Download PDF

J. Embryol. exp. Morph. 82, 147-161 (1984)
Printed in Great Britain © The Company of Biologists Limited 1984
147
Role of transferrin in branching morphogenesis,
growth and differentiation of the embryonic
kidney
By IRMA THESLEFF AND PETER EKBLOM
Department of Pathology, University of Helsinki, Haartmaninkatu 3, 00290
Helsinki 29, Finland
SUMMARY
Our previous work has suggested that transferrin is an important serum component for
differentiation of the kidney. In this study we have analysed more closely the response of
cultured mouse embryonic kidney to exogenous transferrin and the dependence of kidney
tubule induction on transferrin. Our results show that transferrin causes a dose-dependent
increase in cell proliferation in the differentiating kidney mesenchyme, but no stimulation
of cell proliferation in the inductor tissue used, the embryonic spinal cord. In cultures of
whole kidney rudiments a remarkable increase in the amounts of DNA and protein are
caused by transferrin but not by other serum components present in a transferrin-depleted
serum. The morphology of the explants was similar when cultured in the presence of human
serum and in the transferrin-depleted serum supplemented with transferrin. In transferrincontaining chemically-defined medium the explants flattened and spread out, but the morphology of the kidney tubules was similar as in explants cultured in the presence of serum.
Examination of the cultured explants by electron microscopy showed that in all transferrincontaining culture media the mesenchymal cells had differentiated into kidney tubules
consisting of epithelial cells lined by a basement membrane. The experiments with the
transferrin-depleted serum demonstrate that the main mitogen for kidney development is
transferrin, and that other serum factors are mainly required for maintenance of tissue
compactness.
Our earlier studies have shown that exogenous transferrin is not needed for certain changes
preceding overt tubule formation in the kidney mesenchyme, and we suggested that transferrin
responsiveness is acquired during the induction of kidney mesenchyme. Our present results
do not contradict the postulate, although they demonstrate that the acquisition of the responsiveness is more complicated than previously thought. When the mesenchyme is exposed to
inductor tissue for 24 h without transferrin, and then subcultured without the inductor in the
presence of transferrin, morphogenesis fails and there is no proliferation of the mesenchyme.
The experiment shows that the inductor, the mesenchyme and transferrin must all three be
simultaneously present for the acquisition of the transferrin responsiveness. Other experiments show that the induced mesenchyme can be a direct target tissue, since it can proliferate
in response to transferrin also in the absence of the inductor. It is evident that the inductor is
required for the acquisition of the responsiveness, as suggested. However, there is apparently
a large overlap between the transferrin-independent and transferrin-dependent proliferation.
The mesenchyme is not a synchronous cell population and cells do not become induced and
transferrin-responsive at the same time. Therefore, in the organ culture, it is necessary to have
transferrin present also during induction. Although this explanation seems most likely,
we cannot exclude that transferrin has two actions, one measurable direct effect on the
proliferation of induced mesenchymes, and another yet unidentified effect on the induction
process.
148
I. THESLEFF AND P. EKBLOM
INTRODUCTION
Early kidney development is characterized by branching of the ureteric bud
epithelium and differentiation of the metanephrogenic mesenchyme surrounding the ureteric bud into epithelial kidney tubules. These synchronous events are
regulated by interactions between the ureteric epithelium and the mesenchyme
(Grobstein, 1955; Saxen etal. 1968). The mechanisms underlying these developmental processes have been studied using organ cultures of either whole kidney
rudiments or transfilter cultures where the epithelial and mesenchymal tissues
have been separated. In transfilter cultures the epithelial ureteric bud is usually
replaced by spinal cord tissue which is another potent inducer of kidney tubules
(Grobstein, 1955, 1956). It has been shown that kidney tubules are induced in
this system during the initial 24 h culture period and that tubule formation then
proceeds to advanced stages even if the inducer is removed at this time (Grobstein, 1967; Saxen etal. 1968; Ekblom etal. 1981a).
By using cultures of whole kidneys and transfilter explants we have shown that
exogenous transferrin is necessary for differentiation of kidney tubules (Ekblom,
Thesleff, Miettinen & Saxen, 1981ft). The induced kidney mesenchyme responds
to transferrin by enhanced proliferation whereas uninduced mesenchymes
remain unresponsive. Furthermore, transferrin is not needed during the 24 h
induction period for certain characteristic changes which precede overt tubule
formation in the mesenchyme. These include a rapid increase in cell proliferation
between 12 and 24 h of culture and a shift in the extracellular matrix composition
from a mesenchymal to an epithelial type. Based on these observations we have
suggested that the responsiveness to transferrin is acquired during the induction
of kidney mesenchyme (Ekblom et al. 1983a).
In the present study, we have in detail studied two major aspects of the serum
requirements. First, we have compared the effects of transferrin and a
transferrin-depleted serum using a number of quantitative assays for cell growth,
and studied the morphology by electron microscopy. This was done with whole
kidneys since both the growth parameters and the compactness of the tissue
could be well evaluated in these. Our data suggest that transferrin is the main
mitogen for kidney differentiation, while other components are required for the
maintenance of the compactness of the organ. Secondly, we have in detail investigated the relationship between embryonic induction and the acquisition of the
transferrin responsiveness. This was performed in transfilter cultures which
allow a good timing of the events and make it possible to remove the inductor
at various times of culture (Grobstein, 1956).
MATERIALS AND METHODS
Organ culture
Developing kidneys from 11-, 12- and 13-day-old hybrid mouse embryos,
Stimulation of kidney differentiation by transferrin
149
CBA/C57B1 (day of vaginal plug = Day 0) were used. For transfilter cultures
metanephric mesenchymes were dissected from day-11 kidneys in 0-02%
EDTA. Two mesenchymes were used in each explant and a piece of spinal cord
from the same embryos was used as an inductor on the opposite side of the filter
(Grobstein, 1956). A Trowell-type organ culture system was used in which the
transfilter explants as well as whole kidney explants were supported by a
Nuclepore filter (pore size 1-0 jum, thickness lO/im) on a metal grid. Improved
Minimum Essential Medium (I-MEM, Richter, Sanford & Evans, 1972) was
prepared by adding amino acids, minerals and vitamins but no growth factors to
Eagle's Minimum Essential Medium (Gibco, Paisley, Scotland). Glutamine
(4 mM) was added to the medium prior to culture. Human transferrin (iron-free)
was purchased from Sigma Chemical Company (St. Louis, Missouri). Normal
human serum (NHS, Blood Transfusion Center, Finland) was used in control
cultures and for transferrin-depleted serum, which was kindly prepared by A.
Miettinen (Department of Bacteriology, Univ. of Helsinki). Transferrin was
removed by immunoadsorption as previously described (Ekblom et al. 1983<z).
As judged by the double immunodiffusion method, less than 0-3 jug/ml of transferrin is present in cultures when 10 % of the transferrin-depleted serum is used.
Histology
For light microscopy the explants were fixed in Zenker's solution, embedded
in Tissue Prep (Fisher Scientific Co., Fairlawn, N. J.), serially sectioned at 5 jum
and stained with haematoxylin-eosin. For electron microscopy the explants were
fixed with 2-5 % glutaraldehyde in 0-1 M-phosphate buffer (pH7-4). They were
postfixed with 1 % OsO4 in 0-1 % phosphate buffer, dehydrated in ethanol and
embedded in Epon. One micron sections were stained with methylene blue and
examined in light microscope. Silver to grey sections were taken for electron
microscopy. They were poststained with uranyl acetate and lead citrate and
examined with a Hitachi HS-7S electron microscope.
Determinations of thymidine incorporation, and DNA and protein contents
[3H]thymidine (specific activity 18-25 Ci/mmol, The Radiochemical Centre,
Amersham, Buckinghamshire, England) was added to the culture medium for
3h at a concentration of 20juCi/ml. The explants were then placed on ice and
sonicated in 0-5 ml distilled water. Aliquots were taken for the determination of
radioactivity into trichloroacetic acid precipitable material (0-2 ml) and for
measurement of DNA content (0-25ml). In the smaller tissue samples, only
DNA content could be measured (Nordling & Aho, 1971), but in 13-day whole
kidneys, protein content was also determined (Lowry, Rosebrough, Farr &
Randall, 1951). The spinal cord was left in place in transfilter explants during the
[3H] thymidine pulse only in experiments where thymidine incorporation in the
spinal cord tissue was studied (Fig. 2). In all other experiments the spinal cord
was removed prior to the radioactive pulse. In such explants the incorporation
150
I. THESLEFF AND P. EKBLOM
of label to the mesenchymes was higher than in explants in which the spinal cord
was present during labelling.
RESULTS
Effects of transferrin on proliferation in transfilter cultures
Metanephric mesenchymes were cultured transfilter with a piece of spinal cord
tissue. We now examined the effects of transferrin by measuring [3H]thymidine
incorporation in the inducer tissue and in the mesenchyme after 42 h of culture.
Transferrin stimulated kidney tubule differentiation (Fig. 1A, B) and increased
thymidine incorporation in the differentiating mesenchyme in a dose-dependent
manner (Fig. 2). The inductor tissue, the spinal cord, was unresponsive to transferrin at all concentrations tested (Fig. 2).
The effect of transferrin on kidney tubulogenesis can apparently not be replaced
by other serum components, since morphological differentiation does not take
place in the presence of serum from which transferrin has been removed (Ekblom
etal. 1983a). We now analysed thymidine incorporation into kidney mesenchymes
after 42 h in transfilter cultures. In the presence of transferrin-depleted serum
proliferation was stimulated about two-fold as compared to explants cultured in a
transferrin-free, chemically-defined medium. The addition of 50 //g/ml transferrin to the medium containing transferrin-depleted serum caused an increase in
incorporation to the same level as with normal human serum (Fig. 3). These
observations indicated that also in the serum-containing culture media, differentiation of kidney tubules is associated with elevated cell proliferation.
Fig. 1. Effect of transferrin on kidney tubulogenesis in transfilter cultures of spinal
cord and metanephric mesenchyme after 72 h in culture. (A) In chemically defined
medium (I-MEM) without transferrin the mesenchyme on top of the filter has
remained undifferentiated. (B) When the medium was supplemented with 50 jug/ml
transferrin epithelial kidney tubules have differentiated. Haematoxylin-eosin staining. X400.
Stimulation of kidney differentiation by transferrin
2
5
Transferrin (fig/ml)
151
50
Fig. 2. Dose-response curve of the effect of transferrin on [3H]thymidine incorporation in transfilter cultures. A clear dose response is evident in kidney mesenchymes
(—) whereas the inductor tissue (---) is unresponsive to transferrin. 20juCi/ml
[3H]thymidine was present in the medium between 42 and 45 h of culture. One
explant comprised two kidney mesenchymes and a piece of spinal cord tissue as the
inductor. The inductor was present during labelling of the tissue unlike in the other
experiments (Figs 3 and 4) where it was removed before the radioactive pulse. Each
point represents the mean of 3-5 explants.
Dependence of the transferrin responsiveness on embryonic induction
The kidney mesenchyme is induced to form kidney tubules during the first 24 h
of transfilter culture (Grobstein, 1967; Saxen etal. 1968; Ekblom et al. 1981a).
At this time tubules have not yet formed but the inductor tissue can be removed
and tubule formation proceeds in the mesenchyme. During the 24 h induction
period there is an increase in cell proliferation in the mesenchyme which is not
152
I. THESLEFF AND P. EKBLOM
dependent on exogenous transferrin (Ekblom etal. 1983a). To study the dependence of the induction on transferrin we exposed the explants to transferrin at
different times of culture and measured thymidine incorporation at 42 h of culture. When the kidney mesenchymes were exposed to exogenous transferrin as
well as to the inductor tissue throughout the 42 h culture, thymidine incorporation was at SSOOOc.p.m./jUg DNA (Fig. 4A). The removal of the inductor tissue
after 24 h of culture caused a decrease in the proliferation rate at 42 h, suggesting
that when the inductor is present the induction of new tubules proceeds even
after the first 24h of culture (Fig. 4B).
When the explants were cultured without transferrin for the entire culture
period, differentiation did not take place and this was associated with a very low
level of thymidine incorporation (3000c.p.m./jUg DNA, Fig. 4C). If transferrin
was added to the cultures after 24 h and the inductor was left in place, thymidine
incorporation at 42 h was at a higher level than in explants cultured in the
presence of transferrin throughout the culture (Fig. 4D). However, if in such
cultures the inductor was removed at 24h, i.e. at the same time that transferrin
was added to the medium, thymidine incorporation stayed throughout culture at
the same low level as in the explants cultured without transferrin (Fig. 4E) and
the mesenchymes did not differentiate.
Fig. 3. The effect of removal of transferrin from human serum on [3H]thymidine incorporation. The kidney mesenchymes were cultured as transfilter explants and
labelled between 42 and 45 h of culture. Each bar represents the mean of three explants
comprising two mesenchymes. A In I-MEM supplemented with 10 % transferrindepleted serum thymidine incorporation was at a low level although somewhat higher
than in chemically defined medium (compare to Fig. 4C). No differentiation took place
in this medium. B The addition of 50 |Ug/ml transferrin to this medium stimulated differentiation as well as proliferation. C Normal human serum supported thymidine incorporation similar to transferrin-depleted serum supplemented with transferrin (B).
Stimulation of kidney differentiation by transferrin
153
<
Z 5
Q
2
X
E
Q.
Day
Inductor
Transferrin
1+1+
A
Fig. 4. The dependence of the proliferative response in the kidney mesenchymes on
exogenous transferrin and the inductor tissue. 20 juCi/ml [3H]thymidine was given to
the cultures during 42-45 h of culture. Each bar represents the mean of 3-5 explants
comprising two mesenchymes. A When transferrin was present in the medium and
the inductor was left in place for the whole culture period (days 1 and 2), high
thymidine incorporation was evident. B The removal of the inductor tissue after 24 h
(day 1) of culture caused a decrease in incorporation. C The inductor tissue did not
support proliferation in the absence of transferrin. D When the explants were cultured without transferrin for the first day, and transferrin was then added to the
medium, incorporation at 42 h (day 2) was higher than in A. E If the explants were
cultured as in D but the inductor tissue was removed at 24 h (at the same time that
transferrin was added to the medium), very low thymidine incorporation was
measured on day 2 and the mesenchymes did not differentiate.
Effects of transferrin on branching morphogenesis of whole kidneys
Transferrin was shown to be necessary for branching morphogenesis. In the
presence of 10 % human serum the ureteric bud of 11-day kidney rudiments
154
5A
I. THESLEFF AND P. EKBLOM
P '*
B
Fig. 5. The dependence of branching morphogenesis on transferrin. The ureter bud
of the day-11 kidney rudiments had not underwent branching prior to culture. (A)
After 3 days of culture branching is evident when the kidneys were cultured in IMEM supplemented with 10 % human serum. (B) No branching has taken place in
a medium containing 10 % transferrin-depleted serum. (C) The addition of 50 /ig/ml
transferrin to the medium in B has restored the ability of normal human serum to
support branching. (D) In a chemically defined medium supplemented with 50 jUg/ml
transferrin branching morphogenesis is evident although the explant has lost its
compactness and it spreads out.
underwent branching morphogenesis which was accompanied by tubule formation in the surrounding mesenchyme (Fig. 5A). The substitution of the normal
serum by transferrin-depleted serum resulted in inhibition of branching of the
ureter as well as tubular differentiation (Fig. 5B). However, in this transferrindepleted medium the explants were compact even after 2 days of culture and they
Stimulation of kidney differentiation by transferrin
155
were well preserved as seen in light and electron microscopy (Figs 6A, 7B). The
addition of 50 //g/ml transferrin to this medium restored the ability of the serum
to support branching as well as tubulogenesis (Fig. 5C) and the explants were
identical to those cultured in the presence of normal human serum. Also in the
chemically defined medium supplemented with 50 jug/ml transferrin branching
and tubule formation took place but the explants flattened and spread out and
the mesenchymal tissue surrounding the kidney tubules regressed (Fig. 5D). The
morphology of the kidney tubules was, however, indistinguishable from explants
cultured with serum (Fig. 6B). Electron microscopy of the kidney tubules in
explants cultured in the various transferrin-containing culture media showed that
they were surrounded by a normal basement membrane (Fig. 7A, C, D),
whereas the cells in the explants cultured in the transferrin-depleted medium
continued to show a mesenchymal morphology (Fig. 7B).
Transferrin profoundly affected the DNA content of whole kidney explants.
When 12-day kidneys were cultured without transferrin, either in transferrindepleted serum or in a chemically defined medium the DNA content stayed at
the level of the onset of culture for the first 2 days. By day 4 the amount of DNA
decreased indicating degeneration of the explants. The addition of 50/ig/ml
transferrin to either of the media caused a two-fold increase in the DNA content
by day 2 and a three-fold increase by day 4 (Fig. 8). A similar increase is not seen
in transfilter cultures (Ekblom et al. 1983<z).
The effect of transferrin on the protein content was studied in kidneys from
12£-13-day embryos. In these larger-size explants transferrin caused an
approximately six-fold increase in the amount of proteins on day 3 of culture and
this was accompanied by a four-fold increase in the DNA content (Fig. 8). By day
5 the amount of proteins as well as of DNA had somewhat decreased, which was
evidently due to poorer survival of the larger-size explants. The protein content
Fig. 6. Histological picture of a day-11 kidney rudiment cultured for 3 days in the
presence of transferrin-depleted serum (A) and chemically defined medium containing transferrin (SOjUg/ml) (B). In transferrin-depleted serum the ureter bud has
neither branched, nor has the mesenchyme differentiated, but the preservation of
the tissue is good. In the chemically defined medium containing transferrin welldeveloped tubules are seen although the surrounding tissue has flattened. x200.
156
I. THESLEFF AND P. EKBLOM
Stimulation of kidney differentiation by transferrin
A
#
O
A
I-MEM + Tf
serum + Tf
serum - Tf
I-MEM
2
4
157
I
3
Time (days)
5
3
5
Fig. 8. The effect of transferrin on the DNA content of day-12 embryonic kidneys
(left) and on the DNA and protein contents of day 121-13 embryonic kidneys.
Transferrin caused a two-fold increase in the DNA content of day-12 kidneys by day
2 and a three-fold increase by day 4. Similar effect was seen both when transferrin
was added to chemically defined medium and to medium containing 10%
transferrin-depleted serum. Without exogenous transferrin the DNA content first
stayed at the level of the start of culture for 2 days and then decreased by day 4.
In day 12^—13 kidneys the DNA content increased four-fold and the protein content
six-fold by day 3 of culture in the presence of transferrin. Thefindingswere identical
in chemically defined medium and in medium containing serum. In the presence of
transferrin-depleted serum the DNA content stayed at the level of the start of
culture for the first 3 days but the amount of proteins increased three-fold. In all
culture media the amounts of DNA and protein decreased during prolonged
culture.
Fig. 7. Electron micrographs of day 11 whole kidney explants cultured for 3 days in
the same media as the explants in Fig. 5. e = epithelial cell, m = mesenchymal cell,
arrows indicate the basement membrane, x 10 000. (A) Ultrastructure of epithelial
cells of a kidney tubule which has differentiated in the presence of 10 % normal
human serum. The cells are lined by a basement membrane consisting of a typical
basal lamina with associated filamentous material. (B) Undifferentiated mesenchymal cells in an explant cultured in the presence of transferrin-depleted serum. (C)
The ultrastructure of the epithelial cells and the basement membrane are similar in
kidney tubules differentiated in the presence of transferrin-depleted serum
supplemented with transferrin (C) and in a chemically defined medium supplemented with 50jUg/ml transferrin (D), as compared to the explant cultured with normal
human serum (A).
EMB82
158
I. THESLEFF AND P. EKBLOM
of the explants increased somewhat also in the absence of transferrin. This
increase was more marked in the medium containing 10 % transferrin-depleted
serum, indicating a positive effect of serum components on the metabolic activity
of the explants. The DNA content did not increase in the absence of transferrin
(Fig. 8).
DISCUSSION
We have shown that the differentiating kidney mesenchyme responds to transferrin by a dose-dependent increase in thymidine incorporation. This reflects an
increase in cell proliferation, since the amount of DNA in whole kidneys increased significantly during culture when transferrin was present in the medium.
Transferrin concentrations which support proliferation in monolayer cultures
are in the same range (Vogt, Mishell & Dutton, 1969; Taub, Chuman, Saier &
Sato, 1979; Rizzino & Crowley, 1980; Barnes & Sato, 1980) as those stimulating
kidney development in the organ cultures. It is apparent that differentiation and
transferrin-dependent proliferation are closely linked during morphogenesis of
the kidney. The necessary serum component for myoblast fusion, the 'muscle
trophic factor' was recently identified as transferrin (Ii, Kimura & Ozawa, 1982).
We suggest that the primary effect of transferrin in both these developmental
systems is a stimulation of cell proliferation.
Our morphological data show that branching morphogenesis does not take
place in the absence of transferrin although other serum factors are present.
Quantitative measurements of DNA and protein content as well as cell proliferation revealed that the addition of transferrin to the transferrin-depleted serum
enhanced growth to the same levels as seen in the presence of normal serum, and
this was accompanied by branching morphogenesis. Ultrastructural studies
showed that the epithelium and its basement membrane formed just as well in
serum as in the defined medium with transferrin.
The transferrin-depleted serum, however, contained factors which were
beneficial for morphogenesis. Growth was slightly better in it than in the chemically defined, transferrin-free medium. The serum components had a clear effect
on the overall morphology of the explants, since the compactness of the tissues
was well preserved and the explants did not spread as they did in the chemically
defined medium. The unidentified serum components responsible for these
effects were non-dialysable. These factors are, however, not fibronectin or
laminin (Thesleff et al. 1983, unpublished). In conclusion, it appears that the
main mitogen in serum for kidney tubule differentiation is transferrin, while
some other factors participate in the maintenance of the compactness of the
organ.
The nature of the transferrin requirements was studied in detail in transfilter
cultures. The responsiveness of the metanephric mesenchyme to transferrin
seems to be regulated by inductive tissue interactions, since isolated uninduced
Stimulation of kidney differentiation by transferrin
159
mesenchymes do not respond to transferrin by cell proliferation (Ekblom et al.
1983a). Since our present study demonstrated that the inductor tissue, the spinal
cord, also did not respond to transferrin, it appears that the target tissue for the
action of transferrin during kidney tubule development is the induced mesenchyme. The role of the inductor is to make the cells responsive to transferrin.
Previous studies have shown that the inductor tissue affects the composition of
the extracellular matrix, and it is also mitogenic for the mesenchyme. These two
effects of the inductor are in no way affected by transferrin, nor can transferrin
mimick these effects (Ekblom etal. 1983a). Taken together, these results suggest
that transferrin is not required for induction, but rather that the responsiveness
to transferrin is acquired as a consequence of induction. It is possible that the
inductor tissue acts by increasing the number of transferrin-responsive cells.
Indeed, when transferrin is continuously present, the rate of thymidine incorporation at 42 h was higher in mesenchymes cocultured with inductor tissue for
42 h as compared with those explants where the inductor tissue was removed at
24 h. It suggests that the inductor tissue continuously can increase the number of
transferrin-responsive cells also after the first 24h of culture. This effect of the
inductor goes largely undetected in morphometrical studies (Saxen & Lehtonen,
1978).
These results thus seem to provide strong evidence that transferrin has no role
in the acquisition of the transferrin responsiveness, or in the related induction
process. The crucial experiment is therefore to induce the mesenchyme in the
absence of transferrin for 24 h, and then to remove the inductor and add transferrin. In such experiments, there are unexpectedly no signs of morphogenesis,
and the mesenchyme does not respond at 42 h to transferrin by proliferation.
Hence, a transferrin-dependent proliferation does not occur unless the mesenchyme, the inductor and transferrin all three simultaneously are present for some
time. The significance of this is difficult to interpret at present. It could mean that
transferrin nevertheless is needed for the acquisition of transferrin responsiveness. However, this seems very unlikely since serum factors do not induce their
own responsiveness. Another explanation of the result is that transferrin is
required for some yet unknown process during induction. Thirdly, it could be
argued that the inductor has some role in the delivery of transferrin into the
mesenchyme. It would leave unexplained, however, why induced cells at the
later stages can proliferate in the absence of inductor tissue in response to transferrin (Ekblom et al. 1983a). A fourth possibility is that certain cells during the
induction period already require transferrin while others are still induced to
become responsive. Because of this asynchrony, there is a certain overlap between the transferrin-independent and transferrin-dependent periods. It is well
known that some cells are induced already at 12-16 h in vitro (Nordling,
Miettinen, Wartiovaara & Saxen, 1971; Lehtonen, 1976; Saxen & Lehtonen,
1978) and at 24 h a threshold has been reached when sufficient cells have been
induced so that a complete tubular part of the nephron can form (Ekblom et al.
160
I. THESLEFF AND P. EKBLOM
1981&). Measurements of the cell cycle have also shown that the mesenchyme is
not a synchronized cell population at any time (Saxen, Salonen, Ekblom &
Nordling, 1983). Morphogenesis would thus fail in any situation where the
amount of induced cells is not sufficient. In the experiment where transferrin is
added only after removal of the inductor tissue, this would occur since all cells
which were induced early (at 12-16 h) degenerate as they do not receive transferrin early enough. In the experiment where the inductor is not removed, the
inductor continuously induces more transferrin-responsive cells and the
degenerated cells are of no major significance. If one wishes to assume that
transferrin has only one function in the model studied, then the explanation
which considers the asynchrony of the cell population offers the best explanation
for our previous and present results.
It is possible that some of these issues will be resolved by studying the transferrin
receptors. It is known that transferrin exerts its effects by binding to specific cell
surface receptors. The transferrin receptor is found in increased amounts on
proliferating cells and the regulation of the density (Shindelman, Ortmeyer &
Sussman, 1981; Trowbridge & Lopez, 1982; Newman et al. 1982) and
distribution (Ekblom, Thesleff, Lehto & Virtanen, 1983/?) is apparently important for the regulation of cell growth. It could therefore be speculated that the
responsiveness of the kidney mesenchyme to transferrin is regulated by the
expression of this receptor. We are now investigating whether the inductor tissue
acts by turning on the synthesis of the transferrin receptor.
Supported by grants from Finska Lakaresallskapet, Sigrid Juselius Foundation, and the
Paulo Foundation.
REFERENCES
D. & SATO, G. (1980). Serum-free cell culture, a unifying approach. Cell 22,
649-655.
EKBLOM, P. (1981). Formation of basement membranes in the embryonic kidney. Animmunohistological study. /. Cell Biol. 91, 1-10.
BARNES,
EKBLOM, P., MIETTINEN, A., VIRTANEN, I., WAHLSTROM, T., DAWNAY, A. & SAXEN, L.
(1981a). In vitro segregation of the metanephric nephron. Devi Biol. 84, 88-95.
P., THESLEFF, I., MIETTINEN, A. & SAXEN, L. (19816). Organogenesis in a defined
medium supplemented with transferrin. Cell Differ. 10, 281-288.
EKBLOM, P., THESLEFF, I., SAXEN, L., MIETTINEN, A. & TIMPL, R. (1983a). Transferrin as a
fetal growth factor. Acquisition of responsiveness related to embryonic induction. Proc.
natn. Acad. ScL, U.S.A. 80, 2651-2655.
EKBLOM, P., THESLEFF, I., LEHTO, V.-P. & VIRTANEN, I. (19836). Distribution of the transferrin receptor in normal human fibroblasts and fibrosarcoma cells. Int. J. Cancer 31,
111-117.
GROBSTEIN, C. (1955). Inductive interaction in the development of the mouse metanephros.
/. exp. Zool. 130, 319-340.
GROBSTEIN, C. (1956). Trans-filter induction of tubules in mouse metanephrogenic mesenchyme. Expl Cell Res. 10, 424-440.
GROBSTEIN, C. (1967). Mechanisms of organogenetic tissue interaction. Natn. Cane. Inst.
Monographs 26, 279-299.
EKBLOM,
Stimulation of kidney differentiation by transferrin
161
Ii, I., KIMURA, I. & OZAWA, E. (1982). A myotrophic protein from chick embryo extract, its
purification, identity to transferrin, and indispensability for avian myogenesis. Devi Biol.
94, 366-377.
LEHTONEN, E. (1976). Transmission of signals in embryonic induction. Med. Biol. 54,
108-128.
LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent. /. biol. Chem. 193, 265-275.
NEWMAN, R., SCHNEIDER, C , SUTHERLAND, R., VODINELICH, L. & GREAVES, M. (1982). The
transferrin receptor. Trends Biochem. Sci. 7, 397-400.
NORDLING, S. & AHO, S. (1971). Fluorimetric microassay of DNA using a modified thiobarbituric acid assay. Anal. Biochem. 115, 260-266.
NORDLING, S., MIETTINEN, H., WARTIOVAARA, J. & SAX6N, L. (1971). Transmission and spread
of embryonic induction. I. Temporal relationships in transfilter induction of kidney tubules
in vitro. /. Embryol. exp. Morph. 26, 231-252.
RICHTER, A., SANFORD, K. K. & EVANS, V. J. (1972). Influence of oxygen and culture media
on plating efficiency on some mammalian tissue cells. /. natn. Cancer Inst. 47,1705-1712.
RIZZINO, A. & CROWLEY, C. (1980). Growth and differentiation of embryonal carcinoma cell
line Fg in defined media. Proc. natn. Acad. Sci., U.S.A. 77, 457-461.
SAX£N, L., KOSKIMIES, O., LAHTI, A., MIETTINEN, H., RAPOLA, J. & WARTIOVAARA, J. (1968).
Differentiation of kidney mesenchyme in an experimental model system. Adv. Morph. 7,
251-293.
SAX6N, L. & LEHTONEN, E. (1978). Transfilter induction of kidney tubules as a function of the
extent and duration of intercellular contacts. /. Embryol. exp. Morph. 47, 97-109.
SAX£N, L., SALONEN, J., EKBLOM, P. & NORDLING, S. (1983). DNA synthesis and cell generation cycle during determination and differentiation of the metanephric mesenchyme. Devi
Biol. 98, 130-138.
SHINDELMAN, J. E.,ORTMEYER, A. E. & SUSSMAN, H. H. (1981). Demonstration of the transferrin receptor in human breast cancer tissue. Potential marker for identifying dividing cells.
Int. J. Cancer 27, 329-334.
SUTHERLAND, R., DELIA, D., SCHNEIDER, C , NEWMAN, R., KEMSHEAD, J. & GREAVES, M.
(1981). Ubiquitous cell-surface glycoprotein on tumor cells is proliferation-associated
receptor for transferrin. Proc. natn. Acad. Sci., U.S.A. 78, 4515-4519.
TAUB, M., CHUMAN, L., SAIER JR., M. H. & SATO, G. (1979). Growth of Madin-Darby canine
kidney epithelial cell (MDCK) line in hormone-supplemented, serum-free medium. Proc.
natn. Acad. Sci., U.S.A. 76, 3338-3342.
THESLEFF, I., EKBLOM, P., KUUSELA, P., LEHTONEN, E. & RUOSLAHTI, E. (1983). Exogenous
fibronectin is not required for organogenesis in vitro. In Vitro 19, 903-910.
TROWBRIDGE, I. S. & OMARY, M. B. (1981). Human cell surface glycoprotein related to cell
proliferation is the receptor for transferrin. Proc. natn. Acad. Sci., U.S.A. 78, 3040-3043.
TROWBRIDGE, I. S. & LOPEZ, F. (1982). Monoclonal antibody to transferrin receptor blocks
transferrin binding and inhibits human tumor cell growth in vitro. Proc. natn. Acad. Sci.,
U.S.A. 79, 1175-1179.
VOGT, A., MISHELL, R. I. & DUTTON, R. W. (1969). Stimulation of DNA synthesis in cultures
of mouse spleen suspensions by bovine transferrin. Expl Cell Res. 54, 195-200.
(Accepted 16 February 1984)