Foetal haemopoiesis during the hepatic period. I. Relation between in vitro liver organogenesis and erythropoietic functionкод для вставкиСкачать
THE ANATOMICAL RECORD 200:221-230 (1981) Foetal Haemopoiesis During the Hepatic Period. 1. Relation Between In Vitro Liver Organogenesis and Erythropoietic Function ANALIA C. NESSI, CARLOS E. BOZZINI, AND MORRIS V. TIDBALL Instituto Argentino de Biologia Teorica, Calk 32 no 1824,La Plata (A.C.N., M.V.T.), and Catedra de Fisiologia, Facultad de Odontologia, M.T. de Aluear 2142, Buenos Aires, Argentina (C.E.B.) ABSTRACT Cultures of the hepatic bud, under different experimental conditions, show a direct relation between foetal age at the time of dissection and the further organogenesis of the explant. In cultures of the septum transversum plus hepatic bud-with or without splanchnic mesenchyme-obtained from embryos of four to 25 somites, the capacity of endodermic cells to differentiate into hepatocytes appears only in a small number of samples; whereas, in the hepatic bud from older embryos (26 to 40 somites), this differentiation occurs in all cases. The amount of time cultures were allowed to grow was important for hepatic organogenesis, as measured by cord-like organization of hepatocytes plus their storage capacity for glycogen. The possibility of the explants showing complete haemopoiesis was also a condition offoetal age a t the time of explant. Haemopoiesis was not found in cultures from embryos of less than 25 somites. On the other hand, development of haemopoiesis did not show a direct relation to days of culture, since the peak of this activity was observed towards the third day in vitro. Most explants showed a generalized haemopoiesis (both interstitial and vascular) with a decay towards the sixth day. Endodermal cells of the hepatic bud were capable of both proliferation and differentiation into hepatocytes, even i n those mesenchymes considered inadequate such as limb mesenchyme. In the latter case we were unable to find haemopoiesis at any time. The septum transversum, when cultured alone, did not contain haemopoietic cells. Hepatic organogenesis (as a system which depends on endomesodermal interaction for its development) is not a typical example of highly specific mesenchyme, induction-dependent tissue, as compared to the salivary (Lawson, '70, '72) or pulmonary systems (Alescio and Cassini, '62; Wessells, '70; Spooner and Wessells, '70). Le Douarin ('64a,b, '69) considers two classes of mesenchyme for the chick embryo hepatic organogenesis: Those favorable (ventral in general and septum transversum in particular) and those unfavorable (somitic, cephalic, and limb) for hepatic growth and differentiation. The true origin of hepatic haemopoiesis function (HF) is probably one of the most controversial problems in modern haematology. In effect, there is no agreement on whether this function depends on elements formed in situ or whether the endomesodermic component serves as a n 0003-276X/81/2002-0221$03.00 0 1981 ALAN R. LISS, INC. adequate bed for the colonization of migrating elements. The purpose of this work was to study the intrinsic haemopoietic capacity of hepatic bud and mesenchyme, as well as the capacity of these tissue to induce liver organogenesis under different experimental conditions. MATERIALS AND METHODS Animals CF, mouse embryos were used. In calculating the foetal age, the morning after overnight mating (on which the vaginal plug was observed) was considered as day 0. Pregnant females were sacrificed by decapitation between days 9 and 10 after mating, and the Received April 23, 1979 accepted July 31, 1980. Carlos E. Bozzini is Career Investigator for Consejo Nacional de Investigaciones Cientificas y Tecnicas de la Republica Argentina. Address reprint requests to Dr. Analia C. Nessi, Instituto Argentino de Biologia Teorica, Calle 32 no 1824, La Plata (1900),Argentina. 222 A N A L ~ Ac. NESSI, CARLOS E. BOZZINI, AND MORRIS gravid uterus removed in sterile conditions. Embryos were washed in balanced saline and dissected under a binocular microscope in the following way: A first incision was made between the caudal boundary of the cardiac prominence and the cephalic limit of the anterior limb, and a second between the umbilical cord and the caudal limit of the same limb bud (Fig. 1). Figure 2 shows the planes of section for obtaining the septum transversum with the hepatic bud plus coelomic mesoderm and ventral ectoderm. From the portions thus obtained, those for the first group of study were prepared by removal of the ventral mesoderm. The ones for the second group were obtained in the same way, plus a removal of the coelomic mesoderm as shown in Figure 3. Specimens for the third group were first prepared as for the second group, but a further separation was made of their central portions, which contains the hepatic bud. The epithelial bud cannot be separated from the septum transversum solely by mechanical methods. For its isolation, this central part was incubated for five minutes a t 37" C in a 0.25% solution of Trypsin (Whorthington Cristalline Trypsin) and 1%Pancreatin (Difco) in Tyrode solution. Further separation of the endoderm from its surrounding mesenchyme then proceeded using tungsten needles. In embryos with more than 28 somites it is impossible to separate all of the hepatic bud, as a result of its peculiar organogenesis. The hepatic primordium proliferates not by terminal buds as other endoderm-derived organs, but by cellular migration towards the surrounding mesenchyme. Therefore, some of these cells are inevitably lost during digestion. The culture techniques were done as described by Lawson ('70). Histochemical methods Explants were fixed a t days 0, 3, and 6 of culture, in Carnoy solution, at 0°C for 45 min- v. TIDBALL utes. Embedding was done in Paraplast (Sherwood Med. Ind. Inc.). Serial sections were mounted in three alternate groups, thus permitting different staining techniques of analogous zones. The first group was incubated in 0.0% amylase solution in phosphate buffer at pH 6.8 for 30 minutes. Following that they were stained with periodic acid, Schiff (PAS) and Mayer Haemalum. The second group was identically prepared except for the absence of the enzymatic treatment. In order to test not only proliferation, but also red blood cell differentiation, the third group was stained with benzidine, following Pearse technique ('72), Mayer Haemalum, and Giemsa. EXPERIMENTAL DESIGN AND STATISTICAL ANALYSIS The material studied consisted of six experimental groups, divided in the following way: Group 1: Septum transversum with hepatic bud plus coelomic mesoderm in embryos of four to 40 somites. In embryos of less than 15 somites, as the hepatic bud is still not formed, the most-caudal portion of the foregut was dissected, i.e., the hepatic bud presumptive area. Group 2: Septum transversum plus hepatic bud from 16 to 35 somite embryos. Group 3: Hepatic bud from embryos of 21 t o 35 somites. Group 4: Coelomic mesoderm from 24 to 30 somite embryos. Group 5 : Limb mesenchyme from embryos of 21 to 35 somites. This material was obtained by removing a complete limb rudiment by a plane of section perpendicular to the limb axis. The ectoderm of this rudiment was then removed. Group 6: Hepatic bud-plus-limb mesenchyme from embryos of the same age as for group 5. The intrinsic capability for proliferation and differentiation of each explant and their capacity t o support development of the HF were analysed. For this, foetal age a t dissection and length of the in vitro period were considered. Fig. 1. First dissection planes (-----I in a mouse embryo, after removing the whole foetal anexes. al, Anterior limb; h, heart; mp, mandibular processes; uc, umbilical cord. Fig. 2. A threedimensional representation of the cuneiform section from Figure 1, showing the septum transversum (st),hepatic bud (hb),anterior limb rudiments (alr), neural tube (nt),somites ( s ) , and the second dissection planes. Fig. 3. Twenty-eightrsomite stage embryo. h, heart; lm, caudal limb rudiment; mp, mandibular process; nt, neural tube; st, septum transversum; -----,dissection planes. PAS-Haemalum staining; magnification, x 60. Fig. 4. Septum transversum plus hepatic bud from a 27-somite stage embryo. Large arrow-empty wide-lumen blood vessel; small arrows-narrow-lumen sinusoids containing immature blood cells (see discussion). The darkest area beside the hepatic bud corresponds to epithelial migrating cells from this rudiment. PAS-Haemalum staining; magnification, x 125. FOETAL HAEMOPOIESIS DURING THE HEPATIC PERIOD 223 224 ANALfA C. NESSI. CARLOS E. BOZZINI. AND MORRIS V. TIDBALL of hepatocytes plus important storage of glycogen (LO-4)(Fig. 7) appears most prominently in explants from the older embryos after six days of culture (Table 1). Development of pancreatic-like acini was also found among the explants of group 1, with a higher frequency in the younger subgroup (43.3 ? 11.2%, compared to 8.1 ? 1.9%in the older). Optimal HF (HF-4) was directly related to embryo age at the time of dissection and not to the time explants were allowed to grow in vitro (Figs. 9 and 11).Indeed, H F involutes to a local and intravascular haemopoiesis (HF-2) by the sixth day of culture. Blood cells were rarely found in explants from embryos younger than 25 somites (Table 21, and only one case of this age, amongst 210 explants, developed HF. This corresponds to a frequency of 0.48% when, on the other hand, 100.0% of explants with 26 somites or older show one of the three active haemopoietic stages (Tables 2 and 4). No significant difference, both from the point of view of haemopoietic capacity or from that of organogenesis, was observed between groups 1 and 2, as shown in Tables 1 and 3. The third experimental group (isolated hepatic bud) was incapable of proliferation and differentiation. Both in three and six days of culture (in 18and 16explants, respectively) the material was expanded in the clot and was not viable. The fourth experimental group (lateral mesoderm) showed proliferation of connective tissue cells; in nearly all of these explants a cell loss occurred from the periphery, resulting in a smaller size of explants a t the end of the experiment than at their plating. No blood cells were observed at either three or six days of culture. Limb mesenchyme explanted alone (group 5 ) or associated with hepatic bud (group6) showed Liver organogenesis was evaluated according to the following scale: LO-1-complete absence of hepatic tissue; LO-2-endodermic proliferation without discernible differentation; LO3-hepatocytic differentiation, evaluated by the presence of glycogen synthesis and storage with the PAS-amylase histochemical method; and LO-4-cord-like organization of hepatocytes plus important storage of glycogen. Development of HF was evaluated according to t h e following scale: HF-1-Absence of haemopoiesis; HF-2-local, intravascular haemopoiesis; HF-3-local, interstitial and intravascular haemopoiesis; and H F - P w i d e spread haemopoiesis. Statistical analysis was based on the x2test of the categoric variables' independence (Haber and Runyon, '73). RESULTS The hepatic bud first appears at the mouse embryo developmental stage of 15 to 17 somites. Although explants reach a chronologically equivalent age of 13 and 16 days of normal in vivo development, both degree of differentiation and growth are retarded in vitro when compared with normal gestation. Experimental group 1 (septum transversum with heptaic bud plus coelomic mesoderm) shows development of hepatic tissue in 60.0%of embryos with age up to 25 somites, and 83.8% in older embryos, at the third day of culture (Table 1).Differentiation was more extensive in the older groups (at the time of culture) of explants (Figs. 5 through 7, Table 1). Conversely, in the group of younger explants the percentage of those which do not show any hepatocyte differentiation ascends to 36.7% at three days of culture, and to 12.5% a t six days. Absence of hepatic tissue was not observed in older groups (Table 1).Cord-like organization - Fig. 5. Same explant a s in Figure 7, treated with amylase prior to the PAS-Haemalum staining; magnification, x 500. Fig. 6. Septum transversum plus hepatic bud, plus somatopleuric and splachnopleuric mesoderm, from a 33-somite embryo, cultured for three days. The persistence of the coelom (c) divides the picture into two zones: The lower part shows good liver organogenesis and optimal development of haemopoiesis, whereas in the upper region, only a good development of epithelial cells is seen. Bemidine-Haemalum and Giemsa staining; magnification, X 125. Fig. 7. Septum transversum plus heaptic bud from a 30-somite stage embryo, after six days of culture. No interstitial haemopoiesis is seen. Liver organogenesisbelongs to the fourth stage, showing a cord-like organization of hepatocytes and important glycogen storage. PAS-Haemalum staining; magnification, x 500. Fig. 8. Septum transversum plus hepatic bud from a 26-somite stage embryo (six days of culture). A big capillary lumen (cl) with numerous erythroid cells in different maturating stages, is seen. The haemopoietic zone is only seen in close relationship to that blood vessel, showing the specific local nature of this process, hinucleate cells (arrows). Benzidine-Haemalum and Giemsa staining; magnification, X 500. FOETAL HAEMOPOIESIS DURING THE HEPATIC PERIOD 225 226 ANALfA C. NESSI, CARLOS E . BOZZINI. AND MORRIS V. TIDBALL aggregate around or to form a lumen, as observed in normal development. But in day 6 cultures, in cells that show this second behavior, glycogen storage is not comparable in magnitude to the high PAS positivity of cells having a cord-like arrangement. However, in some day 6 cultures, binucleate cells were frequently observed (Fig. 8, arrows). The hepatic bud incapacity to proliferate and differentiate in isolated cultures coincides with the findings reported by Le Douarin ('64a,b, '69). Other isolated epithelial buds, such as the primordium of salivary glands, show similar DISCUSSION behavior (Lawson, '72). Nevertheless, our findGlycogen incorporation by embryo hepato- ing t h a t the hepatic bud has the capacity to cytes is not of great importance until days 17 differentiate-even in a minor way-when in and 18 of in vivo development (Peters et al., association with a mesenchyme considered un'63). As stated by these authors, day 18 corre- favorable, differs from t h e findings of Le sponds to the peak of glycogen synthesis, with Douarin. The difference in experimental modvalues even higher than in normal adult cells. els could be responsible for these differences in In the present study, day 6 explants attained a results. Some authors (Thomas and Yoffey, '62; Yofchronologically equivalent age of 15 and 16 days of normal intrauterine development, and fey, '71) accept a n endodermal origin of it is precisely in these explants where the opti- haemopoietic cells in the foetal liver. Neverthemal liver organogenesis (LO-4) was observed. less, the incapacity of hepatic bud-cultured In monolayer cultures, a s there is a loss in the alone, recombined with its own mesenchyme, original structure of the specimen, the presence or associated with limb mesenchyme-to deof glycogen is not useful for identification of velop H F shows the failure of the endodermic hepatocytes from other endoderm-derived tis- layer for the development of haemopoiesis by sues. Nevertheless, the presence of glycogen itself. The findings of Houssaint ('70, '72) upon conserves its value as an hepatocyte marker in recombining mouse embryo hepatic bud with organotypic cultures. Also, in the latter cul- lung mesenchyme obtained from chick embryos tures, other judgement factors exist in deter- coincide with ours. On the other hand, upon mining the degree of development; i.e., conser- associating the septum transversum with the vation of the general architecture, especially in hepatic bud of young embryos, or utilizing the the few days of culture. In this way, in hepatic lateral mesoderm of the same embryos, HF detissue showing good in vitro organogenesis velopment was not attained on day 3 or 6 of with observable cord-like development, culture, even with growth and differentiation hepatocytes will be highly PAS-positive, of the explants (Tables 2 and 4, Fig. 10). On whereas other epithelial cells derived from the these bases, a n interesting point of discussion is hepatic bud, i.e., pancreatic acini, epithelium of posed by t h e presence of haemopoiesis i n bile ductules, etc., will show a tendency to explants of a given age, as opposed to the total a good proliferation of cartilaginous tissue and general connective tissue elements. The hepatic bud not only proliferates in these conditions, but also shows differentiation (Table 5 , Fig. 12). It does not reach a n optimal organogenesis, however, even in six days of culture, but some ofthe hepatocytes store glycogen (LO-3). Original embryonic age is not critical for the further organogenesis of the explant (Table 5 ) but, as in groups 3 and 4, no H F was observed. Forty-five percent of grafted explants developed pancreatic-like acini. Fig. 9. Detail of Figure 6; compare this picture with Figure 11, and the lower region of Figure 8. Benzidine-Haemalum and Giemsa staining; magnification, x 500. Fig. 10. Septum transversum with the hepatic bud plus coelomic mesoderm from a 19-somite stage embryo ( 3 days of culture). A rather good liver organogenesis exists. Nevertheless, no erythroid cells are seen. c, coelom; um, undifferentiated mesenchyme. Amylase-treated section, prior to the PASHaemalum staining; magnification, X 125. Fig. 11. Septum transversum with the hepatic bud plus coelomic mesoderm from a 33-somite stage embryo, (six days of culture) showing a good liver organogenesis and a decay in the haemopoietic function. Benzidine, Haemalum, and Giemsa staining; magnification, x 500. Fig. 12. Limb mesenchyme plus hepatic bud from a 29-somite embryo, after six days of culture. In the central area, a round epithelial formation (ep)is seen, showing their cells of cord-like organization. In some of them a thin granulation of PAS-positive material, which disappears with the amylase digestion, is seen. Haemopoiesis does not develop under these experimental conditions. pc, protochondral tissue; um, undifferentiated mesenchyme. PAS-Haemalum staining; magnification, x 500. ANALIA C. NESSI, CARLOS E. BOZZINI, AND MORRIS V. TIDBALL 228 TABLE 1. Liver organogenesis in ST + HB + CM explants plated in block (group lJ.* No. of cultures Days of culture No. of somites 30 3 71-25 Stages (%) LO, ? 24 71-25 6 36.7 2.4 3 60.0 5.5 2 3.3 0.3 25.0 2.6 ? 45.8 4.7 83.8 6.8 * 16.2 1.3 2 23.5 2.0 f 12.5 t 1.3 37 ? 0 26-40 ? 6 34 26-40 LO, LO, 2.9 0 t 0.2 ____ ~~~ ~ LO4 0 18.7 t 1.7 0 67.5 t 5.7 ~ *Stages are gwen as the average percent of a t least four experiments (t95% confidence) LO = liver organogenesis, ST = septum transversum, HE = hepatic bud, CM = coelomic mesoderm TABLE 2. Haemopoietic function development in ST plated in block (group l).* No. + HB + C M explants No. of cultures Days of culture of somites 30 3 71-25 100.0 24 6 71-25 -t 95.8 4.3 i f 2.7 0.2 t 37 34 3 6 _ _ _ HFi 26-40 26-40 ~ Stages (%) . HF, 0 ? HF, HF, 0 0 0 0 0 4.2 0.2 2.8 0.2 ? 70.6 6.1 f 16.2 1.3 t 6.4 78.3 20.6 1.8 ? 8.8 0.6 *Stages are given as the average percent of a t least four experiments (t 95% confidence). HF = haemopoietic function; ST = septum transversum; HB = hepatic bud; CM = coelornic mesoderm. TABLE 3 . Liver organogenesis in ST No. of cultures Days of culture No. of somites 57 3 16-25 45 6 16-25 + HB explants plated in block (group 2).* 36 3 6 26-35 26-35 . LO, LO, LO, LO, 0 43.0 f 2.8 57.0 ? 3.8 ? 20.0 2.5 t 2.5 2 27.0 2.0 ? 15.0 0.9 t 85.0 5.6 t 27.0 2.0 63.9 2.8 * 14.0 1.1 f 13.0 i 0.9 27 Stages - _____ -. 0 0 ? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ~ *Stages are mven as the average percent of a t least five experiments ( 2 95% confidence) LO = liver organogeneis, ST = septum transversum, HB = hepatic bud 0 33.1 86.0 0.1 229 FOETAL HAEMOPOIESIS DURING THE HEPATIC PERIOD + H B explants TA B L E 4 . Haemopoietic function deuelopment in S T plated in block (group 21.* No. Days of culture No. of cultures Stages (%) ~~ of somites H FI HF, ~ HF:, HF, ~~ 57 3 16-25 100.0 0 0 0 45 6 16-25 100.0 0 0 0 27 3 26-35 0 0 ? 36 6 26-35 0 ? 63.9 2.8 7.4 0.3 2 14.0 92.6 4.2 11.1 0.1 -c 1.1 f *Stages are given as the average percent of a t least five experiments ( ? 95% confidence), HF haemopoietic function: ST = septum transversum; HI3 = hepatic bud. ~ TABLE 5. Lruer organogenesrs rn HB explants grafted on limb mesenchyme (group 6).* Stages (%I No. of cultures Days of culture No. of somites LO, LO1 LO, 58 6 21-25 0 37.5 i z 2.5 62.5 t 4.1 0 42 6 26-35 0 66.7 3.2 0 LO4 ~~~ & *Stages are given as the average percent of at least seven experiments I,O = liver organogenesis; HB = hepatic bud. (? 33.3 1.8 _f 95% confidence) TABLE 6. Haemopoietic functron deuelopment in H B explants grafted on lrmb mesenchyme (group 61." ~~ Stages (%) No. of cultures Days of culture No. of somites HFr HFI HF, 58 6 21-25 100.0 0 0 0 42 6 26-35 100.0 0 0 0 ~ ~ ~ ~ ~ *Stages are given as the average percent of a t least seven experiments HF = haemopoietic function; HB = hepatic bud. incapacity for such function in specimens obtained from younger embryos. Johnson and Jones ('73) obtained similar results, but these authors suggested that 28 somites is the critical stage for division between explants with or without the capacity to develop haemopoiesis. They suppose the embryonic layers to be too immature before that period to bring about the interactive phenomena that result in active HF. It must be pointed out that the septum (? HF, ~ 95% confidence). transversum is traversed by t h e omphalomesenteric veins in periods prior to the critical age for t h e development of haemopoiesis i n vitro. Why is i t then, if the migrational theory is valid, that HF does not develop earlier? We propose t h a t purely mechanical factors may play a role in these experiments. In effect, in embryos up to 24 somites of development, the omphalomesenteric vein lumens are very wide in relation to the 230 ANALtA C. NESSI, CARLOS E. BOZZINI, AND MORRIS V. TIDBALL septum size, and it is possible that on handling during dissection and successive washings, circulating haemopoietic cells are separated from the explants. Note the empty lumen of a wide capillary in Figure 4 (large arrow). On the contrary, omphalomesenteric veins after giving rise to thinner branches (Fig. 4,small arrows) could facilitate the retention of haemopoietic cells found amid the septum transversum, thus making their loss during processing improbable. Some explants, even those considered in the best conditions of development for both HF and organogenesis, show peripheral areas where haemopoiesis is absent. Similar changes may be observed in the marginal zones of the normally in-vivo-developedhepatic lobes (Nessi and Bozzini, ’79). The presence of these areas (Fig. 6, upper part) in an explant that shows some zones very densely populated by blood cells and some other zones free of these elements, opposes the idea of the embryonic layers being too immature to interact, resulting in an active de novo haemopoiesis. On the contrary, the present findings favor the migrational theory. REFERENCES AIescio, T., and A. Cassini (1962) Induction in vitro of tracheal buds by pulmonary mesenchyme grafted on tracheal epithelium. J. Exp. Zool., 150:83-94. Haber, A., and R.P. Runyon (1973) Estadistica General. Bogota, Caracas, Mexico, etc. Fondo Educativo Interamericano, p. 266. Houssaint, E., N. Le Douarin, and A. Weaver (1970) Mise en evidence de glycogzne dans le foie de l’embryon de poulet des le stade de 4 jours d’incubation. C.R. Acad. Sci. (Paris), 271: 1315-1318. Houssaint, E. (1972) Etude des capacites de differentiation de l’endoderme hepatique de l’embryon de Souris. C.R. Acad. Sci. (Paris), 275t461-464. Johnson, G.R., and R.O. Jones (1973) Differentiation of the m a m m a l i a n hepatic primordium i n vitro. I. Morphogenesis and the onset of haemopoiesis. J. Embryol. Exp. Morphol., 30t83-96. Lawson, K.A. (1970) Morphogenesis and functional differentiation of the rat parotid gland in vivo and in vitro. J. Embryol. Exp. Morphol., 24t411-424. Lawson, K.A. (1972) The role of mesenchyme in the morphogenesis and functional differentiation of rat salivary epithelium. J. Embryol. Exp. Morphol., 27t497-513. Le Douarin, N. (1964a) Isolement experimental du mesenchyme propre du foie et r6le morphogene de la composante mesodermique dans l’organogenese hepatique. J. Embryol. Exp. Morphol., 12.141-160. Le Douarin, N. (1964b) Induction de l’endoderme prehepatique par le mesoderme de l’aire cardiaque chez I’embryon de poulet. J. Embryol. Exp. Morphol., 12t651664. Le Douarin, N. (1971)Les apports recents des cultures et des greffes en biologie animale. Masson, Paris, p. 95. Nessi, A.C., and C.E. Bozzini (1979) Foetal haernopoiesis during the hepatic period. 11. Topographic histology. J. Embryol. Exp. Morphol., 52t13-21. Pearse, A.G.E. (1972) Histochemistry, 3rd Ed. Churchill Livingstone, Edinburgh and London, p. 1387. Peters, V., G. Kelly, and H. Dembitzer (1963) Cytologic changes in fetal and neonatal hepatic cells of the mouse. Ann. N.Y. Acad Sci., 111.87-103. Thomas, D.B., and J.M. Yoffey (19621 Human foetal haemopoiesis. I. The cellular composition of foetal blood. Br. J. Haematol., 8.290-301. Spooner, B.S., and N.K. Wessells (1970) Mammalian lung development: Interactions in primordium formation and bronchial morphogenesis. J. Exp. Zool., 175:445-454. Wessells, N.K. (1970) Mammalian lung development: Interactions in formation and morphogenesis of tracheal buds. J. Exp. Zool., 175.455-466. Yoffey,J.M. (1971)The stem cell problem in the fetus. Isr. J. Med. Sci., 7t825-833.