta and trachea to assess the integrity of the connective tissue. In the aorta, there was no difference in elastin content between the infected and control groups, but tracheal elastin was increased significantly in the B. avium group (Table 2). The amino acid analysis (Table 4) revealed that there was actually a reduced amount of cross-link components (ISDES) in the aorta of the infected poults. The crosslinks establish a quantitative difference specifically for the elastin content. Therefore, the reduced amount of cross-links in the aorta from infected poults suggested a less resilient tissue. At necropsy, the aorta from infected poults was flaccid, distorted, and collapsed and appeared to have lost structural integrity as compared to that of controls. The loss of structural integrity in the aorta from infected poults may be related to the decreased elastin cross-links found in this study. The collagen content (hydroxyproline residues per 1,000 amino acid residues) of the aorta was not decreased in the infected poults (Table 3). The crosslinks analysis revealed, however, that there was actually
FIGURE 2. Photomicrographs (400×) of elastin within tracheas from Bordetella avium-infected (top) and control (bottom) turkey poults. The B. avium infection has altered the normal arrangement of the elastin fibers causing them to be nonuniform and to be concentrated in bunches rather than to be uniformly distributed (indicated by large arrow in each panel). Small arrow indicates ciliated epithelium in the control.
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
an increase in DHLNL, one of the immature derivatives (Table 5). The DHLNL cross-link analysis indicated that collagen had been synthesized recently in response to tissue injury or in response to decreased elastin crosslinks (Table 4). Therefore, at 17 d of age, it appeared that the aorta was under repair and was in the early stages of recovery from the disease. Bordetellosis-associated dyspnea is indicative of reduced oxygen availability due to tracheal occlusion with excessive mucus and/or trachea collapse. These factors cause an increase in cardiac work and hypertrophy of the structures and vessels of the heart. The small increases in the collagen cross-links may herald the development of cardiac hypertrophy in response to bordetellosis. The crude elastin content of the trachea was increased significantly in the infected poults. This finding paralleled the amino acid cross-link analysis for DES and ISDES in infected poults. These observations
BORDETELLOSIS INFLUENCE ON ELASTIN AND COLLAGEN
REFERENCES Arp, L. W., and N. F. Cheville, 1984. Tracheal lesions in young turkeys infected with Bordetella avium. Am. J. Vet. Res. 45: 2196–2200.
Arp, L. W., and J. A. Fagerland, 1987. Ultrastructural pathology of Bordetella avium infection in turkeys. Vet. Pathol. 24:411–418. Association of Official Analytical Chemists, 1970. Official Methods of Analysis. 11th ed. Association of Official Analytical Chemists, Washington, DC. Butler, E. J., and M. J. Curtis, 1973. The effects of Escherichia coli endotoxin and ACTH on plasma zinc concentration in the domestic fowl. Res. Vet. Sci. 15:363–367. Curtis, M. J., and E. J. Butler, 1980. Response of caeruloplasmin to Escherichia coli endotoxins and adrenal hormones in the domestic fowl. Res. Vet. Sci. 28:217–222. Edens, F. W., F. M. McCorkle, D. G. Simmons, and A. G. Yersin, 1987. Brain monoamine concentrations in turkey poults infected with Bordetella avium. Avian Dis. 31: 504–508. Freeman, B. M., A.C.C. Manning, and D. S. Pole, 1973. Factors affecting plasma ceruloplasmin activity in Gallus domesticus. Comp. Biochem. Physiol. 45A:689–698. Gentry-Weeks, C. R., B. T. Cookson, W. E. Goldman, R. B. Rimler, S. P. Porter, and R. Curtiss, III, 1988. Dermonecrotic toxin and tracheal cytotoxin, putative virulence factors of Bordetella avium. Infect. Immun. 56:1698–1707. Gray, J. G., J. F. Roberts, R. C. Dillman, and D. G. Simmons, 1983. Pathogenesis of change in the upper respiratory tracts of turkeys experimentally infected with an Alcaligenes faecalis isolate. Infect. Immun. 42:350–355. Harris, E. D., and R. A. DiSilvestro, 1981. Correlation of lysyl oxidase with p-phenylenediamine oxidase activity (ceruloplasmin) in serum (41102). Proc. Soc. Exp. Biol. Med. 166:528–531. Harris, E. D., J. K. Rayton, and T. E. Degroot, 1977. A critical role for copper in aortic elastin structure and synthesis. Pages 543–565 in: Elastin and Elastic Tissue. I. A. Sanberg, ed. L. B. Plenum Press, New York, NY. Hill, C. H., B. Starcher, and C. Kim, 1967. Role of copper in the formation of elastin. Fed. Proc. 26:129–133. Hodges, R. D., 1974. The respiratory system. Pages 113–139 in: Histology of the Fowl. Academic Press, London, UK. Krane, S. M., 1981. Connective tissue. Pages 755–798 in: Pathophysiology: The Biological Principles of Disease. L. H. Smith, and S. O. Thier, ed. W. B. Saunders Co., Philadelphia, PA. Martin, G. R., S. R. Pinnell, R. C. Siegal, and E. R. Goldstein, 1970. Lysyl oxidase: the enzymatic step in collagen and elastin synthesis. Pages 405–410 in: Chemistry and Molecular Biology of the Intracellular Matrix. Vol. 1. E. A. Balazs, ed. Academic Press, New York, NY. McCorkle, F. M., F. W. Edens, and D. G. Simmons, 1985. Alcaligenes faecalis infection in turkeys: effects on serum corticosterone and serum chemistry. Avian Dis. 29:80–89. Rayton, J. K., and E. D. Harris, 1979. Induction of lysyl oxidase with copper: properties of an in vitro system. J. Biol. Chem. 254:621–626. SAS Institute, 1991. A User’s Guide to the Statistical Analysis System: Statistics. SAS Institute Inc., Cary, NC. Siegal, R. C., S. R. Pinnell, and G. R. Martin, 1970. Crosslinking of collagen and elastin properties of lysyl oxidase. Biochemistry 23:4486–4492. Simmons, D. G., 1991. Turkey Coryza. Pages 251–256 in: Diseases of Poultry. 8th ed. M. S. Hofstad, ed. Iowa State University Press, Ames, IA. Simmons, D. G., J. G. Gray, L. P. Rose, R. C. Dillman, and S. E. Miller, 1979. Isolation of an etiological agent of acute
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
strongly suggest an increase in elastin synthesis during the course of infection. One might argue that the trachea ring distortion, normally observed with the infection, would be better explained by the opposite effect or a decrease in elastin resulting in loss of stability of the trachea ring and ultimate collapse. However, collagen is the primary matrix protein associated with the cartilage structure of the trachea (Tinker and Rucker, 1985). The current study showed that the tracheal collagen content was reduced significantly during Bordetella infection compared to the uninfected Control. Similar to the responses of collagen and elastin in the aorta, when primary connective tissue protein is decreased, the secondary protein appears to respond with increased synthesis. These responses of trachea collagen and elastin appeared to be in response to injury caused by the bacteria or bacterial toxin(s) (Gentry-Weeks et al., 1988). Histological observations of the trachea collagen and elastin show that elastin in the infected poults was not deposited in a regular pattern along the border of the tracheal cartilage. In the infected poults, the elastin fibers in the trachea sections appeared to be knotted and irregular. It is possible that the irregular pattern may be a reflection of increased cross-linkage. The elastin arrangement was similar to the formation of scar tissue in wound healing (Krane, 1981). Bordetella avium produces a tracheal-cytotoxin, but its mechanism of action, other than being toxic to the ciliated epithelium, has not been determined (GentryWeeks et al., 1988). It is possible that the bacteria or toxin(s) induce serum proteases or collagenases to actively degrade the connective tissue matrix, and this degeneration could be the mechanism for the decreased elastin and collagen content of the aorta and trachea, respectively. However, one can not discount the possibility that collagen and elastin content of the trachea and aorta may be antigenic targets of the host immune response to infection. With initiation of the recovery from inflammation associated with bordetellosis, synthesis of new cross-links would be required to re-establish structural integrity of both the aorta and trachea. Results presented herein suggest that in cases of bordetellosis, there is a decrease in tracheal collagen but increases in trachea elastin. The increased elastin content of the trachea represents mature cross-linking. However, there is an increase also in the number of collagen and elastin cross-links in the immature types as well. The disease-associated increases in trachea (HLNL) and aorta (DHLNL) elastin cross linking (Table 5) with decreased collagen in the trachea, and lesser amounts in the aorta (Table 3) suggests a mechanism that may explain disease-associated tracheal collapse and flaccid texture of the aorta.
1659
1660
YERSIN ET AL. Yamauchi, M., E. P. Katz, and G. L. Mechanic, 1986. Intermolecular Cross-linking and Stereospecific molecular packing in Type I collagen fibrils of the periodental ligament. Biochem. 25:4907–4913. Yamauchi, M., D. T. Woodley, and G. L. Mechanic, 1988a. Aging and cross-linking of skin collagen. Biochem. Biophys. Res. Commun. 152:898–903. Yamauchi, M., D. R. Young, G. S. Chandler, and G. L. Mechanic, 1988b. Cross-linking and new bone collagen synthesis in immobilized and recovering primate osteoporosis. Bone 9:415–418. Yersin, A. G., F. W. Edens, and D. G. Simmons, 1991. Bordetella avium-infection effects on electrocardiograms in turkey poults. Avian Dis. 35:668–673.
Downloaded from http://ps.oxfordjournals.org/ at SUNY Health Science Center at Brooklyn - Medical Research Library on April 18, 2015
respiratory disease (rhinotracheitis) of turkey poults. Avian Dis. 23:194–203. Starcher, B. C., 1976. Determination of the elastin content of tissues by measuring desmosine and isodesmosine. Anal. Biochem. 79:11–15. Starcher, B. C., and C. H. Hill, 1965. Hormonal Induction of ceruloplasmin in chicken serum. Comp. Biochem. Physiol. 15:429–434. Tinker, D., and R. B. Rucker, 1985. Role of selected nutrients in synthesis, accumulation, and chemical modification of connective tissue proteins. Physiol. Rev. 65:607–657. Yamauchi, M., and G. L. Mechanic, 1988. Cross-linking of collagen. Pages 157–172 in: Collagen. Vol. 1. M. E. Nimni, ed. CRC Press, Boca Raton, FL.