Beruflich Dokumente
Kultur Dokumente
DNA fingerprints of several Azospirillum strains, belonging to the five known species A. amuwnense, A. brasilense, A.
halopraeferens, A. irakense and A. lipoferum, were obtained by restriction analysis of the amplified 16s rDNA and by
restriction fragment length polymorphism of the histidiue biosynthetic genes. Data obtained showed that amplified rDNA
restriction analysis is an easy, fast, reproducible and reliable tool for identification of Azospirillum strains, mainly at the
species level, whereas restriction fragment length polymorphism could, in some cases, differentiate strains belonging to the
same species. Moreover, both analyses gave congruent results in grouping strains and in the assignment of new strains to a
given species.
Keywords: Azospirillum; Restriction fragment length polymorphism; Histidine operon; 16s rDNA, Amplified rDNA
restriction analysis
unknown function (i.e. randomly cloned DNA) [7,8] 2.3. RFLP analysis of the histidine operon
have been commonly utilized. The rRNA genes can
be easily obtained in high amounts by enzymatic Total DNA of Azospirillum strains was purified
amplification via polymerase chain reaction (PCR); as described [2]. Restriction digestions using
the restriction analysis of rDNA amplified genes Boehringer (Mannheim) enzymes and buffers, were
(ARDRA) has also been utilized for bacterial identi- carried out by treating l-3 ,ug of DNA with lo-30
fication [9,10]. units of enzyme at 37°C for 5-6 h. Fragments were
In the present work we have used the restriction separated by electrophoresis on a 0.8% (w/v) agarose
analysis of amplified 16s rDNA to evaluate the gel.
reliability of this technique for taxonomical and iden- Probes, prepared according to standard protocols
tification purposes in the genus Azospirillum, com- [ll], were labeled and the hybridization signals de-
paring the results with those obtained with RFLP on tected with the Digoxigenin Labeling and Detection
total DNA probed with the histidine biosynthetic Kit (Boehringer, Mannheim) using the chemilumi-
genes. The latter technique has previously been used nescence method following the instructions of the
successfully on a limited number of Azospirillum supplier.
strains [6]. Southern blotting was carried out as described
1111 on nylon membranes (Hybond N, Amersham),
which were hybridized as described 161. The hy-
2. Materials and methods bridization patterns were analysed with a scanner
densitometer GDS2000 (Ultra-Violet Product Ltd.,
Cambridge, UK).
2.1. Bacterial strains and growth conditions
2.4. Cluster analysis
Bacterial strains are listed in Table 1. A. halo-
prueferens was grown at 41’C in Nutrient Broth [ll] Pairwise comparison of the restriction patterns
supplemented with 0.25% NaCI; all other Azospiril- was carried out and the coefficient of similarity S,
lum strains were grown in LB medium [ 1l] at 35°C. between pairs of strains was calculated according to
Fox et al. [13] from the ratio of twice the number of
2.2. Amplification and restriction analysis of 16s common fragments to the total number of fragments.
rDNA (ARDRA) These values were used to cluster the strains by the
unweighted pair group method with arithmetic mean
Amplification of 16s rDNA was performed di- (UPGMA) using the CLUSTER and TREE proce-
rectly on a single bacterial colony as follows: a dures of the SAS package [14].
whole colony, taken from an agar plate with a sterile
toothpick, was resuspended in 20 ~1 of sterile dis-
tilled water and heated to 95°C for 10 min to allow 3. Results and discussion
cell lysis; 1 ~1 of lysed cell suspension was used for
PCR amplification. The reactions were performed as 3.1. Restriction analysis of amplified Azospirillum
described [12] with the two universal primers 27f rDNA (ARDRA)
(~-GAGAGT-I-TGATCCTGGCI-CAG) and 1495r
(5’~CTACGGCTACCITGICGA), which allowed 16s rDNA was amplified as described in Materi-
the amplification of nearly the entire 16s rRNA als and methods by means of PCR from twenty-nine
gene. A 4-6-~1 aliquot of each PCR reaction, con- Azospirillum strains. Most of these strains had been
taining about 1.5 pg of DNA, was treated with 5 previously assigned, as shown in Table 1, on the
units of the restriction enzyme AluI (Boehringer, basis of phenotypic traits and/or molecular data, to
Mannheim) in a total volume of 30 ~1 at 37°C for 3 one of the five known species: A. amazonense, A.
h. The reaction products were analysed by agarose brasilense, A. halopraeferens, A. irakense or A.
gel (2.5% w/v) electrophoresis. lipoferum; in contrast, other strains were not pre-
Fig. 1. Agarose gel electrophoresis of amplified 16s rDNA of 29 Awspirillum strains digested with restriction endonuclease AluI. Lanes
1-16, A. brasilense (A. b.) strains Sp7, SpF94, Cdr, DSM 1859, DSM 2298, DSM 2287, DSM 1858, SpBrl7, SpF2, Tl, T2, F14, 129/l,
129/2, 129/3; lanes 17-24, A. Zipoferum(A. 1.) strains 133/l, 137/S, ATCC 29707, ATCC 29708, ATCC 29731, F, W03, PAl; lanes
25-27, A. amazonense (A. a.) strains: Yl, Y2, Y6; lane 28, A. irakense (A. i.) strain 103312; lane 29, A. halopraeferens (A. h.) strain
Au4. Numbers indicate the size of some of bands of the molecular mass marker (123 bp ladder, Boehringer, Mannheim).
ciscly identified (Table 1). A major strong amplifica- SpF94, Cdr, DSM 1859, DSM 2298, DSM
tion band of about 1450 bp from each Azospirillum 2787, DSM 1858, or A. amazonense strains
strain was obtained (not shown). The endonuclease Yl, Y2 and Y6).
Ah1 was chosen on the basis of a preliminary (ii) Strains SpP2, Tl, T2, F14, 129/l, 129/2,
screening of several four- or six-base cutting restric- 129/3 can be assigned to the species A.
tion endonucleases, performed on a limited number brasdense, and strains 133/l and 137/5 to
of strains. The reproducibility of ARDRA patterns, the species A. Zipoferum. This assignment is in
obtained after Ah1 digestion, was verified on PCR agreement with physiological tests (C. Scotti,
products of different independent amplifications. personal communication) and/or with DNA
The ARDRA from the 29 Azospirihm strains is fingerprints [2,15]. Strains SpBr17 and Sp242,
shown in Fig. 1. The following conclusions can be which had previously been assigned to the
drawn: species A. lipoferum, showed the typical
(i) Strains previously assigned to a given species ARDRA pattern of A. brasilense, which is in
share a very similar or identical restriction agreement with restriction endonuclease analy-
pattern (for example A. brusilense strains Sp7, sis of total DNA [2,15].
RI Pv Pv Pv Pv
1043 I 2150 1 739 1 590 1 1520 y%3~
Fig. 2. Restriction and genetic map of the hi&dine operon and the upstream region of A. brasilense strain SpF94. Symbols: Bg, Fv, P,
recognition sites for the restriction endonucleases BglII, PuuII, PstI, respectively; numbers between restriction sites indicate the length of
the restriction fragments ln bp; Bd, H, 1, A, F, E, 2, genes belonging to the histidine cluster hi&d, hti, ORFl, hM, hiss, his& ORF2,
respectively; black box, transcription terminator; the arrow with closed circle indicates the promoter and the direction of tmnscription of his
genes. 0RF978 and ORFS46 represent two putative ORFs with unknown function (unpublished results). Continuous lines under the genetic
map indicate the fragments used as probes.
(iii) Four main groups of restriction profiles can be relevant one being the presence of an addi-
recognized, three of which typical of the tional band, in A. bradense strain DSM1859
species A. brusihzse (lanes l-161, A. (Fig. 1, lane 4). This fact can be explained by
lipoferum (lane 17-24) or A. amazonense (lane the presence, within the DSM1859 genome, of
25-271, respectively. The fourth group in- rDNA alleles with different sequences, as pre-
cludes strains A. in&we 103312 and A. viously reported for Comamonadaceae [9].
halopraeferens Au4, which show identical The data obtained indicate that ARDRA could
Ah1 restriction patterns; these two strains also represent a simple, extremely rapid, highly repro-
share identical patterns with the enzyme ducible and reliable tool for the identification of
SadAI. Awspirillum species. The two species A. irukmse
(iv) Only few differences are found in the restric- and A. halopraejkrens, however, appear to be closely
tion patterns at intraqecific level, the most related.
--
-3621
-2995
-2896
Fig. 3. Hybridization profiles on total DNA from Awspirillum strains digested with P&I and BgZII. The probe used was a 7679-bp PstI
DNA fragment from A. bradense (see Fig. 2). Identical patterns were found for strains SpF94, Cdr and Sp7 (not shown); for strains
129/l, 129/2, 129/3 (not shown) and for strains PAl, 137/S and 137/6 (not shown). Numbers indicate the size (in bp) of some
hybridization bands. Symbols: A.a., A.b., Ah., A.i., A.I., indicate species A. a mazmense, A. bradense, A. halopraeferens, A. irakeme
and A. lipoferum, respectively.
[7] Grimont, F., Verger, J.M., Comelis, P., Limet, J., Lefevre, [15] Ventura, S., Giovannetti, L., Gori, A., Manta, S. and Mon-
M., Grayon, M., Regnault, B., Van Broeck, J. and Grimont, taini, P. (1990) DNA restriction fingerprinting for the identi-
PAD. (1992) Molecular typing of Bruce/h with cloned fication of Azospirilla. In: Nitrogen Fixation (Polsinelli, M.,
DNA probes. Res. Microbial. 143, 55-65. Materassi, R. and Vincenzini, M., Eds.) pp. 169-174. Kluwer
[8] Scieux, C., Grimont, F., Regnault. B. and Grimont, P.A.D. Academic publishers, Dordrecht.
(1992) DNA fingerprinting of Chlamydia trachomatis by use [16] Fani, R., Alifano, P., Allotta, G., Bazzicalupo, M., Carlo-
of ribosomal RNA, oligonucleotide and randomly cloned magno, M.S., Gallori, E., Rivellini, F. and Polsinelli, M.
DNA probes. Res. Microbial. 143, 755-765. (1993) The histidine operon of Azospirillum brasilense:
[P] Vaneechoutte, M., Rossau, R., De Vos, P., Gillis, M., organization, nucleotide sequence and functional analysis.
Janssens, D., Paepe, N., De Rouck, A., Fiers, T., Claeys, G. Res. Microbial. 144, 187-200.
and Kerster, K. (1992) Rapid identification of bacteria of the [17] Fani, R., Bazzicalupo, M., Damiani, G., Bianchi, A.,
Comamonadaceae with amplified ribosomal DNA-restriction Schipani, C., Sgaramella, V. and Polsinelli, M. (1989)
analysis (ARDRA). FEMS Microbial. Lett. 93, 227-234. Cloning of the histidine genes of Azospirillum brasilense:
[lo] Ralph, D P:\tic, D., Baranton, G., Pretzman, C. and Mc- organization of the ABFH gene cluster and nucleotide se-
Clelland, M. (1993) Species of Borrelia distinguished by quence of the hisB gene. Mol. Gen. Genet. 216, 224-229.
restriction site polymorphisms in 16s rRNA genes. FEMS [18] Fani, R., Bazzicalupo, M., Ricci, F., Schipani, C. and
Microbial. Lett. 111, 239-244. Polsinelli, M. (1988) A plasmid vector for the selection and
[ll] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecu- study of transcription promoters in Azospirillum brasilense.
lar Cloning: A Laboratory Manual, 2nd edn. Cold Spring FEMS Microbial. Lett. 50, 271-276.
Harbor Laboratories, Cold Spring Harbor, NY. [lP] Waschutza, S., Hofmann, N., Niemann, E.G. and Fendrik, I.
1121 Bandi, C., Damiani, G., Magrassi, L., Grigolo, A., Fani, R. (19921 Investigations on root exudates of korean rice. Sym-
and Sacchi, L. (1994) Flavobacteria as intracellular symbiont biosis 13, 181-189.
in cockroaches. Froc. R. Sot. Lond. B. 257, 43-48. [20] Delorme, C., Ehrlich, SD. and Renault, P. (1992) Histidine
[13] Fox, G.E., Stackebrandt, E., Hespell, R.B., Gibson, J., biosynthesis genes in Lactococcus lactis subsp. lactis. J.
Maniloff, J., Dyer, T.A., Wolfe, R.S., Balch, W.E., Tannar, Bacterial. 174, 6571-6579.
R., Magrum, L., Zablen, L.B., Blakemore, R., Gupta, R., [21] Kuenzler, M., Balmelli, T., Egli, C.M., Paravicini, G. and
Bonen, L., Lewis, B.J., Stahl, D.A., Luehrsen, K.R., Chen, Braus, G.H. (1993) Cloning, primary structure, and regula-
K.N. and Woese, CR. (1980) The phylogeny of prokaryotes. tion of the HIS7 gene encoding a bifunctional glutamine
Science 209, 457-463. amidotransferase: cyclase from Saccharomyces cereuisiae. J.
[14] SAS Institute Inc. (1988) SAS/STATTM user’s guide, edn. Bacterial. 175, 5548-5558.
6.03. SAS Institute, Gary, NC.