The effects of microgravity on T lymphocytes was studied on three sounding rockets flights with µg phases lasting 7- 13 min. Automated pre-programmed instruments could inject fluorescent-labelled concanavalin A and/or para-formaldehyde as fixative at given times. A microscope remotely controlled from the ground and connected to a CCD camera recorded cell motility and interactions under µg. Chemically fixed cells were labelled after flight with monoclonal anti-vimentin and anti- tubulin, respectively, followed by fluorescent anti-mouse Ig. First, while binding of concanavalin A to the cell membrane is not affected, patching of the receptors is slightly but significantly retarded during µg. Second, marked alterations of the intermediate filaments of vimentin as well as of the microtubule network are observed after 30 s of µg and change only slightly during the flight. Ground controls conducted at hypergravity show that the changes are not due to the acceleration peak of 13-15 g following launch. Third, cell motility and cell-cell contacts also occur when cells are free- floating in µg. These results indicate that direct effects of µg on the cytoskeleton are likely whereas little changes occur at the membrane level, and are in agreement with recent data showing that gravity controls the pattern of microtubule formation. The existence of active movements and cell-cell contacts supports, but does not prove, the notion that T cells can communicate and transmit signals in µg.
A dramatic depression of T lymphocyte activation in vitro by concanavalin A (Con A) was discovered in an experiment conducted aboard Spacelab 1 in 1983.¹ More sophisticated experiments followed during Spacelab D1 in 1985,² SLS-1 in 1991³,4 and IML-2 in 1994.5 All investigations confirmed the effect. The results indicate that the depression is mainly due to a lack of expression of the interleukin-2 receptor.4, 5 Conversely, when T lymphocytes and monocytes were attached to microcarrier beads, activation was more than doubled under µg compared with controls cultured at 1 g, either in flight or on the ground.3, 4 It was concluded that signal transduction is markedly changed in µg.
The work described in this paper was carried out to clarify aspects of such effects that can be investigated during relatively short µg periods (7-13 min) achieved on sounding rockets using either automatic pre-programmed instruments or a microscope adjusted via telecommand.
The first question addressed was whether binding of Con A to the cell membrane, followed by patching and capping of the membrane proteins involved, would take place normally in µg. It is believed that this is the first signal required for activation.6 Patching and capping are processes occurring within a few minutes after binding. Abnormalities in this process would obviously alter the transduction of the first activation signal. The hypothesis was that exposure to µg could produce changes of certain membrane properties.
The second question was related to structural changes possibly occurring within the cells exposed to µg. Gravity unloading per se is not sufficient to change the shape of a cell, because the microenvironment is subjected to forces that are many orders of magnitude larger than that of gravity.7, 8 However, organelles with a higher density than the cytoplasm (such as ribosomes, centrioles and nucleolus) may exert at 1 g some pressure on cytoskeletal and other structures. This pressure would disappear in µg. A similar effect occurs in the statocytes of the roots of some plants where gravitropism is attributed to the pressure of the statoliths on the endoplasmic reticulum. Structural changes may have an important impact on cellular events. In addition, their occurrence would demonstrate the existence of direct effects of µg on single mammalian cells.
The third question was whether free-floating cells in µg are capable of autonomous motions and, hence, of cell-cell interactions. Cell-cell interactions are an important means of cell communication and signal delivery in T lymphocyte activation. Although in previous experiments cell aggregates were always observed in samples fixed in µg,¹,² one could argue that lack or reduction of cell contacts is the cause of the loss of activity in space. Direct evidence in real time will certainly give a clear answer to the question.
We present here the results of four experiments carried out on the MASER 3, MASER 4 and MAXUS 1B sounding rockets. In MASER 3/4, the binding of fluorescent-labelled Con A (FITC-Con A) to human lymphocytes and the subsequent patching and capping was investigated. In MAXUS 1B, one experiment studied the structure of the microfilaments of vimentin and of the microtubule network in Jurkat cells (a human T cell line) chemically fixed in-flight and labelled after recovery with monoclonal antibodies. In the second experiment, motility and interactions of human lymphocytes were recorded in real time by an onboard microscope. A detailed description of the instruments aboard MASER 3/4 is given in ref. 9, while the equipment used on MAXUS is presented in ref. (6).
All cell cultures were incubated at 37°C during flight. In MASER 3, the experiment was installed in the CIS-1 (Cells in Space) module developed by Fokker Space & Systems with the National Aerospace Laboratory and the Centrum voor Constructie en Mechatronica (CCM) in The Netherlands. The experiment-specific device consisted of a unit provided by CCM. Cell cultures and reagents were stored in three contiguous sections (respectively, fixative/cells/Con A) of a flexible silicon tube closed at both ends and separated by roller bars released at pre-programmed times. See ref. 9 and Section 3 of this volume for details.
In MASER 4, a more sophisticated system, the Plunger Box Mix Unit (PBMU, also provided by CCM), was used in the CIS-2 module. The reagents and, later, the fixative were injected into the cultures by pre-programmed plungers.9
Two instruments, both integrated in the TEM 06-5M module and provided by ERNO Raumfahrttechnik, Bremen, were used in MAXUS 1B (see ref. 6 for a detailed description). To study the cytoskeleton, cell cultures and fixative were sealed in two disposable syringes separated by a valve. At pre-programmed times, the piston of the syringe with the fixative automatically injected the reagent into the culture. The observation of cell movements was performed in a glass cuvette mounted in a phase contrast microscope connected to a CCD camera. To prevent sedimentation of the cells, the cuvette was rotated at 30 rpm/min during the pre-flight and launch phases; rotation was stopped at the onset of µg. In-flight, the selection of the observation field and focusing were operated manually via telecommand from the ground station.
Each flight experiment with the exception of the observation of cell movements was accompanied by a synchronous ground control with the same batch of cells in an equivalent apparatus.
Peripheral blood was drawn from healthy donors 14-18 hr before launch. Cells were purified by gradient centrifugation on Ficoll (Pharmacia, Uppsala) and cultures were prepared as described previously.10 The preparations contained 84±3% of lymphocytes. The remaining were macrophages (13±3%) and polymorphonuclear cells.
Jurkat cells, derived from a human T cell leukemia,11 were kindly provided by A. Lanzavecchia of the Basle Institute of Immunology, Basle. The cells are mycoplasma-free. Cells were grown in RPMI 1640 medium supplemented with 10% (v/v) foetal bovine serum (Biochrom KG- Germany), 20 mM HEPES, 5 mM sodium bicarbonate, 1 mM sodium pyruvate, 4 mM L-glutamine and gentamycine (50 µg/ml).
Monoclonal antibodies anti- vimentin (V-5255) and anti-ß tubulin (T-4026), and anti- mouse polyvalent immunoglobulins FITC-conjugated (fluorescein isothiocyanate) were obtained from SIGMA (St. Louis, Missouri, USA). Con A-FITC was purchased from ICN Immuno Biologicals (Lisle, Illinois, USA).
In all investigations, the experiment-specific units containing the cell cultures and reagents were installed in the modules of the rockets 2 hr before launch. The µg phases of MASER 3 & 4 lasted 7 min, that of MAXUS 1B 12.5 min. µg in a sounding rocket flight is defined as g <10-4. The flight samples were recovered and handed over to the experimenters within 90 min of launch.
In MASER 3, lymphocyte cultures (290 µl, 2x106 cells/ml) were sealed in the middle part of each of the eight silicon tubes of the flexible tube unit. After onset of µg, FITC-Con A (280 µl, 51 µg/ml, final concentration 25 µg/ml) was added. Fixation followed by mixing with paraformaldehyde (final concentration 1%) 0.3, 30.3, 60.3, 90.3, 120.3, 180.3, 270.3 & 390 s after the addition of FITC-Con A. The protocol of the investigation in MASER 4 was analogous to that of MASER 3. In four of the six chambers, FITC-Con A was released 180 s after onset of µg; fixations followed 0.3, 30, 90 & 210 s later. In the other two chambers the mitogen was added after 300 s of µg; fixations followed 30 & 90 s later. All samples were duplicated.
In MAXUS 1B, the labelling of vimentin was carried out on Jurkat cells fixed 30, 420 & 720 s after the onset of µg by means of an automatic pre-programmed device. For tubulin labelling, the cells were fixed after 30 & 720 s. To 1 ml of culture of Jurkat cells (1.7x106 cells/ml), 1 ml of 2% paraformaldehyde was added. The observation of cell motility was made in cultures containing 3x106 lymphocytes/ml. Realtime recording was on magnetic tape in an onboard CCD camera. Analysis was performed on a papercopy (video-printout) showing pictures taken at 13 s intervals.
Evaluation of binding, patching and capping
After recovery, the cells were washed three times with Hanks balanced salt solution (PBS) and fixed on microscope slides by cytocentrifugation (200 g, 8 min). The cells were mounted in glycerol containing p-phenylendiamine as antifading and observed in a Nikon Optiphot microscope. At least three slides were prepared from each sample. About 500 cells were evaluated and counted on each slide. For statistical analysis of the data the one-tailed U-test of Wilcoxon, Mann and Whitney was applied.
Labelling of vimentin and tubulin
After recovery, the cells were rinsed with 10 ml of cytoskeleton- stabilising buffer (SB) containing 50 mM imidazole, 50 mM potassium chloride, 0.5 mM magnesium chloride, 1 mM EGTA, 1 mM 2-mercaptoethanol, 4 M glycerol, pH 6.8. The cell pellet was resuspended in 1 ml of SB containing 1% Triton X-100 and the cells were kept for 30 min at ambient temperature. The cells were rinsed again with 10 ml SB, pelleted and incubated for 30 min at ambient temperature with 200 µl of anti-vimentin or anti- tubulin, diluted 1:100 with SB. At the end of the incubation the cells were rinsed with 10 ml of SB, and the cell pellet incubated with the secondary antibody FITC-conjugated, diluted 1:100 in SB. Incubation and rinsing with SB followed as above. Finally, the cells were cytocentrifuged (200 g, 8 min). The cells, adhering to microscope slides, were mounted in glycerol containing p- phenylendiamine as antifading and observed in epifluorescence in a Nikon Optiphot microscope. Three slides were prepared from each sample and about 400 cells were evaluated on each slide. Micrographs were made on Fujichrome 400 ASA films. For statistical analysis of the data the one-tailed U-test of Wilcoxon, Mann and Whitney was applied.
Binding of the mitogen
The type and number of signals required for T lymphocyte activation have yet to be clarified and are still a matter of controversy.6 Nevertheless, there is agreement in the literature on the nature of the first signal. This is usually provided by the presentation to the T cell of the modified antigen in conjunction with the MHC component of the antigen- presenting cell, usually a monocyte. The presentation of the antigen can be replaced by a mitogen such as Con A. At least part of the effects observed in space could be attributed to alterations of the binding pattern. Therefore, it was essential to establish whether the binding of Con A to the cell membrane, followed by patching and capping of the receptors involved, occurs normally in µg.
Two experiments were conducted to answer this question; preliminary data were presented in conference proceedings.12, 13 In the first investigation (MASER 3), Con A was added to the cells immediately after µg conditions were established. Fixation with paraformaldehyde followed after different preset times of 0.3-390 s. No significant differences in the rate of binding, patching and capping compared to the 1 g control were observed.12 Binding was very fast and completed after 30 s.
The second experiment (MASER 4) was carried out to test if the lack of differences between flight and 1 g control samples arose from a lack of influence of µg or from the fact that the cells did not have sufficient exposure time to 'adapt' to µg. Therefore, Con A was injected into the samples 3 & 5 min after the onset of µg, followed by fixation after different preset times. The data obtained confirm that, even after 5 min in µg, there are no changes in binding of Con A to the cell membrane compared to the ground control.13 Furthermore, there are no differences in the patching of the Con A receptors in the samples where the cells have been exposed for 3 min to µg before the addition of the mitogen (Fig. 1). However, patching is significantly retarded when Con A is added after 5 min of µg (Fig. 1), especially in the samples where the cells were fixed 90 s after the addition of Con A. In the flight samples, 2.9±0.9% of the cells show patching, compared to 7.4±3.4% in the 1 g control (99.5% significance of the difference; U(11.6) = 5, alpha = 0.005). In the samples fixed after 30 s, the difference in patching is statistically significant at the 95% level (1.9±1.0% in the flight sample, compared to 3.9±2.6% in the ground control; U(12,7) = 19, alpha = 0.05). Furthermore, we also observed a significant difference in patching between the samples exposed for 3 or 5 min to µg and fixed after 90 s (97.5% significance level; U(6,6) = 3, alpha = 0.025). In addition, the data in Fig. 1 indicate that, despite a significantly lower rate at µg, patching is a dynamic process that would be completed if the incubation time were longer than 210 s. The question on differences in capping could not be answered with statistically significant data as the number of cells showing capping was too low in this experiment. However, there seems to be a tendency that capping is retarded in the flight samples.
Fig. 1. Patching and capping of Con A receptors following binding of the mitogen to human lymphocytes. The bars represent the percentage of cells showing either patching or capping at different fixation times after the addition of Con A. Standard errors of the means are given.
The fact that the patterns of binding, patching and capping are not changed compared to the ground control when Con A is added immediately after the onset of µg, and the fact that changes in patching - but not in binding - are observed only when Con A is added after 5 min of µg, indicate that the short exposure to hypergravity during the launch phase is not relevant.
The data show that the influence of µg on the delivery of the first activation signal is rather small and that, therefore, rapid processes such as binding of the mitogen, patching (although slightly retarded) and probably also capping are not involved in the depression of the in vitro activation of T lymphocytes observed in the Spacelab experiments.1-5
In connection with the study on cell movements in µg, we investigated the occurrence of cytoskeleton changes as it is well known that the cytoskeleton plays an important role in cell motility. In particular, we analysed the pattern of the intermediate filaments of vimentin and of microtubuli in Jurkat cells. The higher cytoplasmic volume and size of these cells, compared to resting lymphocytes, render them more suitable for this type of analysis.
The cells were fixed with paraformaldehyde 30, 420 & 720 s after the onset of µg. After recovery, the cytoskeleton was marked with monoclonal antibodies anti-vimentin and anti-tubulin. The cytofluorographs of the intermediate filaments of vimentin are shown in Fig. 2; the quantitative evaluation of the changes is presented in Fig. 3.
Fig. 2. Cytofluorographs of the intermediate filaments of vimentin of Jurkat cells. Top: ground control; Bottom: flight sample fixed after 30 s in µg.
Fig. 3. Influence of gravity changes on the structure of Vimentin in Jurkat cells. The bars represent the percentage of cells showing either the appearance of large bundles (solid bars) or the formation of protein aggregates together with discontinuities of filamentous network (cross-hatched bars). Standard errors of the means are given.
The morphological analysis reveals that cytoskeleton structural changes are occurring in µg and that they are most evident in vimentin. We must point out that some of the changes were also detected, although to a significantly smaller degree, in the control samples.
The most important difference is given by the intermediate filaments of vimentin which under µg assemble into thick bundles in 20.9±1.7% of the cells, compared to 9.9±1.3% (95% significance of the difference; U(3,3) = 0, alpha = 0.05) in 1 g. This change is observed after only 30 s of µg. The number of cells showing large bundles is slightly but significantly decreased at µg times of 420 s (17.8±0.4%, 95% significance, U(3,3) = 0, alpha = 0.05) and 720 s (18.8±1.6%, 95% significance, U(3,3) = 0, alpha = 0.05).
The other changes observed in µg consist of the formation of aggregates of protein, appearing as fluorescent points in Fig. 2 (bottom) and suggesting depolimerisation processes, as well as of discontinuities of the filamentous network (11.9±1.3% at 30 s, 12.1±2.3% at 420 s and 11.0±1.3% at 720 s, compared to 8.8±1.6% of the ground control. Also, these changes are significant (95% at 30 s and 90% at 420 & 720 s).
A significant difference in the appearance of large bundles has also been observed in the microtubuli network, where 13.2±1.2% of the cells show these changes after 30 s of µg, compared to 5.9±2.8% in the ground control (95% significance, U(3,3) = 0, alpha = 0.05). After 720 s, 11.8±1.5% of the cells show large bundles. Also, this difference is significant (95%, U(3,3) = 0, alpha = 0.05). The other changes are less evident.
During the sounding rocket's boost phase, the cells were exposed to hypergravitational forces. In a control experiment, cells loaded into flight-type syringes were exposed for 2 min to 13 g at 37°C and than fixed. Analysis of the structure of vimentin showed the appearance of large bundles in 8.9±1.2% of the cells, whereas the formation of aggregates was found in 8.0±1.1%. Thus these cells do not exhibit the effects observed in µg. The same is true for the microtubuli network where 7.55±1.2% of the cells exposed to hypergravity showed large bundles. Therefore, the observed changes in the cytoskeletal structure are due to µg and not to the g-stress caused by the launch or a combination of hypergravity and gravitational unloading. This is also supported by the fact that the number of cells displaying the changes are only slightly reduced after 720 s in µg.
The intermediate filaments of vimentin consist of a protein expressed in cells of mesenchimal origin as well as in other types of cells cultured in vitro.14 The proposed function of the intermediate filaments is to provide a structural network, with characteristics typical of each cell type. This network maintains the shape of the cell, the distribution of its organelles and, based on the connections between cytoplasmic membrane and nucleus, mediates the intracellular signal transduction.
Little is known on the function of vimentin in lymphocytes:
Studies carried out on B cells with anti-immunoglobulins
antibodies, have demonstrated that this protein is involved in
the capping of surface immunoglobulins in the region of the
Of particular interest is the observation that in T lymphocytes, the activation by Con A causes an alteration in the distribution of vimentin that appears to be correlated with the phases of the cellular cycle.16 This indicates that the role of vimentin may change during the progression of the cell through its cycle. In addition, cytoskeleton changes, beside generating the movements, may reveal the beginning of the apoptosis which was observed in micrographs of lymphocytes cultivated in an earlier experiment in space²
A recent article 17 discussed the role of gravity in pattern formation of microtubule in solutions of purified tubulin. The pattern was different depending on whether the reaction containers were in the upright or horizontal position. This phenomenon is related to the bifurcations occurring in non- linear out of equilibrium states. As discussed by the authors in a previous paper,18 the transition from 1 g to µg may result in a bifurcation point at which the biological systems drastically changes its behaviour.
The data presented here, together with the authors' previous findings in space,1-5 support the bifurcation theory and the control of pattern formation of cytoskeletal structures by gravity.
The images recorded under µg clearly show that the free-floating non-activated cells were able to display autonomous motion in random directions. Fig. 4A shows the overall displacement of 11 different cells. A detailed analysis (one image every 13 s) shows that the movements were much more complex. The cells often changed direction, moved back and forwards and sometimes crossed the same point several times (Fig. 4B). The average velocity, calculated from the displacement in the 13 s increment, was 8.4±1.2 µm/min, with a range of 0-29.4 µm/min. Also of interest is the observation that the cells in µg were not all round. They very often exhibited longitudinal forms, rotated around their axis and also showed contraction waves similar to those described in the literature for lymphocytes that move in collagen gels under 1 g conditions.19 All 11 cells in the observation field showed motion capability under µg conditions. This is in contradiction to the behaviour of lymphocytes under 1 g conditions, where mostly activated cells only or cells in the presence of a chemoattractant show this capability.20
Fig. 4. Motility and interactions of human lymphocytes in µg. A: overall movement of 11 different cells; B: detailed motion behaviour of four cells (one measuring point each 13 s). The numbers on the axes, in µm, correspond to the position of the cell in the viewing field.
The origin of the movement can be attributed to two major causes. First, changes of the cytoskeleton, in particular the formation of thick bundles of microfilaments, may induce motility of lymphocytes by contraction and elongation of the free-floating cell. Second, the Marangoni convection due to differences in the concentration of components dissolved in the medium may generate the movements of cells resuspended in µg.21, 22Concentration gradients are generated by the metabolism of the cells, which are consuming nutrients (mainly glucose and glutamine) and producing waste (such as lactate and ammonia). Recent data (Cogoli-Greuter, Dall'Aglio, Bedarida and Cogoli, unpublished results) based on laser diffraction studies clearly show that concentration gradients form in cultures of lymphocytes and Jurkat cells.
In a recent experiment performed in the NIZEMI unit during the Space Shuttle's IML-2 mission, the motility and aggregate formation of human lymphocytes during the in vitro activation with Con A was observed in real time.23 The velocity of free cells (cells not attached to an aggregate) in the presence of Con A was found to be in the same range as in the absence of the mitogen: 6.6±1.4 µm/min. Interestingly, the velocity did not change during the 3 days of incubation, whereas in the ground control, the velocity of single cells was significantly lower and decreased with increasing incubation time.
Nevertheless, although the movements and interactions observed under µg in free-floating cells suggest that cell-cell interactions and signal transmission may work as at 1 g, it is important to remark that the cytoskeleton is altered and that, therefore, signal transmission and transduction may be altered as well. Also, the aggregates formed by lymphocytes are smaller and less frequent under µg than at 1 g 2, 23
The experiments described here have contributed to clarifying certain aspects of the in vitro activation of T lymphocytes in microgravity:
This project was conducted with grants from the Italian Space Agency (ASI), the Swiss National Research Foundation (Grant No. 31-25181.88), ETH Zürich, Contraves A.G., Zürich, and ERNO Raumfahrttechnik, Bremen. The authors gratefully acknowledge the assistance of D. Mesland, W. Herfs and D. Frimout of ESA- ESTEC, Noordwijk, of R. Huijser of Fokker Space & Systems, Amsterdam, of H. van Soest and A. Koppen of CCM, Nuenen (NL), of the team of ERNO Raumfahrttechnik, Bremen, in particular D. Grothe and I. Meyer, and of the team of the Swedish Space Corporation at the ESRANGE launch facility in Kiruna, Sweden. The authors also wish to thank A. Lanzavecchia of the Basle Institute of Immunology for providing the Jurkat cells.