In vivo Characterization of the Effects of Abscisic ... - Semantic Scholar

Annals of Botany 89: 391±400, 2002 doi:10.1093/aob/mcf057, available online at www.aob.oupjournals.org

In vivo Characterization of the Effects of Abscisic Acid and Drying Protocols Associated with the Acquisition of Desiccation Tolerance in Alfalfa (Medicago sativa L.) Somatic Embryos L E K H A S R E E D H A R 1 , ² , W IL LE M F . W O L K E R S 2 , ³ , FO L K E R T A . H O E K S T R A 2 , * and J . D E R EK B EW L EY 1 1Department of Botany, University of Guelph, Guelph, Ontario, Canada N1G 2W1 and 2Department of Plant Sciences, Laboratory of Plant Physiology, Wageningen University, Wageningen, The Netherlands Received: 9 April 2001 Returned for revision: 10 November 2001 Accepted: 15 December 2001

Although somatic embryos of alfalfa (Medicago sativa L.) had acquired some tolerance to desiccation at the cotyledonary stage of development (22 d after plating), additional culturing in 20 mM abscisic acid (ABA) for 8 d induced greater desiccation tolerance, as determined by increased germination. Compared with fast drying, slow drying of the ABA-treated embryos improved desiccation tolerance. However, slow drying of non-ABA-treated embryos led to the complete loss of germination capacity, while some fast-dried embryos survived. An electron paramagnetic resonance spin probe technique and in vivo Fourier transform infrared microspectroscopy revealed that cellular membrane integrity and a-helical protein secondary structure were maintained during drying in embryos cultured in media enriched with 20 mM ABA, but not in embryos cultured in the absence of ABA. Slow-dried, non-ABA-treated embryos had low oligosaccharide to sucrose ratios, an increased proportion of b-sheet protein secondary structures and broad membrane phase transitions extending over a temperature range of more than 60 °C, suggestive of irreversible phase separations. The spin probe study showed evidence of imbibitional damage, which could be alleviated by prehydration in humid air. These observations emphasize the importance of appropriate drying and prehydration protocols for the survival and storage of somatic embryos. It is suggested that ABA also plays a role in suppressing metabolism, thus increasing the level of desiccation tolerance; this is particularly evident under stressful conditions such as slow drying. ã 2002 Annals of Botany Company Key words: Abscisic acid, alfalfa (Medicago sativa L.), desiccation tolerance, electron paramagnetic resonance spectroscopy, Fourier transform IR spectroscopy, germination, membrane integrity, protein secondary pro®le, somatic embryos, sugars.

INTRODUCTION Somatic embryogenesis is an in vitro regeneration system whereby an organized bipolar structure containing a root and shoot apex, and morphologically resembling a zygotic embryo, is formed from one somatic cell. After multiplication of cell suspensions on auxin medium, cytoplasmic-rich cells and cell clusters are formed. Transfer of these cells to hormone-free medium leads to the successive development of globular, heart- and torpedo-shaped somatic embryos, and ®nally to the outgrowth of plantlets. At present, a few hundred plant species can be regenerated through somatic embryogenesis. Because it is dif®cult to keep hydrated somatic embryos alive for extended periods of time, even at near freezing temperatures, a method that renders somatic embryos desiccation tolerant is desirable. Survival times during storage would thus be extended considerably, particularly when using cryogenic storage. Desiccation tolerance of * For correspondence. Fax +31 317 484740, e-mail folkert.hoekstra@ pph.dpw.wau.nl ² Present address: Cell Biology & Biophysics, School of Biological Sciences, University of Missouri-Kansas City, MO 64110, USA. ³ Present address: Section of Molecular and Cellular Biology, University of California, Davis, CA 95616, USA.

somatic embryos has been de®ned as the ability to resume growth after storage for some weeks at room temperature and at water contents as low as 10 % (on a d. wt basis) or less (Tetteroo et al., 1995a). By using recombinants of abscisic acid (ABA)-de®cient and ABA-insensitive mutants of Arabidopsis thaliana, Koornneef et al. (1989) have unequivocally demonstrated that endogenous ABA is involved in the acquisition of desiccation tolerance in seeds. By adding appropriate amounts of ABA to the culture medium at the appropriate stage of development, somatic embryos could be dehydrated to moisture contents of less than 10 % with retention of some viability. Slow drying procedures have also been shown to improve desiccation tolerance (Tetteroo et al., 1995a; Bomal and Tremblay, 1999; Wolkers et al., 1999). The production of desiccation-tolerant somatic embryos has been reported for the following species: alfalfa (McKersie et al., 1989; Senaratna et al., 1989), geranium (Marsolais et al., 1991), soybean (Parrott et al., 1988), spruce (Roberts et al., 1990; Attree et al., 1991; Bomal and Tremblay, 2000), grape (Gray, 1989) and carrot (Iida et al., 1992; Lecouteux et al., 1992; Tetteroo et al., 1995a). However, the conversion/establishment frequency is generally low. Poor storage reserve accumulation and subã 2002 Annals of Botany Company

392

Sreedhar et al. Ð Desiccation Tolerance in Alfalfa Somatic Embryos

optimal culture protocols have been cited as the reasons for poor seedling establishment (Krochko et al., 1992; Lai and McKersie, 1994; Tetteroo et al., 1995a). Somatic embryogenesis is a useful approach by which to study the acquisition of desiccation tolerance because it is possible to produce tolerant and intolerant embryos by manipulating the culture media. Alfalfa somatic embryogenesis involves callus induction and multiplication, followed by embryo induction, maturation, acquisition of desiccation tolerance and desiccation. During zygotic embryogenesis, the ability to tolerate desiccation increases progressively during development, probably due to a series of physiological changes. These include the synthesis of LEA proteins (Galau et al., 1987; Blackman et al., 1992) and non-reducing sugars during late embryogenesis (Blackman et al., 1992), although the cellular mechanisms behind tolerance of dehydration are not fully understood. It has been shown that dehydration alters the phase behaviour of membrane lipids (reviewed in Hoekstra and Golovina, 1999) and may lead to lateral phase separation of membrane components in somatic embryos that are desiccation sensitive (Tetteroo et al., 1996). Furthermore, dehydration may result in irreversible denaturation of proteins in model protein systems (Prestrelski et al., 1993; Wolkers et al., 1998a) and in desiccation-sensitive seeds (Wolkers et al., 1998b). The results of these damaging events can be measured in the dried state using Fourier transform infrared (FTIR) microspectroscopy (Wolkers and Hoekstra, 1995; Wolkers et al., 1998b, 1999). The membrane integrity after rehydration can be measured in very small specimens by a nitroxyl spin probe method (Golovina et al., 1997; Hoekstra et al., 1999). The application of FTIR to proteins is based on the assessment of the amide-I band, located between 1600± 1700 cm±1 (Byler and Susi, 1986; Surewicz and Mantsch, 1988). The structure-sensitive amide-I band in the infrared spectrum can be used to detect changes in the overall protein secondary structure in plant tissues in vivo. a-Helical structures are dominant in desiccation-tolerant plant tissues, whereas b-sheet structured protein aggregates are formed when desiccation-sensitive tissues are dried (Wolkers and Hoekstra, 1995; Wolkers et al., 1998b). Here we describe a protocol to produce fully desiccationtolerant embryos of alfalfa. The importance of developmental stage, ABA, the drying regime and prehydration in maximizing desiccation tolerance is correlated with characteristics such as germination capacity and changes in overall protein secondary structure and membrane integrity, detected using in vivo FTIR spectroscopy and electron paramagnetic resonance (EPR) spectroscopy. MA TE R IA L S A N D ME T H O D S Plant material

Zygotic seeds of alfalfa (Medicago sativa L. `Rangelander', clonal line RL-34) were obtained from the Department of Crop Science, University of Guelph, Ontario, Canada. Plants were grown in chambers at 23 °C with a 16 h light and 8 h dark photoperiod (100 mmol m±2 s±1) at 90 % relative

humidity (RH). Plants were fertilized with 20 : 20 : 20 NPK every 2 weeks. Culture conditions and drying regimes

Callus induction and embryogenesis were carried out using SH (Schenk and Hildebrandt, 1972) medium. In all cases the pH of the medium was adjusted to 5´8 prior to autoclaving at 0´122 MPa for 25 min. Petioles of young leaves were sterilized by ®rst immersing them in 95 % ethanol for 1 min followed by immersion in 30 % (v/v) commercial bleach (5´6 % sodium hypochlorite) containing two drops per 100 ml of the surfactant `Tween 20' (polyoxyethylenesorbitan monolaurate; Sigma, St Louis, MO, USA) for 20 min. The petioles were then rinsed repeatedly in sterile distilled water and sliced into pieces approx. 1 cm in length. Ten explants each were placed in Petri dishes (100 mm diameter 3 15 mm height), each containing 25 ml SH medium supplemented with 3 % sucrose, 1 mg l±1 2,4-D, and 0´2 mg l±1 kinetin (callusinduction medium). The Petri dishes were sealed with Para®lm and incubated in growth cabinets at 25 °C and a 16 h photoperiod (75 mmol m±2 s±1) for 2 weeks. The callused petiole explants were then transferred to liquid SH medium containing 1 mg l±1 2,4-D (callus-multiplication medium) for 1 week at 25 °C and a 16 h photoperiod (35 mmol m±2 s±1). Calli were ®rst sieved through a 500 mm nylon screen and then through a 200 mm nylon screen (Nitex; B & SH Thompson, Toronto, Canada). The resulting cells were plated uniformly onto a 200 mm nylon screen and placed on an SH-based embryo induction (EI) medium containing 5 % sucrose with or without 1 mM ABA. Added ABA at low concentration has been found to promote the accumulation of storage reserves (Sreedhar and Bewley, 1998). After 12 d, following completion of the globular and heart stages, the torpedo-stage embryos were transferred to a glutaminesupplemented (50 mM) SH basal medium with 9 % sucrose for 10 d for conversion to the cotyledonary stage and embryo maturation. Optionally, some of these embryos [22 d after plating (dap)] were fast-dried, slow-dried or dried at an intermediate rate, as described below, and allowed to develop for an additional 8 d. Promotion of the acquisition of desiccation tolerance was achieved by the subsequent transfer of embryos to SH basal medium supplemented with ABA (20 mM) and 5 % sucrose [Desiccation Tolerance Promoting (DTP) medium]. After 8 d, these embryos (30 dap) were dried at three different rates to less than 10 % moisture content of embryo dry weight and stored at ambient temperature and humidity. Embryos were slow-dried at room temperature by placing them sequentially, for 5 d each, in desiccation chambers at a range of decreasing RHs (75, 63, 45 and 30 %) established over saturated salt solutions. An intermediate rate of drying was achieved by exposing the embryos to this range of RHs for 2 d each. For fast drying, embryos were spread on a ®lter paper in a Petri dish and exposed to a rapid air¯ow at room temperature (20±25 °C) and an ambient RH of 40±50 %. Desiccation tolerance was quantitated by measuring germination frequency. Embryos were collected after 22


393

or 30 dap, either ABA treated or non-ABA treated, and were germinated either fresh, or after fast, slow or intermediate rates of drying. Protrusion of the radicle and greening of the cotyledons were taken as an indication of germination and these were scored after 7 d on a moist ®lter paper saturated with sterile water in closed containers. Sugar analysis

Samples were prepared and analysed according to de Bruijn et al. (1997). Brie¯y, individual, pre-weighed and dried somatic embryos, which were ABA-treated (20 mM) or non-ABA treated, and zygotic seeds were ground in small glass Potter tubes in methanol (80 %), containing lactose as the internal standard. There were ®ve replications per treatment. The extracts were heated for 10 min at 75 °C before the solvent was removed by drying in a Speed-Vac (Savant Instruments Inc., Farmingdale, NY, USA) for 2 h. The pellet was resuspended in water and the ®nal volume was brought to 1 ml. The supernatant was collected after centrifugation for 5 min in a table-top centrifuge at maximum speed (15 000 rpm), and analysed for sugars by HPLC. Sugars were separated with a Dionex HPLC system (Dionex Corporation, Sunnyvale, CA, USA) equipped with an ED40 pulsed electrochemical detector. Sugars were chromatographed on a CarboPac PA (4 3 250 mm) column preceded by a guard column (CarboPac PA100, 4 3 50 mm). The ¯ow rate was 1 ml min±1 at 4 °C. The electrodes were maintained at a constant temperature of 4 °C. Identi®cation of carbohydrate peaks was by comparison with retention times of standard solutions. A 50±200 mM gradient of NaOH in water was used as the eluent, and 1´1 M sodium acetate in 100 mM NaOH was used to clean the column after each run. Data were analysed using a Spectra Physics integrator model SP 4400 and Spectra Physics software (Labnet; Chromdat, San Jose, CA, USA). Membrane integrity

A nitroxyl spin probe technique was used to estimate the integrity of plasma membranes in alfalfa somatic embryos. This technique is based on the differential permeability of membranes to the stable 4-oxo-2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPONE) radical and to ferricyanide ions that broaden the signal (Miller, 1978; Golovina and Tikhonov, 1994; Golovina et al., 1997). When plasma membranes are intact, the EPR spectrum is exclusively derived from spin probe molecules within the cell, because the spectral contribution from extracellular TEMPONE molecules is broadened to invisibility by the ferricyanide ions. Inside the cell, TEMPONE molecules partition between the lipid fraction (oil and membranes) and the aqueous cytoplasm. The intracellular EPR signal of TEMPONE therefore originates from these two environments, which are best resolved in the high ®eld (right-hand-side peaks) region of the spectrum. When ferricyanide penetrates the cytoplasm through deteriorated plasma membranes, the aqueous signal is broadened, but the lipid signal remains because the lipid phase is impermeable to ferricyanide. The ratio (W/L)

F I G . 1. EPR spectra of TEMPONE in alfalfa somatic embryos (30-d-old) that were allowed to imbibe in water for 10 min, followed by incubation for 10 min in a TEMPONE/ferricyanide solution (1 mM/120 mM). All embryos were prehydrated in moisture-saturated air for 3 h prior to imbibition. Embryos were cultured in the presence or absence of 20 mM ABA in DTP medium and dried at a slow or intermediate rate. The ratios between the aqueous (W) and hydrophobic (L) peak heights (W/L) are representative of the plasma membrane integrity, and are presented in Table 3.

between the line heights of the aqueous (W) and lipid (L) components of the spectrum was used to quantitatively characterize membrane permeability (Fig. 1). The intensity of the lipid signal serves as the reference for the amount of material under investigation, whereas the intensity of the water component is negatively correlated with the amount of ferricyanide molecules that penetrated the cells. EPR spectra were obtained at room temperature using an X-band EPR spectrometer (model 300E; Bruker Analytik, Rheinstetten, Germany). The EPR settings were: modulation amplitude of 0´5 Gauss, ®eld centre at 3480 Gauss with a scan range of 60 Gauss and a microwave frequency of 9´8 Ghz. The microwave power was 2´02 mW. The Zeeman ®eld modulation was 100 kHz, and the scan time was 80 s. Embryos were ®rst placed in water for 10 min and then in a solution of 1 mM TEMPONE (Sigma) + 120 mM potassium ferricyanide for 10 min. Imbibition before the application of the spin probe solution was meant to allow the plasma membranes to optimally re-establish barrier properties. Optionally, imbibition was preceded by prehydration of the embryos in moisture-saturated air for 3 h to avoid imbibitional injury (Tetteroo et al., 1996). Embryos were loaded into 2-mm diameter glass capillaries (with sample length of approx. 5 mm) and a small amount of the probe solution was added to keep the material hydrated. The

394


TA B L E 1. Germinability of alfalfa somatic embryos in relation to ABA in the growth media and to desiccation protocol Presence of ABA in medium

Germination (%)

DTP-medium

Embryo age (days after plating)

Fresh

Fast drying

Interm. drying

Slow drying

±ABA ±ABA ±ABA

±ABA +ABA

22 30 30

72 80 0

7 27 62

0 15 70

0 0 63

+ABA +ABA +ABA

±ABA +ABA

22 30 30

82 88 0

18 33 65

5 30 83

0 0 78

EI-medium

Somatic embryos were induced in SH medium (EI medium) with or without 1 mM ABA, followed (optionally) by culture in desiccation tolerance promoting (DTP) medium with or without 20 mM ABA. All embryos were placed on moist ®lter paper either fresh, after fast drying, slow drying or after drying at an intermediate (interm.) rate, and germination was recorded after 7 d. The tests were performed on 60 embryos (in batches of ten), and germination counts were expressed as a percentage. Data are signi®cantly different (P < 0´05) when they diverge by 20 % or more (c2-test).

capillaries were accommodated within standard 4-mm diameter quartz tubes. Four scans were accumulated.

Deconvolved amide-I bands were ®tted with ®ve Pearson band functions using Peak®t software (Jandel Scienti®c, San Rafael, CA, USA) (Wolkers et al., 2001).

Fourier transform infrared microspectroscopy

IR spectra were recorded on an FTIR spectrometer (model 1725; Perkin Elmer, Beacons®eld, Buckinghamshire, UK), equipped with a liquid nitrogen-cooled narrow band Mercury/Cadmium/Telluride (MCT) detector and a PerkinElmer microscope as described previously (Wolkers and Hoekstra, 1995). Membrane ¯uidity. To study membrane ¯uidity, FTIR spectra were recorded in a temperature-controlled brass cell. Cross-sections of embryos were pressed slightly between two diamond windows. Forty to ®fty spectra were recorded over a temperature range of ±60 to +60 °C. The acquisition parameters were as described by Wolkers and Hoekstra (1995). A resolution of 4 cm±1, 32 co-added interferograms, 2 cm s±1 mirror speed, 3500±900 cm±1 wave number range and a triangle apodization function were used. The time taken for acquisition and processing of a spectrum was 0´5 min. Membrane ¯uidity was monitored by observing the band position of the CH2 asymmetric stretching vibration around 2852 cm±1. The gel-to-liquid crystalline phase transition temperature can be estimated from possible discrete shifts in these band positions with temperature. Protein secondary structure analysis. Embryos, both somatic and zygotic, were cross-sectioned, pressed slightly between two diamond windows and placed in the brass cell. The spectral region 1800±1500 cm±1 containing the amide-I and amide-II absorption bands of the protein backbones was selected, and deconvolved spectra were calculated using the interactive Perkin-Elmer routine for Fourier selfdeconvolution (smoothing factor 15 and width factor 30 cm±1). Peak positions were determined using Perkin Elmer graphics software with an uncertainty of 6 0´1 cm±1. FTIR spectra for studying the protein secondary structure were recorded at room temperature. Spectral analyses were conducted using the Infrared Data Manager Analytical Software, version 3´5 (Perkin Elmer).

R E S U LT S Germination of desiccated embryos

Somatic embryos were subjected to three different drying regimes: a fast, slow (over a period of 20 d) or intermediate (over a period of 8 d) rate. These treatments were given after 22 d of development in EI medium or after additional culture for 8 d in DTP medium, which was aimed at promoting the acquisition of desiccation tolerance (30-d-old embryos). Irrespective of the presence of ABA (1 mM) in the EI medium, fresh 22-d-old embryos had a germination frequency of >70 % before desiccation (Table 1). This was also true of the fresh 30-d-old somatic embryos cultured without ABA (20 mM) in the DTP medium. However, when cultured in the presence of ABA, fresh 30-d-old embryos did not germinate. The inhibition of germination by ABA was only temporary because after a cycle of drying and rehydration these embryos were able to germinate (Table 1). From Table 1 it is clear that ABA in the DTP medium considerably improved desiccation tolerance, more so after slow/intermediate drying rates than after fast drying. The degree of desiccation tolerance was not improved signi®cantly if culture in the EI medium had occurred in the presence of ABA (1 mM). Nevertheless, 30-d-old embryos cultured in the absence of ABA showed some desiccation tolerance, and the shorter the drying period the better the survival of desiccation. None of the slow-dried embryos survived. This suggests that ABA plays a role in slowing ageing phenomena under conditions of partly reduced moisture content. The 30-d-old embryos cultured in the absence of ABA survived better than the 22-d-old embryos. Sugars and desiccation tolerance

The striking difference in desiccation tolerance between ABA-treated and non-ABA-treated somatic embryos after slow drying prompted us to investigate possible differences


395

TA B L E 2. Sugar composition and contents of slow-dried alfalfa somatic embryos that were cultivated in DTP medium in the presence (desiccation tolerant) or absence (desiccation sensitive) of 20 mM ABA mg per embryo or seed (6 s.d.) Embryo type/treatment

Sucrose

Raf®nose

Stachyose

Ratio oligo/suc

Somatic embryos, +ABA Somatic embryos, ±ABA Zygotic seed

37 6 5 80 6 2 460

24 6 3 15 6 3 562

50 6 7 50 6 4 36 6 5

2´0 6 0´5 0´8 6 0´1 10´3 6 2´0

Total cultivation time was 30 d. Values for zygotic seeds are also given. Data are the average of ®ve replications (6 s.d.). oligo/suc indicates the weight ratio of the oligosaccharides raf®nose + stachyose to sucrose.

TA B L E 3. Membrane integrity of differently treated alfalfa somatic embryos as derived from EPR spectra of a solution of TEMPONE/ferricyanide (1 mM/120 mM), in which the somatic embryos were immersed for 10 min Ratio W/L Mode of rehydration No pretreatment Humid air pretreatment (3 h)

+ABA slow-dried

±ABA slow-dried

±ABA interm. rate

3´2 4´8

0´7 1´1

2´4 3´8

Membrane integrity, expressed as the ratio W/L, was calculated from the amplitudes of the aqueous (W) and hydrophobic (L) peaks at the high-®eld (right-hand-side) region of the EPR spectra (see Fig. 1). Embryos were cultivated in DTP medium in the presence or absence of 20 mM ABA and were dried at a slow or intermediate (interm.) rate. Dried embryos were prehydrated in moisture-saturated air for 3 h prior to imbibition, or were left untreated. Labelling was performed after 10 min of imbibition in water. LSD was 0´88 at P = 0´05.

in sugar composition and content. Table 2 shows the contents of sucrose and oligosaccharides in these somatic embryos as compared with those of zygotic seeds. In all samples, somatic and zygotic, the trisaccharide stachyose was present in relatively high amounts, but the amount of sucrose was particularly low in the zygotic seeds. Desiccation tolerance has been linked to relatively high oligosaccharide : sucrose mass ratios (Steadman et al., 1996). In this experiment, the mass ratio was 2´0 for tolerant somatic embryos, 0´8 for intolerant embryos and 10´3 for zygotic seeds. Membrane integrity on rehydration

A spin probe technique was used to determine membrane integrity. This technique is superior to leakage or conductivity methods because it measures the ability of an external polar ion to penetrate into the cytoplasm, and not the out¯ow of endogenous ions, the extent of which depends on the leakage history (Hoekstra et al., 1999). Figure 1 gives examples of EPR spectra of TEMPONE in 30-d-old somatic embryos that were cultured in DTP medium with ABA (slow-dried) or without ABA (intermediate and slow drying rates). The embryos were prehydrated for 3 h in humid air to reduce injury that may result from abrupt imbibition (Tetteroo et al., 1996). After 10 min of imbibition they were incubated in TEMPONE/ferricyanide solution for 10 min prior to spectrum recording. The ratio (W/L) between the line heights of the polar and hydrophobic peaks to the baseline gives an estimate of the cellular membrane integrity of the sample. The higher the W/L-value, the

more viable are the cells. From the spectra in Fig. 1 it is clear that the slow-dried, ABA-treated embryos (top spectrum) had the highest W/L ratio, but also that the non-ABA-treated embryos had a reasonably high W/L value when they were dried at an intermediate rate (bottom spectrum). However, when dried slowly, non-ABA treated embryos were characterized by a loss of plasma membrane integrity as can be derived from the considerable reduction of the aqueous peak from the spectrum (middle spectrum). These embryos were extremely desiccation sensitive (see Table 1). Accordingly, W/L-values were calculated separately for embryos that were, or were not, prehydrated in humid air (Table 3). The W/L-values for those embryos that were prehydrated in humid air were higher than for those that were not pretreated. Phase transitions of membranes in dried and rehydrated embryos

In vivo FTIR microspectroscopy was used to estimate the temperature range over which phase transitions in the membranes occur. The peak position of the absorption band due to the symmetric CH2 stretching vibration at around 2852 cm±1 was followed over the temperature range ±60 to +60 °C. It has previously been established that the in vivo IR absorbance peak at around 2852 cm±1 in dry seeds, (somatic) embryos and pollen may, in principle, arise from both oil and membrane phospholipids (Hoekstra et al., 1991, 1993; Wolkers et al., 1998b). The oil in alfalfa somatic embryos has completely melted at ±10 °C (Hoekstra

396

Sreedhar et al. Ð Desiccation Tolerance in Alfalfa Somatic Embryos carrot somatic embryos (Tetteroo et al., 1996) and after ageing of pollen (Van Bilsen et al., 1994). We suggest that free radical-induced changes in the lipid composition are responsible for the extension of the range over which phase transition occurs. The peak position of the CH2 symmetric stretch also tended to be higher in desiccation-sensitive than in desiccation-tolerant embryos, particularly at the lower temperature range. We attribute this to ¯uidization of the membranes in the injured embryos as a result of the abovementioned changes in lipid composition. Protein secondary pro®le in dried somatic embryos

F I G . 2. Thermotropic response of endogenous lipids in 30-d-old, slowdried, 20 mM ABA-treated (A) and non-ABA-treated (B) alfalfa somatic embryos. ABA-treated embryos were desiccation tolerant, whereas nonABA-treated embryos were not. Wavenumber vs. temperature plots (FTIR) of the CH2 stretching mode of dried (solid circles) and rehydrated (open circles) embryos are shown.

et al., 1993). This means that any upward shift in peak position above ±10 °C can be attributed to the melting of membrane lipids. Embryos cultivated in DTP medium, with or without ABA, and then dried slowly were compared (Fig. 2). In the dried desiccation-tolerant embryos (+ABA; Fig. 2A) melting was complete above 20 °C. Upon rehydration, this occurred at a slightly lower temperature. An upward shift of the temperature range over which melting occurs is a known result of desiccation in model lipid systems and desiccationtolerant organisms (Hoekstra and Golovina, 1999). In contrast, dried, desiccation-sensitive embryos (±ABA; Fig. 2B) showed an extension of the wavenumber shift to considerably above 20 °C, which remained after rehydration. Such a reduced cooperativity of the phase transition has been observed previously in dried, desiccation-sensitive

Dehydration of desiccation-sensitive organisms may result in irreversible denaturation of proteins (Wolkers et al., 1998b). We studied the effects of an ABA treatment during embryo development and the drying regime on the overall protein secondary structure. Embryos were subjected to fast drying over several hours, slow drying or an intermediate rate of drying. Figure 3 depicts the amide-I band of fast-dried embryos at 22, 26 and 30 d after induction of embryogenesis, with or without ABA in the medium. The amide-I band of the desiccation-tolerant (+ABA) embryos hardly changed as embryogenesis progressed (Fig. 3A). The amide-I band pro®le was dominated by a band around 1656 cm±1, indicative of a-helical structures, and a band around 1638 cm±1, indicative of b-sheet and turn structures. During late embryogenesis of non-ABA-treated embryos, a broadening of the amide-I band was observed (Fig. 3B) and the line height ratio between the bands at 1656 cm±1 and 1638 cm±1 decreased as embryogenesis progressed. After drying at an intermediate rate, broader amide-I bands were observed in non-ABA-treated embryos (30-dold) than in ABA-treated embryos (Fig. 4A). However, after slow drying there were major differences in the amide-I band pro®le between ABA-treated and non-ABA-treated embryos (Fig. 4B). The amide-I band pro®le of the ABAtreated, desiccation-tolerant embryos resembled that of the rapidly dried embryos and of the embryos dried at an intermediate rate. In contrast, slow drying of the non-ABAtreated embryos resulted in a major reduction of the a-helical band at 1656 cm±1 and concomitant increases of bands at 1638 and 1688 cm±1, both indicative of extended b-sheet structures. As an aid to interpreting the FTIR spectra in Figs 3 and 4, the relative proportions of protein secondary structures of the differently dried embryos were quanti®ed by curve ®t analysis (Table 4). However, the method does not allow for great precision of the percentages of the different secondary structures. Fast-dried embryos had a higher relative proportion of a-helical structures after culture in ABA-containing DTP medium than after culture in DTP medium devoid of ABA. The overall protein secondary structure of non-ABAtreated embryos after fast and intermediate rates of drying was composed of approx. 30 % a-helical structures and 35 % b-sheet structures. However, when these embryos were dried slowly, the relative proportion of a-helical structures dropped to 20 % and the proportion of aggregated


F I G . 3. Deconvolved FTIR spectra over the region 1700±1600 cm±1 of developing alfalfa somatic embryos during late embryogenesis, cultured with (A) or without (B) 20 mM ABA in DTP medium. At 22, 26 and 30 d after embryo-induction, embryos were rapidly dried within a period of several hours in a stream of dry air (3 % RH) and IR spectra of the dried embryos were recorded.

b-sheet structures increased to 64 %. We suggest that the high percentage of b-sheet structures in these embryos is the result of proteolysis reactions, which lead to the formation of b-sheet structured protein aggregates. Fast and intermediate rates of drying could apparently reduce the formation of protein aggregates in these embryos (Table 4). D IS C U S SIO N Preservation of membrane integrity and cytosolic components during dehydration is of utmost importance in the survival of desiccation-tolerant organisms. Sugars, particularly, are thought to play a role in this preservation

397

F I G . 4. Deconvolved FTIR spectra over the region 1700±1600 cm±1 of 30-d-old alfalfa somatic embryos, cultured with or without 20 mM ABA in DTP medium. The embryos were subjected to drying at an intermediate (A) or slow rate (B), after which IR spectra of the dried embryos were recorded.

(Crowe et al., 1992). Studies on viviparous mutants have shown that ABA is an important endogenous plant hormone for suppressing precocious germination and for the acquisition of desiccation tolerance of seeds (Koornneef et al., 1989). The presence of ABA in the culture medium is also critical in determining the degree of desiccation tolerance of somatic embryos, as is the drying protocol and the stage of embryo-development at which dehydration is started (Table 1). Although ABA in the embryo induction medium did not impart desiccation tolerance, it was effective in this respect after transferring mature embryos to DTP medium, lasting from 22 to 30 dap. The highest rates of germination of ABA-treated embryos followed drying at an intermediate

398


TA B L E 4. Relative peak areas (%) of the composite amide-I band of the dried alfalfa somatic embryos shown in Figs 3 and 4 +ABA (% of peak area)

Wavenumber (cm±1)

Band assignment

1638 1656 1671 1681 1688

b-sheet a-helix Turn Turn b-sheet

±ABA (% of peak area)

Fast drying

Interm. drying

Slow drying

Fast drying

Interm. drying

Slow drying

33 39 14 7 7

33 43 11 9 4

31 35 15 14 4

30 29 28 10 4

31 30 26 9 2

43 20 9 8 21

Embryos were cultivated in DTP medium (total cultivation time 30 d) in the presence or absence of 20 mM ABA and were dried at a fast, slow or intermediate (interm.) rate. Deconvolved amide-I bands (Figs 3 and 4) were curve-®tted with ®ve Pearson band functions to estimate the contributions from the different protein secondary structures.

rate, followed by slow drying; fast drying was the least effective. However, when cultured in the absence of ABA, the slow drying protocol led to complete intolerance to desiccation, whereas some survival occurred at faster drying rates. An interesting effect of ABA is apparent here, in addition to its effect on desiccation tolerance. During slow drying, which also can be considered a form of accelerated ageing treatment, ABA might have reduced cellular damage by curtailing metabolic activity, as has been demonstrated in carrot somatic embryos (Tetteroo et al., 1995b). Thus, somatic embryos could be comparatively more sensitive to slow drying when grown in the absence of ABA than when grown in its presence. The objective of this study was to maximize desiccation tolerance in developing alfalfa somatic embryos rather than to compare different drying protocols. Hence, desiccation tolerance was quantitated by measuring the germination frequency. In addition, intracellular biophysical changes occurring during the acquisition of desiccation tolerance and after subsequent desiccation and rehydration were observed in vivo, using EPR and FTIR spectroscopy. On the basis of EPR spectra of TEMPONE that was administered in a ferricyanide (broadening agent) solution to dried somatic embryos (Fig. 1; Table 3), it is clear that the loss of the capacity to germinate of the slow-dried, nonABA-treated embryos coincides with a considerable loss of plasma membrane integrity (low W/L ratio). In contrast, the ABA-treated embryos with the capacity to germinate had intact plasma membranes. The in vivo FTIR data on membrane ¯uidity (Fig. 2) indicate that these slow-dried ABA- and non-ABA-treated embryos differed in their thermotropic response, in the dried state. A more or less cooperative phase transition was observed in the desiccation-tolerant, ABA-treated embryos. However, the cooperativity was low in the non-ABA-treated, desiccationsensitive embryos, and instead an extended phase transition over a wide temperature range was observed, indicative of lateral phase separations. This difference was observed before as well as after rehydration. Such phase separations are most likely due to the accumulation of lipid peroxides and free fatty acids in the plasma membranes (reviewed by Leprince et al., 1993). Because of the co-existence of liquid

crystalline and gel phase domains, at room temperature membranes become extensively leaky. Massive leaching of ions, phosphates, sugars and soluble proteins can occur upon unrestricted imbibition, particularly at low temperatures. This phenomenon has been described for pollen (Hoekstra and van der Wal, 1988), yeast (Van Steveninck and Ledeboer, 1974), intact seeds (reviewed in Herner, 1986) and isolated mature and immature zygotic embryos (Dasgupta et al., 1982; Adams et al., 1983). Since somatic embryos are naked (not covered by a seed coat), they might be particularly susceptible to imbibitional damage, as has indeed been demonstrated in carrot somatic embryos (Tetteroo et al., 1996). Using EPR we attempted to ®nd evidence for membrane damage during imbibition of desiccation-tolerant and desiccation-intolerant alfalfa somatic embryos. We found the W/L ratios of embryos that were prehydrated for 3 h, a treatment that is known to reduce imbibitional stress, to be higher than those of the non-treated embryos (Table 3). We thus demonstrated that membrane integrity could be enhanced by applying the proper rehydration protocol. It should be noted that following prehydration the germination frequency also was improved to almost 100 % for 30-d-old ABA-treated somatic embryos that were dried at slow and intermediate rates. The protein secondary structure of dried, desiccationsensitive mutant seeds of Arabidopsis thaliana differs from that of desiccation-tolerant wild-type seeds in the ratio of ahelical to b-sheet structures. This was linked to differences in the properties of the cytoplasmic matrix between the mutants and wild-type (Wolkers et al., 1998b). Such an analysis has become possible by in vivo FTIR spectroscopy, which is one of the most powerful techniques to provide information about the secondary structure of proteins in their native environment (Wolkers et al., 1998a, b, 1999). In an attempt to understand the changes occurring during the different culture and desiccation protocols in alfalfa embryos, the secondary structure of the proteins was studied. As embryogenesis proceeded, the proteins in the embryos treated with ABA in DTP medium maintained their a-helical structures relative to the other structures, whereas those in non-ABA-treated embryos showed a slight decrease

Sreedhar et al. Ð Desiccation Tolerance in Alfalfa Somatic Embryos (Fig. 3; Table 4). Proteins in the ABA-treated embryos maintained their predominantly a-helical structure during slow drying. In contrast, slow drying of the non-ABAtreated embryos resulted in a considerable loss of a-helical structures and a concomitant increase in b-sheet structured protein aggregates (Fig. 4). We suggest that the high percentage of b-sheet structures in these embryos is the result of proteolysis reactions. The amide-I band that is observed in vivo is an average of all proteins present in the embryo. However, storage proteins are most likely dominant in the amide-I band, since approx. 30 % of the weight of the embryo at maturity (22 dap) is constituted by storage proteins (Sreedhar and Bewley, 1998). The extent of b-sheet formation in the non-ABA-treated embryos was reduced by changing the drying protocol, for example, by increasing the rate of drying (intermediate rate; Fig. 4B). In addition to serving as carbon reserves for germination, sugars also act as membrane and cytosol protectants during desiccation (Crowe et al., 1987; Williams and Leopold, 1989; Buitink et al., 1998). The lack of desiccation tolerance of the non-ABA-treated embryos after slow drying could have resulted from inadequate sugar protection, as has been found in carrot somatic embryos (reviewed in Hoekstra et al., 2002). However, there was no sign of a decreased content of sugars in these embryos. This might be the result of a continued uptake of sugars during culture in the DTP medium. In carrot somatic embryos, it has been found that the presence of ABA in the medium leads to a cessation of both sugar uptake from the medium and sugar utilization, whereas in the absence of ABA there is active sugar uptake and utilization (reviewed in Hoekstra et al., 2002). Thus, in the non-ABA-treated carrot somatic embryos, high metabolic activity prevails, leading to reactive oxygen species during drying and the associated membrane damage. A similar scenario might pertain to the non-ABA-treated, slow-dried alfalfa somatic embryos, where the loss of membrane integrity and fast ageing rates would concur with an unrestricted metabolism. To summarize, in contrast to desiccation-tolerant alfalfa somatic embryos, desiccation-sensitive embryos have deteriorated membranes and high amounts of b-sheet structured protein aggregates upon drying. ABA appears to be essential during the later stages of somatic embryogenesis to develop full desiccation tolerance. Non-ABAtreated embryos acquire a germinative mode, in which they are much less able to survive desiccation stress, possibly because of an insuf®cient restriction of metabolism. A C K N O W L ED G EM EN T S L.S. was supported by funding from a Commonwealth Fellowship and by a research grant A2210 from the Natural Sciences and Engineering Research Council of Canada to J.D.B. L ITE R A T U R E C IT E D Adams CA, Fjerstad MC, Rinne RW. 1983. Characteristics of soybean seed maturation: necessity for slow dehydration. Crop Science 23: 265±267.

399

Attree SM, Moore D, Sawhney VK, Fowke LC. 1991. Enhanced maturation and desiccation tolerance of white spruce [Picea glauca (Moench) Voss] somatic embryos: effects of a non-plasmolysing water stress and abscisic acid. Annals of Botany 68: 519±525. Blackman SA, Obendorf RL, Leopold AC. 1992. Maturation proteins and sugars in desiccation tolerance of developing soybean seeds. Plant Physiology 100: 225±230. Bomal C, Tremblay FM. 1999. Effect of desiccation to low moisture content on germination, synchronization of root emergence, and plantlet regeneration of black spruce somatic embryos. Plant Cell, Tissue and Organ Culture 56: 193±200. Bomal C, Tremblay FM. 2000. Dried cryopreserved somatic embryos of two Picea species provide suitable material for direct plantlet regeneration and germplasm storage. Annals of Botany 86: 177±183. Buitink J, Walters C, Hoekstra FA, Crane J. 1998. Storage behavior of Typha latifolia pollen at low water contents: interpretation on the basis of water activity and glass concepts. Physiologia Plantarum 103: 145±153. Byler DM, Susi H. 1986. Examination of the secondary structure of proteins by deconvolved FTIR spectra. Biopolymers 25: 469±487. Crowe JH, Hoekstra FA, Crowe LM. 1992. Anhydrobiosis. Annual Review of Physiology 54: 570±599. Crowe JH, Crowe LM, Carpenter JF, Aurell Wistrom C. 1987. Stabilization of dry phospholipid bilayers and proteins by sugars. Biochemical Journal 242: 1±10. Dasgupta J, Bewley JD, Yeung EC. 1982. Desiccation-tolerant and desiccation-intolerant stages during the development and germination of Phaseolus vulgaris seeds. Journal of Experimental Botany 33: 1045±1057. DeBruijn SM, Ooms JJJ, Karssen CM, Vreugdenhil D. 1997. Effects of abscisic acid on reserve deposition in developing Arabidopsis seeds. Acta Botanica Neerlandica 46: 263±277. Galau GA, Bijaisoradat N, Hughes DW. 1987. Accumulation kinetics of cotton late embryogenesis-abundant mRNAs and storage protein mRNAs: coordinate regulation during embryogenesis and the role of abscisic acid. Developmental Biology 123: 198±212. Golovina EA, Tikhonov AN. 1994. The structural differences between the embryos of viable and nonviable wheat seeds as studied with the EPR spectroscopy of lipid-soluble spin labels. Biochimica et Biophysica Acta 1190: 385±392. Golovina EA, Tikhonov AN, Hoekstra FA. 1997. An electron paramagnetic resonance spin-probe study of membranepermeability changes with seed aging. Plant Physiology 114: 383± 389. Gray DJ. 1989. Effects of dehydration and exogenous growth regulators on dormancy, quiescence and germination of grape somatic embryos. In Vitro Cellular and Developmental Biology 25P: 1173±1178. Herner RC. 1986. Germination under cold soil conditions. HortScience 21: 1118±1122. Hoekstra FA, Golovina EA. 1999. Membrane behavior during dehydration: implications for desiccation tolerance. Russian Journal of Plant Physiology 46: 295±306. Hoekstra FA, van der Wal EW. 1988. Initial moisture content and temperature of imbibition determine extent of imbibitional injury in pollen. Journal of Plant Physiology 133: 257±262. Hoekstra FA, Bomal C, Tetteroo FAA. 1993. Membrane phase transitions in dry seeds as measured by FT-IR spectroscopy. In: CoÃme D, Corbineau F, eds. Proceedings of the fourth international workshop on seeds: Basic and applied aspects of seed biology. Paris: ASFIS, 755±762. Hoekstra FA, Crowe JH, Crowe LM. 1991. Effect of sucrose on phase behaviour of membranes in intact pollen of Typha latifolia L. as measured with Fourier transform infrared spectroscopy. Plant Physiology 97: 1073±1079. Hoekstra FA, Golovina EA, Tetteroo FAA, Wolkers WF. 2002. Induction of desiccation tolerance in plant somatic embryos: how exclusive is the protective role of sugars? Cryobiology (in press). Hoekstra FA, Golovina EA, van Aelst AC, Hemminga MA. 1999. Imbibitional leakage from anhydrobiotes revisited. Plant, Cell and Environment 22: 1121±1131. Iida Y, Watabe K, Kamada H, Harada H. 1992. Effects of abscisic acid on the induction of desiccation tolerance in carrot somatic embryos. Journal of Plant Physiology 140: 356±360.

400


Koornneef M, Hanhart CJ, Hilhorst HWM, Karssen CM. 1989. In vivo inhibition of seed development and reserve protein accumulation in recombinants of abscisic acid biosynthesis and responsiveness mutants in Arabidopsis thaliana. Plant Physiology 90: 463±469. Krochko JE, Pramanik SK, Bewley JD. 1992. Contrasting storage protein synthesis and messenger RNA accumulation during development of zygotic and somatic embryos of alfalfa (Medicago sativa L.). Plant Physiology 99: 46±53. Lai FM, McKersie BD. 1994. Regulation of starch and protein accumulation in alfalfa (Medicago sativa L.) somatic embryos. Plant Science 100: 211±219. Lecouteux C, Tessereau H, Florin B, Courtois D, Petiard V. 1992. Preservation of somatic embryos of carrot (Daucus carota L.) by dehydration after pretreatment with sucrose. Comptes Rendus de l'AcadeÂmie des Sciences Paris 314: 423±428. Leprince O, Hendry GAF, McKersie BD. 1993. The mechanisms of desiccation tolerance in developing seeds. Seed Science Research 3: 231±246. McKersie BD, Senaratna T, Bowley SR, Brown DCW, Krochko JE, Bewley JD. 1989. Application of arti®cial seed technology in the production of hybrid alfalfa (Medicago sativa L.). In Vitro Cellular and Developmental Biology 26P: 1183±1188. Marsolais AA, Wilson DPM, Tsujita MJ, Senaratna T. 1991. Somatic embryogenesis and arti®cial seed production in zonal (Pelargonium 3 hortorum) and regal (Pelargonium 3 domesticum) geranium. Canadian Journal of Botany 69: 1188±1193. Miller RW. 1978. Osmotically induced removal of water from fungal cells as determined by a spin probe technique. Plant Physiology 62: 741± 745. Parrott WA, Dryden G, Vogt S, Hildebrand DF, Collins GB, Williams EG. 1988. Optimization of somatic embryogenesis and embryo germination in soyabean. In Vitro Cellular and Developmental Biology 24P: 817±820. Prestrelski SJ, Tedeschi N, Arakawa T, Carpenter JF. 1993. Dehydration-induced conformational transitions in protein and their inhibition by stabilizers. Biophysical Journal 65: 661±671. Roberts DR, Sutton BCS, Flinn BS. 1990. Synchronous and high frequency germination of interior spruce somatic embryos following partial drying at high relative humidity. Canadian Journal of Botany 68: 1086±1090. Schenk RV, Hildebrandt AC. 1972. Medium and techniques for induction and growth of monocotyledonous and dicotyledonous plant cell cultures. Canadian Journal of Botany 50: 199±204. Senaratna T, McKersie BD, Bowley SR. 1989. Desiccation tolerance of alfalfa (Medicago sativa L.) somatic embryos: In¯uence of abscisic acid, stress pretreatments and drying rates. Plant Science 65: 253± 260. Sreedhar L, Bewley JD. 1998. Nitrogen- and sulfur-containing

compounds enhance the synthesis of storage reserves in developing somatic embryos of alfalfa (Medicago sativa L.). Plant Science 134: 31±44. Steadman KJ, Pritchard HW, Dey PM. 1996. Tissue-speci®c soluble sugars in seeds as indicators of storage category. Annals of Botany 77: 667±674. Surewicz WK, Mantsch HH. 1988. New insight into protein secondary structure from resolution-enhanced infrared spectra. Biochimica et Biophysica Acta 952: 115±130. Tetteroo FAA, Hoekstra FA, Karssen CM. 1995a. Induction of complete desiccation tolerance in carrot (Daucus carota) embryoids. Journal of Plant Physiology 145: 349±356. Tetteroo FAA, Peters AHLJ, Hoekstra FA, Van der Plas LHW, Hagendoorn MJM. 1995b. ABA reduces respiration and sugar metabolism in developing carrot (Daucus carota L.) embryoids. Journal of Plant Physiology 145: 477±482. Tetteroo FAA, de Bruijn AY, Henselmans RNM, Wolkers WF, van Aelst AC, Hoekstra FA. 1996. Characterization of membrane properties in desiccation-tolerant and -intolerant carrot somatic embryos. Plant Physiology 111: 403±412. VanBilsen DGJL, Hoekstra FA, Crowe LM, Crowe JH. 1994. Altered phase behavior in membranes of aging dry pollen may cause imbibitional leakage. Plant Physiology 104: 1193±1199. VanSteveninck J, Ledeboer AM. 1974. Phase transitions in the yeast cell membrane: the in¯uence of temperature on the reconstitution of active dry yeast. Biochimica et Biophysica Acta 352: 64±70. Williams RJ, Leopold AC. 1989. The glassy state in corn embryos. Plant Physiology 89: 977±981. Wolkers WF, Hoekstra FA. 1995. Aging of dry desiccation-tolerant pollen does not affect protein secondary structure. Plant Physiology 109: 907±915. Wolkers WF, Van Kilsdonk MG, Hoekstra FA. 1998a. Dehydrationinduced conformational changes of poly-L-lysine as in¯uenced by drying rate and carbohydrates. Biochimica et Biophysica Acta 1425: 127±136. Wolkers WF, Tetteroo FAA, Alberda M, Hoekstra FA. 1999. Changed properties of the cytoplasmic matrix associated with desiccation tolerance of dried carrot somatic embryos. An in situ Fourier transform infrared spectroscopic study. Plant Physiology 120: 153± 163. Wolkers WF, Alberda M, Koornneef M, LeÂon-Kloosterziel KM, Hoekstra FA. 1998b. Properties of proteins and the glassy matrix in maturation-defective mutant seeds of Arabidopsis thaliana. Plant Journal 16: 133±143. Wolkers WF, McCready S, Brandt W, Lindsey GG, Hoekstra FA. 2001. Isolation and characterization of a D-7 LEA protein from pollen that stabilizes glasses in vitro. Biochimica et Biophysica Acta 1544: 196±206.