P.M. Strózycki, A. Skapska, K. Kolaczkowska-Szczesniak, E. Sobieszczuk, A.B. Legocki
Institute of Bioorganie Chemistry Polish Academy of Sciences, Noskowskiego 12/14, Pozna 61-704, Poland
Iron is one of the most important nutrients for all eukaryotes. However, it is also one of the most dangerous elements. Because of its redox properties, iron is a critical element for such basic processes as DNA and hormone synthesis, respiration and photosynthesis (Briat et al. 1995). Although iron is the fourth most abundant element in the Earth crust, it is not easily available. This is because of low solubility of iron-containing minerals, especially in aerobic and neutral pH environments (Guerinot, Yi 1994). In order to cope with this problem, plants have developed several mechanisms of iron acquisition. Except for morphological changes leading to the extension of active root area, these mechanisms include proton pumping (acidification), secretion of organic acids and phenolics (chelation) and induction of the membrane bound reductase (dicots and nongraminaceous monocots). In addition, high-affinity chelators are used to dissolve Fe(EH) oxides (Fox, Guerinot 1998; Hinsinger 1998; Jones 1998; Marschner, Romheld 1996; Schmidt 1999; Thoiron et al. 1997). However, in the presence of active oxygen species the same desirable iron may catalyze the generation of hydroxyl radicals (OH*) - the most powerful oxidizing agents known thus far (Fenton reaction) (Cadenas 1989; Meneghini et al. 1995; Nappi, Vass 2000). Attack of toxic oxygen species usually leads to severe results like lipid peroxidation, protein and DNA oxidation and eventually cell disintegration. Plant antioxidant defenses include such compounds like glutathione, ascorbate, carotenoids, tocopherols, etc. Antioxidant enzymes like catalase, peroxidases, dismutases and enzymes of the ascorbate-glutathione cycle are also activated (Becana et al. 1998; Larson 1995). It is obvious that there has to be a strict control of "free" iron in the cell, just to prevent generation of reactive oxygen species (ROS) in the first place. Limited generation of ROS, however, may be a part of plant defense systems against pathogens (Wojtaszek 1997). Predominant portion of cell iron is bound in proteins and enzyme cofactors, but it may easily be "set free" under stress situations.
Ferritins, iron storage proteins, are the major and the most effective elements necessary for iron homeostasis control. A ferritin molecule is a hollow protein shell, composed of 24 polypeptide subunits, capable of storing up to 4500 iron atoms in hydrous ferric oxide form (Harrison, Arosio 1996; Trikha et al. 1994). Thus iron biomineralized inside ferritin is safe for cell components. It may be released when needed for their functioning (Moore et al. 1992). Ferritins are widespread in all kingdoms of living organisms, animals, plants and microorganisms, showing high structural conservation (Theil 1987). Amino acid identities are lower except of residues required for a proper structure of the polypeptide (Andrews et al. 1992). Plant ferritin polypeptides consist of two additional (compared to animal ferritins) fragments. One, the so-called transit peptide (TP; 47-91 amino acids) is required for plastid targeting of ferritin precursor, the other, extension peptide (EP; 28-32 amino acids) is probably responsible for the stability of matured protein (Briat et al. 1999; Ragland et al. 1990).
As practically legume nitrogen fixation relies on iron (two of the key elements of nitrogen reduction, plant hemoglobin and bacterial nitrogenase, are iron-proteins), strict control of this metal during nodule development is particularly important. Moreover, leghemoglobin accounts for up to 30% of total soluble proteins in the nodule and has to be considered as a source of dangerous radicals (Moreau et al. 1996).
In our studies on iron management in plant tissues, we use yellow lupin (Lupinus luteus) as a model. Based on the plant hemoglobin sequence analysis we have assumed that lupins belong to one of the oldest species among legume plants (Kass, Wink 1995; Strózycki et al. 2000; Strózycki, Legocki 1995). Lupin plants infected with Bradyrhizobium lupini form nodules of a very characteristic type. As far as the general morphology is concerned, a lupinoid nodule is similar to that of undeterminate type - long-lasting meristems and clearly divided developmental zones. However, because of very early division of nodule meristem and its growth in all directions, it escapes from the cylindrical shape. During further development, meristematic activity remains only in nodule edges, and as a result of the activity of two lateral meristems, the initially spherical nodules grow laterally encircling the root (Golinowski et al. 1987). Because of the shape, this type of nodule is called a "collar type". Additionally, the first cell divisions that initiate nodule formation in lupin take place in the first layer of the primary root cortex, similar to the determinate type nodules, i.e. soybean or beans, and not in the deepest layers of the primary root cortex, which is typical of all other undeterminate-nodule type plants, i.e. pea, clover or alfalfa (Golinowski et al. 1992).
During lupin root nodule development, low "physiological" levels of ferritin polypeptide may be detected. There is a significant temporary increase in ferritin accumulation around the time when nitrogen fixation starts. This increase is correlated in time with massive synthesis of leghemoglobin (12-14 days after inoculation with B. lupini). In fact, strong accumulation of ferritin polypeptides starts with the first symptoms of nodule tissue decay and increases with nodule senescence. This picture of developmental regulation was also confirmed by RNA hybridization experiments and by in situ immuno detection. Apart from the clearly stronger hybridization signal on material from 14-day-old nodules, there is a distinct layer of cells just between the meristematic and matured nodule zones, characterized by a higher ferritin content. It is probably the equivalent of amyloplast containing interzone (or pre nitrogen-fixation zone) of alfalfa. It is also apparent in lupin senescing nodules that the main increase of ferritin signal is shifted to cortex and younger zones, close to meristems. It could be explained as protection of the still active and functional tissues against catalytic iron released from spreading "destruction". In addition, there is a strong increase in ferritin accumulation in root tissues surrounding vascular bundles and even after complete disintegration of a nodule, the adjacent root tissue seems to remain intact. It is likely that the main task of the antioxidant defenses of bacteroid tissues during nodule functioning and senescence repose on other systems (Becana et al. 2000). However, as it was reported before, a majority of the diverse antioxidant systems functioning in an active nodule decline with the aging process (Becana et al. 2000).
Observations presented above are generally in agreement with those of other authors (Lucas et al. 1998) describing ultra structural immuno-localization of ferritin in the cells of soybean, lupin and alfalfa nodules. The exception is the finding that meristematic activity was not observed at all stages of yellow lupin nodule development. According to our observations, meristems exist in L. luteus even in 96-day-old nodules (with clear mitotic figures).
Following the reports on ferritin accumulation in the cells of cortex and parenchyma in nodules of nitrate and dark stressed pea and bean plants, we checked a quantitative contribution of this protein to the general reaction in this type of induced senescence (Escuredo et al. 1996; Gogorcena et al. 1997). Surprisingly, there was no detectable change in ferritin accumulation levels in stems, roots or nodules of either nitrate or dark treated plants. Moreover, in contrast to the analysis performed for bean nodules (Matamoros et al. 1999), a slight decrease was noticed in the leaves of plants kept in the dark for 4-7 days.
To have a complete picture of lupin ferritin, we analyzed ferritin content in other than nodule parts of the unstressed plant. In lupin, we could detect ferritin polypeptide in all tested organs. The only exceptions were cotyledons during plant germination and young, forming seeds. The former is due to the fact that cotyledons are rather a source of iron (for a growing plant) and ferritin is degraded to release it. The latter is due to the fact that ferritin is synthesized at a later stage of seed development. The highest levels of ferritin were found in flowers and leaves. Flowers contain a whole range of iron containing proteins (and pigments), therefore, the presence of ferritin there is of no surprise. The case of lupin leaves is somewhat different from that reported for pea, where ferritins were detected only in the roots and leaves of young plantlets and remained undetectable in the corresponding organs of adult plants (Lobreaux, Briat 1991). In yellow lupin, the level of ferritin is much lower but significant in young leaves, and it increases with the plant age. Additionally, in all the cases tested, ferritin polypeptide levels rise during tissue senescence.
Nevertheless, ferritin may be considered to be a part of the oxidant stabilization system, which slows down the senescence processes by sequestration of released iron.
On the basis of the collected information it may be assumed that animal and plant ferritins are also coded by small gene families. Structures of these genes significantly differ between the families in the number (three vs. seven) and positions of introns (Proudhon et al. 1996).
In animal systems, ferritin synthesis is regulated on the translational level. Changes in conformation of iron regulatory proteins (IRP), caused by iron molecule binding, lead to removal of these proteins (iron sensing) from RNA structures (iron responsive elements; IRE). Freeing of 5' untranslated region of mRNA unblocks it for translation (Theil 1994). Regulation of synthesis of plant ferritins is much less understood. In general, it is believed that it is regulated mainly on the transcriptional level (Briat et al. 1999). Recently, some promoter sequence elements have been proposed to be important in the regulation of iron dependent soybean and maize ferritin gene expression (Petit et al. 2001; Wei, Theil 2000).
We have identified three classes of ferritin genes in yellow lupin tissues (cDNA and genomic). These genes code for polypeptides of 84-89% identity in the parts corresponding to matured protein and only 31-43% of identity of transit peptides. They reveal typical plant ferritin gene organization. We are Currently analyzing promoter sequences of these genes in search for elements homologous to these reported.
The analysis of the expression patterns of ferritin genes in different organs on the RNA level generally has confirmed earlier results on ferritin polypeptides detection. The difference was that we detected ferritin transcripts in all analyzed tissues, even in germinating plant cotyledons and in young seeds. This fact indicates posttranscriptional regulation of ferritin synthesis in these organs. Using class specific probes we could also analyze patterns of expression of each class of lupin ferritin separately in different organs. Generally, one class RNA is almost undetectable, the other two show similar patterns in the above-ground organs and differential expression in nodules.
We also used hydroponically grown plants and tissue cultured cells to test inductivity of lupin ferritin genes. These experiments have revealed that transcription of each ferritin class gene is differently induced by iron and abscisic acid. These last results indicate a possibility of different contribution of each class of ferritin to oxidative stress and ABA mediated defenses. Additionally, ABA is a hormone known to take part in a whole range of stress responses in plants, i.e. water stress and pathogen attack (Giraudat et al. 1994). Apart from its scientific (cognitive) objectives, research into ferritins has also practical purposes.
It has been proposed that withholding iron is one of the mechanisms of plant defense against pathogen invasion. Compounds like phenolics, synthesized by plant tissues, can limit the pathogen growth by restricting vital iron. It is also known that some pathogenic organisms produce toxins which induce iron acquisition, thus promoting the growth of the invader. In experiments with transgenic tobacco, others have demonstrated that plants ecotopically expressing alfalfa ferritin are tolerant to oxidative damage and pathogens (Deak et al. 1999). In our experiments, however, we could not detect any increase in ferritin level after the infection of lupin plants with pathogens. We have analyzed proteins isolated from lupin leaf tissue, from and around necrotic spots after the infection with Pseudomonas syringiae, from crown gall induced on lupin root neck by Agrobacterium tumefaciens and from lupin stem infected by Erwinia chrysantemi. That could mean that ferritin synthesis in response to pathogen attack, at least in the cases and/or attack stages we examined, is not a part of the natural plant defense system.
As iron deficiency is the most common nutritional disorder, affecting more than 30% of the world's population (WHO, UNICEF, ASPP, USD A), iron fortified transgenic plants could help solve this problem. It has been demonstrated that plant ferritin can serve as a source of iron for animals, curing iron deficiency anemia (Beard el al. 1996). Further work is concentrated on improving iron content in plant tissues by overexpressing the ferritin gene. Iron fortification of rice seed by the overexpression of soybean ferritin has been demonstrated (Goto et al. 1999).
The increased iron storage capacity as a result of increased iron sequestration leads to certain changes in iron homeostasis in such transgenic plants. It has been shown that besides activation of iron transport systems (root ferric reductase), plants accumulating ferritin on the highest levels display "iron deficient", chlorotic phenotypes (Van et al. 1999). This result is only apparently contradicting previous observations of the linear relationship between the level of ferritin accumulation and increased iron content (Goto et al. 1901). Later experiments with transgenic tobacco plants overexpressing ferritin have shown high soil-dependent variability of leaf iron accumulation (Vansuyt et al. 2000).
Recently, the obtaining of transgenic plants, which overexpress the ferritin gene (lettuce) and show not only the ability to accumulate excess iron but also the ability to increase their own growth has also been reported (Goto et al. 2000).
In the course of our work we have also generated transgenic tobacco (lettuce) plants with S35-promoter driven lupin ferritin gene. We now have a collection of a whole range of ferritin accumulating plants. There are plants with almost undetectable (immuno-detection) amounts of lupin ferritin subunit as well as such in which the hybridization signal is very strong. Although we can observe a clear tendency toward the iron-starvation-like phenotype, which is strictly correlated with increased levels of detected lupin ferritin, we can easily manipulate the symptoms. In the tissue culture where iron availability becomes limited with time, we can observe chlorosis of diverse intensity, but general levels of iron accumulation in leaf tissues are similar. On the other hand, in artificial soil, where nutrient ratios may be supervised, these symptoms may be leveled by iron and phosphorous supplementation. However, even in these conditions the most severe effects of high ferritin accumulation, observed in flowers and fruits, are difficult to overcome. Additionally, some lines of plants with the highest accumulation of ferritin show strong physiological aberrations. In iron sufficient conditions, leaves of these plants, which are narrow and irregularly shaped, turn much deeper green than the leaves of other plants. Also the flowers are morphologically different, i.e. they have short stamens and long styles.
Information collected thus far proves that there is a natural correlation between metal storage and acquisition systems, and that in order to engineer a fully functional plant for biotechnological applications these systems cannot be evaluated separately.
Detailed data concerning our work on yellow lupin ferritins are currently being prepared for publication.
Was this article helpful?