- Email: [email protected]
Effects of lead and exercise on endurance and learning in young herring gulls
ARTICLE IN PRESS
Ecotoxicology and Environmental Safety 57 (2004) 136–144
Effects of lead and exercise on endurance and learning in young herring gulls Joanna Burger and Michael Gochfeld1 Consortium for Risk Evaluation with Stakeholder Participation, Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ 08854, USA Received 14 June 2002; received in revised form 13 March 2003; accepted 14 March 2003
Abstract In this paper, we report the use of young herring gulls, Larus argentatus, to examine the effect of lead and exercise on endurance, performance, and learning on a treadmill. Eighty 1-day-old herring gull chicks were randomly assigned to either a control group or a lead treatment group that received a single dose of lead acetate solution (100 mg/kg) at day 2. Controls were injected with an equal volume of isotonic saline at the same age. Half of the lead treatment group and half of the control group were randomly assigned to an exercise regime of walking on a treadmill twice each day. The other group remained in their cages. We test the null hypotheses that neither lead nor exercise affected performance of herring gull chicks when subsequently tested on the treadmill at 7, 11, and 17 days post-injection. Performance measures included latency to orient forward initially, to move continuously, forward on the treadmill, and to avoiding being bumped against the back of the test chamber. Also measured were the number of calls per 15 s, and the time to tire out. Latency to face forward and avoiding being bumped against the back of the test chamber were measures of learning, and time to tire out was a measure of endurance. We found significant differences as a function of lead, exercise, and their interaction, and rejected the null hypotheses. For all measures of behavior and endurance, lead had the greatest contribution to accounting for variability. In general, lead-treated birds showed better performance improvement from the daily exercise than did controlled non-lead birds, with respect to endurance and learning. We suggest that in nature, exercise can improve performance of lead-exposed birds by partially mitigating the effects of lead, thereby increasing survival of lead-impaired chicks. r 2003 Elsevier Science (USA). All rights reserved. Keywords: Lead exposure; Exercise; Learning; Endurance; Treadmill; Fatigue
1. Introduction Governmental agencies as well as the public and private sectors are all concerned about the potential adverse effects of contaminants on neurobehavioral development. Although much of the recent discussion has dealt with chemicals that cause disruptions in the endocrine system, the neurodevelopmental effects of lead exposure continue to cause concern for humans as well as other animals. Lead, derived from industrial and urban pollution, from agricultural runoff, as well as Corresponding author. Division of Life Sciences, Nelson Biological Laboratory, Rutgers University, 604 Allison St. Nelson Hall, Piscataway, NJ 08854-8082, USA. Fax: +732-445-5870. E-mail address: [email protected] (J. Burger). 1 Present address: Environmental and Community Medicine, UMDNJ-Robert Wood Johnson Medical School, Piscataway, NJ 08854, USA.
from natural geochemical processes, enters the air, soil, and water, as well as the food chain. The neurotoxicity of lead has been extensively studied (Cory-Slechta and Schaumburg, 2000). In recent years, decreases in the use of leaded gasoline and lead-based paints in the USA have resulted in reductions in both lead exposure and human blood lead concentrations (ATSDR, 1999). However, lead is one of the most common contaminants at Superfund and other contaminated sites (ATSDR, 1999). Moreover, lead levels are increasing in some cohorts of children (ATSDR, 1988, 1999) and in some species of birds (Burger et al., 1994), but not in others (Burger and Gochfeld, 2003a). There is increasing evidence that lead affects a number of cognitive and motor functions in a wide range of vertebrates, including humans. Animal models provide information on the effects of low- and mediumlevel lead effects on neural and cognitive processes
0147-6513/03/$ - see front matter r 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0147-6513(03)00035-6
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
(Eisler, 1988; Lippman, 1990; Alber and Strupp, 1996; Burger and Gochfeld, 2000), and for the most part, the animal models and epidemiological data provide similar results (Rice, 1996). In humans and other primates, lead exposure at low or medium levels leads to auditory, neurobehavioral, hematological, nephrotoxic, and reproductive effects (Rice, 1984, 1996; Bellinger et al., 1987; Payton et al., 1998; Lasky et al., 2001). Even at very low doses, lead exposure can produce serious adverse effects on the central nervous system of children and infants, and these effects last for years (Needleman et al., 1990; Dietrich et al., 2001; Morgan et al., 2001). The relationship between lead exposure and cognitive development in children has engendered lively controversy. A meta-analysis by Pocock et al. (1994), reviewing 26 studies and concluding that lead had only a minimal effect, received extensive criticism for excluding key studies (e.g., Needleman, 1995). As early as 1990, Silbergeld (1990) suggested that the neurodevelopmental effects of lead may not have a detectable threshold, and recent authors support this view (Lidsky and Schneider, 2003). Furthermore, neurotoxic effects have been noted in a variety of other mammals and in birds (Dietz et al., 1979; Cory-Slechta et al., 1983; Burger and Gochfeld, 1985, 1988; Eisler, 1988; Alber and Strupp, 1996; Lilienthal and Winneke, 1996; Newland et al., 1996; Bunn et al., 2000). Lead also disrupts social behavior in a number of species, including rodents (Cutler, 1977; Donald et al., 1981; Burright et al., 1983; Dolinsky et al., 1983) and primates (Bushnell and Bowman, 1979a, b; Laughlin et al., 1991). For 15 years we have been investigating the effects of low-level lead exposure on the behavior of young birds, particularly the herring gull, Larus argentatus, in the laboratory. Lead exposure delays individual recognition of both parents and siblings (Burger and Gochfeld, 1993a; Burger, 1998), locomotion and balance (Burger, 1990), thermoregulation and visual cliff behavior (Burger and Gochfeld, 1995a, b). Moreover, similar effects occur in nature (Burger and Gochfeld, 1994, 1996, 1997, 2000). These experiments have been useful in providing paradigms to understand and study effects of lead in humans and other primates, and to understand the effects of lead on behavior, physiology, and survival of birds in nature. Birds can be useful models for metal toxicity because they share with humans a reliance on visual and vocal communication, in contrast to the primarily ultrasonic, tactile, and olfactory communication of rodents. Herring gull chicks move about the nest within a few hours of hatching, and by 2–3 days of age they walk easily around their parents’ territory of a few square meters (Burger, 1984). In this paper young herring gulls were used to test the null hypotheses that there are no
137
differences in endurance and learning as a function of lead treatment and exercise regime, using a treadmill test. Whereas our previous studies have shown deficits in individual recognition as a function of exposure to lead (Burger and Gochfeld, 1993a), the effects of lead on learning a new skill (running on a treadmill) have not been examined in an avian model. Further more, endurance has not been examined in birds. Running speed and endurance are characteristics often examined in mammals (Djawdan and Garland, 1988; Dohm et al., 1996) and lizards (Bauwens et al., 1995). Treadmill performance is used as a method of integrating physiological and ecological approaches to adaptation (Arnold, 1983). In both mice and lizards, running is an important aspect of their daily behavior used to escape danger (both species) and capture food (lizards in pursuit of insects). Gull chicks are seldom exposed to a situation that challenges endurance or sprint speed because they spend most of their time sitting or standing on the nest waiting to be fed or hiding in nearby vegetation until they are old enough to fly. During this time they are protected by at least one parent, reducing the need for predator avoidance. However, when disturbed, they may need to run vigorously to escape potential predators. In these tests we used a treadmill to examine learning, performance, and endurance. To perform successfully, the gulls must learn to face forward, to walk continuously forward on the moving treadmill, and to avoid being bumped against the back of the test area (a mildly annoying stimulus). It was not clear before our pilot tests that the chicks could learn the task, would learn the task, or would continue to perform. We predicted that gulls that were afforded the opportunity to exercise daily on the treadmill would perform better when tested with the treadmill operating at a faster speed than the exercise speed. Furthermore, we expected that lead would disrupt their ability to learn how to move on the treadmill and would reduce their endurance as well.
2. Materials and methods 2.1. Source and handling of study birds Eighty 1-day-old herring gull chicks were collected from nesting colonies at Captree, Long Island and Barnegat Bay, New Jersey in 1997, under appropriate federal and state permits. To minimize effects on reproductive success, only the first-hatched chick in any nest was collected. After marking with numbered leg bands for individual identification, chicks were randomly assigned to one of four groups: lead treatment or saline control, with or without exercise.
ARTICLE IN PRESS 138
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
Chicks were housed in 50 50 50-cm stainless-steel cages in groups of three to simulate natural broods (Burger, 1984), and were maintained in a warm laboratory at 2772 C with a natural light/dark cycle (15 h light, 9 h dark). Chicks were fed three to five times daily with fresh fish and seafood, supplemented with canned cat and dog food, eggs, and mice. Young gulls do best with a varied diet. Chicks were individually fed by caretakers until they learned to feed on their own; most fed voraciously throughout the study. Weight matching was effective. Research assistants rather than routine animal keepers did all care and handling of the chicks, making daily notes on each chick, including what they ate, whether they ate appropriately, and whether they appeared alert and healthy. Chicks gained weight appropriately for their age, and there were no differences in weight between control and experimental chicks at the end of the experiment (w2 tests). Only chicks that remained healthy were included in the data set. In nature, B30% of herring gulls chicks die in the first week from a variety of causes (Burger, 1984); thus it is not surprising that some chicks become ill or die in the laboratory. Because we were interested in the effects of lead on cognitive function and endurance, and not on physical impairment due to other causes, such chicks were eliminated from the study. Within each treatment group, chicks were assigned to pseudo-sibling groups based on weight. It was essential to have ‘‘siblings’’ similar in weight to avoid undue fighting in the cages (chicks in nature normally fight with their siblings: see Burger, 1984). Furthermore, all three chicks in each sibling group had the same exposure. This was essential to avoid cross-contamination, because chicks may ingest the defecations of their cagemates. This avoided any possibility of controls obtaining lead from siblings. 2.2. Lead exposure On day 2 lead-treated chicks were given an intraperitoneal injection of lead acetate (total dose=100 mg/kg lead in sterile water) (25 mg/mL). Controls received an equivalent volume (based on their weight) of isotonic saline solution. Lead was injected by a technician not otherwise involved in the behavioral tests, and all caretakers were blind to the treatment status of the gulls they handled. Lead was administered by injection rather than in food because chicks that eat different amounts would receive different doses. This dose was selected because it yields tissue levels within the range found in young herring gulls in the wild; this dose results in mean blood lead levels of 16.5–25 mg/dL at fledging (Burger and Gochfeld, 1990). The levels of lead in feathers of lead-injected birds averaged 4.8 ppm, which is similar to amounts reached
in some wild herring gulls in the northeastern USA, which have lead concentrations in their feathers of 0.5– 8.5 ppm (Burger and Gochfeld, 1993b). There is a high correlation between lead concentrations in blood and in feathers (r ¼ 0:79; Burger and Gochfeld, 1990), as well as between those in blood and in the brain (r ¼ 0:83; Burger and Gochfeld, 1990). In the current experiment, blood lead concentrations for controls averaged 2.270.2 mg/dL, and those in lead-treated birds averaged 2477.8 mg/dL. 2.3. Exercise treatment Half of the lead-treated chicks and half of the control chicks were subjected to a daily exercise regime. Nonexercised chicks were not allowed to exercise except for normal movements within their cages. We exercised chicks on a human treadmill set at a speed of 200 cm/ min. For their exercise regime they were placed on the treadmill with their cagemates. To prevent injuries, a cardboard frame was placed over the treadmill so that the chicks did not fall off; thus chicks bumping against the back experienced mild discomfort. Chicks were watched at all times, and when a chick fell over or could not maintain the pace, it was removed from the treadmill. From 4 to 7 days of age, the chicks exercised on the treadmill for 5 min twice a day; from day 8 to 12, they were exercised for 10 min. This was increased incrementally by 5 min every 3 days up to 20 min. 2.4. Testing We conducted performance and endurance tests on the treadmill for all chicks at 7, 11, and 17 days after lead injection when they were 9, 13, and 19 days of age. During the test, chicks were placed individually on the treadmill and observed for 2 min at a faster treadmill speed of 300 cm/min. During that time we recorded the latency to face forward initially, latency to face forward continually for the duration of the test, time touching the back of the apparatus, time walking forward, time to tire out, and number of calls given in the first 15 s. Time to tire out is the time until they fell down or stopped walking, and is a measure of endurance. If they continued to walk for the 2-min test, this value was set at 120 s. Latency to face forward (initial and continued) and time at the back touching the box were measures of learning and motor performance. 2.5. Statistical tests Analysis of variance (ANOVA) was used to determine if treatment or age contributed to differences in behavioral responses (PROC GLM, SAS, 1995). We used lead treatment, exercise status, and test day as well as interactions as the independent variables. We also
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
analyzed the data by date. We compared performance of lead vs. controls and exercise vs. nonexercise with Kruskal–Wallis tests (Siegel, 1956). To compensate for low power, we list P values o0.10.
Control - No Exercise
Lead- No Exercise
Control - Exercise
Lead - Exercise
Time to First Face Forward (s)
70
Face Forward Continually (s)
3. Results Overall, the models indicated that lead treatment had the greatest effect on variations in the response measures, with exercise and date (that is, age) also contributing considerably to most models (except for number of calls, Table 1). Interactions between variables were generally not significant, except for latency to face forward (Table 1). The models for response measures were also significant by days post-injection, with lead treatment generally contributing the most. Exercise did not enter as a significant variable during the first test, when the exercised chicks had experienced the treadmill for only 3 days. However, it had a major impact on the next test 4 days later. The lead-treated chicks with no exercise showed the greatest behavioral deficits on all measures (Table 2), taking the longest to face forward initially and to continue facing forward, spending the greatest amount of time at the back of the box and the least amount of time walking forward freely, and they tired out the quickest. Although there were differences within the control chicks as a function of whether they were
139
60 50 40 30 20 10 0 100 90 80 70 60 50 40 30 20 10 0
7
11
17
Days Post-Injection Fig. 1. Time to first face forward, and to face forward continuously, when placed on a treadmill. Shown are means7SE.
Table 1 Linear models explaining variations in behavioral and endurance variables for herring gull young Behaviors
Latency to turn forward Latency: continuous forward Time at back Time spent walking Time to tire (s) Calls/15 s a
Overall
Model
Variables entering the model
R2
F
P
Lead P
Exercise P
Lead exercise P
Date (age) P
0.19 0.24 0.24 0.23 0.11 0.21
9.54 12.70 12.56 11.99 5.01 10.6
0.0001 0.0001 0.0003 0.0001 0.0002 0.0001
o0.0001 o0.0001 o0.0001 o0.0001 o0.0002 o0.0001
o0.005 o0.0008 o0.0002 o0.0002 NSa o0.0014
o0.011 o0.003 o0.08 NS NS o0.10
o0.013 o0.0001 o0.0006 o0.0005 o0.017 o0.009
NS=not significant.
Table 2 Overall means for behavioral responses on treadmill tests for herring gull chicksa
Number of birds Latency to face forward initially (s) Latency to face forward continually (s) Total time in back (s) Time walking (s) Endurance (up to 120 s) No. calls/15 s a
No lead, no exercise
No lead, exercise
Lead, no exercise
Lead+exercise
Wilcoxon w2 (P)
18 18.673.0 (B) 35.274.6 (B) 51.174.5 (C) 67.574.4 (C) 11172.4 (B) 2.870.16 (B)
17 17.473.2 33.675.6 41.775.5 78.175.5 11172.6 2.570.2
16 51.676.6 (A) 76.976.5 (A) 87.974.6 (A) 33.674.3 (A) 95.074.1 (A) 4.470.31 (A)
17 28.674.3 (B) 42.775.3 (B) 61.075.0 (B) 58.775.0 (B) 10273.5 (A) 3.370.19(B)
30.6 24.8 11.3 21.1 14.5 31.7
Letters indicate significant differences (Duncan multiple range test).
(B) (B) (C) (C) (B) (B)
(0.0001) (0.0001) (0.01) (0.0001) (0.002) (0.0001)
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
exposed to exercise this effect was smaller than in the lead-treated group (Table 2). The latency to face forward initially, and to continue to face forward, are measures of learning how to perform on the moving treadmill. The pattern among the three tests was complicated (Fig. 1), although overall, the lead-treated, nonexercised birds performed the worst (Table 2). In general, after the first test, chicks with exercise experience performed better than those with no regular experience. Latency to face forward generally decreased with age and experience. The exception was the lead-treated, nonexercised birds. During the second test (11 days post-injection) the lead-treated, nonexercised birds performed much worse than on the other two tests. Time (in seconds) to walk forward, without touching the back, was also a measure of learning because the gulls had to learn to continue walking, and to walk at the appropriate speed to keep in the center of the box. In general, birds with previous exercise improved steadily in performance, whether they had lead or not (Fig. 2). Control birds performed better on the first and last tests than on the middle test, and we have no explanation for this deficit. Calling, a measure of anxiety, generally decreased with time, except for the lead-treated, nonexercised birds (Fig. 2). 6
Number of Calls
5
4
3
120 115 110
Endurance (s)
140
105 100 95 90 Control -No Exercise
85
Control -Exercise Lead-No Exercise
80
Lead-Exercise
75 11
7
17
Days Post - Injection Fig. 3. Endurance on a treadmill test. Shown in seconds to tire out (means7SE).
We report two measures of endurance: time to tire (in seconds), and the number of birds that did not complete the 120-s test. Time to tire out was significantly affected by lead at every age (see Fig. 3). The exercised birds generally improved in performance over time. There were also significant differences among the four groups in the percentage of chicks that failed to endure for the 120-s test (Table 3). The differences were greatest for the nonexercised birds; 27% of those not treated with lead failed to reach criterion, whereas 70% of the leadtreated birds failed. For exercised birds, 24% of those not treated with lead failed to reach criterion whereas 47% of lead-treated birds failed. Thus, exercise improved performance of the lead-treated birds more than it improved the performance of the birds not treated with lead.
4. Discussion
2 Control - No Exercise
4.1. Lead effects on behavior
Control - Exercise
1
Lead- No Exercise Lead - Exercise
Walk Forward (s)
100 90 80 70 60 50 40 30 20
7
11 Days Post – Injection
17
Fig. 2. Number of calls, and total time walking forward, when placed on a treadmill. Shown are means7SD.
In these experiments, lead was the most significant factor affecting behavior and endurance on the treadmill test. We tested three key components with the treadmill: learning to orient forward quickly, learning to maintain the appropriate direction and speed, and endurance. Lead disrupted all three of these aspects of behavior. Mean (7SE) blood lead concentrations at the end of the study were: lead-treated (nonexercised) 22.771.7 mg/dL; lead-treated (exercised) 25.271.1 mg/dL; control (nonexercised) 2.170.2 mg/dL; and control (exercised) 2.270.2 mg/dL. However, in a previous experiment with this exercise protocol, lead concentrations in the brain of exposed gulls that did not exercise were nearly twice those of exposed gulls that exercised (measured at 45
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
141
Table 3 Endurance among groups of herring gulls (combining all three tests) Sample size
Endurance timea
Percent tiringb
Mean7SE
w2 (P)
%
w2 (P)
208
102.871.5
22.9 (0.0001)
41
26.9 (0.001)
Lead-exposed birds Exercise Nonexercised Exercised vs. nonexercised
98 53 44
98.972.7 101.673.5 95.474.1
Birds not lead-treated Exercised birds Nonexercised Exercised vs. nonexercised
111 49 62
All lead-exposed vs. all nonexposed All exercised vs. all nonexercised Exercised birds (lead-exposed vs. unexposed) Nonexercised birds (lead-exposed vs. unexposed)
208 208 102 106
Among all four groups
58 47 70 2.7 (NS)
110.771.7 110.772.6 110.672.4
5.3 (0.05) 26 24 27
0.1 (NS)
106.072.2 104.372.3
0.1 0.4 5.1 16.7
(NS) (NS) (0.02) (0.0001)
0.1 (NS)
36 45
21.4 1.7 6.3 12.3
(0.001) (NS) (0.02) (0.001)
NS, not significant. a Kruskal–Wallace nonparametric, one-way ANOVA run on actual endurance times. b Contingency table w2 comparing proportion tiring (o120 s) vs. enduring (running for 120 s).
days of age; Burger and Gochfeld, 2003b). Similar differences were observed for the kidney, but not for muscle, blood, or liver. Lead disrupts expression of synaptic neural cell adhesion molecules in herring gull brains (Dey et al., 2000). Lead was the most significant variable contributing to variations in latency to face forward. Lead disrupted the birds’ ability to learn to face forward immediately, whereas previous exposure to the treadmill (through daily exercise on the treadmill) improved performance in both lead-exposed birds and controls. Experience with the treadmill had a greater effect on lead-exposed birds than on controls, suggesting that birds that are not impaired could learn more quickly the need to face forward, and respond immediately, whereas lead-impaired birds without experience on the treadmill were more seriously affected. The duration of walking forward was a combination of actively walking forward at a sufficient speed not to fall behind and hit the back of the test apparatus. Thus the birds had to adjust their speed to that of the treadmill. Some birds that touched the back continued to walk at a speed which allowed them to perform, but others fell down or tripped when they hit the back. Optimal performance required that they learn to adjust their speed to that of the treadmill. Again, the differences as a function of exercise were greater for the lead-treated birds than the control birds. That is, control nonexercised birds learned the task better than lead-exposed nonexercised birds, consistent with a perceptual, cognitive, and/or motor impairment. Endurance was also reduced most significantly by lead.
Taken together, these results indicate that lead affects learning, endurance, and performance on a treadmill, and that daily experience with the treadmill results in improved performance in all birds, but that control birds are less affected by a lack of daily experience with the treadmill than are lead-exposed birds. We have no explanation for the poorer performance of the leadtreated nonexercise birds on the second test than on the first or third test, although it may be due to toxicokinetic changes in brain lead concentrations. The 2-min experience during test 1 provides a learning opportunity for the chicks but not sufficient practice or conditioning experience to influence their subsequent behavior. We suggest that these experiments indicate that lead disrupts learning and endurance, and that experience with the test (the treadmill) improves performance more for lead-exposed birds than for birds not exposed to lead. Birds not injected with lead seem able to learn more quickly what the task entails, whereas lead-treated nonexercised birds were slower to learn. 4.2. Implications of toxicological effects on chicks in nature In these experiments both lead and exercise affected the learning ability and endurance of the chicks. Exercise seemed to lessen the effect of lead, lessening the latency to respond, increasing the ability to learn to face forward and to walk forward, and increasing endurance. We believe these results suggest a modifying of cognitive or perceptual/motor dysfunction, which may be caused by lower levels of lead reaching the brain
ARTICLE IN PRESS 142
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
in the exercised birds. This could occur if (1) exercise allowed the birds to excrete more lead, (2) exercise allowed the birds to sequester more lead in their bones by stimulating osteogenesis (therefore less lead was available to cross the brain/blood barrier), (3) exercise resulted in their eating more, thus diluting the effect of the lead, or (4) a combination of the above. In 45-dayold chicks, brain lead concentrations of nonexercised chicks are almost twice as high as those of exercised birds (Burger and Gochfeld, 2003b). Exercise accelerates deposition and remodeling of bones (Mykkanen and Wasserman, 1981), and because lead is deposited in bone during active formation and resorption (Hamilton and O’Flaherty, 1995), we expected that more lead would be deposited in the bones of exercised birds, lessening the lead available to reach the brain. Presumably, lower levels of lead reaching the brain would lessen the cognitive effects observed in this study with exercise. In nature, herring gull chicks rarely experience the need for prolonged endurance, except during a predator disturbance when chicks need to find cover quickly. However, in an open colony, without abundant vegetation to provide shelters from predators, young herring gulls escape from intruders by running long distances to find cover. Under these circumstances, lead would impair performance, and previous exercise would lessen the effects. Learning, as indicated by the ability to face forward quickly on the treadmill, was less impaired in leadexposed exercised chicks compared with lead-exposed, nonexercised chicks. Thus, chicks on nests with sufficiently large territories to allow frequent movements within their territory might suffer less from the effects of lead than chicks with small territories. In a larger sense, this suggests that crowding would increase the effects of lead. Overall these experiments show that lead treatment has the greatest effect on treadmill behavior (both learning and endurance), and that exercise can moderate these effects. This has implications for studies with other animals, particularly rodents and lizards for which sprint speed and endurance are critical to understanding the relationships among ecology, behavior, and physiology (Arnold, 1983; Djawdan and Garland, 1988; Bauwens et al., 1995). 4.3. Methodological considerations The selection of a single intraperitoneal dose may be problematic. Although the dose selected produces lead feather concentrations within the range of what we have found in wild chicks (Burger and Gochfeld, 1990), there is probably a difference both in the kinetics and toxicity of the acute vs. subacute exposure and in the effect (Cory-Slechta, 1990). It would be rare for chicks to
receive as large a dose from a single, especially contaminated food source. Additional experiments should be conducted examining the effects of a single vs. a split dose. Moreover, an intraperitoneal dose may be more efficiently absorbed into the bloodstream, resulting in quicker and perhaps more severe effects than an equal dose ingested.
Acknowledgments We thank T. Benson, C. Dixon, J. Ondrof, R. Ramos, T. Shukla, and M. McMahon, as well as a number of undergraduate research assistants, who helped care for the gulls, exercised them, and aided in data analysis and graphics. We thank B.D. Goldstein, C. Powers, and J. Moore for comments on the manuscript. We thank the US Fish and Wildlife Service, the New York Department of Conservation, and the New Jersey Department of Environmental Protection for permits to collect the gulls. This research was approved by the University Animal Review Board and was funded by the Consortium for Risk Evaluation with Stakeholder Participation (CRESP) through the Department of Energy (AI#DE-FC01-95EW55084, DE-FG 26-00NT 40938), and NIEHS (ESO 5022).
References Alber, S.A.B., Strupp, J., 1996. An in-depth analysis of lead effects in a delayed spatial alternation task: assessment of mnemonic effects, side bias, and proactive interference. Neurotoxicol. Teratol. 18, 3–15. Arnold, S.J., 1983. Morphology performance and fitness. Am. Zool. 23, 347–361. ATSDR, 1988. The Nature and Extent of Lead Poisoning in Children in the United States. A Report to Congress. Agency for Toxic Substances and Disease Registry, US Public Health Service, Atlanta. ATSDR, 1999. Toxicological Profile for Lead. Agency for Toxic Substances and Disease Registry, US Public Health Service, Atlanta. Bauwens, D., Garland Jr, T., Castilla, A.M., van Damme, R., 1995. Evolution of sprint speed in lacertid lizards: morphological, physiological, and behavioral covariation. Evolution 49, 848–865. Bellinger, D., Leviton, A., Waternaux, C., Needleman, H., Rabinowitz, M., 1987. Longitudinal analyses of prenatal and postnatal lead exposure and early cognitive development. N. Engl. J. Med. 316, 1037–1043. Bunn, T.L., Marsh, J.A., Dietert, R.R., 2000. Gender differences in developmental immunotoxicity to lead in the chicken: analysis following a single early low-level exposure in ovo. J. Toxicol. Environ. Health Pt A 61, 677–693. Burger, J., 1984. Pattern, mechanism, and adaptive significance of territoriality in herring gulls (Larus argentatus). Ornithol. Monogr. 34, 1–91. Burger, J., 1990. Behavioral effects of early postnatal lead exposure in herring gulls (Larus argentatus). Pharmacol. Biochem. Behav. 36, 7–13.
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144 Burger, J., 1998. Effects of lead on sibling recognition in young herring gulls. Toxicol. Sci. 43, 155–160. Burger, J., Gochfeld, M., 1985. Early postnatal lead exposure: behavioral effects in common tern chicks Sterna hirundo. J. Toxicol. Environ. Health 29, 869–886. Burger, J., Gochfeld, M., 1988. Effects of lead on growth in young herring gulls (Larus argentatus). J. Toxicol. Environ. Health 25, 227–236. Burger, J., Gochfeld, M., 1990. Tissue levels of lead in experimentally exposed herring gull (Larus argentatus) chicks. J. Toxicol. Environ. Health 29, 219–233. Burger, J., Gochfeld, M., 1993a. Lead and behavioral development in young herring gulls: effects of timing of exposure on individual recognition. Fund. Appl. Toxicol. 21, 187–195. Burger, J., Gochfeld, M., 1993b. Lead and cadmium accumulation in eggs and fledgling seabirds in the New York Bight. Environ. Toxicol. Chem. 12, 261–267. Burger, J., Gochfeld, M., 1994. Behavioral impairments of leadinjected young herring gulls in nature. Fund. Appl. Toxicol. 23, 553–561. Burger, J., Gochfeld, M., 1995a. Effects of varying temporal exposure to lead on behavioral development in herring gull (Larus argentatus) chicks. Pharmacol. Biochem. Behav. 52, 601–608. Burger, J., Gochfeld, M., 1995b. Behavior effects of lead exposure on different days for gull (Larus argentatus) chicks. Pharmacol. Biochem. Behav. 50, 97–105. Burger, J., Gochfeld, M., 1996. Lead and behavioral development: parental compensation for behaviorally impaired chicks. Pharmacol. Biochem. Behav. 55, 339–349. Burger, J., Gochfeld, M., 1997. Lead and neurobehavioral development in gulls: a model for understanding effects in the laboratory and the field. Neurotoxicol. 18, 279–287. Burger, J., Gochfeld, M., 2000. Effects of lead on larids: a review of laboratory and field studies. J. Toxicol Environ. Health. Pt. B 3, 59–78. Burger, J., Gochfeld, M., 2003a. Spatial and temporal patterns in metal levels in eggs of common terns (Sterna hirundo) in New Jersey. Science Total Environ., in press. Burger, J., Gochfeld, M., 2003b. Lead in young herring gulls: paradoxical effects of exercise on tissue concentration. J. Toxicol. Environ. Health Pt. A 66, 181–193. Burger, J., Shukla, T., Benson, T., Gochfeld, M., 1994. Lead levels in exposed herring gulls: differences in the field and laboratory. In: Johnson, B.L., Xintaras, C., Andrews Jr, J.S. (Eds.), Hazardous Waste: Impacts on Human and Ecological Health. Princeton Scientific Publ. Co, Princeton, NJ, pp. 115–123. Burright, R.G., Engellenner, W.J., Donovick, P.J., 1983. Lead exposure and agonistic behavior of adult mice of two ages. Physiol. Behav. 30, 285–288. Bushnell, P.J., Bowman, R.E., 1979a. Persistence of impaired reversal learning in young monkeys exposed to low levels of dietary lead. J. Toxicol. Environ. Health 5, 1015–1023. Bushnell, P.J., Bowman, R.E., 1979b. Effects of chronic lead ingestion on social development in infant rhesus monkeys. Neurobehav. Toxicol. 1, 207–219. Cory-Slechta, D.A., 1990. Exposure duration modifies the effect of low-level lead on fixed interval performance. Neurotoxicol. 11, 427–442. Cory-Slechta, D.A., Schaumburg, H.H., 2000. Lead, inorganic. In: Spencer, P.S., Schaumburg, H.H., Ludolph, A.C. (Eds.), Experimental and Clinical Neurotoxicology, 2nd Edn. Oxford University Press, New York, pp. 708–720. Cory-Slechta, D.A., Weiss, B., Cox, C., 1983. Delayed behavioral toxicity of lead with increasing exposure concentration. Toxicol. Appl. Pharmacol. 71, 342–352.
143
Cutler, M.G., 1977. Effects of exposure to lead on social behaviour in the laboratory mouse. Psychopharmacology 52, 279–282. Dey, M., Burger, J., Gochfeld, M., Reuhl, K.R., 2000. Developmental lead exposure disturbs expression of synaptic neural cell adhesion molecules in herring gull brains. Toxicology 146, 137–147. Dietrich, K.M., Ris, M.D., Succop, P.A., Berger, O.G., Bornschein, R.L., 2001. Early exposure to lead and juvenile delinquency. Neurotoxicol. Teratol. 23, 511–518. Dietz, D.D., McMillan, D.E., Mushak, P., 1979. Effects of chronic lead administration on acquisition and performance of serial position sequences by pigeons. Toxicol. Appl. Pharmacol. 47, 377–384. Djawdan, M., Garland Jr, T., 1988. Maximal running speed of bipedal and quadrupedal rodents. J. Mamm. 69, 765–772. Dohm, M.R., Hayes, J.P., Garland Jr, T., 1996. Quantitative genetics of sprint running speed and swimming endurance in laboratory house mice (Mus domesticus). Evolution 50, 1688–1701. Dolinsky, Z.S., Burright, R.G., Donovick, P.J., 1983. Behavioral changes in mice following lead administration during several stages of development. Physiol. Behav. 30, 583–589. Donald, J.M., Cutler, M.G., Moore, M.R., Bradley, M., 1981. Development and social behavior in mice after prenatal and postnatal administration of low levels of lead acetate. Neuropharmacology 20, 1097–1104. Eisler, R., 1988. Lead Hazards to Fish, Wildlife, and Invertebrates: a Synoptic Review. US Fish and Wildlife Service, Washington, DC. Hamilton, J.D., O’Flaherty, E.J., 1995. Influence of lead on mineralization during bone growth. Fund. Appl. Toxicol. 26, 265–271. Lasky, R.E., Luck, M.L., Torre III, P., Laughlin, N., 2001. The effects of early lead exposure on auditory function in rhesus monkeys. Neurotoxicol. Teratol. 23, 638–649. Laughlin, N.K., Bushnell, P.J., Bowman, R.E., 1991. Lead exposure and diet: differential effects on social development in the rhesus monkey. Neurotoxicol. Teratol. 13, 429–440. Lidsky, T.I., Schneider, J.S., 2003. Lead neurotoxicity in children: basic mechanisms and clinical correlates. Brain 126 (Pwt 1), 5–19. Lilienthal, H., Winneke, G., 1996. Lead effects on the brain stem auditory evoked potential in monkeys during and after the treatment phase. Neurotoxicol. Teratol. 18, 17–32. Lippman, M., 1990. Lead and human health: background and recent findings. Environ. Res. 51, 1–24. Morgan, R.E., Garavan, H., Smith, E.G., Driscoll, L.L., Levitsky, D.A., Strupp, B.J., 2001. Early lead exposure produces lasting changes in sustained attention, response initiation, and reactivity to errors. Neurotoxicol. Teratol. 23, 519–531. Mykkanen, H.M., Wasserman, R.H., 1981. Gastrointestinal absorption of lead (203Pb) in chicks: influence of lead, calcium, and age. J. Nutr. 111, 1757–1765. Needleman, H., 1995. Environmental lead and children’s intelligence studies included in the meta–analysis are not representative. Br. Med. J. 310, 1408. Needleman, H.L., Schell, A., Bellinger, D., Leviton, A., Allred, E.N., 1990. The long-term effects of exposure to lead in childhood, An 11 year follow-up report. N. Engl. J. Med. 322, 83–88. Newland, M.C., Yezhou, S., Logdberg, B., Berlin, M., 1996. In utero lead exposure in squirrel monkeys: motor effects seen with schedule-controlled behavior. Neurotoxicol. Teratol. 18, 33–40. Payton, M., Riggs, K.M., Spiro III, A., Weiss, S.T., Hu, H., 1998. Relations of bone and blood lead to cognitive function: the VA normative aging study. Neurotoxicol. Teratol. 20, 18–27. Pocock, S.J, Smith, M., Baghurst, P., 1994. Environmental lead and children’s intelligence: a systematic review of the epidemiological evidence. Br. Med. J. 309, 1189–1197.
ARTICLE IN PRESS 144
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
Rice, D.C., 1984. Behavioral deficit (delayed matching to sample) in monkeys exposed from birth to low levels of lead. Toxicol. Appl. Pharmacol. 75, 337–345. Rice, D.C., 1996. Behavioral effects of lead: commonalities between experimental and epidemiologic data. Environ. Health Perspect. 104, 337–351.
SAS (Statistical Analysis Systems), 1995. SAS Users’ Guide. Statistical Institute, Inc., Cary, NC. Siegel, S., 1956. Nonparametric Statistics. McGraw-Hill, NY. Silbergeld, E.K., 1990. Toward the twenty-first century: lessons from lead and lessons yet to learn. Environ. Health Perspect. 86, 191–196.
Ecotoxicology and Environmental Safety 57 (2004) 136–144
Effects of lead and exercise on endurance and learning in young herring gulls Joanna Burger and Michael Gochfeld1 Consortium for Risk Evaluation with Stakeholder Participation, Environmental and Occupational Health Sciences Institute, Rutgers University, Piscataway, NJ 08854, USA Received 14 June 2002; received in revised form 13 March 2003; accepted 14 March 2003
Abstract In this paper, we report the use of young herring gulls, Larus argentatus, to examine the effect of lead and exercise on endurance, performance, and learning on a treadmill. Eighty 1-day-old herring gull chicks were randomly assigned to either a control group or a lead treatment group that received a single dose of lead acetate solution (100 mg/kg) at day 2. Controls were injected with an equal volume of isotonic saline at the same age. Half of the lead treatment group and half of the control group were randomly assigned to an exercise regime of walking on a treadmill twice each day. The other group remained in their cages. We test the null hypotheses that neither lead nor exercise affected performance of herring gull chicks when subsequently tested on the treadmill at 7, 11, and 17 days post-injection. Performance measures included latency to orient forward initially, to move continuously, forward on the treadmill, and to avoiding being bumped against the back of the test chamber. Also measured were the number of calls per 15 s, and the time to tire out. Latency to face forward and avoiding being bumped against the back of the test chamber were measures of learning, and time to tire out was a measure of endurance. We found significant differences as a function of lead, exercise, and their interaction, and rejected the null hypotheses. For all measures of behavior and endurance, lead had the greatest contribution to accounting for variability. In general, lead-treated birds showed better performance improvement from the daily exercise than did controlled non-lead birds, with respect to endurance and learning. We suggest that in nature, exercise can improve performance of lead-exposed birds by partially mitigating the effects of lead, thereby increasing survival of lead-impaired chicks. r 2003 Elsevier Science (USA). All rights reserved. Keywords: Lead exposure; Exercise; Learning; Endurance; Treadmill; Fatigue
1. Introduction Governmental agencies as well as the public and private sectors are all concerned about the potential adverse effects of contaminants on neurobehavioral development. Although much of the recent discussion has dealt with chemicals that cause disruptions in the endocrine system, the neurodevelopmental effects of lead exposure continue to cause concern for humans as well as other animals. Lead, derived from industrial and urban pollution, from agricultural runoff, as well as Corresponding author. Division of Life Sciences, Nelson Biological Laboratory, Rutgers University, 604 Allison St. Nelson Hall, Piscataway, NJ 08854-8082, USA. Fax: +732-445-5870. E-mail address: [email protected] (J. Burger). 1 Present address: Environmental and Community Medicine, UMDNJ-Robert Wood Johnson Medical School, Piscataway, NJ 08854, USA.
from natural geochemical processes, enters the air, soil, and water, as well as the food chain. The neurotoxicity of lead has been extensively studied (Cory-Slechta and Schaumburg, 2000). In recent years, decreases in the use of leaded gasoline and lead-based paints in the USA have resulted in reductions in both lead exposure and human blood lead concentrations (ATSDR, 1999). However, lead is one of the most common contaminants at Superfund and other contaminated sites (ATSDR, 1999). Moreover, lead levels are increasing in some cohorts of children (ATSDR, 1988, 1999) and in some species of birds (Burger et al., 1994), but not in others (Burger and Gochfeld, 2003a). There is increasing evidence that lead affects a number of cognitive and motor functions in a wide range of vertebrates, including humans. Animal models provide information on the effects of low- and mediumlevel lead effects on neural and cognitive processes
0147-6513/03/$ - see front matter r 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0147-6513(03)00035-6
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
(Eisler, 1988; Lippman, 1990; Alber and Strupp, 1996; Burger and Gochfeld, 2000), and for the most part, the animal models and epidemiological data provide similar results (Rice, 1996). In humans and other primates, lead exposure at low or medium levels leads to auditory, neurobehavioral, hematological, nephrotoxic, and reproductive effects (Rice, 1984, 1996; Bellinger et al., 1987; Payton et al., 1998; Lasky et al., 2001). Even at very low doses, lead exposure can produce serious adverse effects on the central nervous system of children and infants, and these effects last for years (Needleman et al., 1990; Dietrich et al., 2001; Morgan et al., 2001). The relationship between lead exposure and cognitive development in children has engendered lively controversy. A meta-analysis by Pocock et al. (1994), reviewing 26 studies and concluding that lead had only a minimal effect, received extensive criticism for excluding key studies (e.g., Needleman, 1995). As early as 1990, Silbergeld (1990) suggested that the neurodevelopmental effects of lead may not have a detectable threshold, and recent authors support this view (Lidsky and Schneider, 2003). Furthermore, neurotoxic effects have been noted in a variety of other mammals and in birds (Dietz et al., 1979; Cory-Slechta et al., 1983; Burger and Gochfeld, 1985, 1988; Eisler, 1988; Alber and Strupp, 1996; Lilienthal and Winneke, 1996; Newland et al., 1996; Bunn et al., 2000). Lead also disrupts social behavior in a number of species, including rodents (Cutler, 1977; Donald et al., 1981; Burright et al., 1983; Dolinsky et al., 1983) and primates (Bushnell and Bowman, 1979a, b; Laughlin et al., 1991). For 15 years we have been investigating the effects of low-level lead exposure on the behavior of young birds, particularly the herring gull, Larus argentatus, in the laboratory. Lead exposure delays individual recognition of both parents and siblings (Burger and Gochfeld, 1993a; Burger, 1998), locomotion and balance (Burger, 1990), thermoregulation and visual cliff behavior (Burger and Gochfeld, 1995a, b). Moreover, similar effects occur in nature (Burger and Gochfeld, 1994, 1996, 1997, 2000). These experiments have been useful in providing paradigms to understand and study effects of lead in humans and other primates, and to understand the effects of lead on behavior, physiology, and survival of birds in nature. Birds can be useful models for metal toxicity because they share with humans a reliance on visual and vocal communication, in contrast to the primarily ultrasonic, tactile, and olfactory communication of rodents. Herring gull chicks move about the nest within a few hours of hatching, and by 2–3 days of age they walk easily around their parents’ territory of a few square meters (Burger, 1984). In this paper young herring gulls were used to test the null hypotheses that there are no
137
differences in endurance and learning as a function of lead treatment and exercise regime, using a treadmill test. Whereas our previous studies have shown deficits in individual recognition as a function of exposure to lead (Burger and Gochfeld, 1993a), the effects of lead on learning a new skill (running on a treadmill) have not been examined in an avian model. Further more, endurance has not been examined in birds. Running speed and endurance are characteristics often examined in mammals (Djawdan and Garland, 1988; Dohm et al., 1996) and lizards (Bauwens et al., 1995). Treadmill performance is used as a method of integrating physiological and ecological approaches to adaptation (Arnold, 1983). In both mice and lizards, running is an important aspect of their daily behavior used to escape danger (both species) and capture food (lizards in pursuit of insects). Gull chicks are seldom exposed to a situation that challenges endurance or sprint speed because they spend most of their time sitting or standing on the nest waiting to be fed or hiding in nearby vegetation until they are old enough to fly. During this time they are protected by at least one parent, reducing the need for predator avoidance. However, when disturbed, they may need to run vigorously to escape potential predators. In these tests we used a treadmill to examine learning, performance, and endurance. To perform successfully, the gulls must learn to face forward, to walk continuously forward on the moving treadmill, and to avoid being bumped against the back of the test area (a mildly annoying stimulus). It was not clear before our pilot tests that the chicks could learn the task, would learn the task, or would continue to perform. We predicted that gulls that were afforded the opportunity to exercise daily on the treadmill would perform better when tested with the treadmill operating at a faster speed than the exercise speed. Furthermore, we expected that lead would disrupt their ability to learn how to move on the treadmill and would reduce their endurance as well.
2. Materials and methods 2.1. Source and handling of study birds Eighty 1-day-old herring gull chicks were collected from nesting colonies at Captree, Long Island and Barnegat Bay, New Jersey in 1997, under appropriate federal and state permits. To minimize effects on reproductive success, only the first-hatched chick in any nest was collected. After marking with numbered leg bands for individual identification, chicks were randomly assigned to one of four groups: lead treatment or saline control, with or without exercise.
ARTICLE IN PRESS 138
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
Chicks were housed in 50 50 50-cm stainless-steel cages in groups of three to simulate natural broods (Burger, 1984), and were maintained in a warm laboratory at 2772 C with a natural light/dark cycle (15 h light, 9 h dark). Chicks were fed three to five times daily with fresh fish and seafood, supplemented with canned cat and dog food, eggs, and mice. Young gulls do best with a varied diet. Chicks were individually fed by caretakers until they learned to feed on their own; most fed voraciously throughout the study. Weight matching was effective. Research assistants rather than routine animal keepers did all care and handling of the chicks, making daily notes on each chick, including what they ate, whether they ate appropriately, and whether they appeared alert and healthy. Chicks gained weight appropriately for their age, and there were no differences in weight between control and experimental chicks at the end of the experiment (w2 tests). Only chicks that remained healthy were included in the data set. In nature, B30% of herring gulls chicks die in the first week from a variety of causes (Burger, 1984); thus it is not surprising that some chicks become ill or die in the laboratory. Because we were interested in the effects of lead on cognitive function and endurance, and not on physical impairment due to other causes, such chicks were eliminated from the study. Within each treatment group, chicks were assigned to pseudo-sibling groups based on weight. It was essential to have ‘‘siblings’’ similar in weight to avoid undue fighting in the cages (chicks in nature normally fight with their siblings: see Burger, 1984). Furthermore, all three chicks in each sibling group had the same exposure. This was essential to avoid cross-contamination, because chicks may ingest the defecations of their cagemates. This avoided any possibility of controls obtaining lead from siblings. 2.2. Lead exposure On day 2 lead-treated chicks were given an intraperitoneal injection of lead acetate (total dose=100 mg/kg lead in sterile water) (25 mg/mL). Controls received an equivalent volume (based on their weight) of isotonic saline solution. Lead was injected by a technician not otherwise involved in the behavioral tests, and all caretakers were blind to the treatment status of the gulls they handled. Lead was administered by injection rather than in food because chicks that eat different amounts would receive different doses. This dose was selected because it yields tissue levels within the range found in young herring gulls in the wild; this dose results in mean blood lead levels of 16.5–25 mg/dL at fledging (Burger and Gochfeld, 1990). The levels of lead in feathers of lead-injected birds averaged 4.8 ppm, which is similar to amounts reached
in some wild herring gulls in the northeastern USA, which have lead concentrations in their feathers of 0.5– 8.5 ppm (Burger and Gochfeld, 1993b). There is a high correlation between lead concentrations in blood and in feathers (r ¼ 0:79; Burger and Gochfeld, 1990), as well as between those in blood and in the brain (r ¼ 0:83; Burger and Gochfeld, 1990). In the current experiment, blood lead concentrations for controls averaged 2.270.2 mg/dL, and those in lead-treated birds averaged 2477.8 mg/dL. 2.3. Exercise treatment Half of the lead-treated chicks and half of the control chicks were subjected to a daily exercise regime. Nonexercised chicks were not allowed to exercise except for normal movements within their cages. We exercised chicks on a human treadmill set at a speed of 200 cm/ min. For their exercise regime they were placed on the treadmill with their cagemates. To prevent injuries, a cardboard frame was placed over the treadmill so that the chicks did not fall off; thus chicks bumping against the back experienced mild discomfort. Chicks were watched at all times, and when a chick fell over or could not maintain the pace, it was removed from the treadmill. From 4 to 7 days of age, the chicks exercised on the treadmill for 5 min twice a day; from day 8 to 12, they were exercised for 10 min. This was increased incrementally by 5 min every 3 days up to 20 min. 2.4. Testing We conducted performance and endurance tests on the treadmill for all chicks at 7, 11, and 17 days after lead injection when they were 9, 13, and 19 days of age. During the test, chicks were placed individually on the treadmill and observed for 2 min at a faster treadmill speed of 300 cm/min. During that time we recorded the latency to face forward initially, latency to face forward continually for the duration of the test, time touching the back of the apparatus, time walking forward, time to tire out, and number of calls given in the first 15 s. Time to tire out is the time until they fell down or stopped walking, and is a measure of endurance. If they continued to walk for the 2-min test, this value was set at 120 s. Latency to face forward (initial and continued) and time at the back touching the box were measures of learning and motor performance. 2.5. Statistical tests Analysis of variance (ANOVA) was used to determine if treatment or age contributed to differences in behavioral responses (PROC GLM, SAS, 1995). We used lead treatment, exercise status, and test day as well as interactions as the independent variables. We also
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
analyzed the data by date. We compared performance of lead vs. controls and exercise vs. nonexercise with Kruskal–Wallis tests (Siegel, 1956). To compensate for low power, we list P values o0.10.
Control - No Exercise
Lead- No Exercise
Control - Exercise
Lead - Exercise
Time to First Face Forward (s)
70
Face Forward Continually (s)
3. Results Overall, the models indicated that lead treatment had the greatest effect on variations in the response measures, with exercise and date (that is, age) also contributing considerably to most models (except for number of calls, Table 1). Interactions between variables were generally not significant, except for latency to face forward (Table 1). The models for response measures were also significant by days post-injection, with lead treatment generally contributing the most. Exercise did not enter as a significant variable during the first test, when the exercised chicks had experienced the treadmill for only 3 days. However, it had a major impact on the next test 4 days later. The lead-treated chicks with no exercise showed the greatest behavioral deficits on all measures (Table 2), taking the longest to face forward initially and to continue facing forward, spending the greatest amount of time at the back of the box and the least amount of time walking forward freely, and they tired out the quickest. Although there were differences within the control chicks as a function of whether they were
139
60 50 40 30 20 10 0 100 90 80 70 60 50 40 30 20 10 0
7
11
17
Days Post-Injection Fig. 1. Time to first face forward, and to face forward continuously, when placed on a treadmill. Shown are means7SE.
Table 1 Linear models explaining variations in behavioral and endurance variables for herring gull young Behaviors
Latency to turn forward Latency: continuous forward Time at back Time spent walking Time to tire (s) Calls/15 s a
Overall
Model
Variables entering the model
R2
F
P
Lead P
Exercise P
Lead exercise P
Date (age) P
0.19 0.24 0.24 0.23 0.11 0.21
9.54 12.70 12.56 11.99 5.01 10.6
0.0001 0.0001 0.0003 0.0001 0.0002 0.0001
o0.0001 o0.0001 o0.0001 o0.0001 o0.0002 o0.0001
o0.005 o0.0008 o0.0002 o0.0002 NSa o0.0014
o0.011 o0.003 o0.08 NS NS o0.10
o0.013 o0.0001 o0.0006 o0.0005 o0.017 o0.009
NS=not significant.
Table 2 Overall means for behavioral responses on treadmill tests for herring gull chicksa
Number of birds Latency to face forward initially (s) Latency to face forward continually (s) Total time in back (s) Time walking (s) Endurance (up to 120 s) No. calls/15 s a
No lead, no exercise
No lead, exercise
Lead, no exercise
Lead+exercise
Wilcoxon w2 (P)
18 18.673.0 (B) 35.274.6 (B) 51.174.5 (C) 67.574.4 (C) 11172.4 (B) 2.870.16 (B)
17 17.473.2 33.675.6 41.775.5 78.175.5 11172.6 2.570.2
16 51.676.6 (A) 76.976.5 (A) 87.974.6 (A) 33.674.3 (A) 95.074.1 (A) 4.470.31 (A)
17 28.674.3 (B) 42.775.3 (B) 61.075.0 (B) 58.775.0 (B) 10273.5 (A) 3.370.19(B)
30.6 24.8 11.3 21.1 14.5 31.7
Letters indicate significant differences (Duncan multiple range test).
(B) (B) (C) (C) (B) (B)
(0.0001) (0.0001) (0.01) (0.0001) (0.002) (0.0001)
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
exposed to exercise this effect was smaller than in the lead-treated group (Table 2). The latency to face forward initially, and to continue to face forward, are measures of learning how to perform on the moving treadmill. The pattern among the three tests was complicated (Fig. 1), although overall, the lead-treated, nonexercised birds performed the worst (Table 2). In general, after the first test, chicks with exercise experience performed better than those with no regular experience. Latency to face forward generally decreased with age and experience. The exception was the lead-treated, nonexercised birds. During the second test (11 days post-injection) the lead-treated, nonexercised birds performed much worse than on the other two tests. Time (in seconds) to walk forward, without touching the back, was also a measure of learning because the gulls had to learn to continue walking, and to walk at the appropriate speed to keep in the center of the box. In general, birds with previous exercise improved steadily in performance, whether they had lead or not (Fig. 2). Control birds performed better on the first and last tests than on the middle test, and we have no explanation for this deficit. Calling, a measure of anxiety, generally decreased with time, except for the lead-treated, nonexercised birds (Fig. 2). 6
Number of Calls
5
4
3
120 115 110
Endurance (s)
140
105 100 95 90 Control -No Exercise
85
Control -Exercise Lead-No Exercise
80
Lead-Exercise
75 11
7
17
Days Post - Injection Fig. 3. Endurance on a treadmill test. Shown in seconds to tire out (means7SE).
We report two measures of endurance: time to tire (in seconds), and the number of birds that did not complete the 120-s test. Time to tire out was significantly affected by lead at every age (see Fig. 3). The exercised birds generally improved in performance over time. There were also significant differences among the four groups in the percentage of chicks that failed to endure for the 120-s test (Table 3). The differences were greatest for the nonexercised birds; 27% of those not treated with lead failed to reach criterion, whereas 70% of the leadtreated birds failed. For exercised birds, 24% of those not treated with lead failed to reach criterion whereas 47% of lead-treated birds failed. Thus, exercise improved performance of the lead-treated birds more than it improved the performance of the birds not treated with lead.
4. Discussion
2 Control - No Exercise
4.1. Lead effects on behavior
Control - Exercise
1
Lead- No Exercise Lead - Exercise
Walk Forward (s)
100 90 80 70 60 50 40 30 20
7
11 Days Post – Injection
17
Fig. 2. Number of calls, and total time walking forward, when placed on a treadmill. Shown are means7SD.
In these experiments, lead was the most significant factor affecting behavior and endurance on the treadmill test. We tested three key components with the treadmill: learning to orient forward quickly, learning to maintain the appropriate direction and speed, and endurance. Lead disrupted all three of these aspects of behavior. Mean (7SE) blood lead concentrations at the end of the study were: lead-treated (nonexercised) 22.771.7 mg/dL; lead-treated (exercised) 25.271.1 mg/dL; control (nonexercised) 2.170.2 mg/dL; and control (exercised) 2.270.2 mg/dL. However, in a previous experiment with this exercise protocol, lead concentrations in the brain of exposed gulls that did not exercise were nearly twice those of exposed gulls that exercised (measured at 45
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
141
Table 3 Endurance among groups of herring gulls (combining all three tests) Sample size
Endurance timea
Percent tiringb
Mean7SE
w2 (P)
%
w2 (P)
208
102.871.5
22.9 (0.0001)
41
26.9 (0.001)
Lead-exposed birds Exercise Nonexercised Exercised vs. nonexercised
98 53 44
98.972.7 101.673.5 95.474.1
Birds not lead-treated Exercised birds Nonexercised Exercised vs. nonexercised
111 49 62
All lead-exposed vs. all nonexposed All exercised vs. all nonexercised Exercised birds (lead-exposed vs. unexposed) Nonexercised birds (lead-exposed vs. unexposed)
208 208 102 106
Among all four groups
58 47 70 2.7 (NS)
110.771.7 110.772.6 110.672.4
5.3 (0.05) 26 24 27
0.1 (NS)
106.072.2 104.372.3
0.1 0.4 5.1 16.7
(NS) (NS) (0.02) (0.0001)
0.1 (NS)
36 45
21.4 1.7 6.3 12.3
(0.001) (NS) (0.02) (0.001)
NS, not significant. a Kruskal–Wallace nonparametric, one-way ANOVA run on actual endurance times. b Contingency table w2 comparing proportion tiring (o120 s) vs. enduring (running for 120 s).
days of age; Burger and Gochfeld, 2003b). Similar differences were observed for the kidney, but not for muscle, blood, or liver. Lead disrupts expression of synaptic neural cell adhesion molecules in herring gull brains (Dey et al., 2000). Lead was the most significant variable contributing to variations in latency to face forward. Lead disrupted the birds’ ability to learn to face forward immediately, whereas previous exposure to the treadmill (through daily exercise on the treadmill) improved performance in both lead-exposed birds and controls. Experience with the treadmill had a greater effect on lead-exposed birds than on controls, suggesting that birds that are not impaired could learn more quickly the need to face forward, and respond immediately, whereas lead-impaired birds without experience on the treadmill were more seriously affected. The duration of walking forward was a combination of actively walking forward at a sufficient speed not to fall behind and hit the back of the test apparatus. Thus the birds had to adjust their speed to that of the treadmill. Some birds that touched the back continued to walk at a speed which allowed them to perform, but others fell down or tripped when they hit the back. Optimal performance required that they learn to adjust their speed to that of the treadmill. Again, the differences as a function of exercise were greater for the lead-treated birds than the control birds. That is, control nonexercised birds learned the task better than lead-exposed nonexercised birds, consistent with a perceptual, cognitive, and/or motor impairment. Endurance was also reduced most significantly by lead.
Taken together, these results indicate that lead affects learning, endurance, and performance on a treadmill, and that daily experience with the treadmill results in improved performance in all birds, but that control birds are less affected by a lack of daily experience with the treadmill than are lead-exposed birds. We have no explanation for the poorer performance of the leadtreated nonexercise birds on the second test than on the first or third test, although it may be due to toxicokinetic changes in brain lead concentrations. The 2-min experience during test 1 provides a learning opportunity for the chicks but not sufficient practice or conditioning experience to influence their subsequent behavior. We suggest that these experiments indicate that lead disrupts learning and endurance, and that experience with the test (the treadmill) improves performance more for lead-exposed birds than for birds not exposed to lead. Birds not injected with lead seem able to learn more quickly what the task entails, whereas lead-treated nonexercised birds were slower to learn. 4.2. Implications of toxicological effects on chicks in nature In these experiments both lead and exercise affected the learning ability and endurance of the chicks. Exercise seemed to lessen the effect of lead, lessening the latency to respond, increasing the ability to learn to face forward and to walk forward, and increasing endurance. We believe these results suggest a modifying of cognitive or perceptual/motor dysfunction, which may be caused by lower levels of lead reaching the brain
ARTICLE IN PRESS 142
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
in the exercised birds. This could occur if (1) exercise allowed the birds to excrete more lead, (2) exercise allowed the birds to sequester more lead in their bones by stimulating osteogenesis (therefore less lead was available to cross the brain/blood barrier), (3) exercise resulted in their eating more, thus diluting the effect of the lead, or (4) a combination of the above. In 45-dayold chicks, brain lead concentrations of nonexercised chicks are almost twice as high as those of exercised birds (Burger and Gochfeld, 2003b). Exercise accelerates deposition and remodeling of bones (Mykkanen and Wasserman, 1981), and because lead is deposited in bone during active formation and resorption (Hamilton and O’Flaherty, 1995), we expected that more lead would be deposited in the bones of exercised birds, lessening the lead available to reach the brain. Presumably, lower levels of lead reaching the brain would lessen the cognitive effects observed in this study with exercise. In nature, herring gull chicks rarely experience the need for prolonged endurance, except during a predator disturbance when chicks need to find cover quickly. However, in an open colony, without abundant vegetation to provide shelters from predators, young herring gulls escape from intruders by running long distances to find cover. Under these circumstances, lead would impair performance, and previous exercise would lessen the effects. Learning, as indicated by the ability to face forward quickly on the treadmill, was less impaired in leadexposed exercised chicks compared with lead-exposed, nonexercised chicks. Thus, chicks on nests with sufficiently large territories to allow frequent movements within their territory might suffer less from the effects of lead than chicks with small territories. In a larger sense, this suggests that crowding would increase the effects of lead. Overall these experiments show that lead treatment has the greatest effect on treadmill behavior (both learning and endurance), and that exercise can moderate these effects. This has implications for studies with other animals, particularly rodents and lizards for which sprint speed and endurance are critical to understanding the relationships among ecology, behavior, and physiology (Arnold, 1983; Djawdan and Garland, 1988; Bauwens et al., 1995). 4.3. Methodological considerations The selection of a single intraperitoneal dose may be problematic. Although the dose selected produces lead feather concentrations within the range of what we have found in wild chicks (Burger and Gochfeld, 1990), there is probably a difference both in the kinetics and toxicity of the acute vs. subacute exposure and in the effect (Cory-Slechta, 1990). It would be rare for chicks to
receive as large a dose from a single, especially contaminated food source. Additional experiments should be conducted examining the effects of a single vs. a split dose. Moreover, an intraperitoneal dose may be more efficiently absorbed into the bloodstream, resulting in quicker and perhaps more severe effects than an equal dose ingested.
Acknowledgments We thank T. Benson, C. Dixon, J. Ondrof, R. Ramos, T. Shukla, and M. McMahon, as well as a number of undergraduate research assistants, who helped care for the gulls, exercised them, and aided in data analysis and graphics. We thank B.D. Goldstein, C. Powers, and J. Moore for comments on the manuscript. We thank the US Fish and Wildlife Service, the New York Department of Conservation, and the New Jersey Department of Environmental Protection for permits to collect the gulls. This research was approved by the University Animal Review Board and was funded by the Consortium for Risk Evaluation with Stakeholder Participation (CRESP) through the Department of Energy (AI#DE-FC01-95EW55084, DE-FG 26-00NT 40938), and NIEHS (ESO 5022).
References Alber, S.A.B., Strupp, J., 1996. An in-depth analysis of lead effects in a delayed spatial alternation task: assessment of mnemonic effects, side bias, and proactive interference. Neurotoxicol. Teratol. 18, 3–15. Arnold, S.J., 1983. Morphology performance and fitness. Am. Zool. 23, 347–361. ATSDR, 1988. The Nature and Extent of Lead Poisoning in Children in the United States. A Report to Congress. Agency for Toxic Substances and Disease Registry, US Public Health Service, Atlanta. ATSDR, 1999. Toxicological Profile for Lead. Agency for Toxic Substances and Disease Registry, US Public Health Service, Atlanta. Bauwens, D., Garland Jr, T., Castilla, A.M., van Damme, R., 1995. Evolution of sprint speed in lacertid lizards: morphological, physiological, and behavioral covariation. Evolution 49, 848–865. Bellinger, D., Leviton, A., Waternaux, C., Needleman, H., Rabinowitz, M., 1987. Longitudinal analyses of prenatal and postnatal lead exposure and early cognitive development. N. Engl. J. Med. 316, 1037–1043. Bunn, T.L., Marsh, J.A., Dietert, R.R., 2000. Gender differences in developmental immunotoxicity to lead in the chicken: analysis following a single early low-level exposure in ovo. J. Toxicol. Environ. Health Pt A 61, 677–693. Burger, J., 1984. Pattern, mechanism, and adaptive significance of territoriality in herring gulls (Larus argentatus). Ornithol. Monogr. 34, 1–91. Burger, J., 1990. Behavioral effects of early postnatal lead exposure in herring gulls (Larus argentatus). Pharmacol. Biochem. Behav. 36, 7–13.
ARTICLE IN PRESS J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144 Burger, J., 1998. Effects of lead on sibling recognition in young herring gulls. Toxicol. Sci. 43, 155–160. Burger, J., Gochfeld, M., 1985. Early postnatal lead exposure: behavioral effects in common tern chicks Sterna hirundo. J. Toxicol. Environ. Health 29, 869–886. Burger, J., Gochfeld, M., 1988. Effects of lead on growth in young herring gulls (Larus argentatus). J. Toxicol. Environ. Health 25, 227–236. Burger, J., Gochfeld, M., 1990. Tissue levels of lead in experimentally exposed herring gull (Larus argentatus) chicks. J. Toxicol. Environ. Health 29, 219–233. Burger, J., Gochfeld, M., 1993a. Lead and behavioral development in young herring gulls: effects of timing of exposure on individual recognition. Fund. Appl. Toxicol. 21, 187–195. Burger, J., Gochfeld, M., 1993b. Lead and cadmium accumulation in eggs and fledgling seabirds in the New York Bight. Environ. Toxicol. Chem. 12, 261–267. Burger, J., Gochfeld, M., 1994. Behavioral impairments of leadinjected young herring gulls in nature. Fund. Appl. Toxicol. 23, 553–561. Burger, J., Gochfeld, M., 1995a. Effects of varying temporal exposure to lead on behavioral development in herring gull (Larus argentatus) chicks. Pharmacol. Biochem. Behav. 52, 601–608. Burger, J., Gochfeld, M., 1995b. Behavior effects of lead exposure on different days for gull (Larus argentatus) chicks. Pharmacol. Biochem. Behav. 50, 97–105. Burger, J., Gochfeld, M., 1996. Lead and behavioral development: parental compensation for behaviorally impaired chicks. Pharmacol. Biochem. Behav. 55, 339–349. Burger, J., Gochfeld, M., 1997. Lead and neurobehavioral development in gulls: a model for understanding effects in the laboratory and the field. Neurotoxicol. 18, 279–287. Burger, J., Gochfeld, M., 2000. Effects of lead on larids: a review of laboratory and field studies. J. Toxicol Environ. Health. Pt. B 3, 59–78. Burger, J., Gochfeld, M., 2003a. Spatial and temporal patterns in metal levels in eggs of common terns (Sterna hirundo) in New Jersey. Science Total Environ., in press. Burger, J., Gochfeld, M., 2003b. Lead in young herring gulls: paradoxical effects of exercise on tissue concentration. J. Toxicol. Environ. Health Pt. A 66, 181–193. Burger, J., Shukla, T., Benson, T., Gochfeld, M., 1994. Lead levels in exposed herring gulls: differences in the field and laboratory. In: Johnson, B.L., Xintaras, C., Andrews Jr, J.S. (Eds.), Hazardous Waste: Impacts on Human and Ecological Health. Princeton Scientific Publ. Co, Princeton, NJ, pp. 115–123. Burright, R.G., Engellenner, W.J., Donovick, P.J., 1983. Lead exposure and agonistic behavior of adult mice of two ages. Physiol. Behav. 30, 285–288. Bushnell, P.J., Bowman, R.E., 1979a. Persistence of impaired reversal learning in young monkeys exposed to low levels of dietary lead. J. Toxicol. Environ. Health 5, 1015–1023. Bushnell, P.J., Bowman, R.E., 1979b. Effects of chronic lead ingestion on social development in infant rhesus monkeys. Neurobehav. Toxicol. 1, 207–219. Cory-Slechta, D.A., 1990. Exposure duration modifies the effect of low-level lead on fixed interval performance. Neurotoxicol. 11, 427–442. Cory-Slechta, D.A., Schaumburg, H.H., 2000. Lead, inorganic. In: Spencer, P.S., Schaumburg, H.H., Ludolph, A.C. (Eds.), Experimental and Clinical Neurotoxicology, 2nd Edn. Oxford University Press, New York, pp. 708–720. Cory-Slechta, D.A., Weiss, B., Cox, C., 1983. Delayed behavioral toxicity of lead with increasing exposure concentration. Toxicol. Appl. Pharmacol. 71, 342–352.
143
Cutler, M.G., 1977. Effects of exposure to lead on social behaviour in the laboratory mouse. Psychopharmacology 52, 279–282. Dey, M., Burger, J., Gochfeld, M., Reuhl, K.R., 2000. Developmental lead exposure disturbs expression of synaptic neural cell adhesion molecules in herring gull brains. Toxicology 146, 137–147. Dietrich, K.M., Ris, M.D., Succop, P.A., Berger, O.G., Bornschein, R.L., 2001. Early exposure to lead and juvenile delinquency. Neurotoxicol. Teratol. 23, 511–518. Dietz, D.D., McMillan, D.E., Mushak, P., 1979. Effects of chronic lead administration on acquisition and performance of serial position sequences by pigeons. Toxicol. Appl. Pharmacol. 47, 377–384. Djawdan, M., Garland Jr, T., 1988. Maximal running speed of bipedal and quadrupedal rodents. J. Mamm. 69, 765–772. Dohm, M.R., Hayes, J.P., Garland Jr, T., 1996. Quantitative genetics of sprint running speed and swimming endurance in laboratory house mice (Mus domesticus). Evolution 50, 1688–1701. Dolinsky, Z.S., Burright, R.G., Donovick, P.J., 1983. Behavioral changes in mice following lead administration during several stages of development. Physiol. Behav. 30, 583–589. Donald, J.M., Cutler, M.G., Moore, M.R., Bradley, M., 1981. Development and social behavior in mice after prenatal and postnatal administration of low levels of lead acetate. Neuropharmacology 20, 1097–1104. Eisler, R., 1988. Lead Hazards to Fish, Wildlife, and Invertebrates: a Synoptic Review. US Fish and Wildlife Service, Washington, DC. Hamilton, J.D., O’Flaherty, E.J., 1995. Influence of lead on mineralization during bone growth. Fund. Appl. Toxicol. 26, 265–271. Lasky, R.E., Luck, M.L., Torre III, P., Laughlin, N., 2001. The effects of early lead exposure on auditory function in rhesus monkeys. Neurotoxicol. Teratol. 23, 638–649. Laughlin, N.K., Bushnell, P.J., Bowman, R.E., 1991. Lead exposure and diet: differential effects on social development in the rhesus monkey. Neurotoxicol. Teratol. 13, 429–440. Lidsky, T.I., Schneider, J.S., 2003. Lead neurotoxicity in children: basic mechanisms and clinical correlates. Brain 126 (Pwt 1), 5–19. Lilienthal, H., Winneke, G., 1996. Lead effects on the brain stem auditory evoked potential in monkeys during and after the treatment phase. Neurotoxicol. Teratol. 18, 17–32. Lippman, M., 1990. Lead and human health: background and recent findings. Environ. Res. 51, 1–24. Morgan, R.E., Garavan, H., Smith, E.G., Driscoll, L.L., Levitsky, D.A., Strupp, B.J., 2001. Early lead exposure produces lasting changes in sustained attention, response initiation, and reactivity to errors. Neurotoxicol. Teratol. 23, 519–531. Mykkanen, H.M., Wasserman, R.H., 1981. Gastrointestinal absorption of lead (203Pb) in chicks: influence of lead, calcium, and age. J. Nutr. 111, 1757–1765. Needleman, H., 1995. Environmental lead and children’s intelligence studies included in the meta–analysis are not representative. Br. Med. J. 310, 1408. Needleman, H.L., Schell, A., Bellinger, D., Leviton, A., Allred, E.N., 1990. The long-term effects of exposure to lead in childhood, An 11 year follow-up report. N. Engl. J. Med. 322, 83–88. Newland, M.C., Yezhou, S., Logdberg, B., Berlin, M., 1996. In utero lead exposure in squirrel monkeys: motor effects seen with schedule-controlled behavior. Neurotoxicol. Teratol. 18, 33–40. Payton, M., Riggs, K.M., Spiro III, A., Weiss, S.T., Hu, H., 1998. Relations of bone and blood lead to cognitive function: the VA normative aging study. Neurotoxicol. Teratol. 20, 18–27. Pocock, S.J, Smith, M., Baghurst, P., 1994. Environmental lead and children’s intelligence: a systematic review of the epidemiological evidence. Br. Med. J. 309, 1189–1197.
ARTICLE IN PRESS 144
J. Burger, M. Gochfeld / Ecotoxicology and Environmental Safety 57 (2004) 136–144
Rice, D.C., 1984. Behavioral deficit (delayed matching to sample) in monkeys exposed from birth to low levels of lead. Toxicol. Appl. Pharmacol. 75, 337–345. Rice, D.C., 1996. Behavioral effects of lead: commonalities between experimental and epidemiologic data. Environ. Health Perspect. 104, 337–351.
SAS (Statistical Analysis Systems), 1995. SAS Users’ Guide. Statistical Institute, Inc., Cary, NC. Siegel, S., 1956. Nonparametric Statistics. McGraw-Hill, NY. Silbergeld, E.K., 1990. Toward the twenty-first century: lessons from lead and lessons yet to learn. Environ. Health Perspect. 86, 191–196.