J. Embryol. exp. Morph. 94, 189-205 (1986)
Printed in Great Britain © The Company of Biologists Limited 1986
189
Branching morphogenesis in the avian lung: electron
microscopic studies using cationic dyes
BETTY C. GALLAGHER*
Department of Biological Sciences, Stanford University, Stanford, California, USA
SUMMARY
The developing chick lung was examined in the electron microscope for intimate cell contacts
between epithelium and mesenchyme, discontinuities in the basal lamina and substructure of the
basement membrane. Cell filopodia were seen which crossed the basal lamina from both the
epithelial and the mesenchymal cells. Ruthenium red and tannic acid staining of the basal lamina
of the chick lung showed it to be thin and sometimes discontinuous at the tips compared to the
more substantial basal lamina in the interbud areas. The bilaminar distribution of particles seen
with ruthenium red is similar to those seen in the cornea and lens. With tannic acid staining,
filaments could be seen which crossed the lamina lucida and connected with the lamina densa.
Spikes perpendicular to the basal lamina were sometimes seen with a periodicity of approximately 110 nm. Alcian blue staining revealed structure similar to that seen by ruthenium red
staining in the salivary and mammary glands, although the interparticle spacing was closer.
Collagen was located in areas of morphogenetic stability, as has been seen by other investigators in different tissues. Collagen was coated with granules (probably proteoglycan) at
periodic intervals when stained with ruthenium red. The fibrils were oriented circumferentially
around the mesobronchus and were assumed to continue into the bud, but the fibres curve
laterally at the middle of a bud. This orientation is opposite to that seen by another investigator
in the mouse lung.
In general, the observations made in the avian lung are similar to those seen in branching
mammalian tissue. It is likely, therefore, that the chick lung uses strategies in its morphogenesis
that are similar to those that have been elucidated previously in developing mammalian organs.
INTRODUCTION
When an organ progresses from an anlage to a highly organized structure,
interactions between the epithelial and mesenchymal tissues are necessary for
proper development, as seen by the reciprocal dependence of the two tissue types
on each other for their differentiation (Rudnick, 1933; Auerbach, 1960; Taderera,
1967).
The need for cell-to-cell contacts between epithelial and mesenchymal tissues
has been well established in inductive events (Saxen etal. 1976) and several
electron microscope studies have reported their presence in various mouse tissues
(Cutler & Chaudrey, 1973; Coughlin, 1975; Bluemink, Van Maurik & Lawson,
* Present address: Department of Anatomy and Cell Biology, Box 439, University of Virginia
Medical Centre, Charlottesville, Virginia 22908, USA.
Key words: chick lung, basal lamina, ruthenium red, tannic acid, alcian blue, collagen
orientation.
190
B. C. GALLAGHER
1976; Lehtonen, 1975). A second feature of branching morphogenesis is a basement membrane that becomes thinner or discontinuous at the tips of growing buds
compared to interbud areas (Bernfield & Banerjee, 1972; Bernfield, Cohn &
Banerjee, 1973; Coughlin, 1975; Lehtonen, 1975; see Bernfield, Bannerjee, Koda
& Rapraeger, 1984, for review). Tannic acid and ruthenium red have been used to
examine substructures of the basal lamina (the nonfibrillar layer of the basement
membrane) in chick neural tube, notochord and cornea (Trelstad, Hayashi &
Toole, 1974; Hay, 1978), chick lens (Hay & Meier, 1974), mouse salivary gland
(Bernfield, Banerjee & Cohn, 1972; Cohn, Banerjee & Bernfield, 1977), and
mouse mammary gland (Gordon & Bernfield, 1980).
Collagen is distributed preferentially in developmentally stable regions of
various tissues (Kallman, unpublished observations; Grobstein & Cohen, 1965;
Wessells, 1970; Mauger et al. 1982). Its arrangement in mouse lung is well ordered
(Wessells, 1970), and comparisons of collagen distribution and orientation in the
chick lung might relate to stabilizing morphology after initial morphogenetic
events have occurred.
The chick lung is an advantageous system in which to study developmental
events, because the buds emerge in a well-defined and characteristic sequence.
Due to this conformation, a bud can be identified by the number of buds rostral to
it, and its age determined by the number of buds caudal to it. A bud can be
compared to homologous buds in other lungs at various ages, and can also be
compared to buds in the same lung at different stages of development.
The present study was undertaken for the purpose of establishing the incidence
of intimate cell contacts, discontinuities in the basement membrane at the tips of
growing buds and the distribution of GAGs (glycosaminoglycans) and collagen in
the developing chick lung. Ample evidence of close cell-to-cell contact was found.
A correlation was made between patterns seen in the light microscope using lectins
in a previous study and patterns seen in the electron microscope: a basal lamina
that is thinner at the tips of growing buds than in the interbud area. The basal
lamina contains an ordered substructure and collagen is preferentially distributed
in the interbud area. Finally, the orientation of the collagen fibres is perpendicular
to that previously found in the mouse lung.
MATERIALS AND METHODS
Electron microscopy
White Leghorn eggs were incubated for 5-8 days and staged according to Hamburger &
Hamilton (1951). The lungs were removed while the embryos were immersed in 3 % glutaraldehyde and 0-5 % paraformaldehyde in either a 0-llM-Hepes or 0-08M-cacodylate buffer at
pH7-4. The buffer used did not significantly alter the ultrastructural appearance.
Ten lungs were transferred to fresh fixative containing (1) 2 % tannic acid (Electron Microscopy Sciences) which was buffered to a final pH of 5-3 (Singley & Solursh, 1980) and postfixed
in 1 % OsO4. Eleven lungs were placed in (2) 0-2 % ruthenium red (Sigma), pH 7-0 for 2h and
postfixed in 0-05 % ruthenium red and 1 % OsO4 in the dark, also for 2h (Luft, 1971). Seven
lungs werefixedfor 2h in (3) 1 % alcian blue, pH6-35, followed by postfixation in 0-25 % alcian
blue for 2 h (Behnke & Zelander, 1976). The mesothelial surface of lungs treated with these dyes
was gently scraped prior to fixation to ensure the even penetration of ruthenium red and alcian
Electron microscopic studies ofavian lung
191
blue. Seven lungs were processed in (4) the initial fixative without supplement and postfixed in
1% OsC>4 for l h (conventional fixation). The tissues were stained en bloc with 2% uranyl
acetate in 0-05 M-sodium hydrogen maleate, pH5-2, then dehydrated in ethanol, propylene
oxide, and embedded in Epon Araldite.
Specimens were sectioned at 3 fim, collected on glass slides, stained with methylene blue and
examined in the light microscope for areas of optimum interest and orientation, as well as for
penetration of ruthenium red and alcian blue. Sections containing exact cross sections were
selected, by taking the midpoint of the bud or the interbud area, in order to increase the
accuracy of measurements of basal laminar thickness. Only the four earliest formed buds were
examined so that comparisons could be made between buds of known age. A block was then
glued to a chosen section with Devcon '5 minute' Epoxy (Danvers, Mass.), then immersed in
liquid nitrogen until the block snapped off the slide. The block was trimmed and sections were
then cut, collected on slotted grids and stained with lead citrate. The material was viewed in a
Hitachi HU 11E-1 electron microscope at 75 kV which had been calibrated with a carbon grating
replica (Fullam).
Polarizing microscopy
For studies using the polarizing microscope, lungs were dissected out of embryos between
stages 26 and 30 in a medium consisting of 1:1 horse serum (Irvine Scientific): Ham's F-12
(Gibco). The lungs were placed in a 1 % trypsin solution for 2-3 min at room temperature, after
which the mesenchyme was removed using forceps and iridectomy knives.
RESULTS
General description
In a typical longitudinal (parasagittal) section, the mesobronchus is cut along its
length, and the first few buds emerge from its dorsal surface in cranial-caudal
sequence (Fig. 1). The area between two bronchial buds will be referred to as the
interbud area. The most-distal extension of a bronchus will be referred to as a tip,
and when a tip begins to branch the area between the two new branches will be
called the new interbud area. The age of a lung will be referred to as the number of
buds it has produced, e.g. a lung with 1 bud will be a 1-bud lung, a bud with 3 buds
a 3-bud lung, etc.
Cell-to-cell contacts
Direct contacts are occasionally seen between epithelial and mesenchymal cells.
They are located near the tips of newly emerging bronchi, rather than older
bronchi or an interbud area, and consist of an extension from either cell type
through discontinuities in the basal lamina. The tips of the extension come within
5-15 nm of the plasma membrane of the contacted cell. No specialized junctional
structures were seen. In the interbud area, extensions from mesenchymal cells are
frequently directed towards the epithelium, but the basement membrane is always
intact in these cases, and therefore the cell layers are separated by at least its
thickness (Fig. 2). In a lung at the 1-bud stage (stage 24), cell contacts are not seen
in an area of the epithelial surface which 8h later will give rise to the second bud.
However, in this same specimen, the first bud has just begun to emerge, and along
its basal lamina several mesenchymal filopodia come within 15-20 nm of the
192
B. C. GALLAGHER
Electron microscopic studies ofavian lung
193
epithelial cells' plasma membranes. At the 5- to 8-bud stage (stage 28), epithelial
cells at the tips of newly formed bronchi are seen with processes which extend
through the basal lamina towards its mesenchyme. Sometimes the basal lamina
remains over most of the surface of the cell process, but the tip is bare. In these
cases, the epithelial-mesenchymal cell contact is closely apposed, and in one case
is separated by only 5 nm (Fig. 3). At stage 30, intimate cell contacts are not seen
in the four rostral buds.
Basal laminar discontinuities
The basal laminae of buds at different stages of development were compared for
evidence of discontinuities at growing tips. As implied above, newly formed tips
have basal laminae which are discontinuous and contain few structural details
(Fig. 4). At the tips of buds formed 1 day earlier and in the interbud area, the basal
lamina is distinct, continuous and is associated with copious fibrillar material on its
mesenchymal surface (Fig. 5). New interbud basal laminae are similar to those of
old interbud areas. Gaps in the basal laminae were not seen in areas expected to
produce primary or secondary bud outgrowth.
Ultrastructure of basal lamina
In areas where the basal lamina is continuous, ultrastructural details can be
discerned. Ruthenium red enhances the electron density of small particles that are
present on either side of the diffusely staining lamina densa (Fig. 7) in older buds
and interbud areas. The particles and the lamina densa are much less evident at the
tips of new buds (Fig. 6). The particles are approximately 15 nm in diameter and
are separated from adjacent particles in the same layer by 47 nm (±12nm, n = 73
measurements, five lungs). The two layers of particles are approximately 60nm
apart. Periodic structures visualized by staining with alcian blue have an average
separation of 58 nm (±15 nm, n = 52 measurements, three lungs) (Fig. 8). In
tangential sections, the particles appear to be aligned vertically and horizontally in
a tetragonal array (Fig. 9).
With tannic acid staining, the area of the basal lamina immediately adjacent to
the plasma membrane (the lamina lucida) is transversed by thin fibres which
Fig. 1. Semithin section of avian lung at 3-bud stage. The section is oriented dorsal
side up, so that bud outgrowth is upward from the mesobronchus. ep, epithelium; mes,
mesenchyme. x265. Bar, 100 jum.
Fig. 2. The basement membrane (bm) of the interbud area is characterized by an
extensive fibrillar layer and the basal laminar (bl) layer. The basal lamina is composed
of the electron-dense lamina densa (Id) and the electron-lucent lamina lucida (//). The
mesenchyme (mes) is separated from the epithelium (ep) by the basement membrane,
x50000. Bar, 100nm.
Fig. 3. A process from a 7-day epithelial cell (ep) extends through the basal lamina (bl)
to contact a mesenchymal cell (mes). The basal lamina covers the surface of the process
except for the tip, where the two cells are separated by only a 5 nm gap. x82000. Bar,
100 nm.
B. C. GALLAGHER
Electron microscopic studies of avion lung
195
sometimes appear Y- or X-shaped (Fig. 10). The spacing between these filaments
is 42nm (±10nm, n = 48 measurements, five lungs). The difference between the
periodicities seen by ruthenium red and tannic acid is statistically significant
(t-test, P< 0-01), and therefore implies that different structures are bound by the
two stains.
The distance between the centre of the lamina densa and the epithelial
plasma membrane was compared between interbud areas (64 nm ± 13 nm, n = 13
measurements, four lungs) and tips (45nm± 16-17 nm, n = 31, four lungs) in
ruthenium-red-stained tissue as a measure of basal laminar thickness. A significant
difference between these two sets of measurements ( P < 0-001) can be detected,
showing that the thickness of the basal lamina in an area of developmental stability
(interbud area) is thicker than the basal lamina in an area of greater activity (the
tips). Because the radius of curvature of the buds is +70/mi and that of the
interbuds is —10 fim, the geometry conceivably could compress the basal lamina in
the laminar plane to produce a thickening of approximately 0-6 %. The measured
increase in thickness is almost 50 %.
Occasionally, tannic acid staining reveals fibrils with a length of approximately
18nm normal to the surface of the basal lamina (Fig. 11). When two or three
appear close together, their spacing is about 100 nm. Faint indications of similar
structures can also be seen in routinely fixed tissue.
The basal surface of the epithelium is relatively smooth in the interbud area and
tips (Figs 12, 13). Cytoplasmic filaments subjacent to the surface are usually
aligned with respect to each other, but not with respect to an overall morphological feature, e.g. the axis of the bronchus, etc.
Localized electron-dense structures were seen on one mesenchymal cell close to
the basal lamina in a lung stained with alcian blue. Electron-dense plaques are not
apparent on cell membranes stained with ruthenium red. This may be due to
understaining with ruthenium red, although electron-dense granules are present in
the nearby extracellular spaces. Thin (3nm) filaments are delineated in the
extracellular spaces with ruthenium red, along which electron-dense granules,
10-30nm in diameter, are positioned. Ruthenium red granules, 10 nm in diameter, also attach to collagen fibres (Fig. 15, inset).
Fig. 4. The basal lamina at the tip of a newly emerged bud at 7 days is discontinuous
in the areas marked by the arrows. Immediately above the basal lamina lies a
mesenchymal cell. Tannic acid-fixed tissue. X94000. Bar, 100nm.
Fig. 5. The basal lamina at the interbud area of a 7-day lung is continuous, and is
traversed by finefilamentsacross the lamina lucida. Above the basal lamina is fibrillar
material. Tannic acid-fixed tissue, bl, basal lamina; ep, epithelium; mes, mesenchyme.
X94000. Bar, 100nm.
Figs 6, 7. Comparison of tip and interbud area of ruthenium red-fixed tissue, bl, basal
lamina. X94 000. Bar, 100 nm.
Fig. 6. A few large particles cover the tip. No regular substructure is seen.
Fig. 7. The basal lamina of the interbud area includes a diffusely staining band in the
region of the lamina densa bordered on both sides by tiny granules (marked by small
arrows).
196
B. C. GALLAGHER
Electron microscopic studies of avion lung
197
Distribution and orientation of collagen fibres
Collagen fibres with an interperiod spacing of 55 nm are commonly associated
with the mesenchymal surface of the basal lamina. They are more abundant in the
interbud area than at the tips, and the striated fibres constitute a major portion of
the fibrillar material that occurs at the epithelial-mesenchymal junction. The
diameter of individual collagen fibres increases significantly from 18 nm (±4-6 nm,
n = 20, five lungs) in 7-day lungs (stage 28) to 28 nm (±9-0nm, n = 20, five lungs)
one day later (P< 0-001). This pattern is seen even at new interbud areas. In two
instances near tips of bronchi, collagen fibres closely associated with each other
were observed to be in alignment with respect to interperiod spacing (Fig. 14). A
third grouping was found out of alignment, revealing that collagen alignment is
only an occasional feature of the chick lung.
While collagen fibres are not always aligned with each other, they are found to
be roughly oriented with respect to overall tissue morphology. Thus, between
bronchi, the fibres course through the saddle-like depression between the buds in a
direction concentric with the circumference of the cylindrical mesobronchus
(Fig. 15). A section grazing the lateral surface of the interbud mesobronchus
shows the fibres to continue around the side of the structure (Fig. 16).
At the tips, although fewer fibres are found, they appear to be oriented with
respect to the bud. The collagen at a 7-day unbranched tip is oriented mediolaterally with respect to the mesobronchial axis (Fig. 17). A grazing section of a
branched tip shows the collagen fibres parallel to the groove separating the new
buds on either side (data not shown). In the same section, but lateral to the
branching tip, the basement membrane shows little evidence of collagen, demonstrating that the fibres are not concentric around the axis of the growing tip in a
manner analogous to collagen around the mesobronchus. Mesenchymal cells do
not demonstrate any preferred orientation relative to the collagen fibres.
Polarizing microscopy
Polarizing microscopy further demonstrates the orientation of collagen fibres.
When a 3-bud lung is viewed through crossed polarizers, the presence of birefringence is detected by incomplete extinction of light. The compensator is rotated
Figs 8, 9. The appearance of the basal lamina when stained with alcian blue, x94000.
Bar, 100 nm.
Fig. 8. Cross section of the basal lamina is characterized by regular structures
indicated at arrows. The particles seem to form layers, as seen in the left central area.
The particles are in the area of the lamina lucida and lamina densa.
Fig. 9. A grazing section of the basal lamina. Some particles seem organized into
rows (marked by arrows).
Figs 10,11. Substructure in the basal lamina revealed by tannic acid staining, x94000.
Bar, 100 nm.
Fig. 10. The lamina lucida is traversed by thinfilamentswhich sometimes look Y- or
X-shaped (arrows).
Fig. 11. Filaments project from the lamina densa towards the fibrillar meshwork of
the basement membrane (arrows).
198
Bi!
' • • • < .
Electron microscopic studies ofavian lung
199
to a position where the specimen is seen in maximum contrast. As the specimen is
rotated in a plane perpendicular to the microscope axis, the patterns of light and
dark areas on the specimen also rotate, indicating that the light areas are not due
to light scattered by the specimen. In Figs 18,19, the specimen is viewed through
crossed polarizers at two different angles, 90° apart, while the polarizers remain in
a fixed position. The diagrams accompanying the photographs represent an
idealized rendition of the positions of extinction. In the first photograph, a dark
area appears at the tip and stalk of the bud and the lateral portions of the
mesobronchus. When the lung is rotated 90°, these areas are now light. The
anterior and posterior surfaces of the bud are bright in the first circumstance and
dark in the second. This is consistent with a radial distribution of collagen around
the tips, when viewed laterally, and a circumferential distribution around the
mesobronchus. The distal extension of the mesobronchus also shows the same
patterns of birefringence and, therefore, a radial distribution of collagen about the
tip.
DISCUSSION
The bronchial basement membrane regions of 7- and 8-day-old chick lungs were
examined by electron microscopy using tannic acid, ruthenium red and alcian blue
to preserve GAGs and increase their electron density (Luft, 1971; Behnke &
Zelander, 1976; Maupin & Pollard, 1983; Simionescu & Simionescu, 1975). It has
been demonstrated recently that tannic acid stains HA (Singley & Solursh, 1980),
and was employed in this study for that purpose.
Cell-to-cell contact
The importance of epithelial-mesenchymal interactions is well established
(Grobstein, 1954,1967; Cunha, 1976; Wessells, 1977; Goldin, 1980; Bernfield et al.
1984), yet the detailed mechanisms of these interactions are not known. In the
present study, epithelial cells extended cell processes through the basal lamina at
the tips of bronchi and approached mesenchymal cells within 5nm. Specialized
junctional complexes were not seen in the present study and dye-coupling techniques have not discovered any evidence of gap junctions in mammalian lungs in
Figs 12, 13. The basal laminae of tips and interbud area stained with tannic acid. In
both cases, the basal laminae are relatively smooth, ep, epithelium, x 11600. Bar,
Fig. 12. The tip of a bud.
Fig. 13. The interbud area. Slight convolutions are seen.
Fig. 14. Collagen fibrils are aligned in lateral register (arrows). Tannic-acid-fixed
tissue. This arrangement was found in the new interbud area of a branching bud.
X94000. Bar, lOOnm.
Fig. 15. Section demonstrating the alignment of collagen in known areas of the
basement membrane. Ruthenium-red-stained collagen sectioned in the plane indicated
by the diagram: the interbud area. Collagen fibres course through the saddle in a
direction concentric with the mesobronchial cylinder. X51000, inset x 190000. Bar,
100 nm.
200
B. C. GALLAGHER
epithelial-mesenchymal interactions (Ryan etal. 1984). Mesenchymal cell processes contacted the epithelium through gaps in the basal lamina at a nascent tip.
Intimate cell contacts were not seen prior to primary bud formation. They were
also not seen prior to branching at the tips of six stage-30 lungs.
Electron microscopic studies of avion lung
201
Intimate cell contacts have been seen in other organs undergoing branching
morphogenesis: the mouse lung (Bluemink etal. 1976), mouse salivary gland
(Cutler & Chaudrey, 1973; Coughlin, 1975) and mouse kidney during induction of
the mesenchyme by the ureteric bud (Lehtonen, 1975). In mouse lung and
submandibular gland (Bluemink etal. 1976), cell contacts formed after morphogenesis and were not seen until branching was well under way. Thus it seems that,
in these mouse tissues, cell contacts are not needed to initiate branching, but are
necessary for continued differentiation.
Basal laminar folding
The basement membrane of the chick lung is relatively smooth compared to the
highly folded structures seen in the clefts of salivary glands (Bernfield & Wessells,
1970; Spooner & Wessells, 1972). In salivary gland, the rapid formation of clefts is
considered a likely cause of the folds. In addition, the radius of curvature at
salivary clefts is much smaller than at lung branch points; hence the lack of folding
in lung basal laminae is not surprising.
Discontinuities in the basal lamina
The thinning of the basal lamina seen at growing tips has also been seen in
mammalian tissues (Coughlin, 1975; Lehtonen, 1975). The distance between the
lamina densa and the plasma membrane was significantly less at the tips than at the
interbud area. It has been proposed that the function of discontinuities of the basal
lamina is to allow contact between epithelium and mesenchyme (Bluemink et al.
1976; Goldin, 1980). Although the function of the contacts is not known, the close
interaction between epithelial and mesenchymal tissues is necessary for morphogenesis to take place. In the present study, gaps in the basal lamina in areas
expected to produce primary or secondary bud outgrowth were not seen, but if the
time in which basal laminar discontinuity precedes bud outgrowth is short, e.g. 1 h,
one would need to examine a much larger number of lungs in order to detect this
event.
Fig. 16. Section demonstrating the alignment of collagen in known areas of the
basement membrane. Tannic acid-stained collagen sectioned on the lateral surface of
the interbud area as indicated by the diagram. Mesenchymal cells show no particular
orientation, x 11600. Bar, 1/zm.
Fig. 17. Orientation of collagen at the tip of an unbranched bud in conventionally
fixed tissue. Collagen fibres are aligned lateromedially with respect to the mesobronchial axis. X50000. Bar, 100nm.
Figs 18, 19. Polarization microscopy of lungs. X125. Bar in Fig. 18, 0-1 mm.
Fig. 18. Specimen viewed through crossed polarizers. The compensator has been
adjusted so that its fast direction is vertical. An idealized version of the light and dark
areas is depicted in the diagram. Arrow shows where alignment of collagen fibres is
known through electron microscopy.
Fig. 19. The same specimen rotated through 90°.
Fig. 20. A schematic drawing of collagen fibres as interpreted by the results of
polarizing microscopy.
202
B.C.
GALLAGHER
infrastructure of basal lamina
Ruthenium red staining of the avian lung demonstrates the bilaminar arrangement of 15 nm particles along the lamina densa. The average interparticle
distance of 47 nm is somewhat less than the interparticle spacing seen in similarly
stained corneal tissue, 60 nm (Trelstad etal. 1974). Ruthenium red deposits were
seen in the lamina lucida of the mouse salivary gland with a repeat of
approximately 55 nm. In addition, 60-90nm ellipsoids with a period of 100 nm
were seen in the lamina densa (Cohn et al. 1977). While structures of this sort are
not seen in the present study using ruthenium red, they are visualized with alcian
blue staining.
In general, basal laminae of developing tissues seem to share common features:
a lamina densa with a bilaminar distribution of GAG particles on either side of it,
and anchoring filaments in the lamina lucida which connect the lamina densa and
the plasma membrane of the epithelium. This is seen in developing tissues,
branching or otherwise, chick and mouse tissue and is also seen in adult glomerular
basement membrane (Farquhar, 1981).
Collagen disfribution
In the present study, collagen was more abundant in the interbud areas of chick
lung than at the tips. Collagen reduces the degradation of extracellular GAG in
mammary tissue (David & Bernfield, 1979,1981), and it probably serves the same
purpose in the chick lung. The epithelial cytoskeleton may be responding to the
presence of collagen via plasma-membrane-associated complexes with extracellular matrix receptors (Hay, 1985) to stabilize the epithelial configuration.
Collagen orientation
The collagen was oriented in a circumferential direction around the mesobronchus, as demonstrated by grazing sections of the basement membrane.
Further elucidation of the collagen fibres was studied more easily by polarization
microscopy (see Slayter, 1970).
In a specimen such as appears in Fig. 18, the dark areas represent the areas
where collagen is oriented vertically. This interpretation is supported by the
electron microscope observation that collagen is oriented vertically at the arrow.
The dark areas vary continuously with rotation of the specimen, indicating that the
collagen is arranged radially about the bud. The same radial orientation is
observed around the distal extension of the mesobronchus, so that the overall
arrangement of collagen fibres can be interpreted as in Fig. 20.
The organization of collagen fibres is exactly opposite to that seen in the mouse
lung (Wessells, 1970), where a densely packed layer of collagen was oriented
parallel to the tracheal axis. These differences may reflect different tensions in the
developing mouse and chick lungs (Weiss, 1933; Stopak & Harris, 1982). Perhaps
collagen orientation enhances growth along the long axis in a manner analogous to
Electron microscopic studies ofavian lung
203
that seen in the plant Graptopetalum (Green & Poethig, 1982). Finally, some
aspect of 'self-assembly' might contribute to fibre alignment.
CONCLUSIONS
The developing chick lung was examined in the electron microscope to discover
features of branching morphogenesis. These are: close cell contacts between
epithelial and mesenchymal tissues, thin basal laminae at growing tips and thicker
basal laminae at the interbud area. Detailed substructures of the basal lamina are
similar to those seen in other developing tissues. A well-ordered lattice may be a
general rule in basal laminae of developing organs, and perhaps of all basal
laminae. Collagen orientation in the chick lung is perpendicular to that found in
the mouse lung, but the reasons for this are not clear. In general, the chick lung
uses strategies in its morphogenesis that are similar to those which have been
elucidated previously in developing mammalian organs.
I wish to thank Norman K. Wessells, Geoffrey Goldin, Paul Green and Steve Klein for
encouragement, support and thoughtful criticisms. Paul Green provided extensive aid for the
polarizing microscopy. I am grateful to Linda Jones for her expert help in typing this manuscript.
This work was supported by Grant no. HD-04708 to N.K.W. and a predoctoral NSF minority
fellowship.
REFERENCES
R. (1960). Morphogenetic interactions in the development of the mouse thymus
gland. Devi Biol. 2, 271-284.
BEHNKE, O. & ZELANDER, T. (1976). Preservation of intercellular substances by the cationic dye
Alcian blue in preparative procedures for electron microscopy. /. Ultrastruct. Res. 31,
424-438.
BERNFIELD, M. R. & BANERJEE, S. D. (1972). Acid mucopolysaccharide (glycosaminoglycan)
at the epithelial-mesenchymal interface of mouse embryo salivary glands. /. Cell Biol. 52,
664-673.
BERNFIELD, M. R., BANERJEE, S. D. & COHN, R. H. (1972). Dependence of salivary epithelial
morphology and branching morphogenesis upon acid mucopolysaccharide-protein (proteoglycan) at the epithelial surface. /. Cell Biol. 52, 674-689.
BERNFIELD, M., BANERJEE, S. D., KODA, J. E. & RAPRAEGER, A. C. (1984). Remodeling of the
basement membrane as a mechanism of morphogenetic tissue interaction. In The Role of the
Extracellular Matrix in Development (ed. R. L. Trelstad), pp. 545-572. New York: Alan R.
Liss, Inc.
BERNFIELD, M. R., COHN, R. H. & BANERJEE, S. D. (1973). Glycosaminoglycans and epithelial
organ formation. Amer. Zool. 13,1067-1083.
BERNFIELD, M. R. & WESSELLS, N. K. (1970). Intra- and extracellular control of epithelial
morphogenesis. 29th Sympos. Soc. devlBiol. Supplement 4, 195-249.
BLUEMINK, J. G., VAN MAURIK, P. & LAWSON, K. A. (1976). Intimate cell contacts at the
epithelial-mesenchymal interface in embryonic mouse lung. J. Ultrastruct. Res. 55, 257-270.
COHN, R. H., BANERJEE, S. D. & BERNFIELD, M. R. (1977). Basal lamina of embryonic salivary
epithelia. Nature of glycosaminoglycan and organization of extracellular material. /. Cell Biol.
73, 464-478.
COUGHLIN, M. D. (1975). Early development of parasympathetic nerves in mouse submandibular
gland. Devi Biol. 43, 123-139.
CUNHA, G. R. (1976). Epithelial-stromal interactions in development of the urogenital tract.
Int. Rev. Cytol. 47, 137-194.
AUERBACH,
204
B . C.
GALLAGHER
L. S. & CHAUDREY, A. P. (1973). Intracellular contacts at the epithelial-mesenchymal
interface during the prenatal development of the rat submandibular gland. Devl Biol. 33,
229-240.
DAVID, G. & BERNFIELD, M. R. (1979). Collagen reduces glycosaminoglycans degradation by
cultured mammary epithelial cells: possible mechanism for basal lamina formation. Proc.
natn. Acad. Sci. U.S.A. 76, 786-790.
DAVID, G. & BERNFIELD, M. (1981). Type I collagen reduces the degradation of basal lamina
proteoglycan by mammary epithelial cells. /. Cell Biol. 91, 281-286.
DERBY, M. A. & PINTAR, J. E. (1978). The histochemicalspecificity oiStreptomyceshyaluronidase
and chondroitinase ABC. Histochem. J. 10, 529-547.
FARQUHAR, M. G. (1981). The glomerular basement membrane. A selective macromolecular
filter. In Cell Biology of Extracellular Matrix (ed. E. D. Hay), pp. 335-378. New York: Plenum
Press.
GOLDIN, G. V. (1980). Towards a mechanism for morphogenesis in epitheliomesenchymal
organs. Q. Rev. Biol. 55, 251-265.
GORDON, J. & BERNFIELD, M. R. (1980). The basal lamina of the postnatal mamary epithelium
contains glycosaminoglycans in a precise ultrastructural organization. Devi Biol. 74,118-135.
GREEN, P. B. & POETHIG, R. S. (1982). Biophysics of the extension and initiation of plant organs.
In Developmental Order: Its Origin and Regulation. 40th Symposium Soc. devl Biol. (ed. S.
Subtelny & P. B. Green), pp. 485-509. New York: Alan R. Liss.
GROBSTEIN, C. (1954). Tissue interaction in the morphogenesis of mouse embryonic rudiments
in vitro. In Aspects of Synthesis and Order in Growth. Symp. Soc. devl Biol., vol. 13 (ed. D.
Rudnick), pp. 233-256. Princeton, N.J.: Princeton Univ. Press.
GROBSTEIN, C. (1967). Mechanisms of organogenetic tissue interaction. Natn. Cancer Inst.
Monogr. 26, 279-299.
GROBSTEIN, C. & COHEN, J. (1965). Collagenase: effect on the morphogenesis of embryonic
salivary epithelium in vitro. Science 150, 626-628.
HAMBURGER, V. & HAMILTON, H. L. (1951). A series of normal stages in the development of the
chick embryo. /. Morph. 88, 49-92.
HAY, E. D. (1978). Fine structure of embryonic matrices and their relation to the cell surface in
ruthenium red-fixed tissues. Growth 42, 399-423.
HAY, E. D. (1985). Matrix-cytoskeletal interactions in the developing eye. /. Cell Biochem. 27,
143-156.
HAY, E. D. & MEIER, S. (1974). Glycosaminoglycan synthesis by embryonic inductors: neural
tube, notochord and lens. /. Cell Biol. 62, 889-898.
LEHTONEN, E. (1975). Epithelio-mesenchymal interface during mouse kidney tubule induction
in vivo. J. Embryol. exp. Morph. 34, 695-705.
LINKER, A., MEYER, K. & HOFFMAN, P. (1960). The production of hyaluronate oligosaccharides
by leech hyaluronidase and alkali. /. biol. them. 235, 924-927.
LUFT, J. H. (1971). Ruthenium red and violet. I. Chemistry, purification, methods of use for
electron microscopy and mechanism of action. Anat. Rec. 171, 347-368.
CUTLER,
MAUGER, A., DEMARCHEZ, M., HERBAGE, D., GRIMAUD, J. A., DRUGUET, M., HARTMANN, D. &
SENGEL, P. (1982). Immunofiuorescent localization of collagen types I and III, and of
fibronectin during feather morphogenesis in the chick embryo. Devl Biol. 94, 93-105.
P. & POLLARD, T. D. (1983). Improved preservation and staining of HeLa cell actin
filaments, clathrin-coated membranes and other cytoplasmic structures by tannic acidglutaraldehyde-saponin fixation. /. Cell Biol. 96, 51-62.
RUDNICK, D. (1933). Developmental capacities of the chick lung in chorioallantoic grafts. /. exp.
Zool. 66, 125-154.
RYAN, U., SLAVKTN, H., REVEL, J.-P., MASSARO, D. & GAIL, D. (1984). Cell-to-cell interactions in
the developing lung. Report of a conference held 3-5 June 1984. Tissue and Cell 16, 829-841.
MAUPIN,
SAXEN, L., LEHTONEN, E., KARKINEN-JAASKELAINEN, M., NORDLING, S. & WARTIOVAARA, J.
(1976). Are morphogenetic tissue interactions mediated by transmissible signal substances or
through cell contacts? Nature, Lond. 259, 662-663.
SIMIONESCU, N. & SIMIONESCU, M. (1976). Galloglucoses of low molecular weight as mordant in
electron microscopy. Part I. J. Cell Biol. 70, 608-621.
SINGLEY, C. T. & SOLURSH, M. (1980). The use of tannic acid for the ultrastructural localization of
hyaluronic acid. Histochem. 65, 93-102.
Electron microscopic studies of avion lung
205
SLAYTER, E. M. (1970). Optical Methods in Biology, chaps 6 & 15. New York: Wiley-Interscience.
SPOONER, B. S. & WESSELLS, N. K. (1972). An analysis of salivary gland morphogenesis: role of
cytoplasmic microfilaments and microtubules. Devi Biol. 27, 38-54.
D. & HARRIS, A. K. (1982). Connective tissue morphogenesis by fibroblast traction.
Devi Biol. 90, 383-398.
TADERERA, J. V. (1967). Control of lung differentiation in vitro. Devi Biol. 16, 489-512.
TRELSTAD, R. L., HAYASHI, K. & TOOLE, B. P. (1974). Epithelial collagens and glycosaminoglycans in the embryonic cornea. Macromolecular order and morphogenesis in the basement
membrane. /. Cell Biol. 62, 815-830.
WEISS, P. (1933). Functional adaptation and the role of ground substances in development. Am.
Nat. 67, 322-340.
WESSELLS, N. K. (1970). Mammalian lung development: interactions in formation and morphogenesis of tracheal buds. /. exp. Zool. 175, 455-466.
WESSELLS, N. K. (1977). Tissue Interaction and Development. MenloPark, Ca.: W. A. Benjamin,
Inc.
STOPAK,
(Accepted 21 January 1986)