Structural and gene expression analyses of uptake hydrogenases and other proteins involved in nitrogenase protection in Frankia.

The actinorhizal bacterium Frankia expresses nitrogenase and can therefore convert molecular nitrogen into ammonia and the by-product hydrogen. However, nitrogenase is inhibited by oxygen. Consequently, Frankia and its actinorhizal hosts have developed various mechanisms for excluding oxygen from their nitrogen-containing compartments. These include the expression of oxygen-scavenging uptake hydrogenases, the formation of hopanoid-rich vesicles, enclosed by multi-layered hopanoid structures, the lignification of hyphal cell walls, and the production of haemoglobins in the symbiotic nodule. In this work, we analysed the expression and structure of the so-called uptake hydrogenase (Hup), which catalyses the in vivo dissociation of hydrogen to recycle the energy locked up in this 'waste' product. Two uptake hydrogenase syntons have been identified in Frankia: synton 1 is expressed under freeliving conditions while synton 2 is expressed during symbiosis. We used qPCR to determine synton 1 hup gene expression in two Frankia strains under aerobic and anaerobic conditions. We also predicted the 3D structures of the Hup protein subunits based on multiple sequence alignments and remote homology modelling. Finally, we performed BLAST searches of genome and protein databases to identify genes that may contribute to the protection of nitrogenase against oxygen in the two Frankia strains. Our results show that in Frankia strain ACN14a, the expression patterns of the large (HupL1) and small (HupS1) uptake hydrogenase subunits depend on the abundance of oxygen in the external environment. Structural models of the membrane-bound hydrogenase subunits of ACN14a showed that both subunits resemble the structures of known [NiFe] hydrogenases (Volbeda et al. 1995), but contain fewer cysteine residues than the uptake hydrogenase of the Frankia DC12 and Eu1c strains. Moreover, we show that all of the investigated Frankia strains have two squalene hopane cyclase genes (shc1 and shc2). The only exceptions were CcI3 and the symbiont of Datisca glomerata, which possess shc1 but not shc2. Four truncated haemoglobin genes were identified in Frankia ACN14a and Eu1f, three in CcI3, two in EANpec1 and one in the Datisca glomerata symbiont (Dg).

Nitrogen fixation in mixed Hylocomium splendens moss communities.

The pleurocarpus feather moss, Hylocomium splendens, is one of two co-dominant moss species in boreal forest ecosystems and one of the most common mosses on earth, yet little is known regarding its capacity to host cyanobacterial associates and thus contribute total ecosystem N. In these studies, we evaluated the N-fixation potential of the H. splendens-cyanobacteria association and contrasted the N-fixation activity with that of the putative N-fixing moss-cyanobacteria association of Pleurozium schreberi. Studies were conducted to: quantify N-fixation in H. splendens and P. schreberi in sites ranging from southern to northern Fennoscandia; assess N and P availability as drivers of N-fixation rates; contrast season-long N-fixation rates for both mosses; and characterize the cyanobacteria that colonize shoots of H. splendens. Nitrogen-fixation rates were generally low at southern latitudes and higher at northern latitudes (64-69 degrees N) potentially related to anthropogenic N deposition across this gradient. Nitrogen fixation in H. splendens appeared to be less sensitive to N deposition than P. schreberi. The season-long assessment of N-fixation rates at a mixed feather moss site in northern Sweden showed that H. splendens fixed a substantial quantity of N, but about 50% less total N compared to the contribution from P. schreberi. In total, both species provided 1.6 kg fixed N ha(-1) year(-1). Interestingly, H. splendens demonstrated somewhat higher N-fixation rates at high fertility sites compared to P. schreberi. Nostoc spp. and Stigonema spp. were the primary cyanobacteria found to colonize H. splendens and P. schreberi. These results suggest that H. splendens with associated Nostoc or Stigonema communities contributes a significant quantity of N to boreal forest ecosystems, but the contribution is subordinate to that of P. schreberi at northern latitudes. Epiphytic cyanobacteria are likely a key factor determining the co-dominant presence of these two feather mosses across the boreal biome.

Identification of white-rot and soft-rot fungi increasing ethanol production from spent sulfite liquor in co-culture with Saccharomyces cerevisiae.

AIM: To identify fungi that are capable of increasing ethanol production from lignocellulose in spent sulfite liquor. METHODS AND RESULTS: In a batch fermentation study, the fungal mix could produce 24.61 g l(-1) ethanol using spent sulfite liquor as substrate. The fungal mix grew well on glucose, xylose, hemicellulose and cellulose. In addition, we were able to identify the fungal mix by use of PCR-amplification of DNA and sequencing, and they were identified as Chalara parvispora and Trametes hirsuta/T. versicolor. In a reconstitution study, the identified fungi were shown to produce equal amount of ethanol as the fungal mix. We were also able to show that C. parvispora could produce ethanol from xylose. CONCLUSION: The present study has shown that ethanol production from biomass can be increased by use of C. parvispora and T. versicolor when compared with fermentation using only S. cerevisiae. SIGNIFICANCE AND IMPACT OF THE STUDY: The study shows that refining biomass by ethanol production from spent sulfite liquor, a lignocellulose material, can be increased by adding C. parvispora and T. versicolor, and it is thus of great potential economical impact.

Molecular characterization of uptake hydrogenase in Frankia.

A molecular characterization of uptake hydrogenase in Frankia was performed by using two-dimensional gel electrophoresis, matrix-assisted laser-desorption ionization-time-of-flight mass spectrometry, PCR amplification and Southern blotting. A polypeptide of approx. 60 kDa was recognized in Frankia UGL011102, AVCI1 and KB5 on the two-dimensional gel by blotting with Ralstonia eutropha (Hox G) antibody. Further analysis by MS resulted in a peptide 'fingerprint', which was similar to the membrane-bound hydrogenase 2 large subunit (HYD2) in Escherichia coli. In addition, a 127 bp PCR fragment could also be amplified from Frankia AVCI1, which gave a 76% similarity with the large subunit of hydrogenase in, e.g., Azotobacter chrococcum, Bradyrhizobium japonicum and Rhizobium leguminosarum. Although immunological similarity between the small subunit of Frankia hydrogenase and that of other organisms has not yet been found, a PCR product of 500 bp could be amplified from the local source of Frankia, the analysis of which gave 69 and 67% identity with the small subunit of hydrogenases in B. japonicum and R. leguminosarum respectively. A Southern-blot analysis further indicated evidence for the presence of the small hydrogenase subunit in other Frankia strains, i.e. KB5, AvcI1 and CcI3.

Frankia KB5 possesses a hydrogenase immunologically related to membrane-bound.

The immunological relationship of the hydrogenase in Frankia KB5 to hydrogenases in other microorganisms was investigated using antisera raised against holo-[NiFe]-hydrogenases isolated from Alcaligenes latus, Azotobacter vinelandii, Ralstonia eutropha, and the small and large hydrogenase subunits from Bradyrhizobium japonicum. The antisera raised against the A. latus, R. eutropha, and B. japonicum (large subunit) polypeptides were found to recognize two polypeptides, corresponding to the unprocessed and processed forms of the hydrogenase subunit in Frankia KB5. None of the antisera, including the antibodies produced against the small hydrogenase subunit isolated from B. japonicum, recognized any polypeptide related to the small hydrogenase subunit in Frankia KB5. An immunogold localization study of the intracellular distribution of hydrogenase in Frankia KB5, with the cryo-section technique, showed that labeling in the membrane of both hyphae and vesicles was positively correlated with hydrogenase activity.

DNase-resistant DNA in the extracellular and cell wall-associated fractions of Frankia strains R43 and CcI3.

DNases were shown to be present in the extracellular fraction of Frankia strains R43 and CcI3. In spite of this, DNA was found in both the extracellular and cell wall fractions of these strains, and it was shown that extracellular DNA was resistant to the DNases secreted into the culture medium of both Frankia strains. Furthermore, Southern blot analysis under high stringency conditions revealed the chromosomal origin of the cell wall-adsorbed DNA (CW-DNA). Mobility gel band shift assays suggested that the extracellular DNA and the CW-DNA are engaged in complexes with other molecules, most likely proteins, which are probably responsible for the enzymatic resistance observed against extracellular DNase activities. In addition, it was shown that lysis of a small proportion of the cells in the exponential growth phase may account for the DNA being released into the supernatant and adsorbed to the cell wall.

Hydrogenase in Frankia KB5: expression of and relation to nitrogenase.

The localization and expression of the hydrogenase in free-living Frankia KB5 was investigated immunologically and by monitoring activity, focusing on its relationships with nitrogenase and H2. Immunological studies revealed that the large subunit of the hydrogenase in Frankia KB5 was modified post-translationally, and transferred into the membrane after processing. The large subunit was constitutively expressed and no correlation was found between hydrogenase activity and synthesis. Although H2 was not needed for induction of hydrogenase synthesis, exogenously added H2 triggered hydrogen uptake in medium containing nitrogen, i.e., in the hyphae. A correlation between nitrogenase activity and hydrogen uptake was found in cultures grown in media without nitrogen, but interestingly the two enzymes showed no co-regulation.

A simple, rapid and non-destructive procedure to extract cell wall-associated proteins from Frankia.

A simple cell fractionation procedure was developed to extract cell wall-associated proteins from the nitrogen-fixing actinomycete Frankia. The method was based on washing Frankia mycelia in 62.5 mM Tris-HCl (pH 6.8) buffer supplemented with 0.1% Triton X-100 as solubilizing agent. Cell wall-associated proteins were efficiently extracted in less than 10 min, recovering approximately 94.5+/-7.44 microg protein per extraction procedure from exponentially growing cells corresponding to 50 ml of culture. The amount of cell lysis occurring during the cell wall extraction was estimated to be 1.50+/-0.51%. Three peptidoglycan hydrolases with apparent molecular masses of 4.7, 12.1, and 17.8 kDa were detected by zymography in the cell wall-associated protein fraction. On the contrary, no cell wall lytic enzyme was detected in the cytoplasmic protein fraction. These results indicate that the present method enables a specific extraction of cell wall-associated proteins. Moreover, fluorescein isothiocyanate (FITC) labelling of the cell surface proteins showed an efficient removal of cell wall-associated proteins. Growth of the treated Frankia cells (i.e. cells from which the cell wall-associated proteins were removed) in semi-solid media suggested that these cells were still viable. This technique is of importance for functionality studies of cell wall-associated proteins, particularly for bacteria where traditional cell fractionation methods are difficult to be applied.

DNase Activities of the Extracellular, Cell Wall-Associated, and Cytoplasmic Protein Fractions of Frankia Strain R43.

DNase activities in different protein fractions of Frankia strain R43 were studied. The extracellular and the cell wall-associated fractions revealed the presence of exo- and endonucleolytic enzymes, but none was detected in the cytoplasmic fraction. The strongest DNase hydrolysis was found in the extracellular fraction, in which six DNases were detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Specificity and effectivity in nodulation by Frankia on southern hemisphere actinorhiza.

Nodulation ability was tested for Frankia strains HFPCcI3 and EL1, and Frankia sources A.t. and G.a. from Allocasaurina torulosa and Gymnostoma australianum, respectively, on A. torulosa Miq., Casuarina cunninghamiana Miq., G. australianum L. Johnson and Elaeagnus triflora Roxb. It was shown that A. torulosa and C. cunninghamiana formed nodules only with the Frankia sources obtained from their own host plant, while E. triflora formed nodules with three of the four Frankia sources tested. All nodules formed were effectively fixing nitrogen. Specific nitrogenase activity was highest in E. triflora inoculated with the Frankia strain isolated from nodules of the same species. Identification of Frankia sources in the nodules was performed by use of PCR amplification of DNA with a random primer. PCR amplification of DNA isolated from nodules of G. australianum and E. triflora inoculated with Frankia strain EL1 revealed, when compared with DNA amplified from free living Frankia strain EL1, that there was only one Frankia strain causing the observed nodules.

Identification of Casuarina-Frankia strains by use of polymerase chain reaction (PCR) with arbitrary primers.

Free-living N2-fixing Frankia strains isolated from Casuarina sp. were investigated for genomic polymorphism. We used six 10-mer oligonucleotides as single arbitrary primers (AP) for the polymerase chain reaction (PCR) in order to amplify random DNA fragments in the genome of free-living Frankia strains. Agarose-gels of the amplified genomic DNA revealed that two of the six arbitrary primers showed polymorphism in the eight different Frankia genomes. Analysis of the AP-PCR products showed 9 polymorphic bands ranging from 4.1-0.60 kb. We conclude that single arbitrary primers can be used to amplify genomic DNA, and that polymorphism can be detected between the amplification products of the different Frankia genomes.

Acetylene, Not Ethylene, Inactivates the Uptake Hydrogenase of Actinorhizal Nodules during Acetylene Reduction Assays.

Acetylene reduction assays were shown to inactivate uptake hydrogenase activity to different extents in one Casuarina and two Alnus symbioses. Inactivation was found to be caused by C(2)H(2) and not by C(2)H(4). Acetylene completely inactivated the hydrogenase activity of intact root systems of Alnus incana inoculated with Frankia strain Avcl1 in 90 minutes, as shown by a drop in the relative efficiency of nitrogenase from 1.0 to 0.73. The hydrogenase of Frankia preparations (containing vesicles) and of cell-free extracts (not containing vesicles) from the same symbiosis was much more susceptible to acetylene inactivation. Cell-free extracts lost all hydrogenase activity after 5 minutes of exposure to acetylene. The hydrogenase activity of intact root systems of Casuarina obesa was less sensitive to acetylene than that of root systems of A. incana, since the relative efficiency of nitrogenase changed only from 1.0 to 0.95 over 90 minutes. Frankia preparations and cell-free extracts of C. obesa still retained hydrogenase activity after a 10 minute-exposure to acetylene.

Activities, occurrence, and localization of hydrogenase in free-living and symbiotic frankia.

Symbiotic and free-living Frankia were investigated for correlation between hydrogenase activities (in vivo/in vitro assays) and for occurrence and localization of hydrogenase protein by Western blots and immuno-gold localization, respectively. Freshly prepared nodule homogenates from the symbiosis between Alnus incana and a local source of Frankia did not show any detectable in vivo or in vitro hydrogenase uptake activity, as also has been shown earlier. However, a free-living Frankia strain originally isolated from these nodules clearly showed both in vivo and in vitro hydrogenase activity, with the latter being approximately four times higher. Frankia strain Cpl1 showed hydrogen uptake activity both in symbiosis with Alnus incana and in a free-living state. Western blots on the different combinations of host plants and Frankia strains used in the present study revealed that all the Frankia sources contained a hydrogenase protein, even the local source where no in vivo or in vitro activity could be measured. The 72 kilodalton protein found in the symbiotic Frankia as well as in the free-living Frankia strains were immunologically related to the large subunit of a dimeric hydrogenase purified from Alcaligenes latus. Recognitions to polypeptides with molecular masses of about 41 and 19.5 kilodaltons were also observed in Frankia strain UGL011101 and in the local source of Frankia, respectively. Immunogold localization of the protein demonstrated that in both the symbiotic state and the free-living nitrogen-fixing Frankia, the protein is located in vesicles and in hyphae. The inability to measure any uptake hydrogenase activity is therefore not due to the absence of hydrogenase enzyme. However, the possibility of an inactive hydrogenase enzyme cannot be ruled out.

Acetylene reduction, H2 evolution and (15)N 2 fixation in the Alnus incana-Frankia symbiosis.

Acetylene reduction, (15)N2 reduction and H2 evolution were measured in root systems of intact plants of grey alder (Alnus incana (L.) Moench) in symbiosis with Frankia. The ratios of C2H2: (15)N2 were compared with C2H2:N2 ratios calculated from C2H2 reduction and H2 evolution, and with C2H2:N2 ratios calculated from accumulated C2H4 production and nitrogen content. It was possible to calculate C2H2:N2 ratios from C2H2 reduction and H2 evolution because this source of Frankia did not show any hydrogenase activity. The ratios obtained using the different methods ranged from 2.72 to 4.42, but these values were not significantly different. It was also shown that enriched (15)N could be detected in the shoot after a 1-h incubation of the root-system. It is concluded that the measurement of H2 evolution in combination with C2H2 reduction represents a nondestructive assay for nitrogen fixation in a Frankia symbiosis which shows no detectable hydrogenase activity.

Biomass production and nitrogen utilization by Alnus incana when grown on N2 or NH 4 (+) made available at the same rate.

A single clone of Alnus incana (L.) Moench was grown in a controlled-environment chamber. The plants were either inoculated with Frankia and fixed atmospheric nitrogen or were left uninoculated but received ammonium at the same rate as the first group fixed their nitrogen. Nitrogen fixation was calculated from frequenct measurements of acetylene reduction and hydrogen evolution. The diurnal variation of acetylene reduction was also taken into account. The relative efficiency of nitrogenase could be used in the calculations of fixed nitrogen since the Frankia used did not show any detectable hydrogenase activity. Alders fixing nitrogen developed more biomass, longer shoots, larger leaf areas and contained more nitrogen than alders receiving ammonium. In one experiment, almost all ammonium given to the non-nodulated alders was taken up and 15% of the nitrogen taken up was excreted. In the other experiment, 34% of the ammonium was left in the nutrient solution and 8% of the nitrogen taken up was excreted. Alders inoculated with Frankia did not excrete any detectable amount of nitrogen. It seems that the energy demand for nitrogen fixation is not so high that biomass production in alders is retarded. The symbiotic system of A. incana and Frankia seems to be more efficient in utilizing its nitrogen than non-symbiotic A. incana receiving ammonium.

Ammonium effects on function and structure of nitrogen-fixing root nodules of Alnus incana (L.) Moench.

Cloned plants of Alnus incana (L.) Moench were inoculated and grown without combined nitrogen for seven weeks. The effects of ammonium on the function and structure of the root nodules were studied by adding 20 mM NH4Cl (20 mM KCl=control) for four days. Nitrogenase activity decreased to ca. 50% after one day and to less than 10% after two days in ammonium treated plants, but was unaffected in control plants. The results were similar at photon flux densities of 200 and 50 µmol m(-2) s(-1). At the higher light level the effect was concentration dependent between 2 and 20 mM NH4Cl. The recovery was slow, and more than 11 d were needed for plants treated with 20 mM ammonium to reach initial activity. The distribution of (14)C to the root nodules after assimilation of (14)CO2 by the plants was not changed by the ammonium treatment. Microscopical studies of root nodules showed high frequencies of endophyte vesicles being visually damaged in nodules from ammonium-treated plants, but not in nodules from control plants. When nitrogenase activity was restored, visually damaged vesicles were again few, whereas young developing vesicles were numerous. The slow recovery, the (14)C-translocation pattern, and the structural changes of the endophyte indicate a more complex mechanism of ammonium influence than simply a short-term reduction in supply of carbon compounds to the nodules.