Isolation And Characterization Of Diazotrophic Endophytes From Grasses And Their Effects On Plant Growth

P.J. Riggs, R.L. Moritz, M.K. Chelius, Y. Dong, A.L. Iniguez, S.M. Kaeppler,

Dept of Agronomy, University of Wisconsin-Madison, Madison, WI 53706, USA 1. Introduction

The discovery of nitrogen fixation by diazotrophic endophytes in sugarcane by Dobereiner, Kennedy and coworkers (for reviews see Boddey et al. 2000; Sevilla, Kennedy 2000) has encouraged several laboratories around the world to identify nitrogen-fixing associations between other grasses and bacteria. Here, we review our efforts to discover such associations, present some new data on the characterization of some of these associations, and describe our plans for the future.

We first isolated a diazotrophic endophyte from a field-grown maize plant collected from the University of Wisconsin Agricultural Experimental Farm late in the growing season (Palus et al. 1996). Using the procedure of Dong et al. (1994), apoplastic fluid was collected from the stem and was plated on a modified LGI medium of Cavalcante and Dobereiner (1988). Three isolates were capable of growth on an N-free medium and could reduce acetylene as well as incorporate 15N2 into cells in vitro. These isolates were identified as members of the genus Klebsiella. We estimated that the number of cells in these stems was approximately 2000 per gram fresh weight of tissue. This was about 2-3 orders of magnitude lower than the number of Gluconacetobacter diazotrophicus in sugarcane. Since that time, we have made four other independent isolations of Klebsiella strains from the interior of maize (Chelius, Triplett 2000a, 2000b, 2001). Klebsiella appear to be common endophytes of maize.

We next wanted to show that these Klebsiella endophytes were inhabiting the interior of maize and that they could re-enter the plant upon inoculation (Chelius, Triplett 2000a). After labeling the Klebsiella cells with a constitutively expressed green fluorescent protein gene, the cells were found to re-enter maize upon inoculation and inhabit the intercellular spaces of the root cortex in large numbers. They are present in stems but in much lower numbers than in roots. In addition, these Klebsiella would produce NifH protein in planta provided that sucrose was added to the nutrient solution for the plants (Chelius, Triplett 2000b). Egener et al. (1998, 1999) previously showed that the expression of nifH by Azoarcus sp. in rice seedlings required the addition of a carbon source.

We have also shown by culture-dependent and culture-independent methods that the phylogenetic diversity of prokaryotes within maize is very extensive (Chelius, Triplett 2000c, 2001). Several divisions of bacteria are present in maize with the largest proportion of the bacteria belonging to the proteobacteria. Two divisions of archaea, the euryarchaea and crenarchaea, are also present in maize but given that nested PCR was required to detect the presence of archaea in maize, we suspect that their numbers in planta are very low compared to the bacteria. Very novel bacteria have also been cultured from the interior of maize. This includes a new genus and species, Dyadobacter fermentens (Chelius, Triplett 2000c).

The frequency with which Klebsiella have been found in maize has encouraged us to characterize these endophytic diazotrophs more thoroughly. We are particularly interested in comparing these endophytes to clinical isolates of K pneumoniae. To begin this work, we used genomic interspecies microarray hybridization to identify 3000 genes in common with E. coli that are present in one of our endophytic Klebsiella (Dong et al. 2001). Genes present in one of our endophytes but absent in a clinical isolate of K. pneumoniae are being isolated by subtraction hybridization.

In addition to isolating endophytes from maize, we have also isolated many endophytes from switchgrass (Panicum virgatum L.). Switchgrass was collected from remnant prairies in the central sands region of Wisconsin. These native plants have never seen nitrogen fertilizer. The isolation and characterization of several diazotrophs collected from these plants will be described here as well as their effect on the growth of wheat and rice.

We have discovered that several of these diazotrophic endophytes enhance maize growth in the presence of nitrogen fertilizer both in greenhouse and field experiments (Riggs et al. 2001). A novel approach to discover the molecular basis for these growth increases will be presented here.

2. Procedure

2.1. Collection and culture of switchgrass. Switchgrass plants were collected (WS98.1) from Buena Vista Quarry Prairie, a native prairie remnant near Plover, Wisconsin on an extremely sandy soil. From each of 10 random plants, we collected 10 tillers, a sample of soil surrounding the plant, and a sample of open-pollinated seed from the plant. Tillers were planted in about 200 ml of prairie soil surrounded by sand in the greenhouse. Plants were watered with a minus-N nutrient solution.

2.2. Isolation of endophytes from switchgrass and maize. Endophytes were isolated from maize and switchgrass as described previously (Chelius, Triplett 2000a, 2000b, 2001). Klebsiella pneumoniae 342 was isolated from an N-efficient line of maize obtained from CIMMYT, the International Maize and Wheat Improvement Center, in Mexico. Pantoea sp. P101 and P102 were isolated from switchgrass lines collected near Plover, Wisconsin.

2.3. Inoculation, presence, and NifH expression of K. pneumoniae in wheat. These experiments were done as described previously for our work on maize (Chelius, Triplett 2000a, 2000b).

2.4. Culture and harvest of rice, wheat, maize and Arabidopsis thaliana in the greenhouse. The rice and wheat lines used in this work were Oryza sativa spp. japonica cv. Nipponbare and Triticum aestivum L. cv. Trenton. Various ecotypes of Arabidopsis thaliana were obtained from the Arabidopsis Biological Resource Center (Ohio State University, Columbus, OH, USA). The seeds were germinated and plants cultured using the protocols of the Arabidopsis Biological Resource Center ( Grasses were cultured in a 1:1 sand:vermiculite mix in two liter plastic pots. Plants were given a nutrient solution that contains the following ingredients: 5 [iM CaCl2, 1.25 ^M MgS04, 5 (iM KC1, 1 ^M KH2P04, 0.162 |iM FeS04, 2.91 nM H3BO3, 1.14 nM MnS04, 0.76 nM ZnS04, 0.13 nM NaMo04, 0.14 nM NiCl2, 0.013 nM C0CI2 and 0.19 nM CuS04. Plants were given supplemental lighting. To obtain the dry weight data presented for all of these plant species, the above-ground portions of the plants were placed in paper bags and dried at 65°C for at least 48 h prior to weighing. Nitrogen content of wheat tissue was determined after grinding the tissue in a Wiley mill. Nitrogen concentration was measured by rapid combustion at 850°C followed by conversion of all N-combustion products to N2 and subsequent measurement by a thermoconductivity cell (LECO Model FP-528; LECO Corp., St Joseph, MO).

2.5. 16S rDNA amplification and sequencing. PCR amplification and sequencing of the 16S rRNA gene from grass endophytes was done as described previously (Chelius, Triplett 2000c). The accession numbers for the 16S rRNA genes of K. pneumoniae 342, Pantoea sp. P101 and Pantoea sp. P102 are AF394537, AF394538 and AF394539, respectively.

2.6. Construction of nifH insertion mutant of Klebsiella peumoniae 342. Amplification and sequence analysis showed that nifH from strain 342 has 88% sequence identity with nifH from K. pneumoniae M5al. Primers nifH If (5 '-GCCTGCAGATGACC ATGCGTCAATGCGCC-3') and nifH876r (5 '-GCGAATTCCGCGTTTTCTTCGGCGGCGGT-3') were designed based on the nifH sequence of K. pneumoniae M5al (GenBank accession number XI3303). One hundred ng of strain 342 DNA was used as the template in a 25 |il PCR mixture containing IX PCR Buffer (Promega), 2.5 mM MgCl2, 0.2 mM dNTPs and 0.5 U of Taq polymerase. Thermal cycle conditions were 4 min of denaturation at 95°C; 30 cycles of 94°C for 30 s, 54°C for 30 s, and 72°C for 1 min; and then an extension at 72°C for 7 min. The PCR product was purified using a Qiagen PCR purification kit and then ligated to pGEM-T Easy vector (Qiagen). White colonies were selected and plasmids were isolated. ABI's cycle sequencing kit was used by the University of Wisconsin-Madison Biotechnology Center to sequence two of those plasmids using primers T7 and Sp6. As a result of this high identity, plasmid pSA30 containing nifHDKYE from K. peumoniae M5al (Brown, Ausubel 1984) was used to make the insertion and subsequent marker exchange. A 1.7 kb fragment containing nifH gene (882 bp) and part of nifD (632 bp) was excised from pSA30 by double digestion with EcoKl and BamHl. This fragment was inserted into EcoKI/BamHI doubly digested vector pUC18, resulting in plasmid pHl. A 1.4 kb fragment from pKJRPll (Reece, Phillips 1995) containing a kanamycin resistance gene downstream of a constitutive promoter was excised with Hindlll and then blunted with Klenow. Following Bglll digestion of pHl and subsequent blunting with Klenow, the fragment from pKRPll was inserted into the Bglll site of pHl. The BglII site in pHl was located in the middle of nifH gene. This new plasmid was named as pll. To exchange the inserted allele of nifH for the wild type allele on the K. pneumoniae 342 chromosome, the 3.1 kb fragment containing nifHD'-Km was excised from pll by double digestion with £coRI and Pstl, and EcoRl site was blunted. This fragment was ligated into the PstUSmal doubly digested suicide plasmid pJQ200KS+ and marker exchanged as described previously (Scupham, Triplett 1997). Nif isolates were confirmed by Southern hybridization with an nptH probe

3. Results and Discussion

The effects on plant growth of three strains isolated in this laboratory from N-efficient lines of maize or switchgrass are described here. These strains are compared with two strains that are nitrogen-fixing endophytes of sugarcane. These strains often increase the productivity of maize when N fertilizer is supplied but they do not relieve N-deficiency conditions of those plants not given N fertilizer (Riggs et al. 2001). Here we tested the growth responses of these strains on rice and wheat in the absence of N and the growth effect of these strains on A. thaliana with the addition of fixed N. As wheat and rice plants are much smaller than maize plants, our hope was that these strains might provide enough fixed N to relieve the nitrogen deficiency of smaller grasses that could not be observed on maize (Figures 1 and 2).

None of the strains used here relieved nitrogen deficiency symptoms in wheat since N- „ , _ , . „ ^ , ^

. , . Figure 1. Growth of Trenton spring wheat six fertilized plants grown simultaneously were weeks after plantmg. Seeds were inoculated with much larger and more vigorous. However, K pneumoniae 342 or a nifti mutant of stram 342.

without addition of fixed N, the growth, ^ unmoculated control is also shown. Plants nitrogen concentration, and total N per plant of were cultured m the absence of added N.

uninoculated Kp342 Kp342nifH

strain 342-inoculated plants was significantly higher (4.22 mg N/plant) than in the uninoculated plants (3.54 mg N/plant) as well as those plants inoculated with the nifll mutant of strain 342 (3.53 mg N/plant). Thus, this growth increase in N-starved wheat upon inoculation with strain 342 can be attributed to nitrogen fixation by this endophyte (Figure 1). Strain 342 is an endophyte of wheat (Dong, Triplett, unpublished) in addition to its ability to survive endophytically in maize (Chelius, Triple« 2000a, 2000b).

As these strains can enhance maize growth with added N (Riggs et al. 2001), we are interested in the mechanism of these responses. We have decided to determine whether any of these strains can enhance the growth of Arabidopsis in the presence of fixed N. If Arabidopsis is increased in size, we can then screen mutants of Arabidopsis for those that do not respond to the inoculum. The non-responsive mutants will then be used to identify host genes involved in this response whose identity should give us clues as to the role of the bacteria in the response. We have found several ecotypes whose growth is dramatically increased upon inoculation with one or more of the endophytes used here (Figure 3) so we can now begin to screen EMS mutants of Ws-2. Hormone-insensitive mutants of Arabidopsis will also be tested.

We continue to maintain in the greenhouse the switchgrass plants that were collected in the fall of 1998 from remnant prairies near Polver, Wisconsin. Some of these plants continue to grow, although very slowly, despite having no fixed nitrogen source since November, 1998. Native grasses growing under low nitrogen conditions appear to be a rich source of diazotrophic bacteria that may provide fixed N to plants as well as increase plant growth by mechanism(s) independent of nitrogen fixation. In future work, we intend to test the ability of the diazotrophs collected from maize and switchgrass to provide fixed N to wheat and rice by 15N2 reduction assays.

Nitrogen Fixation Plants

uninoculated P102K p342

Figure 3. Growth of Arabidopsis thaliana Ws-2 following inoculation with the switchgrass isolate Pantoea sp. PI02 or the maize isolate K. pneumoniae 342. An uninoculated control is also shown. Plants were cultured with added N.

uninoculated P102K p342

Figure 3. Growth of Arabidopsis thaliana Ws-2 following inoculation with the switchgrass isolate Pantoea sp. PI02 or the maize isolate K. pneumoniae 342. An uninoculated control is also shown. Plants were cultured with added N.

uninoculated P101

Figure 2. Growth of rice 13 weeks following inoculation with Pantoea sp. P101. An uninoculated control is also shown. Plants were cultured m the absence of added N.

Was this article helpful?

0 0

Post a comment