Protoplasts of corn coleoptiles and Arabidopsis hypocotyls respond to the plant hormone auxin with a rapid change in volume. We checked the effect of antibodies directed against epitopes of auxin-binding protein 1 from Arabidopsis thaliana (AtERabp1) and Zea mays (ZmERabp1), respectively. Antibodies raised against the C-terminus of AtERabp1 inhibited the response to auxin, while antibodies raised against a part of box a, the putative auxin-binding domain, induced a swelling response similar to that caused by auxin treatment. Synthetic C-terminal oligopeptides of ZmERabp1 also caused a swelling response. These effects occurred regardless of whether the experiments were carried out with homologous (anti-AtERabp1 antibodies on Arabidopsis protoplasts or anti-ZmERabp1 antibodies in maize protoplasts) or heterologous immunological tools. The results indicate that the auxin signal for protoplast swelling is perceived by extracellular ABP1.
The classical effect of the plant hormone auxin is the stimulation of elongation growth (Went, 1928). There is still no consensus as to which receptor molecule perceives the auxin signal in that system (Hertel, 1995; for review see Lüthen et al., 1999; Venis, 1995). However, electrophysiological techniques on protoplasts support the hypothesis of a receptor role of auxin-binding protein 1 (ABP1). The hypothesis is that extracellular ABP1 binds auxin at the apoplastic side of the plasma membrane. The ABP1–auxin complex attaches to a hypothetical transmembrane docking protein (Klämbt, 1990; MacDonald, 1997; Napier, 1995), from where the signalling chain is started. This model is supported by the following findings: Anti-ABP antibodies inhibit auxin-induced electrophysiological effects on protoplasts (Barbier-Brygoo et al., 1989), while antibodies specifically directed against the putative auxin-binding pocket of the protein (e.g. D16 antibodies) have auxin agonist activity (Leblanc et al., 1999a; Rück et al., 1993; Venis et al., 1992). Synthetic C-terminal oligopeptides, which include the domain supposed to attach to the docking protein, induced auxin-like effects on the modulation of K+ channels in guard cells (Thiel et al., 1993), and are also active in the hyperpolarization assay (Leblanc et al., 1999b).
However, there is still no proof that ABP1 is involved in the growth response. Experiments using auxin efflux inhibitors (Claussen et al., 1996; Davies, 1974; Vesper and Kuss, 1990) and the apparent 2,4-D insensitivity of a mutant defective in an auxin-uptake carrier (Bennett et al., 1996) suggest, in contrast, that the auxin signal for growth is perceived intracellularly. Apart from the view of Rainer Hertel (1995), who termed ABP1 a ‘red herring’, it seems to be possible that some responses to auxin are mediated by ABP1 while others are not. For instance, D16 antibodies have auxin agonist activity in the electrophysiological protoplast system, but do not cause the expression of pCNT103-like auxin-induced genes in the same cells (Boot et al., 1996), and there are to our knowledge no studies reporting the induction of other auxin-induced genes by D16 antibodies. Therefore it appears crucial to discriminate between auxin effects perceived by extracellular ABP1 and those possibly mediated via other signalling chains.
Anti-ABP1 antibodies (as Fab fragments) do not show any effect in growth measurements (Lüthen et al., 1999); this might be due to the diffusion barrier imposed by the cell wall, preventing access of the antibodies to the cell surface. The most significant difference between the auxin-induced electrophysiological responses on the protoplast level and the growth effects on the organ level are their vastly different time scales. Auxin-induced growth and proton pumping in intact plant organs commence after a lag phase of 10–20 min, whereas ABP-mediated electrophysiological responses in single cells require only seconds of auxin administration. The purpose of the present study is to examine another physiological auxin response on protoplasts, the time courses of which resembles the growth response more closely, and to study the effects on them of anti-ABP antibodies and oligopeptides.
The auxin-induced increase of protoplast volume, the so-called swelling response (Keller and Van Volkenburgh, 1996), resembles the growth response in some important details. Just as in auxin-induced growth (Claussen et al., 1997), the volume change depends on plasma membrane potassium uptake through potassium channels (Keller and Van Volkenburgh, 1996). The effect is also reasonably specific for active auxins (IAA, 4-Cl-IAA, NAA, but not the inactive analogue 2-NAA; Steffens and Lüthen, 2000). However, in all previous studies protoplast swelling was recorded hours after auxin application. Therefore these experiments do not cover the time window relevant for the understanding of auxin-induced growth.
We recently established a novel technique to monitor the time course of auxin-induced protoplast volume changes with corn coleoptile protoplasts at high temporal resolution (Steffens and Lüthen, 2000). Immediately after auxin addition, a transient shrinkage of the protoplasts was observed, followed by a long-term swelling response. This effect occurred in the same time window as the typical auxin growth response of coleoptiles, which also consists of a transient growth inhibition (‘wrong start’) followed by an increase of growth rate. Fusicoccin, a potent promoter of the PM-H+-ATPase and a strong inducer of growth, also induced a rapid swelling response. Using this method it was possible to compare the effect of auxin at the single-cell level with the growth response in intact organs. This system is appropriate for analysing mechanisms of auxin signal transduction because it is possible to use antibodies and peptides in volumes as small as 20 µl.
In the present paper, we use this technique to address the following questions:
1 Do antibodies against ABP1 affect auxin-induced protoplast swelling?
2 Do antibodies directed against auxin-binding domain box a (e.g. D16 antibodies) have auxin agonist activity on protoplast swelling?
3 Do C-terminal peptides induce auxin-like responses in the system?
Virtually all previous electrophysiological studies have used heterologous immunological tools, for example, antibodies raised against ABP1 from maize were used in tobacco protoplasts (Barbier-Brygoo et al., 1989). As this has raised objections, we additionally critically compared the physiological activities of the antibodies in homologous systems (antibodies raised against AtERabp1 in Arabidopsis protoplasts; ZmERabp1 antibodies used with maize protoplasts) and heterologous systems (anti-ABP antibodies from Arabidopsis used in maize protoplasts).
Origin and specificity of antibodies and peptides
The antibodies were directed against two synthetic peptides of the AtERabp1 sequence (Figure 1a). One peptide (internal peptide) is located in the region of the so-called box a, and shows 87% identity to the peptide from maize (Table 1). The second antibody used in this study (anti-C-term antibody) is directed against a C-terminal peptide of AtERabp1 lacking the last three amino acids DEL, and which shows 50% identity to the maize sequence (Table 1).
Table 1. Comparison of the sequences of the oligopeptides used for raising antibodies or for investigating their physiological effects
The highlighted letters are identical with the maize sequence. The D16 sequence (Venis et al., 1992) is also shown.
Internal peptide from Arabidopsis ABP1
D16 from maize ABP1
C-terminal peptide T11/12 from maize ABP1
C-terminal peptide from Arabidopsis ABP1
Control antibody (goat anti-rabbit)
(ABP1 protein from maize)
Anti-ABP1 antibody from maize ABP1
Other peptides used were the C-terminal peptide of ZmERabp1 (referred to as C-terminal peptide T11/12), and a control peptide with a random sequence unrelated to ABP1 (Table 1).
The specificity of the antibodies used in this study is demonstrated by two immunoblots. Figure 1(b) shows that the anti-intern antibody specifically detects ABP1 from Zea mays (lane 1). Addition of the antigen completely inhibits binding to ABP1 (lane 2). In Figure 1(c), the anti-C-term antibody recognizes recombinantly expressed ZmERabp1 (with an N-terminal ubiquitin fusion), although comparison of the Arabidopsis thaliana and Z. mays sequences reveals only 50% identity. The specificity is demonstrated by inhibition with the C-terminal peptide T11/12 deduced from the Z. mays sequence. We conclude that it is reasonable to use these antibodies in functional studies of ZmERabp1 on maize protoplasts.
Anti-ABP1 antibodies of maize and D16 antibodies (corresponding to anti-intern antibody, but raised against box a of ZmERabp1) were kindly supplied by Dr Richard Napier (Wellesbourne, UK). They have been extensively characterized in previous publications (e.g. Venis et al., 1992).
Time courses of IAA-induced protoplast swelling and growth responses
IAA-induced changes in the net volume of maize protoplasts are shown in Figure 2(a). Immediately after IAA addition there was a transient decrease in net volume: the protoplast began to shrink. After this ‘wrong start’, the protoplast began to swell and increased in volume. The subsequent IAA-induced protoplast swelling occurred on a time scale similar to the growth response of maize coleoptiles (Figure 2b).
Figure 2(c) shows the net volume change of Arabidopsis hypocotyl protoplasts. Just after IAA application, the protoplast increased its volume. There was no transient ‘wrong start’ in Arabidopsis. The lag phase in the time course of the growth effect of intact Arabidopsis hypocotyls (Figure 2d) was much shorter than in corn coleoptiles. Again, protoplast swelling and growth occurred on roughly similar time scales.
Anti-intern and D16 antibodies induce protoplast swelling in both heterologous and homologous test systems
The time course of the anti-intern antibody effect on maize coleoptile protoplasts is shown in Figure 3(a). Immediately after application of 10−8m of the anti-intern antibody, a transient shrinkage of the protoplasts was observed. After a lag phase of 10 min, the protoplast began to increase its volume. The response was similar to that induced by IAA, but of larger amplitude. In the auxin response (Figure 2a), there was a 15 min negative response, followed by swelling. The amplitude of the antibody effect is higher than the auxin response. A slight difference in the general shapes of the curves is that the auxin curve levels off after 50 min, whereas the antibody curve does not.
The anti-intern antibody was found to induce protoplast swelling at concentrations ranging from 10−10 to 10−7m, with an optimal concentration around 10−8m (Figure 3b). To exclude any unspecific effect of the antibody, a control antibody was given to maize protoplasts (Figure 3a). This control antibody slightly increased the endogenous shrinkage of the protoplasts, but did not induce any swelling response on its own.
The protoplasts in homologous systems – D16 antibodies on maize coleoptile protoplasts (Figure 4a) and anti-intern antibody on Arabidopsis hypocotyl protoplasts (Figure 4b) – displayed an increase in net volume. This response started just after application of the antibodies, and was transient in the Arabidopsis system (Figure 4b). After 30 min the protoplasts began to shrink. In the maize protoplast system, an increased steady-state volume is established (Figure 4a).
Anti-C-term and anti-ABP1 antibodies inhibit auxin-induced protoplast swelling
Figure 5(a) depicts the effect of a pretreatment with anti-C-term antibody (10−8m) on auxin-induced swelling of maize protoplast. The antibodies were delivered in a concentrated phosphate-buffered saline solution (PBS). This caused an osmotic effect, leading to slight shrinkage of the protoplasts. The effect of PBS (without antibodies) is shown as a control in Figure 5(b). Another control experiment demonstrates that this PBS effect alone did not prevent the auxin-induced volume change (Figure 5a). We conclude that the anti-C-term antibody effectively renders the protoplast system insensitive for the applied auxin concentration. The anti-ABP1 antibody, raised against the entire ABP1 protein, had a similar effect (Figure 5c). As it was delivered at a slightly higher PBS concentration, the buffer-induced shrinkage is more pronounced in this experiment.
Micromolar concentrations of the C-terminal peptide T11/12 induced an increase of net volume of maize coleoptile protoplasts (Figure 6). The amplitude of the peptide effect was similar to that of anti-intern antibodies, and exceeded the response to optimal auxin concentrations. In contrast to the effects of both auxin and anti-intern antibody, the peptide response showed neither a lag phase nor a transient negative response. A control peptide with a sequence not related to ABP1 did not induce any swelling, but caused a transient decrease in net volume (Figure 6).
Antigen inhibition of anti-intern antibody induced protoplast swelling
Figure 7(a) shows the effect of competition experiments in which a mixture of 10−8m anti-intern antibody and 10−6m internal peptide was used. There was no characteristic anti-intern antibody-induced protoplast swelling response. Instead, a slow decrease in protoplast volume was observed. In order to exclude any nonspecific interference between antibodies and peptide-induced swelling, a mix of 10−8m anti-intern antibody and 10−6m C-terminal peptide T11/12 was used as a control in a further experiment (Figure 7b). This treatment resulted in an instantaneous swelling response, the time course of which was virtually identical to the effect of C-terminal peptide T11/12 alone (Figure 7b).
Protoplast swelling is a cellular response controlled by auxins in a rapid and specific manner (Figure 2a,c; Steffens and Lüthen, 2000). It occurs in the same time window as the classical growth response at the organ level (Figure 2b,d). Although the match between the time courses of growth and protoplast swelling is only partial, both share some common overall features. In the maize growth response, the lag phase is much longer than in Arabidopsis. The same is true for the transient inhibition of maize and Arabidopsis protoplast swelling.
Antibodies against ABP1 affected protoplast swelling in a specific manner: those directed against the whole ABP protein or against the C-terminus (Figure 1) prevented auxin-induced swelling (Figure 5); whereas anti-intern and D16 antibodies, directed against the putative auxin-binding domain box a, by themselves induced protoplast swelling in a manner resembling the auxin response (Figure 4). We conclude from these findings that the auxin signal leading to protoplast swelling is perceived by auxin binding to extracellular ABP1.
Similar effects of D16 and anti-ABP antibodies to those reported here have been shown in membrane hyperpolarization assays in tobacco mesophyll protoplasts (Barbier-Brygoo et al., 1989; Venis et al., 1992) and in a patch-clamp measurement of an ATP-driven current (Rück et al., 1993). The present study is the first demonstration of ABP1 signalling in a physiological (rather than electrophysiological) system in cells isolated from organs that display auxin-induced growth (Figure 2).
This is not just a matter of a different technical approach: it appears that some auxin responses are mediated by ABP1 while others are not, and it is therefore necessary to establish which auxin responses depend on ABP1 and which do not. We see this work as a contribution to this.
As the few mV of hyperpolarization detected in the tobacco mesophyll protoplast cannot account for the volume changes reported in the present study, protoplast swelling may reflect a response at the level of the whole cell, rather than a subtle modulation at the level of membrane transport. As auxin is known to modulate a large number of effects, this tentatively suggests a significant role of ABP signalling in auxin action.
Nearly all previously published work on the physiological activity of anti-ABP antibodies has been done in heterologous systems. Anti-maize ABP antibodies were tested in the tobacco mesophyll protoplast hyperpolarization assay, although the sequences of maize and tobacco ABP are generally of limited homology. This problem has been a matter of criticism for years. Even the D16 peptides, against which the D16 antibodies were raised, contain flanking regions to box a which are not conserved (Table 1). The relevance of the published data has been questioned on the basis of this argument. The present paper addresses this problem by a rigorous comparison of heterologous (Figures 3 and 5a) and homologous (Figures 4 and 5c) immunological tools in the same experimental set-up. Although the time courses differed in detail, the general direction of the effect was always the same in both arrangements. The fact that comparable effects were detected in both arrangements is also relevant for the interpretation of previous electrophysiological work by other groups, and suggests that their interpretation – that ABP1 is a relevant auxin receptor – is basically correct. This, however, does not necessarily mean that all known auxin effects are mediated via extracellular ABP1.
We could also demonstrate that the synthetic C-terminal oligopeptides T11/T12 induce protoplast swelling in a manner similar to auxin (Figures 6 and 7). Similar peptide effects have been shown in electrophysiological responses of guard cells (Thiel et al., 1993), and there is one report of membrane hyperpolarization induced by C-terminal peptides in tobacco mesophyll protoplast (Leblanc et al., 1999b). Both guard cells and tobacco leaves are far from being classical systems of auxin physiology. Therefore it appears daring to relate these observed effects to the typical auxin-induced growth responses. Here we demonstrate for the first time such effects in coleoptile and hypocotyl cells. In terms of time scales, origin of the cells and phenomenon studied, the gap between the ABP1-mediated cellular responses and classical auxin-induced growth becomes narrower.
Generally, the present study strengthens the view that extracellular ABP1 is a physiologically relevant auxin receptor, perceiving the auxin signal in at least some auxin responses. However, the involvement of extracellular ABP1 in the growth response remains far from proven, especially as the biophysical causes for auxin- and fusicoccin-induced growth responses, and the volume changes, are thought to be quite different. This matter may be clarified by a critical comparison of the responses of auxin signal-transduction mutants at the single cell level and at the organ level, in order to explore the similarities between auxin-induced and antibody-induced effects.
Anti-ABP1 and D16 antibodies from maize
Both antibodies were kindly supplied by Dr R. Napier (Wellesbourne, UK). The antibodies were stored in PBS.
Synthesis of all other peptides and antibodies
The peptides were produced by standard solid-phase procedures using the peptide synthesizer EPS221 (Abimed, Langenfeld, Germany). After coupling the peptides to keyhole limpet hemocyanine, two rabbits were immunized (Eurogentec, Belgium). The antibodies were affinity purified against the immobilized peptides and characterized on immunoblots.
SDS–PAGE and immunoblot analysis
SDS–PAGE was performed using 12% polyacrylamide gels. The proteins were transferred to a PVDF membrane. After blocking with 2% fat-free milk, membranes were incubated for 1 h with the anti-intern antibody or the anti-C-term antibody, respectively. After washing in PBS plus 0.02% Tween, bound antibodies were detected by goat anti-rabbit IgG–peroxidase conjugate and the Super Signal Substrate kit (Pierce, Rockford, IL, USA). For competition, the antibodies were pre-incubated for several minutes with a 1 : 100 molar excess of peptide.
Preparation of protoplasts from maize coleoptiles
Seeds of maize (Zea mays L. cv. Garant) were soaked overnight and grown on well watered filter paper at 26°C in the dark. After 5 days, eight coleoptiles were harvested and abraded (SiC powder with 1200 mesh from K. Schriever, Hamburg, Germany). The apical 3 mm of the coleoptiles were cut and discarded.
The isolation of protoplasts and volume change measurements took place in a washing solution containing 1 mm CaCl2·2H2O, 10 mm MES and 10 mm KCl pH 6.5. A Roebling osmometer was used to control the osmotic pressure. The osmolarity of all solutions employed was adjusted to an osmolarity of 330 mM.
Segments (1 cm) directly below the first excision (without the primary leaf) were collected and cut into very small pieces. The pieces were digested for 4.5 h by gentle shaking in a digesting solution (pH 5.6) containing 1% cellulase ‘Onozuka’ RS (Yakult Honsha Co., Tokyo, Japan), 0.075% Pectolyase Y-23 (Kikkoman Corporation, Tokyo, Japan) and some macro- and micronutrients (Krautwig, 1993). After digestion, the protoplasts were separated by using sieves with 100, 50 and 30 µm mesh. The protoplasts were stored in washing solution. In the preparation, the diameters of the protoplasts varied from 25 to 120 µm. Initial experiments showed that all intact protoplasts in the size range 50–60 µm responded to auxin. This fraction was therefore used for all experiments reported here.
Preparation of protoplasts from Arabidopsis hypocotyls
The protoplasts from Arabidopsis hypocotyls were isolated as described above, except that (i) seeds of A. thaliana (Wt, Columbia) were placed on well watered filter paper on Petri dishes; and (ii) seeds were stored in the dark at 4°C, then given 16 h light at 25°C, after which hypocotyls were grown in the dark for 2 days at 25°C. The next day, 40 hypocotyls were harvested and cut into small pieces.
The size distribution of Arabidopsis protoplasts was found to be similar as in maize. A fraction of protoplasts with a diameter of 40–60 µm proved to be auxin-sensitive, and was used in this study.
Measurement of protoplast cross-sectional areas
The method used to detect protoplast swelling at a high time resolution has been described elsewhere (Steffens and Lüthen, 2000). Briefly, protoplasts were observed on microtitre plates containing 20 µl washing solution per well. Typically, 20 protoplasts were placed in each well. To prevent any changes in osmolarity due to evaporation, the plates were covered with a lid. Applications were made by means of a microlitre syringe through five small perforations in the lid. All experiments were performed at room temperature (22 ± 2°C) at a quantum flow of 150 µm photons s−1.
Individual, vital protoplasts were photographed with an inverted microscope (objective ×40) at regular time intervals. The resulting slides were projected with an enlarger (DurstM301, Nevoneg, Italy) onto the surface of a graphics tablet (GT-1212B, Genius, Langenfeld, Germany). By drawing the outline of the protoplast with the digitizing pen of the graphics tablet, about 120 x–y coordinates were determined and stored in ASCII format. In Hamburg we developed software to derive protoplast volumes from these data allowing for deviation from the spherical shape. For each protoplast, relative volumes (percentage of initial volume) were calculated. We found that the relative volumes of untreated control protoplasts decreased slowly with a linear time function (r ≥ 0.95). To correct for this endogenous shrinkage, we performed a linear regression of the time function until the addition of an effector. This function was extrapolated and subtracted from the measured volumes, yielding net volumes; details of the numerical treatment are described in an earlier paper (Steffens and Lüthen, 2000). Plots throughout the present paper show the change in net volume as a function of time, and are typical results of four to ten individual experiments from two to five protoplast preparations. There was generally no significant deviation between experiments performed with protoplasts from different preparations. The vitality was tested with neutral red. Only protoplasts that accumulated the dye were taken for experiments. Vital protoplasts maintained cytoplasmic streaming for the duration of the experiments.
The rapid growth effects of maize coleoptiles were measured using positional angular transducers (Lüthen and Böttger, 1992). This set-up permits six simultaneous experiments at a temporal resolution of 3 min. Each measuring cuvette contained 4 ml buffer [5 mm MES, 1.25 mm Ca(OH)2, 10 mm KCl pH 6.04]. Data were logged by means of an AD-converter card and processed on a personal computer.
The growth responses of Arabidopsis hypocotyls were measured by a CCD-auxanometer. At time intervals of 3–5 min, a computer-controlled CCD camera (LCCD 14, OES, Egloffstein, Germany) took images of a growing hypocotyl segment placed in a special flow-through cuvette perfused with an aerated incubation solution (10 mm KCl, 1 mm CaCl2 pH 5). The data were analysed using custom and self-written software. Details of the methods are published elsewhere (Christian and Lüthen, 2000).
Growth substance, antibodies and peptides
IAA stock solutions were freshly prepared from IAA potassium salt (Merck, Darmstadt, Germany) in buffer or protoplast washing solution. Antibodies and peptides were stored in PBS. After dilution in washing solution, aliquots were added to create the final concentrations indicated.
For the competition experiments, aliquots of the antibodies were mixed with a 100-fold excess of peptides. This solution was incubated for 1 h and applied. The final concentration was 10−8 and 10−6m for the antibodies and peptides, respectively.
This paper is based in part on a doctoral thesis by Bianka Steffens and May Christian at the University of Hamburg, and by Christian Feckler at the Max-Delbrück-Laboratory of Cologne. We are very indebted to Richard Napier for his generous gift of D16 and anti-ABP1 antibodies. Funding by DFG (grant no. Bo537.16) is gratefully acknowledged. Bianca Steffens wishes to acknowledge the support by a Phd student’s grant of the University of Hamburg.