Hydrotaxis of Cyanobacteria in Desert Crusts
O. Pringault1 and F. Garcia-Pichel2
(1) Centre IRD de Nouméa, BP A5, 98 848 Nouméa Cedex, New Caledonia
(2) Department of Microbiology, Arizona State University, Tempe, AZ 85287, USA
Received: 21 October 2002 / Accepted: 4 March 2003 / Online publication: 12 November 2003
Abstract
We studied the migration of cyanobacteria in desert
crusts from Las Bárdenas Reales (Spain). The crusts were
almost exclusively colonized by the filamentous cyanobacterium Oscillatoria, which formed a dense layer approximately 600 lm thick located between 1.5 and 2.1
mm deep. Laboratory and field experiments showed that
saturation of the crust with liquid water induced a migration of the cyanobacteria leading to a significant greening of the surface within a few minutes. Under light
and rapid evaporation, the green color rapidly disappeared and the crust surface was completely devoid of
filaments within 60 min. In contrast, 260 min was required to recover the original white color of the crust
when slow evaporation was experimentally imposed. The
up and down migration following wetting and drying
occurred also in the dark. This demonstrates that light
was not a required stimulus. Addition of ATP synthesis
inhibitors prevented the cyanobacterium from migrating
down into the crust, with filaments remaining on the
surface. Therefore, the disappearance of the green color
observed during desiccation can only be attributed to an
active cyanobacterial motility response to the decrease in
the water content. The simplest explanation that can
account for the evidence gathered is the presence of a
mechanism that links, directly or indirectly, these motility responses to gradients in water content, namely a
form of hydrotaxis.
Introduction
Microbial crusts are found on arid soil surfaces
throughout the world, from polar regions to extreme hot
deserts [2, 5]. In such inhospitable environments where
Correspondence to: O. Pringault; E-mail: olivier.pringault@noumea.
ird.nc
366
higher plant vegetation is often absent, cyanobacteria
represent the dominant component of soil photosynthetic communities [6]. These topsoil formations are
initiated by the growth of cyanobacteria during episodic
events of available moisture. In most cases, they are filamentous and able to excrete exopolysaccharide. Despite
a long period of study (over 30 years), the ecological
importance of such microcommunities is still not clearly
defined [16]. However, they obviously have a strong
impact both on the biology and physics of arid soils.
Presumably, one of the most significant roles played by
biological crusts is the stabilization of surface soil and the
consequent reduction of soil erosion in areas which are
otherwise highly erodible [16]. When higher plants are
absent, phototrophic microorganisms inhabiting crusts
represent the sole source of primary production for the
ecosystem. When they are capable of fixing atmospheric
nitrogen (N2), as it is often the case, they can also contribute to increase the nitrogen content of the soil [27].
Using microsensor techniques, Garcia-Pichel and Belnap
[6] have shown that oxygen dynamics in desert crusts
were in the same order of magnitude as those observed in
microbial mats, which are known to be among the most
productive aquatic ecosystems on Earth [4]. In desert
crusts, the capability of cells to find refuge below the
surface by a migration into the crust has been reported as
an essential mode of survival against extreme conditions
and erosional abrasion that characterize the topmost soil
layers [6]. However, the mechanisms for this behavior are
not well understood.
Among the different types of microbial taxes and
motility responses observed in nature, phototaxis and
chemotaxis are the most common in microorganisms.
Photomovements of microorganisms are often observed
in benthic environments (see [8] for a review). As a
general rule, light stimulation induces a positioning of
motile photosynthetic microorganisms within the benthic structure at a depth corresponding to the light intensity required to optimize their metabolism. However,
UV radiation exposure can modify this rule by forcing
DOI: 10.1007/s00248-002-0107-3
d
Volume 47, 366–373 (2004)
d
Springer-Verlag New York, LLC 2003
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
367
CYANOBACTERIAL CRUSTS
microorganisms to migrate downward [1, 17], and consequently decrease the net production of the ecosystem
[1]. In benthic environments, microorganisms can also
migrate in response to a chemical stimulus. This has been
often observed in microbial biofilms where, for example,
the sulfide-oxidizing bacterium (Beggiatoa sp.) followed
the daily vertical fluctuations of the oxygen–sulfide
interface [10, 23]. Recently, Thar and Kuhl [28] have
shown that the purple sulfur bacterium, Marichromatium
gracile, exhibited complex migration patterns as a function of both the light regime and oxygen-sulfide gradients.
In desert crusts, migration of cyanobacteria up to the
surface can be induced by incubation under wet conditions, which leads to a consequent greening of the crust
[3, 6]. Such a phenomenon can also be observed in the
field, where the greening occurs after long rainy and
cloudy conditions. In these environments, the water
potential obviously plays a key role in the control of
growth and activity of microorganisms [3, 6, 15]. Recently, we have shown that cyanobacteria in desert soils
can track water, suggesting that cyanobacterial migration
was controlled by the water content [11]. In the present
work we present a more comprehensive view of this
phenomenon. For that purpose, the migration of cyanobacteria was monitored under different conditions of
moisture, light, and energy availability.
Materials and Methods
Experimental Procedure.
Crusts were collected in Las
Bárdenas Reales, a desert region located in Navarre (NE
Spain). In order to facilitate the experimental approach,
samples with apparently flat surfaces were selected. Crust
pieces of approximately 5 · 5 cm and 1 cm of thickness
were collected and stored in plastic boxes with soft paper
to prevent cracking and crumbling. Boxes were sealed,
transported to the laboratory, and kept dry in the dark
until analysis. They were then carefully retrieved and
fixed to the bottom of PVC dishes (6 cm diameter) using
modeler’s clay (Fig. 1). Light was provided vertically by a
halogen lamp (Schott, Wiesbaden, Germany). The
downwelling irradiance in the photosynthetically active
region (400–700 nm) was 400 lmol photons m)2 s)1.
The lamp’s internal infrared cutoff filter was removed in
order to attain sufficient irradiance in the infrared for the
control of the migration (see below). Thereafter, all operations were visually monitored by observing the samples through a binocular microscope. The various
experiments were performed in a room thermostated at
20C.
Monitoring of the Migration.
The migration of the
cyanobacteria within the crusts was followed under wet
and dry conditions by monitoring the color change of the
crust surface. A radiance microprobe [19] was placed at a
distance of ca. 5 mm from the crust surface at an angle of
45 to the vertical (Fig. 1). The radiance microprobe was
coupled to a photodiode array–based detector system
(optical spectra multichannel analyzer, Spectroscopy Instruments). This unit and the relevant optical couplings
have been previously described in detail [19]. Spectra
were recorded at time intervals of 2–20 min (as a function of light conditions), and two specific wavelengths
were selected, 675 and 750 nm. The wavelength 675 nm
corresponds to the in vivo maximum absorption of
chlorophyll a (Chl a). The wavelength 750 nm is not
absorbed significantly by cyanobacteria, and therefore
was chosen as a control for the reflectance as a function
of the moisture conditions. From those two wavelengths
the following ratio, K, was calculated:
K¼
IR;750 nm
IR;675 nm
ð1Þ
where IR represents the reflectance. The cyanobacterial
migration is inferred from the variation of K values. An
increase of K indicates that cyanobacteria were moving
up to the surface and vice versa.
The migration was first monitored under illumination during sequential wetting and drying conditions.
Distilled water was added so that the samples were watersaturated and a ca. 2 mm thin film of water covered the
crust surface. The crusts were kept water saturated for 20–
50 min to allow the appearance of the filaments on the
crust surface leading to a significant change of the color
surface. The water film was then removed by suction and
light spectra were recorded until the K ratio exhibited the
same values as those measured before wetting, indicating
the complete desiccation of the crust. A stream of air
directed toward the crust surface via a Pasteur pipette was
used to speed up water evaporation (Fig. 1).
The migration as a function of moisture conditions
was also followed in the dark. The crusts were kept under
dark conditions for wetting and drying; however, a brief
light exposure of 5 s every 10–20 min was necessary to
allow the spectra recording by the light detector. This
short light exposure was probably insufficient to stimulate a light response by the cyanobacteria, as judged by
nonquantitative, visual observations of the phenomenon
carried out in complete darkness.
In order to test the energy requirement for the migration, crust samples were wetted with distilled water
amended with inhibitors of ATP formation, carbonylcyanide m-chlorophenylhydrazone (CCCP) and oligomycin, at a final concentration of 2 lM and 0.05 mg L)1,
respectively. CCCP is an uncoupler of oxidative phosphorylation, and oligomycin inhibits the transport of
protons required for ATP synthetase [29].
368
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
CYANOBACTERIAL CRUSTS
Figure 1. Schematic drawing of the
experimental setup used to monitor
cyanobacterial migration in desert crusts.
Vertical Distribution.
The vertical distribution of
phototrophic microorganisms can be inferred from radiance microprofiles for the wavelengths corresponding
to the in vivo absorption of specific pigments, such as 675
nm for Chl a. This technique has been used to estimate
the distribution of phototrophic micro-organisms in
microbial mats [20] or other phototrophic biofilms [24].
In dry crusts, microprofiles of radiance were extremely
difficult to perform because of the strong resistance offered by the mineral particles to the advance of the
microprobe, which resulted in a breaking of the tip. In
order to prevent this, the crust was first wetted with
distilled water to facilitate the penetration of the light
probe into the crust. CCCP and oligomycin were
also added to prevent the migration of the cyanobacteria.
By adding the ATP inhibitors, we can then assume
that the position of the cyanobacteria within the wetted crust was identical as for dry conditions. Polynomia
(7th or 8th order) were fitted to the radiance profiles at
675 nm, and the first derivative was then calculated to
obtain the depth distribution of the attenuation coefficient. Details of this calculation procedure are given
elsewhere [20, 25]. From the depth distribution of the
attenuation coefficient for 675 nm, we can assess the
original biomass distribution of the cyanobacteria before
wetting.
Results
Crust Description.
Samples were collected from the
lowest elevations of the Bárdenas region, in the so-called
Bárdena Blanca, or ‘‘white Bardena,’’ were soils are finegrained, carbonate-rich, gypsiferous marl and typically
display desiccation polygons. They were almost exclusively colonized by a motile filamentous cyanobacterium
that was identified on the basis of morphology as Oscillatoria sp. (Fig. 2). 16S rRNA analyses of field samples
and cultures from this cyanobacterium demonstrated its
relatedness to filamentous oscillatorian cyanobacteria of
marine origin, which is somewhat unusual for a desert
crust organism [9]. This was probably a result of selection
due to the presence of sulfate in the soils. The surface of
the crust was devoid of Oscillatoria filaments and appeared bright white, but a soft scratch of the dry crust
revealed a dense, homogenous subsurface green layer of
Oscillatoria. In the field and during daylight, green spots
appeared on the surface within 30 min of gentle wetting
with liquid water.
Monitoring of the Migration.
In order to follow the
cyanobacterial migration, radiance reflectance spectra
were measured every 2–5 min after wetting the crust
samples and during the desiccation phase. A selection of
spectra is presented in Fig. 3. The addition of water
strongly decreased the reflectance as compared to dry
crusts. This was expected and was the result of a purely
physical phenomenon of change in refractive index between mineral particles and their surrounding medium
(i.e., wet sand appears darker than dry sand). A few
minutes after wetting, strong attenuation at the surface
was observed for the photosynthetic active radiation
(PAR, 400–700 nm) with specific signals at 620 and 675
nm corresponding to phycocyanin (a phycobilin) and
Chl a, respectively (Fig. 3A). Under wet conditions, the
reflectance for the near infrared radiation (NIR, 700–900
nm) was constant with time and consequently unaffected
by the cyanobacterial migration. After 23 min of wetting,
the film of water was removed and spectra were measured
with the same procedure as previously described. The
reflectance for PAR and the NIR gradually increased with
time, and reflectance values corresponding to the original
dry values were reached after 88 min (Fig. 3B).
At the end of the desiccation period, reflectance of
the PAR and the NIR were similar. Microscopic observation confirmed that the surface was totally devoid of
filaments, the cyanobacteria having retreated from the
surface. As a general rule, crust samples were kept water
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
CYANOBACTERIAL CRUSTS
369
Figure 2. Photomicrograph of a cultivated
isolate from Oscillatoria sp. obtained from
Las Bárdenas samples. The inset is a
neighbor-joining distance tree based on
16S rRNA gene sequences showing the
phylogenetic relationship of this isolate
(in bold) to its most closely related
cyanobacterium (simplified from [9]).
Its closest known relative is a marine
Oscillatoria isolate. Scale represents 5%
sequence divergence.
saturated for 20–50 min. An interesting observation was
that if wet conditions were maintained for a long time,
and a dense Oscillatoria spp. biofilm developed on the
crust surface, the retreat toward the crust interior caused
by desiccation could not be effectively triggered, with
most of the population being trapped, dry, right at the
surface. Therefore, to maintain the potential of the filaments to migrate down during desiccation, short wetting
periods (i.e., <50 min) were experimentally imposed.
Migration of the cyanobacteria was monitored by
calculating the reflectance ratio between 750 and 675 nm
(see Methods) for different conditions of light and desiccation. Results are presented in Fig. 4. The migration
down into the crust was faster when the rate of desiccation was improved by gentle air flow above the crust.
Under these conditions, the crust surface recovered its
initial white color within 60 min (Fig. 4A). In contrast, in
the absence of air flow, 260 min required (Fig. 4B) to
obtain the same result. The migration up and down also
occurred in the dark (Fig. 4C). During wetting and
drying, the kinetics of K were comparable to those observed under light conditions.
Migration and Inhibitors.
To demonstrate that the
color change observed during desiccation was exclusively
due to a cyanobacterial response and not to a possible
dragging of filaments by surface tension of a retreating
desiccation front, migration was monitored with addition
of inhibitors for ATP synthesis (oligomycin) and electron
transport uncouplers (CCCP) (Fig. 5). For each experiment, a control migratory cycle was first performed in the
absence of inhibitors; the inhibitors were then added as
soon as the surface recovered its initial white color (K =
1) (Fig. 5A,B,C). When CCCP was used as sole inhibitor
(at t = 215 min), migration toward the surface was halted
compared to the control (Fig. 5A), but some residual
downward migratory activity could be measured. At the
end of the desiccation period (330 min), the K ratio was
greater than 1 (1.12), indicating that some cyanobacteria
had remained on the surface, which was confirmed by
microscopic observation. Oligomycin was also effective in
slowing the upward migration during wetting (Fig. 5B).
Oligomycin was added at t = 180 min, and a few minutes
after the inhibitor addition, the increase in K stopped and
a plateau at 1.65 was measured (Fig. 5B). Under drying
conditions, K slowly decreased and then stabilized at a
value of 1.42 (i.e., 42% more than the control), indicating
that there was also some residual downward migration.
Optical observations revealed that Oscillatoria filaments
were present on the crust surface. The simultaneous
presence of both inhibitors stopped the upward migration completely during the wetting, and the K ratio re-
370
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
CYANOBACTERIAL CRUSTS
Figure 3. Reflectance variations from the crust surface induced by
the wetting (A) and desiccation (B) processes. The numbers on the
graphs indicate the time (min) after water addition (A) and water
removal (B). Chl a indicates the spectral signal for the chlorophyll
a absorption. In both panels, the value at 100% represent the
spectra under dry conditions before wetting (A) and after water
removal (B).
mained constant with time under the desiccation phase
(Fig. 5C), indicating that virtually no downward migratory activity could be detected either.
Vertical Distribution.
The concomitant addition of
CCCP and oligomycin during wetting showed that cyanobacterial migration was completely stopped (see
above). We used this to visualize the depth distribution
of cyanobacteria within the crust. Radiance profiles were
measured in crust wetted with distilled water amended
with CCCP and oligomycin. In this way, profiling with a
radiance microprobe was possible and cyanobacteria
were kept at a position presumably the same as in the
original dry crust. Radiance profiles were measured at
675 nm, corresponding to the in vivo maximum absorption of Chl a (Fig. 6). The radiance at 675 nm slowly
decreased in the first millimeters, and strong attenuation
was observed from 1.5 to 2.1 mm (Fig. 6A). Radiance at 2
mm represented 0.04% of the surface value. Maximal
attenuation (almost 14 mm)1) was detected at 1.8 mm
depth (Fig. 6B). From the attenuation profile, the cyanobacteria biofilm must have been localized between 1.5
and 2.1 mm depth with a maximum of biomass at 1.8
mm.
Figure 4. Reflectance ratio of the crust between 750 and 675 nm
under different moisture and light conditions. The gray panels
indicate the period of time where the crusts were water saturated.
(A) The crust sample was permanently exposed to light (400 lmol
photons m)2 s)1) and desiccation was stimulated by directing an
air stream at the crust surface. (B) Same as A, but the air stream
was not provided. (C) Same as A, but the crust was kept under
dark conditions.
Discussion
Cyanobacterial migration
was deduced by the color variation of the crust surface
measured with a light microprobe coupled to a spectrophotometer. This technique was originally applied to
study the migration of M. chthonoplastes in microbial
mats as a function of UV exposure [1]. However, in this
work, migration was monitored by the reflectance variation with time of a single wavelength (577 nm) corresponding to the in vivo maximum absorption of
phycoerythrocyanin, a characteristic pigment of M. chthonoplastes. In our study, we were confronted with reflectance variations caused by change of water content
within the crust. For a given wavelength, reflectance from
dry particles is higher than from wet particles [18].
Therefore, it was necessary to obtain a neutral signal of
Note on the Methodology.
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
CYANOBACTERIAL CRUSTS
371
Figure 6. (A) Profile of the reflectance at 675 nm in wetted crust.
The water was amended with specific inhibitors for the ATP synthesis (see Methods for more details) to avoid cyanobacterial migration. (B) Profile of attenuation coefficient for 675 nm calculated
from the reflectance profile presented in (A).
Figure 5. Effect of inhibitors of the ATP synthesis on the
reflectance ratio of the crust between 750 and 675 nm. The gray
panels indicate the period of time where the crusts were water
saturated. The addition of inhibitors is marked with arrows.
the reflectance fluctuations induced by the changes in the
water content of the crusts. This neutral signal was provided by the reflectance measured in the NIR at 750 nm,
a wavelength not absorbed by cyanobacteria. During wet
periods the reflectance for the NIR was constant with
time (Fig. 3A). Therefore, the color change of the crust
surface measured by the reflectance ratio between 750
and 675 nm could be used to follow the migration of the
Oscillatoria filaments as a function of the moisture conditions.
Evidence for Migration.
In their natural habitats,
dry crusts from Las Bárdenas Reales have a white color. A
short period of rain is enough to induce, in a few minutes, the appearance of green spots at the surface [11].
During evaporation, the green color progressively disappears and the surface recovers its white color. The time
required for this recovery is strongly dependent on desiccation conditions. We also observed alternating color
changes on crust surfaces after wet and dry exposure in
crusts sampled in the Colorado Plateau (USA). In those
crusts, the greening of the surface subsequent to wetting
occurred only after few hours and was less pronounced
compared to the greening observed for the Las Bárdenas
crusts. Similar observations in others desert crusts have
been also reported [3, 6]. From both field and laboratory
observations, evidence for cyanobacterial migration
caused by change of water content is clearly shown. The
role of the water potential on growth and activity of
cyanobacteria in desert crusts has been demonstrated
previously [3, 6]. Therefore, a control of migration by
water potential could also be postulated [11]. However,
other parameter should be considered as potential stimuli
before assessing the role of the water potential on migration in desert crusts.
Control of the Migration.
Evidence for photomovements has been previously demonstrated in microbial communities under visible, NIR, or UV exposure
[1, 10, 20]. In our study, cyanobacterial migration was
observed in the light, but also in the dark during wetting
and drying phases (Fig. 4C). To follow this migration
under dark conditions, we used short light exposures to
allow measurement by the light probe. These short light
exposures were assumed to be insufficient to stimulate a
light response by the cyanobacteria, which was confirmed
by microscopic observation of the cyanobacterial migration in crusts maintained permanently in the dark
under successive wetting and drying conditions. In the
dark, wetting the crust was followed by a significant
greening of the surface in 50 min, and then, under
evaporation, the cyanobacteria migrated down into the
crust 150 min after the removal of the water film in the
372
absence of any light cues. This clearly indicates that light
was not required to induce migration of the cyanobacteria in response to the change in the water content.
Obviously, exposure to very strong irradiance may interfere and inhibit the migration up to the surface subsequent to the wetting, and probably the ecologically
significant phenomenon involves the interplay of both
stimuli.
Preliminary experiments had shown that cyanobacteria were not able to leave the surface during the drying
process when glutaraldehyde was added concomitantly
with the water (data not shown). Using inhibitors of ATP
formation, we showed that migration was also affected.
CCCP dissipates the electrochemical gradient across
membranes and, in cyanobacteria, uncouples not only
oxidative phosphorylation but also photophosphorylation. ATP can only be then produced by substrate-level
phosphorylation. Similarly, oligomycin inhibits the
translocation of protons for ATP synthetase, and all energy-generating processes but substrate-level phosphorylation should be hindered. Using either of these
inhibitors affected the ability for migration in our cyanobacteria, but residual activity could still be detected.
The concurrent addition of both inhibitors was much
more effective, completely abolishing the migratory capacity of the cyanobacteria. This clearly demonstrates
that energy generation beyond what little could be provided by substrate-level phosphorylation was necessary to
sustain the migration, and by extension, that the migration was not caused merely by a physical phenomenon. In
conclusion, the color change observed during wetting
and drying is due to an active cyanobacterial response to
fluctuations of the water potential for which no light cues
are necessary.
Ecological Significance of Migration.
In desert
crusts of Las Bárdenas, Oscillatoria sp. formed a dense
layer between 1.5 and 2.1 mm depth (Fig. 6). This distance from the surface may be small, but it is certainly
environmentally very significant. UV radiation at 2 mm
depth is attenuated more than two orders of magnitude,
and, if the soil is stabilized, there is no abrasion by windblown dust. In this regard, by following the desiccation
gradient, cyanobacteria are migrating into a refuge habitat. Wetting and exposure at 400 lmol photons m)2 s)1
induced a greening of the surface. Those conditions of
light and moisture are similar to those observed in the
field where rain events are concomitant with low irradiance levels. In arid regions, because of the lack of water,
cyanobacterially dominated biological desert crusts remain biologically inactive most of the time [6]. However,
when water is available through precipitation, active biological processes such as photosynthesis and respiration
can start within a few minutes [6]. Similarly, we showed
that migration to the surface also occurred within a few
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
CYANOBACTERIAL CRUSTS
minutes after water addition. Under low irradiance exposure (i.e., less than 2000 lmol photons m)2 s)1), the
migration up to the surface during the wetting could be
interpreted as a biological response to the light climate.
Oscillatoria would position itself at a depth where photosynthesis can be optimal, such as close to or at the
surface. Similar behavior has been observed in microbial
mats with migrating cyanobacteria [17]. Light-harvesting
pigments such as phycobilin and chlorophyll are very
sensitive to strong irradiances. We showed that migration
to the surface also occurred after wetting under dark
conditions. The ecological interpretation of such behavior remains difficult because a photoresponse to the light
climate cannot be invoked. However, this suggests a close
coupling between water content and cyanobacterial
movements. Similarly, migration of benthic diatoms in
tidal sediment is strongly influenced by the tide level [13,
26], but the causes of the rhythm are not yet clearly
understood [13]. The upward migration in the presence
of water could simply be a relaxation of a tightly bound
population that initiates a random walk; thus part of it
reaches the surface in the absence of stimulus.
The migration down into the crust induced by
evaporation might be explained as a preventative biological mechanism to protect cells against solar UV and
visible radiation. This Oscillatoria has no photoprotective
pigments such as scytonemin or MAAs (mycosporinelike amino acids), which are used by some cyanobacteria
as ultraviolet sunscreens [7, 12]. Therefore, being trapped
at the surface in a desiccated state may represent the
demise of the population. By contrast, cyanobacteria
belonging to the genera Scytonema and Nostoc, also inhabiting crusts, are typically surface dwellers and nonmotile, but they have a high scytonemin and MAA
content [7].
The Role of Vertical Gradients.
Under incipient
desiccation, a gradient of water potential with depth can
be expected to form, with low water content on the
surface and high values deeper in the crust. A certain
vertical component is expected in the orientation of the
Oscillatoria filaments in order to be able to sense these
differences in water potential. Oscillatoria excretes exopolysaccharides (EPS) as it glides, and it forms filaments
which can reach several millimeters of length. Those EPS,
which are essential for the crust cohesion by trapping
mineral particles [21], can act as guides to help the vertically directed gliding of the cyanobacteria [14, 22]. This
would explain the fast responses encountered. Therefore,
under desiccation, cyanobacteria would be able to get
information about the moisture conditions and would be
attracted toward the area where the water content is
higher, i.e., deeper in the crust. This hypothesis also helps
explain the observation that, under sufficiently long
moisture exposure, many cyanobacteria got trapped at
O. PRINGAULT, F. GARCIA-PICHEL: HYDROTAXIS
IN
373
CYANOBACTERIAL CRUSTS
the crust surface during desiccation. This is because many
of the filaments had completely reached the surface and
were gliding about horizontally on it, which would have
prevented an efficient sensing of a gradient that was orthogonal to their gliding axis.
Concluding Remarks
The migration of motile cyanobacteria in desert crusts
seems essential for they survival in order to escape desiccation at the surface [11]. After examining the likely
factors which might regulate this migration, water potential remains as the most likely parameter controlling
cyanobacterial downward taxis in desert soil crusts. By
analogy to other motility stimuli, we can call this
hydrotaxis. In fact, it might be based physiologically on a
‘‘hydrophobic response,’’ again by analogy to well-known
photophobic cyanobacterial responses [6]. However, the
physiological mechanisms of such migration mode are
still obscure, and further investigations must be carried
out to determine them.
Acknowledgments
We thank G. Holst and B. Grunwald for their help with
the light acquisition and stimulating discussions. JeanPascal Torréton and Ben Moreton and two anonymous
reviewers are thanked for their critical comments on the
earlier version of the manuscript.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
References
1. Bebout, B, Garcia-Pichel, F (1995) UV B-induced vertical migrations of cyanobacteria in a microbial mat. Appl Environ Microbiol
61: 4215–4222
2. Belnap, J, Langue, OL (2001) Biological Soil Crusts: Structure,
Function and Management. Springer-Verlag, Heidelberg
3. Brock, TD (1975) Effect of water potential on a Microcoleus (Cyanophyceae) from a desert crust. J Phycol 11: 316–320
4. Canfield, DE, Des Marais, DJ (1993) Biogeochemical cycles of
carbon, sulfur, and free oxygen in microbial mat. Geochim Cosmochim Acta 57: 3971–3984
5. Garcia-Pichel, F (2002) Desert environments: biological soil crust.
In: Britton, G (Ed.) Encyclopedia of Environmental Microbiology.
Wiley, San Diego, pp 1019–1023
6. Garcia-Pichel, F, Belnap, J (1996) Microenvironments and microscale productivity of cyanobacterial desert crusts. J Phycol 32: 774–
782
7. Garcia-Pichel, F, Castenholz, RW (1991) Characterization and
biological implications of scytonemin, a cyanobacterial sheath
pigment. J Phycol 27: 395–409
8. Garcia-Pichel, F, Castenholz, RW (2001) Photomovements of
microorganisms in benthic and soil microenvironments. In: Häder,
D-P, Lebert, M (Eds.) Photomovements. Elsevier, Amsterdam, pp
200–215
9. Garcia-Pichel, F, Lópes-Córtes, A, Nübel, U (2001) Phylogenetic
and morphological diversity of cyanobacteria in soil desert crusts
21.
22.
23.
24.
25.
26.
27.
28.
29.
from the Colorado Plateau. Appl Environ Microbiol 67: 1902–
1910
Garcia-Pichel, F, Mechling, M, Castenholz, RW (1994) Diel migrations of microorganisms within a benthic, hypersaline mat
community. Appl Environ Microbiol 60: 1500–1511
Garcia-Pichel, F, Pringault, O (2001) Cyanobacteria track water in
desert soils. Nature 413: 380–381
Garcia-Pichel, F, Wingard, CE, Castenholz, RW (1993) Evidence
regarding the UV-sunscreen role of a mycosporine-like compound
in the cyanobacteria Gloeocapsa sp. Appl Environ Microbiol 59:
170–176
Guarini, JM, Blanchard, GF, Gros, P, Gouleau, D, Bacher, C (2000)
Dynamic model of the short-term variability of microphytobenthic
biomass on temperate intertidal mudflats. Mar Ecol Prog Ser 195:
291–303
Hoiczyk, E (2000) Gliding motility in cyanobacteria: observations
and possible explanations. Arch Microbiol 174: 11–17
Jeffries, DL, Link, SO, Klopatek, JM (1993) CO2 fluxes of cryptogamic crusts. 2. Response to dehydratation. New Phytol 125: 391–
396
Johansen, JR (1993) Cryptogamic crust of semiarid and arid lands
of north America. J Phycol 29: 140–147
Kruschel, C, Castenholz, RW (1998) The effects of solar UV and
visible irradiance on the vertical movements of cyanobacteria in
microbial mats of hypersaline waters. FEMS Microbiol Ecol 27: 53–
72
Kühl, M, Jørgensen, BB (1994) The light field of microbenthic
communities: radiance distribution and microscale optics of sandy
coastal sediments. Limnol Oceanogr 39: 1368–1398
Kühl, M, Jørgensen, BB (1992) Spectral light measurements in
microbenthic phototrophic communities with a fiber-optic
microprobe coupled to a sensitive diode array detector. Limnol
Oceanogr 37: 1813–1823
Kühl, M, Lassen, C, Jørgensen, BB (1994) Optical properties of
microbial mats: Light measurements with fiber-optic microprobes.
In: Stal, L, Caumette, P (Eds.) Microbial Mats: Structure, Development and Environmental Significance. NATO ASI Series G, vol
35. Springer-Verlag, Berlin, pp 149–166
Mazor, G, Kidron, G, Vonshak, A, Abeliovich, A (1996) The role of
cyanobacterial exopolysaccharides in structuring desert microbial
crusts. FEMS Microbiol Ecol 21: 121–130
Mc Bride, M (2001) Bacterial gliding motility: multiple mechanisms for cell movement over surfaces. Ann Rev Microbiol 55: 49–
75
Nelson, DC, Jørgensen, BB, Revsbech, NP (1986) Growth pattern
and yield of a chemoautotrophic Beggiatoa sp. in oxygen-sulfide
microgradients. Appl Environ Microbiol 52: 225–233
Pringault, O, De Wit, R, Kühl, M (1999) A microsensor study of
the interaction between purple sulfur and green sulfur bacteria in
experimental benthic gradients. Microb Ecol 37: 173–184
Pringault, O, Kühl, M, De Wit, R, Caumette, P (1998) Growth of
green sulphur bacteria in experimental benthic oxygen, sulphide,
pH and light gradients. Microbiology 144: 1051–1061
Serôdio, J, Da Silva, J, Ctarino, F (1997) Non-destructive tracing of
migratory rhythms of intertidal benthic microalgae using in vivo
chlorophyll a fluorescence. J Phycol 33: 542–553
Snyder, JM, Wullstein, LH (1973) The role of desert cryptogams in
nitrogen fixation. Am Mid Nat 90: 257–265
Thar, R, Kuhl, M (2001) Motility of Marichromatium gracile in
response to light, oxygen, and sulfide. Appl Environ Microbiol 67:
5410–5419
Zollner, H (1989) Handbook of Enzyme Inhibitors. VCH, Verlagsgesellschaft MBH, Weinheim, Germany