Metabolic cooperation between argininosuccinate synthetase and argininosuccinate lyase deficient human fibroblasts

Experimental

Cell Research 150 (1984) 367-378

Metabolic Cooperation between Argininosuccinate Synthetase and Argininosuccinate Lyase Deficient Human Fibroblasts J. S. DAVIDSON,

I. M. BAUMGARTEN

and E. H. HARLEY

Department of Chemical Pathology, University of Cape Town, Medical School, Observatory 7925, Cape Town, South Africa

Human tibroblasts deficient in either argininosuccinate synthetase or argininosuccinate lyase show low levels of incorporation of [‘4C]citrulline into protein. However, when these two cell types are co-cultured [14C]citrulline incorporation is restored to the levels found in control cultures. This metabolic cooperation is cell-density-dependent and does not occur by diffusion of argininosuccinate into the medium. Our results indicate that argininosuccinate passes between the two cell types via intercellular junctions, and this system provides a simple and accurately quantifiable model for the study of intercellular communication.

Many types of eukaryotic cells in organs and tissues, as well as in culture form intercellular junctions that allow the direct passage of small molecules from the cytoplasmic compartment of one cell to that of a neighbouring cell [l, 21. This permeability is reported to be selective only with respect to molecular size, and permits the transfer between cells of molecules of molecular weight less than about 1OOOD [3, 41. There is much evidence that the structure responsible for this intercellular transfer of small molecules is the gap junction, and electron microscopic studies show that gap junctions are composed of six protein subunits arranged hexagonally to form a tube which spans the plasma membranes of the two adjacent cells

r51* Normal human libroblasts contain both argininosuccinate synthetase (ASS; EC6.3.4.5) and argininosuccinate lyase activity (ASL; EC4.3.2.1) and are therefore able to convert citrulline into arginine [6]. The autosomal recessive diseases citrullinemia (McKusick 21570) and argininosuccinic aciduria (McKusick 20790) result from deficiencies of ASS and ASL respectively [7] and fibroblasts from patients with these disorders show lower rates of incorporation of [‘4C]citrulline into protein than do normal fibroblasts [8]. This has been used in the prenatal diagnosis of these disorders [9]. Here we report a system using measurement of [i4C]citrulline incorporation relative to [3H]leucine by co-cultures of these mutant tibroblasts. In this system [i4C]citrulline incorporation depends on the transfer of metabolites between the two cell types via intercellular junctions. The amount of [‘4C]citrulline incorporCopyright @I 1984 by Academic Press, Inc. All rights of reproduction in any form reserved 0014-4827/84 $03.00

368 Davidson,

Baumgarten

and Harley

ation therefore provides a quantifiable measure of the degree of intercellular junctional communication between the two cell lines. Advantages of this system are the ease and precision of quantification, together with a very low background measurement. MATERIALS

AND

METHODS

Cell Cultures Fibroblast line F25 was established from a skin biopsy from a neonate who died in the first week of life from citrullinemia. The biochemical features included hyperammonemia, raised urinary erotic acid and raised serum citrulline concentration and will be reported fully elsewhere. Fibroblast line F199 was obtained from a skin biopsy from an infant who died aged 8 days from argininosuccinic aciduria. The biochemical findings, to be reported elsewhere, included hyperammonemia, raised urinary erotic acid and raised serum and urinary argininosuccinic acid. Lymphocytes from 5 ml of this patient’s blood were transformed with Epstein Barr Virus [lo] and the resulting lymphoblast line was designated L199. Control tibroblast lines FG, F139, F140 and F200 were grown from skin biopsies from normal volunteers in plastic tissue culture flasks (Falcon). All cell cultures were maintained on Ham’s FlO medium (Gibco) supplemented with 15% fetal calf serum (FCS) (Gibco) and penicillin (30 mgil), streptomycin (50 mg/l) and neomycin (25 mg/l), in 5% CO2 at 37°C. All cell lines were routinely checked for mycoplasma contamination by culture and r3H]thymidine autoradiography. Some 1-7 days prior to using the cells for labelling experiments or autoradiography, the medium was changed to Eagle’s Basal Medium (BME, Flow Laboratories) with 15 % FCS and antibiotics as above. This was done to reduce the intracellular arginine pool to the minimum required for growth. Ham’s FlO medium contains 1 mmol/l arginine, while BME contains 0.1 mmokl.

Preparation

of Co-cultures

Confluent monolayers of F25 and F199 were treated with trypsin (0.1%) and EDTA (0.5 mmohl) in phosphate-buffered saline to detach the cells. The cell suspensions were agitated by pipetting to disrupt clumps of cells and an appropriate volume of Eagles’ Basal Medium with 15% FCS and antibiotics was added. Except where otherwise stated the F25 and F199 suspensions were mixed in a ratio to give equal numbers of cells. Where the two cell types were mixed in different ratios, aliquots of each cell suspension were counted in quadruplicate using a Coulter counter (model ZF), and the volumes of each suspension to be mixed were adjusted accordingly to give the desired ratio. The mixed suspensions were then pipetted into 25 cm’ flasks at a density of approx. 1.4~ lo6 cells/flask. Radio-isotope labelling experiments were performed on the following day, except where otherwise stated.

Incorporation

of Radioisotopes

into TCA-precipitable

Material

Four different labelling media were employed. Media A, B and C consisted of serum-free BME with antibiotics as described above and the following isotopes were added: (A) [‘4C](ureido)-citrulline (0.5 t&i/ml, 0.01 mmol/l, [‘Hlleucine (0.5 @ml, 0.2 mmol/l); (B) [‘4C]citrulline (2 @i/ml, 0.04 mmol/l), [3H]leucine (2 $X/ml, 0.2 mmohl); (C) [‘4C]leucine (0.125 @i/ml, 0.2 mmol/l), [3H]arginine (1 @/ml, 0.1 mmolil). Medium (D) was arginine-free Minimal Essential Medium; (Flow Laboratories) without serum and with the following isotopes: [14C]citrulline (0.25 uCi/ml, 0.805 mmol/l), [3H]leucine (1 $i/ml, 0.4 mmohl). Incubations in media A, C and D were for 4 or 5 h and for 1 h in medium B, at 37°C and 5 % CO2 in 2 ml of labelling medium. Medium A was used for all experiments except those shown in table 1 (medium C), fig. 3 (medium D), and fig. 4 (medium B). The molar incorporation ratio (moles citrulline incorporated per mole of leucine) was calculated as follows: Exp

Cell

Res 150 (1984)

Metabolic

cooperation

in mutantfibroblasts

Table 1. Incorporation cultures of the mutant

of [‘4C]leucine relative to [3H]arginine cell lines, their co-cultures, and two normal

Cell line

3H dpm

14C dpm

14C dpm ‘H dpm

F25 (ASS-deficient)

53 575 44 661

26 228 22 736

0.490 0.509

Fl99 (ASL-deficient)

41 448 33 591

19 468 15 819

0.470 0.471

F25iF199 Co-culture

44 182 54 381

22 387 26 419

0.507 0.486

F139 (Control)

23 577 20 875

13 128 11 309

0.557 0.542

F140 (Control)

28 162 34 604

15 902 18 233

0.565 0.527

molar incorporation

369

in duplicate cell lines

ratio = 14C dpm x S.A. (leu) S.A. (tit) ‘H dpm ’

where S.A. (tit) and S.A. (leu) are the final specific activities in the labelling medium of citrulline and leucine respectively. Isotopes were from The Radiochemical Centre, Amersham, England. No unlabelIed citrulline was added, except where stated. After incubation in labelled medium, the cells were washed three times in normal saline and harvested with 2 ml of trypsin/EDTA. 8 ml of 10% w/v trichloracetic acid (TCA) was added to each cell suspension. The precipitate was washed twice in 10 ml of 10% w/v TCA, dissolved in 0.2 ml of 0.1 N NaOH, and mixed with 10 ml of Instagel Scintillant (Packard Instruments, Downers Grove, Ill.) and counted in a Beckman Model LS250 scintillation spectrometer. Disintegrations per minute (dpm) were determined for each isotope with windows optimized for double-labelling conditions and using external standardization.

Autoradiography Cell suspensions of F25 and F199 and mixtures of these suspensions were plated into chambers of 8-chambered tissue culture slides (Lab-Tek Products, Naperville, Ill.). After 2 h the medium was replaced with BME (serum-free) containing 2 @i/ml of [‘4Clcitrulline and incubated overnight. The cells were then washed twice with serum-free BME and incubated in serum-free BME with 1 mmol/l unlabelled citrulline for 2 h. After washing the cells with phosphate-buffered saline, the slides were placed in 10% formaldehyde in normal saline for 15 min and washed in running water for 30 min. After air-drying, the slides were coated with Kodak NTB-2 emulsion, exposed at -70°C for 2 weeks, developed, and stained lightly with methylene blue.

RESULTS Double-labelling

Method

The incorporation ratio of [‘4C]leucine to 13Hlarginine was found to be similar in the two mutant cell lines, F25 and F199, in co-cultures of these cells, and in two normal cell lines F139 and F140 (table 1) and varied by less than 8% with different degrees of confluence of the cultures (data not shown). Differences in the incorporation ratio of [14C]citrulline to L3Hlleucine should therefore reflect differences in the rate of conversion of citrulline to arginine. Erp

Cell

Res

150 (1984)

370

Davidson,

Baumgarten

and Harley

Table 2. Metabolic co-operation in co-cultures of differing proportions of ASSdeficient and ASL-deficient human fibroblasts, measured by incorportion of [‘4C]citrulline relative to [3H]leucine into TCA-precipitable material F25 (ASS-deficient)

F199 (ASL-deficient)

3H dpm

14C dpm

14C dpm ‘H dpm ratio

100 98 96 94 90 75 50 25 10 6 4 2 0

0 2 4 6 10 25 50 75 90 94 96 98 1loo

30 31 31 27 36 38 28 34 32 36 32 35 36

17 240 501 608 1 253 2 825 3 308 4 356 3 656 3 024 2 253 1 346 25

0.0006 0.0077 0.0159 0.0223 0.0342 0.0739 0.1145 0.1260 0.1112 0.0826 0.0690 0.0384 0.0007

931 116 444 220 664 221 889 585 871 620 639 087 170

In all experiments in which cells were labelled with [‘4C]citrulline and [‘Hlleutine, the t4C dpmPH dpm ratio in TCA-precipitable material was found to be measurable with high precision within experiments (e.g., see table 4). The mean difference between duplicates in the 14C dpm/3H dpm ratio in 26 pairs of duplicate flasks in six different experiments was 5.3 %+4.4 (mean + SD). Metabolic Co-operation between F25 and F199 Cells

Both F25 (ASS-deficient) and F199 (ASL-deficient) cells showed very low levels of incorporation of [14C]citrulline relative to [3H]leucine when cultured separately. However, when these cells were co-cultured [‘4C]citrulline incorporation increased by an average of 130-fold, in seven experiments performed on different occasions (fig. 1). By mixing the two cell types in varying proportions and plotting the 14C dpm13H dpm ratio against percentage of cell type, a curve was obtained which showed that complementation reached a maximum with a mixture of 25 % F25 and 75 % F199 cells (table 2, fig. 2 a). This maximum level was of the same order as and often higher than the level found in normal fibroblasts (fig. 1). The curve of metabolic co-operation as a function of varying proportions of the two cell types (fig. 2a) was skewed towards the right, indicating that the admixture of as little as 4 % ASS-deficient cells with a culture of ASLdeficient cells was capable of restoring [14C]citrulline incorporation relative to [3H]leucine by the whole culture to half-maximal levels. The assay was sensitive enough to detect metabolic co-operation in co-cultures consisting of 98 % of one cell type and 2% of the other cell type. The data of table 2 could also be expressed as mean flux of citrulline-derived label relative to leucine-derived label Exp Cell

Res

150 (1984)

Metabolic

cooperation

in mutant fibroblasts

37 1

l

3

t

Fig. 1. Incorporation of [‘4C]citrulline relative to [3H]leucine into TCA-precipitable material from cocultures of F25 (ASS-deficient) and FlW (ASL-deficient) cells compared with separate cultures of these cells and normal fibroblasts. This represents cumulated data from experiments performed on seven different occasions. Fig. 2. (a) Graphic representation of data shown in table 2: Metabolic cooperation between F25 and Fi99 fibroblasts as a function of the proportion of each cell type in the co-culture; (b) Mean citrullinederived relative flux per cell as a function of the percentage (p) of ASL-deficient cells in the coculture. Mean flux per ASS-deficient ceil (0) is calculated as (14C dpm/‘H dpm) x [100/(100/p)]. Mean flux per ASL-deficient cell (0) is calculated as (14C dpm/3H dpm) x (100/p). The bars in the centre of the figures represent the mean + SD of the values found in normal fibroblasts (the data for normal tibroblasts is shown in fig. 1).

per cell of a given type (fig. 2 b). The resulting curves showed that when ASSdeficient cells constituted 2 % of the co-culture, the relative citrulline-derived flux per ASS-deficient cell was 20-fold greater than in normal cells under the same conditions. When ASL-deficient cells constituted 2% of the co-culture the relative citrulline-derived flux per ASL-deficient cell was 4-fold greater than the flux per normal cell under the same conditions. Kinetics

of Citrulline

Incorporation

The maximal rate of citrulline incorporation was determined for a normal cell line and for a co-culture of 25 % ASS-deficient and 75 % ASL-deficient cells by adding increasing amounts of unlabelled citrulline to the labelling medium. The maximal rate of citrulline incorporation in the co-culture was similar to that in a normal cell line (fig. 3). At a citrulline concentration of 2 mmol/l the rate of citrulline incorporation was 2.8~ IO5 molecules of citrulline per ASS-deticient cell per second. Both the co-culture and the normal cell line showed saturable kinetics, with half-maximal citrulline incorporation at a citrulline concentration of about 0.2 mmol/l (fig. 3). Exp Cell Res I50 (1984)

372 Davidson,

Baumgarten

and Harley

TIME (hours1

Fig. 3. Incorporation of citrulline as a function of citrulline concentration in the medium, expressed as moles of citrulline per mole of leucine incorporated into TCA-precipitable material. 00, Normal fibroblasts (F140); O-O, co-culture consisting of 25% ASS-deficient (F25) cells and 75% ASLdeficient (Fl99) cells. Fig. 4. Time course of establishment of metabolic co-operation. Co-cultures of F25 and F199 tibroblasts were incubated in duplicate with serum-free BME containing [‘4C]citrulline (2 @/ml) and [3H]leucine (2 @i/ml) for 1 h at the indicated times after mixing the cell suspensions, and radioactivity was determined in TCA-precipitable material.

Time of Establishment of Metabolic

Co-operation

Serial measurements at varying times after mixing the cell suspensions (fig. 4) showed that metabolic co-operation was well established after 1 h of co-culture and was maximal by 2-3 h. Dependence of Metabolic Co-operation on Cell Density

A suspension of F25 and F199 cells in equal numbers was plated onto five tissue culture dishes of different sizes, using the same volume of cell suspension for each dish, so that in the smallest dishes the cells were confluent and in the largest dishes they were sparse. Metabolic co-operation as measured by the 14C dpm3H dpm ratio increased with increasing cell density (fig. 5). Common Medium

To test whether the metabolic co-operation was due to release of a metabolic intermediate into the medium, four 2%cm* dishes were divided into two compartments with partitions, sealed with agarose. F25 and Fl!B cells were seeded into adjacent halves of the dishes at confluent density. Co-culture dishes were identical, except that there were no partitions, and were seeded with a mixed suspension of F25 and F199 cells in equal numbers. On the following day the partitions were removed from two of the dishes, and all dishes were incubated with 2 ml of labelling medium, with gentle shaking (60 r-pm) on a rotary platform. In the dishes with the partitions removed therefore, the two cell types, although physically separated, were bathed in the same medium. No increase in 14C dpm3H dpm ratio was seen in the dishes with the partitions removed (table 3) indicating that Exp Cell

Res I50 (1984)

Metabolic

a

“C DPM o.Dn

38 0.230

cooperation

293

5m

430

144

713

2.32

3.41

464

5.26

1.11

in mutant fibroblasts

373

a “CDPM

RAllD

5

4

xim

FISS (ASL-I

F25 (ASS-1

7

Fig. 5. Dependence of metabolic co-operation on cell density. A suspension of approximately equal numbers of F25 and F199 cells was plated into five dishes of different sizes, the same volume of suspension being used for each dish. After incubation for 4 h with [‘4C]citrulline and [3H]leucine, 14C dpm and 3H dpm were measured in TCA-precipitable material. Fig. 7. Proposed mechanism of metabolic co-operation between F25 and F199 cells.

metabolic co-operation in this system is not due to release of a metabolic intermediate into the medium. Competition by Argininosuccinate

in the Medium

The addition of 1 mmol/l of unlabelled argininosuccinate to the labelling medium caused a small but significant decrease in the incorporation of [‘4C]citrulline in co-cultures of F25 and F199 cells. A similar small decrease was seen in a normal fibroblast cell line (table 4).

Table 3. Absence of metabolic co-operation when F25 and F199 cells are physically separated and incubated in duplicate in the same labelling medium, containing [‘4CJcitrulline and [3HJleucine

14Cdpmloc)

‘H dpm

t4C dpm

F25 and Fl99 Cells separate, media separate

35 183 24 658

32 17

0.09 0.07

F25 and F199 Cells separate, common medium Co-culture in common medium

35 736 35 334

24 49

0.07 0.14

42 004 35 053

4244 3599

‘H dpm

10.1 10.3 Exp Cell Res 150 (1984)

374 Davidson,

Baumgarten

and Harley

Table 4. Effect of adding unlabelled argininosuccinate to labelling mediuiiz A bn citrulline incorporation in co-cultures of F2.5 and F199 cells and on normal fibroblasts (FG) Argininosuccinate (mmol/l)

Cell line FG (control)

0 1

F2YF199

0

Co-culture

1

‘H dpm

14C dpm

14C dpm x 100 ‘H dpm

50 836 57 721 52 993 14 184 63 241 53 236 41 601 48 141

5 192 5 190 4 154 1 069 5 851 5 078 3 533 3 990

10.2 10.0 7.84 7.54 9.25 9.53 8.49 8.29

Lack of Metabolic Co-operation between F25 and ASL-deficient Tra&ormed Lymphoblasts

When the Epstein-Barr Virus-transformed ASL-deficient lymphoblast cell line L199 was co-cultured with F25 fibroblasts, metabolic co-operation did not occur, in contrast to the metabolic co-operation which was seen between F25 cells and F199 fibroblasts under the same conditions (table 5). Autoradiography

F25 cells were not distinguishable from F199 cells morphologically. When cultured separately in [14C]citrulline, both F25 and F199 tibroblasts showed only negligible labelling above background, which was present equally in all cells (fig. 6A, D). When 97 % F25 were co-cultured with 3 % F199 cells, most cells showed no difference in labelling from the F25 culture, while isolated groups of 5-15 adjacent cells were more heavily labelled. One such group is shown in fig. 6B. Within these groups the few central cells were equally heavily labelled, while other cells at the edge of the group were less heavily labelled. When 3 % F25 were co-cultured with 97 % F199, groups of 5-15 heavily labelled cells were again seen (fig. 6 C). In these latter groups there was usually a single cell at the centre of the group which was very heavily labelled, with surrounding adjacent cells less heavily labelled. These central cells were present at a frequency of approx. 3 % and are presumed to be the F25 synthetase-deficient cells. DISCUSSION It has been previously reported [8, 111 that ASS-deficient and ASL-deficient human fibroblasts show complementation with respect to citrulline incorporation, Exp Cell

Res 150 (1984)

Metabolic Table 5. Absence of metabolic and ASL-deficient lymphoblasts

cooperation

co-operation

in mutant fibroblasts

between ASL-deficient

375

Jibroblasts

1 x lo6 Epstein-Barr Virus-transformed ASL-deficient lymphoblasts (L199) were co-cultured with 6xld ASS-deficient tibroblasts (F25), and [‘4C]citrulline incorporation relative to [3H]leucine was compared with that in a co-culture of 6x 16 F25 and the same number of F199 cells

14Cdpml~ Cell line

3H dpm

14C dpm

‘H dpm

F25

40900 41 910

23 25

0.06 0.06

F2YF199 Co-culture

34 797 30 146

3 571 3 242

F199

20 381 20 336

19 20

0.09 0.10

F25IL199 Co-culture

41 190 46 750

21 26

0.06 0.06

L199

25 057 20 781

15 14

0.06 0.07

10.3 10.7

without the necessity for cell fusion. However, the mechanism of this complementation has not been studied in detail. A likely mechanism for the complementation observed between these human mutant fibroblast cell lines can be inferred from the results presented above. The dependence of complementation on cell density (fig. 5) shows that it requires intercellular contact. The absence of demonstrable transfer of labelled intermediates through the medium (table 3) implies that if transfer of a metabolic intermediate between the two cell types occurs, it is through intercellular junctions which are not leaky to the outside. The obvious candidate for such an intermediate is argininosuccinate, and large amounts of added unlabelled argininosuccinate caused only a small decrease [‘4C]citrulline incorporation (table 4). This indicates that the [‘4C]argininosuccinate passes directly between cells and not via the medium in any significant quantity, where it would be diluted by the unlabelled argininosuccinate at least lOO-fold. Unlabelled argininosuccinate caused a similar small decrease in [‘4C]citrulline incorporation in a normal cell line (table 4) showing that the decrease seen in the co-culture is probably due to intracellular dilution of [ 14C]argininosuccinate. The mechanism shown in fig. 7 is therefore proposed: in the F199 cells [‘4C]citrulline is converted to [14C]argininosuccinate which passes through intercellular junctions to the F25 cells where it is converted into [14C]arginine. The [14C]arginine is incorporated into protein in the F25 cells, and some of the [r4C]arginine passes back through the junctions to the F199 cells to be incorporated into protein. The evidence for the latter is the lighter labelling of neighbouring cells relative to the central heavily labelled cells seen in the autoradiographs (fig. 6C). Exp

Cell

Res

150 (19&f)

376 Davidson,

Baumgarten

and Harley

Fig. 6. Autoradiography of F25/F199 co-cultures after incubation with [14C]citrulline. (A) F25 cells only; (B) 97% F25 and 3 % F199 cells; (C) 3 % F25 and 97% F199 cells; (D) F199 cells only.

Exp Cell

Res 150 (1984)

Metabolic

cooperation

in mutant Jibroblasts

377

The finding that LlW (ASL-deficient) lymphoblasts did not show metabolic cooperation with F25 fibroblasts supports the conclusion that metabolic co-operation does not occur by simple diffusion of a metabolite into the medium, and indicates that EBV-transformed lymphoblasts do not from intercellular junctions with fibroblasts. In this respect EBV-transformed lymphoblasts do not differ from unstimulated or PHA-stimulated lymphocytes, which have been reported not to show metabolic co-operation 1121. An interesting feature of this metabolic co-operation system is the fact that the rate of citrulline incorporation relative to leucine can be as high in co-cultures as in normal fibroblasts, even when only 6% of the cells in the co-culture contain ASL. According to the model proposed above, in co-cultures containing only a small percentage of ASS-deficient cells, these latter cells have the capacity to convert to arginine all the argininosuccinate supplied by the excess of ASLdeficient cells. The maximal rate of citrulline incorporation was as high in the coculture as in a culture of normal cells (fig. 3), indicating that the flux of citrulline to arginine was not limited by the capacity of the intercellular junctions to transmit argininosuccinate. In co-cultures containing a small percentage of ASLdeficient cells, the flux of citrulline to argininosuccinate per ASL-deficient cell was 4-fold greater than in normal cells. This could be a consequence of transfer of citrulline from ASS-deficient cells via intercellular junctions to ASL-deficient cells, if transport of citrulline into cells is normally rate-limiting for this pathway. Alternatively increased removal of argininosuccinate by the excess of ASSdeficient cells could result in increased flux, since ASS is known to be inhibited by argininosuccinate [ 131. Many different model systems using cultured cells have been used to study intercellular communication, and these have been recently reviewed [2]. The system presented here has the advantages that the method is simple and rapid to perform, and the results are accurately quantifiable. The use of a second isotope ([3H]leucine in this case) corrects for errors due to differences in the number of cells in each flask and due to losses of TCA-precipitable material during the washing procedure. Because incorporation of radioactivity occurs only as a result of metabolic cooperation, the background measurement is very low. In addition, the method does not require autoradiography or selective cell-killing techniques. ASS-deficient cells have previously been used in studies of metabolic cooperation [ 141. These authors used Chinese hamster Don cells which are ASS-deficient and suffer arginine starvation when placed in citrulline-supplemented argininefree medium. Cells containing ASS are able to relieve the arginine starvation in the Don cells by transferring arginine and/or argininosuccinate via intercellular junctions. It was necessary for arginase and ASL to be present in the medium in order to prevent non-junctional transfer of these metabolites. In the system described here non-junctional transfer is negligible, as shown by the absence of metabolic cooperation when cells which were not in contact were incubated in a common medium. Exp

Cell

Res

150 (1984)

378 Davidson,

Baumgarten

and Harley

Recently the tumour-promoting phorbol esters [ 15, 161, saccharin [ 171, and retinoic acid [18], have been reported to inhibit metabolic co-operation in other systems. This system may be valuable for the study of such compounds which inhibit intercellular communication, and further studies along these lines are in progress. REFERENCES 1. Loewenstein, W R, Biochim biophys acta 560 (1979) 1. 2. Hooper, M L, Biochim biophys acta 651 (1982) 85. 3. Schwarzmann, G, Wiegandt, H, Rose, B, Zimmerman, A, Ben-Haim, D & Loewenstein, W R, Science 213 (1981) 551. 4. Finbow, M E & Pitts, J D, Exp cell res 131 (1981) 551. 5. Unwin, P N T & Zampighi, G, Nature 283 (1980) 545. 6. Tedesco, T A & Mellman, W J, Proc natl acad sci US 57 (1967) 829. 7. Walser, M, The metabolic basis of inherited disease (ed J B Stanbury, J B Wyngaarden, D S Frederickson, J L Goldstein & M S Brown) 5th edn, p. 402. McGraw-Hill, New York (1983). 8. Kennaway, N G & Curtis, H C, J inher metab dis 4 (1981) 23. 9. Fleisher, L D, Rassin, D K, Desnick, R J, Salwen, H R, Rogers, P, Bean, M & Gaull, G E, Am j hum genet 31 (1979) 439. 10. Sugden, B & Mark, W, J virol(1977) 503. 11. Cathelineau, L, Dinh, L P, Briand, P & Kamoun, P, Hum genet 57 (1981) 282. 12. Cox, R P, Krauss, M R, Balis, M E & Dancis, J, Exp cell res 101 (1976) 411. 13. Tadaka, S, Saheki, T, Igarashi, Y & Katsunuma, T, J biochem 85 (1979) 1309. 14. Hooper, M L & Morgan, R H M, Exp cell res 119 (1979) 410. 15. Murray, A W & Fitzgerald, D J, Biochem biophys res commun 91 (1979) 395. 16. Mosser, D D & Bols, N C, Carcinogenesis 3 (1982) 1207. 17. Trosko, J E, Dawson, B, Yotti, L P & Chang, C C, Nature 285 (1980) 109. 18. Pitts, J D, Biirk, R R & Murphy, J P, Cell biol int rep 5 suppl. A (1981) 45. Received June 30, 1983 Revised version received August 22, 1983

Exp

Cell

Res IS0 (1984)

Printed

in Sweden