Studies on the Comparative Physiology of Digestion: III.—Secretion, Digestion, and Assimilation in the Gut of Ciona Intestinalis

This research has been carried out in order to extend our extremely scanty knowledge regarding the physiology of digestion in the Tunicate. The mechanism of feeding in these animals has already been thoroughly examined by Orton ³  and his results have been entirely confirmed. Material was obtained and all experimental work performed at the Zoological Station at Naples during the early months of 1924. The writer wishes to express his thanks to the British Association for the privilege of occupying their table, and to Dr Lancelot Hagen for kindly reading through the MS. of this paper. The work was completed in the Zoological Department of the University of Edinburgh.

The expenses of the research have been defrayed by grants from the Earl of Moray fund of the University of Edinburgh and from the Government Grant Committee of the Royal Society.

Roule⁴  has described and figured the anatomy and histology of the alimentary tract of Ciona intestinalis. It is necessary only to complete Roule’s description.

The best histological results were obtained by fixing material with Rouen’s fluid, cutting sections 6 μ thick and staining with Heidenheimer’s iron-hæmatoxylin or Delafield’s hæmatoxylin and erythrosine for general purposes and with mucicarmine for the detection of mucus.

The œsophagus is a narrow tube about 1 cm. in length leading from the branchial sac into the stomach. It is lined internally with a cylindrical epithelium consisting of high narrow cells, and is thrown into a series of grooves two of which have distinct histological characters. One of them, the continuation of the dorsal lamina, possesses cells up to 64 μ in height which bear long cilia of about half that length. Passing away from this region the cells and cilia are both smaller white mucous glands, which carry rounded vesicles of mucus at their outer extremities, are increasingly developed. Finally, there is a groove opposite to the one already mentioned, which represents the continuation of the endostyle. Here the cells are only some 30 μ high, there are no cilia, and mucous glands are everywhere present. Cell-inclusions in the form of granules which stain darkly with osmic acid are found in the longer epithelial cells, but are absent in the shorter ones.

Cilia and mucous glands are better developed in the oesophagus than in any other part of the gut. There is no evidence of any secretion of digestive enzymes in this region, and the sole function of the œsophagus is that of conducting the mucus-laden strings of food from the dorsal lamina into the stomach.

Unlike many tunicates, Ciona possesses no-“liver” or stomach cæca of any kind. The so-called pyloric gland, a mass of fine tubules which ramify over the outer surface of the stomach and mid-gut, has been considered a digestive gland, but Route, by injecting coloured fluids into it, found that it did not open into the stomach but into the heart and must,, therefore, be considered as a part of the circulatory system.

Apparently the stomach performs the double functions of stomach and digestive gland. The stomach wall is thrown into a series of furrows and ridges, the epithelium varying in height from 57 μ in the furrows to a third of this on the ridges. Cilia are present, particularly in the grooves where they may reach a length of 11 μ, but they are shorter on the sides of the ridges and absent on their summits.

Roule distinguished two types of cells in the epithelium—ordinary epithelial cells which contain many yellow granules which reduce osmic acid and which he thought consisted of cholesterol and bile acids ; and occasional “cellules calicinales “which he described as containing mucus and occurring chiefly, overhung by the other epithelial cells, in depressions in the ridges.

Undifferentiated epithelial cells (fig. 1, A.C.) which contain granules which are blackened by osmic acid certainly form the mass of the epithelium, but tests for cholesterol and bile acids are invariably negative. Scattered among these cells are others (S.C.) which contain a secretion, and are also to be distinguished by the character of their nuclei which are large and contain a single conspicuous mass of chromatin, whereas those of the other cells are small and contain only a few specks of chromatin. The secretion is often found being discharged into the lumen of the stomach, and there seems no reason to doubt that those are the secretory cells which elaborate the digestive enzymes. They occur equally in the furrows and on the ridges. Small depressions in the epithelium are not uncommon, especially on the ridges, and contain small colourless cells and occasionally, though by no means always, secretory cells. These depressions are probably the site of the formation of young cells to replace those which are destroyed.

It is difficult to discover exactly what Roule means by his “cellules calicinales.” There are no mucous cells in the stomach and the description he gives does not apply to the secretory cells.

Large round cells are also occasionally found in the epithelium, but as they are particularly characteristic of the mid-gut thev will be described later.

This region is distinguished by the presence of testicular acini which ramify within its walls and occupy the interior of the typhlosole (or testicular ridge as Roule prefers to call it) which runs the whole length of the mid-gut. The epithelium varies in thickness between 30 and 60 μ and possesses cilia which are most developed (12 μ) at the base of the typhlo-sole, and are either absent or very slightly developed (it is difficult to say which) in the regions remote from this.

Three types of cell (all of them shown in fig. 2) have been distinguished in the epithelium: 1. Absorption cells (A.C.) which contain round nuclei and vacuolated protoplasm. These cells form the bulk of the epithelium. 2. Mucous cells (M.C.) which are very narrow and contain a mass of mucus at their outer ends and whose protoplasm has a granular appearance. Their nuclei are elongated and are smaller than those of the absorption cells. Like the cilia they are most developed in the channels which lie on either side of the typhlosole and along which the bulk of the food-stream passes; 3. Large round or elliptical cells (G.C.) which lie against the basement membrane but are usually (though not always) covered by overlapping epithelial cells on their outer surfaces so that they have no free surface projecting into the lumen. Their average diameter is about 24 μ. Their cytoplasm is granular and stains darkly with hæmatoxylin while within the nuclear membrane lies a single round mass of chromatin some 4μ in diameter. After staining with iodine or Best’s carmine these cells take on a dark brown or red coloration respectively—in vivid contrast to the remainder of the section which remains colourless—thus showing the presence of large deposits of glycogen. As will be shown later the mid-gut and stomach in the order named are the absorptive regions of the gut and it is in them alone that these glycogen cells are found, and much more frequently in the mid-gut than in the stomach. All the evidence available, therefore, points to their playing an important part in the assimilation, storage, and metabolism generally of carbo-hydrates.

In this connection it is interesting to note the work of Wagner (quoted by Seeliger⁶ ) on Cynthia echinata. He found in the “liver,” and to a less extent in the stomach, very large cells which contained clear concentrically striated starch granules which stained blue with iodine and which he was quite convinced developed in the epithelial cells. Apparently, there-fore, Cynthia has a similar mode of carbohydrate storage to Ciona, but that in that case the “liver “is the principal site of assimilation and starch the form in which carbohydrate is stored.

As in the stomach there are clusters of short colourless cells in the epithelium of the mid-gut.

In cross section the hind-gut is not unlike the mid-gut, but the walls are more folded and the cavity of the typhlosole is no longer occupied by testicular acini but by the sexual ducts. This is the only region of the gut, as Roule has pointed out, in which muscle fibres are present; they occur as occasional longitudinal bundles and doubtless aid deæecation.

The epithelium (fig. 3) consists of low cells, everywhere ciliated, which possess a uniform height of about 30 μ Sections stained with mucicarmine show that mucus is secreted by nearly all the cells, and as there is no apparent difference (such as is seen in the mid-gut) between those which secrete and those which do not, possibly the latter represent cells which are temporarily inactive. There are no glycogen cells but instead there are occasional large epithelial cells (S.C.) which secrete a very darkly staining mass which is often found being dis-charged into the lumen. The secretion is not stained red with mucicarmine and the cells are quite unlike the secretory cells of the stomach. The nature and action of this secretion have not been ascertained.

Krukenburg (quoted by Jordan¹  and Seeliger⁶ ) found a very weak “tryptic” enzyme and an amylase in the extract of the gut of Ciona canina. Owing to the weakness of the proteolytic enzyme he thought the feeding processes of ascidians represented a connecting link between the exclusively intra-cellular and the enzymatic secretory processes. There is, however, no trace of intracellular digestion in Ciona.

Extracts of the gut of Ciona were prepared by grinding the tissue with sand and extracting for 3 days with sea-water or distilled water to which toluol had been added. When testing for amylolytic or lipolytic enzymes 15 per cent, extracts were employed, but on account of the weakness of the proteo-lytic enzyme 25 per cent, extracts were employed in testing for this. Rigorous controls consisting of boiled extracts were set up. Unless otherwise stated, all digests were incubated at a temperature of 35° C.

Table I. gives the results of the experiments on carbohydrate digestion.

It will be seen that the only carbohydrates which are digested are starch, glycogen, and sucrose. There is no action on inulin, raffinose, cellulose, maltose, or lactose.

Experiments to determine the temperature of destruction of the amylolytic enzyme revealed the fact that whereas 10 c.c. of extract which had been heated to 63° C. for 25 minutes was able to digest starch after an incubation at 35° C. for 18 hours, an extract which had been heated at 64° C. for the same period was not able to do so. The enzyme is apparently destroyed at the latter temperature.

A series of tests was made to determine the optimum temperature of the amylolytic enzyme. In each case 10 c.c. extract and 10 c.c. of 1 per cent, starch were taken and the mixture stood for $212$ hours. Each digest was then boiled, filtered, made up to exactly 20 c.c. and titrated into 10 c.c. of Benedict’s solution. The following table gives the results :—

The result is expressed in the form of a graph in fig. 4 and the optimum temperature shown to lie between 42 and 43° C.

Table II gives the results of experiments on fat digestion. Methyl acetate and, to a less extent, an emulsion of olive oil were both split up after a lengthy incubation and the butyrin in milk was converted into butyric acid.

The proteolytic enzyme is very weak. Small flakes of fibrin were dissolved in neutral and alkaline media (never in acid media) after 5 days, and the biuret reaction revealed traces of albuminoses after 8 days’ incubation, but even after 14 days no trace of tryptophane or tyrosin could be found. Calcified milk was coagulated within two hours, the control experiment showing no change.

Attempts were made to find the optimum medium for the enzyme, but this was almost impossible owing to its weakness. 70 c.c. of a 25 per cent, extract in distilled water was divided into 7 equal parts and 10 c.c. of a 5 per cent, casein solution added to each. Distilled water and caustic soda were then added so that all were made up to 30 c.c. and the concentrations arranged thus: N/10, N/20, N/30, N/40, N/50, N/80 NaOH, and neutral. All were incubated at 35° C. and 2 c.c. removed every day and treated with 2 c.c. of 25 per cent, acetic acid so as to show to what extent the casein had been dissolved. It was, however, impossible to distinguish any difference in the rate of dissolution in the concentrations between neutral and N/30 NaOH, but it was certainly not so rapid at the two highest concentrations of alkali. The same results were obtained when equal flakes of fibrin were substituted for casein.

The action of the extract was also tested on the glucosides, salicin and amygdalin. The results of the experiments in Table III. show that both were split up by the extract.

It was Seeliger’s⁶  opinion that the stomach with its folds plays the chief part in both absorption and secretion, although he did not deny that the mid-gut may take part in both to a smaller degree. He advanced no experimental proof, however, of this statement. The only person who has worked on absorption in the tunicates is Schneider,⁵  who found that iron was absorbed under ordinary living conditions in both gut and stomach.

Animals were kept in sea-water containing ferrum oxydatum saccharatum in suspension and after various periods of time the gut was removed and fixed in 95 per cent alcohol containing 5 per cent, of ammonium sulphide. Sections were cut and the iron precipitated by the ferrocyanide method as described in a previous paper (Yonge⁸ ). Absorption was found in all regions of the gut though only very slightly in the œsophagus, minute particles of iron being distinguished lying within vacuoles in the epithelial cells.

In order the better to determine the relative importance for absorption of the chief regions of the gut, animals were starved for three weeks and then fed with olive oil The oil was injected into the branchial sac by means of a pipette and the animals placed in large test-tubes, which were then placed in tanks of sea-water and inverted so that the olive oil could not leave the branchial sac except by way of the gut After periods of 6, 15, 24, and 48 hours, animals were removed and the stomach, mid-gut, and hind-gut fixed with Flemming’s strong solution. Any fat which had been absorbed was then easily seen in the form of black globules. Table IV. shows the degree of absorption which was observed. The mid-gut showed by far the greatest power of absorption, 15 hours after feeding the epithelial cells being practically filled with fat. There were also traces of fat after 6 hours and 24 hours. The stomach and hind-gut only showed indications of assimilation 15 hours after feeding, the stomach much better than the hind-gut, which only showed the very faintest traces of fat.

In order to test the permeability of the gut to sea-water and to dissolved substances, a number of experiments were carried out with excised guts which were filled with various fluids, ligatured carefully at either end, and placed in glass vessels containing fluid of a known composition.

Table V. gives the results with sea-water of various concentrations. In the first three experiments the fluids within and without the gut were isotonic, in the next four the contained fluid was hypotonic to the surrounding fluid, and in the last three the contained fluid was hypertonic. The guts were dried on filter papers and weighed three times—before the experiment, after 5 hours, and again after 22 hours. It will be seen that there was an average decrease in weight of 22.7 per cent, when the fluids were isotonic, of 26.5 per cent, when the contained fluid was hypotonic, and of 7.3 per cent, when it was hypertonic. The loss in weight when the fluids were isotonic cannot be due to the action of osmosis but to the physiological action of the absorbing epithelium of the gut. The increased loss of weight when the contained fluid was hypotonic and the decreased loss when this was hypertonic show that osmotic pressure may influence the rate of absorption. Indeed in the third series of experiments two showed a decided increase in weight after 5 hours, and the other only a slight loss; but after 22 hours, however, only one showed an increase and that very slight. These results are similar to those obtained for the absorbing mid-gut of Nephrops (Yonge⁸ ), and in opposition to those obtained for the non-absorbing fore-gut of that animal, which behaves as a semi-permeable membrane. They are most easily explained on the hypothesis that adverse osmotic pressure may temporarily overcome the absorptive power of the epithelium, but rather that after a certain equilibrium has been obtained the latter regains the upper hand and fluid flows out of the gut.

Four experiments were set up to determine the permeability of the gut to dissolved substances :—

• (1) Three guts were filled with sea-water containing 1 per cent, of potassium ferrocyanide and each placed in a separate vessel of sea-water. After 3 hours the surrounding fluid was tested with ferric chloride and in all three cases a blue coloration showed that iron had passed into it.

• (2) Three guts were treated as above only a 10 per cent. solution of glucose in 50 per cent, sea-water was employed instead of the potassium ferrocyanide solution. After 3 hours the surrounding fluid was tested with Fehling’s solution and in all cases the presence of reducing sugar was shown.

• (3) Five guts containing sea-water were placed in a beaker of sea-water containing 1 per cent, of potassium ferrocyanide. After 3 hours the contents of all five gave the test for iron.

• (4) Four guts containing sea-water were placed in a beaker containing a 10 per cent, solution of glucose in 50 per cent sea-water. After 3 hours the contents of all four showed the presence of glucose.

Judging from these experiments, the gut allows dissolved substances to diffuse freely in and out of it. Here again the results are the same as those obtained for the mid-gut, and contrary to those obtained for the fore-gut, of Nephrops.

Jordan and Begemann²  state that an absorbing gut, provided it is alive, is characterised by its power to absorb iron in small vacuoles into its cells, by its independence of osmosis and by a polarisation of the transport of fluid which is always from within outwards. The gut of Ciona, like the mid-gut of Nephrops, shows the first two characteristics, but not the third, and it is impossible, therefore, to agree with these authors as to the presence of this polarity.

Ciona intestinalis stores reserve food in the form of glycogen and fat. In order to demonstrate the presence of glycogen microscopically, material was fixed with Carnoy’s fluid and the sections stained with iodine and with Best’s carmine. By these methods the presence of glycogen was demonstrated in the following places: the branchial sac, in masses under the epithelium and in the blood-lacunæ, and very occasionally in large cells; the stomach, in the glycogen cells and in small quantities in the connective tissue under the epithelium; the mid-gut in the numerous glycogen cells; the hind-gut in small amounts beneath the epithelium and in the lacunæ; and the ovary, particularly in the cytoplasm surrounding the nuclear membrane. Fat was demonstrated by sectioning material fixed in Flemming’s strong solution and was found in the epithelial cells of the œsophagus and stomach, and in the cytoplasm and surrounding test-cells of the ova.

Quantitative estimations of the amount of glycogen and fat in (1) the whole animal without the test; (2) the gut, and (3) the ovary were also made. In order to estimate the fat content a weighed quantity of wet drained tissue was dried in a steam oven, weighed, and extracted with ether in a Soxhlet apparatus for one day and the weight of the extract determined. This extract was then redissolved in ether and the lecithin precipitated with acetone. The precipitate was collected in a filter, dried, and weighed. Glycogen was estimated by Pfluger’s method. Tables VI. and VII. give the results of these estimations.

It will be seen that both gut and ovary have more than the average tissue-content of both fat (including lecithin) and glycogen, the gut possessing the greater percentage of fat and the ovary of glycogen. With regard to the glycogen estimations it must be noted that Starkenstein,⁷  who worked on Phallusia mammillata, found that up to 50 per cent of the glycogen was adsorbed by the iron hydroxide formed by the interaction of the iron, which is very plentiful in the tissues of tunicates, and the caustic potash employed to dissolve the tissues in Pflüger’s method, and that, therefore, estimations by this method must be corrected for this amount

No trace of bile constituents or lipochrome pigments was found, and both the Liebermann-Burchard and the Salkowski tests for cholesterol gave negative results.