Dissociated cells from immature (15-day) rat testes maintained in primary culture were shown to undergo morphogenesis. Dissociated cells (1–4 ×106) were cultured in 2 ml medium NCTC-135 containing 10% rat serum at 32·5 °C under an air: CO2 atmosphere (95%:5%). During the initial culture period (1–3 days) the dissociated cells attached to the culture dish and formed a monolayer. After 4–5 days of culture, aggregates of cells formed at discrete foci within the monolayer; these structures averaged 50 μ m in diameter and numbered 3–4 × 103/dish. Within 5–6 days of culture some of these aggregates (2–3 × 102) released their contact with the substratum. Histological examination of released aggregates indicated a more regular, tissue-like organization with increasing time in culture. Aggregates at earlier time periods (4–5 days) exhibited concentric rings of cell while by day 6 some aggregates exhibited a vesicular appearance with a single layer of columnar epithelium surrounding a fluid-filled lumen. The type of structure formed varied with plating density; at densities less than 2 × 10G/dish the structures were spheroid while at greater densities tubes were formed. Time lapse observations indicated that spheres formed by movements of cells into an aggregating center while tubes formed by a rolling from the edge of the monolayer. The cells comprising the vesicle epithelium appeared to be predominantly one cell type as determined by light microscopy. These results therefore indicate that dissociated testicular cells have the ability to reform tissue-like associations in culture.

Dissociated cells from the immature rat testis form colonies in response to follicle-stimulating hormone in primary culture (Davis & Schuetz, 1975a). Although the testis contains a number of germinal and somatic cell types, these colonies contained primarily one cell type. Recent evidence has suggested that these colonies are formed by cell-specific aggregation and not cell division and are Sertoli cells (Davis & Schuetz, 1977). In the present study, further differentiation of dissociated rat testicular cells in primary culture was observed, namely the reorganization of monolayers of rat testicular cells to form tissue-like structural units resembling those seen in situ.

In vitro cell-specific aggregations have been noted with cells from a number of non-testicular tissues of several species (Galtsoff, 1925,Moscona, 1957,Steinberg, 1963). Such aggregations have, in some cases, differentiated into tissue-like structures in primary culture (Okada, 1965; McGrath, Nandi & Young, 1973). These structures have been typically spheroid or ‘dome-shaped’. Depending on initial cell concentrations, dissociated rat testicular cells form spheroids or tubes. These observations indicate that dissociated rat testicular cells have the capacity for morphogenesis in vitro.

I. Cell dissociation

Testes from five 15-day-old Sprague-Dawley rats (Flow Laboratories) were removed and decapsulated. The teased tubules were incubated in 15 ml of 0 · 1 % trypsin (Difco 1:250) in Ca, Mg-free phosphate-buffered saline (pH 7 · 35) in a 50 ml trypsinizing flask. The tissue-free supernatant was collected every 10 min, diluted with 8 % fetal calf serum (1:1), and filtered through 28 μ Nitex cloth (TET-Kressilk) (Davis & Schuetz, 1975b). After 50 min of dissociation, the pooled cells were collected by centrifugation at 1000 g for 10 min.

II. Culture conditions

Cells were cultured in 2 ml NCTC-135 containing 10% rat serum (Difco), at 32 · 5° under air:CO2 (95 %:5 %) with penicillin 100 i.u./ml and streptomycin 100 μ ml in 35 mm dishes. In most cases, cells were plated at 1 – 6 × 106 cells/ dish. The cell suspension contained 1·16 ± 0·05 cells/clump (n = 4, mean± S.E.M., 300 cells counted); viability (trypan blue dye exclusion) ranged from 93 to 96 %. In all cases new media was added on day 4; care was taken not to pipet aggregates during media changes. Cell number was determined on a Biophysics Cytograf; only particles with sizes greater than red blood cells were counted.

Detached aggregates were collected and fixed in 4 % formalin (12 h), embedded in 5 % agar, and the agar block embedded in paraffin and sectioned at 5 μ m. Sections were stained for 4 min in hemotoxylin. In one experiment movement of cells within the monolayer was followed by time-lapse cinephoto-microscopy.

Description of aggregation

Single-cell suspensions of rat testis were plated at 1 × 106 cells/dish in NCTC-135 medium containing 10 % rat serum. After 24 h of culture, cells were attached to the culture dish; however, no cells were seen in closely packed groups, colonies, with culture in 10 % rat serum as has been noted in media containing 10 % fetal calf serum (Davis & Schuetz, 1975 a). By day 3 of culture, aggregates of cells were seen attached to the dish substratum (Fig. 1). Unlike the ‘colonies’ noted in the previous study (Davis & Schuetz, 1975a), which contained closely-packed groups of cells attached to the monolayer, the aggregated cells formed mounds or multiple layers with some cells losing contact with the substratum. The cells were loosely arranged within the aggregate. At 4 days of culture, aggregates had enlarged and were spheroid in shape (Fig. 2). Cell orientations within all aggregates examined appeared to be random. The number of attached aggregates varied between 3 – 4 × 103/dish (n = 6).

Fig. 1.

Phase-contrast photomicrograph of cells in the monolayer of a 4-day culture. Cells were plated at 1 × 106/dish. Cells can be seen in multiple layers at the center of the figure (arrow). Bar = 10 μm.

Fig. 1.

Phase-contrast photomicrograph of cells in the monolayer of a 4-day culture. Cells were plated at 1 × 106/dish. Cells can be seen in multiple layers at the center of the figure (arrow). Bar = 10 μm.

Fig. 2.

Phase-contrast photomicrograph of cells in an aggregate in a 5-day culture. Cells were plated at 1 × 106/dish. Bar = 10 μ m.

Fig. 2.

Phase-contrast photomicrograph of cells in an aggregate in a 5-day culture. Cells were plated at 1 × 106/dish. Bar = 10 μ m.

After 4 – 5 days in culture, most aggregates detached from the dish bottom. Histological examination of the released aggregates showed some internal organization (Fig. 3) with concentric layers of cells around a central lumen. By 5 days (Fig. 4), an epithelial-like layer of cells surrounding a central lumen, similar to that noted in the germ-cell-depleted testis in situ (Means & Huckins, 1974), was seen. The cells forming the epithelium appear to be of one cell type. Other released vesicles showed an epithelial-like layer surrounded by concentric rings of cells; in others, the concentric layer was seen inside the epithelial layer. Some aggregates resembling those seen at day 4 (Fig. 3) were still present at day 5.

Fig. 3.

Section of a detached vesicle at 4 days of culture. At this point little organization is seen within the structure. Bar = 10 μm.

Fig. 3.

Section of a detached vesicle at 4 days of culture. At this point little organization is seen within the structure. Bar = 10 μm.

Fig. 4.

Section of detached vesicle at 5 days of culture. An epithelial arrangement with a clear lumen is seen at this time. Bar = 10 μ m.

Fig. 4.

Section of detached vesicle at 5 days of culture. An epithelial arrangement with a clear lumen is seen at this time. Bar = 10 μ m.

At cell concentrations below 2 × 106/dish, vesicles were formed ; at greater cell concentrations, tubules were formed (Fig. 5). The time course of tubule formation was identical to that of vesicles. At concentrations greater than 4 × 106/dish, a concentric ring of tubes was seen at the outside of the dish with vesicles in the middle (Fig. 5). Size of the tubes varied from 500 μ m to 1 · 2 cm in length. The number of tubes varied from 1 to 15 per dish, number was inversely proportional to size, and the time course of tube formation was identical to aggregate formation. Whereas the aggregates were formed by movement of cells into one center, tubes were formed by rolling of the edge of the monolayer. The rolling edge was always initiated at the point where the confluent monolayer was in contact with the dish edge. The absence of tube formation at plating densities below 2 × 106/dish is probably due to the lack of monolayer confluence.

Fig. 5.

Dish (35 mm) containing a concentric ring of tubes (arrows) at 6 days of culture.

Fig. 5.

Dish (35 mm) containing a concentric ring of tubes (arrows) at 6 days of culture.

To rule out the possibility that the detached vesicles formed from epithelial cells of the tubuli recti, the tubulus portion of testis from 15-day-old rats was removed and discarded prior to cell dissociation. Under these conditions, aggregation still occurred, indicating that the tubulus cannot be the sole source of reaggregating cells.

It is now well established that embryonic cells from several tissues can form tissue-specific associations in vitro (Moscona, 1957; Steinberg, 1963). Abraham (1960) has noted specific growth patterns in monolayers of embryonic chick testis cells. Tissue-like reassociations of adult mammary cells have also been noted (McGrath et al. 1973). The in vitro reassociations reported in the present paper exhibit several characteristics identical to testicular cells in situ, including (1) formation of a somatic (Sertoli) cell epithelium, (2) lumen formation, and (3) formation of tubular structure. The testis cells were obtained after birth, but at an age when the cells involved are still differentiating (Flickinger, 1967).

The ability of a mixture of cells to form an epithelium comprising predominately one cell type clearly implies the ability for cell recognition, either by cell-surface or extrinsic signals (Moscona, 1960). The geometry of the resulting structures may be explained by the hypothesis of Steinberg (1963), according to which cell aggregates form spheres in vitro because this maximizes the cell-cell and minimizes the cell-media interactions. Tubes, however, are initiated at the cell-dish interface, the only position at which a cell in the monolayer lacks an interaction with the surface of another cell. Since the cells in the axis parallel to that interface are already in direct contact with one another, this puts a constraint on movement in that direction, so that the resulting structure of least free energy is a cylinder or tube.

The testis of the 15-day-old rat contains germinal and non-germinal cells. However, only non-germinal (somatic) cells attach to the culture dishes (Steinberger, 1965). The cells comprising the epithelial layer of the vesicles appear to be Sertoli cells (Davis & Schuetz, 1977). Close scrutiny of the Sertoli cell in situ offers some insight into the forces involved in the movement. The Sertoli layer in the seminiferous tubules in vivo is polarized, the basal portion lying in a plasma-like environment and the apex in the unique ionic environment of the tubule (Tuck, Setchell, Waites & Young 1970); this polarization may be involved in the movements noted.

I would like to thank Drs Allen Schuetz, Zenus Sykes, and Young Kim for their helpful discussions; Dr Allen Cohen’s help with the time-lapse photomicroscopy is greatly appreciated. This research supported by a contract HD3-2794 from the National Institutes of Child Health and Human Development. Funds for some equipment were provided by a grant from the Rockefeller Foundation (grants awarded to A. W. Schuetz).

Abraham
,
M.
(
1960
).
Processus de reorganisation dans les agregets formes per les cullules des gonades dissociées d’Embryon de poulet
.
Archs Anat. microsc. Morph
.
49
,
333
344
.
Davis
,
J. C.
&
Schuetz
,
A. W.
(
1975a
).
Follicle stimulating hormone enhances attachment of rat testis cells in culture
.
Nature, Lond
.
254
,
611
612
.
Davis
,
J. C.
&
Schuetz
,
A. W.
(
1975b
.
Separation of germinal cells from immature rat testes by sedimentation at unit gravity
.
Expl Cell Res
.
91
,
79
86
.
Davis
,
J. C.
&
Schuetz
,
A. W.
(
1977
).
Purification of colony forming cells from the immature rat testis
.
Expl Cell Res
.
106
,
253
260
.
Flickinger
,
C. J.
(
1967
).
The postnatal development of the Sertoli cells of the mouse
.
Z. Zellforsch. mikrosk. Anat
.
78
,
92
113
.
Galtsoff
,
P. S.
(
1925
).
Regeneration after dissociation (an experimental study on sponges)
.
J. exp. Zool
.
42
,
183
221
.
Mcgrath
,
C. M.
,
Nandi
,
S.
&
Young
,
L.
(
1973
).
Relationship between organization of mammary tumors and the ability of tumor cells to replicate mammary tumor virus and to recognize growth-inhibitory contact signals in vitro
.
J. Virol
.
9
,
367
376
.
Means
,
A. R.
&
Huckins
,
C.
(
1974
).
In Hormone Binding and Target Cell Activation in the Testis, Current Topics in Molecular Endocrinology
1
,
145
165
.
New York
:
Plenum Press
.
Moscona
,
A.
(
1957
).
Development in vitro of climaric aggregates of dissociated embryonic chick and mouse cells
.
Proc. natn. Acad. Sei. U.S.A
.
43
,
184
194
.
Moscona
,
A.
(
1960
).
Patterns mechanisms of tissue reconstruction from dissociated cells
.
In Developing Cell Systems and their Controls
(ed.
D.
Rudnick
) pp.
45
70
.
New York
:
Ronald Press
.
Okada
,
T. S.
(
1965
).
Immunohistological studies on the reconstitution of nephric tubules from dissociated cells
.
J. Embryol. exp. Morph
.
13
,
299
307
.
Steinberg
,
M. S.
(
1963
).
Reconstruction of tissues of dissociated cells
.
Science, N.Y
.
141
,
401
408
.
Steinberger
,
A.
(
1965
).
In vitro growth of mammalian testicular cells
.
Anat. Rec
.
151
,
420
.
Tuck
,
R. R.
,
Setchell
,
B. P.
,
Waites
,
G. M. H.
&
Young
,
J. A.
(
1970
).
The composition of fluid collected by micropuncture and catheterization from the seminiferous tubules and rete testis of rats
.
Zur. J. Physiol
.
318
,
225
243
.