Co-action of temperature and phosphate in inducing turion formation in Spirodela polyrhiza (Great duckweed)


  • K.-J. Appenroth

    Corresponding author
    1. Institut für Allgemeine Botanik und Pflanzenphysiologie, Friedrich-Schiller-Universität Jena, D-07743 Jena, Germany
    Search for more papers by this author

  • *Dedicated to my academic teacher Professor Dr Helmut Augsten.

Klaus-J. Appenroth. Fax: +49 364 194 9232; e-mail:


Increased phosphate concentration, higher temperature and addition of glucose all increased the number of fronds and turions of the duckweed Spirodela polyrhiza formed under in vitro conditions. Increasing the number of turions by increasing the plant biomass does not mean that the developmental process (switch of the programme of the primordia from vegetative fronds toward resting turions) has been specifically influenced. The specific turion yield (STY; number of turions formed by one frond) and the time of onset of turion formation have been used as more specific measures of turion induction. At more than 30 µm initial phosphate the STY was increased by lower temperature (15 °C) and became independent of the phosphate concentration. Between 10 and 30 µm and at higher temperatures (25  °C) the STY was increased by lower phosphate levels. The stimulatory effect of lower temperature was more pronounced than that of lower phosphate concentrations. Decreased phosphate concentration highly accelerated the formation of the first turions. The influence of low temperature was small at lower phosphate concentration but became dominant at higher concentrations (especially in autotrophic cultures). Low phosphate levels (e.g. 10 µm) and low temperatures (e.g. 15 °C) both represent specific turion-inducing factors having significant interactive effects. In S. polyrhiza, these signals may replace the interactive effects of photoperiods and low temperature known from other hydrophytes in turion induction under natural conditions.


specific turion yield.


Hydrophytes are water plants with perennating buds below the water surface (Bartley & Spence 1987). Such plants show a strong tendency for vegetative propagation. A range of devices, such as rhizomes, root tubers, stem tubers or turions are employed by hydrophyte species to sustain themselves through periods that are unfavourable to growth. The environmental control of induction of vegetative propagules has been investigated in a broad spectrum of hydrophytes, e.g. Potamogeton pectinatus (van Wijk 1989), Potamogeton crispus (Chambers, Spence & Weeks 1985), Myriophyllum verticillatum (Weber & Nooden 1976), Utricularia sp. (Adamec 1999) and Hydrilla verticillata (Thullen 1990; Netherland 1997). The major environmental factors controlling the formation of such propagules are (Bartley & Spence 1987): water temperature, photoperiod, nutrient availability, light quality and photon fluence rate. In many species more than one factor controls the switch of the developmental programme from vegetative propagation to the formation of resting propagules. The photoperiod is of special importance because this signal permits the anticipation of seasonal changes of growth conditions (Netherland 1997).

This paper focuses on investigations of Spirodela polyrhiza, a member of Lemnaceae (duckweeds). Duckweeds represent a family of floating hydrophytes (Liliopsida; Les et al. 2002) characterized by a special mode of vegetative propagation (Landolt 1986; Landolt & Kandeler 1987). Under conditions of normal vegetative growth, mother fronds (F0) produce daughter fronds (F1) from two meristematic pockets at their proximal end. The daughter fronds themselves contain meristematic pockets from which further daughter fronds (F2) develop leading to clonally related plants. In 15 out of 37 species (Landolt 1998) the primordia can follow a second, specialized developmental path forming turions (Smart & Fleming 1993) or turion-like fronds (Landolt & Kandeler 1987) instead of normal vegetative fronds. Turions of S. polyrhiza belong to the ‘true turions’ because they are morphologically different from normal fronds, sink to the bottom of the water body, and do not grow any further (Landolt 1986). They are distinguished from normal fronds by smaller size, lack of aerenchyma, thicker cell wall (Jacobs 1947; Smart & Trewavas 1983; Appenroth & Bergfeld 1993) and accumulation of starch (Henssen 1954; Appenroth & Gabrys 2001). Turions also contain two meristematic pockets from which new vegetative fronds can develop following germination (Appenroth & Bergfeld 1993). Turions play an important role in the survival strategy of this species (Landolt & Kandeler 1987) because fronds cannot tolerate low temperatures, and thus die during late autumn in temperate climate. In spring, after the temperature rises, turions germinate in a phytochrome-mediated response (Appenroth & Augsten 1990). The vegetative mode of reproduction, mixotrophic and fast growth make S. polyrhiza a superior system for the study of dormant organ induction in a photosynthetic organism.

The developmental programme to form turions is induced by external signals, some of them representing environmental conditions which are unfavourable for vegetative growth. In experiments with S. polyrhiza under controlled laboratory conditions, the following turion-inducing factors have been revealed: low temperature (Perry 1968), abscisic acid (Steward 1969; Smart et al. 1995), shortage of nutrients such as phosphate (Appenroth et al. 1989), nitrate (Sibasaki & Oda 1979; Malek & Cossins 1983) or sulphate (Malek & Cossins 1983), exogenously applied sugars (Henssen 1954; Newton et al. 1978) and high light intensity (Jacobs 1947; Appenroth, Teller & Horn 1996). The effect of light is partially mediated by phytochrome (Appenroth, Hertel & Augsten 1990) and a specific blue light receptor is involved under autotrophic conditions (Appenroth et al. 1996).

In contrast to the results in other hydrophytes, turion formation in S. polyrhiza is not controlled by the daily light period. Extending the duration of illumination from 4 to 24 h per day resulted in a linear increase of the turion yield, but no critical day length, usually crucial for photoperiodic responses, was detected (Appenroth et al. 1990). Therefore, co-action of the effects of short days and low temperature (the most common mechanism for the induction of dormant propagules in hydrophytes) cannot be responsible for the induction of turions in S. polyrhiza.

The occurrence of duckweed species has been followed for several years (1993–98) in the artificial water reservoir ‘Bleilochtalsperre’ in the German federal state of Thuringia (Loth et al. 1995). The results of these research projects indicated that turions of S. polyrhiza might be induced: (1) by low temperature around 15 °C regardless of phosphate concentrations; and (2) by limitation of phosphate at temperatures higher than 19 °C (H.-P. Liebert, University of Jena, personal comm.). Therefore we decided to test the following hypothesis under laboratory conditions: at 15 °C the influence of phosphate is not significant and thus turion formation is induced by decreasing temperature. At higher temperature (20 °C, 25 °C) limiting phosphate concentrations becomes important for the induction of turion formation. On a physiological level we addressed the question of which external signal(s) switch(es) the developmental programme of the primordia in the two meristematic pockets from the formation of daughter fronds to the formation of turions. As shown before for the effect of abiotic stress on turion formation (Srivastava & Jaiswal 1989), this developmental switch is best indicated by the specific turion yield (STY) defined as the ratio of the number of turions to the number of fronds. In general, external factors may provide good growth conditions, which may subsequently (after turion induction) lead to the production of large number of turions. The total turion yield may thus indicate the indirect effects of external factors and may be of limited value in evaluating the significance of turion-inducing factors. Therefore we have also calculated the STY and determined the time of the onset of turion formation under defined conditions of temperature and phosphate concentrations to account for specific turion-inducing factors.

Materials and methods

Plant material and nutrient media

Fronds of Spirodela polyrhiza (L.) Schleiden, strain SJ, were cultivated under axenic conditions as described (Appenroth et al. 1996). The standard nutrient solution contained 8 mm KNO3, 1 mm Ca(NO3)2, 1 mm MgSO4, 25 µm Fe(III)EDTA, 5 µm H3BO3, 13 µm MnCl2 and 0·4 µm Na2MoO4 (Appenroth et al. 1996). The phosphate concentration (in standard experiments 60 µm KH2PO4) was varied between 0 and 120 µm (0, 4, 10, 30, 60 or 120 µm) when indicated. For mixotrophic cultivation 50 mm d-glucose was added. 300 mL Erlenmeyer flasks containing 180 mL nutrient solution were used for plant cultivation.

Irradiation conditions

White fluorescence tubes TLD 18 W/86 (Philips, Eindhoven, The Netherlands) were used within temperature-controlled growth chambers. Photosynthetic active radiation was measured with the radiometer LI 250 (Li-Cor, Lincoln, NB, USA). The growth rate increased with photosynthetic active radiation augmented from 35 to 90 µmol m−2 s−1, but decreased at 300 µmol m−2 s−1. At the highest fluence rate used, the plants formed high amounts of anthocyane indicating light stress (not shown). For technical reasons we have chosen 60 µmol m−2 s−1 continuous white light (approximately 9 W m−2 total fluence rate) for all further experiments.

Plant cultivation

Plants were pre-cultivated for 8 d in solutions containing 60 µm phosphate with or without d-glucose (50 mm) for subsequent mixotrophic cultivation or for subsequent autotrophic cultivation, respectively. Temperature was kept at 25 ± 0·1 °C. Two three-frond colonies, obtained from this pre-cultivation were then used as starting material for experiments. In order to acclimate the plants to the changed nutrient solution, these cultures were kept for 4 d at 25 ± 0·1 °C and thereafter transferred to 15 ± 0·1 °C or 20 ± 0·1 °C or kept at 25 ± 0·1 °C. The day when the cultures were transferred to the three different temperatures was defined as time zero. The cultures were inspected twice a day for possible formation of turions. The time when the first immature turions became visible, still attached to the mother frond, was taken as the onset of turion formation. The plants were cultivated for 28 d until no further formation of turions could be expected. Fronds and turions, both immature and mature (i.e. still attached to mother plants or already at the bottom of the Erlenmeyer flasks) were then counted. The STY was defined as the number of turions formed by one vegetative frond.


Each data point represents the results of five independent samples evaluated in two independent experiments (total: n = 5). The results are given as means ± standard error. When required, the significance of the difference of averages was assessed using the two-sided Student's t-test (5% level). Growth rates (k) were estimated by non-linear least squares fitting of the equation Nt = Nt = 0 exp(kt) to the observed number of fronds (N) at time t. In these experiments the number of fronds was measured each day for 7 d. The significance of the influence of phosphate concentrations and temperature on frond number, turion number and STY was tested by the univariate analysis of variance, anova (Underwood 1997). To improve the homogeneity of variances the data was square root- or log-transformed before analysis. For parallel samples, the onset of turion formation proceeded usually on the same day. Therefore, no errors were given in Fig. 3.

Figure 3.

Influence of initial phosphate concentration at three different temperatures on the onset of turion formation in Spirodela polyrhiza. Turions were counted as immature when still connected with the mother plants. No errors were given because parallel samples formed first visible turions normally on the same day.


The influence of phosphate concentrations at different temperatures on the number of fronds, turions, and on the STY is shown in Fig. 1 (autotrophic cultivation) and Fig. 2 (mixotrophic cultivation). The results of anova analysis demonstrated significant effects of the two independent parameters, temperature and phosphate concentration, on all three measured parameters, i.e. number of fronds, turions and STY. It also demonstrated strong interaction between temperature and phosphate concentration in the control of growth and development of S. polyrhiza. This interaction was observable in autotrophic as well as mixotrophic cultivation conditions (not shown).

Figure 1.

Influence of initial phosphate concentration and temperature on different yield parameters of Spirodela polyrhiza under autotrophic conditions. (a, b), Frond number per flask; (c, d), turion number per flask; (e, f), turion number divided by frond number. After pre-cultivation (see Materials and methods) plants were kept in continuous white light and number of fronds and turions were evaluated 28 d after transfer to the conditions indicated. Both mature and immature turions were considered.

Figure 2.

Influence of initial phosphate concentration and temperature on different yield parameters of Spirodela polyrhiza under mixotrophic conditions. For further explanations see Fig. 1.

The number of fronds increased with increasing phosphate concentration (Fig. 1a). At the highest initial concentration (120 µm) the response apparently did not reach saturation. Increasing the temperature from 15 to 20 or 25 °C also increased the number of fronds. Glucose was added to simulate the presence of metabolizable organic compounds in aquatic environment (Fig. 2). The effect of phosphate was enhanced in the presence of glucose (Fig. 2a) and the number of fronds was increased at all three temperatures (Fig. 2b). The influence of the phosphate concentrations (PHOS) and temperature (TEMP) on the number of fronds (FN) formed under autotrophic and mixotrophic conditions can be described by the following linear equations:


inline image


inline image

The same set of independent variables (phosphate concentrations, temperature) was used under autotrophic and mixotrophic conditions. Therefore, it can be concluded that under mixotrophic conditions the relative importance of the phosphate concentration increased in comparison with temperature. Similar equations to those described above cannot be derived for the formation of turions or the STY because of the non-linear relationships.

The number of turions produced after 28 d was increased by increasing phosphate concentration and increasing temperature (Fig. 1c & d). The addition of glucose also increased the number of turions formed (Fig. 2c & d).

Under autotrophic conditions at 25 and 20 °C, STY had maxima at approximately 10 µm phosphate (Fig. 1e). Above this optimum, the STY decreased with increasing phosphate concentration. At 15 °C the stimulating effect of lower phosphate concentrations was not apparent. The decreasing of temperature resulted in increased STY. The highest value of 3·3 turions per frond was obtained at 15 °C. Figure 1f shows that the stimulation effect of lower temperature can only be observed at higher phosphate concentrations. Without applied phosphate approximately one turion was formed per frond regardless of the temperature. Under mixotrophic conditions, the stimulating effect of lower temperature on STY (Fig. 2e & f) required higher phosphate concentrations and could not be observed at 10 and 4 µm or without phosphate. At 25 °C lower concentrations of phosphate strongly enhanced STY, in the investigated range by the factor of 2·3 (Fig. 2e). At lower temperatures (20 °C, 15 °C; Fig. 2e) the stimulatory effect of low phosphate concentrations was barely detectable.

Decreased phosphate concentrations clearly accelerated the onset of turion formation (Fig. 3). Whereas the difference between the effect at 25 and 20 °C was very small, 15 °C had a strong effect in decreasing the time before turion formation was first observed. Turion formation started later in mixotrophic conditions, particularly at low phosphate and low temperature (cf. Fig. 3a & b).


The yield of green fronds as well as the turion yield of S. polyrhiza increased with increasing phosphate concentration and with increasing temperature. The application of glucose had a further enhancing effect. These results can be expected because the yield of the survival organs depends on the yield of green fronds from which they are produced. This means turion yield is indirectly influenced by phosphate, temperature and glucose via increasing the total yield (biomass) of the system. In order to find out whether these exogenous factors have also specific effects on turion formation, the value of STY has been considered as a possible criterion.

In hydrophytes, the developmental switch from the formation of fronds to the formation of resting turions is inducible by chemical/metabolic effectors (salts, sugars), as well as physical/environmental (photoperiod, light quality, fluence rate, temperature) signals. The most common mechanism is the induction by photoperiod and low temperature (Bartley & Spence 1987). As an example, short days were required by Hydrilla verticillata (Steward 2000) and 20 short day periods were necessary to induce turion formation (Thakore, Haller & Shilling 1997). The paper of Perry (1968) is often cited to show the influence of a short-day treatment also in S. polyrhiza (e.g. Smart 1996). However, analysis of the data published so far does not support this interpretation. Perry (1968) used four different clones, and, in each experiment, the temperature or nitrate supply were changed simultaneously with different photoperiods. Both factors are known to also influence turion formation (Malek & Cossins 1983; Appenroth et al. 1989; Smart & Fleming 1993). Spirodela polyrhiza, strain SJ has been characterized as a day-neutral plant in turion formation (Appenroth et al. 1990). Consequently, factors other than photoperiod must be responsible for the induction of turion formation in this species. Our initial hypothesis, derived from outdoor experiments (see Introduction), favours the influence of phosphate concentration and temperature as the dominating factors inducing the developmental switch to form turions. This was confirmed by current experiments and a specific co-action of both factors was demonstrated by statistical analysis (anova) whereas the meaning of other possible factors, e.g. nitrate or sulphate concentrations, has not been excluded.

Phosphate starvation became a well-investigated stress response in plants (Plaxton & Carswell 1999). As occurs with many other environmentally mediated perturbations, phosphate stress results in modified gene expression. A set of vegetative storage protein genes are expressed only at low levels of phosphate (Sadka et al. 1994). Phosphate deficiency has often been associated with starch accumulation. In Lemna gibba and Phaseolus vulgaris, the response to phosphate deficiency is characterized by starch accumulation and has been associated with high activity of the pathway of sucrose synthesis (Thorsteinsson & Tillberg 1987; Ciereszko & Barbachowska 2000). A rapid increase in the translocation of assimilated carbon from shoot to root has been observed during phosphate deficiency (Ciereszko et al. 1996). Such effects of low phosphate concentrations on the carbohydrate metabolism and partitioning may also be involved in the formation of turions, which are starch storage organs (Appenroth & Gabrys 2001). Lower levels of phosphate in the nutrient solutions increased the STY in S. polyrhiza at 25 °C but not at 15 °C. This effect was restricted to a concentration window of phosphate, especially under autotrophic conditions. The STY represents an average over the whole cultivation period. Most probably, very low phosphate concentrations are not sufficient to support the formation of turions over the whole period of cultivation. The effect of the phosphate level was most evident with the time of turion emergence taken as a measure (Fig. 3). The first turions emerged faster when the phosphate concentration was lower. The influence of phosphate was detectable at all three temperatures studied. The onset of turion formation depends less on the capacity of biomass formation and therefore better represents specific effects of environmental factors on turion formation. It can be concluded that the limitation of phosphate is a specific inducing factor of turion formation.

Plants perceive low temperature and respond with changes in membrane fluidity, cytoskeleton rearrangement and calcium influxes (Browse & Xin 2001). Cold acclimation and freezing tolerance are often connected with changes in carbohydrate spectrum (Palonen, Buszard & Donnelly 2000). The molecular mechanism of cold treatment on the induction of turion formation in S. polyrhiza is not yet clear. The STY was mainly defined by the effect of temperature when the initial phosphate concentration was higher than 30 µm. Turion formation is apparently induced by low temperature. This conclusion is further supported by the effect of temperature on the time of onset of turion formation. Here, the difference between the effect of 25 and 20 °C was small but 15 °C had a specific inducing effect. At lower phosphate concentrations this specific effect of low temperature was less pronounced. At low phosphate levels (10 µm and lower) the STY depends only to a small degree on temperature under autotrophic conditions; under mixotrophic conditions, the highest STY values were unexpectedly obtained at 25 °C. The results concerning the influence of temperature and phosphate concentration on the time of onset of turion formation (Fig. 1) confirm these conclusions. Thus, temperature has opposite effects on the total turion yield on the one side, and on the specific turion yield as well as the time of onset of turion formation on the other side. This supports the idea that indirect effects via biomass formation have to be distinguished from direct inducing effects on turion formation. Low temperature is clearly a specific turion-inducing factor.

Our initial hypothesis can now be formulated more precisely: At 15 °C, temperature is the dominating turion-inducing factor but this process requires sufficient phosphate (at least initially 30 µm). At higher temperatures (25 °C) the turion-inducing effect of low phosphate levels is obvious, especially under mixotrophic conditions.

Different sugars have often been described as turion-inducing factors (e.g. Henssen 1954; Newton et al. 1978). However, this conclusion was based on the total turion yield. Whereas the number of fronds and the number of turions was strongly increased by mixotrophic cultivation (cf. Figs 1 & 2), the ratios STY(mixotrophic)/STY (autotrophic), especially at 15 and 20 °C, are close to 1 (data not shown). These results exclude a specific effect of glucose on turion formation.

Landolt (1986) reviewed several turion-inducing factors in duckweeds, which can now be divided into two classes. Some of them (class 1) affect the turion number via the total yield of the system (most probably sugars, high photon fluence rates or high CO2). Class 2 contains those factors, which serve as signals to switch the developmental programme of the primordia from vegetative fronds to resting turions (most probably low temperature, low levels of phosphate, nitrate, or sulphate, abscisic acid and photomorphogenic effects of light via phytochrome or blue light receptors).


The author is grateful for the skilful technical assistance of Ms G. Lenk, the discussion of Professor H. Gabrys, Jagiellonian University of Krakow, Poland, the personal communication of Dr H.-P. Liebert, University of Jena, Germany, and the statistical advice of Dr J. Schumacher, University of Jena. I thank Mike Chick, CELT Cardiff, Wales for linguistic revision of the manuscript and Dr L. Adamec, Trebon, Czech Republic, for his support.