Long-term effects of microgravity on the swimming behaviour of young rats
1 Department of Physiology and Neuroscience, New York University School of Medicine, New York, NY 10016, USA
Abstract
The postnatal development of sensory systems has been shown in studies over the last four decades to be influenced by experience during critical periods of development. We report here that similar experience-dependent development can be observed in the swimming behaviour of young rats reared from postnatal day 14 (P14) to P30 in the reduced gravitational field of low earth orbit. Animals flown in space when placed in the water on the day of landing maintained their head and forelimbs in a balanced posture. However, until the animals began to swim, their hindquarters showed little lateral postural control resulting in rotation about the longitudinal axis (60°± 4 deg). Such results suggest an ‘unlinking’ of postural control of the forequarters from the hindquarters in the early hours after landing. Similar instability seen in animals age-matched to the day of launch (97 ± 7 deg) and in ground control animals (9 ± 3 deg) was corrected within one or two rotations, even in the absence of swimming. Animals flown in space began to swim sooner after being placed in the water, and the duration of swimming strokes was shorter than in control animals. Motion analysis revealed a difference in the swimming style on landing day. In flight animals, the knee joint was more flexed throughout the stroke, there was a narrower range of movement, and the linear velocity of the tip of the foot was faster throughout most of the stroke than in age-matched control animals. Thus, posture in the water as well as swimming speed and style were altered in the animals flown in space. Some of these characteristics persisted for as long as the animals were followed (30 days). These included the short pre-swimming interval and short stroke duration in flight animals. These findings clearly show that an altered gravitational field influences the postnatal development of motor function. The nature of the differences between animals reared in space for 16 days and those remaining on the ground reflects an adaptation of the flight animals to the microgravity environment. The data suggest that the most fundamental of these adaptations is a resetting of the basic motor rhythm to a higher frequency.
, http://www.100md.com
Introduction
It is well established that environmental stimuli play an important role in ensuring the proper development of sensory function. For example, Hubel and Weisel showed that visual deprivation during a period after birth can permanently affect the function and structure of the visual system from the retina to the occipital cortex (Wiesel & Hubel, 1963a,b; Hubel & Wiesel, 1970; Sosula & Glow, 1971). Their studies introduced the concepts of ‘sensitive’ and ‘critical’ periods of development (Hubel & Wiesel, 1970; Wiesel, 1982). The need for environmental stimuli during critical periods has also been clearly shown for the postnatal development of several other sensory systems (Meisami, 1978; Conlee & Parks, 1981; Gray et al. 1982; Fox, 1992; Schlaggar et al. 1993; O'Leary et al. 1994; Hillman et al. 1996; Penn & Shatz, 1999). Nevertheless, studies in the somatosensory system suggest that such critical periods of development may not be absolute for all systems, in that plasticity is seen in the adult human (Mogilner et al. 1993). However, study of the role of environmental stimuli in motor system development has proved to be more complex. One reason is that the environmental parameter most clearly associated with posture and movement, gravity, can be increased by centrifugation, but cannot be removed or reduced for periods longer than seconds.
, 百拇医药
Studies using a model of weightlessness in which the animal is lifted by its tail until its hindlimbs leave the ground (tail suspension) revealed marked effects on motor system development, depending on the age of the animal and suspension duration (Walton et al. 1992). Animals suspended between postnatal day 8 (P8) and P13 had significantly increased stroke duration and an altered swimming style compared to controls. Despite these marked differences, the animals recovered after the end of the suspension period. Animals suspended between P13 and P31 showed significant differences in walking style and vestibular reflexes that persisted into adulthood (Walton et al. 1992; Walton, 1998).
, 百拇医药
Since intermittent unloading of the hindlimbs had such profound effects on motor development, we proposed that similar, but more marked changes would be seen in animals that experienced complete unloading during the same period of development. While there have been many opportunities to study the effects of microgravity on adult rats, only recently have neonatal rats been included in space missions (Liskowsky et al. 1996; Homick et al. 1998; Maese & Ostrach, 2002). Litters of neonatal rats were included aboard Neurolab (STS-90), a 16-day space shuttle mission. Here we report on the effect of this experience on postflight swimming in animals that were in microgravity from P14 to P30. In a companion paper we report on surface righting (Walton et al. 2005). Swimming was studied for several reasons: (1) since the limbs do not bear the weight of the animal as during walking, the possible effects due to weakness in antigravity limb muscles following space flight are minimized (Adams et al. 2000a,b; Fitts et al. 2000); (2) swimming allows us to evaluate the effects of reduced gravity on the development of postural control as well as coordinated movement. This is because balance and postural control are needed for swimming which itself requires coordinated movements; (3) although rats are able to swim at birth, adult swimming characteristics do not develop fully until the third postnatal week (Beckoff & Trainer, 1979), and may be influenced by a change in the gravitational field. Analysing postflight swimming allows us to test our hypothesis that the postnatal development of motor skills is influenced by environmental stimuli such as the gravitational field.
, 百拇医药
Methods
The experimental design compares motor function in young rats launched into low-earth orbit with age-matched control animals that remained on earth. The data set comprises measurements of pre-swimming and swimming in animals from Neurolab (STS-90), a 16-day space shuttle mission.
Animals
Sprague-Dawley dams and litters were obtained from Taconic Farms (Germantown, NY, USA). Litters of a dam and seven female neonates were delivered to Kennedy Space Center when the neonates were P7 (±12 h). Upon arrival at the life sciences support facility at the Kennedy Space Center, each dam and litter was individually housed in a standard vivarium cage. The animals were put on the same diet as during the subsequent space flight (food bars, Teklad Diet, American Institute of Baking). The specific pathogen-free status of each dam was verified.
, 百拇医药
Groups and housing. Four groups of animals were formed: (1) eight flight animals flown in space and housed in flight cages; (2) eight age-matched ground control animals housed in flight-like cages; (3) seven age-matched ground control animals housed in standard vivarium cages; and (4) eight animals that were studied at P14 and housed in standard vivarium cages. This last group was needed since animals in the other three groups could not be studied on the day of launch when they were P14. Data from these animals are included in Fig. 4. Animals were selected for these experiments as described in the accompanying paper (Walton et al. 2005).
, 百拇医药
A, rotation of the hindquarters about the horizontal axis was greatest in the P14 control animals. Rotation in the flight animals was closer to that of P14 than age-matched ground controls (P30). B, rotation about the vertical axis of the hindlimb closest to the water surface. Flight values were more similar to P14 than to age-matched ground control animals. C, rotation about the vertical axis of the hindlimb furthest from the water surface. There was no difference between flight and control animals. Data are mean ± S.E.M. *P < 0.05, P < 0.001, P < 0.0001; unpaired t test; n = 10 P14; n = 6 flight; n = 8–10 control animals. Measurements were made as shown in Fig. 1C and D.
, 百拇医药
Although the control animals remained at the Kennedy Space Center, the terms ‘launch’ and ‘landing’ are applied to all groups of animals. Animals were P14 on the day of launch and P30 on the day of landing. The mission launched after a 24 h delay during which the animals remained in the orbiter (flight animals) or life science support facility (all ground controls).
The experiments reported here were carried out at the Kennedy Space Center life sciences support facility. Local government authorities and NASA approved all experimental procedures. The animals were cared for according to NIH, NASA, and New York University School of Medicine guidelines. During the postflight testing period the animals ate normally and increased in body weight. The mean rate of weight gain over the first 30 days after landing was 4.0 g day–1, not significantly different from that of control animals. At the end of the experiment the animals were used for anatomical studies after humane killing (DeFelipe et al. 2002).
, http://www.100md.com
Data acquisition
Individual animals from each group were video taped on P14 or from the day of landing (return day 0, R0) until R30. Each animal was placed individually in a swimming lane formed by a Plexiglas partition in the front a 76 x 30 cm rectangular tank of water maintained at 32–35°C. The animal swam until it reached a platform at the other end of the tank, where it left the water. The animal was then taken from the platform and placed in the water at the opposite end of the tank again. Each time an animal was placed in the water was called a trial. Animals were able to complete from 8 to 12 trials during the two-minute period allotted for swimming on each test day. After swimming, the animals were towel-dried and returned to their cage where they groomed, ate normally and were quickly accepted by the other animals. These brief periods of swimming did not interfere with weight gain. Swimming was videotaped at 60 frames s–1 (fps). Not all of the animals were included in all the measurements each day because they were participating in other experiments (see Results).
, http://www.100md.com
Data analysis
Trials were selected for such analysis in which (1) the animal was not moving its limbs (floating) or maintained forward motion (swimming); (2) there was no interference with the side of the tank; and (3) each stroke could be clearly seen (swimming).
Frame-by-frame analysis. When the animals were first placed in the water on R0 frame-by-frame analysis was used to evaluate (a) their activity before they began to swim; (b) the movements made to initiate swimming; and (c) the time interval between their placement in the water and the first clear swimming stroke. Once the animals began to swim, frame-by-frame analysis was used to measure the stroke duration. Each stroke was divided into a power phase and a return phase. The power phase was defined as those frames during which the hindlimb moved back. The return phase was defined as those frames during which the hindlimb moved forward. The sum of these gave the number of frames in each stroke. This was converted to milliseconds. The temporal resolution is limited by the duration of one frame (16.67 ms at 60 fps).
, 百拇医药
Computer-based analysis. A computer-based motion analysis system (Peak Motus 5.0) was used to digitize the positions of the animals when they were floating and during swimming. All points to be digitized (Fig. 1), except the knee, were marked on the rats with a permanent marker. The points were entered into the computer manually.
A, drawing of the right side of a swimming rat showing the points and line segments used for postural and motion analysis. B, the relative joint angles (knee, ankle, foot) calculated from the lateral aspect of the animal. C, stick figure of a control animal as viewed from behind. The f–f line segments connecting the heads of the femurs, the absolute angle between the upper hindlimb (f–mg line segment) and the vertical axis (HL°) are shown. D, stick figure of the hindlimbs of a flight animal as viewed from behind. The absolute angle between the f–f line segment and the horizontal axis gives hindquarter rotation (HQ°). Note the rotation of the hindquarters and limbs about the vertical axis. Abbreviations: t, tip of the toes; mp, metatarsophalangeal joint; a, ankle; k, knee; f, head of the femur; it, ischial tuberosity; ic, and iliac crest; f°, foot angle.
, 百拇医药
Floating
The drawing in Fig. 1A illustrates the position of the digitized points with respect to the outline of the animal as viewed from the side. Seven points were used to define the position of each hindlimb: tip of the toes (t), metatarsophalangeal joint (mp), ankle (a), knee (k), head of the femur (f), ischial tuberosity (it), and iliac crest (ic). Line segments, foot (f deg), ankle (a deg), and knee (k deg) joints were defined as shown in Fig. 1B.
, 百拇医药
Five of the seven points listed above were digitized when the animal was viewed from behind as shown in Fig. 1C and D. Two line segments were added: f–mp (between the femur head and metatarsophalangeal joint) and f–f (between the right and left femurs heads) (Fig. 1C). The absolute angle between the f–mp segment and the vertical axis (with the head of the femur as the vertex) was used to indicate the position of each hindlimb (HL deg, Fig. 1C and D). The f–mp segment was chosen because this gives a HL deg 0 deg when the hindlimb is directly under the animal. Hindlimb angles >0 deg indicate rotation of the limb from the vertical. To obtain average absolute angles that were independent of the direction of rotation (right or left), the hindlimbs were distinguished from each other as being closer to (Fig. 1C, upper HL) or further from (Fig. 1D, lower HL) the water surface. The absolute angle between the f–f segment and the horizontal axis was used to indicate the position of the hindquarters (HQ, Fig. 1D). Hindquarter angles were measured counterclockwise with the femur head closest to the water surface as the vertex. HL angles were measured clockwise or counterclockwise to give values <180 deg. Relative angles (those measured between limb segments) were measured counterclockwise.
, http://www.100md.com
Swimming
Swimming trials on R0 and on R7 (two trials per animal per day) were digitized for five age- and cage-matched ground control animals and five animals flown in space. The trials to digitize were chosen using the criteria listed above. Eight points were digitized from the lateral aspect of the animal (Fig. 1A). The ankle and knee angles as well as the horizontal and vertical linear velocity of the tip of the toe were calculated from these points. The joint angles were filtered (Butterworth, cut-off frequency, 10 Hz) to reduce the effects of camera interlace, pixel jitter, and human error. This filtering did not distort the waveform or reduce the amplitude of the angle values. This is shown in Fig. 2A where the ankle and knee angles are plotted as a function of time for one trial. The unfiltered (continuous lines) and filtered values (broken lines) are superimposed for the knee joint and for the ankle joint. The linear velocities for the same trial are shown in Fig. 2B. Because the amplitude of the velocity was very sensitive to filtering, these were not filtered.
, 百拇医药
A, knee and ankle joint angles during a swimming trial in a control animal on the day of landing. Unfiltered (continuous line) and filtered (broken line) traces are superimposed for the knee joint and for the ankle joint. The trial was divided into five complete strokes and the maximum (*) and minimum () values identified and marked automatically using Matlab. B, the linear velocity of the tip of the toe in the x-axis (broken line) and y-axis (continuous line) for the trial in panel A. C, superposition of ankle (continuous line) and knee (broken line) joint angles for individual strokes in panel A. D, strokes shown in panel C normalized to 100% time. E, superposition of one trace before () and after () normalization. (These traces are also included in panels C and D.)
, 百拇医药
Calculations
The maximum and minimum values for the ankle angle, knee angle, toe x-axis velocity, and toe y-axis velocity were calculated for each stroke using Matlab (The Math Works, Inc.). These are marked on the ankle angle (Fig. 2A) and x-axis velocity (Fig. 2B) traces. The peak ankle angle was the main marker used to automatically separate each trial into its component strokes because it was easily and reliably measured in all recordings. This process generated a set of strokes of different lengths as shown in Fig. 2C for the trial in Fig. 2A. The automatic division of a trial into individual strokes was verified visually. The duration of the individual strokes was found to be independent of the ankle or knee joint maximum value, the range of joint movement, and the toe linear velocity (correlation coefficient 0.001).
, 百拇医药
The individual strokes were next normalized with respect to time. The individual strokes contained between 8 and 20 points (mean, 13.4 ± 2.1 points), one point occurring every 1/60th of a second. Each stroke was resampled to comprise 100 points (Fig. 2D). The resample points were linearly interpolated from the original data. All strokes started and ended with the original data points. The original (14 from Fig. 2C) and normalized (100 from Fig. 2D) ankle angle values for one stroke are superimposed in Fig. 2E. Note that the normalization did not change the amplitude or waveform of the angle angles. The mean and S.E.M. were obtained for each parameter in flight and control animals on R0 and R7 (Figs 8, 10 and 11).
, 百拇医药
A, day of landing mean values for five flight (n = 40 strokes) and five control (n = 39 strokes) animals. Note the narrow range for the knee joint and that the loop is shifted to the left. There is a significant difference between the groups of animals throughout the stroke. B, similar to panel A for 7 days after landing. Significant differences remain between flight (n = 40 strokes) and control (n = 36 strokes) animals in the maximum ankle angle.
A, control on day of landing. Note long early return phase. B, flight on day of landing. Note difference from control in angle positions during phase transitions. C, control 7 days after landing. The major difference from the day of landing is the shift in the beginning of the late return phase. D, flight 7 days after landing. Note shift in late return phase from R0. Flight and control diagrams are more similar 7 days after landing than on landing day because of a change in the control animals. ER, early return stroke; LR, late return stroke; EP, early power stroke; LP, late power stroke.
, http://www.100md.com
Using only one camera introduces error since the three-dimensional movement of the limb is flattened into a two-dimensional plane. Segment lengths appear to wax and wane during each stroke as the limb moves toward and away from the camera and as the animal rotates about its longitudinal axis. (This is apparent in Fig. 7A and B) The largest change was in the length of the longest segment (ankle to knee). However, there were no significant differences between flight and control animals for any segment.
, http://www.100md.com
A and B, stick figures representing the position of the hindlimb and toe trajectory during 730 ms of a swimming trial in one control (A) and one flight (B) animal on the day of landing. Swimming is from left to right. Note that there are more strokes in the flight than control and that the circle transcribed by the toe is smaller in the flight animal. C, ankle, knee joint–joint plot for five consecutive strokes in the control animal. The points corresponding to the hindlimb positions during the first stroke in panel A are indicated (). D, similar to panel C for the flight animal. The points correspond to the third stroke in panel B.
, 百拇医药
Statistical methodology
For categorical data (pre-swimming activity), the mean for each animal was calculated. The mean of these means for flight and control animals was compared. In all other cases, results are expressed as the means of individual measurements. StatView and JMP (SAS Institute Inc.) were used for statistical analysis. The unpaired Student's t test was used to determine statistical significance. Fisher's r to z was used to obtain P values for correlation coefficients and Pearson's chi-square was used to evaluate distribution. Unless otherwise indicated, all values are mean ± S.E.M. Differences with P < 0.05 were considered to be significant.
, 百拇医药
Results
The behaviour of the flight animals in the water differed clearly from that of age-matched control animals. There were no differences, however, between the two groups housed in flight-like and standard vivarium cages. For this reason data from these animals were combined into one control group in Figs 3–6. Age-matched ground control animals housed in flight-like cages are included in Figs 7 and 8.
A, percentage of trials on day of landing in which flight (filled bars) and control (open bars) animals swam immediately, floated, or tried to leave the water. There were no significant differences between flight and control animals. B, similar to A for one day after landing. C, pooled data from R2, R3, R10 and R23 showing that the majority of animals swam immediately upon entering the water. Data are mean ± S.E.M. P < 0.001, P < 0.0001 unpaired t test; A and B n = 8 flight, 15 control animals; C n = 28 flight and 70 control animals.
, 百拇医药
Swimming has not been evaluated in neonatal or adult animals after space flight. For this reason, a detailed analysis was made of all aspects of the behaviour of the animals beginning when they were first placed in the water. The Results section is organized to answer questions comparing the animals flown in space with those that remained on the ground. (1) Are there differences in the behaviour when the animals are first placed in the water on the day of landing (2) Does the initiation of swimming differ in the two groups of animals (3) Does swimming in the two groups of animals differ with respect to its form or its speed (4) If differences are seen on the day of landing, do they persist
, 百拇医药
Motor control on the day of landing
Behaviour when entering the water. Observing the behaviour of animals when they were placed in the water within hours of landing provided insight into the influence of experience on the development of motor control, especially the control of posture. The behaviour of both groups of animals when they were placed in the water fell into three categories: (a) floating with little limb movement (b) immediate swimming (c) attempting to leave the water. This last behaviour was not considered to be swimming because the animals did not make forward progress, the hindlimbs did not make full swimming strokes, and the forelimb movements were more suited to attempting to climb a vertical surface than to swimming. Trials in which the entire animal was not in view, and those that did not fall into one of the three main categories were excluded from these analyses.
, http://www.100md.com
On the day of landing, the distribution of pre-swimming activity for flight animals was random while the control animals were twice as likely to try to leave the water as to swim or float (Pearson chi-square, P = 0.0150). Accordingly, the flight animals were more likely to swim immediately or float than were the control animals (Fig. 3A). There were no statistically significant differences between the two groups when each behaviour was compared, however.
The trials in which flight and control animals swam immediately increased significantly between landing day (Fig. 3A) and one day later (Fig. 3B) (P < 0.05). Pooled data from R2, R3, R10 and R23 showed no further significant changes from R1 values (Fig. 3C). Flight animals swam immediately in a larger percentage of trials than control animals. Control animals were more likely to float when they were placed in the water than were flight animals. Both flight and control animals tried to leave the water in just over 20% of the trials.
, 百拇医药
Floating posture. Both groups of animals looked the same when they were attempting to leave the water on the day of landing but not when they were floating. Although the posture of the head and forelimbs was similar, the hindquarters of flight animals were rotated about the longitudinal axis to a greater degree than those of control animals. Although the hindquarters of both groups of animals rotated when they were first placed in the water, the control animals quickly made a correction to reduce their rotation while the flight animals did not do this. As a result, the hindlimbs of control animals tended to be directly beneath their bodies (Fig. 1C), while those of the flight animals were more likely to be rotated laterally (Fig. 1D).
, 百拇医药
To quantify this observation, we identified the stable floating posture achieved soon after the animals were placed in the water. Since the animals moved as they floated, and floated in more than one trial, we were able to view and digitize the posture of most of the animals from both the side and the back (Fig. 1). Three relative hindlimb joints (foot, ankle, and knee) were calculated from the points digitized from the side view (Fig. 1B). There were no differences in these joint angles between the day of launch (P14) and the day of landing (P30, flight and control) as shown in Table 1.
, 百拇医药
The angles of the hindquarters and hindlimbs were calculated from the points digitized from the back view (Fig. 1C and D). As shown in Fig. 4A, the absolute angle of the hindquarters with respect to the horizontal was clearly greatest before launch (P14), and smallest in the P30 ground control animals. Values for the flight animals at landing (P30) were closer to launch (P14) values than to P30 control animals. There were significant differences among all three groups of animals as shown in Fig. 4A. The range also differed; being 45 deg to 122 deg at P14, 46 deg to 73 deg in P30 flight, and 0 deg to 22 deg in P30 ground control animals.
, 百拇医药
A similar pattern was seen in the position of the hindlimbs as shown in Fig. 4B and C. For the hindlimb closest to the surface (Fig. 4B), significant differences were seen between the P14 and P30 control and between the P30 flight and P30 control animals. For the lower hindlimb, there were significant differences between P14 and P30 flight and between P14 and P30 control animals. Rotation of the hindlimbs was secondary to hindquarter rotation. This is indicated by a high correlation of these two parameters. The correlation coefficients for the hindquarters–upper hindlimb pair, and hindquarters–lower hindlimb pair were both > 0.8 (P < 0.0001). The rotations of the two limbs were also highly correlated (>0.7, P < 0.0001).
, 百拇医药
In answer to the first question there were clear differences in the behaviour of flight and control animals when they were first placed in the water. Two differences stand out: (1) control animals tried to leave the water more than the flight animals and (2) the hindquarter and hindlimb position appeared to be independent of head and forelimb position in flight but not control animals.
Swimming initiation
Swimming in flight and control animals on the day of landing was characterized by hindlimb-only propulsion as is typical of adult animals (Gruner & Altman, 1980). Forelimb movement was also present in some strokes as is seen in younger animals (Beckoff & Trainer, 1979), but this was not typical. It appeared during the postflight videotaping that swimming was essentially the same in both groups of animals. However, differences were seen upon careful analysis of the stroke duration and kinematics of hindlimb movement.
, 百拇医药
Initial swimming movements. On the day of landing, every flight animal floated before making any movement the first time it was placed in the water. Since floating was the only activity in all of the flight animals before they began to swim, we were able to analyse the initial swimming movement. Swimming was initiated with hindquarter rotation and subsequent correction of the hindlimb displacement. In six of the flight animals this was followed by a short kick brought about by extension of one hindlimb. The other two flight animals extended both hindlimbs together. Swimming was initiated during the first (three animals) or second (five animals) trial.
, http://www.100md.com
Of the control animals, six floated when they were placed in the water on the first trial (eight attempted to leave the water and one swam immediately). Coordinated swimming began during the first (nine animals) or second (six animals) trial. The transition from floating to swimming followed a similar pattern to that seen in the flight animals. This was the case even if the animal did not begin the first or second trial by floating. The first swimming movement was a short kick brought about by extension of one hindlimb in ten control animals. (One control animal began with a forward hindlimb movement while two began swimming with simultaneous extension of both hindlimbs.)
, 百拇医药
Delay before swimming. We analysed how long it took the animals to begin swimming after they were placed in the water. The flight animals () began to swim sooner than control () animals as shown in Fig. 5 where the delay before swimming is plotted as a function of days after landing. This delay decreased in both groups of animals after the day of landing, but flight animals continued to begin to swim sooner than the control animals throughout the test period. When the trials in which the animals swam immediately are included (delay = 0), both plots in Fig. 5 are shifted downward preserving the differences.
, 百拇医药
This interval was greatest on the day of landing (day 0) in both groups of animals. The flight animals () began swimming before the control animals () on every test day. Data are mean ± S.E.M., *P < 0.05, P < 0.001, P < 0.0001; unpaired t test, n = 104–165 in control, n = 57–87 in flight animals.
In answer to the second question, both groups of animals initiate swimming with the same movements, but the flight animals do this sooner after being placed in the water than the control animals.
, 百拇医药
Swimming stroke duration, style, and speed
Stroke duration. A swimming stroke comprises two phases: the power phase that propels the animal through the water, and the return phase which brings the hindlimb forward in preparation for another power phase. The durations of both phases were measured and summed to give the stroke duration. Strokes were included for analysis only if the difference between the power and return phases was ± 1 frame (±16 ms). This eliminated irregular swimming and ‘coasting’ where, after a power phase, the animal held its limbs still before beginning the return phase. Coasting was seen in 10% of the strokes in flight, in 5% of the strokes in control, and in 2% of the strokes in the P14 animals.
, http://www.100md.com
The mean stroke cycle duration in P14, control, and flight animals is plotted as a function of time in Fig. 6. (S.E.M. values are smaller than the circles in the plot.). As shown in the figure, on landing day (day 0) the mean stroke duration in both flight () and control () animals was significantly shorter than on P14 (P < 0.0001). The durations of both the power and return phases were significantly shorter in the flight animals on every test day than the age-matched ground control animals. Thus, the flight animals' movements during both phases contributed to the short stroke duration. Since plots of the power stroke and return stroke duration had the same shape and relationship between flight and control animals as the total stroke duration plot (Fig. 6), they are not shown.
, http://www.100md.com
Note that on the day of landing (day 0), stroke duration of both flight () and control () animals were shorter than on the day of launch ( ). Throughout the postflight period the stroke duration was significantly shorter in the flight animals. Data are mean, ± S.E.M., *P < 0.05, P < 0.001, P < 0.0001; unpaired t test. Number of observations n = 146–443 except flight R23 = 76. Error bars are smaller than symbols.
The persistence of short stroke duration for 30 days prompted us to examine the swimming stroke in more detail on R0 and R7 in five flight and five control animals housed in flight-like cages. Return day 7 was chosen for the high quality of the data and because there were no significant long-term differences in flight animals after that day.
, http://www.100md.com
Trajectories and joint angles. The position of the hindlimb in an individual control and an individual flight animal swimming from left to right on the day of landing are represented as stick figures in Fig. 7A and B, respectively. The trajectory of the toe is also shown. Examination of the stick figures reveals that the toe trajectory was smaller and less circular in the flight (Fig. 7B) than in the control (Fig. 7A) animal. To investigate this observation and the movement of the hindlimb more closely, we measured the ankle and knee joint angles. These angles were chosen because they have the largest range of movement during swimming.
, http://www.100md.com
Angle–angle diagrams for these individual animals are shown in Fig. 7C and D where the ankle joint angle is plotted as a function of the knee joint angle. These diagrams illustrate the range of movement of each joint as well as how the joints work together as the hindlimb moves through the water. The ankle-knee angle-angle plot for six consecutive strokes in the control animal is shown in Fig. 7C (three of these strokes are shown in panel Fig. 7A). The numbers on the thick line in the plot correspond to limb positions marked in Fig. 7A. The power phase begins as the knee and ankle joints both extended to bring the limb down and back (positions 2–5). In this stroke, maximum knee extension was reached at position 4 and maximum ankle extension at position 5. Knee flexion then brings the limb up as it continues to move back through positions 5 and 6. Flexion of the ankle and knee together bring the limb up and forward from position 7–13 during the return phase. Finally, the knee extends, bringing the limb past position 13, to complete the return phase.
, http://www.100md.com
A similar ankle–knee angle–angle plot for five strokes in a flight animal is shown in Fig. 7D. (Three of these strokes are shown in Fig. 7B) The numbers correspond to those for the third stroke in Fig. 7B. This stroke was chosen for comparison with the first stroke in the control animal because it has a similar length and toe trajectory. Visual comparison of the plots from the two animals on R0 (Fig. 7C and D) reveals that the flight plot is shifted to the left.
, 百拇医药 To find if the differences noted between these two animals was typical, mean values were calculated. Superposition of mean angle–angle plots of flight and control animals on R0 (Fig. 8A) revealed clear differences in the form of the hindlimb movement that are similar to those seen for the individual animals in Fig. 7C and D. Throughout the stroke, the loop for the flight animals (solid line) was shifted to the left toward knee flexion. The narrow loop reflects the small range of mean knee movement compared with the control (broken lines) animals. The narrow range of mean knee movement is consistent with the short stroke duration (Fig. 5).
, 百拇医药
The opposite is the case in for the ankle. Here, flight animals have a greater mean range of movement than controls with no clear shift along the joint angle axis. Note that in addition to the difference in the absolute values for the angles, there is a difference in the shapes of the two ankle–knee angle–angle plots (Fig. 8A), reflecting a difference in the relationship between the two joints.
Over the first week after landing the joint–joint relationship changed in both groups of animals (Fig. 8B) but this was significant only for the flight animals. There was a rightward shift along the mean knee joint axis in the flight animal plot (solid line, Fig. 8B). This was due to increases in both the minimum (P < 0.01) and maximum (P < 0.0001) joint values (see Table 2). There was no change in the range of mean knee movement. There was no shift along the ankle axis since the minimum (P < 0.0001) and maximum (P < 0.0001) ankle angles both decreased between R0 and R7. This resulted in a decreased range of ankle movement on R7.
, 百拇医药
Values for ankle and knee minimum (most flexed angle) and maximum (most extended angle) and range of movement are summarized in Table 2. Significant differences between flight and control animals are marked (*). The R7 value for the maximum ankle angle in flight animals reached the R0 control value. However, a significant difference remained between flight and control animals on R7 because the control value for this parameter decreased between R0 and R7.
The narrow range of knee movement and small toe trajectory seen in the flight animals on R0 is consistent with short stroke duration. However, these differences in swimming style were much less pronounced on R7 (Fig. 8B) while the short stroke duration persisted (Fig. 5). This finding prompted us to search for another variable that might be linked to the persistent short stroke duration. Toward this end, we analysed the velocity of hindlimb movement since an increased velocity could contribute to short stroke duration.
, http://www.100md.com
Linear velocity. To examine directly the speed of hindlimb movement, the linear velocity of the digitized points was calculated for the x- and y-axes. The speed of the movement of the tip of the toe was chosen for analysis since this gave the best indication of effective hindlimb movement. The first stroke in Fig. 7A A and the third stroke in Fig. 7B are shown in Fig. 9A and B, respectively. The linear velocity of the toe movement for the trial containing the strokes in Fig. 9A and B are shown in the y-axis–x-axis velocity–velocity plots in Fig. 9C and D, respectively. (The thick lines correspond to the stroke shown in Fig. 9A and B)
, http://www.100md.com
A and B, stick figure showing the position of a hindlimb and toe trajectory for the control (A) and flight (B) animal (same stroke as in Fig. 7). C, y-axis–x-axis velocity–velocity plot for the tip of the toe during five consecutive strokes in the control animal. The velocity of the toe at each frame in panel A is marked as the hindlimb moves from the early power (EP) through the late power (LP), early return (ER), and late return (LR) phase of the stroke (see text). D, similar to panel C for the flight animal.
, 百拇医药
By convention, upward and forward velocity is positive, downward and backward velocity is negative. Thus, during a swimming stroke the y-axis–x-axis velocity–velocity curve crosses through zero four times. Characterizing the power phase by movement of the foot back, and the return phase as movement forward, the swimming stroke can be divided into four phases beginning with the return phase (as marked in Fig. 9C): (1) early return phase (ER) as the foot moves forward and up; (2) late return phase (LR) as the foot moves forward and down; (3) early power phase (EP) as the foot moves back and down and; (4) late power phase (LP) as the foot moves back and up. In the control example stroke (Fig. 9C), the return phase began just after point 6 as the limb moved forward. During this phase the limb moved up (points 7–13) then down (points 1–2). (Because there was little change in the linear velocity in either the x- or y-axis at the end of the early return phase, points 11–13 are clustered together.) The EP phase began from position 2 as the limb moved downward and back to position 4. The maximum downward velocity was reached midway during this movement (point 3) and the maximum velocity back occurred at the end (point 4). During the LP phase (points 5–7) the limb moved upward and back.
, 百拇医药
A similar pattern was seen in the flight animal (Fig. 9B and D), but the power phase had fewer frames (the limb moved faster). The early power phase began between points 1 and 2 and continued to point 3. At point 3, the foot had attained maximum velocity back. This velocity was maintained as the foot moved back and up to point 4. The maximum upward velocity was reached near the end of the late power phase (point 5). During the early return phase (point 5–7) the foot moved forward and up. In the late return phase (points 8–12), the foot is brought forward and down in preparation for the next power phase. The mean values for the maximum velocity in all four directions are given in Table 3.
, http://www.100md.com
The mean y-axis–x-axis velocity–velocity plots for five flight (continuous line) and five control (broken line) animals on R0 are superimposed in Fig. 10A. The fastest absolute movements occurred in the horizontal axis near the midpoints of the power and return phases. The foot of the flight animals moved faster at both of these points (Table 3). Mid-power phase corresponds to when the most propulsion is generated (maximum x-axis velocity). This increased speed could contribute to the short stroke duration in the flight animals. The mean maximum movement down was faster in the control animals (Table 3). As can be seen in Fig. 10A, this is associated with a difference in the shape of the y-axis–x-axis velocity–velocity plot. In the control animals the foot accelerated down during the transition from the return phase to the power phase (Fig. 10A, LR–EP). In contrast, the flight animals tended to maintain a constant downward velocity during this movement. There was no difference in the velocity of the movement upward.
, http://www.100md.com
A, on the day of landing there was a significant difference in toe velocity in the flight and control animals except in the transition from the LP to the ER phase. B, similar to panel A for 7 days after landing. Significant differences remain in the maximum forward LR and downward EP velocity. Number of strokes as in Fig. 8
Since difference in stroke duration persisted after R0, the velocity of the tip of the toe was calculated on R7 (Table 3). Between R0 and R7 the speed of movement of the foot downward (P < 0.0001) and forward (control P < 0.0001; flight P < 0.01) increased in both groups of animals. In the flight animals, upward movement (P < 0.0001) also increased. The increased speed was greater in the control than in the flight animals, but differences remained on R7 (Fig. 10B). Control animals moved their limbs faster than flight animals downward at the beginning of the stroke and flight animals moved faster at the mid-point of the return stroke (Table 3).
, 百拇医药
Swimming phases
Values for the ankle and knee angles (Figs 7 and 8) provide information about the position of these joints, while the linear velocity of the toe (Figs 9 and 10) provides information about movement direction and speed. Combing these two data sets allowed us to link knee and ankle joint position to limb movement in a unique way. To accomplish this, the points at which the mean toe velocity crossed zero were identified and marked on the mean angle–angle diagrams (Fig. 11). This allowed us to divide the angle–angle plots into the four phases of limb movement identified by the velocity data.
, http://www.100md.com
The combination of these data sets demonstrates that the phases of the swimming stroke are associated with different mean joint angle values in control (Fig. 11A and C) and flight (Fig. 11B and D) animals. Beginning with the return phase (ER), both joints flexed as the hindlimb began to move forward (it was already moving up). In the control animals (Fig. 11A), the ankle joint flexed throughout this phase and the knee began to extend before the next phase (LR) began. In contrast, this phase was cut short in the flight animals and both joints flexed (Fig. 11B).
, http://www.100md.com
The late return phase showed the most striking difference in joint position between the two groups of animals. In control animals, both joints extended throughout this phase as the limb moved forward and down. In flight animals the ankle flexed rather than extended, with little change in the knee angle, and both joints extended only at the end of the movement. This more flexed position of the ankle and knee reduced the forward toe trajectory since the toe was not brought as far forward as in the control animals before beginning to move back to begin the power stroke. (Compare Fig. 11A and B).
, 百拇医药
Propulsion for forward movement occurs during the power phase. In the control animals (Fig. 11B) the ankle extends throughout this movement (EP) but the knee begins to flex and flexes throughout the late power phase (LP). In flight animals (Fig. 11B), both joints extend throughout the early power phase (EP). The knee begins to flex only in mid–late power phase. The values at the beginning of each phase and the net movement during each phase are given in Table 4. Note that at each transition in movement direction the ankle and knee joint angles were significantly different in the two groups of animals except for the early return phase. The change in joint angle during each phase was significantly different in all the phases.
, http://www.100md.com
Changes in both control and flight animals over the first week after landing worked together to reduce the differences that were seen on R0. The angle–angle loops for control animals on R7 (coloured) and R0 (open, with dots marking the phase transitions) are superimposed in Fig. 11C. Two changes may be noted; (1) there was a clockwise shift in the beginning of the early return (ER), late return (LR) and late power (LP) phases (Figs 11C and 2) maximum knee extension is less on R7. These changes all work to make the relationship between the two angles more similar to that of the flight animals on R0.
, 百拇医药
Similar superposition of the R0 and R7 angle–angle plots for flight animals reveal three differences between R0 and R7 (Fig. 11D): (1) there was a shift of the loop to the right; (2) the beginning of the late return (LR) and early power (EP) phases was shifted counterclockwise; (3) the beginning of the early return (ER) phase was shifted clockwise. The overall effect was to reduce the range of movement during the power phase of the stroke. Significant differences remained between flight and control animals on R7, however. This was most notable in the ankle during the return phases and in the knee in the early power phase (Table 4).
, http://www.100md.com
Discussion
These experiments show that exposure of young animals (P14–P30) to microgravity deeply influences motor function in a protracted fashion. Indeed, on the day of landing, the motor behaviour of flight animals in the water was clearly different from that of ground control animals. Some behaviours seen on the day of landing, such as an apparent rostro-caudal ‘unlinking’ of postural control, quickly reverted to control values. However, behaviours that fall within the time domain (pre-swimming interval, stroke duration, and stroke kinematics) were not transient, but persisted and were still present in adult animals 30 days after landing. Both the transitory and persistent differences between flight and control animals are clearly related to their experience between P14 and P30 (in flight).
, 百拇医药
It can be argued that motor system development of animals on the ground proceeds to optimize performance in 1 g, while animals living in a microgravity environment may develop different motor control solutions to achieve the most efficient motor execution, i.e. that requiring the least time and energy to achieve a goal. This is consistent with the idea that development proceeds to meet the needs of the animal in terms of survival, safety, motivation, social, and other expectations. While these basic needs are similar in any gravitational environment, the methods used to achieve them will differ, especially when locomotion is involved.
, http://www.100md.com
Both internal (e.g. bone and muscle characteristics) and external (e.g. gravitational field and temperature) parameters will influence adaptation of an animal to its environment, and space flight introduces others such as radiation (see Walton et al. 2005). While the development of all systems is affected to some extent by a 16-day space flight and must be considered in interpretation of these data, our interpretation of these findings places emphasis on nervous system development.
, http://www.100md.com
Adaptation of the motor system to microgravity
Initial movements. Adaptation of the nervous system of flight animals to microgravity is reflected in their behaviour when they were placed in the water. Thus, while control animals tried to leave the water on the majority of trials on R0, the flight animals did not show a clear preference between this and floating or swimming (Fig. 3). This suggests that the buoyancy afforded by the water was closer to the environmental parameters of flight than to those of ground control animals. Thus, the flight animals may have been less fearful or less stressed when placed in the water. Another factor that may influence an animal's behaviour in the water is its own buoyancy. This is provided in part by air bubbles trapped in the animal's fur, and cleaner fur accumulates more bubbles. Because of cage movement during re-entry and landing, control animals tended to be cleaner than the flight animals on R0. This is probably not a major factor since control animals were less likely to float than the flight animals (Fig. 3).
, 百拇医药
After R0, swimming was the preferred initial activity in both groups of animals (Fig. 3B and C), however, swimming occurred in a larger percentage of trials in the flight animals. This was due to the persistence of pre-swimming floating in control animals after R0 (Fig. 3B). The dominance of swimming is not surprising since this activity helps the animals to achieve their goal – to leave the water (as indicated by their preference for swimming towards the platform and their leaving the water every time they reached the platform).
, http://www.100md.com
Throughout the study, flight animals began to swim sooner after entering the water than control animals (Fig. 4). This was not accounted for by a difference in the pre-swimming behaviour of the two groups animals. Flight animals just spent less time doing each activity before beginning to swim. The behaviour of the flight animals is consistent with the interpretation offered above, namely, that the buoyancy of water provides a familiar environment and they make appropriate movements. However satisfactory this explanation may be for R0, it does not predict that the difference would persist for 30 days since by then both groups of animals had almost daily swimming experience. The short interval may reflect a more general property of flight animals – they move through the water faster than control animals.
, http://www.100md.com
It is difficult to know if cognitive factors may have contributed to the differences in the behaviour of these two animal groups. However, a study of these same rats showed no differences in their ability to learn or remember spatial tasks (Morris water maze), in long-term potentiation, or hippocampal morphology (Temple et al. 2003).
Floating. When they were placed in the water for the first time in their experience, all of the flight animals floated quietly before making any movements. Since the otolithic system had experienced a reduced gravitational input for 16 days, such well-controlled floating was unexpected. Postural control of the head indicates that the peripheral vestibular apparatus (vestibular labyrinths, Scarpa's ganglia, vestibular nerves) and vestibulocolic reflexes were fully functional within hours of landing. This is consistent with the fact that development of this system was largely complete at launch (Dememes et al. 2001). Control of the forequarters further indicates that, among others, the vestibulospinal and corticospinal circuits were functional, at least to the level of the cervical spinal cord. In fact, in this respect, flight animals were similar to ground controls. This suggests that vestibular function, as reflected in the ability of the animals to maintain balance in the water, did not deteriorate in the reduced gravitational input of microgravity. It is also consistent with our finding that flight animals were able to right themselves from a supine to a prone position on the first trial on R0 (Walton et al. 2005).
, http://www.100md.com
While the floating posture of the head and forequarters was similar in flight and control animals, the position of the hindquarters differed (Fig. 4). This may reflect a lack of maturation of earth-based postural control in the flight animals. The control animals moved their hindlimbs under the body, resulting in a posture that was appropriate and necessary for quadruped stance as well as swimming. This correction did not occur in the flight animals until they began to swim.
, http://www.100md.com
This finding suggests that the uncorrected hindquarter rotation reflects experience-dependent development of midline control that is adaptive to the microgravity environment. Hindlimb rotation and the rostro-caudal ‘unlinking’ both reflect the characteristics of locomotion in microgravity where the forelimbs were often used as the principal form of propulsion. Since the hindlimbs did not necessarily participate in locomotion, their posture could be largely independent of the forelimbs and often floated away from the surface (DeFelipe et al. 2002; see Supplemental material). As a result, we proposed that positions in which the hindquarters are rotated with respect to the forequarters are included in the set of ‘normal’ postures for flight animals. Uncorrected hindquarter rotation while floating does not reflect a lack of postural control because the animals were stable in the water, and before making the first stroke every flight animal rotated its hindquarters to bring its hindlimbs into the position appropriate for swimming.
, 百拇医药
Swimming. When they were launched into space, the eyes of the flight animals were open, they had achieved quadruped stance, interlimb coordination, quadruped locomotion, and were able to move about freely. That these motor skills were modified as the nervous system adapted to microgravity over the following two weeks is reflected in the way the flight animals swam, as well as in their pre-swimming behaviour.
Stroke duration is a key parameter in swimming analysis because it reflects both the form (Figs 7 and 8) and velocity (Figs 9 and 10) of hindlimb movement. During normal terrestrial development this parameter changes over the first two weeks of life, decreasing from over 700 ms at birth to near 300 ms by P14 (Beckoff & Trainer, 1979). These experiments have shown a subsequent decrease in control animals from a mean of 284 ms P14 (Fig. 6, P14) to a mean of 217 ms in control and 205 ms in flight animals on P30 (Fig. 6, day 0). Although the decrease after P14 is small when compared to the normal range of development, it is nevertheless significant (P < 0.0001) as is the difference between control and flight animals on P30 (Fig. 6). The difference in stroke duration (Fig. 6) and pattern of limb movement (Fig. 11) between control and flight animals on R0 demonstrates for the first time that establishment of adult swimming parameters is influenced by environmental factors.
, 百拇医药
Comparison with hindlimb suspension
Comparison of swimming in flight animals and in those experiencing hindlimb suspension address the question of which mechanisms may underlie the findings reported here. The short stroke duration in flight animals is not due to the lack of hindlimb weight bearing in microgravity, per se. Indeed, hindlimbs do not bear the animal's weight under either condition, yet the effects on stroke duration are in opposite directions. Although in the tail suspension paradigm, the animals do bear weight for one hour in 24 (while they are feeding), it is unlikely that this is long enough to influence their adaptation to the suspension paradigm. Also, we have found increases in stroke duration after only 5 h of suspension with no weight bearing before testing (K.W., personal observation). The difference in stroke duration is directly related to limb position during the stroke, which itself is directly related to the different use of the hindlimbs during suspension and in microgravity.
, http://www.100md.com
Suspended animals show an ‘extensor bias’ and keep their hindlimbs extended behind them unless they are asleep. Similarly, they swim with their hindlimbs extended, resulting in a larger trajectory and longer stroke duration than control animals (Walton et al. 1992). In contrast, animals in microgravity show a flexor bias and use their limbs to pull themselves about the cage. Accordingly, they swim with their hindlimbs in a more flexed position resulting in a smaller trajectory and shorter stroke duration than control animals (Fig. 7, Table 2). The finding of such flexor bias in animals adapted to microgravity is not surprising. This is consistent with a ‘flexor bias’ that has been reported for quadrupedal locomotion in-flight or postflight in non-human primates (Recktenwald et al. 1999) and in humans (Bloomberg & Mulavara, 2003; McCall et al. 2003).
, 百拇医药
Thus, our results indicate that in the absence of a need to counter gravity during standing and for extensor activity to move the animal forward prior to foot lift (Gruner & Altman, 1980), the motor programme for locomotion is altered in flight animals. This is consistent with other studies of motor activity during and after space flight that have been interpreted in terms of changes in the brain's internal model of gravity upon which movement is based (see McIntyre et al. 2001; Courtine et al. 2002).
, 百拇医药
Role of the musculoskeletal system
It might be expected that changes in muscle properties may underlie the differences seen between flight and control animals (see Fitts et al. 2001). The changes in hindlimb muscle properties under conditions of tail suspension and microgravity have been well documented. The postnatal development of antigravity muscles is selectively suppressed in both tail-suspended (Edgerton & Roy, 1994; Riley et al. 1995) and flight animals (Adams et al. 2000a, b). Yet, the effects of tail suspension and space flight on swimming style as well as on stroke duration are in opposite directions as summarized above. Thus, similar changes in muscle properties are associated with opposite changes in swimming.
, 百拇医药
Critical period hypothesis
One question posed at the start of these experiments was whether postnatal activity-dependent modifications of sensory-motor function remain fixed or are subject to alteration throughout the life span. Our findings clearly show that an altered gravitational field influences the postnatal development of sensory–motor function. Some of these characteristics persisted for as long as the animals were followed (one month). These included the short pre-swimming interval and short stroke duration in flight animals. The nature and persistence of the changes suggests that they (1) are due to the particular experience of the animals during a critical period of development and (2) reflect a long lasting adaptation of the animals to the microgravity environment. The data suggest that the most fundamental of these adaptations is a resetting of the basic motor rhythm to a higher frequency.
, http://www.100md.com
In conclusion, then, the data presented in this and the accompanying paper (Walton et al. 2005) indicate a deep level of CNS adaptability. Thus, in the absence of gravitation one of the primordial sensory inputs required for the organization of up–down and left–right is missing. The CNS adapts just sufficiently to allow proper movement in space. This is done without modifying the load-unburdening systems such as the oculomotor and directional head orientation. Equally remarkable is the fact that if such changes occur at a time in development where connectivity is being optimized, the changes remain throughout the life span of the animal, suggesting the possibility of a motor critical period as seen in sensory systems.
, 百拇医药
Supplemental material
The online version of the companion paper (Walton et al. 2005) can be accessed at: DOI: 10.1113/jphysiol.2004.074385
http://j.p.physoc.org/cgi/content/full/jphysiol.2004.074385/DC1
and contains supplemental material consisting of a movie entitled: A neurolab experiment: The effect of gravity on postnatal motor system development.
This material can also be found as part of the full text HTML version available from http://www.blackwellsynergy.com
, 百拇医药
References
Adams GR, Haddad F, McCue SA, Bodell PW, Zeng M, Qin L, Qin AX & Baldwin KM (2000a). Effects of spaceflight and thyroid deficiency on rat hindlimb development. II. Expression of MHC isoforms. J Appl Physiol 88, 904–916.
Adams GR, McCue SA, Bodell PW, Zeng M & Baldwin KM (2000b). Effects of spaceflight and thyroid deficiency on hindlimb development. I. Muscle mass and IGF-I expression. J Appl Physiol 88, 894–903.
, 百拇医药
Beckoff A & Trainer W (1979). The development of interlimb co-ordination during swimming in postnatal rats. J Exp Biol 83, 1–11.
Bloomberg JJ & Mulavara AP (2003). Changes in walking strategies after spaceflight. IEEE Eng Medical Biol March/April, 58–62.
Conlee JW & Parks TN (1981). Age- and position-dependent effects of monaural acoustic deprivation in nucleus magnocellularis of the chicken. J Comp Neurol 202, 373–384.
, 百拇医药
Courtine G, Papaxanthis C & Pozzo T (2002). Prolonged exposure to microgravity modifies limb endpoint kinematics during the swing phase of human walking. Neurosci Lett 332, 70–74.
DeFelipe J, Arellano JI, Merchán-Perez A, González-Albo MC, Walton K & Llinás R (2002). Spaceflight induces changes in the synaptic circuitry of the postnatal developing neocortex. Cortex 12, 883–891.
Dememes D, Dechesne CJ, Venteo S, Gaven F & Raymond J (2001). Development of the rat efferent vestibular system on the ground and in microgravity. Brain Res Devel Brain Res 128, 35–44.
, 百拇医药
Edgerton VR & Roy RR (1994). Neuromuscular adaptation to actual and simulated weightlessness. Adv Space Biol Med 4, 33–67.
Fitts RH, Riley DR & Widrick JJ (2000). Physiology of a microgravity environment invited review: microgravity and skeletal muscle. J Appl Physiol 89, 823–839.
Fitts RH, Riley DR & Widrick JJ (2001). Functional and structural adaptations of skeletal muscle to microgravity. J Exp Biol 204, 3201–3208.
, http://www.100md.com
Fox K (1992). A critical period for experience-dependent synaptic plasticity in rat barrel cortex. J Neurosci 12, 1826–1838.
Gray L, Smith Z & Rubel EW (1982). Developmental and experimental changes in dendritic symmetry in n. laminaris of the chick. Brain Res 244, 360–364.
Gruner JA & Altman J (1980). Swimming in the rat: analysis of locomotor performance in comparison to stepping. Exp Brain Res 40, 374–382.
, http://www.100md.com
Hillman DE, Chen S, Bing R, Penniston JT & Llinás R (1996). Ultrastructural localization of the plasmalemmal calcium pump in cerebellar neurons. Neurosci 72, 315–324.
Homick JL, Delaney P & Rodda K (1998). Overview of the Neurolab Spacelab mission. Acta Astronaut 42, 69–87.
Hubel DH & Wiesel TN (1970). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol 206, 419–436.
, 百拇医药
Liskowsky DR, Frey MA, Sulzman FM, White RJ & Likowsky DR (1996). The Neurolab mission and biomedical engineering: a partnership for the future. BME 10, 11–25.
McCall GE, Goulet C, Boorman GI, Roy RR & Edgerton VR (2003). Flexor bias of joint position in humans during spaceflight. Exp Brain Res 152, 87–94.
McIntyre J, Zago M, Berthoz A & Lacquanti F (2001). Does the brain model Newton's Law Nature Neurosci 4, 693–694.
, 百拇医药
Maese C & Ostrach L (2002). Neurolab. Final report for the Ames. Research Center Payload NASA/TM-2002–211841, NASA, Moffett Field.
Meisami E (1978). Influence of early anosmia on the developing olfactory bulb. Prog Brain Res 48, 211–229.
Mogilner A, Grossman JA, Ribary U, Joliot M, Volkmann J, Rapaport D, Beasley RW & Llinás RR (1993). Somatosensory cortical plasticity in adult humans revealed by magnetoencephalography. Proc Natl Acad Sci U S A 90, 3593–3597.
, 百拇医药
O'Leary DD, Ruff NL & Dyck RH (1994). Development, critical period plasticity, and adult reorganizations of mammalian somatosensory systems. Curr Opin Neurobiol 4, 535–544.
Penn AA & Shatz CJ (1999). Brain waves and brain wiring: the role of endogenous and sensory-driven neural activity in development. Pediatr Res 45, 447–458.
Recktenwald MR, Hodgson JA, Roy RR, Riazanski S, McCall GE, Kozlovskaya I, Washburn DA, Fanton JW & Edgerton VR (1999). Effects of spaceflight on rhesus quadrupedal locomotion after return to 1G. J Neurophysiol 81, 2451–2463.
, 百拇医药
Riley DA, Thompson JL, Krippendorf BB & Slocum GR (1995). Review of spaceflight and hindlimb suspension unloading induced sarcomere damage and repair. Basic Appl Myol 5, 139–145.
Schlaggar BL, Fox K & O'Leary DD (1993). Postsynaptic control of plasticity in developing somatosensory cortex. Nature 364, 623–626.
Sosula L & Glow PH (1971). Increase in number of synapses in the inner plexiform layer of light deprived rat retinae: quantitative electron microscopy. J Comp Neurol 141, 427–451.
, 百拇医药
Temple MD, Denslow MJ, Kosik K & Steward O (2003). Neural development under conditions of spaceflight. In The Neurolab Spacelab Mission: Neuroscience Research in Space, ed. Buckey JC and Homick JL, pp. 161–168. NASA, Washington DC.
Walton K (1998). Postnatal development under conditions of simulated weightlessness and space flight. Brain Res Brain Res Rev 28, 25–34.
Walton KD, Harding S, Anschel D, Harris YT & Llinás R (2005). The effects of microgravity on the development of surface righting in rats. J Physiol 565, 593–608.
, 百拇医药
Walton KD, Lieberman D, Llinás A, Begin M & Llinás RR (1992). Identification of a critical period for motor development in neonatal rats. Neurosci 51, 763–767.
Wiesel TN (1982). Postnatal development of the visual cortex and the influence of the environment. Nature 299, 583–591.
Wiesel T & Hubel D (1963a). Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 26, 1003–1017.
Wiesel TN & Hubel DH (1963b). Effects of visual deprivation on morphology and physiology of cells in the cat's lateral geniculate cortex. J Neurophysiol 26, 978–993., 百拇医药(Kerry D. Walton, Louis Be)
Abstract
The postnatal development of sensory systems has been shown in studies over the last four decades to be influenced by experience during critical periods of development. We report here that similar experience-dependent development can be observed in the swimming behaviour of young rats reared from postnatal day 14 (P14) to P30 in the reduced gravitational field of low earth orbit. Animals flown in space when placed in the water on the day of landing maintained their head and forelimbs in a balanced posture. However, until the animals began to swim, their hindquarters showed little lateral postural control resulting in rotation about the longitudinal axis (60°± 4 deg). Such results suggest an ‘unlinking’ of postural control of the forequarters from the hindquarters in the early hours after landing. Similar instability seen in animals age-matched to the day of launch (97 ± 7 deg) and in ground control animals (9 ± 3 deg) was corrected within one or two rotations, even in the absence of swimming. Animals flown in space began to swim sooner after being placed in the water, and the duration of swimming strokes was shorter than in control animals. Motion analysis revealed a difference in the swimming style on landing day. In flight animals, the knee joint was more flexed throughout the stroke, there was a narrower range of movement, and the linear velocity of the tip of the foot was faster throughout most of the stroke than in age-matched control animals. Thus, posture in the water as well as swimming speed and style were altered in the animals flown in space. Some of these characteristics persisted for as long as the animals were followed (30 days). These included the short pre-swimming interval and short stroke duration in flight animals. These findings clearly show that an altered gravitational field influences the postnatal development of motor function. The nature of the differences between animals reared in space for 16 days and those remaining on the ground reflects an adaptation of the flight animals to the microgravity environment. The data suggest that the most fundamental of these adaptations is a resetting of the basic motor rhythm to a higher frequency.
, http://www.100md.com
Introduction
It is well established that environmental stimuli play an important role in ensuring the proper development of sensory function. For example, Hubel and Weisel showed that visual deprivation during a period after birth can permanently affect the function and structure of the visual system from the retina to the occipital cortex (Wiesel & Hubel, 1963a,b; Hubel & Wiesel, 1970; Sosula & Glow, 1971). Their studies introduced the concepts of ‘sensitive’ and ‘critical’ periods of development (Hubel & Wiesel, 1970; Wiesel, 1982). The need for environmental stimuli during critical periods has also been clearly shown for the postnatal development of several other sensory systems (Meisami, 1978; Conlee & Parks, 1981; Gray et al. 1982; Fox, 1992; Schlaggar et al. 1993; O'Leary et al. 1994; Hillman et al. 1996; Penn & Shatz, 1999). Nevertheless, studies in the somatosensory system suggest that such critical periods of development may not be absolute for all systems, in that plasticity is seen in the adult human (Mogilner et al. 1993). However, study of the role of environmental stimuli in motor system development has proved to be more complex. One reason is that the environmental parameter most clearly associated with posture and movement, gravity, can be increased by centrifugation, but cannot be removed or reduced for periods longer than seconds.
, 百拇医药
Studies using a model of weightlessness in which the animal is lifted by its tail until its hindlimbs leave the ground (tail suspension) revealed marked effects on motor system development, depending on the age of the animal and suspension duration (Walton et al. 1992). Animals suspended between postnatal day 8 (P8) and P13 had significantly increased stroke duration and an altered swimming style compared to controls. Despite these marked differences, the animals recovered after the end of the suspension period. Animals suspended between P13 and P31 showed significant differences in walking style and vestibular reflexes that persisted into adulthood (Walton et al. 1992; Walton, 1998).
, 百拇医药
Since intermittent unloading of the hindlimbs had such profound effects on motor development, we proposed that similar, but more marked changes would be seen in animals that experienced complete unloading during the same period of development. While there have been many opportunities to study the effects of microgravity on adult rats, only recently have neonatal rats been included in space missions (Liskowsky et al. 1996; Homick et al. 1998; Maese & Ostrach, 2002). Litters of neonatal rats were included aboard Neurolab (STS-90), a 16-day space shuttle mission. Here we report on the effect of this experience on postflight swimming in animals that were in microgravity from P14 to P30. In a companion paper we report on surface righting (Walton et al. 2005). Swimming was studied for several reasons: (1) since the limbs do not bear the weight of the animal as during walking, the possible effects due to weakness in antigravity limb muscles following space flight are minimized (Adams et al. 2000a,b; Fitts et al. 2000); (2) swimming allows us to evaluate the effects of reduced gravity on the development of postural control as well as coordinated movement. This is because balance and postural control are needed for swimming which itself requires coordinated movements; (3) although rats are able to swim at birth, adult swimming characteristics do not develop fully until the third postnatal week (Beckoff & Trainer, 1979), and may be influenced by a change in the gravitational field. Analysing postflight swimming allows us to test our hypothesis that the postnatal development of motor skills is influenced by environmental stimuli such as the gravitational field.
, 百拇医药
Methods
The experimental design compares motor function in young rats launched into low-earth orbit with age-matched control animals that remained on earth. The data set comprises measurements of pre-swimming and swimming in animals from Neurolab (STS-90), a 16-day space shuttle mission.
Animals
Sprague-Dawley dams and litters were obtained from Taconic Farms (Germantown, NY, USA). Litters of a dam and seven female neonates were delivered to Kennedy Space Center when the neonates were P7 (±12 h). Upon arrival at the life sciences support facility at the Kennedy Space Center, each dam and litter was individually housed in a standard vivarium cage. The animals were put on the same diet as during the subsequent space flight (food bars, Teklad Diet, American Institute of Baking). The specific pathogen-free status of each dam was verified.
, 百拇医药
Groups and housing. Four groups of animals were formed: (1) eight flight animals flown in space and housed in flight cages; (2) eight age-matched ground control animals housed in flight-like cages; (3) seven age-matched ground control animals housed in standard vivarium cages; and (4) eight animals that were studied at P14 and housed in standard vivarium cages. This last group was needed since animals in the other three groups could not be studied on the day of launch when they were P14. Data from these animals are included in Fig. 4. Animals were selected for these experiments as described in the accompanying paper (Walton et al. 2005).
, 百拇医药
A, rotation of the hindquarters about the horizontal axis was greatest in the P14 control animals. Rotation in the flight animals was closer to that of P14 than age-matched ground controls (P30). B, rotation about the vertical axis of the hindlimb closest to the water surface. Flight values were more similar to P14 than to age-matched ground control animals. C, rotation about the vertical axis of the hindlimb furthest from the water surface. There was no difference between flight and control animals. Data are mean ± S.E.M. *P < 0.05, P < 0.001, P < 0.0001; unpaired t test; n = 10 P14; n = 6 flight; n = 8–10 control animals. Measurements were made as shown in Fig. 1C and D.
, 百拇医药
Although the control animals remained at the Kennedy Space Center, the terms ‘launch’ and ‘landing’ are applied to all groups of animals. Animals were P14 on the day of launch and P30 on the day of landing. The mission launched after a 24 h delay during which the animals remained in the orbiter (flight animals) or life science support facility (all ground controls).
The experiments reported here were carried out at the Kennedy Space Center life sciences support facility. Local government authorities and NASA approved all experimental procedures. The animals were cared for according to NIH, NASA, and New York University School of Medicine guidelines. During the postflight testing period the animals ate normally and increased in body weight. The mean rate of weight gain over the first 30 days after landing was 4.0 g day–1, not significantly different from that of control animals. At the end of the experiment the animals were used for anatomical studies after humane killing (DeFelipe et al. 2002).
, http://www.100md.com
Data acquisition
Individual animals from each group were video taped on P14 or from the day of landing (return day 0, R0) until R30. Each animal was placed individually in a swimming lane formed by a Plexiglas partition in the front a 76 x 30 cm rectangular tank of water maintained at 32–35°C. The animal swam until it reached a platform at the other end of the tank, where it left the water. The animal was then taken from the platform and placed in the water at the opposite end of the tank again. Each time an animal was placed in the water was called a trial. Animals were able to complete from 8 to 12 trials during the two-minute period allotted for swimming on each test day. After swimming, the animals were towel-dried and returned to their cage where they groomed, ate normally and were quickly accepted by the other animals. These brief periods of swimming did not interfere with weight gain. Swimming was videotaped at 60 frames s–1 (fps). Not all of the animals were included in all the measurements each day because they were participating in other experiments (see Results).
, http://www.100md.com
Data analysis
Trials were selected for such analysis in which (1) the animal was not moving its limbs (floating) or maintained forward motion (swimming); (2) there was no interference with the side of the tank; and (3) each stroke could be clearly seen (swimming).
Frame-by-frame analysis. When the animals were first placed in the water on R0 frame-by-frame analysis was used to evaluate (a) their activity before they began to swim; (b) the movements made to initiate swimming; and (c) the time interval between their placement in the water and the first clear swimming stroke. Once the animals began to swim, frame-by-frame analysis was used to measure the stroke duration. Each stroke was divided into a power phase and a return phase. The power phase was defined as those frames during which the hindlimb moved back. The return phase was defined as those frames during which the hindlimb moved forward. The sum of these gave the number of frames in each stroke. This was converted to milliseconds. The temporal resolution is limited by the duration of one frame (16.67 ms at 60 fps).
, 百拇医药
Computer-based analysis. A computer-based motion analysis system (Peak Motus 5.0) was used to digitize the positions of the animals when they were floating and during swimming. All points to be digitized (Fig. 1), except the knee, were marked on the rats with a permanent marker. The points were entered into the computer manually.
A, drawing of the right side of a swimming rat showing the points and line segments used for postural and motion analysis. B, the relative joint angles (knee, ankle, foot) calculated from the lateral aspect of the animal. C, stick figure of a control animal as viewed from behind. The f–f line segments connecting the heads of the femurs, the absolute angle between the upper hindlimb (f–mg line segment) and the vertical axis (HL°) are shown. D, stick figure of the hindlimbs of a flight animal as viewed from behind. The absolute angle between the f–f line segment and the horizontal axis gives hindquarter rotation (HQ°). Note the rotation of the hindquarters and limbs about the vertical axis. Abbreviations: t, tip of the toes; mp, metatarsophalangeal joint; a, ankle; k, knee; f, head of the femur; it, ischial tuberosity; ic, and iliac crest; f°, foot angle.
, 百拇医药
Floating
The drawing in Fig. 1A illustrates the position of the digitized points with respect to the outline of the animal as viewed from the side. Seven points were used to define the position of each hindlimb: tip of the toes (t), metatarsophalangeal joint (mp), ankle (a), knee (k), head of the femur (f), ischial tuberosity (it), and iliac crest (ic). Line segments, foot (f deg), ankle (a deg), and knee (k deg) joints were defined as shown in Fig. 1B.
, 百拇医药
Five of the seven points listed above were digitized when the animal was viewed from behind as shown in Fig. 1C and D. Two line segments were added: f–mp (between the femur head and metatarsophalangeal joint) and f–f (between the right and left femurs heads) (Fig. 1C). The absolute angle between the f–mp segment and the vertical axis (with the head of the femur as the vertex) was used to indicate the position of each hindlimb (HL deg, Fig. 1C and D). The f–mp segment was chosen because this gives a HL deg 0 deg when the hindlimb is directly under the animal. Hindlimb angles >0 deg indicate rotation of the limb from the vertical. To obtain average absolute angles that were independent of the direction of rotation (right or left), the hindlimbs were distinguished from each other as being closer to (Fig. 1C, upper HL) or further from (Fig. 1D, lower HL) the water surface. The absolute angle between the f–f segment and the horizontal axis was used to indicate the position of the hindquarters (HQ, Fig. 1D). Hindquarter angles were measured counterclockwise with the femur head closest to the water surface as the vertex. HL angles were measured clockwise or counterclockwise to give values <180 deg. Relative angles (those measured between limb segments) were measured counterclockwise.
, http://www.100md.com
Swimming
Swimming trials on R0 and on R7 (two trials per animal per day) were digitized for five age- and cage-matched ground control animals and five animals flown in space. The trials to digitize were chosen using the criteria listed above. Eight points were digitized from the lateral aspect of the animal (Fig. 1A). The ankle and knee angles as well as the horizontal and vertical linear velocity of the tip of the toe were calculated from these points. The joint angles were filtered (Butterworth, cut-off frequency, 10 Hz) to reduce the effects of camera interlace, pixel jitter, and human error. This filtering did not distort the waveform or reduce the amplitude of the angle values. This is shown in Fig. 2A where the ankle and knee angles are plotted as a function of time for one trial. The unfiltered (continuous lines) and filtered values (broken lines) are superimposed for the knee joint and for the ankle joint. The linear velocities for the same trial are shown in Fig. 2B. Because the amplitude of the velocity was very sensitive to filtering, these were not filtered.
, 百拇医药
A, knee and ankle joint angles during a swimming trial in a control animal on the day of landing. Unfiltered (continuous line) and filtered (broken line) traces are superimposed for the knee joint and for the ankle joint. The trial was divided into five complete strokes and the maximum (*) and minimum () values identified and marked automatically using Matlab. B, the linear velocity of the tip of the toe in the x-axis (broken line) and y-axis (continuous line) for the trial in panel A. C, superposition of ankle (continuous line) and knee (broken line) joint angles for individual strokes in panel A. D, strokes shown in panel C normalized to 100% time. E, superposition of one trace before () and after () normalization. (These traces are also included in panels C and D.)
, 百拇医药
Calculations
The maximum and minimum values for the ankle angle, knee angle, toe x-axis velocity, and toe y-axis velocity were calculated for each stroke using Matlab (The Math Works, Inc.). These are marked on the ankle angle (Fig. 2A) and x-axis velocity (Fig. 2B) traces. The peak ankle angle was the main marker used to automatically separate each trial into its component strokes because it was easily and reliably measured in all recordings. This process generated a set of strokes of different lengths as shown in Fig. 2C for the trial in Fig. 2A. The automatic division of a trial into individual strokes was verified visually. The duration of the individual strokes was found to be independent of the ankle or knee joint maximum value, the range of joint movement, and the toe linear velocity (correlation coefficient 0.001).
, 百拇医药
The individual strokes were next normalized with respect to time. The individual strokes contained between 8 and 20 points (mean, 13.4 ± 2.1 points), one point occurring every 1/60th of a second. Each stroke was resampled to comprise 100 points (Fig. 2D). The resample points were linearly interpolated from the original data. All strokes started and ended with the original data points. The original (14 from Fig. 2C) and normalized (100 from Fig. 2D) ankle angle values for one stroke are superimposed in Fig. 2E. Note that the normalization did not change the amplitude or waveform of the angle angles. The mean and S.E.M. were obtained for each parameter in flight and control animals on R0 and R7 (Figs 8, 10 and 11).
, 百拇医药
A, day of landing mean values for five flight (n = 40 strokes) and five control (n = 39 strokes) animals. Note the narrow range for the knee joint and that the loop is shifted to the left. There is a significant difference between the groups of animals throughout the stroke. B, similar to panel A for 7 days after landing. Significant differences remain between flight (n = 40 strokes) and control (n = 36 strokes) animals in the maximum ankle angle.
A, control on day of landing. Note long early return phase. B, flight on day of landing. Note difference from control in angle positions during phase transitions. C, control 7 days after landing. The major difference from the day of landing is the shift in the beginning of the late return phase. D, flight 7 days after landing. Note shift in late return phase from R0. Flight and control diagrams are more similar 7 days after landing than on landing day because of a change in the control animals. ER, early return stroke; LR, late return stroke; EP, early power stroke; LP, late power stroke.
, http://www.100md.com
Using only one camera introduces error since the three-dimensional movement of the limb is flattened into a two-dimensional plane. Segment lengths appear to wax and wane during each stroke as the limb moves toward and away from the camera and as the animal rotates about its longitudinal axis. (This is apparent in Fig. 7A and B) The largest change was in the length of the longest segment (ankle to knee). However, there were no significant differences between flight and control animals for any segment.
, http://www.100md.com
A and B, stick figures representing the position of the hindlimb and toe trajectory during 730 ms of a swimming trial in one control (A) and one flight (B) animal on the day of landing. Swimming is from left to right. Note that there are more strokes in the flight than control and that the circle transcribed by the toe is smaller in the flight animal. C, ankle, knee joint–joint plot for five consecutive strokes in the control animal. The points corresponding to the hindlimb positions during the first stroke in panel A are indicated (). D, similar to panel C for the flight animal. The points correspond to the third stroke in panel B.
, 百拇医药
Statistical methodology
For categorical data (pre-swimming activity), the mean for each animal was calculated. The mean of these means for flight and control animals was compared. In all other cases, results are expressed as the means of individual measurements. StatView and JMP (SAS Institute Inc.) were used for statistical analysis. The unpaired Student's t test was used to determine statistical significance. Fisher's r to z was used to obtain P values for correlation coefficients and Pearson's chi-square was used to evaluate distribution. Unless otherwise indicated, all values are mean ± S.E.M. Differences with P < 0.05 were considered to be significant.
, 百拇医药
Results
The behaviour of the flight animals in the water differed clearly from that of age-matched control animals. There were no differences, however, between the two groups housed in flight-like and standard vivarium cages. For this reason data from these animals were combined into one control group in Figs 3–6. Age-matched ground control animals housed in flight-like cages are included in Figs 7 and 8.
A, percentage of trials on day of landing in which flight (filled bars) and control (open bars) animals swam immediately, floated, or tried to leave the water. There were no significant differences between flight and control animals. B, similar to A for one day after landing. C, pooled data from R2, R3, R10 and R23 showing that the majority of animals swam immediately upon entering the water. Data are mean ± S.E.M. P < 0.001, P < 0.0001 unpaired t test; A and B n = 8 flight, 15 control animals; C n = 28 flight and 70 control animals.
, 百拇医药
Swimming has not been evaluated in neonatal or adult animals after space flight. For this reason, a detailed analysis was made of all aspects of the behaviour of the animals beginning when they were first placed in the water. The Results section is organized to answer questions comparing the animals flown in space with those that remained on the ground. (1) Are there differences in the behaviour when the animals are first placed in the water on the day of landing (2) Does the initiation of swimming differ in the two groups of animals (3) Does swimming in the two groups of animals differ with respect to its form or its speed (4) If differences are seen on the day of landing, do they persist
, 百拇医药
Motor control on the day of landing
Behaviour when entering the water. Observing the behaviour of animals when they were placed in the water within hours of landing provided insight into the influence of experience on the development of motor control, especially the control of posture. The behaviour of both groups of animals when they were placed in the water fell into three categories: (a) floating with little limb movement (b) immediate swimming (c) attempting to leave the water. This last behaviour was not considered to be swimming because the animals did not make forward progress, the hindlimbs did not make full swimming strokes, and the forelimb movements were more suited to attempting to climb a vertical surface than to swimming. Trials in which the entire animal was not in view, and those that did not fall into one of the three main categories were excluded from these analyses.
, http://www.100md.com
On the day of landing, the distribution of pre-swimming activity for flight animals was random while the control animals were twice as likely to try to leave the water as to swim or float (Pearson chi-square, P = 0.0150). Accordingly, the flight animals were more likely to swim immediately or float than were the control animals (Fig. 3A). There were no statistically significant differences between the two groups when each behaviour was compared, however.
The trials in which flight and control animals swam immediately increased significantly between landing day (Fig. 3A) and one day later (Fig. 3B) (P < 0.05). Pooled data from R2, R3, R10 and R23 showed no further significant changes from R1 values (Fig. 3C). Flight animals swam immediately in a larger percentage of trials than control animals. Control animals were more likely to float when they were placed in the water than were flight animals. Both flight and control animals tried to leave the water in just over 20% of the trials.
, 百拇医药
Floating posture. Both groups of animals looked the same when they were attempting to leave the water on the day of landing but not when they were floating. Although the posture of the head and forelimbs was similar, the hindquarters of flight animals were rotated about the longitudinal axis to a greater degree than those of control animals. Although the hindquarters of both groups of animals rotated when they were first placed in the water, the control animals quickly made a correction to reduce their rotation while the flight animals did not do this. As a result, the hindlimbs of control animals tended to be directly beneath their bodies (Fig. 1C), while those of the flight animals were more likely to be rotated laterally (Fig. 1D).
, 百拇医药
To quantify this observation, we identified the stable floating posture achieved soon after the animals were placed in the water. Since the animals moved as they floated, and floated in more than one trial, we were able to view and digitize the posture of most of the animals from both the side and the back (Fig. 1). Three relative hindlimb joints (foot, ankle, and knee) were calculated from the points digitized from the side view (Fig. 1B). There were no differences in these joint angles between the day of launch (P14) and the day of landing (P30, flight and control) as shown in Table 1.
, 百拇医药
The angles of the hindquarters and hindlimbs were calculated from the points digitized from the back view (Fig. 1C and D). As shown in Fig. 4A, the absolute angle of the hindquarters with respect to the horizontal was clearly greatest before launch (P14), and smallest in the P30 ground control animals. Values for the flight animals at landing (P30) were closer to launch (P14) values than to P30 control animals. There were significant differences among all three groups of animals as shown in Fig. 4A. The range also differed; being 45 deg to 122 deg at P14, 46 deg to 73 deg in P30 flight, and 0 deg to 22 deg in P30 ground control animals.
, 百拇医药
A similar pattern was seen in the position of the hindlimbs as shown in Fig. 4B and C. For the hindlimb closest to the surface (Fig. 4B), significant differences were seen between the P14 and P30 control and between the P30 flight and P30 control animals. For the lower hindlimb, there were significant differences between P14 and P30 flight and between P14 and P30 control animals. Rotation of the hindlimbs was secondary to hindquarter rotation. This is indicated by a high correlation of these two parameters. The correlation coefficients for the hindquarters–upper hindlimb pair, and hindquarters–lower hindlimb pair were both > 0.8 (P < 0.0001). The rotations of the two limbs were also highly correlated (>0.7, P < 0.0001).
, 百拇医药
In answer to the first question there were clear differences in the behaviour of flight and control animals when they were first placed in the water. Two differences stand out: (1) control animals tried to leave the water more than the flight animals and (2) the hindquarter and hindlimb position appeared to be independent of head and forelimb position in flight but not control animals.
Swimming initiation
Swimming in flight and control animals on the day of landing was characterized by hindlimb-only propulsion as is typical of adult animals (Gruner & Altman, 1980). Forelimb movement was also present in some strokes as is seen in younger animals (Beckoff & Trainer, 1979), but this was not typical. It appeared during the postflight videotaping that swimming was essentially the same in both groups of animals. However, differences were seen upon careful analysis of the stroke duration and kinematics of hindlimb movement.
, 百拇医药
Initial swimming movements. On the day of landing, every flight animal floated before making any movement the first time it was placed in the water. Since floating was the only activity in all of the flight animals before they began to swim, we were able to analyse the initial swimming movement. Swimming was initiated with hindquarter rotation and subsequent correction of the hindlimb displacement. In six of the flight animals this was followed by a short kick brought about by extension of one hindlimb. The other two flight animals extended both hindlimbs together. Swimming was initiated during the first (three animals) or second (five animals) trial.
, http://www.100md.com
Of the control animals, six floated when they were placed in the water on the first trial (eight attempted to leave the water and one swam immediately). Coordinated swimming began during the first (nine animals) or second (six animals) trial. The transition from floating to swimming followed a similar pattern to that seen in the flight animals. This was the case even if the animal did not begin the first or second trial by floating. The first swimming movement was a short kick brought about by extension of one hindlimb in ten control animals. (One control animal began with a forward hindlimb movement while two began swimming with simultaneous extension of both hindlimbs.)
, 百拇医药
Delay before swimming. We analysed how long it took the animals to begin swimming after they were placed in the water. The flight animals () began to swim sooner than control () animals as shown in Fig. 5 where the delay before swimming is plotted as a function of days after landing. This delay decreased in both groups of animals after the day of landing, but flight animals continued to begin to swim sooner than the control animals throughout the test period. When the trials in which the animals swam immediately are included (delay = 0), both plots in Fig. 5 are shifted downward preserving the differences.
, 百拇医药
This interval was greatest on the day of landing (day 0) in both groups of animals. The flight animals () began swimming before the control animals () on every test day. Data are mean ± S.E.M., *P < 0.05, P < 0.001, P < 0.0001; unpaired t test, n = 104–165 in control, n = 57–87 in flight animals.
In answer to the second question, both groups of animals initiate swimming with the same movements, but the flight animals do this sooner after being placed in the water than the control animals.
, 百拇医药
Swimming stroke duration, style, and speed
Stroke duration. A swimming stroke comprises two phases: the power phase that propels the animal through the water, and the return phase which brings the hindlimb forward in preparation for another power phase. The durations of both phases were measured and summed to give the stroke duration. Strokes were included for analysis only if the difference between the power and return phases was ± 1 frame (±16 ms). This eliminated irregular swimming and ‘coasting’ where, after a power phase, the animal held its limbs still before beginning the return phase. Coasting was seen in 10% of the strokes in flight, in 5% of the strokes in control, and in 2% of the strokes in the P14 animals.
, http://www.100md.com
The mean stroke cycle duration in P14, control, and flight animals is plotted as a function of time in Fig. 6. (S.E.M. values are smaller than the circles in the plot.). As shown in the figure, on landing day (day 0) the mean stroke duration in both flight () and control () animals was significantly shorter than on P14 (P < 0.0001). The durations of both the power and return phases were significantly shorter in the flight animals on every test day than the age-matched ground control animals. Thus, the flight animals' movements during both phases contributed to the short stroke duration. Since plots of the power stroke and return stroke duration had the same shape and relationship between flight and control animals as the total stroke duration plot (Fig. 6), they are not shown.
, http://www.100md.com
Note that on the day of landing (day 0), stroke duration of both flight () and control () animals were shorter than on the day of launch ( ). Throughout the postflight period the stroke duration was significantly shorter in the flight animals. Data are mean, ± S.E.M., *P < 0.05, P < 0.001, P < 0.0001; unpaired t test. Number of observations n = 146–443 except flight R23 = 76. Error bars are smaller than symbols.
The persistence of short stroke duration for 30 days prompted us to examine the swimming stroke in more detail on R0 and R7 in five flight and five control animals housed in flight-like cages. Return day 7 was chosen for the high quality of the data and because there were no significant long-term differences in flight animals after that day.
, http://www.100md.com
Trajectories and joint angles. The position of the hindlimb in an individual control and an individual flight animal swimming from left to right on the day of landing are represented as stick figures in Fig. 7A and B, respectively. The trajectory of the toe is also shown. Examination of the stick figures reveals that the toe trajectory was smaller and less circular in the flight (Fig. 7B) than in the control (Fig. 7A) animal. To investigate this observation and the movement of the hindlimb more closely, we measured the ankle and knee joint angles. These angles were chosen because they have the largest range of movement during swimming.
, http://www.100md.com
Angle–angle diagrams for these individual animals are shown in Fig. 7C and D where the ankle joint angle is plotted as a function of the knee joint angle. These diagrams illustrate the range of movement of each joint as well as how the joints work together as the hindlimb moves through the water. The ankle-knee angle-angle plot for six consecutive strokes in the control animal is shown in Fig. 7C (three of these strokes are shown in panel Fig. 7A). The numbers on the thick line in the plot correspond to limb positions marked in Fig. 7A. The power phase begins as the knee and ankle joints both extended to bring the limb down and back (positions 2–5). In this stroke, maximum knee extension was reached at position 4 and maximum ankle extension at position 5. Knee flexion then brings the limb up as it continues to move back through positions 5 and 6. Flexion of the ankle and knee together bring the limb up and forward from position 7–13 during the return phase. Finally, the knee extends, bringing the limb past position 13, to complete the return phase.
, http://www.100md.com
A similar ankle–knee angle–angle plot for five strokes in a flight animal is shown in Fig. 7D. (Three of these strokes are shown in Fig. 7B) The numbers correspond to those for the third stroke in Fig. 7B. This stroke was chosen for comparison with the first stroke in the control animal because it has a similar length and toe trajectory. Visual comparison of the plots from the two animals on R0 (Fig. 7C and D) reveals that the flight plot is shifted to the left.
, 百拇医药 To find if the differences noted between these two animals was typical, mean values were calculated. Superposition of mean angle–angle plots of flight and control animals on R0 (Fig. 8A) revealed clear differences in the form of the hindlimb movement that are similar to those seen for the individual animals in Fig. 7C and D. Throughout the stroke, the loop for the flight animals (solid line) was shifted to the left toward knee flexion. The narrow loop reflects the small range of mean knee movement compared with the control (broken lines) animals. The narrow range of mean knee movement is consistent with the short stroke duration (Fig. 5).
, 百拇医药
The opposite is the case in for the ankle. Here, flight animals have a greater mean range of movement than controls with no clear shift along the joint angle axis. Note that in addition to the difference in the absolute values for the angles, there is a difference in the shapes of the two ankle–knee angle–angle plots (Fig. 8A), reflecting a difference in the relationship between the two joints.
Over the first week after landing the joint–joint relationship changed in both groups of animals (Fig. 8B) but this was significant only for the flight animals. There was a rightward shift along the mean knee joint axis in the flight animal plot (solid line, Fig. 8B). This was due to increases in both the minimum (P < 0.01) and maximum (P < 0.0001) joint values (see Table 2). There was no change in the range of mean knee movement. There was no shift along the ankle axis since the minimum (P < 0.0001) and maximum (P < 0.0001) ankle angles both decreased between R0 and R7. This resulted in a decreased range of ankle movement on R7.
, 百拇医药
Values for ankle and knee minimum (most flexed angle) and maximum (most extended angle) and range of movement are summarized in Table 2. Significant differences between flight and control animals are marked (*). The R7 value for the maximum ankle angle in flight animals reached the R0 control value. However, a significant difference remained between flight and control animals on R7 because the control value for this parameter decreased between R0 and R7.
The narrow range of knee movement and small toe trajectory seen in the flight animals on R0 is consistent with short stroke duration. However, these differences in swimming style were much less pronounced on R7 (Fig. 8B) while the short stroke duration persisted (Fig. 5). This finding prompted us to search for another variable that might be linked to the persistent short stroke duration. Toward this end, we analysed the velocity of hindlimb movement since an increased velocity could contribute to short stroke duration.
, http://www.100md.com
Linear velocity. To examine directly the speed of hindlimb movement, the linear velocity of the digitized points was calculated for the x- and y-axes. The speed of the movement of the tip of the toe was chosen for analysis since this gave the best indication of effective hindlimb movement. The first stroke in Fig. 7A A and the third stroke in Fig. 7B are shown in Fig. 9A and B, respectively. The linear velocity of the toe movement for the trial containing the strokes in Fig. 9A and B are shown in the y-axis–x-axis velocity–velocity plots in Fig. 9C and D, respectively. (The thick lines correspond to the stroke shown in Fig. 9A and B)
, http://www.100md.com
A and B, stick figure showing the position of a hindlimb and toe trajectory for the control (A) and flight (B) animal (same stroke as in Fig. 7). C, y-axis–x-axis velocity–velocity plot for the tip of the toe during five consecutive strokes in the control animal. The velocity of the toe at each frame in panel A is marked as the hindlimb moves from the early power (EP) through the late power (LP), early return (ER), and late return (LR) phase of the stroke (see text). D, similar to panel C for the flight animal.
, 百拇医药
By convention, upward and forward velocity is positive, downward and backward velocity is negative. Thus, during a swimming stroke the y-axis–x-axis velocity–velocity curve crosses through zero four times. Characterizing the power phase by movement of the foot back, and the return phase as movement forward, the swimming stroke can be divided into four phases beginning with the return phase (as marked in Fig. 9C): (1) early return phase (ER) as the foot moves forward and up; (2) late return phase (LR) as the foot moves forward and down; (3) early power phase (EP) as the foot moves back and down and; (4) late power phase (LP) as the foot moves back and up. In the control example stroke (Fig. 9C), the return phase began just after point 6 as the limb moved forward. During this phase the limb moved up (points 7–13) then down (points 1–2). (Because there was little change in the linear velocity in either the x- or y-axis at the end of the early return phase, points 11–13 are clustered together.) The EP phase began from position 2 as the limb moved downward and back to position 4. The maximum downward velocity was reached midway during this movement (point 3) and the maximum velocity back occurred at the end (point 4). During the LP phase (points 5–7) the limb moved upward and back.
, 百拇医药
A similar pattern was seen in the flight animal (Fig. 9B and D), but the power phase had fewer frames (the limb moved faster). The early power phase began between points 1 and 2 and continued to point 3. At point 3, the foot had attained maximum velocity back. This velocity was maintained as the foot moved back and up to point 4. The maximum upward velocity was reached near the end of the late power phase (point 5). During the early return phase (point 5–7) the foot moved forward and up. In the late return phase (points 8–12), the foot is brought forward and down in preparation for the next power phase. The mean values for the maximum velocity in all four directions are given in Table 3.
, http://www.100md.com
The mean y-axis–x-axis velocity–velocity plots for five flight (continuous line) and five control (broken line) animals on R0 are superimposed in Fig. 10A. The fastest absolute movements occurred in the horizontal axis near the midpoints of the power and return phases. The foot of the flight animals moved faster at both of these points (Table 3). Mid-power phase corresponds to when the most propulsion is generated (maximum x-axis velocity). This increased speed could contribute to the short stroke duration in the flight animals. The mean maximum movement down was faster in the control animals (Table 3). As can be seen in Fig. 10A, this is associated with a difference in the shape of the y-axis–x-axis velocity–velocity plot. In the control animals the foot accelerated down during the transition from the return phase to the power phase (Fig. 10A, LR–EP). In contrast, the flight animals tended to maintain a constant downward velocity during this movement. There was no difference in the velocity of the movement upward.
, http://www.100md.com
A, on the day of landing there was a significant difference in toe velocity in the flight and control animals except in the transition from the LP to the ER phase. B, similar to panel A for 7 days after landing. Significant differences remain in the maximum forward LR and downward EP velocity. Number of strokes as in Fig. 8
Since difference in stroke duration persisted after R0, the velocity of the tip of the toe was calculated on R7 (Table 3). Between R0 and R7 the speed of movement of the foot downward (P < 0.0001) and forward (control P < 0.0001; flight P < 0.01) increased in both groups of animals. In the flight animals, upward movement (P < 0.0001) also increased. The increased speed was greater in the control than in the flight animals, but differences remained on R7 (Fig. 10B). Control animals moved their limbs faster than flight animals downward at the beginning of the stroke and flight animals moved faster at the mid-point of the return stroke (Table 3).
, 百拇医药
Swimming phases
Values for the ankle and knee angles (Figs 7 and 8) provide information about the position of these joints, while the linear velocity of the toe (Figs 9 and 10) provides information about movement direction and speed. Combing these two data sets allowed us to link knee and ankle joint position to limb movement in a unique way. To accomplish this, the points at which the mean toe velocity crossed zero were identified and marked on the mean angle–angle diagrams (Fig. 11). This allowed us to divide the angle–angle plots into the four phases of limb movement identified by the velocity data.
, http://www.100md.com
The combination of these data sets demonstrates that the phases of the swimming stroke are associated with different mean joint angle values in control (Fig. 11A and C) and flight (Fig. 11B and D) animals. Beginning with the return phase (ER), both joints flexed as the hindlimb began to move forward (it was already moving up). In the control animals (Fig. 11A), the ankle joint flexed throughout this phase and the knee began to extend before the next phase (LR) began. In contrast, this phase was cut short in the flight animals and both joints flexed (Fig. 11B).
, http://www.100md.com
The late return phase showed the most striking difference in joint position between the two groups of animals. In control animals, both joints extended throughout this phase as the limb moved forward and down. In flight animals the ankle flexed rather than extended, with little change in the knee angle, and both joints extended only at the end of the movement. This more flexed position of the ankle and knee reduced the forward toe trajectory since the toe was not brought as far forward as in the control animals before beginning to move back to begin the power stroke. (Compare Fig. 11A and B).
, 百拇医药
Propulsion for forward movement occurs during the power phase. In the control animals (Fig. 11B) the ankle extends throughout this movement (EP) but the knee begins to flex and flexes throughout the late power phase (LP). In flight animals (Fig. 11B), both joints extend throughout the early power phase (EP). The knee begins to flex only in mid–late power phase. The values at the beginning of each phase and the net movement during each phase are given in Table 4. Note that at each transition in movement direction the ankle and knee joint angles were significantly different in the two groups of animals except for the early return phase. The change in joint angle during each phase was significantly different in all the phases.
, http://www.100md.com
Changes in both control and flight animals over the first week after landing worked together to reduce the differences that were seen on R0. The angle–angle loops for control animals on R7 (coloured) and R0 (open, with dots marking the phase transitions) are superimposed in Fig. 11C. Two changes may be noted; (1) there was a clockwise shift in the beginning of the early return (ER), late return (LR) and late power (LP) phases (Figs 11C and 2) maximum knee extension is less on R7. These changes all work to make the relationship between the two angles more similar to that of the flight animals on R0.
, 百拇医药
Similar superposition of the R0 and R7 angle–angle plots for flight animals reveal three differences between R0 and R7 (Fig. 11D): (1) there was a shift of the loop to the right; (2) the beginning of the late return (LR) and early power (EP) phases was shifted counterclockwise; (3) the beginning of the early return (ER) phase was shifted clockwise. The overall effect was to reduce the range of movement during the power phase of the stroke. Significant differences remained between flight and control animals on R7, however. This was most notable in the ankle during the return phases and in the knee in the early power phase (Table 4).
, http://www.100md.com
Discussion
These experiments show that exposure of young animals (P14–P30) to microgravity deeply influences motor function in a protracted fashion. Indeed, on the day of landing, the motor behaviour of flight animals in the water was clearly different from that of ground control animals. Some behaviours seen on the day of landing, such as an apparent rostro-caudal ‘unlinking’ of postural control, quickly reverted to control values. However, behaviours that fall within the time domain (pre-swimming interval, stroke duration, and stroke kinematics) were not transient, but persisted and were still present in adult animals 30 days after landing. Both the transitory and persistent differences between flight and control animals are clearly related to their experience between P14 and P30 (in flight).
, 百拇医药
It can be argued that motor system development of animals on the ground proceeds to optimize performance in 1 g, while animals living in a microgravity environment may develop different motor control solutions to achieve the most efficient motor execution, i.e. that requiring the least time and energy to achieve a goal. This is consistent with the idea that development proceeds to meet the needs of the animal in terms of survival, safety, motivation, social, and other expectations. While these basic needs are similar in any gravitational environment, the methods used to achieve them will differ, especially when locomotion is involved.
, http://www.100md.com
Both internal (e.g. bone and muscle characteristics) and external (e.g. gravitational field and temperature) parameters will influence adaptation of an animal to its environment, and space flight introduces others such as radiation (see Walton et al. 2005). While the development of all systems is affected to some extent by a 16-day space flight and must be considered in interpretation of these data, our interpretation of these findings places emphasis on nervous system development.
, http://www.100md.com
Adaptation of the motor system to microgravity
Initial movements. Adaptation of the nervous system of flight animals to microgravity is reflected in their behaviour when they were placed in the water. Thus, while control animals tried to leave the water on the majority of trials on R0, the flight animals did not show a clear preference between this and floating or swimming (Fig. 3). This suggests that the buoyancy afforded by the water was closer to the environmental parameters of flight than to those of ground control animals. Thus, the flight animals may have been less fearful or less stressed when placed in the water. Another factor that may influence an animal's behaviour in the water is its own buoyancy. This is provided in part by air bubbles trapped in the animal's fur, and cleaner fur accumulates more bubbles. Because of cage movement during re-entry and landing, control animals tended to be cleaner than the flight animals on R0. This is probably not a major factor since control animals were less likely to float than the flight animals (Fig. 3).
, 百拇医药
After R0, swimming was the preferred initial activity in both groups of animals (Fig. 3B and C), however, swimming occurred in a larger percentage of trials in the flight animals. This was due to the persistence of pre-swimming floating in control animals after R0 (Fig. 3B). The dominance of swimming is not surprising since this activity helps the animals to achieve their goal – to leave the water (as indicated by their preference for swimming towards the platform and their leaving the water every time they reached the platform).
, http://www.100md.com
Throughout the study, flight animals began to swim sooner after entering the water than control animals (Fig. 4). This was not accounted for by a difference in the pre-swimming behaviour of the two groups animals. Flight animals just spent less time doing each activity before beginning to swim. The behaviour of the flight animals is consistent with the interpretation offered above, namely, that the buoyancy of water provides a familiar environment and they make appropriate movements. However satisfactory this explanation may be for R0, it does not predict that the difference would persist for 30 days since by then both groups of animals had almost daily swimming experience. The short interval may reflect a more general property of flight animals – they move through the water faster than control animals.
, http://www.100md.com
It is difficult to know if cognitive factors may have contributed to the differences in the behaviour of these two animal groups. However, a study of these same rats showed no differences in their ability to learn or remember spatial tasks (Morris water maze), in long-term potentiation, or hippocampal morphology (Temple et al. 2003).
Floating. When they were placed in the water for the first time in their experience, all of the flight animals floated quietly before making any movements. Since the otolithic system had experienced a reduced gravitational input for 16 days, such well-controlled floating was unexpected. Postural control of the head indicates that the peripheral vestibular apparatus (vestibular labyrinths, Scarpa's ganglia, vestibular nerves) and vestibulocolic reflexes were fully functional within hours of landing. This is consistent with the fact that development of this system was largely complete at launch (Dememes et al. 2001). Control of the forequarters further indicates that, among others, the vestibulospinal and corticospinal circuits were functional, at least to the level of the cervical spinal cord. In fact, in this respect, flight animals were similar to ground controls. This suggests that vestibular function, as reflected in the ability of the animals to maintain balance in the water, did not deteriorate in the reduced gravitational input of microgravity. It is also consistent with our finding that flight animals were able to right themselves from a supine to a prone position on the first trial on R0 (Walton et al. 2005).
, http://www.100md.com
While the floating posture of the head and forequarters was similar in flight and control animals, the position of the hindquarters differed (Fig. 4). This may reflect a lack of maturation of earth-based postural control in the flight animals. The control animals moved their hindlimbs under the body, resulting in a posture that was appropriate and necessary for quadruped stance as well as swimming. This correction did not occur in the flight animals until they began to swim.
, http://www.100md.com
This finding suggests that the uncorrected hindquarter rotation reflects experience-dependent development of midline control that is adaptive to the microgravity environment. Hindlimb rotation and the rostro-caudal ‘unlinking’ both reflect the characteristics of locomotion in microgravity where the forelimbs were often used as the principal form of propulsion. Since the hindlimbs did not necessarily participate in locomotion, their posture could be largely independent of the forelimbs and often floated away from the surface (DeFelipe et al. 2002; see Supplemental material). As a result, we proposed that positions in which the hindquarters are rotated with respect to the forequarters are included in the set of ‘normal’ postures for flight animals. Uncorrected hindquarter rotation while floating does not reflect a lack of postural control because the animals were stable in the water, and before making the first stroke every flight animal rotated its hindquarters to bring its hindlimbs into the position appropriate for swimming.
, 百拇医药
Swimming. When they were launched into space, the eyes of the flight animals were open, they had achieved quadruped stance, interlimb coordination, quadruped locomotion, and were able to move about freely. That these motor skills were modified as the nervous system adapted to microgravity over the following two weeks is reflected in the way the flight animals swam, as well as in their pre-swimming behaviour.
Stroke duration is a key parameter in swimming analysis because it reflects both the form (Figs 7 and 8) and velocity (Figs 9 and 10) of hindlimb movement. During normal terrestrial development this parameter changes over the first two weeks of life, decreasing from over 700 ms at birth to near 300 ms by P14 (Beckoff & Trainer, 1979). These experiments have shown a subsequent decrease in control animals from a mean of 284 ms P14 (Fig. 6, P14) to a mean of 217 ms in control and 205 ms in flight animals on P30 (Fig. 6, day 0). Although the decrease after P14 is small when compared to the normal range of development, it is nevertheless significant (P < 0.0001) as is the difference between control and flight animals on P30 (Fig. 6). The difference in stroke duration (Fig. 6) and pattern of limb movement (Fig. 11) between control and flight animals on R0 demonstrates for the first time that establishment of adult swimming parameters is influenced by environmental factors.
, 百拇医药
Comparison with hindlimb suspension
Comparison of swimming in flight animals and in those experiencing hindlimb suspension address the question of which mechanisms may underlie the findings reported here. The short stroke duration in flight animals is not due to the lack of hindlimb weight bearing in microgravity, per se. Indeed, hindlimbs do not bear the animal's weight under either condition, yet the effects on stroke duration are in opposite directions. Although in the tail suspension paradigm, the animals do bear weight for one hour in 24 (while they are feeding), it is unlikely that this is long enough to influence their adaptation to the suspension paradigm. Also, we have found increases in stroke duration after only 5 h of suspension with no weight bearing before testing (K.W., personal observation). The difference in stroke duration is directly related to limb position during the stroke, which itself is directly related to the different use of the hindlimbs during suspension and in microgravity.
, http://www.100md.com
Suspended animals show an ‘extensor bias’ and keep their hindlimbs extended behind them unless they are asleep. Similarly, they swim with their hindlimbs extended, resulting in a larger trajectory and longer stroke duration than control animals (Walton et al. 1992). In contrast, animals in microgravity show a flexor bias and use their limbs to pull themselves about the cage. Accordingly, they swim with their hindlimbs in a more flexed position resulting in a smaller trajectory and shorter stroke duration than control animals (Fig. 7, Table 2). The finding of such flexor bias in animals adapted to microgravity is not surprising. This is consistent with a ‘flexor bias’ that has been reported for quadrupedal locomotion in-flight or postflight in non-human primates (Recktenwald et al. 1999) and in humans (Bloomberg & Mulavara, 2003; McCall et al. 2003).
, 百拇医药
Thus, our results indicate that in the absence of a need to counter gravity during standing and for extensor activity to move the animal forward prior to foot lift (Gruner & Altman, 1980), the motor programme for locomotion is altered in flight animals. This is consistent with other studies of motor activity during and after space flight that have been interpreted in terms of changes in the brain's internal model of gravity upon which movement is based (see McIntyre et al. 2001; Courtine et al. 2002).
, 百拇医药
Role of the musculoskeletal system
It might be expected that changes in muscle properties may underlie the differences seen between flight and control animals (see Fitts et al. 2001). The changes in hindlimb muscle properties under conditions of tail suspension and microgravity have been well documented. The postnatal development of antigravity muscles is selectively suppressed in both tail-suspended (Edgerton & Roy, 1994; Riley et al. 1995) and flight animals (Adams et al. 2000a, b). Yet, the effects of tail suspension and space flight on swimming style as well as on stroke duration are in opposite directions as summarized above. Thus, similar changes in muscle properties are associated with opposite changes in swimming.
, 百拇医药
Critical period hypothesis
One question posed at the start of these experiments was whether postnatal activity-dependent modifications of sensory-motor function remain fixed or are subject to alteration throughout the life span. Our findings clearly show that an altered gravitational field influences the postnatal development of sensory–motor function. Some of these characteristics persisted for as long as the animals were followed (one month). These included the short pre-swimming interval and short stroke duration in flight animals. The nature and persistence of the changes suggests that they (1) are due to the particular experience of the animals during a critical period of development and (2) reflect a long lasting adaptation of the animals to the microgravity environment. The data suggest that the most fundamental of these adaptations is a resetting of the basic motor rhythm to a higher frequency.
, http://www.100md.com
In conclusion, then, the data presented in this and the accompanying paper (Walton et al. 2005) indicate a deep level of CNS adaptability. Thus, in the absence of gravitation one of the primordial sensory inputs required for the organization of up–down and left–right is missing. The CNS adapts just sufficiently to allow proper movement in space. This is done without modifying the load-unburdening systems such as the oculomotor and directional head orientation. Equally remarkable is the fact that if such changes occur at a time in development where connectivity is being optimized, the changes remain throughout the life span of the animal, suggesting the possibility of a motor critical period as seen in sensory systems.
, 百拇医药
Supplemental material
The online version of the companion paper (Walton et al. 2005) can be accessed at: DOI: 10.1113/jphysiol.2004.074385
http://j.p.physoc.org/cgi/content/full/jphysiol.2004.074385/DC1
and contains supplemental material consisting of a movie entitled: A neurolab experiment: The effect of gravity on postnatal motor system development.
This material can also be found as part of the full text HTML version available from http://www.blackwellsynergy.com
, 百拇医药
References
Adams GR, Haddad F, McCue SA, Bodell PW, Zeng M, Qin L, Qin AX & Baldwin KM (2000a). Effects of spaceflight and thyroid deficiency on rat hindlimb development. II. Expression of MHC isoforms. J Appl Physiol 88, 904–916.
Adams GR, McCue SA, Bodell PW, Zeng M & Baldwin KM (2000b). Effects of spaceflight and thyroid deficiency on hindlimb development. I. Muscle mass and IGF-I expression. J Appl Physiol 88, 894–903.
, 百拇医药
Beckoff A & Trainer W (1979). The development of interlimb co-ordination during swimming in postnatal rats. J Exp Biol 83, 1–11.
Bloomberg JJ & Mulavara AP (2003). Changes in walking strategies after spaceflight. IEEE Eng Medical Biol March/April, 58–62.
Conlee JW & Parks TN (1981). Age- and position-dependent effects of monaural acoustic deprivation in nucleus magnocellularis of the chicken. J Comp Neurol 202, 373–384.
, 百拇医药
Courtine G, Papaxanthis C & Pozzo T (2002). Prolonged exposure to microgravity modifies limb endpoint kinematics during the swing phase of human walking. Neurosci Lett 332, 70–74.
DeFelipe J, Arellano JI, Merchán-Perez A, González-Albo MC, Walton K & Llinás R (2002). Spaceflight induces changes in the synaptic circuitry of the postnatal developing neocortex. Cortex 12, 883–891.
Dememes D, Dechesne CJ, Venteo S, Gaven F & Raymond J (2001). Development of the rat efferent vestibular system on the ground and in microgravity. Brain Res Devel Brain Res 128, 35–44.
, 百拇医药
Edgerton VR & Roy RR (1994). Neuromuscular adaptation to actual and simulated weightlessness. Adv Space Biol Med 4, 33–67.
Fitts RH, Riley DR & Widrick JJ (2000). Physiology of a microgravity environment invited review: microgravity and skeletal muscle. J Appl Physiol 89, 823–839.
Fitts RH, Riley DR & Widrick JJ (2001). Functional and structural adaptations of skeletal muscle to microgravity. J Exp Biol 204, 3201–3208.
, http://www.100md.com
Fox K (1992). A critical period for experience-dependent synaptic plasticity in rat barrel cortex. J Neurosci 12, 1826–1838.
Gray L, Smith Z & Rubel EW (1982). Developmental and experimental changes in dendritic symmetry in n. laminaris of the chick. Brain Res 244, 360–364.
Gruner JA & Altman J (1980). Swimming in the rat: analysis of locomotor performance in comparison to stepping. Exp Brain Res 40, 374–382.
, http://www.100md.com
Hillman DE, Chen S, Bing R, Penniston JT & Llinás R (1996). Ultrastructural localization of the plasmalemmal calcium pump in cerebellar neurons. Neurosci 72, 315–324.
Homick JL, Delaney P & Rodda K (1998). Overview of the Neurolab Spacelab mission. Acta Astronaut 42, 69–87.
Hubel DH & Wiesel TN (1970). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol 206, 419–436.
, 百拇医药
Liskowsky DR, Frey MA, Sulzman FM, White RJ & Likowsky DR (1996). The Neurolab mission and biomedical engineering: a partnership for the future. BME 10, 11–25.
McCall GE, Goulet C, Boorman GI, Roy RR & Edgerton VR (2003). Flexor bias of joint position in humans during spaceflight. Exp Brain Res 152, 87–94.
McIntyre J, Zago M, Berthoz A & Lacquanti F (2001). Does the brain model Newton's Law Nature Neurosci 4, 693–694.
, 百拇医药
Maese C & Ostrach L (2002). Neurolab. Final report for the Ames. Research Center Payload NASA/TM-2002–211841, NASA, Moffett Field.
Meisami E (1978). Influence of early anosmia on the developing olfactory bulb. Prog Brain Res 48, 211–229.
Mogilner A, Grossman JA, Ribary U, Joliot M, Volkmann J, Rapaport D, Beasley RW & Llinás RR (1993). Somatosensory cortical plasticity in adult humans revealed by magnetoencephalography. Proc Natl Acad Sci U S A 90, 3593–3597.
, 百拇医药
O'Leary DD, Ruff NL & Dyck RH (1994). Development, critical period plasticity, and adult reorganizations of mammalian somatosensory systems. Curr Opin Neurobiol 4, 535–544.
Penn AA & Shatz CJ (1999). Brain waves and brain wiring: the role of endogenous and sensory-driven neural activity in development. Pediatr Res 45, 447–458.
Recktenwald MR, Hodgson JA, Roy RR, Riazanski S, McCall GE, Kozlovskaya I, Washburn DA, Fanton JW & Edgerton VR (1999). Effects of spaceflight on rhesus quadrupedal locomotion after return to 1G. J Neurophysiol 81, 2451–2463.
, 百拇医药
Riley DA, Thompson JL, Krippendorf BB & Slocum GR (1995). Review of spaceflight and hindlimb suspension unloading induced sarcomere damage and repair. Basic Appl Myol 5, 139–145.
Schlaggar BL, Fox K & O'Leary DD (1993). Postsynaptic control of plasticity in developing somatosensory cortex. Nature 364, 623–626.
Sosula L & Glow PH (1971). Increase in number of synapses in the inner plexiform layer of light deprived rat retinae: quantitative electron microscopy. J Comp Neurol 141, 427–451.
, 百拇医药
Temple MD, Denslow MJ, Kosik K & Steward O (2003). Neural development under conditions of spaceflight. In The Neurolab Spacelab Mission: Neuroscience Research in Space, ed. Buckey JC and Homick JL, pp. 161–168. NASA, Washington DC.
Walton K (1998). Postnatal development under conditions of simulated weightlessness and space flight. Brain Res Brain Res Rev 28, 25–34.
Walton KD, Harding S, Anschel D, Harris YT & Llinás R (2005). The effects of microgravity on the development of surface righting in rats. J Physiol 565, 593–608.
, 百拇医药
Walton KD, Lieberman D, Llinás A, Begin M & Llinás RR (1992). Identification of a critical period for motor development in neonatal rats. Neurosci 51, 763–767.
Wiesel TN (1982). Postnatal development of the visual cortex and the influence of the environment. Nature 299, 583–591.
Wiesel T & Hubel D (1963a). Single-cell responses in striate cortex of kittens deprived of vision in one eye. J Neurophysiol 26, 1003–1017.
Wiesel TN & Hubel DH (1963b). Effects of visual deprivation on morphology and physiology of cells in the cat's lateral geniculate cortex. J Neurophysiol 26, 978–993., 百拇医药(Kerry D. Walton, Louis Be)