434
Eur. J. Immunol. 1979. 9: 434-439
H. B. Steen and T. Lindmo
Harald B. Steen and Tore Lindmo Department of Biophysics, Norsk Hydro’s Institute for Cancer Research, Montebello, Oslo
Initiation of the blastogenic response of lymphocytes by hyperoptimal concentrations of concanavalin A The blastogenic response of human lymphocytes in vitro to hyperoptimal concentrations of concanavalin A (Con A ) has been studied by means of volume spectroscopy (measuring cellular and nuclear volume), flow cytofluorometry (measuring cellular DNA content) and incorporation of [3H]thymidine ([3H]dThd). The optimal Con A dose with respect to [3H]dThd incorporation was about 30 pg/ml. In cultures given hyperoptimal doses, e.g. 100 pg/ml, [3H]dThd incorporation was strongly inhibited, whereas the number of cells entering S-phase and significantly increasing their cellular and nuclear volume was considerably larger than with 30 pg/ml. With 200 pg/ml Con A , which induced negligible rH]dThd incorporation, the percentage of responding cells was even larger. Hence, doses of Con A, which were hyperoptimal with regard to 13H]dThd incorporation, induced blastogenic response, including DNA synthesis, in a larger percentage of the cells than did the optimal dose. However, in cultures with hyperoptimal Con A doses, the progression of the cell cycle stagnated mainly during S- and G2-phase and few cells completed mitosis. Thus, the blocking effect of hyperoptimal doses was not confined to any particular point of the cell cycle. The reduced [3H]dThd incorporation, seen with hyperoptimal doses, is attributed partly to a failure of this assay under such conditions.
1 Introduction The mechanisms by which the blastogenic response of lymphocytes is triggered by antigens and mitogens binding to appropriate receptors on the plasma membrane are poorly understood. Various effects of stimulation have been reported to occur within minutes of exposure to mitogen (review in [l]). Some of these effects have been observed also in other types of cells, supporting the assumption that important features of the mechanisms of the receptor-mediated control of lymphocyte proliferation are common to cells of higher organisms in general [2, 31. It is not known, however, which of the observed effects, if indeed any of them, acts as a triggering signal, and how this signal is transferred from the receptor to the inner cell. One of the early effects of mitogen stimulation is aggregation of stimulant-binding receptors into “patches” which may subsequently merge to form a “cap”. The observation that all stimulatory agents acting through receptors are at least divalent, whereas the corresponding monovalent forms have no mitogenic effect, has been taken to indicate that receptor cross-linking is crucial for the triggering of lymphocyte activation [l]. However, recent reports that monovalent Fab fragments may cause blastogenic response [4, 51 seem to show that this is not always the case. It has been suggested [6] that some of the surface receptors are connected with the cytoplasmic system of microtubuli, possibly via actin and myosin-containing filaments [7], and that the initiation of mitogen response is caused by alterations in this system which are dependent upon redistribution of mitogen receptors. Thus, receptor mobility is considered being of [I 22761
crucial importance for the initiation of blastogenic response. An essential part of the basis for this hypothesis is the observation that hyperoptimal concentrations of concanavalin A (Con A) inhibit both patching [8] and capping [9] of receptors, while at the same time the blastogenic response, as measured by the incorporation of [3H]thymidine (13H]dThd), is considerably suppressed [6]. Furthermore, similar concentrations of the divalent succinyl-Con A, which does not inhibit patching and capping significantly, do not suppress blastogenic response either 161. However, the nature of the inhibitory effects of large mitogen concentrations is not clear. For example, it has been reported that hyperoptimal concentrations of Con A lead to extensive cell death indicating that reduced response may be a trivial effect of cytotoxicity [lo]. Others reported [ l l ] that hyperoptimal doses of Con A do not inhibit the initiation of blastogenesis, but only some later step in the process. Using flow cytometric methods, i.e. volume spectroscopy and flow cytofluorometry, which provide more detailed information on lymphocyte response than does the [3H]dThd assay, we have studied the inhibitory effects of high mitogen concentrations. It is found that in cultures given hyperoptimal concentrations of Con A, blast transformation, including DNA synthesis, is initiated in a larger proportion of the lymphocytes than in cultures with Con A concentrations which are optimal with regard to [3H]dThd incorporation. Thus, it appears that hyperoptimal doses of Con A do not inhibit the triggering of blastogenesis. But the subsequent stages of the blastogenesis are progressively suppressed and the frequency of mitosis is strongly reduced.
2 Materials and methods
Correspondence: Harald B. Steen, Department of Biophysics, Norsk Hydro’s Institute for Cancer Research, Montebello, Oslo 3, Norway
2.1 Cell culture
Abbreviations: amM: a-Methyl-D-mannoside Con A: Concanavalin A PBS: Phosphate-buffered saline PHA: Phytohemagglutinin [3H]dThd [MethylL3H]thymidine
Mononuclear leukocytes were separated from fresh, heparinized, human blood by centrifugation on a Ficoll-Triosil gradient (Lymphoprep, Nyegaard & Co., Oslo, Norway) [12],
0014-2980/79/0606-0434$02.50/0
0Verlag Chemie, GmbH, D-6940 Weinheim, 1979
Eur. J. Immunol. 1979. 9: 4 3 6 4 3 9
washed three times by centrifugation at 160 X g to remove platelets, and resuspended in medium RPMI 1640 (Gibco, Grand Island, NY), with antibiotics and 10% AB serum to a concentration of about 1.5 X lo6 cells/ml. The cells were cultured in 0.1 ml aliquots in microculture trays (Linbro Chemicals, New Haven, CT) at 37”C, 5% COz, and high humidity. Con A (Sigma, Chemical Co., St. Louis, MO), dissolved in RPMI 1640 to ten times the final concentration, was added to the cultures shortly after seeding. Agglutinated cells and cells adhering to the wells were dispersed by adding to each well 3 h before harvesting 0.1 ml of a solution containing 40 m~ EDTA (Sigma) and 100 m~ a-methyl-Dmannoside (ctmM, Sigma) in Ca and Mg-free phosphatebuffered saline (PBS, Gibco). Cell nuclei were prepared by suspending the cells in a counting solution (131 consisting of 1% formaldehyde (Merck, Darmstadt, FRG), 0.5 % glacial acetic acid (Merck) and 1 mg/ml Cetrimide (Sigma) in 0.85 % NaCI. Cells used in the three different assays were cultivated on the same tray.
Blastogenic response of lymphocytes
435
percentage of the cells continued to grow until cell division which was most frequent around day 3 (Fig. 1B). The nonresponding cells remained with constant volume, i.e. that of unstimulated controls, for the entire culture period. The average volume of cells, when they reached mitosis, was about four times that of the nonresponding cells. Hence, newly divided lymphoblasts had approximately twice the volume of resting lymphocytes. Thus, the large central peak in Fig. 1 B (96 h) is primarily due to newly divided cells. Monocytes were very scarce in cultures stimulated with the present doses of Con A, possibly due to agglutination which was not resolved by EDTA plus amM. In cultures with 100 pg/ml Con A (Fig. lC), the percentage of responding cells was considerably larger than in cultures with the optimal dose of Con A, i.e. 30 pg/ml. Furthermore, volume growth commenced earlier and proceeded at a greater rate with the higher concentration of mitogen. These effects -0 h
2.2 Volume spectroscopy
2.L
The Coulter-type volume spectrometer, which yields the distribution of cellular or nuclear volume within a cell culture, has been described elsewhere [14]. The full contents of each culture well were suspended in 10 ml Isoton I1 counting solution (Coulter Electronics Ltd., Harpenden, GB) for measurement of cell volume or in 10 ml of the counting solution described above for nuclear volume spectra.
1.6
IA)
20
1.2
08 0.L
1.2
1 .O 0.8
0.6
2.3 Flow cytofluorometry
01
For determination of cellular DNA content, the cells were stained with mithramycin (Pfizer) [15, 161, and DNA histograms were obtained by the flow cytophotometer described previously [17]. The output of this instrument was fed to a computer for calculation of the distribution of cells between the various phases of the cell cycle [18].
f? z
0.2
8
06
3
0.1
0.2
2.4 [3H]dThd assay Cell cultures were given a 65-min pulse of [3H]dThd (7 pCi/ ml; 5 Wmmol = 185 GBq/mmol), harvested by a microculture harvester (Skatron, Lierbyen, Norway) and counted in a liquid scintillation counter [19].
3 Results 3.1 Volume spectra Volume spectra, obtained after different periods of incubation with various concentrations of Con A, are shown in Fig. 1. The main peak at the lower volume represents the small, resting lymphocytes. The peak at higher volumes in the spectra of unstimulated cultures (Fig. 1A) is due to monocytes. It can be seen that the volume of the monocytes grew considerably during the first 2-3 days of culture as they matured into macrophages. In stimulated cultures, significant volume growth of the lymphocytes was observed after 24 h. In cultures with 30 pg/ml Con A, which was found to be the optimal concentration as measured by the [3H]dThd assay, a certain
0.L
0.2
CHANNELNUMBER(PROPORTIONAL
1112)6.0
TO CELL VOLUME)
Figure 1. Distributions of cellular volume in human lymphocyte cultures incubated with different doses of Con A for various periods of time (indicated). The cultures were made 20 m~ in EDTA and 50 m~ in amM 3 h before harvesting in order to disperse adherent and agglutinated cells. Each spectrum represents 3 X lo4- 5 X lo4 cells. The peak at about channel 26 represents unstimulated lymphocytes, whereas the peak at higher volumes in (A) is due to monocytes. The broad peak at about channel 60 in (B) (96 h) represents newly divided cells. The feature peaked at channel 15 in (D) is attributed to dead cells. Due to membrane defects, the cytoplasm of these cells is in electrolytic contact with the extracellular medium resulting in a reduced “electrical volume”. The assumed spectrum of nonresponding cells, i.e. a normalized spectrum of unstimulated cells, is indicated separately in (D).
436
Eur. J. Immunol. 1979. 9: 4 3 4 4 3 9
H. B. Steen and T. Lindmo
were further enhanced in cultures with 200 pg/ml (Fig. 1D). After day 1, however, the growth rate in cultures with the higher doses of Con A decreased drastically, so that after day 2 growth was negligible. The peak which is characteristic of newly divided cells does not appear in the spectra of cultures with hyperoptimal doses of Con A (Figs. 1C and D) indicating that little cell division took place. The shorter response time and the larger percentage of responding cells in cultures with hyperoptimal Con A doses appear more clearly from the data in Fig. 2, which shows the percentage of cells with enlarged volume as a function of time. It can be seen that the percentage of enlarged cells leveled out after about 30 h, indicating that growth had commenced in the majority of responding cells at this time. It appears that in cultures given the optimal dose of Con A, i.e. 30 pg/ml, approximately 35 5% of the cells responded with volume growth as compared to about 60% in cultures with 100 pg/ml. The percentage of responding cells was even higher in cultures with 200 pg/ml Con A, although an accurate estimate is complicated by extensive cell death in these cultures.
appeared in both types of spectra. The percentage of enlarged nuclei was in accordance with that of enlarged cells, showing a significant increase with mitogen dose also in the hyperoptimal dose range. An advantage of nuclear spectra is that the present method of preparing nuclei completely dispersed all agglutinated cells. Thus, microscopic investigations of nuclear suspensions showed negligible aggregation, i.e. less than 5 % of nuclei, confirming the original morphological observations of Stewart and Ingram [13]. Hence, the similarity of cellular and nuclear spectra eliminates the possibility that small aggregates of cells or debris were confused with enlarged cells to any significant extent. In comparison, up to 20% of whole cells from stimulated cultures could be found in a few larger aggregates after treatment with EDTA plus amM. Consequently, cell numbers have been based on nuclear volume spectra. Control
-0-0
30pglml con A IM)pg/mlcon A 2W&ml con A
--&-A -X-X
-0-
/A-A /A
0
I
0
lI2276.3!
INCUBATION TIME (HOURS)
Figure 2. The percentage of lymphocytes with enlarged volume as a function of time in cultures incubated with different doses of Con A (indicated). The intermediate plateau from about 30 h represents the situation where growth has commenced in the great majority of responding cells while cell division, giving rise to the secondary increase after 48 h, has not yet started. Data for cultures with 200 pg/ml Con A obtained after 24 h have been omitted because of extensive cell death which appears preferentially to exclude large cells (see Fig. 1D).
The subsequent increase in the number of enlarged cells beginning at about 48 h (Fig. 2) is due to cell division, as is evident from the increase in the total number of cells/culture beginning at this time (Fig. 3). It can be seen that this increase was much smaller in cultures with 100 pg/ml Con A than in cultures with the optimal dose. From the data in Figs. 2 and 3, we calculate that in cultures with 100 pg/ml Con A no more than 22% of the responding cells completed mitosis as compared to more than 80% in cultures with 30 pg/ml Con A. In the cultures with the highest dose of Con A no cell division could be discerned, and the number of cells with intact nuclei fell drastically indicating extensive cell death. The volume spectra of nuclear suspensions were similar to those of whole cells, except that the relative increase of nuclear volume during blastogenesis was somewhat smaller than that of cellular volume, i.e. nuclear volume increased by a factor of about 3 up to mitosis* [14]. Thus, the same spectral features
* Steen, H. B. and Nielsen, V., Scand. J. Immunol., in press.
I
I
I
26
I
I
I
12
40
I
I 96
INCUBATION TIME (HOURS)
Figure 3. The total number of intact nuclei per culture as a function of the time of incubation with various concentrations of Con A (indicated).
3.2 DNA histograms DNA histograms, recorded after various times of incubation with the different concentrations of Con A, are shown in Fig.4. The histograms obtained for cells incubated with 30 yg/ml Con A are typical of proliferating cultures, with cells evenly distributed over the entire S-phase. In contrast, the histograms of cultures with 100 yg/ml Con A show an abnormally high percentage of cells in early S, whereas fewer cells reached G 2 . In comparing these histograms, one should keep in mind that in cultures with 30 pg/ml Con A cells were lost from G 2 because of cell division, whereas in cultures with 100 pg/ml Con A little division occurred.
+
+
The distribution of cells between GO G I , S and G2 M in the various cultures is depicted in Fig. 5. It can be seen that with 100 pg/ml Con A a significantly larger number of cells entered S-phase than with the optimal concentration of 30 yg/ ml. Whereas with the optimal concentration of Con A the percentage of cells in S and G 2 M decreased after day 3 as a result of cell division, little change is seen for cultures with the higher concentration of Con A.
+
3.3 C3H]dThd incorporation The rate of DNA synthesis, as measured by the [3H]dThd assay, is shown as a function of time and Con A concentration in
Eur. J. Immunol. 1979. 9: 434-439
Blastogenic response of lymphocytes
437
100 gglml con A
Lt
200pglml con A
lI2275.5(
INCUBATION TIME (HOURS)
ICI Figure5. The percentage of cells in the various phases of the cell cycle as a function of the time of incubation with various doses of Con A (indicated). The data were obtained by computer analysis of DNA histograms.
1.0'
,
20
*.i
2Lh
LO
60
80
I 100
CHANNEL NUMBER (PROPORTIONALTO CELLULAR DNA CONTENT) j122)1.41
Figure 4 . DNA histograms obtained after various incubation times (indicated) with different doses of Con A. The cell cultures were made 20 m~ in EDTA and 50 m~ in amM 3 h before harvest to disperse adherent and agglutinated cells, washed in PBS and stained with 0.1 mg/ml mithramycin in 25 % aquous ethanol. Each spectrum comprises approximately 2 X lo4 cells. The origos of the spectra have been shifted for clarity. The peaks at channel number 37 and 78 represent cells in G 1 and G 2 M, respectively, with cells in S-phase in between. The solid line represents the distribution of S-phase cells as estimated by a computer fitting of a mathematical model [18]. Cultures without Con A remained with a negligible fraction of cells outside G 1 (see Fig. 5). Counts below approximately channel 30 (C) are attributed primarily to the partly degraded nuclei of dead cells.
+
Fig, 6 and 7, respectively. These results are similar to those reported by others [20], showing a significantly suppressed [3H]dThd incorporation in cultures with the hyperoptimal concentrations of Con A. Note, however, that in cultures with 100 pg/ml Con A the [3H]dThd incorporation was relatively large at an early stage. There is an obvious discrepancy between the data on [3H]dThd incorporation on the one hand and those on cellular and nuclear volumes and cellular DNA content on the other. Whereas ['HIdThd incorporation in cultures with 100 yg/ml Con A was suppressed by a factor of 3.7 relative to that induced by the optimal Con A dose (30 pg/ml), the percentage of blast cells was about 1.7 times larger with the higher dose (Fig. 2 ) , and the number of cells in S and G2 M was significantly larger than in cultures with 30 yg/ml (Fig. 5 ) . This discrepancy
+
0
If2275.Sj
2L
48
12
96
lNCU64TlON TIME (HOURS)
Figure 6. The rate of [3H]dThd incorporation as a function of incubation time with various doses of Con A (indicated). The cultures were given a 65-min pulse of [3H]dThd (7 pCi/ml) before harvest. Each point represents the mean of three cultures.
is even more striking for cultures with 200 pg/ml which produced the highest percentage of blast cells, whereas [3H]dThd incorporation was barely above that found for unstimulated controls.
3.4 Viability
Viability was not significantly affected by hyperoptimal concentrations of Con A up to 100 pg/ml, remaining above 80% over the four-day experimental period. In cultures with 200 pg/ml, however, more than 60% of the cells were lost from the cultures during this period. A high percentage of
438
H. B. Steen and T. Lindmo
Eur. J. Immunol. 1979. 9: 434439
the lymphocyte population than does the Con A dose which is optimal with regard to [3H]dThd incorporation.
Figure 7. The total incorporation of [3H]dThdbetween 24 and 92 h of incubation as a function of the Con A concentration. The data were obtained by integration of the curves in Fig. 6.
dead cells is evident from the volume spectra in Fig. 1D and is indicated also by the large amount of debris with subnuclear size in the DNA histograms (Fig. 4 C).
3.5 Reproducibility The experiment described above has been performed five times with cells from different donors. The percentage of responding cells was found to vary considerably from one experiment to the other, i.e. from 35 to above 50% in cultures with 30 pg/ml Con A, apparently reflecting differences between the donors. Qualitatively, however, the results were always the same, showing in all experiments a significantly larger percentage of responding cells in cultures with 100 pg/ml Con A than in cultures with 30 pg/ml, which was the optimal dose with regard to [3H]dThd uptake.
3.6 Phytohemagglutinin The same experiments have also been carried out with phytohemagglutinin (PHA, P-type, Gibco) stimulation. The results were quite similar to those obtained with Con A. Thus, a PHA concentration five times that producing maximum incorporation of [3H]dThd was found to inhibit mitosis completely, whereas the percentage of cells where blast transformation, including DNA synthesis, was initiated, was larger than with the optimal PHA concentration.
4 Discussion The present experiments show that the inhibition of lymphocyte proliferation by concentrations of Con A (and PHA), which are hyperoptimal with regard to [3H]dThd incorporation, is not due to blocking of processes that trigger blastogenesis. Thus, the upper declining limb of the typical mitogen dose response curves obtained with the [3H]dThd assay does not seem to reflect a reduction in the number of responding cells as has usually been taken for granted. On the contrary, it appears that hyperoptimal doses of Con A (and PHA), which strongly suppress incorporation of [3H]dThd, initiate volume growth and DNA synthesis in a larger percentage of
This conclusion corroborates work [ll]showing that the fraction of cells responding by growth increased with the concentration of Con A even in the hyperoptimal dose range. On the basis of [3H]dThd measurements it was suggested that hyperoptimal Con A doses block blastogenesis 6 to 12 h before the G 1/S boundary. The present experiments, however, demonstrate that the blocking effect is not confined to this part of the cell cycle. On the contrary, it appears that the great majority of cells responding to hyperoptimal doses enter well into S-phase before growth stagnates, and a significant portion even reach G2. Thus, the block does not seem to be associated with any particular part of the cell cycle, but may occur anywhere between late G 1 and mitosis. The factors which determine at which point a cell is stopped are not known. It seems pertinent, however, that when the mitogen concentration was increased from the optimal to a hyperoptimal level( 124 pg/ml) at t = 24 h of incubation, i. e. when most responding cells were at the beginning of S-phase, the blocking effect was largely the same as when the hyperoptimal dose was given at t = O (unpublished results). Hyperoptimal doses had significant blocking effects when given as late as at t = 48 h. The DNA histograms (Fig. 4) demonstrate that one reason for the reduced ['HIdThd incorporation at high Con A concentrations is that DNA synthesis stagnated during S-phase in a majority of the responding cells. However, the main reason for the reduced ['HIdThd incorporation with hyperoptimal doses of mitogen appears to be that the ['HIdThd assay fails to operate in a quantitative way under such conditions. This is particularly obvious from the results obtained with the highest dose of Con A which suppressed [3H]dThd incorporation almost completely (Fig. 7), whereas according to the DNA histograms a significant proportion of the cells entered S-phase (Figs. 4 and 5). The reason for the failure of the [3H]dThd assay may either be that the membrane transport of ['HIdThd is inhibited o r that some biochemical step, viz.. phosphorylation, necessary for its incorporation into DNA, is blocked. It should be noted, in this respect, that exogenic thymidine is not a prerequisite for DNA synthesis in lymphocytes. The stagnation of DNA synthesis in cultures given hyperoptima1 doses of Con A is accompanied by a similar reduction in the growth of cellular and nuclear volume as shown by the volume spectra (Fig. 1). It has been found [14] that cellular volume is approximately proportional to protein content, which means that volume growth represents a corresponding net protein synthesis. Hence, the stagnation of volume growth reflects a corresponding stagnation of protein synthesis. It may thus appear that hyperoptimal doses of Con A, after inducing blastogenesis in a large percentage of the cell population, progressively inhibit the entire synthetic activity as these cells proceed through the cell cycle. This observation may indicate that some quite general process is affected, the most obvious possibility perhaps being membrane transport. Stimulation of lymphocytes by mitogens in optimal doses has been found to be accompanied by an increase of the membrane permeability for a variety of substances including nucleosides, sugars, amino acids, K+ and Ca++ [ 11. The increased permeability is assumed to depend on conformational changes in surface glycoproteins. It is conceivable that the extensive receptor cross-linking, which is likely to occur with hyperoptimal doses of Con A, may inhibit such conformational changes and there-
Eur. J. Immunol. 1979. 9: 4 3 4 4 3 9
by affect the uptake of essential nutrients. This explanation may apply also to the reduced uptake of [3H]dThd. On the other hand, receptor cross-linking takes effect shortly after the application of the mitogen, whereas the stagnation of D N A and protein synthesis occurs only after about two days. Hence, if the stagnation of synthesis is due to reduced membrane permeability, this reduction should occur fairly late and not as an immediate result of cross-linking and immobilization of receptors. The suppression of the [jH]dThd incorporation in lymphocyte cultures by hyperoptimal doses of Con A is reported to be accompanied by inhibition of patching and capping of receptors, including receptors that are not engaged by the mitogen [6, 8, 91. Thus, it was suggested that mobility and redistribution of receptors is a prerequisite for the initiation of blastogenic response [6]. This hypothesis has been questioned [21]. It is not supported by the present results which show that hyperoptimal doses of Con A enhance the response rather than supress it. On the contrary, it may be concluded that to the extent that the hyperoptimal Con A doses used in our experiments inhibit receptor mobility and redistribution, initiation of response does not depend on these phenomena. The observation that drugs known to interfere with microtubuli, including colchicine and vinblastine, reduce lymphocyte response to optimal doses of mitogen, whereas the response to hyperoptimal doses may be increased [22, 231, prompted the hypothesis that the mobility and redistribution of surface receptors are controlled by the microtubuli system, and that the microtubuli may serve as signal regulators for mitogenesis [ 6 ] . Hyperoptimal mitogen doses were assumed to induce a polymerized state of the microtubuli system, thereby increasing the ancorage of the receptors with a resultant reduction of their mobility and blocking of the initiation of blastogenesis. The present results indicate that such an alteration of the microtubuli system, if it occurs at all, does not play an essential role in transmitting the signal that triggers blastogenesis. This conclusion is supported by our finding [24] that colchicine and colcemid in concentrations which strongly suppressed [3H]dThd incorporation, did not significantly reduce the percentage of lymphocytes which responded to Con A stimulation by growth of cellular and nuclear volume, indicating that the initiation of blastogenesis is not much affected by these drugs. The same conclusion was reached in a recent report [25] showing that colchicine and vinblastine failed to affect events which occur early in blastogenesis including increased RNA synthe-
Blastogenic response of lymphocytes
439
sis, synthesis of lymphotoxin and increased turnover of membrane phospholipids. It seems possible, however, that the stagnation of protein and DNA synthesis seen with hyperoptimal Con A doses at the later stages of blastogenesis is associated with reduced receptor mobility and alterations in the microtubuli system. Received September 27, 1978; in final revised form December 27, 1978.
5 References 1 Resch, K., in Cuatrecasas, P. and Greaves, M. F. (Eds.), Receptors and Recognition, John Wiley and Sons, New York 1976. 1: 59. 2 Greaves, M. F., in Cuatrecasas, P. and Greaves, M. F. (Eds.), Receptors and Recognition, John Wiley and Sons, New York 1976. 1: 1. 3 Nicolson, G. L., Int. Rev. Cyt. 1974.39: 89. 4 Ringdtn, 0. and Johansson, B. G., Scand. J . Immunol. 1977. 6: 281. 5 Sela, B.-A,, Wang, J. L. and Edelman, G. M., J. Exp. Med. 1976. 143: 665. 6 Edelman, G. M., Science 1976.192: 218. 7 Ash, J. F., Louvard, D. and Singer, S. J., Proc. Nat. Acad. Sci. USA 1977. 74: 5584. 8 Yahara, I. and Edelman, G. M., Exp. Cell Res. 1973. 81: 143. 9 De Petris, S., J. Cell Biol. 1975. 65: 123. 10 Moller, G., Scand. J. Immunol. 1976. 5: 583. 11 McClain, D. A. and Edelman, G. M., J. Exp. Med. 1976. 144: 1494. 12 Boyum, A,, Scand. J . Clin. Lab. Invest. 1968. 21: 21. 13 Stewart, C. C. and Ingram, M., Blood 1967.29: 628. 14 Steen, H. B. and Lindmo, T., Cell Tissue Kinet. 1978. 11: 69. 15 Crissman, H. A. andTobey, R. A., Science 1974.184: 1297. 16 Lindmo, T. and Pettersen, E. O., Cell Tissue Kinet. 1978, in press. 17 Lindmo, T. and Steen, H. B., Biophys. J . 1977.18: 173. 18 Lindmo, T. and Aarnaes, E., J . Histochem. Cytochem. 1979. 27: 297. 19 Heier, H. E., Klepp, R., Gundersen, S., Godal, T. and Norman, T., Scand. J . Haematol. 1977. 18: 137. 20 Wang, J. L., McClain, D. A. and Edelman, G. M., Proc. Nat. Acad. Sci. U S A 1975. 72: 1917. 21 Loor, F., Prog. Allergy 1977.23: 1 . 22 Gery, I. and Eidinger, D., Cell. Immunol. 1977.30: 147. 23 Wang, J. L., Gunther, G. R. and Edelman, G. M., J . Cell Biol. 1975. 66: 128. 24 Steen, H. B. and Lindmo, T., Eur. J. Immunol. 1978.8: 667. 25 Resch, K., Bouillion, D., Gemsa, D. and Averdunk, R., Nature 1977.265: 349.
Eur. J. Immunol. 1979. 9: 434-439
H. B. Steen and T. Lindmo
Harald B. Steen and Tore Lindmo Department of Biophysics, Norsk Hydro’s Institute for Cancer Research, Montebello, Oslo
Initiation of the blastogenic response of lymphocytes by hyperoptimal concentrations of concanavalin A The blastogenic response of human lymphocytes in vitro to hyperoptimal concentrations of concanavalin A (Con A ) has been studied by means of volume spectroscopy (measuring cellular and nuclear volume), flow cytofluorometry (measuring cellular DNA content) and incorporation of [3H]thymidine ([3H]dThd). The optimal Con A dose with respect to [3H]dThd incorporation was about 30 pg/ml. In cultures given hyperoptimal doses, e.g. 100 pg/ml, [3H]dThd incorporation was strongly inhibited, whereas the number of cells entering S-phase and significantly increasing their cellular and nuclear volume was considerably larger than with 30 pg/ml. With 200 pg/ml Con A , which induced negligible rH]dThd incorporation, the percentage of responding cells was even larger. Hence, doses of Con A, which were hyperoptimal with regard to 13H]dThd incorporation, induced blastogenic response, including DNA synthesis, in a larger percentage of the cells than did the optimal dose. However, in cultures with hyperoptimal Con A doses, the progression of the cell cycle stagnated mainly during S- and G2-phase and few cells completed mitosis. Thus, the blocking effect of hyperoptimal doses was not confined to any particular point of the cell cycle. The reduced [3H]dThd incorporation, seen with hyperoptimal doses, is attributed partly to a failure of this assay under such conditions.
1 Introduction The mechanisms by which the blastogenic response of lymphocytes is triggered by antigens and mitogens binding to appropriate receptors on the plasma membrane are poorly understood. Various effects of stimulation have been reported to occur within minutes of exposure to mitogen (review in [l]). Some of these effects have been observed also in other types of cells, supporting the assumption that important features of the mechanisms of the receptor-mediated control of lymphocyte proliferation are common to cells of higher organisms in general [2, 31. It is not known, however, which of the observed effects, if indeed any of them, acts as a triggering signal, and how this signal is transferred from the receptor to the inner cell. One of the early effects of mitogen stimulation is aggregation of stimulant-binding receptors into “patches” which may subsequently merge to form a “cap”. The observation that all stimulatory agents acting through receptors are at least divalent, whereas the corresponding monovalent forms have no mitogenic effect, has been taken to indicate that receptor cross-linking is crucial for the triggering of lymphocyte activation [l]. However, recent reports that monovalent Fab fragments may cause blastogenic response [4, 51 seem to show that this is not always the case. It has been suggested [6] that some of the surface receptors are connected with the cytoplasmic system of microtubuli, possibly via actin and myosin-containing filaments [7], and that the initiation of mitogen response is caused by alterations in this system which are dependent upon redistribution of mitogen receptors. Thus, receptor mobility is considered being of [I 22761
crucial importance for the initiation of blastogenic response. An essential part of the basis for this hypothesis is the observation that hyperoptimal concentrations of concanavalin A (Con A) inhibit both patching [8] and capping [9] of receptors, while at the same time the blastogenic response, as measured by the incorporation of [3H]thymidine (13H]dThd), is considerably suppressed [6]. Furthermore, similar concentrations of the divalent succinyl-Con A, which does not inhibit patching and capping significantly, do not suppress blastogenic response either 161. However, the nature of the inhibitory effects of large mitogen concentrations is not clear. For example, it has been reported that hyperoptimal concentrations of Con A lead to extensive cell death indicating that reduced response may be a trivial effect of cytotoxicity [lo]. Others reported [ l l ] that hyperoptimal doses of Con A do not inhibit the initiation of blastogenesis, but only some later step in the process. Using flow cytometric methods, i.e. volume spectroscopy and flow cytofluorometry, which provide more detailed information on lymphocyte response than does the [3H]dThd assay, we have studied the inhibitory effects of high mitogen concentrations. It is found that in cultures given hyperoptimal concentrations of Con A, blast transformation, including DNA synthesis, is initiated in a larger proportion of the lymphocytes than in cultures with Con A concentrations which are optimal with regard to [3H]dThd incorporation. Thus, it appears that hyperoptimal doses of Con A do not inhibit the triggering of blastogenesis. But the subsequent stages of the blastogenesis are progressively suppressed and the frequency of mitosis is strongly reduced.
2 Materials and methods
Correspondence: Harald B. Steen, Department of Biophysics, Norsk Hydro’s Institute for Cancer Research, Montebello, Oslo 3, Norway
2.1 Cell culture
Abbreviations: amM: a-Methyl-D-mannoside Con A: Concanavalin A PBS: Phosphate-buffered saline PHA: Phytohemagglutinin [3H]dThd [MethylL3H]thymidine
Mononuclear leukocytes were separated from fresh, heparinized, human blood by centrifugation on a Ficoll-Triosil gradient (Lymphoprep, Nyegaard & Co., Oslo, Norway) [12],
0014-2980/79/0606-0434$02.50/0
0Verlag Chemie, GmbH, D-6940 Weinheim, 1979
Eur. J. Immunol. 1979. 9: 4 3 6 4 3 9
washed three times by centrifugation at 160 X g to remove platelets, and resuspended in medium RPMI 1640 (Gibco, Grand Island, NY), with antibiotics and 10% AB serum to a concentration of about 1.5 X lo6 cells/ml. The cells were cultured in 0.1 ml aliquots in microculture trays (Linbro Chemicals, New Haven, CT) at 37”C, 5% COz, and high humidity. Con A (Sigma, Chemical Co., St. Louis, MO), dissolved in RPMI 1640 to ten times the final concentration, was added to the cultures shortly after seeding. Agglutinated cells and cells adhering to the wells were dispersed by adding to each well 3 h before harvesting 0.1 ml of a solution containing 40 m~ EDTA (Sigma) and 100 m~ a-methyl-Dmannoside (ctmM, Sigma) in Ca and Mg-free phosphatebuffered saline (PBS, Gibco). Cell nuclei were prepared by suspending the cells in a counting solution (131 consisting of 1% formaldehyde (Merck, Darmstadt, FRG), 0.5 % glacial acetic acid (Merck) and 1 mg/ml Cetrimide (Sigma) in 0.85 % NaCI. Cells used in the three different assays were cultivated on the same tray.
Blastogenic response of lymphocytes
435
percentage of the cells continued to grow until cell division which was most frequent around day 3 (Fig. 1B). The nonresponding cells remained with constant volume, i.e. that of unstimulated controls, for the entire culture period. The average volume of cells, when they reached mitosis, was about four times that of the nonresponding cells. Hence, newly divided lymphoblasts had approximately twice the volume of resting lymphocytes. Thus, the large central peak in Fig. 1 B (96 h) is primarily due to newly divided cells. Monocytes were very scarce in cultures stimulated with the present doses of Con A, possibly due to agglutination which was not resolved by EDTA plus amM. In cultures with 100 pg/ml Con A (Fig. lC), the percentage of responding cells was considerably larger than in cultures with the optimal dose of Con A, i.e. 30 pg/ml. Furthermore, volume growth commenced earlier and proceeded at a greater rate with the higher concentration of mitogen. These effects -0 h
2.2 Volume spectroscopy
2.L
The Coulter-type volume spectrometer, which yields the distribution of cellular or nuclear volume within a cell culture, has been described elsewhere [14]. The full contents of each culture well were suspended in 10 ml Isoton I1 counting solution (Coulter Electronics Ltd., Harpenden, GB) for measurement of cell volume or in 10 ml of the counting solution described above for nuclear volume spectra.
1.6
IA)
20
1.2
08 0.L
1.2
1 .O 0.8
0.6
2.3 Flow cytofluorometry
01
For determination of cellular DNA content, the cells were stained with mithramycin (Pfizer) [15, 161, and DNA histograms were obtained by the flow cytophotometer described previously [17]. The output of this instrument was fed to a computer for calculation of the distribution of cells between the various phases of the cell cycle [18].
f? z
0.2
8
06
3
0.1
0.2
2.4 [3H]dThd assay Cell cultures were given a 65-min pulse of [3H]dThd (7 pCi/ ml; 5 Wmmol = 185 GBq/mmol), harvested by a microculture harvester (Skatron, Lierbyen, Norway) and counted in a liquid scintillation counter [19].
3 Results 3.1 Volume spectra Volume spectra, obtained after different periods of incubation with various concentrations of Con A, are shown in Fig. 1. The main peak at the lower volume represents the small, resting lymphocytes. The peak at higher volumes in the spectra of unstimulated cultures (Fig. 1A) is due to monocytes. It can be seen that the volume of the monocytes grew considerably during the first 2-3 days of culture as they matured into macrophages. In stimulated cultures, significant volume growth of the lymphocytes was observed after 24 h. In cultures with 30 pg/ml Con A, which was found to be the optimal concentration as measured by the [3H]dThd assay, a certain
0.L
0.2
CHANNELNUMBER(PROPORTIONAL
1112)6.0
TO CELL VOLUME)
Figure 1. Distributions of cellular volume in human lymphocyte cultures incubated with different doses of Con A for various periods of time (indicated). The cultures were made 20 m~ in EDTA and 50 m~ in amM 3 h before harvesting in order to disperse adherent and agglutinated cells. Each spectrum represents 3 X lo4- 5 X lo4 cells. The peak at about channel 26 represents unstimulated lymphocytes, whereas the peak at higher volumes in (A) is due to monocytes. The broad peak at about channel 60 in (B) (96 h) represents newly divided cells. The feature peaked at channel 15 in (D) is attributed to dead cells. Due to membrane defects, the cytoplasm of these cells is in electrolytic contact with the extracellular medium resulting in a reduced “electrical volume”. The assumed spectrum of nonresponding cells, i.e. a normalized spectrum of unstimulated cells, is indicated separately in (D).
436
Eur. J. Immunol. 1979. 9: 4 3 4 4 3 9
H. B. Steen and T. Lindmo
were further enhanced in cultures with 200 pg/ml (Fig. 1D). After day 1, however, the growth rate in cultures with the higher doses of Con A decreased drastically, so that after day 2 growth was negligible. The peak which is characteristic of newly divided cells does not appear in the spectra of cultures with hyperoptimal doses of Con A (Figs. 1C and D) indicating that little cell division took place. The shorter response time and the larger percentage of responding cells in cultures with hyperoptimal Con A doses appear more clearly from the data in Fig. 2, which shows the percentage of cells with enlarged volume as a function of time. It can be seen that the percentage of enlarged cells leveled out after about 30 h, indicating that growth had commenced in the majority of responding cells at this time. It appears that in cultures given the optimal dose of Con A, i.e. 30 pg/ml, approximately 35 5% of the cells responded with volume growth as compared to about 60% in cultures with 100 pg/ml. The percentage of responding cells was even higher in cultures with 200 pg/ml Con A, although an accurate estimate is complicated by extensive cell death in these cultures.
appeared in both types of spectra. The percentage of enlarged nuclei was in accordance with that of enlarged cells, showing a significant increase with mitogen dose also in the hyperoptimal dose range. An advantage of nuclear spectra is that the present method of preparing nuclei completely dispersed all agglutinated cells. Thus, microscopic investigations of nuclear suspensions showed negligible aggregation, i.e. less than 5 % of nuclei, confirming the original morphological observations of Stewart and Ingram [13]. Hence, the similarity of cellular and nuclear spectra eliminates the possibility that small aggregates of cells or debris were confused with enlarged cells to any significant extent. In comparison, up to 20% of whole cells from stimulated cultures could be found in a few larger aggregates after treatment with EDTA plus amM. Consequently, cell numbers have been based on nuclear volume spectra. Control
-0-0
30pglml con A IM)pg/mlcon A 2W&ml con A
--&-A -X-X
-0-
/A-A /A
0
I
0
lI2276.3!
INCUBATION TIME (HOURS)
Figure 2. The percentage of lymphocytes with enlarged volume as a function of time in cultures incubated with different doses of Con A (indicated). The intermediate plateau from about 30 h represents the situation where growth has commenced in the great majority of responding cells while cell division, giving rise to the secondary increase after 48 h, has not yet started. Data for cultures with 200 pg/ml Con A obtained after 24 h have been omitted because of extensive cell death which appears preferentially to exclude large cells (see Fig. 1D).
The subsequent increase in the number of enlarged cells beginning at about 48 h (Fig. 2) is due to cell division, as is evident from the increase in the total number of cells/culture beginning at this time (Fig. 3). It can be seen that this increase was much smaller in cultures with 100 pg/ml Con A than in cultures with the optimal dose. From the data in Figs. 2 and 3, we calculate that in cultures with 100 pg/ml Con A no more than 22% of the responding cells completed mitosis as compared to more than 80% in cultures with 30 pg/ml Con A. In the cultures with the highest dose of Con A no cell division could be discerned, and the number of cells with intact nuclei fell drastically indicating extensive cell death. The volume spectra of nuclear suspensions were similar to those of whole cells, except that the relative increase of nuclear volume during blastogenesis was somewhat smaller than that of cellular volume, i.e. nuclear volume increased by a factor of about 3 up to mitosis* [14]. Thus, the same spectral features
* Steen, H. B. and Nielsen, V., Scand. J. Immunol., in press.
I
I
I
26
I
I
I
12
40
I
I 96
INCUBATION TIME (HOURS)
Figure 3. The total number of intact nuclei per culture as a function of the time of incubation with various concentrations of Con A (indicated).
3.2 DNA histograms DNA histograms, recorded after various times of incubation with the different concentrations of Con A, are shown in Fig.4. The histograms obtained for cells incubated with 30 yg/ml Con A are typical of proliferating cultures, with cells evenly distributed over the entire S-phase. In contrast, the histograms of cultures with 100 yg/ml Con A show an abnormally high percentage of cells in early S, whereas fewer cells reached G 2 . In comparing these histograms, one should keep in mind that in cultures with 30 pg/ml Con A cells were lost from G 2 because of cell division, whereas in cultures with 100 pg/ml Con A little division occurred.
+
+
The distribution of cells between GO G I , S and G2 M in the various cultures is depicted in Fig. 5. It can be seen that with 100 pg/ml Con A a significantly larger number of cells entered S-phase than with the optimal concentration of 30 yg/ ml. Whereas with the optimal concentration of Con A the percentage of cells in S and G 2 M decreased after day 3 as a result of cell division, little change is seen for cultures with the higher concentration of Con A.
+
3.3 C3H]dThd incorporation The rate of DNA synthesis, as measured by the [3H]dThd assay, is shown as a function of time and Con A concentration in
Eur. J. Immunol. 1979. 9: 434-439
Blastogenic response of lymphocytes
437
100 gglml con A
Lt
200pglml con A
lI2275.5(
INCUBATION TIME (HOURS)
ICI Figure5. The percentage of cells in the various phases of the cell cycle as a function of the time of incubation with various doses of Con A (indicated). The data were obtained by computer analysis of DNA histograms.
1.0'
,
20
*.i
2Lh
LO
60
80
I 100
CHANNEL NUMBER (PROPORTIONALTO CELLULAR DNA CONTENT) j122)1.41
Figure 4 . DNA histograms obtained after various incubation times (indicated) with different doses of Con A. The cell cultures were made 20 m~ in EDTA and 50 m~ in amM 3 h before harvest to disperse adherent and agglutinated cells, washed in PBS and stained with 0.1 mg/ml mithramycin in 25 % aquous ethanol. Each spectrum comprises approximately 2 X lo4 cells. The origos of the spectra have been shifted for clarity. The peaks at channel number 37 and 78 represent cells in G 1 and G 2 M, respectively, with cells in S-phase in between. The solid line represents the distribution of S-phase cells as estimated by a computer fitting of a mathematical model [18]. Cultures without Con A remained with a negligible fraction of cells outside G 1 (see Fig. 5). Counts below approximately channel 30 (C) are attributed primarily to the partly degraded nuclei of dead cells.
+
Fig, 6 and 7, respectively. These results are similar to those reported by others [20], showing a significantly suppressed [3H]dThd incorporation in cultures with the hyperoptimal concentrations of Con A. Note, however, that in cultures with 100 pg/ml Con A the [3H]dThd incorporation was relatively large at an early stage. There is an obvious discrepancy between the data on [3H]dThd incorporation on the one hand and those on cellular and nuclear volumes and cellular DNA content on the other. Whereas ['HIdThd incorporation in cultures with 100 yg/ml Con A was suppressed by a factor of 3.7 relative to that induced by the optimal Con A dose (30 pg/ml), the percentage of blast cells was about 1.7 times larger with the higher dose (Fig. 2 ) , and the number of cells in S and G2 M was significantly larger than in cultures with 30 yg/ml (Fig. 5 ) . This discrepancy
+
0
If2275.Sj
2L
48
12
96
lNCU64TlON TIME (HOURS)
Figure 6. The rate of [3H]dThd incorporation as a function of incubation time with various doses of Con A (indicated). The cultures were given a 65-min pulse of [3H]dThd (7 pCi/ml) before harvest. Each point represents the mean of three cultures.
is even more striking for cultures with 200 pg/ml which produced the highest percentage of blast cells, whereas [3H]dThd incorporation was barely above that found for unstimulated controls.
3.4 Viability
Viability was not significantly affected by hyperoptimal concentrations of Con A up to 100 pg/ml, remaining above 80% over the four-day experimental period. In cultures with 200 pg/ml, however, more than 60% of the cells were lost from the cultures during this period. A high percentage of
438
H. B. Steen and T. Lindmo
Eur. J. Immunol. 1979. 9: 434439
the lymphocyte population than does the Con A dose which is optimal with regard to [3H]dThd incorporation.
Figure 7. The total incorporation of [3H]dThdbetween 24 and 92 h of incubation as a function of the Con A concentration. The data were obtained by integration of the curves in Fig. 6.
dead cells is evident from the volume spectra in Fig. 1D and is indicated also by the large amount of debris with subnuclear size in the DNA histograms (Fig. 4 C).
3.5 Reproducibility The experiment described above has been performed five times with cells from different donors. The percentage of responding cells was found to vary considerably from one experiment to the other, i.e. from 35 to above 50% in cultures with 30 pg/ml Con A, apparently reflecting differences between the donors. Qualitatively, however, the results were always the same, showing in all experiments a significantly larger percentage of responding cells in cultures with 100 pg/ml Con A than in cultures with 30 pg/ml, which was the optimal dose with regard to [3H]dThd uptake.
3.6 Phytohemagglutinin The same experiments have also been carried out with phytohemagglutinin (PHA, P-type, Gibco) stimulation. The results were quite similar to those obtained with Con A. Thus, a PHA concentration five times that producing maximum incorporation of [3H]dThd was found to inhibit mitosis completely, whereas the percentage of cells where blast transformation, including DNA synthesis, was initiated, was larger than with the optimal PHA concentration.
4 Discussion The present experiments show that the inhibition of lymphocyte proliferation by concentrations of Con A (and PHA), which are hyperoptimal with regard to [3H]dThd incorporation, is not due to blocking of processes that trigger blastogenesis. Thus, the upper declining limb of the typical mitogen dose response curves obtained with the [3H]dThd assay does not seem to reflect a reduction in the number of responding cells as has usually been taken for granted. On the contrary, it appears that hyperoptimal doses of Con A (and PHA), which strongly suppress incorporation of [3H]dThd, initiate volume growth and DNA synthesis in a larger percentage of
This conclusion corroborates work [ll]showing that the fraction of cells responding by growth increased with the concentration of Con A even in the hyperoptimal dose range. On the basis of [3H]dThd measurements it was suggested that hyperoptimal Con A doses block blastogenesis 6 to 12 h before the G 1/S boundary. The present experiments, however, demonstrate that the blocking effect is not confined to this part of the cell cycle. On the contrary, it appears that the great majority of cells responding to hyperoptimal doses enter well into S-phase before growth stagnates, and a significant portion even reach G2. Thus, the block does not seem to be associated with any particular part of the cell cycle, but may occur anywhere between late G 1 and mitosis. The factors which determine at which point a cell is stopped are not known. It seems pertinent, however, that when the mitogen concentration was increased from the optimal to a hyperoptimal level( 124 pg/ml) at t = 24 h of incubation, i. e. when most responding cells were at the beginning of S-phase, the blocking effect was largely the same as when the hyperoptimal dose was given at t = O (unpublished results). Hyperoptimal doses had significant blocking effects when given as late as at t = 48 h. The DNA histograms (Fig. 4) demonstrate that one reason for the reduced ['HIdThd incorporation at high Con A concentrations is that DNA synthesis stagnated during S-phase in a majority of the responding cells. However, the main reason for the reduced ['HIdThd incorporation with hyperoptimal doses of mitogen appears to be that the ['HIdThd assay fails to operate in a quantitative way under such conditions. This is particularly obvious from the results obtained with the highest dose of Con A which suppressed [3H]dThd incorporation almost completely (Fig. 7), whereas according to the DNA histograms a significant proportion of the cells entered S-phase (Figs. 4 and 5). The reason for the failure of the [3H]dThd assay may either be that the membrane transport of ['HIdThd is inhibited o r that some biochemical step, viz.. phosphorylation, necessary for its incorporation into DNA, is blocked. It should be noted, in this respect, that exogenic thymidine is not a prerequisite for DNA synthesis in lymphocytes. The stagnation of DNA synthesis in cultures given hyperoptima1 doses of Con A is accompanied by a similar reduction in the growth of cellular and nuclear volume as shown by the volume spectra (Fig. 1). It has been found [14] that cellular volume is approximately proportional to protein content, which means that volume growth represents a corresponding net protein synthesis. Hence, the stagnation of volume growth reflects a corresponding stagnation of protein synthesis. It may thus appear that hyperoptimal doses of Con A, after inducing blastogenesis in a large percentage of the cell population, progressively inhibit the entire synthetic activity as these cells proceed through the cell cycle. This observation may indicate that some quite general process is affected, the most obvious possibility perhaps being membrane transport. Stimulation of lymphocytes by mitogens in optimal doses has been found to be accompanied by an increase of the membrane permeability for a variety of substances including nucleosides, sugars, amino acids, K+ and Ca++ [ 11. The increased permeability is assumed to depend on conformational changes in surface glycoproteins. It is conceivable that the extensive receptor cross-linking, which is likely to occur with hyperoptimal doses of Con A, may inhibit such conformational changes and there-
Eur. J. Immunol. 1979. 9: 4 3 4 4 3 9
by affect the uptake of essential nutrients. This explanation may apply also to the reduced uptake of [3H]dThd. On the other hand, receptor cross-linking takes effect shortly after the application of the mitogen, whereas the stagnation of D N A and protein synthesis occurs only after about two days. Hence, if the stagnation of synthesis is due to reduced membrane permeability, this reduction should occur fairly late and not as an immediate result of cross-linking and immobilization of receptors. The suppression of the [jH]dThd incorporation in lymphocyte cultures by hyperoptimal doses of Con A is reported to be accompanied by inhibition of patching and capping of receptors, including receptors that are not engaged by the mitogen [6, 8, 91. Thus, it was suggested that mobility and redistribution of receptors is a prerequisite for the initiation of blastogenic response [6]. This hypothesis has been questioned [21]. It is not supported by the present results which show that hyperoptimal doses of Con A enhance the response rather than supress it. On the contrary, it may be concluded that to the extent that the hyperoptimal Con A doses used in our experiments inhibit receptor mobility and redistribution, initiation of response does not depend on these phenomena. The observation that drugs known to interfere with microtubuli, including colchicine and vinblastine, reduce lymphocyte response to optimal doses of mitogen, whereas the response to hyperoptimal doses may be increased [22, 231, prompted the hypothesis that the mobility and redistribution of surface receptors are controlled by the microtubuli system, and that the microtubuli may serve as signal regulators for mitogenesis [ 6 ] . Hyperoptimal mitogen doses were assumed to induce a polymerized state of the microtubuli system, thereby increasing the ancorage of the receptors with a resultant reduction of their mobility and blocking of the initiation of blastogenesis. The present results indicate that such an alteration of the microtubuli system, if it occurs at all, does not play an essential role in transmitting the signal that triggers blastogenesis. This conclusion is supported by our finding [24] that colchicine and colcemid in concentrations which strongly suppressed [3H]dThd incorporation, did not significantly reduce the percentage of lymphocytes which responded to Con A stimulation by growth of cellular and nuclear volume, indicating that the initiation of blastogenesis is not much affected by these drugs. The same conclusion was reached in a recent report [25] showing that colchicine and vinblastine failed to affect events which occur early in blastogenesis including increased RNA synthe-
Blastogenic response of lymphocytes
439
sis, synthesis of lymphotoxin and increased turnover of membrane phospholipids. It seems possible, however, that the stagnation of protein and DNA synthesis seen with hyperoptimal Con A doses at the later stages of blastogenesis is associated with reduced receptor mobility and alterations in the microtubuli system. Received September 27, 1978; in final revised form December 27, 1978.
5 References 1 Resch, K., in Cuatrecasas, P. and Greaves, M. F. (Eds.), Receptors and Recognition, John Wiley and Sons, New York 1976. 1: 59. 2 Greaves, M. F., in Cuatrecasas, P. and Greaves, M. F. (Eds.), Receptors and Recognition, John Wiley and Sons, New York 1976. 1: 1. 3 Nicolson, G. L., Int. Rev. Cyt. 1974.39: 89. 4 Ringdtn, 0. and Johansson, B. G., Scand. J . Immunol. 1977. 6: 281. 5 Sela, B.-A,, Wang, J. L. and Edelman, G. M., J. Exp. Med. 1976. 143: 665. 6 Edelman, G. M., Science 1976.192: 218. 7 Ash, J. F., Louvard, D. and Singer, S. J., Proc. Nat. Acad. Sci. USA 1977. 74: 5584. 8 Yahara, I. and Edelman, G. M., Exp. Cell Res. 1973. 81: 143. 9 De Petris, S., J. Cell Biol. 1975. 65: 123. 10 Moller, G., Scand. J. Immunol. 1976. 5: 583. 11 McClain, D. A. and Edelman, G. M., J. Exp. Med. 1976. 144: 1494. 12 Boyum, A,, Scand. J . Clin. Lab. Invest. 1968. 21: 21. 13 Stewart, C. C. and Ingram, M., Blood 1967.29: 628. 14 Steen, H. B. and Lindmo, T., Cell Tissue Kinet. 1978. 11: 69. 15 Crissman, H. A. andTobey, R. A., Science 1974.184: 1297. 16 Lindmo, T. and Pettersen, E. O., Cell Tissue Kinet. 1978, in press. 17 Lindmo, T. and Steen, H. B., Biophys. J . 1977.18: 173. 18 Lindmo, T. and Aarnaes, E., J . Histochem. Cytochem. 1979. 27: 297. 19 Heier, H. E., Klepp, R., Gundersen, S., Godal, T. and Norman, T., Scand. J . Haematol. 1977. 18: 137. 20 Wang, J. L., McClain, D. A. and Edelman, G. M., Proc. Nat. Acad. Sci. U S A 1975. 72: 1917. 21 Loor, F., Prog. Allergy 1977.23: 1 . 22 Gery, I. and Eidinger, D., Cell. Immunol. 1977.30: 147. 23 Wang, J. L., Gunther, G. R. and Edelman, G. M., J . Cell Biol. 1975. 66: 128. 24 Steen, H. B. and Lindmo, T., Eur. J. Immunol. 1978.8: 667. 25 Resch, K., Bouillion, D., Gemsa, D. and Averdunk, R., Nature 1977.265: 349.
