Characterization of the spreading process in human T lymphocytes

Experimental Cell Research 149 (1983) 201-213

Characterization of the Spreading Human T Lymphocytes

Process in

P. OTTESKOG and K.-G. SUNDQVIST Department of Periodontology and Department of Clinical Immunology, Karolinska Institutet, and Huddinge University Hospital, S-14104 Huddinge, Sweden

Substratum-bound concanavalin A (conA) caused attachment and spreading of human T lymphocytes identified by monoclonal anti-T cell antibodies and sheep erythrocyte rosette formation. The simultaneous presence of conA in the medium increased the spreading, whereas preincubation of the cells with conA inhibited spreading. The tendency of conA to induce spreading was dependent on the concentration used, the higher the conA concentration the more pronounced was the spreading. For example, conA at 10 ug ml-’ triggered the formation of prominent substratum-attached filopodia with a length of l-10 urn in 6040% of T-enriched lymphocytes obtained from separate individuals. At the same conA dose the filopodia were, in 10-20% of the lymphocytes, accompanied by development of lamellipodia. With conA at 100 ug ml-’ the number of cells that underwent pronounced spreading was 5540% in separate individuals. Observation of T-enriched cells fixed at different times after initiation of spreading induced by conA at 100 ug ml-’ indicated that tilopodia formation represented the initial morphological alteration during the spreading process. This process thereafter proceeded with development of lamellipodia, extensive cytoplasmic spreading and flattening of the central cell mass. Quiescent and mitogenactivated cells exhibited the same sequence of changes during spreading. Spreading led to disappearance of the microvilli with a length of 0.1-0.7 urn present on lymphocytes in suspension, although some microvilli persisted over the cell center.

Interaction with the environment is a central function of the lymphocyte. This cell circulates outside the lymphoid tissues and moves to specific areas within the peripheral lymphoid organs. Antigenic and inflammatory stimuli modify the migration pathway of lymphocytes [I]. Understanding of problems related to the homing of lymphocytes most likely requires basic knowledge of the components and mechanisms involved in cell-cell and cell-substratum interactions. In nonlymphoid cells there exists extensive literature on cell attachment to various adhesive surfaces as a model for cell-ceil and cell-substratum interactions in multicellular systems [2-8]. There are a number of reports which describe the interaction of lymphocytes with surfaces coated with ligands that bind to surface receptors on the lymphocyte [!&13]. One of them shows that antigen-antibody complexes induce cytoplasmic spreading in 5-15 % of human blood mononuclear cells probably via Fc-receptors [12]. We have shown that concanavalin A (conA) and phytohemagglutinin (PHA) can trigger spreading of a major portion of normally non-adherent human mononuclear cells [14]. The study of spreading capacity-a contact-induced active cellular process-may be a useful experimental approach for analysis of the interaction between lymphocytes and their Copyright 0 1983 by Academic Press, Inc. AU rights of reproduction in any form reserved 0014-4827/83 $03.00

202 Otteskog and Sundqvist environment. The present experiments were performed in order to study the details of the spreading process in quiescent and blast-transformed thymusderived lymphocytes (T cells). The results obtained further define the role of substratum and conA for the induction of spreading and demonstrate that the spreading process in human T lymphocytes consists of a sequence of dissociable membrane events. MATERIAL

AND METHODS

Cells Lymphocytes were obtained from peripheral blood of healthy human donors. The mononuclear cells were isolated by sedimentation in gelatin followed by iron powder treatment [15]. T-lymphocyteenriched cells were obtained by using sheep erythrocyte rosette sedimentation (E-rosetting) [15]. These cells comprising 87-% % sheep erythrocyte rosette-forming cells were used in the experiments performed to study spreading. Erythrocytes were lysed by addition of 0.83 % NH&l. The cells were cultivated (1.5~ 106/culture) in RPM1 1640medium with 10% fetal calf serum (FCS) in the absence or presence of conA 25 ug ml-’ for 3 days in Falcon 1008 Petri dishes. The number of spontaneously adherent cells was generally less than 1%.

Spreading In order to induce spreading T-lymphocyte-enriched cells were allowed to settle on glass coverslips in RPM1 medium containing 10% FCS and conA 100 ug ml-‘. Alternatively the glass coverslips were incubated with conA 100 ug ml-’ for 60 min at 37°C and subsequently washed in PBS. The cells were then induced to spread in medium without conA. Before spreading the mitogen activated T-enriched cells were incubated with alpha-methyl-o-mannoside (Sigma Chemicals) at a final concentration of 0.1 M in RPM1 medium for 2 h at 37°C. The cells were then washed in medium without alpha-methyl-omannoside.

E-rosetting by Substratum-Attached Cells Lymphocytes spread on conA-coated coverslips were washed in medium without conA and thereafter covered with 2.5 % (V/V) sheep red cells in medium containing 25 % FCS. After incubation at room teperature for 1 h and incubation at 4°C over night or for 4 h at room temperature, the cells remaining adherent on the glass were washed carefully in PBS and the percentage of E-rosetting cells were determined in a Leitz Diavert microscope.

Light Microscopy An inverted phase contrast microscope (Leitz Diavert) was used for observations of T cells in contact with glass coverslips in Falcon Petri dishes (1008) in the absence or presence of conA. Nonspread cells were phase bright and spread cells appeared phase dark.

Immunojluorescence (IF) One million T-enriched cells were reacted with 50 ul of monoclonal mouse anti-human T-cell antibodies (OKT3 from Ortho Pharmaceuticals, Raritan, N.J.) for 30 min at 4°C. The cells were then induced to spread in the presence of conA for 60 min at 37°C and fixed in 4% paraformaldehyde for 10 min at 4°C. Thereafter the cells were reacted with fluorescein isothiocyanate (FITC)-conjugated sheep anti-mouse immunoglobulin (SBL, 111 610, F/P molar ratio 3.5, protein concentration 4.8 mg ml-‘) diluted %o. The morphology and IF staining of the cells were examined simultaneously in a Leitz Orthoplan fluorescence microscope with interference contrast equipment and a Ploemopak containing filters for FITC. Photographic recordings were made on Kodak ‘B-i-X and Echtachrome 400 films. Exp Cell Res 149 (1983)

Spreading process in human T lymphocytes

203

Fig. 1. Surface morphology of T cells in suspension Bar, 2 urn.

Scanning Electron

Microscopy

(SEM)

One million T-enriched cells suspended in 0.3 ml of RPM1 medium were fixed by addition of an excess of 2.5% glutaraldehyde (in PBS). These cells were referred to as lymphocytes in suspension. After incubation in glutaraldehyde for at least 60 min at room temperature the cells were attached to glass coverslips preincubated with an aqueous solution of poly-1-lysine (Sigma Chemicals) at a concentration of 1 mg ml-r. Cells incubated on glass in the absence and presence of conA were fixed in situ after removal of the medium by addition of an excess of 2.5% glutaraldehyde. The fixed lymphocyte were rinsed in distilled water and dehydrated in a series of acetone dilutions followed by incubation in absolute acetone and critical point-drying. The preparations were covered with gold (70 A) using a Polaron gold sputter and then examined in a Philips S501 scanning electron microscope (grid voltage 15 kV).

Terminology Short cylindrical and linger-like extensions of the lymphocyte surface of uniform diameter and length 0.2-l urn have been termed microvilli. Broad-based extensions with a length exceeding 1 urn were termed lilopodia. Flattened lamellar projections were termed lamellipodia. Broad-based undulating lamellar extensions have been termed ruffles.

RESULTS Morphology of T-enriched Lymphocytes after Contact with a Glass Surface

in Suspension and

The mean diameter of T-enriched lymphocytes fixed in suspension was 4.6 pm (4-5.1 pm) and M-97% of the cells (variation in seven individuals), carried microvilli with a length of 0.1-0.7 pm (fig. 1). The surface of the rest of the cells Exp Cell Res 149 (1983)

204 Otteskog and Sundqvist

Fig. 2. Morphology of T cells after incubation on glass in serum-containing medium for 60 min at 37°C. (a) Close view of the area of cell-substratum contact. Notice the attachment to substratum of several villouslike projections. (b) Development of filopodia. Bar, 1 w.

(5-16%) carried few or no microvilli but showed r&Yes or ridges. Cells allowed to settle on a glass surface for 60 min at 37°C and then fixed in situ were generally round and carried microvilli like cells in suspension. Examination of the cell surface in close contact with the substratum revealed villous-like substratumExp Cell Res 149 (1983)

Spreading process in human T lymphocytes

Fig.

205

3. Sheep red cell rosette formation by spread T cell. Bar, 2 pm.

attached projections with a length of 1-2 urn in 63-89 % of the cells obtained from separate individuals (fig. 2 a). More prominent cylindrical projections (filopodia) with a length of 1-5 urn were developed in 25-65% of the cells obtained from separate individuals (fig. 2b). Such projections were not seen on cells fixed in suspension. Substratum-bound

ConA Induces Spreading of T Lymphocytes

In order to further elucidate the conA-induced spreading T-enriched lymphocytes were allowed to settle onto a glass surface under the following conditions: (1) The cells were allowed to settle on glass precoated with conA 100 ug ml-‘. (2) The cells, preincubated for 30 min with conA 100 ug ml-‘, were allowed to settle on non-coated glass in medium without conA. (3) The cells were preincubated with conA as in (2) but were allowed to settle on glass precoated with conA. (4) The cells were allowed to settle on non-coated glass in medium containing conA 100 ug ml-‘. After incubation for 60 min under these conditions the number of spread cells in a representative experiment was (1) 68 %; (2) 3 %; (3) 15%; (4) 76%. This indicated that conA bound to the substratum was responsible for the spreading. The spread cells were characterized by E-rosette formation and the monoExp Cell Res 149 (1983)

clonal anti-T-cell antibody OKT3. Unfractionated T-cell-enriched and T-celldepleted mononuclear cells were induced to spread and then incubated with sheep red cells. The spread cells formed rosettes (fig. 3) and the number of rosettes obtained was 75, 93 and 7% respectively. In addition, 77, 92 and 10% respectively of the spread cells carried the T3 antigen (fig. 4). The formation of Erosettes taken together with the fact that they carried T3 antigen indicated that the spread cells were T lymphocytes. Concentration of ConA Determines the Number of Spread Cells and the Degree of Spreading T-enriched lymphocytes were incubated on glass coverslips in the presence of conA 10, 25, 50, 75 or 100 ug ml -’ for 60 min at 37°C. Examination of the cells using phase contrast microscopy showed that the higher the conA concentration, the more cells exhibited spreading (fig. 5). For example, conA at 100 yg ml-’ induced spreading of 55-92% of T-enriched cells from 7 individuals, whereas conA 10 ug ml-’ induced spreading of 5-20% of the cells from the same individuals. The degree of spreading of T-enriched lymphocytes induced by different conA doses was compared by means of SEM after incubation of the cells for 60 min. The figures represent the variation in four separate individuals. With conA at 10 f-usml-‘, 67-94% of the cells showed 1-13 substratum-attached tilopodia/cell with a length of I-10 urn. The tilopodia were thus increased in number and size compared to those developed by lymphocytes settled on glass in the absence of conA. The tilopodia were often branched (fig. 6a). In addition to filopodia conA 10 ug ml-’ caused lamellipodia formation in approx. 10% of the cells from different individuals. The morphology of lymphocytes induced to spread by conA at 25 ug ml-’ was virtually indistinguishable from that of cells spread by conA at 10 ug ml-‘. A fraction of the filopodia-bearing cells (approx. 10% in different individuals) carried the filopodia on the terminal end of the uropod. The number of cells with lamellipodia was g-32%. ConA at 50 ug ml-’ induced lamellar cytoplasmic spreading in 34-65 % of the cells (fig. 6 c) and at conA at 75 and 100 Exp CeNRes 149 (1983)

Spreading process in human T lymphocytes 100

207

Fig. 5. Spreading of T cells at different doses of conA. T-enriched lymphocytes were added to Falcon 1008 dishes (1 x Id/dish) containing glass coverslips. The percentage of spread cells (phase dark) was recorded in an inverted phase contrast microscope (Leitz Diavert) by counting at least two hundred cells in randomly selected fields.

i

10

25

50

7'5

i-44ml-‘, 30-81% and 61-83 % respectively of the cells exhibited a pronounced lamellar cytoplasmic spreading and nuclear flattening. On cells showing pronounced lamellar spreading microvilli disappeared, although some microvilli persisted over the central cell mass containing the nucleus (fig. 66). Kinetics

of Spreading

of Quiescent and Mitogen-activated

Cells

The morphology of “quiescent” (unstimulated, O-3 days) and mitogen-stimulated T-lymphocyte-enriched cells (conA, 25 ug ml-‘, 3 days) induced to spread by conA at 100 ug ml -’ for 2, 5, 15 and 60 min was compared. Four main morphology types were distinguished: (I) Spherical non-spread cells indistinguishable from cells in suspension and 4.3 urn (3.9-5.7 urn) in diameter (fig. 7a). These cells were representative for cultures not stimulated by mitogen. The corresponding mitogen-stimulated cells were 6.3 urn (4.1-9.2 urn) and exhibited a more villous surface (fig. 7b). (II) Spherical villous cells with substratum-attached tilopodia. In non-stimulated cultures these lymphocytes showed l-7 filopodiakell with a length of l-5 ym (fig. 7 c), whereas the corresponding mitogen-stimulated lymphocytes showed l-9 filopodiakell with a length of 2-8 urn (tig. 74. (III) Lymphocytes showing lamellar cytoplasmic spreading from the cell base and absence of microvilli in the flattened areas. Unstimulated cells often showed ruffles on the peripheral lamellae (fig. 7 e), whereas the mitogen-stimulated cells showed numerous peripheral filopodia (fig. 7f). (IV) Lymphocytes showing lamellar cytoplasmic spreading and nuclear flattening (fig. 7g). In addition, approx. 50% of the cells in mitogen-stimulated cultures showed prominent peripheral projections (fig. 7 h and ref. [16]). Fig. 8 shows the number of cells of each morphology type (I-IV) after 2, 5, 15 and 60 min. It appears that there was a time-dependent transition from type I to type II, III and IV. This indicates that the morphological variations I-IV represent different stages of the spreading process which is common for mitogenstimulated and quiescent T cells. In addition, the results summarized in fig. 8 14-838338

Exp Cell Res 149 (1983)

208 Otteskog and Sundqvist

Fig, 6. Morphology of T cells incubated for 60 min at 37°C in serum-containing medium supplemented

with different doses of conA. (a) 10 ug ml-‘, notice formation of prominent filopodia; (b) 25 ug ml-‘, notice formation of tilopodia and lamellipodia; (c) 50 ug ml-‘, observe pronounced lamellar spreading; (d) 100 ug ml-‘, notice the extensive lamellar spreading and flattening of the central cell mass. Bar, 2 urn. Exp Cell Res 149 (1983)

Spreading process in human T lymphocytes

209

Exp Cell Res 149 (1983)

210 Otteskog and Sundqvist

Fig. 7. Sequence of morphological changes during spreading of (a, c, e, g) quiescent; (b, d, f, h)

mitogen-activated (conA 25 pg ml-‘, 3 days) cells. (I) Suspension morphology; (II) Initial contact and formation of tilopodia; (III) Lamellar cytoplasmic spreading; (IV) Flattening of the central cell mass. Bar, 1 km.

points to the possibility that mitogen-stimulated T cells have acquired a capacity to spread at a higher rate than quiescent T cells. However, this requires further inspection. DISCUSSION The present results indicate that the spreading process in human T lymphocytes, identified as E-rosette-forming and T3-antigen-bearing cells, is initiated by filopodia formation accompanied by lamellar cytoplasmic spreading and flattening of the central cell mass. The sequence of events during the spreading process appeared to be the same at all conA concentrations (10, 25, 50, 75 and 100 ug Exp CeNRes 149 (1983)

Spreading process in human T lymphocytes

211

IV

ml-‘), although the degree of spreading was dependent on the concentration of conA used. The spreading process in fibroblasts has previously been demonstrated to consist of a similar sequence of changes [17-191. The rate of spreading in fibroblasts, even in the presence of conA, is much slower than in lymphocytes

cw. Is spreading a “normal” physiological state of the lymphocyte? This question merits inspection because lymphocytes in the absence of appropriate ligands are non-adherent or weakly adherent cells. Furthermore, in order to observe the lymphocyte surface and avoid the production of artefacts it is considered to be necessary to fix the cells in suspension [21]. However, under certain conditions the lymphocyte shows a morphology reminiscent of the spreading characterized here. For example, Sanderson & Glauert, studying ultrastructural features of the interaction between lymphocytes and target cells, showed that a fraction of the lymphocytes accommodated to the shape of the target cells [22]. The developExp Cell Res 149 (1983)

212 Otteskog and Sundqvist I

%

II

100

50

2

5

15

60

2

5

15

60

2

5

Time of ConA incubation

15

60

2

5

15

60

(min)

Fig. 8. Kinetics of conA-induced spreading of O-O,

quiescent; @-@, activated T cells (conA 25 pg ml-‘, 3 days). The cells were incubated with conA on glass coverslips for 2, 5, 14 and 60 min and the percentage of cells showing morphology types I, II, III or IV (see fig. 7) were determined after each incubation period.

ment of fibrous projections on lymphocytes during rosetting with erythrocytes [23] may correspond to the initial phase of a spreading process as described here. Following activation by mitogens [24] or allogenic stimuli [25-261, l&20% of the cells exhibit a polarized asymmetric morphology (uropod formation). It appears that uropods often are formed in relation to contact with other cells or substratum. In the present study uropod-forming cells were noticed preferentially under “non-optimal” spreading conditions. It is conceivable that spreading is an important lymphocyte function which may determine the capacity of these cells to interact with components of the extracellular matrix and possibly with other cells. Furthermore, spreading is probably a prerequisite for lymphocyte migration. Thus, the active extension of filopodia and lamellipodia, which characterize spreading of lymphocytes, are considered to be required for cell migration in general [27]. ConA attached to substratum appeared to be responsible for the induction of spreading. In contrast, binding of conA to the cells inhibited a subsequent spreading process by substratum-bound conA. It is worth noting in this context that preincubation of effector cells with conA does not induce T-cell-mediated cytolysis, whereas conA on target cells enhances cytolysis [28]. Filopodia formation constituted the initial event of the spreading process. Although the filopodia present on substratum-attached cells are distinct from the microvilli on cells in suspension it is likely that the filopodia represent a contactinduced extension of microvilli (e.g., see fig. 2a). The extent of filopodia formation may explain the differences in spreading rate between quiescent and mitoExp Cell Res 149 (1983)

Spreading process in human T lymphocytes

213

gen-activated cells. Structures with a low radius of curvature, such as microvilli and filopodia, overcome electrostratic repulsive forses which may counteract adhesion and spreading [29]. The present investigation has been concerned with spreading of T lymphocytes. It should be emphasized that conA also induces spreading of non-T lymphocytes [14]. There is some indication that the relationship between conA concentration and the degree of spreading may be different in T and B lympocytes. We are currently studying this possible difference and the migration of lymphocytes on substratum in relation to spreading by means of time-lapse cinematography. This work was supported by the Swedish MRC project 16X-04774 and Karolinska Institutet. Scanning electron microscopy was performed at the Research Center, Huddinge Hospital. We thank Eva Bergdahl for excellent help with the cells and Susan Otteskog for skilful preparation of the manuscript.

REFERENCES 1. 2. 3. 4. 5. 6. 7.

de Sousa, M, Lymphocyte circulation. John Wiley & Sons, New York (1981). Gulp, L, Curr top membr transp 2 (1978) 327. Culp, L A, Murray, B A & Rollins, B J, J supramol struct 2 (1979) 401. Frazier, W & Glaser, L, Ann rev biochem 48 (1979) 491. Grinnell, F, Int rev cytol.53 (1978) 65. Marchase, R B, Vosbeck, K & Roth, S, Biochim biophys acta 457 (1976) 385. Moscona, A A, Neuronal recognition (ed S H Barondes) pp 205-226. Plenum Press, New York (1976). 8. Yamada, K & Olden, K, Nature 275 (1978) 179. 9. Rutishauser, U, Yahara, I t Edelman, G, Proc natl acad sci US 71 (1974) 1149. 10. Wong, S Y, Longenecker, B M, Pazderka, F & Ruth, R F, Exp cell res 92 (1975) 428. 11. Chess, L, MacDermot, R P & Chlossman, S F, J immunol 113 (1974) 1113. 12. Alexander, E & Henkart, P J, J exp med 142 (1976) 329. 13. Choi, T K, Sleight, D R & Nisonoff, A, J exp med 139 (1974) 761. 14. Sundqvist, K-G, Otteskog, P, Wanger, L, Thorstensson, R & Utter, G, Exp cell res 130 (1980) 327. 15. Natvig, J B, Perlmann, P & Wigzell, H, Stand j immunol, suppl. 5 (1976). 16. Otteskog, P, Friskopp, J 62 Sundqvist, K-G, Exp cell res 137 (1982) 111. 17. Rajaraman, R, Rounds, D E, Yen, S P S & Rembaum, A, Exp cell res 88 (1974) 327. 18. Witkowski, J A & Brighton, W D, Exp cell res 68 (1971) 372. 19. Taylor, A C, Exp cell res 8 (1961) 154. 20. Hughes, R C, Pena, D J, Clark, J & Dourmashkin, R R, Exp cell res 121 (1979) 307. 21. Newell, D G, Roath, S & Smith, J L, Brit j haematol 32 (1976) 309. 22. Sanderson, C J & Glauert, A M, Proc roy sot London, ser. B 192 (1977) 315. 23. Jerrells, T R & Hinrichs, D J, Res reticuloendothelial sot 17 (1975) 84. 24. Biberfeld, P & Perlmann, P, Exp cell res 62 (1970) 433. 25. Biberfeld, P, Biberfeld, G, Perlmann, P & Holm, G, Cell immunol 7 (1973) 60. 26. McFarland, W & Heilman, D H, Nature 205 (1%5) 887. 27. Bessis, M & de Boisfleury, A, Blood cells 2 (1976) 365. 28. Berke, G, Hu, V, McVey, E & Clark, W R, J immunol 127 (1981) 776. 29. Pethica, B A, Exp cell res, suppl. 8 (1961) 123. Received February 16, 1983 Revised version received June 23, 1983 Printed

in Sweden

E.xp Cell Res I49 (1983)