Intercellular transfer of tritium-labelled uridine nucleotides has been used to detect junctional communication between various cell types in primary culture. Epidermal kératinocytes, melanocytes and dermal fibroblasts from new-born-mouse skin, and epithelial cells from baby mouse kidney form communicating junctions in all possible homologous and heterologous combinations. This lack of detectable communication specificity between cells in primary culture contrasts with the specificity shown by some established cell lines.

Cells in animal tissues are commonly connected by permeable intercellular junctions, which permit the free exchange between cells of small molecules and ions but not macromolecules (Pitts & Simms, 1977; Flagg-Newton, Simpson & Loewenstein, 1979). The physical basis of this permeability is believed to be the gap junction (Gilula, Reeves & Steinbach, 1972). Permeable junctions have been found between cells in most tissues in vivo and are formed between many cells in culture.

In culture, the formation of permeable junctions shows little species specificity. Mammalian, avian, amphibian and fish cells can form junctions with one another (Pitts, 1977 and unpublished data) but not with cells from arthropods, which possess morphologically distinct gap junctions (Epstein & Gilula, 1977).

In coupled tissues, the individuality of a cell is maintained by its nucleic acid and protein content but the small molecular weight cellular components are shared between all the cells. This results in a limited syncytial rather than a strictly cellular type of tissue organization and the appreciation of this fact has important implications for understanding tissue function, which have been discussed elsewhere (Loewenstein, 1979; Finbow & Yancey, 1980; Pitts, 1980).

Of the several important questions still left to be answered about this form of intercellular communication, one concerns the extent of the metabolic continuum. Does it extend through all the junction-forming cells of an animal (cf. the symplast concept in plants, Münch, 1930) or is it restricted to organs or smaller regions, marked or not by histologically identifiable boundaries? The discovery of selective junctional communication between certain cell types in culture (Pitts & Biirk, 1976; Fentiman, Taylor-Papadimitriou & Stoker, 1976; Gaunt & Subak-Sharpe, 1979) suggested that such restriction may occur. Cells of the hamster fibroblast line BHK21/13 (Macpherson & Stoker, 1962) and of the Buffalo rat-liver line BRL (Coon, 1968) always form permeable junctions at homologous (BHK-BHK or BRL-BRL) contacts but only rarely at heterologous (BHK-BRL) contacts (Pitts & Biirk, 1976). Breast epithelial cells and breast fibroblasts show similar selectivity but there is no general rule of epithelial-fibroblastic selectivity because calf lens cells form junctions with human fibroblasts as well as they do among themselves (Fentiman et al. 1976). Conversion of normal cells to tumour cells, either in vivo (Fentiman & Taylor-Papadimitriou, 1977) or by transformation with oncogenic viruses in culture (Shaw & Pitts, unpublished data), can result in the loss of selectivity.

Communication selectivity is not thought to be due to differences in junctional components but to variations in the frequencies with which the cell membranes come close enough together to allow junction formation (Pitts & Biirk, 1976; Gaunt & Subak-Sharpe, 1979).

Little is known about the role of junctional communication in animals, particularly in the non-excitable tissues. A knowledge of the patterns of communication between and within tissues will provide a guide to the kinds of function that should be considered. This paper presents a quantitative analysis of junction formation between 2 cell types from the epidermis of baby mouse skin (keratinocytes and melanocytes), fibroblasts from the dermis of the same tissue, epithelial cells from mouse kidney and cells of several established cell lines. None of the cells in primary culture showed the selectivity characteristic of the cell lines.

Materials

Collagenase (Clostridium perfringens type IV ; EC 3.4.24.3), hyaluronidase (ovine testicular; EC 3.2.1.36), L-(β)3,4-dihydroxyphenylalanine (DOPA) and p-bis(o-methylstyryl)-benzene (bisMSB) were obtained from British Drug House Ltd, London; uridine and D-valine from Sigma Chemical Co. Ltd, Poole, England; 2,5-diphenyloxazole from Koch-Light Laboratories Ltd, Coinbrook, England; Amfix from May and Baker Ltd, Dagenham, England; DePeX mounting medium and Giemsa stain from G. D. Searle and Co. Ltd, London; [5-3H]-uridine (sp. act. 30 Ci.mmol−1) from the Radiochemical Centre, Amersham, England. Other reagents were BDH ‘Analar’ grade or obtained from Flow Laboratories Ltd.

Methods

Preparation of primary cultures from newborn mouse skin

Cultures of epidermal keratinocytes and dermal fibroblasts from newborn mouse skin were prepared by the method of Regnier, Delescluse & Prunieras (1973). Full thickness dorsal skin from 1 to 3-day-old mice was incubated in 0·2% trypsin in phosphate-buffered saline (PBS, Flow Laboratories Ltd) at 37 °C for 1 h. Dermis and epidermis were separated using forceps and basal epidermal keratinocytes were liberated by agitation of both tissues in the culture medium. The kératinocyte suspension was filtered through sterile gauze, centrifuged at 500 g for 10 min at 4 °C and the pellet resuspended in fresh culture medium.

Dermal fibroblasts were produced by digestion of the dermis in 0·1% collagenase, 0·1% hyaluronidase in GKN solution (0·1% glucose in physiological saline; Hinz & Syverton, 1959) followed by filtration, centrifugation and resuspension as above. Viable cell numbers were estimated by Trypan Blue exclusion. Four-ml aliquots of keratinocytes or fibroblasts were plated at 2·5 × 106 cells per ml in 50-mm diameter plastic Petri dishes.

Melanocytes occurred as a contaminant of the kératinocyte preparation and were prepared from this by the method of Moreno, Vinzens & Prunieras (1978). Epidermal cells (5 × 106) were sedimented by centrifugation and resuspended in 2 ml 0·8% sodium citrate in 0·9% NaCl for 2 min. The suspension was diluted in culture medium to 5 × 105 cells per ml and 4-ml aliquots plated out in 50-mm diameter dishes. Six h later the medium was replaced with serum-free medium and 18 h thereafter this was replaced with normal medium. DOPA staining of melanocyte cultures was performed by the method of Riley (1970).

All cultures were grown in Glasgow Modification of Eagle’s Medium (Flow Laboratories Ltd) containing too units.ml−1 penicillin and 100 μg.ml−1 streptomycin supplemented with 10% foetal calf serum (EFC10) under an atmosphere of 5% carbon dioxide/95% air.

Preparation of primary epithelial cell cultures from baby mouse kidney

Baby mouse kidney epithelial cells were prepared by the method of Gilbert & Migeon (1975). Twelve kidneys from 9 to 15-day-old mice were minced finely with scissors and dissociated by stirring in 10 ml 0·2% trypsin in PBS at 37 °C for 15 min, centrifuged at 500 g at 4 °C for 10 min and resuspended in culture medium. Four-ml aliquots of 2 5 × 10s viable cells per ml were plated in 50-mm diameter dishes. The cells were grown in EFC10 in which L-valine (normal concn 46 mg.ml−1) was replaced by D-valine (92 mg.ml−1). Kidney fibroblasts lack the racemase to interconvert the 2 isomers and no fibroblastic cells were observed in these cultures.

Assay of junction formation by following the transfer of uridine nucleotides between cells in contact

Junctional communication between cells in culture was assayed by the method of Pitts & Simms (1977). Donor cell cultures were labelled with [5H]uridine for 3 h, washed 3 times with unlabelled medium, washed twice with trypsin/EDTA (Flow Laboratories Ltd) and suspended in EFC10. Donor cells (4 × 104) were mixed with 1 ·6 × 106 unlabelled (recipient) cells and plated in 35-mm plastic Petri dishes containing 3 sterile 13-mm diameter glass coverslips in 2 ml EFC10 supplemented with 10−3 M non-radioactive uridine. After 3 h coculture, the cells were fixed in formal-saline for 1 h at 4 °C. Washing with cold 5% trichloroacetic acid to remove cellular nucleotides and processing for autoradiography were carried out as described by Pitts & Simms (1977). Autoradiographs were analysed and photographed using a Leitz Orthomat photomicroscope fitted with plan optics.

Nucleotide transfer was quantitated by counting autoradiographic grains over 50 recipient cells in contact with donor cells and over 50 isolated recipient cells. The 2 populations of grain counts were compared using Student’s t-test.

For control experiments to show that RNA was not transferred (confirming normal junctional permeability), donor cells were labelled as above and then incubated in non-radioactive medium for 24 h. The chased donor cells were then cocultured with unlabelled recipient cells as above.

RNA synthesis and changes in cellular uridine nucleotide pools were measured in replicate donor cell cultures immediately after labelling, and 3 h and 24 h thereafter. Nucleotides were extracted by washing the cell monolayers with 1-ml aliquots of 5% trichloroacetic acid. After washing twice with water, residual acid-insoluble material was extracted in 1 ml 0·1 M-NaOH (for 1 h at room temperature) and acidified with 0·2 ml 1 M-HCI. Extracts were added to 10 ml Triton/toluene scintillation fluid (70% (v/v) toluene, 30% Triton X100, 0 ·5% (w/v) PPO, 0·05% bisMSB) and counted for 4 min in a Nuclear Chicago Isocap 400 liquid scintillation spectrometer. Channels ratios showed that counts varied by less than 5% due to differences in efficiency.

Donor cell cultures were labelled with [3H]uridine (1 μ Ci ml−1) for 3 h, then washed 3 times with sterile BSS (British Standard Saline). The cells were suspended in EFC10 medium after washing with trypsin/EDTA, and mixed with suspensions of unlabelled recipient cells in a donor : recipient ratio of 1:4. Cell mixtures were plated out at 2 × 106 cells per dish in 35-mm plastic Petri dishes containing 3 sterile glass coverslips in 2 ml EFC 10 containing 10−3 M nonradioactive uridine. After 3 h coculture the medium was discarded and the cells were fixed with 2 ml 10% formal-saline at 4 °C for 1 h. Acid-soluble material was removed by washing the coverslips for 5 min in ice-cold 5% trichloroacetic acid twice, then for 5 min in distilled water twice, and dipped in methylated spirit. The dried coverslips were mounted on glass microscope slides using DePeX and processed for autoradiography using Ilford L4 nuclear research emulsion by the method of Pitts & Simms (1977). After 48 h, autoradiographs were developed, fixed and stained with Giemsa. Further coverslips were mounted on top with DePeX.

Autoradiographic grains were counted over 50 recipient cells in contact with donor cells and 50 recipient cells not in contact with donors. The means and standard deviations of the grain counts for the 2 populations were used to calculate t-values according to Student’s t-test, and values of P derived from statistical tables. C.R., recipient cells in contact with donor cells. N.C.R., recipient cells not in contact with donor cells, d.f., degrees of freedom.

Nucleotide transfer between epidermal keratinocytes and dermal fibroblasts

The autoradiographic grain counts over recipient cells in contact with prelabelled donor cells (contacting recipients, C.R.) and over isolated recipient cells (noncontacting recipients, N.C.R.) were analysed statistically rather than presenting them as nomographs as in previous studies. The grain counts over the non-contacting recipients represent autoradiographic background (due to the state of the emulsion and the conditions of processing) plus any incorporation of radioactivity from the medium (due to incomplete washing or release of radioactive components from the donor cells into the medium, neither of which are normally detectable). The grain counts over the contacting recipients represent the same background and incorporation from the medium plus the incorporation of 3H-labelled nucleotides transferred through junctions from the donor cells. A significant difference between the 2 populations of grain counts (C.R. and N.C.R.) indicates junction formation.

Epidermal keratinocytes and dermal fibroblasts form junctions in both homologous (EK-EK and DF-DF) and heterologous (EK-DF and DF-EK) combinations (Table 1 and Fig. 1). In each case the probability of the C.R. and N.C.R. populations being the same is less than 0·001 (using Student’s t-test). The patterns of incorporation after heterologous (DF-EK) transfer are similar to those reported previously (Pitts & Simms, 1977). The donor cells are heavily labelled, particularly over the cytoplasm due to the ‘chase’ during the coculture period of label in nuclear RNA into longer-lived cytoplasmic forms. Contacting recipients are less heavily labelled and show relatively greater nuclear (particularly nucleolar) labelling. Similar results (not shown) have also been obtained with keratinocytes and fibroblasts derived from adult guinea-pig ear skin.

Table 1.

Uridine nucleotide transfer between epidermal kératinocytes and dermal fibroblasts

Uridine nucleotide transfer between epidermal kératinocytes and dermal fibroblasts
Uridine nucleotide transfer between epidermal kératinocytes and dermal fibroblasts
Fig. 1.

Uridine nucleotide transfer between dermal fibroblasts and epidermal kératinocytes. Experimental details were as described in the legend to Table 1, except that the donor cell cultures were labelled with [3H]uridine at a concentration of 10 μ Ci.ml−1, and the autoradiograph was exposed for 3 weeks prior to developing. Donors, dermal fibroblasts; recipients, epidermal kératinocytes, × 1460.

Fig. 1.

Uridine nucleotide transfer between dermal fibroblasts and epidermal kératinocytes. Experimental details were as described in the legend to Table 1, except that the donor cell cultures were labelled with [3H]uridine at a concentration of 10 μ Ci.ml−1, and the autoradiograph was exposed for 3 weeks prior to developing. Donors, dermal fibroblasts; recipients, epidermal kératinocytes, × 1460.

When chased donor cells (see Materials and methods) are used instead of donor cells there is little transfer of activity to the contacting recipients confirming that transfer occurs via junctions permeable to nucleotides but not to RNA. The changes in radioactivity in the nucleotide pools (acid-soluble counts) and RNA (acid-insoluble counts) when labelled donor cells are cultured in unlabelled medium for 24 h are shown in Table 2 and the transfer from chased donor cells is shown in Table 3. The activity in the RNA increases during the 24-h period (Table 2), while the activity in the nucleotide pools falls by 88% (EK) or 91% (DF). The effect of the chase on the transfer is shown by the fall in the net (C.R. — N.C.R.) grain counts over the contacting recipients (Tables 1, 3). This fall is on average 94% (range 93 – 95%). This agrees very well with the expected fall (88 –91%) from the measurements of the pool activities when it is remembered that the donor cells will have divided during the 24-h chase period. The differences in nucleotide transfer in the various cell combinations (Table 1) are consistent with the differences in available nucleotide pool activities in the 2 cell types (Table 2).

Table 2.

Distribution of radioactivity between medium and cellular acid-soluble and acid-insoluble pools after labelling of newborn-mouse skin cells with [3H]uridine

Distribution of radioactivity between medium and cellular acid-soluble and acid-insoluble pools after labelling of newborn-mouse skin cells with [3H]uridine
Distribution of radioactivity between medium and cellular acid-soluble and acid-insoluble pools after labelling of newborn-mouse skin cells with [3H]uridine
Table 3.

Uridine nucleotide transfer between pulse-chased epidermal keratinocytes and dermal fibroblasts

Uridine nucleotide transfer between pulse-chased epidermal keratinocytes and dermal fibroblasts
Uridine nucleotide transfer between pulse-chased epidermal keratinocytes and dermal fibroblasts

Nucleotide transfer between primary skin cells and established cell lines

Primary cultures of epidermal keratinocytes and dermal fibroblasts from new-born mouse skin were tested for junction formation with cells of the cell lines C13, BRL and A9. BHK21/13 (C13) is a junction-forming fibroblast line (Macpherson & Stoker, 1962; Pitts, 1971) and BRL is a junction-forming liver epithelial cell line (Coon, 1968; Pitts & Bürk, 1976), but C13 and BRL cells only form junctions with low frequency in mixed cultures (Pitts & Bürk, 1976). Cells of the A9 line (derived from L929) do not form junctions (Marin & Littlefield, 1968; Pitts, 1971) and are used as a negative control.

Epidermal keratinocytes form junctions with C13 and BRL cells but not with A9 (Table 4). Counts over contacting recipients in the EK-BRL and BRL-EK cocultures are lower than those in the EK-C13 and C13-EK cocultures but are still significant at the P < 0·001 level. Nucleotide transfer between C13 and BRL cells is much lower than that between cells in either homologous coculture though it is still significant at the P < 0·01 level. However, the high standard deviations observed for the heterologous cocultures indicate the non-unimodal distribution of the counts (i.e. some contacting recipients show normal transfer while most show none).

Table 4.

Uridine nucleotide transfer between epidermal keratinocytes and cells of established cell lines

Uridine nucleotide transfer between epidermal keratinocytes and cells of established cell lines
Uridine nucleotide transfer between epidermal keratinocytes and cells of established cell lines

Dermal fibroblasts also form junctions with C13 and BRL cells but not with A9 (Table 5). Again, the extent of transfer between the DF and C13 cells appears greater than that between DF and BRL but both are highly significant (P < 0·001).

Table 5.

Uridine nucleotide transfer between dermal fibroblasts and cells of established cell lines

Uridine nucleotide transfer between dermal fibroblasts and cells of established cell lines
Uridine nucleotide transfer between dermal fibroblasts and cells of established cell lines

The selective communication observed with C13 and BRL cells is not seen in any mixtures of EK or DF cells with EK, DF, C13 or BRL cells.

Formation of junctions by epidermal melanocytes

Melanocytes were prepared as described (see Materials and methods) and identified by their dendritic appearance and by DOPA staining (Fig. 2). Melanocytes form junctions with EK, DF, C13 and BRL cells but not with A9 (Table 6). In this experiment, because melanocytes cannot be subcultured successfully, they were prelabelled and unlabelled cells (recipients) were added to them. This difference in experimental procedure may account for the higher grain counts of the contacting recipients in this experiment compared with the others.

Table 6.

Uridine nucleotide transfer between epidermal melanocytes, epidermal keratinocytes, dermal fibroblasts, and cells of established cell lines

Uridine nucleotide transfer between epidermal melanocytes, epidermal keratinocytes, dermal fibroblasts, and cells of established cell lines
Uridine nucleotide transfer between epidermal melanocytes, epidermal keratinocytes, dermal fibroblasts, and cells of established cell lines
Fig. 2.

DOPA staining of epidermal melanocytes. Cultures of epidermal melanocytes were established from newborn-mouse skin (see Materials and methods) and 2 days later were stained with dihydroxyphenylalanine by the method of Riley (1970). × 1460.

Fig. 2.

DOPA staining of epidermal melanocytes. Cultures of epidermal melanocytes were established from newborn-mouse skin (see Materials and methods) and 2 days later were stained with dihydroxyphenylalanine by the method of Riley (1970). × 1460.

Fig. 3.

Uridine nucleotide transfer between renal epithelial cells and C13 cells. For experimental details, see the legend to Fig. 1. Donors, renal epithelial cells; recipients, Ci 3 cells, × 1460.

Fig. 3.

Uridine nucleotide transfer between renal epithelial cells and C13 cells. For experimental details, see the legend to Fig. 1. Donors, renal epithelial cells; recipients, Ci 3 cells, × 1460.

Fig. 4.

Lack of uridine nucleotide transfer between MDCK cells and BRL cells. For experimental details, see the legend to Fig. 1. Donors, BRL cells; recipients, MDCK cells. × 455.

Fig. 4.

Lack of uridine nucleotide transfer between MDCK cells and BRL cells. For experimental details, see the legend to Fig. 1. Donors, BRL cells; recipients, MDCK cells. × 455.

Melanocytes, which are neurectoderm cells derived from the neural crest during embryogenesis, show no selectivity in junctional communication with the epithelial and mesodermal-derived cell types tested.

Junction formation by primary renal epithelial cells and renal epithelial cells of an established cell line

Primary epithelial cells from baby mouse kidney (RE cells) form junctions among themselves and with both C13 and BRL cells (Table 7). Selective junction formation is not apparent.

Table 7.

Uridine nucleotide transfer between renal epithelial cells and cells of established cells lines Ci 3 and BRL

Uridine nucleotide transfer between renal epithelial cells and cells of established cells lines Ci 3 and BRL
Uridine nucleotide transfer between renal epithelial cells and cells of established cells lines Ci 3 and BRL

Epithelial cells of the canine kidney cell line MDCK (Rindler, Chuman, Shaffer & Saier, 1979), on the other hand show a new level of selectivity. They form junctions among themselves but not with either BRL or C13 cells (or, as expected, with the A9 cells; Table 8).

Table 8.

Uridine nucleotide transfer between canine kidney epithelial cell line MDCK and cell lines C13, BRL and A9

Uridine nucleotide transfer between canine kidney epithelial cell line MDCK and cell lines C13, BRL and A9
Uridine nucleotide transfer between canine kidney epithelial cell line MDCK and cell lines C13, BRL and A9

The formation of permeable junctions between cells can be detected in various ways. Electrophysiological methods (Furshpan & Potter, 1959; Loewenstein & Kanno, 1964) are very sensitive but are technically difficult and the data obtained require careful interpretation (Sheridan, 1976). Injection of fluorescent probes for junctional permeability (Loewenstein, 1976) also requires sophisticated apparatus and the observations are not amenable to quantitative analysis. The method also appears to be less sensitive than electrophysiology as electrical coupling can be detected in some systems in the absence of observable transfer of fluorescence (Bennett, Spira & Pappas, 1978; Lo & Gilula, 1979). The method used in this study, the detection of 3H-labelled metabolite transfer by autoradiography is much simpler but its sensitivity has not yet been related to either of the other methods. There is good agreement between the results of all 3 methods when cells are clearly coupled or clearly do not form junctions. In situations where cells are poorly coupled (e.g. heterologous coupling between cell types showing selectivity) it is not known how the sensitivity of this method compares with the other two.

With autoradiography, variations in coupling efficiency (and variations in available metabolite pools or variations in macromolecular synthesis, if these occur) and heterogeneity of coupling are defined with increasing significance as more grain counts are made. In this study we have attempted to reach a sensible compromise between the time taken to make large numbers of grain counts (in a large number of different cell combinations) and adequate statistical significance.

With counts over only 10 contacting recipients and 10 non-contacting recipients, a reproducible, significant (P < 0·001) difference is obtained between the 2 populations (C.R. and N.C.R.). if the cells form normal levels of permeable junctions (e.g. C13C13 or BRL-BRL). However, if the cells do not form junctions (e.g. L-L), the probability that the 2 populations are the same is variable (the background grain count usually ranges between 1 and 7 grains per cell). Differences between no junction formation and infrequent junction formation (e.g. C13-BRL) are therefore uncertain.

With counts over 50 cells of each type (C.R. and N.C.R.) the 2 populations are significantly different (always P < 0·05, usually P < 0·001) even in the C13-BRL selective communication combinations. When no junctions are formed, the probability that the 2 populations are the same rises to more than 0·5. This number (50 C.R. and 50 N.C.R.) was therefore counted in the reported analyses. However, differences between no junctions and very few junctions (less than between C13 and BRL) may not be apparent using this method.

The lack of selectivity shown by the primary cells is surprising and contrasts with the selectivity seen between breast epithelial cells and breast fibroblasts in primary culture (Fentiman et al. 1976). There are other reports in the literature (based on different techniques) of heterologous junction formation. Primary cultures of mouse myocardial cells and rat-ovarian granulosa cells respond coordinately to hormonal stimuli, which, because they are coupled, are most likely due to junctional transfer of cAMP (Lawrence, Beers & Gilula, 1978). Synchrony of beat of mouse embryonic myocardial cells can be transmitted through heart fibroblasts or through cells of amniotic or kidney epithelial cell lines by junctional transfer of ions (Goshima & Tonomura, 1969).

The observations, in this study, that epidermal keratinocytes form junctions in culture with dermal fibroblasts, extend the earlier reports of junction formation between keratinocytes from human skin (demonstrated by electron microscopy and electrophysiology; Cavoto & Flaxman, 1972) and the indication of metabolic coupling of keratinocytes in vivo by the absence of mosaicism in epidermal cells of Lesch-Nyhan syndrome heterozygotes labelled with [3H]guanine (Frost, Weinstein & Nyhan, 1970). Junctional communication between fibroblasts and keratinocytes may be involved in the inductive dermal-epidermal interactions by which the differentiation of the epidermis is regulated by the underlying dermis (Billingham & Silvers, 1967; Dhouailly, Rogers & Sengel, 1978). Occasional contacts between dermal and epidermal cells occur across the basement membrane (Briggaman & Wheeler, 1975) and increase in frequency in the hyperproliferative disease, psoriasis (Cox, 1969).

The lack of communication selectivity between primary cell types in culture suggests that, if dermal and epidermal cell layers are metabolically segregated in vivo, this is more likely to be due to the intervening basement membrane than phenotypic inability to form heterologous permeable junctions.

This work was supported by a Cancer Research Campaign grant to J.D.P. and a University of Glasgow, Medical Faculty Scholarship to G.K.H.

Bennett
,
M. V. L.
,
Spira
,
M. E.
&
Pappas
,
G. D.
(
1978
).
Permeability of gap junctions between embryonic cells of Fundulus-. a re-evaluation
.
Devi Biol
.
65
,
114
120
.
Billingham
,
R. E.
&
Silvers
,
W. K.
(
1967
).
Studies on the conservation of epidermal specificities of skin and certain mucosas in adult mammals
.
J. exp. Med
.
125
,
429
446
.
Briggaman
,
R. A.
&
Wheeler
,
C. E.
(
1975
).
The epidermal-dermal junction
.
J. Invest. Dermatol
.
65
,
71
84
.
Cavoto
,
F. V.
&
Flaxman
,
B. A.
(
1972
).
Communication between normal human epidermal cells in vitro
.
J. invest. Dermatol
.
59
,
370
374
.
Coon
,
H. G.
(
1968
).
Clonal culture of differentiated rat liver cells
.
J. Cell Biol
.
39
,
29
a.
Cox
,
A. J.
(
1969
).
The dermal-epidermal junction in psoriasis
,
J. invest. Dermatol
.
53
,
428
435
.
Dhouailly
,
D.
,
Rogers
,
G. E.
&
Sengel
,
P.
(
1978
).
The specification of feather and scale protein synthesis in epidermal-dermal recombinations
.
Devi Biol
.
65
,
58
68
.
Epstein
,
M. L.
&
Gilula
,
N. B.
(
1977
).
A study of communication specificity between cells in culture
.
J. Cell Biol
.
75
,
769
787
.
Fentiman
,
I.
&
Taylor-Papadimitriou
,
J.
(
1977
).
Cultured human breast cancer cells lose selectivity in direct intercellular communication
.
Nature, Lond
.
269
,
156
157
.
Fentiman
,
I.
,
Taylor-Papadimitriou
,
J.
&
Stoker
,
M.
(
1976
).
Selective contact-dependent cell communication
.
Nature, Lond
.
264
,
760
762
.
Finbow
,
M. E.
&
Yancey
,
S. B.
(
1980
).
The roles of intercellular junctions
.
In The cell surface
(ed.
E. P.
Knox
).
Illinois
:
CRC Press
. (In Press.)
Flagg-Newton
,
J. L.
,
Simpson
,
I.
&
Loewenstein
,
W. R.
(
1979
).
Permeability of the cellcell membrane channels in mammalian cell junctions
.
Science, N.Y
.
205
,
404
409
.
Frost
,
P.
,
Weinstein
,
G. D.
&
Nyhan
,
W. L.
(
1970
).
Diagnosis of Lesch-Nyhan syndrome by direct study of skin specimens
.
J. Am. Med. Ass
.
212
,
316
319
.
Furshpan
,
E. J.
&
Potter
,
D. D.
(
1959
).
Transmission at giant motor synapses of the crayfish
.
J. Physiol
.
143
,
289
325
.
Gaunt
,
S. J.
&
Subak-Sharpe
,
J. H.
(
1979
).
Selectivity in metabolic cooperation between cultured mammalian cells
.
Expl Cell Res
.
120
,
307
320
.
Gilbert
,
S. F.
&
Migeon
,
B. R.
(
1975
).
D-valine as a selective agent for normal human and rodent epithelial cells in culture
.
Cell
5
,
11
17
.
Gilula
,
N. B.
,
Reeves
,
O. R.
&
Steinbach
,
A.
(
1972
).
Metabolic coupling, ionic coupling and cell contacts
.
Nature, Lond
.
235
,
262
265
.
Goshima
,
K.
&
Tonomura
,
Y.
(
1969
).
Synchronous beating of embryonic mouse myocardial cells mediated by FL cells in monolayer culture
.
Expl Cell Res
.
56
,
387
392
.
Hinz
,
R. W.
&
Syverton
,
J. T.
(
1959
).
Mammalian cell cultures for the study of influenza virus: preparation of monolayer cultures with collagenase
.
Proc. Soc. exp. Biol. Med
.
101
,
19
22
.
Lawrence
,
T. S.
,
Beers
,
W. H.
&
Gilula
,
N. B.
(
1978
).
Hormonal stimulation and cell communication in cocultures
.
Nature, Lond
.
272
,
501
506
.
Lo
,
C. W.
&
Gilula
,
N. B.
(
1979
).
Gap junctional communication in the post-implantation mouse embryo
.
Cell
18
,
411
422
.
Loewenstein
,
W. R.
(
1976
).
Permeable junctions
.
Cold Spring Harbor Symp. quant. Biol
.
40
,
49
63
.
Loewenstein
,
W. R.
(
1979
).
Junctional intercellular communication and the control of growth
.
Biochim. biophys. Acta
560
,
1
65
.
Loewenstein
,
W. R.
&
Kanno
,
Y.
(
1964
).
Studies on an epithelial gLond junction
.
J. Cell Biol
.
22
,
565
586
.
Macpherson
,
LA.
&
Stoker
,
M. G. P.
(
1962
).
Polyoma transformation of hamster cell clones - an investigation of genetic factors affecting cell competence
.
Virology
16
,
147
151
.
Marin
,
G.
&
Littlefield
,
J. W.
(
1968
).
Selection of cell lines from polyoma-transformed BHK21/13 hamster fibroblasts
.
J. Virol
.
2
,
69
77
.
Moreno
,
G.
,
Vinzens
,
F.
&
Prunieras
,
M. J.
(
1978
).
UV induced unscheduled DNA synthesis in guinea pig skin melanocytes isolated in culture
.
J. invest. Dermatol
.
70
,
21
24
.
Münch
,
E.
(
1930
).
Die Stoffbewegung in die Pflanze
.
Jena
:
Gustav Fischer
.
Pitts
,
J. D.
(
1971
).
Molecular exchange and growth control in tissue culture
.
Ciba Found. Symp.: Growth control in cell cultures
(ed.
G. E. W.
Wolstenholme
&
J.
Knight
), pp.
89
105
.
Edinburgh and London
:
Churchill Livingstone
.
Pitts
,
J. D.
(
1977
).
Direct communication between animal cells
.
In International cell biology
(ed.
B. R.
Brinkley
&
K. R.
Porter
), pp.
43
49
.
New York
:
Rockefeller University Press
.
Pitts
,
J. D.
(
1980
).
The role of junctional communication in animal tissues
.
In Vitro (in Press)
.
Pitts
,
J. D.
&
BüRk
,
R. R.
(
1976
).
Specificity of junctional communication between animal cells
.
Nature, Lond
.
264
,
762
764
.
Pitts
,
J. D.
&
Simms
,
J. W.
(
1977
).
Permeability of junctions between animal cells; intercellular transfer of nucleotides but not macromolecules
.
Expl Cell Res
.
104
,
153
163
.
Regnier
,
M.
,
Delescluse
,
C.
&
Prunieras
,
M.
(
1973
).
Studies on guinea pig skin cell cultures
.
Acta dermatovener
53
,
241
247
.
Riley
,
P. A.
(
1970
).
Mechanism of pigment cell toxicity produced by hydroxyanisole
.
J. Path. Bact
.
101
,
163
169
.
Rindler
,
M. J.
,
Chuman
,
L. M.
,
Shaffer
,
L.
&
Saier
,
M. H.
(
1979
).
Retention of differentiated properties in an established dog kidney epithelial cell line (MDCK)
.
J. Cell Biol
.
81
,
63S
648
.
Sheridan
,
J. D.
(
1976
).
Cell coupling and cell communication during embryogenesis
.
In The cell surface in animal embryogenesis and development
(ed.
G.
Poste
&
G. L.
Nicholson
), pp.
409
447
.
Amsterdam
:
North-HolLond
.