ABSTRACT
Continuous, long-term observations of amniote embryos have always been difficult. Special culture techniques for young avian and mammalian embryos have been developed (New, 1967) and these have helped to visualize the early stages of development. But studies of normal and abnormal development during the major period of organogenesis have been made largely by tedious indirect methods, such as the examination of a series of embryos preserved at different time intervals. Transitory responses to toxic stimuli have been particularly difficult to detect in this manner. To observe the visible initial effects of terato-genic agents, a photographic time-lapse study of chick embryos in their natural, in ovo, state was initiated. This report compares the changes in normal and hypoxia-treated embryos during the third day of development.
Of the many agents which produce abnormal development, oxygen deficiency is one of the better known, since (1) it is readily induced in the laboratory by a variety of means, and (2) it is generally considered to be a significant cause of spontaneously occurring anomalies (Rubsaamen, 1952; Ingalls, 1952). Dareste, in 1877, and many others after him, produced oxygen lack in the chick embryo by covering the egg with impervious material. More recently, the effects of hypoxia on the chick embryo have been studied utilizing partial vacuum and gaseous mixtures (Gallera, 1951; Naujoks, 1953; Biichner, 1955; Grabowski& Paar, 1958; Grabowski, 1961, 1964). The last two studies in particular indicated that the effects of hypoxia were more varied than generally suspected. For instance, some of the abnormal development induced by hypoxia is caused by cytotoxic effects, but most is mediated through a complex ‘edema syndrome’. In our photographic study, the edematous state, its development and some of its consequences are described in detail
MATERIALS AND METHODS
Eggs of the Kimber strain of white leghorn chickens were incubated at 39 °C for 3 days prior to use. Observations of the embryo were made through an overlying coverglass window sealed with paraffin to the shell (see New, 1967). The eggs were kept in a transparent chamber maintained at 39 °C ± 1° through which pre-warmed air or a mixture of nitrogen and air flowed. Photographs were taken on Kodak Plus-X film with a Nikon F 35 mm camera, equipped with either a 55 mm Micro Nikkor lens and bellows attachment or a 500 mm Medical Nikkor lens. The definition obtained with both lenses was equally good, but the latter, a modified telephoto lens, was more convenient to use because of the greater working distance. A green filter (Wratten B) was used to enhance the visibility of the circulatory system. An electronic flash gun beneath the egg provided illumination. The initial magnification was x 2-5; the prints for study purposes were uniformly made at x 15. With this system observations could be made on embryos 2-5 days old with exposures taken as rapidly as 7 sec. apart.
Many soft structures such as the brain and spinal cord, eyes, somites, limb buds, and even visceral arches were clearly seen on these photographs even though no vital stain was used (Plates 1-3). The embryonic and extra-embryonic blood vessels were particularly clear. A planimeter measured the area of the whole embryo or parts of it. Such a measurement essentially represents the area of an optical saggital section. In some cases, changes in shape were studied by com-paring tracings made from successive prints (Text-fig. 1). In several experiments measurements were taken of the angle between the head and trunk of the embryo, using the dorsal aorta of both regions as a guide.
Hypoxia-induced enlargement of heart and adjacent blood vessels. Tracings taken from time-sequence photographs of hearts at same stage of beat, i.e. atrium contracted, ventricle just starting to contract and filling the truncus arteriosus. All three embryos at stage 17 at start of observations. Embryo A is a control; embryos B and C were exposed to 6 % O2 for 6 h. Note enlargement of all parts in treated embryos still evident at 9 h, 3 h after return to air. In embryo B the heart has returned to near-normal size by 13 h. Note also that aortic arch 4 is not open in the first drawing of each series, but is present by 7 h. Aortic arch 2 shows signs of closing by 13 h. Abbreviations: v.v., vitelline vein; v., ventricle; t.a., truncus arteriosus; 2, 3, 4, aortic arches; atrium is hidden behind ventricle.
Hypoxia-induced enlargement of heart and adjacent blood vessels. Tracings taken from time-sequence photographs of hearts at same stage of beat, i.e. atrium contracted, ventricle just starting to contract and filling the truncus arteriosus. All three embryos at stage 17 at start of observations. Embryo A is a control; embryos B and C were exposed to 6 % O2 for 6 h. Note enlargement of all parts in treated embryos still evident at 9 h, 3 h after return to air. In embryo B the heart has returned to near-normal size by 13 h. Note also that aortic arch 4 is not open in the first drawing of each series, but is present by 7 h. Aortic arch 2 shows signs of closing by 13 h. Abbreviations: v.v., vitelline vein; v., ventricle; t.a., truncus arteriosus; 2, 3, 4, aortic arches; atrium is hidden behind ventricle.
The development of normal 3-day embryos was compared to those exposed to 4-6 % oxygen for 4-8 h and then returned to 21 % oxygen. The usual period of observation was 18 h with photographs taken every 15-30 min. Additional photographs of survivors were made at 24 h. Usually 30-40 successive photo-graphs were available for each embryo. On every photograph of a series, measurements were taken of: (1) crown-rump length, measured from the top of the mid-brain to the tip of the tail; (2) the diameter of the aorta a short distance below the aortic arches ; (3) the diameter of the aortic arches at their mid-points; (4) the width of the anterior cardinal vein at a point just before it crosses over the dorsal aorta; (5) the diameter of the anterior vitelline vein just above the head of the embryo; and (6) the diameter of the vitelline arteries and veins at a point close to the embryo, whenever separated enough to be dis-tinguishable. These points of measurements are shown in Plate 1, fig. B.
Measurements of the total area of the embryo were made on every second or third print of a series. Measurements of other blood vessels, of the angle between head and trunk and other structures also were made, but not on every embryo. Presenting all the data in detail is neither necessary nor desirable, and only the averages of the measurements made at the start of the experiment (zero hours), at 7 h, and at 13 h of the area of the embryo and diameters of several blood vessels are presented in Tables 1 and 2. Some data on other structures are included. Altogether, this study is based on 10 control and 34 treated embryos for which complete records were available, and is supplemented with observa tions on 18 additional embryos that were either younger or older than the majority, or for which only partial records were available.
Hematocrit values were obtained on blood samples drawn with glass micro-needles from the vitelline arteries of normal 3-day embryos and embryos exposed to 6 % O2 for 6 h. From the latter group the blood was drawn within 1 h following treatment. All samples were then placed in glass capillaries and spun in a Micro-hematocrit centrifuge for a standard length of time. Relative volumes of serum and packed cells were recorded. The rate of heart beat was measured by tapping a blood cell counter in synchrony with the heart.
RESULTS
Observations on normal embryos
Since the 18 h period of observation is one of considerable growth and morphogenesis, the development of treated embryos must be carefully com-pared with that of controls. Embryos would normally progress by 1-1·5 Hamburger & Hamilton (1951) stages in the first 13 h of observation. Further, the measurements of embryos at stage 17 at the onset of observations were significantly different than those of stage 18 at the start (Tables 1, 2).
Two major changes observed in the neural tube were a gradual increase in size and progressive development of cephalic flexure. Progress of cephalic flexure could be followed by measuring the angle between the head and trunk. In embryos at stage 17 this angle is approximately 100-110°; at stage 18 it is 85-90° and becomes reduced to 60-70° by stage 19 (Text-fig. 2). In the photo-graphs numerous changes in the circulatory system could be followed, such as the rapid development of the capillaries of the anterior cardinal system and steady increases in size of the aorta, anterior cardinal vein, and other embryonic and extra-embryonic blood vessels (Table 2). In the aortic arch region the appearance of the fourth arch and, somewhat later, disappearance of the second, could be followed with ease (Text-fig. 1 and Plate 1). Other structures that exhibited changes during this period of observation were the heart, limb buds, and tail bud. The allantois became visible in the prints usually toward the end of the third day.
Effects of moderate hypoxia (6 % O2) on c.R. length and diameters of aorta and anterior cardinal vein (A.C.V.) of 3-day chick embryos. Embryos A (control) and C were at Hamburger & Hamilton stage 18 at start of study, embryo B at stage 17. Photographs taken every 30 min. All measurements made on x 15 prints. The figures shown associated with the C.R. length curves are measurements of the angle between head and trunk. See text for details. Photographs of embryo B are shown in Plate 2, of embryo C in Plate 3.--- A ;, B ; —, C ; ⦿, hypoxia started; Δ, return to air; H, hematoma; E, exsanguination.
Effects of moderate hypoxia (6 % O2) on c.R. length and diameters of aorta and anterior cardinal vein (A.C.V.) of 3-day chick embryos. Embryos A (control) and C were at Hamburger & Hamilton stage 18 at start of study, embryo B at stage 17. Photographs taken every 30 min. All measurements made on x 15 prints. The figures shown associated with the C.R. length curves are measurements of the angle between head and trunk. See text for details. Photographs of embryo B are shown in Plate 2, of embryo C in Plate 3.--- A ;, B ; —, C ; ⦿, hypoxia started; Δ, return to air; H, hematoma; E, exsanguination.
An estimate of the embryo’s growth was made by measurements of the crown-rump length and of the area of the embryo on the prints. The area of all normal embryos increased 25 % in the first 7 h and 43 % by 13 h (Table 1). Despite this increase in embryo size, the crown-rump length actually decreased slowly, since the progressive development of cephalic flexure gradually brings the brain closer to the tail. The crown-rump length characteristically decreases in waves (Text-fig. 2).
Effects of moderate hypoxia (6 h at 6 % O2) on 3-day embryos
The embryos in this group were exposed to 6 % oxygen either continuously for 6 h or discontinuously, i.e. 4 h of hypoxia, 1 h of room air and another 2 h of hypoxia. The effects of the discontinuous treatment tended to be somewhat more pronounced and prolonged than those of the continuous treatment, but since these differences were relatively minor, all the embryos in these two groups were considered together. Good records on eight embryos at stage 17 at onset of the experiment and on 11 embryos at stage 18 were obtained. Of these 19 embryos, six (32%) died within 24 h after the treatment was started; most of the remainder survived to the fifth day when they were sacrificed. Some were allowed to live until the seventh or eighth day of development. Although a slight degree of retardation in growth and development was apparent in the treated embryos within the first 24 h, this difference usually was still not apparent by the fifth day.
Changes in size and shape
A pronounced swelling of the entire embryo was a conspicuous and consistent effect of the treatment. This was apparent not only on casual inspection of the photographs (Plates 2 and 3), but also from measurements of area (Table 1). The average increase in area of all 6 h treated embryos over the first 7 h of observation was 41 or 57 % (according to age) compared to 25 % in the con-trols. By 13 h the experimental embryos were nearly normal in size. As measured from prints, it is apparent that hypoxia produces a transient but significant increase in the area of the embryo which, in turn, reflects a volume increase in the embryo. This increase can be detected 30 min to 1 h after the treatment begins. It usually reaches a maximum shortly after the treatment is terminated and is followed by a gradual return to normal. Occasional embryos, however, continued in the edematous state for as long as 5 or 10 h after termination of treatment.
The crown-rump length of experimental embryos always increases approxi-mately 10-15 % in the first 2 h of hypoxia, remains constant, and then starts to diminish after the treatment is ended (Text-fig. 2). This increase, which is in marked contrast to the steady decrease found in normal embryos, may be due partly to the general size increase of treated embryos. To a greater extent, how-ever, it is caused by transitory reversal of normal flexion movements, since these changes in C.R. length are closely paralleled by changes in the angle between the trunk and head. In the control embryos this angle gradually becomes reduced as flexion progresses. In hypoxia-treated embryos, the angle increases by as much as 10° during the first or second hour of treatment and normally does not begin to decrease again until treatment is terminated (Text-fig. 2). This ‘straightening’ is also seen in the photographs (Plates 2 and 3). In a few embryos measurements of the area of the forebrain and mid-brain were made. In the treated embryos this area increased or decreased along with the over-all increase or decrease of the rest of the embryo, indicating that hypoxia also induces swelling within the neural tube. Since the neural tube at this stage is a closed system, it would seem likely that this transitory reversal of normal flexion move-ments is due to an increase in turgidity within the distended neural tube. Whether this temporary interference with a normal morphogenetic movement can have any permanent effect has not yet been ascertained.
Effects of hypoxia on the circulatory system
The most conspicuous feature of these observations was the dramatic increase in the size of the heart and diameter of the embryonic blood vessels (Text-fig. 1 ; Plates 2 and 3). Although the over-all size increase in hypoxia-treated embryos was only 16-33% above normal, the major embryonic blood vessels always increased to double, triple, or in some individual cases, even quadruple their initial diameter. Extra-embryonic vessels were only slightly affected. These increases were first detectable 0·5-1 h after treatment was started and continued for the duration of exposure to hypoxia. Return to 21 % oxygen usually, though not always, started a gradual but irregular return to normal diameter (Text-fig. 2).
The aorta, third aortic arch, and the anterior cardinal vein in normal embryos increase in diameter approximately 25 % in the first 7 h of observation, and another 25 % by 13 h. On the other hand, the average increase of the aorta in the hypoxia-treated embryos was 73 % in the first 7 h (Table 2). In some individuals the aorta increased as much as times its original diameter. The caudal aortae, which are clearly seen in the photographs, showed proportional increases (Plates 2 and 3). By 13 h these vessels usually returned to nearly normal diameter (Table 2).
The third aortic arch is relatively stable in contrast to the second arch, which normally disappears. It is a small artery compared to the aorta, but displays the same pattern, i.e. a maximum increase at 7 h and a return to normal diameter at 13 h (Table 2; Text-fig. 1). The average increase of this vessel over the first 7 h was from to 3 times its initial diameter.
The anterior cardinal vein is very sensitive to hypoxia, especially its large sinus dorsolateral to the aorta. The average increase in this vessel was 3 times normal and, in some individuals, this vein increased fivefold over the first 7 h of observation (Table 2). The capillaries of the anterior cardinal system also became engorged (Plate 3).
The extra-embryonic vessels on the other hand, did not follow the same pattern of change as the embryonic vessels. The major vitelline arteries and veins showed a slight to moderate degree of distention, but accurate measurements were seldom possible because these vessels overlap each other. The best measure-ments were made on the relatively small anterior vitelline vein, and this vessel showed no significant change in diameter during hypoxia beyond that found in the controls. Perhaps the thicker, and presumably stronger, walls of the extra-embryonic vessels help to prevent their distention. However, the embryonic (terminal) portion of the vitelline vein showed considerable distention (Text-fig. 1).
The rapid increases in blood vessel size, which are not accompanied by any corresponding decrease in size of other vessels, demonstrate that exposure to moderate hypoxia produces a sharp increase in blood volume (hypervolemia). This was further checked by measuring changes in hematocrit. In normal 3-day embryos, the blood cells composed an average of 22 % of the total volume of blood (8 determinations, range 20·0-26·6 %). In blood samples obtained from embryos that had just been exposed to 6 % O2 for 6 h, this hematocrit value had decreased to an average of 17% (9 determinations, range 15·7-22-0%). As-suming that the number and size of red blood cells has remained constant in the embryo during the 6 h of treatment, the fluid portion of the blood would have to increase by at least 37% to account for this increase in the relative volume of red blood cells. (See Grabowski, 1966, p. 202, for explanation of calculations. In that previous study a plasma volume increase of 60 % was found in 5-day embryos exposed to 10 % O2 for 5 h.) It should be noted that extra-embryonic vessels expand very little, so that most of this increase in blood volume would be contained within the embryonic blood vessels. This hypoxia-induced hypervolemia is a reversible effect since return to 21 % oxygen usually starts a gradual return to normal blood vessel diameter, and, presumably, to normal blood volume.
Hemorrhage was the most common visible result of exposure to hypoxia. Of the embryos exposed to 6 h of hypoxia, 8 out of 19 bled to some significant degree. Typically, hemorrhage does not begin until several hours after termina-tion of treatment (Text-fig. 2). The most common site of bleeding in the embryo was from capillaries of the anterior cardinal vein in the brain region (Plate 3). Almost as frequent was the occurrence of bleeding in and near the tail and extremities, where the escaped blood usually formed distinct hematomas (Plates 2 and 3). Such hematomas are teratogenically important since, if they persist, they can lead to the abnormal development of adjacent structures (Gluecksohn-Schoenheimer, 1945; Jost, 1951; Giroud, Lefebvres, Prost & Dupuis, 1955; Grabowski, 1964). Some of the hematomas formed abruptly and remained constant in size (Plate 2, fig. C); others gradually increased in size (Plate 3). Not infrequently, hematomas and areas of diffuse bleeding were reabsorbed. Of the 19 embryos treated for 6 h, five developed extensive bleeding at embryonic or extra-embryonic sites, resulting in exsanguination and death (Plate 2, fig. D). Such bleeding was the most common cause of death following exposure to moderate hypoxia.
Conspicuous increases in the size of the heart accompanied the blood vessel changes in treated embryos. This change is shown in the silhouette tracings of Text-fig. 1 in which several hearts are shown at the same stage of contraction. Exposure to hypoxia also affects the rate of heart beat. The normal rate in 3-day embryos is 164 beats/min (12 cases, S.D. 19-4). After 1 h of hypoxia, the rate in these same embryos decreased to 126 beats/min. (S.D. 11-9; decrease significant at 1 % level, t/-test). During the last 3 h of treatment, the average rate was 121 beats/min (S.D. 20-9) and in some embryos it dropped to 80/min. On return to 21% O2 the rate immediately increased in every embryo, and after h in a normal atmosphere the rate increased to 174/min (S.D. 24-7). During treatment the rate of heart beat in some embryos became erratic, alternating between periods of fast and slow rates. Cardiac arrest sometimes occurred during the last 2 h of treatment, usually followed after 5-30 s by spontaneous recovery. Atrio-ventricular arhythmias (e.g. three beats of atrium to one of the ventricle) were occasionally observed during treatment. It is evident that the heart is profoundly affected in several different ways by exposure to moderate hypoxia.
Specific case histories
The foregoing generalizations can be illustrated by two case histories. Embryo B, solid Une in Text-fig. 2 (see also Plate 2), was exposed to 6 % O2 for 6 h. A rapid increase in C.R. length accompanied the 5° increase in the angle between head and trunk. After a 45 min lag, the size of the anterior cardinal vein gradu-ally increased times, the aorta almost twofold (Plate 2, fig. B). Blood vessel diameter, C.R. length, and the angle between head and trunk all began to decrease as soon as the embryo was returned to 21 % oxygen, indicating a restoration of the homeostatic mechanisms for maintenance of fluid balance. However, the decline in blood vessel diameter was irregular and at
h a moderately large (0-5 mm) hematoma abruptly appeared in the tail (Plate 2, fig. C). At
h a large extra-embryonic vessel ruptured and rapid exsanguination occurred (Plate 2, fig. D). Although most blood vessels collapsed at this time, the tail hematoma persisted.
Embryo C, the dotted line in Text-fig. 2 (see also Plate 3), was exposed to 6 % oxygen for 4 h, returned to air for 1 h, and then exposed to hypoxia for 2 h. The parameters measured reflected this pattern inasmuch as they showed a slight dip after 4h and then continued to increase again after return to hypoxia. However, the increases continued for several hours after the second treatment was concluded, and return to normal size did not begin until 10 h after observa-tion had begun. At this time the anterior cardinal vein was 4 times larger than at the start of the experiment and the aorta times larger (Plate 3, fig. B). Bleeding over the forebrain occurred at 6 h (Plate 3, fig. B) and over the midbrain at 13 h. Small hematomas appeared simultaneously in the right wing and leg buds at
h. Unlike the tail hematoma in embryo B, which appeared abruptly, these started small and gradually increased in size. The wing hematoma was resorbed by 18 h, but the leg hematoma at that time was 0-35 mm in diameter (Plate 3, fig. C). The embryo died a few hours later.
Effects of various other degrees of hypoxia on the chick embryo
Fifteen other cases were studied. These included five embryos exposed to 6 % oxygen for 8 h, six exposed to 4 % oxygen for 6 h, and four embryos exposed to pure nitrogen for 4 h. Of the embryos exposed to 6 % oxygen for 8 h, two died at 8 and 9 h from extensive extra-embryonic bleeding. The other four survived without any apparent ill effects. In general, the distention of the embryo and blood vessels was comparable to, but somewhat less pronounced than, that of embryos exposed for 6 h to 6 % oxygen (Table 2). One embryo of this series deserves special mention. In this specimen the over-all size increase was main-tained for 7 h after the treatment ceased. Blood vessel distention persisted and the vessels continued to increase in diameter up to 15 h. For example, the anterior cardinal vein showed dimensions of 0·07 mm at the start, 0·27 mm at 7 h and 0·40 mm at 13 h. Similarly, the third aortic arch showed measurements of, respectively, 0·10, 0·20 and 0·27 mm. At 15 h there were numerous small hemor-rhages over the surface of the embryo. The smaller blood vessels had irregular outlines and the blood within them was dark in appearance, more characteristic of a dead rather than a living embryo. However, by the next morning, apparently complete recovery had occurred, and the embryo lived until day 7 when it was sacrificed and examined. No anomalies were apparent. This embryo and several similar, though less dramatic, cases vividly illustrate the recovery power of embryos exposed to a normally teratogenic situation.
Four of the six embryos exposed to 4 % oxygen for 6 h died, three from extensive hemorrhage. The two survivors of the treatment showed no apparent ill effects. For the most part, only moderate swelling of the embryo and its blood vessels occurred. In one specimen, however, the anterior cardinal vein expanded to times its normal width by the end of the treatment.
Four embryos were exposed to pure nitrogen for 4 h. Three of these died within the first 2 h of treatment, all by exsanguination caused by embryonic or extra-embryonic bleeding. This bleeding was preceded by only slight or moderate swelling of the blood vessels. The single survivor of this series showed moderate increase in blood vessel size, i.e. a twofold increase in the size of the anterior cardinal vein. The C.R. length increased from 95 to 115 mm at 4 h and then began to decrease. This embryo lived for 30 h after the treatment then died without any obvious cause.
These experiments showed that maximum swelling of the 3-day embryo and its blood vessels was usually achieved with an exposure of 6 h to 6% oxygen. Exposure to more severe conditions of hypoxia can precipitate a variable amount of swelling, embryonic bleeding, and death.
DISCUSSION
On the basis of the changes induced by moderate hypoxia on the chemistry of the blood plasma and a consideration of the osmotic relationships of the embryo to its surrounding environments, Grabowski (1966) concluded that: (1) the young chick embryo has an osmoregulatory problem, and (2) exposure to moderate hypoxia interferes with the osmoregulatory activity of the embryo, producing swelling and concomitant ionic changes in the blood stream. The teratological consequences of these fluid imbalances have been described to some extent and the entire sequence referred to as ‘the edema syndrome’ (Grabowski, 1964). The present study illustrates the initial stages of this edema syndrome on living specimens. Also revealed are some features previously unsuspected, such as turgidity effects on flexion, and the extent to which the embryonic heart and blood vessels are affected.
Since the major embryonic vessels increase in diameter from two-to threefold, it is apparent that the volume of blood must also increase several-fold during exposure to hypoxia. Hematocrit data support this conclusion. This extreme hypervolemia may be highly significant to the development of the embryo, since the cardiovascular system is being molded at this time and hemodynamic factors are generally considered of primary importance to the process (Jaffee, 1965, 1966; Rychter, 1962). Some vascular anomalies following hypoxia have been reported (Tedeschi & Ingalls, 1956; Grabowski & Paar, 1958) but these are difficult to detect unless specifically sought, although the present study suggests they may be more numerous than heretofore suspected. We have examined our photographs to see if any abnormal changes in vascular pattern could be detected as a result of the distention of these vessels. The second aortic arch, which normally disappears during the period under observation (Text-fig. 1), was studied to see if the hypoxia-induced distention might have delayed its closure. Careful scrutiny of many records suggests that in some cases closure of this arch may be delayed by as much as 2 h. But, at best, this is a slight effect difficult to estabfish. In embryos exposed to hypoxia at stages 18 and 19, closure of the second arch occurred even while this vessel was distended 2 to 3 times normal size. In such cases the closure was more abrupt than in control embryos. Even though this particular attempt to find a permanent effect of hypoxia-induced distention on a blood vessel was not successful, a combination of hypoxia and time-lapse photography may prove useful in studying the signifi-cance of hemodynamic factors in the moulding of other parts of the cardio-vascular system.
Persistent hematomas in embryonic tissues can cause abnormal development in adjacent structures (Gluecksohn-Schoenheimer, 1945; Jost, 1951; Wadding-ton & Carter, 1953; Giroud et al. 1955; Grabowski, 1964). The initial objective of these time-lapse studies was to determine, if possible, how and why these hematomas form as a consequence of hypoxia. It seems reasonable to assume that the hemorrhage is a result of the distention of the blood vessels beyond their elastic limits, but the correlation is not precise. Some embryos with great enlargement of blood vessels seemed to recover completely, and other embryos with only moderately distended vessels bled to death. It is apparent that more than simple distention of vessels is involved in the rupturing of blood vessels. One possibility is that lack of oxygen may cause degeneration of the vascular epithelium. The high incidence of hemorrhage, preceded by only moderate swelling, in embryos exposed to pure nitrogen would support this notion. Another clue may come from the puzzling observation that most hemorrhage occurs several hours after treatment (Text-fig. 2; see also Grabowski, 1964). It usually occurs not when blood vessels are maximally distended, but as they are returning to normal size. A possible explanation of this phenomenon is emerging from blood-pressure studies. Toben (1967) measured the mean ventricular blood pressure in normal 5-day chick embryos and in embryos exposed to 10% oxygen for 5h. The normal level was equivalent to 20-1 mm of water. An increased blood pressure (to 22-5 mm water) was noted immediately following treatment and a return to normal levels 5 h afterward. But the maximum increase (to 25-2 mm of water) did not occur until 2 h following treatment. We have recently measured the mean ventricular blood pressure in 3-day chicks. Moderate hypoxia (6 h at 6 %) increases the blood pressure from a normal level of 11-6 mm of water to an average of 16-5 mm during the second hour after treatment, with the pressure in some embryos reaching 25-27 mm of water. The relationship between the tension (T) in a blood vessel wall, the radius (r) of the vessel, and the pressure (P) within it, is expressed by the formula of Laplace (1841): T = P×r. Since both P and r can increase twofold in some embryos during exposure to moderate hypoxia, the tension within the vessel walls can increase fourfold. It is possible, therefore, that all three factors, namely, (1) blood vessel distention, (2) degenerative effects on blood vessel walls, and (3) increased blood pressure, play a combined role in this teratogenically important process of hemorrhage and hematoma formation.
The various ways in which hypoxia affects the embryonic heart—distention, changes in rate of contraction, arhythmias, and arrest—are interesting from the physiological standpoint. Some of these effects are probably related to the increased levels of serum potassium, another consequence of exposure to hypoxia (Grabowski, 1963, 1966). The gross distention of the heart (Text-fig. 1) during a period of rapid morphogenesis also raises the question of whether or not there may be permanent effects on heart structure.
The rapid increase in the volume of the neural tube has been correlated with a temporary reversal of normal flexion movements. These could be the result of an increase in the volume of cerebrospinal fluid. Were such an increase to persist, it could theoretically lead to a hydrocephalus-like condition. Such distended neural tubes may be caused by other agents that produce swelling in embryos, such as trypan blue in the mouse (Waddington & Carter, 1953; Turbow, 1966). It is also feasible that this distention of the neural tube could lead to its rupturing just as distention sometimes leads to rupture in blood vessels. On three occasions in this study a sudden reduction in the size of distended neural tubes was ob-served, particularly in the forebrain-midbrain areas. This sudden reduction was clearly localized in the neural tube, and was not seen in other parts of the embryo. Presumably it was caused by the rupture of the neural wall and loss of cerebrospinal fluid, although this could not be established by subsequent examination of the embryo. However, in a different experiment, an embryo treated with dimethylsulfoxide (DMSO) at 4 days displayed a marked, persistent distention of the neural tube. Thirty-six hours after treatment the embryo was preserved and a jagged wound was visible on the left side of a deflated midbrain, clearly a ruptured brain wall.
This study also illustrates another important aspect of embryonic life not always considered in teratological studies, recuperative power. Despite mani-festations of physiological and morphological stress in every treated embryo (swelling, straightening of neural tube, distention of major blood vessels, hyper-volemia), almost half of them apparently recovered completely by 24 h after the start of the experiment.
This study shows that the initial stages of teratogenic action can be examined by a direct, continuous observation of embryos in their natural environment. The modified time-lapse technique used here has several advantages over con-ventional techniques based on the use of movie film, in that (1) detailed measure-ments on a variety of structures are considerably easier to obtain, (2) the cost of film and equipment is a fraction of that needed for conventional techniques, and (3) in the low magnification range, the precision lenses commercially avail-able for 35 mm cameras are easier to use than low power photomicrographic equipment.
SUMMARY
The visible reactions of 3-day chick embryos (in ovó) to moderate hypoxia (mostly 6 % O2 for 6 h) were quantitatively studied, utilizing a modified time-lapse technique. These observations were then correlated with the known tera-togenic and lethal effects of this treatment.
Considerable swelling of the entire embryo occurs during treatment, followed by a gradual, but irregular, return to normal size over several hours after return to 21 % O2. Swelling of the neural tube also occurs and apparently the increased turgidity within the tube results in a temporary reversal of normal flexure movements.
The major embryonic blood vessels as well as the heart itself become reversibly distended 2-3 times normal size. This distention suggests the develop-ment of a transient increase in blood volume, which was confirmed by a study of changes in hematocrit. An apparent result of blood vessel distention is their rupturing which, in turn, leads to the development of teratogenically significant hematomas and, sometimes, death by exsanguination. Possible effects of this hemodynamic disturbance on blood vessel formation and heart development are considered.
Despite manifestations of physiological and morphological stress in all treated embryos (swelling, hypervolemia, reversal of cephalic flexure, etc.) about half of them recover within 24 h after the experiment, illustrating the recuperative capacity of embryos exposed to a potentially teratogenic treatment.
RÉSUMÉ
Etude sur la séquence photographique d’embryons de poulet soumis aux doses tératogènes de l’hypoxie.
Les auteurs étudient quantitativement les effets visibles provoqués par une hypoxie modérée (6 % O2 pendant 6 h dans la plupart des cas) sur des embryons de poulet Ågés de 3 jours, in ovo, à l’aide d’une technique modifiée de séquence photographique. Ils comparent ces observations avec les effets tératogènes et létaux connus, obtenus après ce traitement.
On observe un gonflement de l’embryon entier pendant le traitement; plusieurs heures après retour à 21 % d’oxygène, l’embryon reprend progressive-ment sa taille normale. On constate également un gonflement du tube nerveux ; la turgescence accrue à l’intérieur du tube résulte d’une inversion temporaire des mouvements normaux de flexion.
La plupart des vaisseaux sanguins ainsi que le cœur se dilatent de manière réversible et atteignent 2 à 3 fois leur taille normale. Cette dilatation suggère une augmentation transitoire du volume sanguin; cette hypothèse est confirmée par l’étude à l’hématocrite des modifications visibles. La dilatation des vaisseaux provoque leur rupture, cause de l’apparition d’hématomes tératologiques signi-ficatifs et parfois de la mort par exsanguination. On envisage les effets possibles de cette altération hémodynamique sur la formation des vaisseaux sanguins et le développement du cœur.
Malgré les manifestations d’agressions morphologiques ou physiologiques observées chez tous les embryons traités (gonflement, augmentation du volume sanguin, inversion de la flexion céphalique), la moitié d’entre eux récupèrent durant les 24 h qui suivent le traitement. Ces résultats montrent la capacité de récupération des embryons soumis à un traitement potentiellement tératogène.
ACKNOWLEDGEMENTS
This work was generously supported by grant HD 00641 from the National Institute of Health, National Institute of Child Health and Human Development. We gratefully acknow-ledge the technical assistance of Miss M. Milan, Mrs J. S. Bennett, and the’night shift’ of J. Browne and N. Chernoff.
REFERENCES
Plate 1
Time sequence photos of a normal embryo, at stage 17 at start of observations, taken (A) at start (0 time); (B) at h; (C) 14 h; (D) 23 h. The arrows in (B) indicate the points at which blood vessel measurements were made. Note particularly the clarity of the vascular system, the increase in size of the embryo and its blood vessels and the progress of cephalic flexure.
Time sequence photos of a normal embryo, at stage 17 at start of observations, taken (A) at start (0 time); (B) at h; (C) 14 h; (D) 23 h. The arrows in (B) indicate the points at which blood vessel measurements were made. Note particularly the clarity of the vascular system, the increase in size of the embryo and its blood vessels and the progress of cephalic flexure.
Plate 2
Time sequence photos of a chick embryo exposed to 6 % O2 for 6 h and then returned to air. Photos taken at (A) start of experiment, (B) 6 h, (C) h, (D) 12 h. See text for explanation (embryo ‘B’, p. 356). Note particularly the large changes between (A) and (B), and compare with Plate 1, figs. A and B. Note also the development of the tail hematoma in (C) and extra-embryonic bleeding in (D).
Time sequence photos of a chick embryo exposed to 6 % O2 for 6 h and then returned to air. Photos taken at (A) start of experiment, (B) 6 h, (C) h, (D) 12 h. See text for explanation (embryo ‘B’, p. 356). Note particularly the large changes between (A) and (B), and compare with Plate 1, figs. A and B. Note also the development of the tail hematoma in (C) and extra-embryonic bleeding in (D).
Plate 3
Time sequence photos of an embryo exposed to 6 % O2 for 6 h (discontinuously). (A) Taken at start of experiment; (B) at 7 h; (C) at 18 h. This is embryo ‘C’ of p. 356. Note particularly the increase in size of embryo and blood vessels, ‘straightening’, and hemorrhage over forebrain in ‘B’ and the hematoma in right leg bud in ‘C’.
Time sequence photos of an embryo exposed to 6 % O2 for 6 h (discontinuously). (A) Taken at start of experiment; (B) at 7 h; (C) at 18 h. This is embryo ‘C’ of p. 356. Note particularly the increase in size of embryo and blood vessels, ‘straightening’, and hemorrhage over forebrain in ‘B’ and the hematoma in right leg bud in ‘C’.