ABSTRACT
Dissociated cells of metanephri of 16-day chick embryos were reaggregated immediately or several days after culturing at a glass-medium interface to encourage spreading.
The reaggregates were organ-cultured at the air-medium interface on nutrient agar for up to 4 days. They were examined immunohistologically using fluorescein conjugated antiserum-globulin directed against semi-purified ‘kidney-specific’ antigens. The specific antigens (‘K-antigens’) are localized only in the secretory tubules of kidney, and the conjugated antiserum stains only the cells originating from this particular tissue.
In the reaggregates of fresh suspension, sorting out of the positive and negative cells started after 1 day in organ culture and the sorting out and selective localization were essentially completed after 4 days. Almost all of the positive cells were incorporated in the reconstituted tubular structures within the reaggregates, while the outer wall of the reaggregates consisted of fluorescence negative epithelial tissue.
In the reaggregates of 30 hr. monolayer culture localization was somewhat less selective: Epithelial cells with positive fluorescence very often formed patches of epithelial tissue in the outer wall.
Reconstitution of tubular structures hardly occurred in reaggregates after 96 hr. in monolayer. Almost all of the cells of the organ-cultured reaggregates were negative cells.
The results indicate that fluorescent antibody technique can be a potent tool for identifying cells derived from a single tissue. The gradual loss of the histoformative capacity in monolayer culture was discussed in relation to the antigenic changes of the cells occurring in such culture conditions.
INTRODUCTION
Since Moscona & Moscona (1952) and Townes & Holtfreter (1955) demonstrated the phenomena of sorting-out and selective localization of cells into a compound (heterotypic) reaggregate, several workers have taken a growing interest in the mechanism of these processes (e.g. Curtis, 1960, 1962; Steinberg, 1962a, 1962b, 1962c, 1963). In approaching this subject experimentally, a basic requirement is that the cells derived from different sources possess markers which can be followed throughout the culture period, since it is impossible to discuss the relative movement and displacement of cells which do not possess any marking. Thus xenoplastic combinations of mouse and chick cells or the use of radioisotopes have been adopted to mark the cells (Moscona, 1957; Trinkhaus & Gross, 1961; Zwilling, 1963). However, even such apparently ingenious conti ivances are not perfect solutions, since the organs used, e.g. liver, kidney, limbbud etc., are complexes of various kinds of tissue and it does not follow that these labels mark the cells derived from a single tissue type. To eliminate this defect, the potential usefulness of pigment cells in heterotypic combination has been investigated recently (Trinkhaus, 1963; Trinkhaus & Lantz, 1964), but this limits the variety of heterotypic combinations which can be examined.
The author has been interested in the possibility of using ‘tissue-specific’ antigen(s) of the cells belonging to a particular tissue type as marker(s) at the tissue level, and possibly at the organ level. Recently considerable advances have been made in the study of the antigenic constituents of tissue cells, and it has become possible to extract ‘kidney-specific’ antigens in a semi-purified form from the microsomal fraction of adult chicken kidneys (Okada & Sato, 1963a, 1963b). In fact, the antisera prepared against this extract (K45) is highly ‘kidney-specific ‘, and can be utilized in immunohistological procedures as a potent analytical reagent to localize the antigen(s) and identify the cells which carry it (Okada, 1964). The distribution of these antigen(s), which will be simply called ‘K-antigens is confined to the epithelium of the proximal secretory tubules. Therefore, the ‘K-antigens’, in theory, could be an excellent marker for cells derived from a single tissue and organ source.
In this paper the potential usefulness of fluorescent antibody staining to identify the cells with ‘K-antigens’ is described with respect to the reconstitution of nephric tubules from dissociated kidney cells. The histoformative ability of kidney cells which had been previously cultured in a monolayer on glass is also examined by this same technique.
MATERIALS AND METHODS
Materials
Metanephri from hybrid white chicken embryos at 16 days’ incubation were removed and cleaned of extraneous tissue. The experiments described in this paper are concerned only with this organ. It must be remembered that our markers, the’ K-antigens’ are localized only in the cells of a single tissue type, and at the tissue level kidney itself is already a heterotypic system. Therefore, no efforts have yet been made to mix the kidney cells with cells derived from other organs.
Dissociation
Dissected pieces of the cleaned metanephri were placed in a medium consisting of 0·5 per cent, trypsin (Difco, 1:250) and 0·03 per cent. EDTA dissolved in Rinaldini’s Ca-and Mg-free saline (Rinalidini, 1959) for about 45 min. ; they were then rinsed thoroughly with balanced saline and incubated with 0-01 per cent. DNA-ase (Steinberg, 1963). This produced soft masses which were rinsed again, carefully transferred into the standard culture medium, and then the suspension of discrete cells was prepared by repeatedly passing the tissue through a pipette of small bore.
General design of culture experiments
The cell suspension of the embryonic kidneys was divided into two portions. The first portion was immediately centrifuged for about 5 min. at ca. 1500 x g to make a packed pellet, which was cut into expiants of ca. 0·3−1·0 mm. to be organ-cultured. The second portion was directly inoculated into flat prescription bottles. After 20 hr. the cells attached to the glass and began to spread as a monolayer. At the 30th or 96th hr., the cells in the monolayer were released from the glass by trypsin treatment and centrifuged. The pellets thus obtained were organ-cultured to determine the histoformative capacities of the spreading cells. The organ-cultured expiants (reaggregates) of each series of experiments were usually prepared for immunohistology at days 0,1,2 and 4.
Organ culture (OC)
The pellets were organ-cultured on nutrient agar (Wolff & Haffen, 1952), the composition of which was 6 parts full strength medium 199, in which 1 − 5 per cent, agar was dissolved, 2 parts horse serum and 1 part 10-day chick embryo extract.
Spreading culture (MC)
Fresh cell suspensions contained 8 × 105 to 1 × 106 cells/ml. 6−7 ml. of such a suspension were inoculated into each bottle. Two kinds of media were used; (1) standard medium (SM) consisting of 8 parts Earle’s saline with 0·5 per cent, lactoalbumin hydrolysate and 0-1 per cent, yeast extract (YLH) and 2 parts horse serum, and (2) growth medium (GM), 6 parts YLH, 2 parts horse serum and 2 parts 10-day chick embryo extract. All the solutions contained a penicillinstreptomycin preparation at a concentration of 50 i.u. per ml. In SM the cell number remained about constant at least until the 8th day, while in GM it increased about 2·5−3·0 times by the 5th day. Details of the characteristics of the spreading cells will be reported subsequently.
Immunohistological procedures
The procedures involved fixation, embedding in paraffin, sectioning and rehydration in such a manner as to preserve the immunological specificity of ‘K-antigens’ (Okada, 1964, 1965). After immunohistological observations the slides were re-fixed with Bouin, washed and stained for ordinary histological observations.
The procedures for isolating the K45 sub-fraction, which consists mainly of ‘K-antigens’, as well as those for preparing highly ‘kidney-specific’ antisera (anti-K45), were given in the author’s earlier publications (Okada, 1962; Okada & Sato, 1963a, 1963b). In this particular study the antisera prepared in rabbits were used with identical results. Details of the conjugation of the antibody globulins of anti-K45 and the preparation of the paraffin sections for immunohistological studies have been described in the author’s latest paper (Okada, 1965; also Nairn, 1962).
RESULTS
Reaggregates from fresh suspensions
The immunohistological observations revealed that the packed pellet obtained immediately after dissociation (0-day reaggregates) is a random assemblage of cells which fluoresce (‘K-antigens-containing cells’ which will be abbreviated as ‘K-cells’) and of cells which do not fluoresce (‘K-antigens-free cells’ which will be designated as negative cells) (Plate, Fig. B). As reported elsewhere (Okada, 1964, 1965), the anti-K-45-conjugate stains only the apical cytoplasm of the cells in intact secretory tubules. This distribution pattern of the ‘K-antigens’ is maintained after dissociation and reaggregation, fluorescence being limited to a cap of cytoplasm (Plate, Figs. A & B). The percentage of ‘K-cells’ in the whole cell population was between 55−60 per cent, in each of several experimental runs. This number usually remained unchanged up to 4 days of OC of the reaggregates.
By 24 hr. OC, the segregation of ‘K-cells’ from negative cells has already started (Plate, Fig. C). A number of small homonomic epithelial vesicles or rosettes composed solely of ‘K-cells’ were found inside the reaggregate (Plate, Fig. C, KV). These vesicles were small, usually consisting of only five to ten cells, but often had a lumen. The fluorescent cap of these ‘K-cells’ was always directed toward the lumen, thus the polarity of the constituent cells composing each vesicle was already established. At this stage of OC, however, fairly large numbers of individual ‘K-cells’, roughly 10-20 per cent, of the total ‘K-cells’ population in a reaggregate, were still dispersed singly among the negative cells. By the 4th day the processes of sorting-out of the cells and selective positioning of the reconstituted tissue were nearly completed (Plate, Fig. D). Almost all of the ‘K-cells’ were incorporated into the formation of nephric tubules, which were distributed inside the reaggregates, but positive fluorescence was situated only along the inner border of the reconstituted tubules, a similar distribution to that of original tissue (Plate, Fig. E). The outer walls of the reaggregates consisted of epithelial tissue, which was composed wholly of negative cells (Plate, Fig. D). In addition the reaggregates included masses of negative cells. Therefore, these observations indicate not only the segregation of the epithelial cells from the cells belonging to such tissues as glomeruli, endothelium and mesenchyme, but also the sorting-out of the two types of epithelial cells, i.e., the fluorescent positive and negative cells.
Reaggregates after 30 hr. in MC
At 30 hr. the SM and GM monolayer cultures contained about 30 and 50 per cent. ‘K-cells’ respectively. The immunohistological picture of the fresh pellets obtained by centrifugation of cells which had been cultured in monolayer for 30 hr. was the same as described for the reaggregates from the fresh suspensions. After 4 days’ OC, most ‘K-cells’ participated in the reconstitution of the nephric tubules (Plate, Fig. G). At this stage, however, there was an important difference from the situation which was obtained in the reaggregates of the fresh suspensions, in that the outer walls of the present reaggregates did not consist only of negative epithelial cells as seen in the previous cases, but the peripheral epithelium contained patches of positive cells with their fluorescent cytoplasmic caps directed externally (Plate, Fig. F). This peripheral positive tissue was often contiguous with the positive internal tubular structure.
Reaggregates after 96 hr. in MC
At 96 hr. the SM and GM monolayer cultures contained about 27 and 49 per cent. ‘K-cells’, respectively. The fluorescence of these ‘K-cells’ was quite feeble and diffusely distributed throughout the cytoplasm. The histological features of the reaggregates at the 4th day of OC were quite different from those found in the previous two series in that they were simple vesicles containing only free mesenchymal cells (Plate, Fig. H). Immunohistological observations revealed few positive cells either in the epithelium constituting the outer wall or the mesenchyme sparsely dispersed inside the vesicles. In rare cases, one or two tiny tubular structures, consisting of ten to twenty positive cells, were detectable.
DISCUSSION
In considering the usefulness of immunohistology for identifying cells derived from a single tissue source in a heterotypic reaggregate, it is essential to establish the following two points: First, that the fluorescent antibody actually stained the desired ‘tissue-specific’ antigens, and secondly, that during the period of reaggregation the ‘tissue-specific’ antigens were neither lost from the originally positive cells nor gained by the originally negative cells. The first point, has been satisfactorily answered (see Okada, 1964, 1965). The problem raised by the second point, namely, that the ‘tissue-specific’ antigens are a valid marker for the cells, was favourably answered by the following facts : (1) Even in MC, the proportion of positive to negative cells indicated that almost all of the originally positive cells maintain a sufficient quantity of the ‘K-antigens’ to be detected by antibody staining at least until the 8th day (for a fuller discussion, see the author’s next paper) ; and (2) during OC, the percentage of the positive cells in the total population of the reaggregate remained nearly constant, except in reaggregates formed after 96 hr. in MC. Therefore, it may be concluded for the present experiments that behaviour of the cells originating from a particular kind of tissue was successfully traced by the immuno-histological method. The serious exception which occurred in the reaggregates after 96 hr. in MC will be discussed later.
The sorting-out of the cells derived from different tissues of the kidney and the selective positioning of the reconstituted tissues within reaggregates occurred completely in the reaggregates formed by fresh cell suspension from 16-day-old chicken embryos (e.g. Plate, Figs. D & E). The results indicate that, of the various tissues of the kidney at this age, fluorescent positive and negative cells can be characterized as of the ‘internally segregating type’ and the ‘externally segregating type’, respectively (Steinberg, 1962a).
The process of selective positioning was incomplete in the reaggregates after 30 hr. in MC. Epithelium containing ‘K-cell’ patches was in direct continuity with patches of the negative cells. It has been shown that for the maintenance of epithelial tubular structure appropriate mesenchyme is important as an epithelial maintenance factor (Drew, 1923; Grobstein, 1953a, 19536; Okada, 1960; Grover, 1962; Hilfer, 1962). In the present cases the lack in such a factor cannot be the cause for involvement of the patches of ‘K-cells’ in the peripheral wall, because mesenchymal cells are contained within the reaggregates (e.g. Plate, Fig. G). If an assumption of Steinberg (1962a) that selective positioning of reconstituted tissues is due to difference in the adhesiveness of the different cell types were applicable to the reaggregation of kidney cells, tubular formation would represent a stronger adhesiveness of the ‘K-cells’ than does peripheral epithelium formation of the negative cells. In the present situation the loss of adhesiveness by some of the ‘K-cells’ would seem to indicate some type of cell differentiation during the short period in MC. Since fresh cells from metanephri of older embryos than those used here were known to display complete reconstitution (the author’s unpublished experiment), this change is a function of culture in monolayer for 30 hr. and not that of normal development in situ.
The histological picture of the reaggregates formed after 96 hr. in MC was utterly different from those of the other series. They contained few reconstituted tubular structures and fluorescent positive cells (Plate, Fig. H). Simiar phenomena, i.e. the loss of histoformative capacity as well as the loss of the ability to synthesize specific cell products in reaggregates of old MC, were previously reported for several tissues such as cartilage (Holtzer et al., 1960) and thyroid (Hilfer, 1962). Could the changes be attributable to differential replacement of cells equivalent to ‘K-cells’ by the equivalent negative cells during the period in MC? Fluorescent antibody observations demonstrated that after 96 hr. in the MC the cell suspension contained on the average 27 and 49 per cent.
‘K-cells’ in the SM and GM respectively. These figures are scarcely different from the percentages observed at the 30th hr. of MC. The fact shows that a drastic replacement of the dominant cell types had not occurred during 96 hr. MC (for full discussion, see the author’s next paper). However, during organ-culturing of reaggregates formed after more than 96 hr. MC, the loss of considerable numbers of the cells was observed. Thus, for these later cultures, the possibility does exist that the failure of normal histoformation can be attributed to cell replacement phenomena in the course of OC. However, a possibility still remains that cell changes reflected by the loss of ‘K-antigens ‘may be more significant than simple replacement.
RÉSUMÉ
Recherches immunohistologiques sur la reconstitution de tubules néphrétiques à partir de cellules dissociées
Des cellules dissociées de métanéphros d’embryons de poulet de 16 jours ont été régrégées immédiatement ou plusieurs jours après culture sur une interface verre-milieu pour faciliter leur étalement.
Les réagrégats ont été mis en culture d’organes à l’interface air-milieu, sur de l’agar nutritif, pendant une durée pouvant atteindre 4 jours. Leur examen immunohistologique a été pratiqué à l’aide d’une globuline d’antisérum conjugué à la fluorescéine, se rapportant à des antigènes semi-purifiés spécifiques du rein. Les antigènes spécifiques (‘Antigènes K’) sont localisés seulement dans les tubules sécréteurs du rein, et l’antisérum conjugué colore seulement les cellules provenant de ce tissu particulier.
Dans les réagrégats de suspension fraîche, la ségrégation des cellules positives et négatives (c’est-à-dire fluorescentes et non fluorescentes) a commencé après une journée en culture d’organes ; la ségrégation et la localisation sélective étaient achevées pour l’essentiel au bout de 4 jours. Presque toutes les cellules positives étaient incorporées dans la structure tubulaire reconstituée à l’intérieur des réagrégats, tandis que la paroi externe de deux-ci consistait en tissu épithélial non fluorescent.
Dans les réagrégats d’une culture en couche monocellulaire de 30 heures, la localisation était un peu moins sélective : les cellules épithéliales avec fluorescence positive ont très souvent formé des fragments de tissu épithélial dans la paroi externe.
La reconstitution de structures tubulaires n’a presque pas eu lieu dans les réagrégats obtenus après culture de 96 heures en couche monocellulaire. Presque toutes les cellules des réagrégats en culture d’organes étaient des cellules non fluorescentes.
Les résultats indiquent que la technique des anticorps fluorescents peut constituer un outil de valeur pour identifier les cellules provenant d’un tissu unique. On discute la perte graduelle des capacités histoformatrices pendant la culture en couche monocellulaire, en rapport avec les modifications antigéniques survenant dans les cellules dans de telles conditions de culture.
ACKNOWLEDGEMENTS
I am grateful to Dr G. W. Nace for his criticism of the manuscript and to Prof. C. H. Waddington for reading this paper before publication. I thank Miss Hiroko Seki for her cheerful assistance throughout all the phases of the present work. This work was in part supported by the Scientific Research Fund from the Japan Ministry of Education to Prof. T. Fujii.
REFERENCES
EXPLANATION OF PLATE
FIG. A. Fluorescence photomicrograph of fresh cell suspension of metanephri of 16-day chicken embryos; ‘K-cell’ with fluorescence in the apical cytoplasm (K) and negative cells (N) being shown, × 1200.
FIG. B. Fluorescence photomicrograph of the paraffin section of a reaggregate after 0 day in OC of the fresh cell suspensions; ‘K-’ (K) and negative cells (N) being randomly assembled, × 800.
FIG. C. Fluorescence photomicrograph of the paraffin section of a reaggregate after 1 day in OC of the fresh cell suspensions; some homonomic vesicles composed of ‘K-cells’ (KV) being seen; some ‘K-cells’ (K) being dispersed singly amidst the negative cells. Strong fluorescent portion indicated by arrow is due to artefact, × 600.
FIG. D. Fluorescence photomicrograph of the paraffin section of a reaggregate after 4 days in OC of the fresh cell suspensions; a number of the tubules consisting of ‘K-cells’ being reconstituted inside the explant. Note the peripheral wall consists of only negative cells. Weak fluorescent portions in the peripheral wall (e.g. a portion indicated by arrow) are not due to specific staining but with non-specific blue colour, × 300.
FIG. E. Fluorescence photomicrograph of reconstituted tubules contained in a reaggregate of the same experimental series as shown in Fig. D. × 600.
FIG. F. Fluorescence photomicrograph of a reaggregate after 30 hr. in MC and 4 days in OC ; ‘K-cell’ patch constituting the peripheral wall being shown. Note fluorescent caps of these cells are directed externally, × 800.
FIG. G. Section of a reaggregate after 30 hr. in MC and 4 days in OC with ordinary histological staining; a number of tubules being reconstituted amongst mesenchymal cells, × 450.
FIG. H. Section of a reaggregate after 96 hr. in MC and 4 days in OC with ordinary histological staining. ×450.
FIG. A. Fluorescence photomicrograph of fresh cell suspension of metanephri of 16-day chicken embryos; ‘K-cell’ with fluorescence in the apical cytoplasm (K) and negative cells (N) being shown, × 1200.
FIG. B. Fluorescence photomicrograph of the paraffin section of a reaggregate after 0 day in OC of the fresh cell suspensions; ‘K-’ (K) and negative cells (N) being randomly assembled, × 800.
FIG. C. Fluorescence photomicrograph of the paraffin section of a reaggregate after 1 day in OC of the fresh cell suspensions; some homonomic vesicles composed of ‘K-cells’ (KV) being seen; some ‘K-cells’ (K) being dispersed singly amidst the negative cells. Strong fluorescent portion indicated by arrow is due to artefact, × 600.
FIG. D. Fluorescence photomicrograph of the paraffin section of a reaggregate after 4 days in OC of the fresh cell suspensions; a number of the tubules consisting of ‘K-cells’ being reconstituted inside the explant. Note the peripheral wall consists of only negative cells. Weak fluorescent portions in the peripheral wall (e.g. a portion indicated by arrow) are not due to specific staining but with non-specific blue colour, × 300.
FIG. E. Fluorescence photomicrograph of reconstituted tubules contained in a reaggregate of the same experimental series as shown in Fig. D. × 600.
FIG. F. Fluorescence photomicrograph of a reaggregate after 30 hr. in MC and 4 days in OC ; ‘K-cell’ patch constituting the peripheral wall being shown. Note fluorescent caps of these cells are directed externally, × 800.
FIG. G. Section of a reaggregate after 30 hr. in MC and 4 days in OC with ordinary histological staining; a number of tubules being reconstituted amongst mesenchymal cells, × 450.
FIG. H. Section of a reaggregate after 96 hr. in MC and 4 days in OC with ordinary histological staining. ×450.