![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
* Department of Animal Science;
Department of Medicine and Epidemiology, School of Veterinary Medicine; and
California Animal Health and Food Safety Lab, University of California–Davis, Davis, CA 95616
4To whom correspondence should be addressed. E-mail: kcklasing{at}ucdavis.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
3000 µg VA/kg diet can cause toxicity. It is possible that disturbances in VA nutrition contribute to the widespread incidence of behavioral problems reported in companion birds.
KEY WORDS: • cockatiel • deficiency • requirement • toxicity • vitamin A
Vitamin A (VA) is an essential micronutrient involved in vision, reproduction, immunity, membrane integrity, growth and embryogenesis. In avian diet formulations, VA can be one of the most challenging nutrients because the range between deficiency and excess is the most narrow of any of the vitamins and the amount found in foods is extremely variable (1,2). Furthermore VA is very heat and light sensitive, and diet processing and storage are associated with a loss of VA activity. For these reasons, various commercial diets may have either deficient or excess concentrations of VA.
The minimum VA requirements for growth and egg production of chickens, turkeys, domestic ducks and Japanese quail have been determined (3), although the requirements of poultry at maintenance (adult, nonreproductive, healthy, nongeriatric animals) are unknown. Virtually no research has been done to determine the dietary concentrations of VA that are required by, or are toxic to, companion avian species, so data from poultry serve as the basis for estimating the needs of companion birds. Most companion avian species differ from poultry species by being granivorous instead of omnivorous and having an altricial instead of a precocial mode of development. Consequently, extrapolations from poultry to companion avian species are potentially erroneous because of substantial differences between digestive physiology, reproductive performance and rates of embryonic and posthatch development. Additionally, the goals for poultry production (e.g., fast growth rate, maximal egg production) differ markedly from those for companion species (e.g., optimal body condition and feathering, maximal disease resistance and longevity). Because the VA requirements of companion birds are not known and grains used in their diets are very low in VA, feed manufacturers supplement VA to pelleted diets and many, but not all, seed mixtures. Thus, VA concentrations in commercial diets range from potentially deficient to potentially toxic. Characterization of the response of companion birds to excessive and deficient VA concentrations is needed for accurate diagnoses of VA malnutrition.
The purpose of this experiment was to characterize the physiological responses of one species of companion birds, the cockatiel (Nymphicus hollandicus), to various dietary VA concentrations, and to provide rough guidelines for the approximate dietary VA concentration resulting in deficiency and toxicity. Cockatiels were chosen as a reference species based on their altricial development pattern, small size and experimental malleability. Additionally, cockatiels are an extremely common companion bird and represent the family Psittacidae, which includes parrots, lorikeets, lovebirds, cockatoos and many other popular companion birds.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
Plasma retinol was assessed after monthly jugular venipuncture and liver retinol was assessed from samples collected at necropsy. The capacity to mount a primary and secondary antibody response was assessed after subcutaneous vaccination of 0.25 mL of sheep red blood cells (SRBC, 2% v/v) on d 211, 218 and 225. At 9 d after the first and last injections (d 211 for primary response, d 234 for secondary response), antibody titers were measured by agglutination procedures as previously described (5). Additionally, bimonthly assessments of oral/nasal epithelium cell types were made by analysis of nasal flushes, obtained by flushing the nasal cavity with 3 mL sterile saline then collecting oral cavity exudates. Subsequently, exudates were centrifuged (5 min at 800 x g), concentrated onto slides (Cytospin 2; Shandon Lipshaw, Pittsburgh, PA), stained with Wright’s stain and evaluated microscopically for the presence of basal epithelial cells (ciliated or nonciliated), squamous cells (keratinized or nonkeratinized) and mucin-containing cells.
As the experiment progressed, we noted diet-dependent changes in behavior during handling for nasal flushes, so vocalization analysis was performed on experimental d 135, 163, 198, 449 and 623. Birds were individually placed in a cage in a soundproof room, and after a 2-min acclimation period, birds were held gently while vocalizations were recorded (SpectraPLUS; Pioneer Hill Software, Poulsbo, WA) for 2 min. Recordings were analyzed for the number of vocalizations within a 2-min period, the number within each 30-s interval, the average length of each vocalization and the peak frequency, amplitude and power of vocalizations.
After 269 d of consuming experimental diets, birds fed 30,000 µg VA showed poor body condition, and therefore all birds fed 30,000 µg VA as well as two birds fed 3000 µg and two birds fed 600 µg VA were killed by CO2 inhalation, after which necropsies were performed. Remaining birds (n = 3 each for 3000 µg VA and 600 µg VA; n = 6 for 0 µg VA) were fed dietary treatments for a total of 706 d, at which time they were killed and necropsied. Liver, spleen, kidney, eye, trachea, syrinx, oropharynx, pancreas, bursa, duodenum, sinus and brain were fixed in 10% buffered neutral formalin and processed for histopathology. Birds were coded so that the treatment group was not known during evaluation.
Retinol analysis.
All chemicals were purchased from Sigma or Fisher Scientific (Springfield, NJ), unless otherwise noted. Briefly, under amber light and/or in amber tubes, liver sections were homogenized in phosphate-buffered saline (PBS) at a ratio of 
2 parts tissue: 1 part PBS (w/v). A 1-g sample of tissue homogenate or 300–500 µL plasma was used to extract retinol. Samples were first vortexed with methanolic potassium hydroxide [5% KOH in methanol (w/v), 1x sample volume; methanol]. Butanol:acetenotrile (1:1 v/v; butanol) was added at twice the sample volume, then samples were vortexed vigorously for 5 min. Hexane:chloroform (2:1 v/v) was then added at 1x sample volume; samples were again vortexed for 5 min and then K2HPO4 (saturated solution of K2HPO4 in deionized H2O) was added at 0.1x sample volume. Samples were vortexed vigorously for 5 min, then centrifuged (5000 x g) for 15 min. The top organic phase was removed, dried under N2 and frozen. Before HPLC analysis, samples were reconstituted into the mobile phase consisting of 75% methanol, 12.5% distilled deionized water, 9% acetonitrile, 3% tetrahydrofuran, 0.5% isopropanol, and 0.46 g ammonium acetate/L. The mobile phase was maintained at pH 7.15 with glacial acetic acid.
HPLC analysis was performed by use of a C18 reverse-phase column (5 µm, 300 nm, 4.6 mm I.D. x 250 mm L; Vydac 201TP54, Hesperia CA) and a high performance guard column (5 µm; Vydac 201GD54T). The isocratic mobile phase (described above) was maintained at a flow rate of 1.0 mL/min (Waters 510 pump, Waters Associates, Milford MA), and automated injections (Waters WISP 712) of 75 µL were made. A UV/visible detector (Waters 484) monitored at 234 nm, and peak identification and quantitation (Waters Millenium software) were made by comparing samples to a purified all-trans retinol standard. Three injections of the retinol standard at different volumes were made before each sample set, and standard curves were generated by use of linear regression.
The analysis of dietary retinol concentrations was kindly provided by Roche Vitamins (Parsippany, NJ). Each diet was analyzed in duplicate, and identification of calculated diet concentrations was not supplied to the laboratory. Diets were analyzed to contain the following (mean ± SD): 0 µg diet, no detectable retinol; 600 µg diet, 835 ± 36 µg/kg; 3000 µg diet, 2815 ± 70 µg/kg; 30,000 µg diet, 24,549 ± 397 µg/kg.
Statistical analysis.
Data were analyzed by a general linear model (JMP; SAS Institute, Cary, NC). Those data collected at a single time (antibody titers, liver retinol) were analyzed by one-way ANOVA for the main effect of the dietary treatment. Data collected at multiple times (body weight change, plasma retinol, nasal epithelial flushes, vocalizations) were analyzed by two-way ANOVA for the main effects of the dietary treatment, days fed the dietary treatment and their interactions. Additionally, this model nested individual bird observations within the dietary treatment, which accounts for repeated observations on a single individual. When appropriate, means comparisons were made by preplanned orthogonal contrasts in which the 0, 3000 and 30,000 µg/kg treatments were individually compared to the 600 µg/kg treatment. Histopathology data were scored for the presence or absence of pathology and then values were analyzed by chi-square for the main effect of the dietary treatment. Finally, liver and plasma retinol data were also analyzed by linear regression to determine the correlation between diet and liver or plasma retinol concentrations. For all statistical analyses, significance was set at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
After consuming experimental diets for 269 d, birds fed 30,000 µg/kg VA had reduced body condition, based on reduced feather quality and atrophy of the pectoral muscle, although the body weight resulting from the dietary treatment did not change (P = 0.30). At experimental d 697, cockatiels fed 0 and 3000 µg/kg VA had body weights (P = 0.86) and conditions that were indistinguishable from birds fed 600 µg/kg. A significant day effect (P < 0.01) indicated that body weights fluctuated throughout the experiment. Overall, changes in body weight for birds fed 0, 3000 or 30,000 µg/kg VA did not differ from those of birds fed 600 µg/kg (P > 0.1).
Plasma and liver retinol.
After consuming experimental diets for 235 d, birds fed 30,000 µg/kg VA had greater plasma retinol concentrations than birds fed 600 µg/kg VA (P < 0.05, Fig. 1). After 697 d, birds fed 3000 µg/kg VA had greater plasma retinol concentrations than birds fed 600 µg/kg VA (P < 0.01). Plasma retinol concentrations did not differ between (P < 0.05) birds fed 0 or 600 µg/kg VA at any time during the trial. Based on the regression model (calculated by use of final plasma retinol concentrations for each treatment group), plasma retinol concentrations increased in a diet-dependent manner [plasma retinol (mmol/L) = 0.112 + (1.96 x diet VA); R2 = 0.60; P < 0.001]. Liver retinol increased with increasing dietary VA concentration (0 µg/kg VA at d 706 = 19.0 ± 5.4; 600 µg/kg VA at d 706 = 34.3 ± 12.2; 3000 µg/kg VA at d 706 = 48.2 ± 14.1; 30,000 µg/kg VA at d 269 = 61.3 ± 20.5 mmol/kg), as demonstrated by linear regression [liver retinol (mmol/kg) = 27.77 + 0.18 x diet VA; R2 = 0.28, P < 0.02].
|
Dietary VA concentration had no impact on the primary antibody response to SRBC (P = 0.7, data not shown). After 269 d, birds fed 0 µg/kg VA had reduced secondary anti-SRBC antibody titers compared to birds fed 600 µg/kg VA (P < 0.05, Fig. 2), whereas secondary anti-SRBC antibody titers of birds fed 3000 or 30,000 µg/kg VA did not differ from birds fed 600 µg/kg VA (P > 0.1).
|
There was no consistent effect of dietary VA concentration on the percentage of any cell type throughout the duration of the trial. However, on d 490, the percentage of keratinized squamous cells was greater for birds fed 3000 µg/kg VA than for those fed 600 µg/kg VA (P < 0.05, data not shown).
Vocalization analysis.
The total number of vocalizations provoked by physical restraint was much greater in birds fed 30,000 µg/kg VA than in birds fed 600 µg/kg VA (P < 0.01, Fig. 3). This difference occurred within the 1st min and the last 30 s of a 2-min observation period (P < 0.05). In contrast, birds fed 0 or 3000 µg/kg VA had fewer numbers of vocalizations compared to birds fed 600 µg/kg VA (P < 0.03), primarily in the first 30 s (P < 0.05). The average length of vocalizations was reduced in birds fed 0 or 3000 µg/kg VA compared to that in birds fed 600 µg/kg VA (P < 0.01). The peak frequency (Hz) of vocalizations (Table 2) was reduced in birds fed 0 (P < 0.01) or 30,000 µg/kg VA (P < 0.02) compared to that in birds fed 600 µg/kg VA, and the peak amplitude (dB) and total power (dB) were reduced in birds fed 0 µg/kg VA compared to that in birds fed 600 µg/kg VA (P < 0.01).
|
|
Necropsies were completed on birds from all treatment groups on d 269 and on birds in the 0, 600 and 3000 µg/kg VA groups on d 706. Birds fed 30,000 µg/kg VA were more likely to have pancreatitis than were birds fed 600 or 3000 µg/kg VA (P < 0.01). Additionally, increases in multifocal accumulations of lymphocytes within the duodenal lamina propria were greater in birds fed 30,000 µg/kg VA than in birds fed 600 or 3000 µg/kg VA (P < 0.05). No significant differences in pathology of other tissues were noted. After 697 d, birds fed the 3000 µg/kg VA diet had increased incidence of splenic hemosiderosis compared to that in birds fed 600 µg/kg VA (P < 0.03), although there were no other differences between birds fed 600 µg/kg VA and those fed 0 or 3000 µg/kg VA.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Adult cockatiels at maintenance were much more susceptible to dietary VA toxicity than to dietary VA restriction. In fact, birds fed 30,000 µg/kg VA developed dramatic changes in vocalization patterns and sufficient deterioration in feather quality and breast muscle mass that euthanasia and necropsy was warranted after <1 y. In contrast, birds fed 0 µg/kg VA for almost 2 y did not show clinically identifiable signs of VA deficiency. The cockatiels used in this experiment previously had been fed a VA-adequate diet (based on successful growth and breeding of this colony for >20 y) formulated to contain 2363 µg/kg VA. Presumably liver VA stores were sufficient to provide VA to the birds for the entire trial. In fact after almost 2 y of consuming a diet with no detectable VA, cockatiels had liver retinol concentrations of 19.0 ± 5.4 mmol/kg. In other parrot species, squamous metaplasia (a common symptom of VA deficiency) was associated with liver retinol concentrations < 7.3 mmol/kg (6), suggesting that further depletion of the cockatiels in the current study would have been required to achieve deficiency. These data indicate that cockatiels are very adept at storing VA for use in times of low dietary VA and conserve their VA efficiently. Considering their wild-type diet, which is seed-based and would likely contain low concentrations of preformed VA (1,7), this type of adaptation and efficient storage of VA is not surprising. However, birds fed 0 µg/kg VA were unable to mount a secondary antibody responses to SRBC, which is also seen in chickens fed a VA-deficient diet (8,9). Thus, changes in immunocompetence were a better indication of diminishing VA status in cockatiels than changes in body condition, plasma retinol, organ pathology or classical signs of deficiency such as squamous metaplasia of nasal epithelia. We did not observe decreased antibody responses in cockatiels fed 30,000 µg/kg VA, although chickens fed extremely high concentrations of dietary VA (>900,000 µg/kg) had reduced antibody titers, accompanied by signs of acute toxicity (8).
In general plasma retinol does not change markedly with marginal VA status, and reduced concentrations are not seen until animals reach a severe deficiency (10). Our results showed that birds fed 0 µg VA for almost 2 y did not have reduced plasma retinol concentrations, suggesting that plasma retinol values are not a reliable predictor of dietary VA deficiency; although the low number of birds used in this study may have made small differences undetectable. In contrast, plasma retinol concentrations were increased (P < 0.05) in birds fed 30,000 µg/kg VA (235 d) and 3000 µg/kg (697 d), suggesting that high plasma retinol concentrations are diagnostic of excessive or toxic dietary VA concentrations.
Signs of VA toxicity included poor feathering, pancreatitis, multifocal accumulations of lymphocytes within the duodenal lamina propria and hyperexcitability. In particular, birds developed highly exaggerated vocalizations in response to restraint. There are several potential mechanisms for this behavioral response. First, impaired production of corticosteroids and resulting impaired stress tolerance may contribute to behavioral changes (11). Second, changes in vocalizations may be a result of the pathology of tissues associated with vocalizations, such as specific regions of the brain (12), respiratory and tracheal muscles (13) or the syrinx (14), although we did not observe changes in the epithelium of the syrinx or consistent pathologies in muscle or brain. Third, VA toxicosis in humans has been implicated in idiopathic intracranial hypertension, resulting in headaches, papilledema and elevated intracranial pressure, without evidence of intracranial lesions (15,16), bone and joint pain and "psychiatric" symptoms (17). Finally, it is also possible that alterations in vocalization patterns with VA malnutrition are adaptive and not pathological. Several nutritional conditions, including amino acid imbalances and calcium deficiencies, result in behavioral changes that alter food selection and serve to improve the nutrient balance of the diet (2). The exact mechanism by which dietary VA concentrations alter cockatiel vocalization patterns is unclear, and warrants further investigation, given that it is possible that disturbances in VA nutrition contribute to the widespread incidence of behavioral problems reported in companion birds.
Birds fed 3000 µg/kg VA had liver retinol concentrations similar to those of birds fed 30,000 µg/kg VA, and plasma retinol concentrations of birds fed 3000 µg/kg VA for 697 d surpassed that of birds fed 30,000 µg/kg VA for 235 d. Furthermore, birds fed 3000 µg/kg VA had a greater incidence of splenic hemosiderosis and differences in vocalization compared to those of birds fed 600 µg/kg VA. Although the behavioral changes caused by 3000 µg/kg VA were subtle compared to those caused by 30,000 µg/kg VA, it is likely that the 3000 µg/kg VA diet is marginally toxic. If this is the case, commercial vitamin supplements and some commercial pelleted feeds have excessive concentrations of vitamin A. Cockatiel chicks express 15,15'-dioxygenase mRNA and can use ß-carotene as a source of VA (unpublished data). Thus, it may be prudent to supplement diets with a combination of retinol and ß-carotene to avoid VA toxicity, yet ensure a margin of safety to circumvent deficiencies arising from uncertainty in requirements and the instability of retinol during diet storage.
In conclusion, adult female cockatiels at maintenance are more susceptible to VA toxicity than to VA deficiency. Clinically, this observation is very important because anecdotal evidence often suggests that Psittacines are most likely to be VA deficient. In the case of adults, VA toxicity may be more prevalent than currently reported, given that vitamin supplements and commercially available diets often contain very high concentrations of VA. Assuming that cockatiels were previously fed diets containing adequate dietary VA concentrations, producing a VA deficiency would presumably require accelerated losses, such as by egg production, or very long term (>2 y) dietary depletion. Additionally, cockatiels fed 3000 µg/kg VA approached the plasma and liver retinol concentrations of birds that had an identifiable VA toxicity, indicating that dietary VA concentrations 3000 µg/kg VA may cause toxicity. Therefore, optimal dietary VA concentrations for adult cockatiels appear to be <3000 µg VA/kg diet (diet analyzed to contain 2815 µg VA/kg diet), whereas diets formulated to contain 600 µg VA/kg diet (diet analyzed to contain 835 µg/kg diet) appear to meet all VA needs at maintenance.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by Morris Animal Foundation grant 98AV-04.
3 Present address: Animal Science Department, California Polytechnic State University, San Luis Obispo, CA.
Manuscript received 12 January 2003. Initial review completed 11 February 2003. Revision accepted 7 March 2003.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Koutsos, E. A., Matson, K. D. & Klasing, K. C. (2001) Nutrition of birds in the order Psittaciformes: a review. J. Avian Med. Surg. 15:257-275.
2. Klasing, K. C. (1998) Comparative Avian Nutrition 1998 CAB International New York, NY.
3. National Research Council (NRC) (1994) Nutrient Requirements of Poultry 9th ed. 1994 National Academy Press Washington, DC.
4. Koutsos, E. A., Smith, J., Woods, L. & Klasing, K. C. (2001) Adult cockatiels (Nymphicus hollandicus) undergo metabolic adaptation to high protein diets. J. Nutr. 131:2014-2020.
5. Klasing, K. C. (1988) Influence of acute feed deprivation or excess feed intake on immunocompetence of broiler chicks. Poult. Sci. 67:626-634.[Medline]
6. Zwart, P., Schreurs, W.H.P. & Dorrestein, G. M. (1979) Vitamin A Deficiency in Parrots. Fowler, M. E. Ippen, R. Schroder, H. D. eds. Proceedings of the XXI International Symposium on Zoo Animal Diseases 1979:47-52 Akademie-Verlag Berlin .
7. Jones, D. (1987) Feeding ecology of the cockatiel, Nymphicus hollandicus, in a grain-growing area. Aust. Wild. Res. 14:105-115.
8. Friedman, A. & Sklan, D. (1989) Antigen specific immune response impairment in the chick as influenced by vitamin A nutrition. J. Nutr. 119:790-795.
9. Leutskaya, Z. K. & Fais, D. (1977) Antibody synthesis stimulation by vitamin A in chickens. Biochim. Biophys. Acta 475:207-216.[Medline]
10. Bondi, A. & Sklan, D. (1984) Vitamin A and carotene in animal nutrition. Prog. Food Nutr. Sci. 8:165-191.[Medline]
11. Wells, A., Jones, A., Williams, M. & Geist, C. R. (1974) Noise, vitamin A deficiency, and emotional behavior in rats. Percept. Mot. Skills 38:392-394.[Medline]
12. Phillips, R. E., Youngren, O. M. & Peek, F. W. (1972) Repetitive vocalizations evoked by local electrical stimulation of avian brains. I. Awake chickens (Gallus gallus). Anim. Behav. 20:689-705.[Medline]
13. Peek, F. W., Youngren, O. M. & Phillips, R. E. (1975) Repetitive vocalizations evoked by electrical stimulation of avian brains. IV. Evoked and spontaneous activity in expiratory and inspiratory nerves and muscles of the chicken (Gallus gallus). Brain Behav. Evol. 12:1-41.[Medline]
14. Gaunt, A. S. & Gaunt, L. L. (1977) Mechanics of the syrinx in Gallus gallus. II. Electromyographic studies of ad libitum vocalizations. J. Morphol. 152:1-19.[Medline]
15. Lombaert, A. & Carton, H. (1976) Benign intracranial hypertension due to A-hypervitaminosis in adults and adolescents. Eur. Neurol. 14:340-350.[Medline]
16. Selhorst, J. B., Waybright, E. A., Jennings, S. & Corbett, J. J. (1984) Liver lover’s headache: pseudotumor cerebri and vitamin A intoxication. J. Am. Med. Assoc. 252:3365.
17. Muenter, M. D., Perry, H. O. & Ludwig, J. (1971) Chronic vitamin A intoxication in adults. Hepatic, neurologic and dermatologic complications. Am. J. Med. 50:129-136.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
