Embryonic Growth relative to the Vitelline Membrane, and its Role in Determining the Level of Post-operative Abnormalities, as Measured in the Chick

In a paper concerned with the fate of the first somite in the chick embryo, Hinsch & Hamilton (1956) drew attention to the incidence of post-operative abnormalities encountered, namely, 14 out of 33 in one of their series and 22 out of 50 in another. The technique used by these workers was to make an opening in the vitelline membrane (V.M.) at the 5-6 somite stage, mark the first somite with carbon, and then allow development to proceed. The abnormalities observed included spina bifida (46-8 per cent.), ‘twisted embryos’, heterotaxis, and small eyes in which the lens was displaced on one or both sides of the head. They suggested that the first three abnormalities mentioned were due to the embryo pressing against the edge of the opening made in the V.M. With regard to the level at which spina bifida was observed they made the following interesting statement: ‘the defect occurred much later and in a posterior part of the tube rather than at the level of the operation, so that trauma from the marking needle would seem an unlikely explanation’. It is with this observation that the present communication is mainly concerned. Data will be presented below to demonstrate first, the changing relationship between the first somite and the overlying V.M. as the embryo grows in length, and second, how this discrepancy between the level of the operation and that of the lesion can come about. Observations concerning the backward movement of Hensen’s node are also included: these findings are in conformity with the well-known work of Wetzel (1929), Pasteels (1937), and Spratt (1947).

The experiments recorded in Tables 1 and 2 were undertaken with white leghorn chick embryos. The results are based on 37 consecutive experiments. No post-operative abnormalities were seen affecting either embryo or membranes.

In order to study the movements of the growing embryo in relation to the overlying V.M. the following very simple procedure was employed. A small puncture (01 mm. diam.) was made in the V.M. with a tungsten needle about 0·3 to 0·4 mm. either to the right of the 1st visible somite or to the right of Hensen’s node. Minimal staining with neutral red was employed in certain cases.

Having replaced the window, the egg was topped up (Grabowski, 1956) and rolled 180° on the candler. Development was allowed to proceed for a further 24 or 48 hours, i.e. to an approximate total incubation time of 48 or 72 hours, when a new window was cut over the blastoderm. The sites of V.M. puncture were located by heavy staining with Nile blue sulphate and their relationship to the embryo recorded in terms of somite levels.

It was considered desirable at the same time to measure the distances separating the somite opposite the puncture-mark in the V.M. from the 1st somite or from Hensen’s node, but in the former case, owing to the disappearance of the 1st somite, measurements were made instead to the most proximal part of the otocyst. This measurement in fact did not give an exaggerated indication of the degree to which movement had occurred owing to the flexed position of the embryo in the head region. In the case of Hensen’s node no problem arose up to the time of development of the tail fold: but once the fold had developed the node was, of course, no longer visible in ovo. Accordingly measurements were made to the most distal visible part of the tail.

The present study was not extended beyond the end of the 3rd day since the V.M. ruptures soon afterwards. In any case by this time the embryo is enveloped in the amniotic sac and the relationship of embryo and V.M. is no longer of practical importance.

The data relating to the forward movement of the 1st somite relative to the V.M. are shown in Table 1 and Text-fig. 1; those relating to the backward movement of Hensen’s node are shown in Table 2.

The first column in Table 1 shows that of 21 experiments 14 were carried out on embryos of six or seven somites, i.e. at the stage when the mid-brain folds come together; in the other seven experiments one embryo had four somites, three had eight, and three had nine.

In the second column the results have been arranged serially in each group according to the stage reached when the results were examined. If attention is directed to those embryos in which the results were examined on the day subsequent to the operation (by which time stages with from 23 to 27 somites had been reached), it will be seen that the opening made in the V.M. no longer lay opposite the 1st somite but was situated somewhere between the 10th and 14th somites (column 3). The actual distances measured from the somite opposite the opening to the otocyst varied from 1·5 mm. to 2·3 mm. (column 4). Two further cases included in the table were examined days after the experiment, and the openings were seen at the level of the 17th somite in both cases.

In the embryos examined 2 days after the operation (Hamburger & Hamilton, 1951, stages 18 or 19) the opening was situated between the 15th and 20th somites, i. e. opposite the right forelimb bud. The distances from the somite opposite the opening in the V.M. to the otocyst now ranged from 2’8 to 4 9 mm. The considerable variation in the last figures is brought about by differing degrees of head flexion, size of embryo, &.c.

Examining Table 2 in the same way we see that on the day subsequent to the operation (stages reached varying from 20 to 26 somites), the opening lay between 1 ·8 to 3 ·1 mm. cranial to Hensen’s node, separated from it by the whole of the undifferentiated paraxial mesoderm and the caudal 2 to 10 somites. On the second day subsequent to the operation (stages 17 to 19) the tail fold had formed and the node was no longer visible, but the opening lay between the 17th and 27th somites, a distance of between 1·9 and 5 mm. from the most caudal part of the tail fold.

The legend for Text-fig. 2 draws further attention to the relative movements of the Hensen’s node region and V.M.

There is no question that adhesion to the edge of the opening in the V.M., as Hinsch & Hamilton suggest, is a cardinal cause of abnormalities in embryos operated on during the first 2 days of development. Grabowski (1956) says that this opening should not exceed 0’2 mm. diameter. The point which is brought out in the present communication is the forward movement of the cranial end of the growing embryo in relation to such an opening. If an abnormality occurs due to an adhesion, the level of the abnormality depends on the time which elapses between the operation and the development of the adhesion. Thus in the case we have concentrated on (the 1st somite), an immediate adhesion would produce abnormality in the hind-brain region; an adhesion occurring the next day would cause mischief in the neck, while another day later it would produce a lesion in the thoracic region, giving rise to a scoliotic or ‘twisted embryo’, by preventing the normal processes of rotation. Text-fig. 3 exemplifies such a lesion.

But Hinsch & Hamilton also refer to abnormalities in the pelvic region. Assuming that their openings in the V.M. were not excessively large, these cannot have been due to adhesion to the opening in the V.M. in their experiments but were nevertheless almost certainly due to adhesion to the intact v.M. Naturally it will be asked how this could happen, particularly as a considerable time must have elapsed between the operation and the development of the abnormality. The following remarks may explain how this can come about. During the actual experiment, even if the whole area of exposed blastoderm is kept moist, it is impossible to avoid damaging the thin layer of albumen which normally separates the V.M. from the shell membrane. The significance of this is shown by considering the secretion of sub-blastodermal fluid (New, 1956), which takes place continuously and depends on the integrity of the layer of albumen over the V.M. from which water can be drawn. Thus, as this supra-blastodermal fluid is continuously being drawn upon, dehydration is continuously taking place with the threat of adhesion to the V.M. When the egg has been returned to the incubator it is significant that adhesion between the V.M. and the shell membrane (which invariably accompanies adhesion between V.M. and the blastoderm) does not occur for several hours, as is easily shown by candling; in fact it is not until much later when the embryo is completely enclosed in the amnion and the circulation is well established, that the danger of drying can be said to have passed. The earlier the stage of the embryo at operation the longer is the time which must elapse before the amnion is completed, and there is thus greater probability of abnormality in experiments at say 24 than 48 hours. This danger is specially relevant in discussing abnormalities in the pelvic region because the caudal part of the embryo is exposed for the greatest length of time.

It follows from the foregoing argument, in the view of the present author, that the solution of the problem of operating in ovo during the early stages, so as to avoid a high abnormality rate (see Deuchar, 1958), consists not only in providing anti-drying measures during the experiment itself but specifically in providing such measures after the egg has been returned to the incubator, to maintain the water content of the supra-blastodermal fluid. This matter will be explored in a later communication.

Lastly, when planning the site of the opening in the V.M. SO SS to avoid adhesion formation, it is useful to be able to picture in the mind’s eye the position of the embryo a day or two after the operation when considerable growth will have occurred. It will be seen from Text-fig. 1, drawings A, B, and c, that, when development proceeds normally, the anterior intestinal portal maintains approximately its relationship to the opening in the V.M., and thus provides a region of orientation.

I wish to record my thanks (1) to the Central Research Fund of the University of London for a grant for the purchase of equipment, (2) to Messrs. C. L. Jarrett and P. Runnieles, (3) to Mr. John Spurbeck for Text-fig. 2, and Miss E. D. Hewland for redrawing the other figures. I wish especially to express my indebtedness to Dr. M. E. Rawles for her guidance and encouragement over a long period, and to Professor E. W. Walls.

The subject of this paper was broached while the author was a member of the 2nd International Team at the Hubrecht Laboratorium, Utrecht. I wish to thank Dr. Pieter Nieuwkoop and his staff for their assistance at that time.