Fig. 1: Prototype of the Random Positioning Machine at Fokker Space and Systems in Leiden, the Netherlands.
With the advent of laboratories in Earth orbit, starting in the early seventies with the American Skylab, the condition of weightlessness has been applied for the study of physical and biological processes, expected or speculated to be influenced by weight (¹).
(¹) This text will use the term weight rather than gravity and weightlessness or microweight instead of microgravity.
Many of these expectations proved to be at variance with the results obtained in orbit and a wealth of new observations were made in a range of disciplines. In turn, these observations and their possible interpretations provided a new basis on which the effects of weight could be explored in different fields.
In general, the new basis also gave rise to the recognition of additional research potential for applying weight as a variable. Firstly, many scientific questions that followed this initial phase required only short-duration conditions of weightlessness. These conditions can be provided by relatively inexpensive droptowers (seconds of free fall), parabolic flight in aircraft (tens of seconds) and sounding rockets (minutes), instead of by expensive orbital facilities. Secondly, reflections on the possible underlying mechanisms of the observed weight effects allowed the conception of ground-based devices that would be able to correctly simulate weightlessness in the particular system studied.
However optimised an orbital facility could be made to be, for carrying out scientific experiments, if a line of research into the effects of weight depended exclusively on in-orbit experimentation, it would inevitably be too costly, and the pace of research too slow by any ground-based standard. Therefore, it was realised rather soon following the first well-controlled in- orbit experiments, that a lot of related research can be done in ground laboratories, and that devices to simulate weightlessness may be essential for the success of this research.
One such device, the Free Fall Machine (FFM), is being presented here. The machine is thought to be of particular interest for weight-related research in cell biology, but its application might be wider than that. Based on similar biological reasoning that resulted in the conception of the FFM, another machine, called the Random Positioning Machine (RPM), has been conceived, following a totally different working principle. The RPM will be briefly introduced as well.
Before Spacelab flew for the first time in 1983, those scientists who had thought about the possible effects of weightlessness on biological cells had the expectation that weight in the microworld of cells would be a negligeable force, and that therefore weightlessness would make no difference in the function or architecture of a biological cell. Preliminary results proving the contrary in the pre-Spacelab period, had been reasoned away as being the consequence of other spaceflight effects such as radiation, mechanical stress during launch and landing, or suboptimal life support conditions for the organisms flown.
Presently, we know that microweight does evoke a number of effects on cells, and a number of well-controlled experiments have shown that the absence of weight can have profound effects on fundamental biological processes active within the cell (see Table 1). The machinery that controls DNA replication and cell division, that transduces chemical signals from outside the cell to inside the cell, and the machinery that controls the dynamic architecture of the cell, all of them can be disturbed in weightlessness.
Table 1: Some Important Results with Cells (Single or Tissue) in Weightless Conditions
Effect Organism Spacecraft
--------------------------------------------------------------------------------------------
- Enhanced proliferation Paramecium Spacelab
Chlamydomonas Spacelab / Bion
E. coli Spacelab
- Enhanced conjugation / recombination E. coli / Yeast Spacelab
- Decreased receptor-mediated Lymphocytes Spacelab/Sounding Rocket
activation / differentiation Jurkat T cells Bion / Spacelab
THP 1 cells Bion / Spacelab
A431 cells Sounding Rocket
Osteogenic MN7 cells Bion
Protoplast cells Bion / Spacelab
- Decreased protein kinase C Jurkat T cells Bion / Spacelab
signalling THP 1 cells Bion / Spacelab
A431 cells Sounding Rocket
- Reorganisation Cytoskeleton A431 cells Sounding Rocket
Osteoblast cells Bion
Lentil statocytes Spacelab
Cress statocytes Spacelab / Sounding Rocket
Chara Sounding Rocket / Spacelab
In trying to understand possible mechanisms that contribute to these effects, it was helpful not to think of the cell in transition from normal weight to microweight conditions, but to think of it in transition from microweight to a normal weight condition. How can it perceive weight? Weight would somehow cause a situation inside the cell of differential mass displacement, due to differences in the mass of cell-internal organelles and supramolecular structures. Such a condition of displacement should be of a sufficient magnitude to trigger the adoption by the cell of a new internal equilibrium. This very condition could be called the 'weight signal', the energy level of which will have to be above the local thermal noise inside the cell.
The concept of weight signal, and its relation to mass displacements inside the cell, is put here in very general terms on purpose, because the detailed sequence of effects that causes the signal is not really important. What is important though, is that a minimal energy level is required to evoke any reaction in the cell. This translates into a minimal displacement of mass inside the cell and therefore to the notion that the weight vector has to have a constant direction for some minimal duration.
Plant biologists call this minimal duration the perception time, and attempts have been made to measure this time for the specialised weight-sensing cells located in the tip of plant roots, at normal ground conditions. Values found in these studies are in a range between 0.5 sec and 13 sec, dependent on the species used and the limitations of methods applied.
The important conclusion from these considerations is:
With this knowledge, simulation of microweight under normal ground conditions could be obtained in two ways, both based on the 'requirement' that the weight vector should act for at least seconds in a constant direction to generate an effect in cells. Conditions in which this requirement is not met actually prevent the cell from feeling weight at all; the weight vector escapes its detection machinery. The machines described are based on the hypothesis that sensing no weight would have similar effects as being weightless.
One approach is to provide a condition in which the weight vector never has a constant direction for seconds or longer. This condition is generated by the Random Positioning Machine (RPM). In the other approach a condition is provided in which the weight vector has a constant direction, but its duration is much shorter than seconds at any time. This condition is generated by the Free Fall Machine (FFM).
What the clinostat achieves in two- dimensional space, the RPM achieves in three-dimensional space: the weight vector experienced by the organisms of an experiment continuously changes its direction in three-dimensional space with a frequency of more than 1 Hz (Fig. 1). An experiment is
Fig. 1: Prototype of the Random Positioning Machine at Fokker
Space and Systems in Leiden, the Netherlands.
An experiment is mounted at the centre of a cardanic framework, the rotations of which are driven by two independent motors. The movement of the experiment is controlled by software providing four basic modes of operation:
In all cases, the so-called walk speed, defined as the speed of a given point on the experiment projected on a sphere of fixed diameter, can be set at a constant value. By adding selection criteria of the random positions generated, the machine can be instructed to avoid a succession of positions in the same space segment, which might distort the time-averaged microweight condition. Similarly, the machine can be instructed to spend more accumulated time in a particular space segment, resulting in average weight vectors of some pre-established values. Such a capability allows the investigator to impose onto the experiment values in between normal weight and microweight for the purpose of establishing threshold levels of the organism under study.
The RPM has been built by Fokker Space and Systems in Leiden, the Netherlands, and a pilot experiment will be performed by Dr J.P. Veldhuijzen of the Free University of Amsterdam, repeating a space-flown experiment on bone mineralisation.
The principle of the Free Fall Machine has been patented in 1987 (ESA patent 170, 19.05.87, 'Ground-based microgravity generator'). The operating principle is that of a bouncing ball: the ball is in a condition of continuous free fall, interrupted only by the bounces. During these bounces the ball experiences short periods of weight, the magnitude of which decreases with each bounce, from several G's to 1 G (normal weight), when the ball comes to rest. If there was no friction in the ball's trajectory, neither during free fall nor during the bounces, the movement of the ball would never cease. The bounce would have to be completely elastic with no dissipation of heat. Such a situation cannot exist, but a steady movement of the ball can be obtained by compensation for the loss of kinetic energy during each cycle, given by an extra push at the time of each bounce. This steady state situation can be described by
where a is the mean acceleration during the bounce, t(sub b) is the duration of the bounce, g is the acceleration of free fall (g = 9.81 ms(exp -2 ) and t(sub f) is the duration of the free fall period for each cycle.
A machine has been constructed by the Centre for Construction and Mechatronics (CCM) in Nuenen, the Netherlands, in which such a situation can be created for an experiment (Figures 2 and 3). It provides a guided free fall and a bouncing mechanism that sustains a steady bouncing condition of the experiment. The dimensions of the machine allow a maximum free fall time during each cycle to be in the order of one second. According to formula (1), the product at(sub b) in this machine will therefore be in the order of 1 gsec. With the machine, the bouncing time t(sub b) can be varied to some extent, but was left constant at a value of some 0.05 seconds, resulting in an average a = 20 g for the bouncing acceleration.
Fig. 2: The Free Fall Machine located at the BioCentrum of the
University of Amsterdam
Fig. 3: Schematic of the Free Fall Machine. The experiment is
mounted on a piston that runs along a guiding shaft. Starting
from an upper position the piston falls into a cylinder and
compresses the air (airspring). On its path downwards, a sensor
signals its position just before entering the cylinder. This
signal is used to regulate an electric valve that allows a small
volume of additional air to enter the cylinder, giving an extra
push to eject the piston again to its predetermined height along
the shaft. Then the cycle starts again and so on, as long as
required by the experiment. The maximal distance between the
cylinder edge and the piston is 1.2 m, giving a maximal free fall
time of 1 sec per cycle. Except for the implemented dimensioning
of the machine, there is no a priori limitation of the mass or
shape of the experiment.
Consequently, the condition generated by the FFM is one of continuous, real weightlessness (free fall), interrupted every second by 50 millisecond peaks of 20 G working in one direction. According to the hypothesis above, these very short peaks should go unnoticed by the cell and therefore the cell should experience a continuous microweight environment.
To test the hypothesis, a pilot experiment was performed by Prof. H. van den Ende, of the University of Amsterdam, with the unicellular green alga Chlamydomonas (Fig. 4). These cells, of about 7 µm diameter, live in fresh water and use light as their energy source. They move around by means of two so-called flagella (whip-like appendages). The cells can grow and multiply by cell division as long as there are sufficient nutrients in the medium. Cell growth and cell multiplication have been observed to be more rapid in experiments performed in space, resulting in the production of more cells after a finite time in space compared to ground controls.
Fig. 4: The unicellular green alga Chlamydomonas, a fresh-water,
photo-synthetizing organism of about 7 mmin diameter, actively
swimming by means of a pair of whip-like appendages called
flagella. If sufficient nutrients are available these cells grow
and then multiply by cell division. Before each division cycle
the flagella are resorbed, giving rise to immotile cells that
sink.
The repetition of such an experiment on the FFM yielded similar results to those obtained in space.
This pilot experiment was succeeded by a series of experiments performed in Van den Ende's laboratory by an ESA Research Fellow, Dr A.Anton. Dr Anton confirmed the initial observation and showed consistently that Chlamydomonas in the FFM grew faster than the controls, started earlier with their cell division and hence gave rise to approximately 1.25 x more cells after one day, and approximately 1.9 x more cells after two days under the experimental conditions applied (see Fig. 5).
Fig. 5: Cell multiplication of Chlamydomonas monoica in the Free
Fall Machine as compared to a 'ground' control run simultaneously
under identical conditions. Each o is an individual experiment.
Approximately 1.9 as many cells are found in the FFM after 48
hours.
This result seemed to confirm the hypothesis that at least Chlamydomonas does not feel the repitition of short 20 G 'kicks'. However, although the results were similar to those obtained in space, it could not be excluded that the effect was rather due to the kicks than to the free fall periods, and that kicks coincidentally produced the same results.
To test this possibility, it was necessary to construct a machine that would produce identical kicks, but followed by normal G conditions instead of weightlessness. Such a machine should give the normal ground- control results if the kicks had no effect.
This Centrifuge Free Fall Machine (CFFM) has been built by CCM (Figures 6 and 7). It is essentially the same as the FFM, the difference being that in addition to the bouncing movement imposed on an experiment, the experiment can be continuously rotated at the same time, at a pre-established speed. In other words, as is nowadays routinely applied in space, a centrifuge takes care of the desired G value in the free fall period. With a centrifugal speed set at zero the CFFM provides exactly the same condition as the FFM, that means weightlessness. Any value higher than zero can be selected to impose G levels between 0 and 2 G on the experiment.
Fig. 6: The Centrifuge Free Fall Machine at the Centre for
Construction and Mechatronics in Nuenen, the Netherlands, prior
to delivery to the BioCentrum of the University of Amsterdam.
Fig. 7: Schematic of the Centrifuge Free Fall Machine. The
principle is the same as described for the FFM (Fig. 3), but with
additional features. The most important feature is the centrifuge
rotation that can be imposed on the experiment simultaneously
with the bouncing movement. Levels between 0 and 2 G can be
obtained. Also the free fall time (values lower than one second),
and the reverse (bounce) time (values lower than 100
milliseconds) are made really adjustable in this machine. All
these parameters can be accurately measured. Furthermore, the
machine features electrical connections to the experiment,
allowing space-flown hardware to run on the machine without
modification. The machine is physically integrated in a
temperature-controlled cabinet.
A quick, pre-delivery test has been performed with this machine, the result of which strongly suggests that, indeed, the cells do not detect the 50 millisecond peaks of 20 G. With the centrifuge speed set at values to simulate 1 G and 0.45 G no differences were found with the ground control.
Elaborate testing, after delivery of the CFFM, has still to be performed in order to fully prove its genuine microweight- generating capability for Chlamydomonas cells. Nevertheless, these initial results seem not to invalidate the hypothesis that cells require time to sense weight and secondly, that if time is too short they react as if being in continuous weightlessness conditions.
If the principle applied in the FFM is working for cells, one might wonder where the limitations are for such machines. One could conjecture that for any physical process in the molecular world, a change in weight can only cause a change in the system's steady state if it is lasting for some minimal time. How short is that time? Can FFMs be built that work with shorter than millisecond 'kicks', followed by free fall periods a factor smaller than seconds? It seems worthwile to explore these possibilities.
On the other hand, the limitations as to the size of the biological objects (and may be for the physical object) that can be meaningfully applied on the machine, still needs to be explored.
It may be clear that the CFFM and the results obtained so far generate a variety of research and development avenues, that can all be explored on ground, but are very relevant to weightlessness research in space. It is this type of approach that will have to be developed with vigour to ascertain that the best research will be performed in space with dedicated facilities. Only then, really competitive microweight research can be developed and maintained in the in-orbit infrastructure.
ESA Microgravity News Vol. 9 No. 1