INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The mechanism of plant gravity sensing is widely hypothesized to be plastid based (Sack, 1997; Kiss, 2000; Tasaka et al., 2001; Chen et al., 2002). In this view, the mass that acts in sensing during gravitropism is intracellular. This mass is presumed to be that of amyloplasts that sediment in specific locations such as in the root cap or stem endodermis (Blancaflor et al., 1998; Weise et al., 2000; Yoder et al., 2001; Morita et al., 2002). Sedimenting amyloplasts may also act in sensing in apical cells of the moss Ceratodon purpureus (Sack et al., 1998, 2001; Kuznetsov et al., 1999). Although the mechanism is not known, amyloplasts may pull or compress an intracellular receptor (Fig. 1A).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Diagram showing two hypotheses for gravitropic sensing in a horizontal apical cell of the moss C. purpureus. The thick lines show the outer limits of the relevant masses. A, Intracellular mass shown as amyloplasts that sediment (gray fill), e.g. which might compress (arrows) a receptor. n, Nucleus. B, Entire cell (cell membrane shown as thick line) may compress and/or pull on (arrows) a receptor located outside the cell membrane. According to this model, a pressure differential between the top and the bottom of the cell can be detected even in the presence of high turgor (Staves et al., 1997a). The gravity vector is toward the bottom of the figure.

An alternative hypothesis is that pressure exerted by the entire cell triggers gravity sensing (Fig. 1B; Wayne and Staves, 1996; Staves, 1997; Hemmersbach et al., 1999). Support for this mechanism derives primarily from studies of algae and protists for phenomena other than gravitropism. The gravity vector influences the polarity of cytoplasmic streaming in large green algal cells (Wayne et al., 1990). This polarity reverses when upright cells are immersed in media denser than the cell, leading to the hypothesis that sensing is whole-cell based and that the receptors are located at the interface between the cell membrane and wall (Staves et al., 1997a). A similar model has been proposed for the directed motility of unicellular protists (gravitaxis) because media denser than the cell reverse the direction of movement (Hemmersbach et al., 1999).

The mass of the entire cell has also been hypothesized to function in gravitropic sensing (Wayne and Staves, 1996; Staves, 1997; Chen et al., 2002). This possibility is supported by a reduction in the rate of gravitropic curvature of rice (Oryza sativa) roots exposed to high-density media (Staves et al., 1997b). However, it is not known whether the density of the media used in this study was greater than the density of the presumptive sensing cells in the root cap.

The effects of high-density media on the gravitropism of single cells have not been reported. Compared with cells in tissues, single-cell systems should allow better media penetration and a more direct determination of cell density. Moss protonemata, the early haploid phase of the life cycle, extend by tip growth of the apical cell. Gravity orients moss tip growth when cultures are maintained in darkness (Schwuchow et al., 2002).

The density of C. purpureus apical cells has been determined previously (Schwuchow et al., 2000). Here, we analyze the effects of high-density media on moss gravitropism. If sensing were whole-cell based, then media denser than the apical cell should cause downward gravitropism. We report that robust upward curvature occurs in media denser than the cell, suggesting that an intracellular based mechanism operates in moss gravitropic sensing.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Cell Wall Porosity

For a high-density medium to affect the pressure exerted by a cell, the medium must penetrate the cell wall and contact the cell membrane. Thus, we determined the size of apoplastic channels in the cell wall using methods previously applied to higher plants (Baron-Epel et al., 1988; Titel et al., 1997). We used molecules of different sizes with a fluorescent tag, fluorescein isothiocyanate (FITC), including FITC-dextrans and FITC-bovine serum albumin (BSA). Because apical cell walls are very thin (Schwuchow et al., 2000), wall penetration was determined by following dye entry into detergent-permeabilized apical cells.

Figure 2 shows the relative fluorescence intensity from cells treated for 18 h. Control cells immersed in water displayed weak cytoplasmic and plastidic autofluorescence. Most conjugates tested produced fluorescence that was much brighter than the control after 1 h and these levels increased after 18 h of immersion. The degree of fluorescence correlated roughly with molecular size (molecular mass and diameters shown in Table I). For example, the smallest molecule tested, 10-kD dextran-FITC (4.6 nm in diameter), resulted in medium to high fluorescence in 72% of all apical cells after 1 h, and 88% after 18 h. This category showed the highest proportion of very bright cells of all FITC-tagged molecules tested. FITC-BSA (about 7.2 nm in diameter) did not penetrate the wall as well as smaller compounds, but one-half of all cells showed above-background fluorescence after 18 h. The largest molecule tested that penetrated the cell wall was 40-kD dextran-FITC, which is about 9.0 nm in diameter; here, 48% (1 h) and 63% (18 h) of cells showed fluorescence brighter than controls, but fluorescence was not as bright as in the 10-kD dextran-FITC treatment. Cells treated with the 150-kD dextran-FITC exhibited only autofluorescence, suggesting that this large conjugate (17 nm) was excluded from the cell and did not penetrate the cell wall. This shows that BSA and dextrans up to about 40 kD can penetrate the wall, although both penetrate more slowly and to a lesser extent than the 10-kD dextran. Thus, some apical cell wall pores are at least 9 nm in diameter (Table I), a size similar to that reported for other plant taxa (Baron-Epel et al., 1988; Shedletzky et al., 1992; Read and Bacic, 1996). Our results are also consistent with the above reports showing that various size classes of channels exist at different frequencies.

View larger version (26K): [in this window] [in a new window]   Figure 2.   Estimation of diameter of cell wall pores reported by the entry of fluorescent compounds of different sizes into apical cells treated for 18 h. Low fluorescence is equivalent to autofluorescence in control cells. Brighter fluorescence was present in the majority of treated cells except with 150-kD FITC-dextran. n = 200 cells for each treatment.

Penetration of Iodixanol

The above data establish that BSA can penetrate the moss cell wall. We also estimated the rate at which a smaller molecule, iodixanol, crosses the wall. Two stereoisomers of iodixanol have been detected by x-ray spectrometry (Tonnessen et al., 1995). Each isomer is ovoid with a long and a short molecular axis. The lengths of the short and long axes were determined to be 1.3 and 1.9 nm for isomer one, and 1.4 and 2.1 nm for isomer two. These diameters are much smaller than that of the 10-kD dextran-FITC (4.6 nm), which was shown above to pass freely through the apical cell wall.

Next, we estimated how long it takes for iodixanol to diffuse through the wall and equilibrate at the cell membrane. Because the cell wall matrix is highly hydrated (Carpita and McCann, 2000), water-soluble molecules smaller than wall pores should diffuse rapidly. Iodixanol is water soluble and uncharged. Thus, iodixanol should encounter little resistance to free diffusion through the wall, and values for the diffusion time should reasonably estimate the time needed to reach equilibrium between the outer and inner sides of the cell wall.

According to Einstein, Smoluchowski, and Stokes (Kuhn and Försterling, 2000), the time needed for the diffusion of a particle or molecule in solution can be calculated as t = 3   x²/kT, where x² is the mean square of particle displacement,  is the viscosity of the medium, r is the particle radius; k is the Boltzmann constant (1.38 · 10²³ JK¹), and T is the absolute temperature. At 20°C, a solution of 60% (w/v) iodixanol has a viscosity of 10.5 mPa s (Accurate Chemical and Scientific Corporation, Westbury, NY), which is about 10 times higher than the viscosity of water at the same temperature. Assuming a 250-nm path length for diffusion (mean cell wall thickness in C. purpureus apical cells; Schwuchow et al., 2000) and a maximal molecular radius of 1.05 nm, the diffusion time was calculated to be 1.6 ms. Because this calculation was based on maximal values for particle radius and viscosity (solutions more dilute than 60% [w/v] were also used), actual diffusion times are probably shorter.

Assuming the very unlikely case that the molecule must diffuse over a distance of 10 µm and that the cell wall impedes its free diffusion by a factor of 100, then the diffusion time of the iodixanol molecule would be about 4.3 min. This is still much shorter than the pre-incubation times used in the experiments (1, 4, and 16 h). Thus, it is very likely that the iodixanol will have traversed the cell wall and established concentration equilibrium between the cell membrane and the bulk solution during the pre-incubation period before gravistimulation (see "Materials and Methods").

Iodixanol does not appear to be deleterious to cells. It is used in density gradients to isolate viable cells and is employed as a radiological contrasting agent in humans (Homo sapiens; McCann and Chantler, 2000; Flinck and Gottfridsson, 2001). Based upon size (Table I), it is unlikely that iodixanol penetrates through the cell membrane. In moss cells, amyloplast sedimentation occurred and even appeared enhanced in horizontal cells that were immersed in 40% to 60% (w/v) iodixanol (data not shown). This suggests that there was no dramatic increase in the density of the cytosol. However, it cannot be ruled out that some iodixanol entered the cell, perhaps through fluid phase endocytosis (Kjeken et al., 2001).

Gravitropism

Negative gravitropic curvature occurred in all concentrations of BSA and iodixanol that were tested. Iodixanol had little effect on gravitropism in cells horizontal for 4 h (Fig. 3A). After 8 h in iodixanol, curvature was inhibited 16% to 32%. BSA inhibited gravitropism more than iodixanol, especially at higher concentrations and after 8 h of exposure (Fig. 3B). Figure 4 shows examples of gravitropism in control and treated cells. Upward curvature also was observed by time-lapse videomicroscopy after treatment with 40% to 60% (w/v) iodixanol and with 10% to 50% (w/v) BSA. Pretreatment with iodixanol (10%-60% [w/v]) and BSA (up to 40% [w/v]) for 1 and 4 h before gravistimulation had little effect on curvature after turning to the horizontal (data not shown). Although a 16-h pretreatment with 40% to 60% (w/v) iodixanol cut curvature in half compared with a 1-h pretreatment, cells still curved upward after the long pretreatment.

View larger version (19K): [in this window] [in a new window]   Figure 3.   Gravitropism in media of different concentrations. Upright cells were pretreated for 1 h with the media shown, rotated to the horizontal, and fixed after 4 or 8 h. A, Iodixanol. B, BSA. Controls (0%) were immersed in water. Each point represents the mean degrees of upward curvature (±SE) from 50 apical cells.

View larger version (73K): [in this window] [in a new window]   Figure 4.   Light micrographs showing gravitropic curvature of apical cells in different media. A, Control. Upright cell in water for 1 h, then reoriented to the horizontal for 4 h. B, As in A, except in 60% (w/v) iodixanol throughout. C, Horizontal for 8 h in 50% (w/v) BSA, starch stained with IK2I. The gravity vector (arrow) is toward the bottom of all micrographs. Arrowheads show regions of plastid sedimentation. Scale bars = 20 µm.

To explore the basis for reduced gravitropic curvature at higher media concentrations, we quantified the effects of concentration on growth rate (Fig. 5). Concentrations of 50% to 60% (w/v) inhibited growth considerably. The osmolality of solutions that inhibited growth was up to 264 mmol kg¹ (Table II). This level is close to that of a 4.9% (w/v) solution of mannitol (277 mmol kg¹), which induces incipient plasmolysis. Thus, it is likely that reduced curvature in high media concentrations results from a reduction in turgor and growth rate. Other possible indirect effects might also operate, including altered metabolism and ion uptake (Galtung et al., 2002).

View larger version (25K): [in this window] [in a new window]   Figure 5.   Effects of high-density media on cell growth. Percent of the control (water) growth rate. Means + SE. n = 35 (control), 102 (BSA), and 71 (iodixanol) living cells monitored for 5 h using time-lapse videomicroscopy.

Because the sensing of cell mass might be greater when gravity acts along the cell length than width, we tested the effects of BSA and iodixanol on gravitropism in upright and inverted cells (Fig. 6). This test is relevant because the polarity of streaming in Chara sp. internodal cells is influenced by gravity only when the cell is in a vertical, but not in a horizontal, position (Wayne et al., 1990). Control cells maintained in an upright position continued to grow straight up (right column in Fig. 6, A and B). Upright cells in high-density media diverged slightly from the vertical, but net upward gravitropism was maintained. When upright cells grown on normal media were inverted, they subsequently curved upward 19.2° ± 1.0° (mean ± SE) after 8 h (left column in Fig. 6, A and B). Upright cells exposed to high-density media and then inverted curved upward 17% to 43% less than control cells. Together, these data show that upward gravitropism is not substantially affected by the length of the pretreatment period, the duration of gravistimulation, the angle of gravistimulation, or the concentration of the medium.

View larger version (22K): [in this window] [in a new window]   Figure 6.   The effects of high-density media on gravitropism in vertical cells. Dishes were filled with media and then either inverted (INV) or left upright (UP) for 8 h. A, BSA. B, Iodixanol. Cultures in B were pretreated with iodixanol for 1 h. Means + SE. n = 50 to 114 cells for each treatment.

Using several methods, we previously estimated the density of the apical cell to be between 1.004 and 1.085 g cm³ (Schwuchow et al., 2000). These densities are equivalent to BSA solutions of 1% to 23% (w/v) and iodixanol solutions of 1% to 16% (w/v). Therefore strong, gravitropism took place in media that were presumably denser than the cell, i.e. in 20% to 60% (w/v) iodixanol and 30% to 50% (w/v) BSA.

Our results with C. purpureus do not conflict with the conclusions of Wayne and Staves regarding gravity sensing in characean cells (Staves, 1997; Staves et al., 1997a). The effect of gravity on cytoplasmic streaming is an entirely different phenomenon than the effect of gravity on differential growth, i.e. on gravitropism. Plants and algae respond to gravity in many ways, and it is likely that different types of gravity sensing have evolved (Sack, 1991, 1997). Chara sp. itself may employ several types of sensing. Gravity sensing in internodal cell streaming seems to be whole-cell based, whereas protonemal and rhizoid gravitropic sensing apparently relies upon the mass of intracellular, sedimenting statoliths (barium sulfate-containing vesicles; Braun, 2001).

However, our data do argue against the view that sedimenting amyloplasts merely provide ballast that contributes to whole-cell based gravitropic sensing (Wayne et al., 1990). The finding of gravitropism under conditions where the cell is effectively lighter than the surrounding medium rules out the possibility that the cell mass acts in gravity sensing in moss protonemata. Instead, these data establish that the mass that acts in sensing is not influenced by differences between the densities of the cell and the medium and, therefore, is intracellular.

In moss protonemata, the likeliest candidate for that intracellular mass is that of amyloplasts that sediment (Sack et al., 1998; Kern et al., 2001). Several lines of evidence support this hypothesis. First, sedimentation occurs before upward curvature in all moss species investigated (Walker and Sack, 1990; Chaban et al., 1998; Schwuchow et al., 2002). Second, all moss genera with gravitropic protonemata have a conserved subapical amyloplast sedimentation zone (Schwuchow et al., 1995, 2002). Third, the recovery of gravitropism after the basipetal centrifugation of C. purpureus protonemata correlates with the return and sedimentation of amyloplasts (Walker and Sack, 1991). These centrifugation data also argue against the idea that the mass that acts in sensing is that of the whole cell because after basipetal centrifugation, gravitropism is temporarily inactivated, but neither the growth nor the mass of the cell is affected. Finally, the application of a high-gradient magnetic field specifically displaces starch and induces amyloplast movement in the correct subapical zone. This displacement is, in turn, associated with differential growth in a direction that correctly mimics gravitropism in both wild-type moss as well as in the wrong-way response mutant (Kuznetsov et al., 1999). Collectively, these data support the view that amyloplast sedimentation provides spatial information that orients tip growth in moss protonemata and reinforce the idea of plastid-based gravitropic sensing in plants.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Growth and Immersion of Cultures

Wild-type cultures (strain WT3) of Ceratodon purpureus (Hedw.) Brid. were vegetatively propagated as in Schwuchow and Sack (1994). For experiments, protonemal filaments were transferred from stock cultures into small petri dishes (35 mm in diameter) using sterile technique and fine forceps. A piece of canning cellophane was positioned nearby so that the filaments would grow onto the cellophane by negative gravitropism. The dishes were sealed, wrapped in opaque material, and maintained in darkness with the surface of the agar vertical. Cultures were grown for 5 to 6 d in darkness, after which protonemata were oriented parallel and upright on the cellophane.

Cultures were immersed in solutions of BSA (Wayne et al., 1990) or in iodixanol. Solutions were injected through a small hole in the dish that was made before sowing and that had been sealed with plastic wrap. The dishes were completely filled to ensure immersion. Because the filaments were anchored in the agar at sowing, they adhered to the cellophane after immersion.

Iodixanol {5,5'-[(2-hydroxy-1,3-propanediyl)-bis(acetylamino)] bis-[N, n = bis(2,3-dihydroxypropyl)-2,4,6-triiodo-1,3 benzenecarboxamide]} is known commercially as OptiPrep (Accurate Chemical and Scientific Corporation). Iodixanol is nonionic and water soluble and was supplied as a sterile 60% (w/v) aqueous solution (Ford et al., 1994). Lower iodixanol concentrations (10%-50% [w/v]) were obtained by dilution with sterile distilled water. BSA (no. A-7906, Sigma) was dissolved in distilled water to final concentrations of 10% to 50% (w/v). The osmolalities shown in Table II were determined using a 5500 Vapor Pressure Osmometer (Wescor Inc., Logan, UT) or were obtained from the manufacturer.

Gravitropism

After the application of BSA or iodixanol, the dishes were reoriented to the horizontal or were inverted. Some dishes were maintained in a vertical, upright orientation after the application of BSA or iodixanol to allow for equilibration before reorientation. At the end of the gravistimulation period, the cultures were chemically fixed in place as in Walker and Sack (1990). Curvature was quantified from video or photographic images of fixed cells. Only protonemata growing on the cellophane were scored.

In some experiments, living cells were studied using a horizontal videomicroscope as in Schwuchow and Sack (1994). The culture chambers described by Schwuchow and Sack (1994) and by Wagner et al. (1997) were used for the BSA and iodixanol experiments, respectively. The solutions were either pipetted directly onto protonemata (BSA) or injected into the dishes (iodixanol) in dim-green light. The responses of 102 (BSA) and 71 (iodixanol) cells were archived using time-lapse video recorders. Growth rates and gravitropic curvature were quantified from the video monitor.

Measurement of Cell Wall Pores and Size of Iodixanol

Porosity was estimated by analyzing the penetration of compounds of different sizes that had been labeled with FITC. FITC-dextrans (10, 20, 40, and 150 kD) and FITC-BSA were obtained from Sigma and from Molecular Probes (Eugene, OR). Both were purified as in Preston et al. (1987) and Cole et al. (1990). Insoluble impurities were removed by the microcentrifugation of 5% (w/v) solutions in sodium phosphate buffer at 8,000g for 1 min. Unbound FITC was removed by loading the supernatants onto prewashed Sephadex columns (G50 or G100, 8 × 240 mm). On elution (flow rate of 300 µL min¹), FITC-dextrans and FITC-BSA were collected as the first fluorescent band. Purified probes were diluted with 20 mM sodium phosphate buffer (pH 7.5) to a final concentration of 0.1% (w/v). FITC (C₂₁H₁₁NO₅S) has a molecular mass of 389 D and, thus, constitutes less than 4% of the weight of FITC-10 kD dextran and less than 0.6% of the weight of FITC-BSA.

The pore size of the cell walls of the apical cells was estimated using the method of Shedletzky et al. (1992). Protonemal mats were cut out from dark-grown cultures. Filaments were treated with 0.05% (w/v) Triton X-100 in nutrient medium for 5 min to allow the penetration of fluorescent markers into the cell. Detergent was applied in dim-green light to prevent light-induced redifferentiation (Kern and Sack, 1999). The filaments were then rinsed with water and immersed in purified solutions of FITC-dextran or FITC-BSA for 1 to 18 h. After an additional rinsing, the cells were examined by fluorescence microscopy (IM 35 microscope, 450-490-nm exciter filter, 510-nm dichroic beam splitter, 530-565-nm transmission filter, Zeiss, Jena, Germany). Cells were visually assigned to one of three levels of relative fluorescence intensity. The lowest was equivalent to autofluorescence seen in control cells immersed in water where both the cytoplasm and plastids were weakly fluorescent. In the medium category, cytoplasmic fluorescence was brighter, but individual plastids could still be distinguished. In the high category, cytoplasmic fluorescence was so bright that individual plastids were hard to detect.

Dimensions of the long and short axes of iodixanol were measured from the three-dimensional molecular structure using teXsan crystallographic software (Rigaku Molecular Structure Corporation, The Woodlands, TX).