Paper - The embryogenesis of human bile capillaries and ducts

From Embryology
Revision as of 00:34, 9 June 2017 by Z8600021 (talk | contribs)
Embryology - 19 Jun 2021    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Bloom W. The embryogenesis of human bile capillaries and ducts. (1926) Amer. J Anat.

Online Editor 
Mark Hill.jpg
This historic 1926 paper by Bloom describes early liver development.

Modern Notes:

Template:Gastrointestinal links}}

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

The Embryogenesis Of Human Bile Capillaries And Ducts

William Bloom

The Snydacker Fund of the Michael Reese Hospital and the Nelson Morris Mcinorial Imtitute for Medicarl Research, Chicago, Illinois

Five Figures

The use of the Eppinger‘ method of staining bile capillaries in the study of normal and pathological human and animal livers has raised a number of questions concerning the relationship of the intralobular bile capillaries to the interlobular bile ducts. It was thought that a knowledge of the developmental relationship of these two groups of structures might throw some light on the connection between these two portions of the excretory system of the liver in such conditions where there is marked distortion of the normal configuration such as occurs in cirrhosis of the liver. A. search of the literature revealed a number of ambiguous statements regarding the embryology of the liver and some diametrically opposed views as to the origin of the duct system in this organ. It seemed advisable, therefore, to investigate the embryogenesis of the human liver, with the aid of the Eppinger method for the demonstration of the earliest appearance of bile capillaries.

In the human, according to the majority of observers, the duodenum gives rise to a hollow diverticulum which shortly develops two main branches. The caudal of these develops independently into the cystic duct and the gall bladder; the cranial elongates into the hepatic duct and becomes crowned with the hepatic bud proper. In this bud cords of embryonic liver cells develop which are separated by large endo 451


thelial-lined blood spaces. These latter will become the intrahepatic veins and the sinusoids of Minot. However, some authors claim that the hepatic cords invade the omphalomesentcric vein, thereby covering themselves with endothelium.

The mode of development of the interlobular ducts is a moot question. According to Amy,” “at eight weeks hollow interlobular ducts appear, spreading inward from the hepatic duct along the larger branches of the portal vein.” Minot?’ states, “the origin of the interlobular ducts is as yet but imperfectly understood, but we may surmise that they arise as evaginations of the hepatic duct; certain it is that they are always distinct from the hepatic cells.”

Hertwig‘ describes the network of hollow or solid liver cylinders developing in a double manner; one part is to become excretory ducts.

According to Schaferf‘ “the portal and interlobular bile ducts are formed by the hollowing out of some of the anastomosing cell-cylinders, so that a lumen is produced within them surrounded by hepatic cells, which lose their original polyhedral character, and become changed into the columnar epithelium of the ducts, the anastomoses between the cellcylinders here disappearing. The remaining cylinders form the secreting substance of the liver. The biliary canaliculi appear as minute passages between the cells and come into continuity with the bile ducts.” It is to be noted in the descriptions of both Hertwig and Schafer that the exact manner in which the liver cells change into ductal epithelium is not described.

Hering“ described the transition of interlobular ducts into live»r—ccll cords as consisting of a canal bordered on one side by ductal epithelium and 011 the other by liver cells. In his work there is a diagram illustrating this in the liver of a threemonth child. The transition, as he describes it, is very abrupt. These have since been known as the canals of Hering.

Policard7 lays great stress on the difference in appearance between the smaller and the larger interlobular ducts~—the EMBRYOGENESIS OF BILE CAPILLARIES 453

former have no cuticular membrane. Recently Von Meyen— bergs has regarded the interlobular ducts as arising in a twofold manner: the smaller ducts develop from the liver cords, while the larger periportal ducts develop from the hepatic duct. He uses this manner of the development of the ducts as an explanation of the appearance of liver cysts.

The description of the process by Lewis9 is as follows: “In an embryo of 22.8 mm. (Fig. 293, A) the spread of the bile-ducts along the main branches of the portal vein has begun. The trabeculae form cords extending along the surface of the periportal mesenchyma, and in them a lumen is formed. In places the cells on the mesenchymal side of the lumen are distinctly flatter than those toward the portal capillaries. As seen in the figure, the trabeculae connect freely with these ducts. In a later stage (29 mm., Fig. 293, B.) the mesenchyma has increased, so that it surrounds the ducts which were seen forming along its surface. Their epithelium has become regularly cuboidal or columnar. On the upper side of the vein in Fig. 293, B, the ducts are in the earlier stage of development. The periportal ducts clearly form a plexus. The larger ducts, which have become surrounded by mesenchyma, are also plexiform, although with the enlargement of the liver their anastomoses become less numerous. However, the plexiform arrangement of the main branches of the hepatic duct, which was clearly seen in a single frontal section at 29 mm., persists throughout life, as has long been known.”

Toldt and Zuckerkandl” describe the development and connection of the embryonic hepatic trabeculae with the interlobular ducts as follows: “Those hepatic trabeculae which are found in the immediate vicinity of the relatively very large portal branches, almost without exception are perpendicular to the latter, so that in cross sections of a portal branch they show a radial arrangement, and in longitudinal sections they appear in parallel rows. They open into the ducts almost at right angles, and their cuboidal cells are inserted directly into the flat epithelium of the ducts.” 454 VVILLIAM BLOOM

The origin of the bile capillaries has been incompletely studied. Toldt and Zuckerkandl, using the ordinary morphologic stains, were frequently able to demonstrate in cross—sections of the liver cords of a four—week human embryo a lumen much larger than that of the future bile capillaries. This lumen was surrounded by from three to six cells. They believed that as the embryo grew the liver cords would be stretched very much in the manner of a rubber tube, so that the number of cells in cross—section of a liver cord at birth would diminish to two and the lumen would contract to the size of the bile capillary of the adult. Hendrickson,“ using the Golgi method, was unable to demonstrate bile capillaries in the livers of human embryos shorter than 5 cm. In embryos of this size he found bile capillaries scattered through the organ, but more frequently in the parenchyma about the portal veins. In larger embryos—those from 100 mm. up—he found the meshwork of capillaries very well developed.

The various observations to date 011 the development of the excretory system in the liver may be summarized as follows: the head of the hepatic duct, after giving rise to the hepatic bud proper, divides into smaller ducts. The branches of these serve, according to some investigators, as interlobular ducts; according to others, however, the interlobular ducts develop from liver—cell cords. The method of connection of the larger branches of the hepatic duet with the small embryonic interlobular ducts is, as yet, but imperfectly understood.


The Eppinger stain for bile capillaries can be carried out using paraffin as the embedding material, so that thin sections can be cut. The method gives general cytological results somewhat similar to those obtained by Weigert’s iron hematoxylin. It was soon found that by the Eppinger method, in our hands at least, bile capillaries are not demonstrable in embryonic rat and chicken livers. The human EMBRYOGENESIS or BILE CAPILLARIES 455

material studied by us consisted of a series of well—preserved human embryos varying in size from 10 to 90 mm. in length. The whole embryos were fixed in formalin for two weeks, then placed in Weigert’s fluorochrome mordant for one week in an incubator at 37.500. They were then carefully dehydrated through a long series of graded alcohols, cleared in cedar oil, and embedded in paraffin. Serial sections of the embryos were made at 6 and 8 u. VVhen there were two embryos of the same size, one was fixed in formalin and stained with iron hematoxylin-eosin and the other stained for bile capillaries. In carrying out the bile capillary stain, a 1 per cent aqueous solution of Merck’s hematoxylin was used. The sections were stained for a few minutes in this, instead of the several hours advised by Eppinger. The method is rather capricious, but the pictures obtained are excellent when the method is successfully carried out. Some of the sections were counterstained with aqueous eosin after previous mordanting in Fle1ning’s osmic—aeid mixture.


In a 10~mm. embryo the liver is a large organ consisting of a network of cords of large polygonal cells with large, clear nuclei, in which one to three nucleoli are prominent. The cords are supported by endothelial-lined vascular spaces which contain many nucleated blood cells. There are frequent collections of nucleated blood cells in the liver cords proper; as these collections of cells disappear shortly after birth, they were erroneously described by Toldt and Zucker— kandl as a younger stage of liver cells. In sections stained by hematoxylin and eosin, in an embryo of this size, a few of the liver cords contain a large clear lumen which corresponds with the embryonic capillary described by Toldt and Zucker— kandl. These spaces which are usually seen in cross—sections of liver cords are irregular in outline and are surrounded by three to six or seven cells. These lumina can be traced through two or three sections and then disappear. In sections of a 10—mm. embryo stained by the Eppinger method 456 WILLIA M BLOOM

many of the liver cords are traversed by double—walled canals which in places branch and which, because of their situation and appearance, are morphologically indistinguishable from the bile capillaries of the adult human liver. They differ from those of the adult, however, in that in the latter the capillary is usually surrounded by two liver cells, whereas in the former it is bounded by three to six or seven cells. The liver-cell membrane at this stage is quite hazy. The large lumina described by Toldt and Zuckerkandl are seen to have a very dark cuticular membrane when demonstrated by the Eppinger method. As yet, the hepatic duct consists of a single layer of deeply staining columnar epithelium ranged about a central lumen which is bounded by a dark blue—black cuticular membrane. The duct is completely outside the hepatic pulp proper. The appearance of the capillaries will be discussed below in greater detail in connection with the 14-mm. embryo. The study of this 10—mm. embryo indicates that true bile capillaries are present at a much earlier period than is shown by the Golgi method. Moreover, the lumen described by Toldt and Zuckerkandl is present at the stage they mentioned, but is not to be considered a true bile canaliculus.

In a l4—mm. embryo the liver-cell membrane is still quite hazy. Nothing suggestive of intrahepatic bile ducts is seen. As mentioned in the description of the previous embryo, many of the liver cords in cross—section are traversed by large lumina. These spaces, which are lined by a definite black cuticula.r membrane, in Eppinger preparations, stand out ver_v sharply as independent structures, particularly when contrasted with the hazy cell membrane surrounding the remaining portions of the liver cells. Practically every liver-cell cord not hollowed out by a large lumen contains a thin, doublewalled, true bile capillary, the diameter of which is slightly larger than that of the bile capillary of the adult. In tracing these canaliculi one finds them frequently emptying into the large lumina described above in such a Way that the cuticular membrane of the large open space is continuous with the wall EMBRYOGENESIS or BILE CAPILLARIES 457

of the bile capillary. It is not unusual to find five and six bile capillaries emptying into each large lumen. The hepatic end of the hepatic duct at this stage is confined to the mesen— chyma surrounding the liver; it apparently has not, as yet, wandered into the liver parenchyma. The relationship be ABBREVIATIONS B.C., bile capillary II.D., hepatic duct B13, branch of hepatic duct I .I)., interlobular duct Br.’, branch of hepatic duct in another L.0.C., liver—cell cord level L.H.D., lumen of hepatic duct

13.17., blood vessel containing blood cells L.I.D., lumen of interlobular duct 0., change from ductal to liver-cell epi- M., mesenchyma thelium

Fig. 1 Liver from 14-mm. human embryo, showing several bile capillaries emptying into large lumina. Blood sinuses filled with young blood cells. Eppinger stain. Camera lucida. Ocular 5 with oil immersion.

tween the large lumina and the bile capillaries is shown in figures 1 and 2.

In a 16—mm. embryo the picture is the same as that seen at 14 mm. Bile capillaries and the large lumina are practically universally present throughout the liver. The hepatic duct, which still is outside of the liver, has six main branches and Fig. 2 At L.L. is a branclring large lumen i11to which empty several small bile capillaries. At R0. are several small portions of bile capillaries. Eppinger stain. Human embryo 14-. mm. Photomierograph, X 750.

Fig. 3 Connection between a branch of the hepatic duct (B12) and 9. liver-cell cord. At Br’ is the start of a branch of the hepatic duct which passes out of the level of the section. In the space C the change in the epithelium from that of hepatic parenchyma to that of the duct can be seen. The differences of the

4:38 cells on the two sides of the bile capillary are quite marked. Eppinger stain. Human 16-mm. embryo. Ph0tomicrograpl1, X 560.

Fig. 4. Bile capillary. Eppinger stain. Human 40-mm. embryo. Photomicr0graph, X 750.

Fig. 5 Clnzmgc in hepatic p:11'encl1yma as it develops into an embryonic bile duct. Note the difference in size of the cells surrounding the lumen. The large cells have not as yet been separated from the hepatic pn1'enchymu by connective tissue. Eppinger stain. Human -10-min. embryo. Pllotoniicrogmpli, X 750.


contains a very definite lumen surrounded by short, columnar epithelium. The branches run up to the liver parenchyma, but stop short with the mesenchyma surrounding the liver bud. In this embryo a particularly instructive picture is seen. Under low power the hepatic duct appears as a deeply bluestaining structure embedded in rather dense young fibrous tissue. The branches of the hepatic duct are continuous with liver—cell cords. In one of these, as shown in figure 3, the lumen of the bile duct is continuous with a characteristic bile capillary. Moreover, those cells of the duct which abut directly against connective tissue are compact and darkly staining, while the opposing cells on the other side of the lumenbile capillary—which are not in direct contact with fibrous tissue are lightly staining and polygonal in outline. In this figure the transition from the polygonal embryonic liver cells to cuboidal ductal epithelium is clearly demonstrated. This process, which is quite similar to the mode of development of the intrahepatic ducts, will be referred to again. The cytoplasm of the liver cells at this stage is quite finely granular and stains lightly; that of the ductal epithelium is much more compact and stains quite darkly by the Eppinger method. The nuclei of the ducts, too, are much darker and more compact than those of the liver—cell cords. It is to be noted that at this stage there is no connective tissue about any of the intrahepatic vessels. Nothing suggestive of the embryonic liver cysts described by Elze” and Lewis” is seen.

In embryos of 22 and 30 mm. evidences of beginning intrahepatic bile ducts appear. Sections through a 40—mm. embryo show the processes of the development of the bile ducts very nicely and will be described in some detail. The large lumina, rather inconspicuously present in younger embryos, l1a.Ve become very numerous; four or five may be found to an oil-immersion field. The bile capillaries are practically universally present and form an interlacing network closely resembling that found in adult human liver. They are still surrounded, however, by three to seven cells. The interest— ing development at this period, however, is the appearance of EMBRYOGENESIS or BILE CAPILLARIES 461

true intrahepatic ducts in the connective tissue growing into the liver about the branches of the portal vein. In the young connective tissue at the hilum of the liver, the hepatic duct, which consists of a fairly large lumen surrounded by columnar epithelium, divides into several branches. Emptying into these are many smaller ducts, each consisting of a small lumen surrounded by a darkly staining cuticular membrane, resting against which are small cubical cells. These stain quite darkly by the Eppinger method and in a low—power field are in sharp contrast to the clear cytoplasm of the large hepatic cells and the loose embryonic connective tissue. The nuclei of the cells of the duct are quite small and compact and usually one—half to one—third the diameter of the nuclei of the liver cells. As one follows these embryonic ducts into the liver tissue, it is found that the connective tissue surrounding the duct disappears on the liver side of the ducts before it does on the side of the duct toward the branch of the portal vein. Concomitantly with the disappearance of this connective tissue the cubical epithelium surrounding the lumen of the duct becomes replaced by the large, pale-staining, polygonal cells of the liver. These cells are directly continuous with the remainder of the liver parenchyma. They are frequently traversed by bile capillaries, the walls of which are continuous with the cuticular membrane lining the lumen of the bile duct. These embryonic ducts appear to run along the long axis of the portal—vein branches. At very frequent intervals the ducts dilate, and at these points branching ducts originate. These can be followed in serial sections and are seen to form a wide—meshed network about the portal vein and its branches. This is beautifully illustrated in the paper by Mall.” The origin of this network of bile ducts is easily visualized when one realizes that the bile ducts apparently metamorphose in situ from a portion of the network of liver—cell cords.

As yet we have found no exceptions to the following observation: Those portions of the lumina of embryonic bile ducts surrounded by cubical epithelium always have young connective tissue resting against the cubical epithelium; 462 WILLIAM BLOOM

when the connective tissue is absent the epithelium surrounding the lumen of the tube is continuous with and morphologically indistinguishable from the remainder of the hepatic parencliyma. This is clearly shown in figure 5. It can be seen from this why some of the previous workers on the subject have considered the bile ducts as growing into the liver along with the ingrowth of fibrous tissue about the branches of the portal vein. From our studies, however, this does not seem to be the case. \Vhat actually happens is that the ingrowth of fibrous tissue about the branches of the portal vein seems to convert the polygonal cells of the hepatic parenchyma into the compressed cubical. cells of the eXtra- and intrahepatic ducts. As the connective tissue grows into the liver, those tubes which were first surrounded by it will become more and more compressed, so that the picture is that of the main branches of the hepatic duct growing into the mese11chyma with the younger, lightly staining cells at the line of advancing growth of mescnchyma.

This process extends back to the hepatic duct proper. As described under the 16—mm. embryo, the hepatic duct is lined with densely staining cubical or cylindrical epithelium as long as it is surrounded by fibrous tissue. When this disappears, the epithelium becomes lightly staining, much larger and polygonal in shape, and continuous with a liver—cell cord. There are, moreover, no evidences of intrahepatic ducts until mesenchyma has penetrated the hepatic pulp along with the portal vein and its branches. Then those liver cells adjacent to the mesenchyma become cuboidal. When the mesenchyma has surrounded the liver cord completely, the cord will have changed its appearance so that it has the characteristics of a bile duct. From this it follows that the lumen, present as the bile capillary in the embryonic liver cords, will become, in those cords surrounded by fibrous tissue, the lumen of the extrahepatic and interlobular ducts draining those portions of liver circumjacent to them. EMBRYOGENESIS or BILE CAPILLARIES 463


The summary of the findings, then, on the application of the Eppinger stain to embryonic human livers is that, in the first place, bile capillaries can be demonstrated by this method in embryos 1 cm. in length. This is much earlier than they have previously been demonstrated. In the second place, the large lnmina, described by Toldt and Zuckerkandl as embryonic capillaries, are present in very young embryos and are certainly to be distinguished from the true bile capillaries. They are probably embryonic canaliculi from which the true bile capillaries arise or they may be dilated spaces due to the confluence of many capillaries. Thirdly, the system of excretory ducts develops in situ from liver-cell cords and seems to assume its typical cubical epithelium under the influence of the ingrowth of young connective tissue along the branches of the portal vein.

In none of our sections have We seen evidences of intracellular branches of the bile capillaries.


Although, because of the lack of Very young human embryos, we have not been able to trace a continuous development of bile capillaries from the lumen of the hepatic duct, the presence of this continuity in an embryo as small as 16 mm. leads to the conclusion that the bile capillaries are probably continuations of the lumen of the hepatic duct. Apparently, bile capillaries do not arise independently in solid liver-cell cords.

According to Kolliker,” the liver begins to secrete bile at the third month. Some observers have claimed that the bile capillaries of the adult are merely intercellular spaces stained with bile excreted by the liver cells. In our specimens bile capillaries are present at 10 mm.—a period when no histological or physiological evidences of secretion of bile have been found. VVe can assume from this that the bile capillaries originate as independent structures and are, therefore, probably not demonstrable as the result of secretion. 4641; WILLIAM BLOOM

We believe that an apparent demonstration of the continuity of hepatic~duct epithelium and liver cords, together with the direct continuity of the of the bile duct with the walls of the bile capillaries and the mode of origin of the intrahepatic ducts, form a fairly definite argument against the hypothesis of Gcraudel” to the effect that the bile ducts arise from the endoderm and the hepatic parenchyma from the mesoderm surrounding the hepatic diverticulum.

Non—parasitic cysts of the liver occur quite rarely, usually in conjunction with cysts of the kidney, but the latter are of much more frequent occurrence. From theoretical considerations, if the liver parenchyma arose from one anlage and the system of interlobular ducts from another, it would follow that in the process of anastomosis the failure of execution of this process might be of much more frequent occurrence and the condition accordingly should be present approximately as frequently as cysts of the kidney. In this organ two anlagen develop and have to anastomose. As a result of our concept of the development of the bile capillaries and ducts, we cannot agree with von Meyenberg’s explanation of the development of liver cysts, for he considers the bile ducts to have two separate anlagen; cysts, therefore, arise when the ducts from these two sources fail to unite. The liver and its ducts arise as the result of repeated dichotomous branching of the distal end of the hepatic duct. In an organ developing in such a manner, cysts, theoretically, should develop no more frequently than they do in other glands which have the same general type of embryogenesis. It is to be considered, however, that the adult liver is more complex structurally than the other compound tubular glands of the body. In a rather cursory examination of organs, such as the pancreas, lungs, and salivary glands, we believe this analogy to be fairly accurate. EMBRYOGENESIS or BILE CAPILLARIES 465


1. In human embryos stained with the Eppinger method, bile capillaries are present in large number at 10 mm.

2. These bile capillaries are to be sharply differentiated from the large lumina described by previous Workers.

3. The embryonic hepatic duct seems to branch continu ously in a dichotomous fashion to form the embryonic liver parenchyma.

4. Certain of these embryonic liver cords metamorphose,

apparently under the influence of connective tissue, into bile ducts.

5. Bile capillaries are present in human embryonic livers before there is evidence of secretion on the part of the liver.


EPPINGER, H. 1902 Zur Pathogenesc des Ikteru. Beitr. z. path. Anat. u. z. Allg. Path., Bd. 31, S. 230. AREY, L. B. 1924 Developmental anatomy, Philadelphia. MINOT, C. S. 1892 Human embryology. New York. IIEP.'rW1G, 0. 1915 Lehrbuch der Entwicklungsgeschichte der Menschen und der Wirbeltiere, 10th edition. Jena. SCHKFER, E. A. 1898 ‘Quain’s elements of anatomy, vol. 1, part I. London. 6 HERING, E. 1871 Kapitel 18 in Stricker’s Handbuch der Lehre von den Geweben. Leipzig. 7 POLICARD, A. 1914 Recherches histophysiologiques sur les voies biliaires intrahépatiques. J our. de physiol. et de path. gen., T. 16, p. 623. 8 VON IVIEYENBERG, H. 1918 Tiber die Oystenleber. Beit. z. path. Anat., u. z. allg. Path., Bd. 64, S. 477. 9 LEWIS, F. T. 1912 Keibel and Mall, Manual of human embryology, vol. 2, p. 403. Philadelphia. 10 TOLDT, 0., UND ZUCKERKANDL, E. 1875 Tiber die Form und Textur veranderungen dcr mcnschlichen Leber wiihrend des Wachstum. Sitz. Berichte der Kais. Akademie der Wiss. Wien., Bd. 72, Abt. 3, S. 241. 11 HENDRICKSON, W. F. 1898 The development of the bile capillaries as revealed by Go1gi’s method. Johns Hopkins Hosp. Bul1., vol. 9, p. 220. 12 ELZE, C. 1907 Beschreibung menschlichen Embryo von ca. 7 mm. griisster Léinge. Anat. Hefte, Bd. 35, H. 106, S. 409. 13 LEWIS, F. T. 1902 The gross anatomy of a 12-mm. pig. Am. Jour. Anat, vol. 2, p. 211. 14 MALL, F. P. 1906 A study of the structural unit of the liver. Am. Jour. Anat., vol. 5, p. 227. 15 KGLLIKER, A. 1879 Entwicklungsgeschichtc dos Menschen und der htiheren Tierc. Leipzig. 16 GERAUDEL, E. 1907 Le parenchyme hépatique et Ins voies biliaires sont deux formations génétiquement indépendants. Journ. de l’Anat. et de la Phys., '1‘. 43, p. 410.


Cl )F- 00 N)