A proline analogue, azetidine-2-carboxylic acid, which has been reported to selectively depress collagen biosynthesis, affects lung morphogenesis in vitro by determining differential modifications of the rate of epithelial growth and ramification activity.

Low concentrations of the analogue induce a sharp decline of the number of newly formed terminal buds, leaving the rate of global growth of the epithelial tree almost unaffected. This effect is reversed by increasing doses which appear more effective on global epithelial growth than on budding. DNA and protein content at the end of the treatment are unequally reduced. Protein concentration is more depressed with consequent rise of the DNA/protein ratio.

These results show that epithelial budding is very sensitive to the treatment with the analogue, and draw attention to collagen biosynthesis as one of the possible mechanisms by which the mesenchyme controls the spatial organization of the epithelial tree. This type of control appears relatively independent on the actual rate of epithelial cell replication, since budding is much more sensitive to the action of the drug.

A number of reports in the last few years (Bernfield & Wessells, 1970) have emphasized the importance of collagenase-sensitive materials in the morphogenesis of several embryonic organs whose developmental mechanics are known to be based on the so called ‘epithelio-mesenchymal interaction’ (Fleischmajer & Billingham, 1968). Most of that work is based on the effect of a collagenase treatment of the epithelial component of the rudiments previously isolated by trypsinization and cultured in a transfilter system in the presence of mesenchyme (Grobstein, 1967; Wessells & Cohen, 1968). Since this procedure involves drastic modifications on the normal developmental conditions, the observed morphogenesis has been said to be ‘reminiscent’ (Wessells & Cohen, 1968) of the normal developmental pattern. Consequently, when a molecular marker of morphogenesis and/or differentiation is not available, morphological evaluation of the observed effect is often questionable. In particular, when the involved developmental event is epithelial budding it is difficult to assess what is morpho-genetically ‘normal’ budding and what represents mere distortion of the epithelial profile due to the removal of its investing mesenchyme.

Moreover, the presence of enzymic activities other than collagenolitic that are detectable in several collagenase preparations, raises the possibility that ‘the dependence of morphogenesis on collagenase-susceptible materials does not necessarily implicate collagen’ (Bernfield & Wessells, 1970).

It is therefore of obvious interest to use an entirely different approach to the problem of the morphogenetic significance of collagen biosynthesis, such as the one recently indicated by the work of Lane, Dehm & Prockop (1971) and Lane, Parkes & Prockop (1971).

They have shown that the proline analogue azetidine-2-carboxylic acid inhibits collagen accumulation in growing chick embryos down to a level of approximately 20% of the untreated controls. Therefore a morphogenetic effect of the proline analogue on the developmental pattern would directly draw attention to collagen, and other proline- and hydroxyproline-rich protein species.

The experiments reported in the current paper deal with the effect of azetidine-2-carboxylic acid on the morphogenesis in vitro of embryonic mouse lungs. This effect seems not to be entirely attributable to the depressed collagen biosynthesis, since other accompanying modifications, probably not dependent on collagen depletion, are present in the epithelial compartment of the rudiments when treated with the analogue.

The rationale of the experiments has been as follows: we assume after Wessells (1970) that oriented collagen bundles act as a physical barrier to the budding movement, thus stabilizing the epithelial structure; the sequence of morphogenetic events should therefore be as depicted in Fig. 1 A, control proper: the oriented collagen deposition selects along the epithelial surface a number of special sites which are left open for budding movement, and the final result is the production of a given number of terminal buds, accompanied by and strictly correlated to an increase of the epithelial surface area; this represents what may be defined as morphogenetic growth.

Fig. 1

Plan of the experiments and expected results.

Fig. 1

Plan of the experiments and expected results.

Fig. 2

Lung rudiments explanted at the 11th day of gestation and cultured for 45 h. (a) controls; (b) – (d) azetidine-2-carboxylic acid (μg/ml) 40, 60, 80 respectively. Photographs taken in the living state, × 40.

Fig. 2

Lung rudiments explanted at the 11th day of gestation and cultured for 45 h. (a) controls; (b) – (d) azetidine-2-carboxylic acid (μg/ml) 40, 60, 80 respectively. Photographs taken in the living state, × 40.

When the proline analogue is added to the culture medium, a remarkable decline of collagen synthesis should ensue, as expected from the results by Lane et al. (1971).

Then the expected morphogenetic result is represented in Fig. 1B, experiment proper. The scarcity of oriented collagen bundles will prevent the formation of definite budding sites, hence generalized enlargement of the epithelial surface area and absence of new buds (i.e. non-morphogenetic growth) will be observed. This result would ultimately demonstrate dissociation and relative independence of the two developmental components, budding activity and global epithelial growth.

This can be quantified by comparing the number of newly formed terminal buds to the size of the epithelial surface area, and by estimating the degree of correlation between the two, which is expected to decline, if the above assumption would prove correct, in the presence of the proline analogue.

Lung rudiments from 11-day first-generation mouse hybrids (♀ C57BL × ♂ BALB/c) were cultured in hanging drop at the surface of a plasma clot composed by equal volumes of chicken plasma and 9-day chick embryo extract.

Azetidine-2-carboxylic acid (Calbiochem, Los Angeles) was introduced in the culture medium (40, 60, and 80 μg/ml). For measurements of morphogenetic and non-morphogenetic growth all lung rudiments were drawn by camera lucida immediately and after 21 and 45 h in culture. The number of terminal buds and the measure of the epithelial surface area have been determined on the drawings after the procedures previously described (Alescio & Colombo Piperno, 1967; Alescio & di Michele, 1968). The data are reported in the form of total number of buds produced in culture (i.e. number of buds at 21 and 45 h minus number of buds at zero time), and of net increase of the epithelial area in culture (i.e. size of the epithelial area, in arbitrary planimetric units, at 21 and 45 h minus size of the epithelial area at zero time) in order to use the same arithmetical treatment for both variables.

Histological examinations were carried out on Carnoy-fixed rudiments, serially sectioned at 5 μm and stained by the periodic acid-Schiff procedure after McManus (Pearse, 1968).

Other rudiments were used for DNA and protein measurements respectively after the method of Ceriotti (1952) as modified by Gwatkin & Biggers (1963), and after the method of Lowry, Rosebrough, Farr & Randall, (1951).

The developmental pattern in vitro of the 11-day lung rudiments and the effect of azetidine-2-carboxylic acid is exemplified in Fig. 2a –d. After a 45-h treatment with 40 μg/ml of the analogue, the rudiments (Fig. 2b) tend to produce less but larger terminal buds as compared with the untreated controls. The branching activity is even more depressed with increasing doses of the analogue (Fig. 2c, d) as the terminal portions of the epithelial tree are somewhat enlarged with very little further branching. Simultaneously, however, the overall growth of the epithelial tree is also clearly reduced with increasing doses as the rudiments are much smaller in epithelial size as compared to the controls (Fig. 2 a).

This is more clearly shown in the diagrams reproduced in Fig. 3, where the epithelial profile of the distal part of the main right bronchus (caudally to the origin of the infra-cardiac branch) is represented.

Fig. 3

Camera lucida drawings representing part of the main right bronchus of the same rudiments as in Fig. 2.

Fig. 3

Camera lucida drawings representing part of the main right bronchus of the same rudiments as in Fig. 2.

We have therefore some evidence that the developmental pattern of the epithelial tree tends to shift, under the effect of the analogue, from morphogenetic growth (prominent budding activity) to non-morphogenetic growth (global enlargement of the epithelial structures with fewer new branches). This effect appears most perceivable when 40 μ g/ml of the analogue are used, while with increasing doses non-morphogenetic growth also appears increasingly affected (Fig. 3).

The above result is quantified by the data reported in Figs. 4 and 5, where the average number of newly formed terminal buds and the average increase of the epithelial surface area in the presence of different doses of azetidine-2-carboxylic acid are plotted as a function of time in culture. They show that the general trend of the response of both budding activity and global growth to the analogue is similar, in the sense that both these developmental parameters undergo an inhibitory effect directly dependent on time in the presence of the analogue and its concentration. However the magnitude and the temporal pattern of the observed inhibitory effect show that budding activity is more sensitive to the analogue, being more depressed than global growth at 40μ g/ml; the last in turn, when the analogue concentration becomes adequate, is affected more precociously than budding activity. The above information is summarized by the data in Table 1, where budding activity and global growth at 21 and 45 h under differing analogue concentrations are reported as percentage of controls; they show that budding activity possesses a lower threshold of sensitivity to the analogue, but the events required for the expression of this effect demand a longer time interval than those necessary to establish a reduced rate of global growth at higher concentrations.

Table 1

Effect of azetidine-2-carboxylic acid on budding activity and global growth

Effect of azetidine-2-carboxylic acid on budding activity and global growth
Effect of azetidine-2-carboxylic acid on budding activity and global growth
Fig. 4

Average number of terminal buds produced in culture as a function of time. 0: controls; A –C: azetidine-2-carboxylic acid (μg/ml) 40, 60, 80 respectively. Uncertainties are standard errors.

Fig. 4

Average number of terminal buds produced in culture as a function of time. 0: controls; A –C: azetidine-2-carboxylic acid (μg/ml) 40, 60, 80 respectively. Uncertainties are standard errors.

Fig. 5

Average net increase of the epithelial surface area in culture as a function of time, in arbitrary planimetric units. 0: controls; A –C: azetidine-2-carboxylic acid (μ g/ml) 40, 60, 80 respectively. Uncertainties are standard errors.

Fig. 5

Average net increase of the epithelial surface area in culture as a function of time, in arbitrary planimetric units. 0: controls; A –C: azetidine-2-carboxylic acid (μ g/ml) 40, 60, 80 respectively. Uncertainties are standard errors.

The interdependence of budding activity and global growth may be considered next, by plotting the average number of newly formed terminal buds after 45 h in culture as a function of the corresponding measures of surface area (Fig. 6). The resulting curve shows, as already implicated by the data in Fig. 4 and 5, a clearly diphasic trend with a definite change of slope dependent on analogue concentration. This provides further evidence that a dose of 40 μg/ml of the analogue determines a marked inhibition of budding accompanied by relatively little depression of global growth; the effect is reversed at higher analogue concentrations, which appear to have maximal inhibitory effect on global growth. Independent fluctuations of budding and growth are therefore definitely substantiated.

Fig. 6

Average number of newly made terminal buds after 45 h in culture as a function of the corresponding average increase of the epithelial surface area, in controls and in the presence of increasing analogue concentrations.

Fig. 6

Average number of newly made terminal buds after 45 h in culture as a function of the corresponding average increase of the epithelial surface area, in controls and in the presence of increasing analogue concentrations.

This relative independence of the two variables brought about by the experimental treatment may then be quantitatively assessed by estimating the degree of statistical correlation between the two. This has been done using the pooled data at both time points, 21 and 45 h, to obtain a more complete evaluation of the dependence on time of the involved fluctuations. The estimated values of the correlation coefficient (r) are plotted as a function of dose in Fig. 7. The highly significant (P < 0·01) statistical correlation existing in normal development (controls) is strongly depressed under the effect of the analogue, and falls below the level of statistical significance when its concentration is of 60 μg/ml or more. Complete statistical dissociation of budding activity and global epithelial growth is therefore gradually achieved with increasing doses.

Fig. 7

Estimated values of r, correlation coefficient, of number of newly made terminal buds to corresponding net increase of epithelial surface area, as a function of the analogue concentration. Data at 21 and 45 h have been pooled. Transverse dashed lines indicate the levels of significance of r. N.S. = not significant.

Fig. 7

Estimated values of r, correlation coefficient, of number of newly made terminal buds to corresponding net increase of epithelial surface area, as a function of the analogue concentration. Data at 21 and 45 h have been pooled. Transverse dashed lines indicate the levels of significance of r. N.S. = not significant.

The histological examination has failed to show noticeable structural modifications related to the presence of the analogue. The histological structure, including the peculiar glycogen distribution (Alescio & Dani, 1971), is apparently unchanged (Fig. 8,a –d), and no appreciable indications of noteworthy cell degeneration are present over the entire dose range used. Finally, some hints on the molecular mechanisms of the described results may be derived from DNA and protein measurements, performed on samples of randomly pooled lung rudiments from several litters, and cultured for 45 h in the absence and in the presence of 60 μg/ml of azetidine-2-carboxylic acid. Both DNA and protein content per lung rudiment is reduced in the presence of the analogue (Table 2). However, protein concentration is more affected than DNA, with consequent rise of the ratio DNA/protein per lung rudiment. The interpretation of the above result must, however, consider that it is based on the assumption that the average content of DNA and protein per lung rudiment is the same in the control and experimental groups at explantation, so that the observed final modifications are representative of modified accumulation rates due to the experimental treatment.

Table 2

Effect of azetidine-2-carboxylic acid on DNA and protein content

Effect of azetidine-2-carboxylic acid on DNA and protein content
Effect of azetidine-2-carboxylic acid on DNA and protein content
Fig. 8

Histological sections of a control culture (a), and of a rudiment treated with 40 μg/ml of the analogue, after 45 h in culture (b). PAS-haematoxylin. × 190. (c, d) Higher magnifications of the same rudiments as in (a) and (b) respectively. PAS-haematoxylin. × 390.

Fig. 8

Histological sections of a control culture (a), and of a rudiment treated with 40 μg/ml of the analogue, after 45 h in culture (b). PAS-haematoxylin. × 190. (c, d) Higher magnifications of the same rudiments as in (a) and (b) respectively. PAS-haematoxylin. × 390.

This work has been based on the assumption that two distinct developmental components are operating on the epithelial compartment in lung morphogenesis. The first one is increase in size of the epithelial tree, mostly but probably not entirely due (Alescio & di Michele, 1968) to cell replication, particularly intense in lung morphogenesis (Wessells, 1970). Its function is to provide materials for growth, which in turn are rendered morphogenetically significant by the second developmental component, that is budding movement. The two processes must be closely integrated for the normal course of development.

We have asked whether it is possible to experimentally dissociate the two developmental components, and the answer is positive but only to a limited extent.

The first morphogenetically noticeable effect of low doses of the proline analogue azetidine-2-carboxylic acid, which is known to inhibit collagen accumulation (Lane, Dehm & Prockop, 1971), is in fact a reduced branching activity, leading to partial uncoupling of growth and ramification, with prevailing non-morphogenetic growth. The rudiment continues to grow as the epithelial structures are enlarged, but less new branches are produced because of the reduced number of bifurcation points.

The mechanism of this effect may well be depressed collagen biosynthesis, and this would confirm the morphogenetic importance attributed to oriented collagen deposition (Wessells, 1970). Since collagen is mostly of mesenchymal origin (Bernfield, 1970), the mesenchyme must be regarded as the site of this effect.

However, when the analogue concentration increases the epithelial growth undergoes more powerful and more precocious depression than budding activity. The difference between relative inhibition of growth and relative inhibition of budding is maximal after 21 h, and then decreases after 45 h of treatment (Table 1). The anticipated involvement of growth is a particularly relevant point, showing that the effect on global growth is probably not a consequence of the prevented budding.

We may therefore assume two rather independent mechanisms of action for the proline analogue. Part of the available analogue may be immediately and preferentially incorporated in precollagen polypeptides, thereby inhibiting collagen accumulation and finally preventing further budding activity; this effect may demand relatively longer time for perceivable final expression. When the concentration of the analogue is sufficiently high, a residual fraction, not involved in collagen metabolism, may become available for interfering in other, possibly also protein biosynthetic pathways, which found a more precocious final expression in a reduced rate of cell replication; hence the noticed reduction of the DNA content per rudiment. Lane, Dehm & Prockop (1971) have shown, in fact, that the inhibitory effect on collagen biosynthesis is not entirely specific, since the net accumulation of non-collagen proteins is also somewhat reduced. A final conclusion on this particular point awaits, however, measurements of extent of the inhibition of collagen synthesis in this particular system, and a more detailed study not only of the modifications of protein and DNA content under the effect of the drug, but also of their rates of synthesis in the presence of the drug. Anyhow whatever the mechanism, it clearly entails loss of correlation of global growth to budding activity, directly dependent on analogue concentration, and opens a new possibility for further study of the reciprocal role of cell replication and budding movement in lung morphogenesis.

This work has been limited, at the moment, to whole lung rudiments, so that the observed results may well be a composite of mesenchymal and epithelial effects variously combined to give the described final pattern. Nothing therefore can be said up to now about the relevance of this experimental system for further understanding of the nature of the epithelio-mesenchymal interaction, besides that partial inhibition of one specific mesenchymal function such as collagen biosynthesis has morphogenetic consequences on the spatial organization of the epithelial tree. Further helpful information may come from knowledge of individual responses of epithelium and mesenchyme to the analogue, and of possible modifications of patterns of macromolecular synthesis in both tissues.

Effetto di un análogo della prolina sulla morfogenesi in vitro del polmone embrionale di topo

Ln analogo della prolina, azetidine-2-carboxylic acid, che è stato dimostrato capace di ridurre in maniera specifica la biosintesi del collagene, influenza la morfogenesi del polmone in vitro determinando modificazioni differenziali del tasso di accrescimento epiteliale e della attività di ramificazione.

Basse concentrazioni di analogo determinano una netta riduzione del numero di gemme terminali prodotte, lasciando relativamente indenne il tasso di crescita complessiva dell’albero epiteliale. Tale effetto è capovolto sotto 1’influsso di dosi più elevate, che risultano più attive sulla crescita che sulla ramificazione. 11 contenuto di DNA e proteine al termine del trattamento è ridotto in maniera diversa, con maggiore riduzione del contenuto proteico e conseguente aumento del valore del rapporto DNA/proteine.

Questi risultati mostrano che la gemmazione epiteliale è molto sensibile al trattamento con l’analogo, e richiamano 1’attenzione sulla biosintesi del collagene come uno dei possibili meccanismi con cui il mesenchima controlla 1’organizzazione spaziale dell’albero epiteliale. Questo tipo di controllo risulta relativamente indipendente dal tasso effettivo di replicazione delle cellule epiteliali, perché la gemmazione è molto più sensibile all’azione della sostanza.

This research has been performed with a contribution from the Italian C.N.R.

Alescio
,
T.
&
Colombo Piperno
,
E.
(
1967
).
A quantitative assessment of mesenchymal contribution to epithelial growth rate in mouse embryonic lung developing in vitro
.
J. Embryol. exp. Morph
.
17
,
213
227
.
Alescio
,
T.
&
Dani
,
A. M.
(
1971
).
The influence of mesenchyme on the epithelial glycogen and budding activity in mouse embryonic lung developing in vitro
.
J. Embryol. exp. Morph
.
25
,
131
140
.
Alescio
,
T.
&
Di Michele
,
M.
(
1968
).
Relationship of epithelial growth to mitotic rate in mouse embryonic lung developing in vitro
.
J. Embryol. exp. Morph
.
19
,
227
237
.
Bernfield
,
M. R.
(
1970
).
Collagen synthesis during epitheliomesenchymal interactions
.
Devi Biol
.
22
,
213
231
.
Bernfield
,
M. R.
&
Wessells
,
N. K.
(
1970
).
Intra- and extracellular control of epithelial morphogenesis
.
Devi Biol. Suppl
.
4
,
195
249
.
Ceriotti
,
G.
(
1952
).
A microchemical determination of deoxyribonucleic acid
.
J. biol. Chem
.
198
,
297
303
.
Fleischmajer
,
R.
&
Billingham
,
R. E.
(
1968
).
Epithelial-Mesenchymal Interactions
.
Baltimore
:
Williams and Wilkins Co
.
Grobstein
,
C.
(
1967
).
Mechanisms of organogenetic tissue interaction
.
Natn. Cancer Inst. Monogr
.
26
,
279
299
.
Gwatkin
,
R. B. L.
&
Biggers
,
J. D.
(
1963
).
Synthesis of deoxyribonucleic acid in embryonic chick tibiotarsi during development in vitro and in vivo
.
J. exp. Zool
.
152
.
Lane
,
J. M.
,
Dehm
,
P.
&
Prockop
,
D. J.
(
1971
).
Effect of the proline analogue azetidine-2-carboxylic acid on collagen synthesis in vivo. I. Arrest of collagen accumulation in growing chick embryos
.
Biochim. biophys. Acta
236
,
517
527
.
Lane
,
J. M.
,
Parkes
,
L. J.
&
Prockop
,
D. J.
(
1971
).
Effect of the proline analogue azetidine-2-carboxylic acid on collagen synthesis in vivo. II. Morphological and physical properties of collagen containing the analogue
.
Biochim. biophys Acta
236
,
528
541
.
Lowry
,
O. H.
,
Rosebrough
,
N. J.
,
Farr
,
A. L.
&
Randall
,
R. J.
(
1951
).
Protein measurement with the Folin phenol reagent
.
J. biol. Chem
.
193
,
265
275
.
Pearse
,
A. E. G.
(
1968
).
Histochemistry. Theoretical and Applied
.
London
:
Churchill
.
Wessells
,
N. K.
(
1970
).
Mammalian lung development: interactions in formation and morphogenesis of tracheal buds
.
J. exp. Zool
.
175
,
455
466
.
Wessells
,
N. K.
&
Cohen
,
J. H.
(
1968
).
Effects of collagenase on developing epithelia in vitro: lung, ureteric bud, and pancreas
.
Devi Biol
.
18
,
294
309
.