A Detailed Phenotypic Characterisation of the Type Strains of Halomonas Species
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
System. Appl. Microbiol. 25, 360–375 (2002)
© Urban & Fischer Verlag
http://www.urbanfischer.de/journals/sam
A Detailed Phenotypic Characterisation of the Type Strains
of Halomonas Species
JUAN ANTONIO MATA, JOSÉ MARTÍNEZ-CÁNOVAS, EMILIA QUESADA, and VICTORIA BÉJAR
Exopolysaccharide Research Group, Department of Microbiology, Faculty of Pharmacy, University of Granada, Spain
Received: June 12, 2002
Summary
We have made a detailed phenotypic characterisation of the type strains of 21 species within the genus
Halomonas and have also studied any possible intraspecific variation of strains within H. eurihalina, H.
halophila, H. maura and H. salina. We used 234 morphological, physiological, biochemical, nutritional
and antimicrobial susceptibility tests. Nutritional assays were carried out using both classical and minia-
turized (BIOLOG system) identification methods. Two different numerical analyses were made using the
TAXAN program; the first included the differential data from all the tests carried out whilst the second
used only the 57 tests with the highest diagnostic scores (≥ 0.5). The results of both analyses were quite
similar and demonstrated the phenotypic heterogeneity of the Halomonas species in question. At a 62%
similarity level the type species were grouped into three phena, the main difference between them being
the capacity of those included within phenon A (H. aquamarina, H. meridiana, H. cupida, H. pantelle-
riensis and H. halmophila) to produce acids from sugars. The species grouped in phenon C (H. camp-
isalis, H. desiderata and H. subglasciescola) used fewer organic substrates than the others. The remaining
strains were included in phenon B. H. marisflavi was clearly distinct and thus was not included in any of
the three phena. High phenotypic similarity (more than 88%) was found between Halomonas campisalis
and Halomonas desiderata.
The results of our work should allow researchers to minimise the tests required to arrive at a reliable
phenotypic characterisation of Halomonas isolates and to select those of most use to differentiate
Halomonas species from each other.
Key words: Halomonas – halophilic bacteria – phenotypic characterisation – BIOLOG – numerical
taxonomy
Introduction
Extremophiles are microorganisms that grow under only one species, H. elongata, but currently Halomonas
conditions hostile to most organisms (HORIKOSHI, 1997). includes 22 species, some of them previously assigned to
Some of them, such as bacteria which thrive in hyper- other genera, such as Arthrobacter (COBET et al., 1970),
saline environments, have been recognised for their po- Deleya (AKAGAWA and YAMASATO, 1989; BAUMANN et al.,
tential use in biotechnological applications (GALINSKI and 1972; BAUMANN et al., 1983; QUESADA et al., 1984;
TINDALL, 1992; VENTOSA and NIETO, 1995; AGUILAR, VALDERRAMA et al., 1991) Flavobacterium (VOLCANI,
1996). Thus extensive studies have been made in recent 1940), Halovibrio (FENDRICH, 1988), Paracoccus (KOCUR
years into hypersaline environments, resulting in a large 1984), Pseudomonas (ROSENBERG, 1983; HEBERT and
number of new halophilic species being isolated. VREELAND, 1987) and Volcaniella (QUESADA et al., 1990).
The family Halomonadaceae included initially two Other species, however, have only been described more
genera of halophilic bacteria, Halomonas and Chromo- recently as members of the genus Halomonas (FRANZ-
halobacter, and the genus Zymobacter, which comprised MANN et al., 1987; JAMES et al., 1990; BERENDES et al.,
non-halophilic microorganisms (FRANZMANN et al., 1988; 1996; ROMANO et al., 1996; MORMILE et al., 1999; DUCK-
DOBSON and FRANZMANN, 1996). Very recently, however, WORTH et al., 2000; BOUCHOTROCH et al., 2001; YOON et
ARAHAL et al. (2002) have proposed that the genus Carni- al., 2001; YOON et al., 2002). Halomonas canadensis,
monas should be included in this family. VREELAND et al. Halomonas israelensis, and Halomonas elongata ATCC
(1980) originally described the genus Halomonas with 33174 and DSM 3043 have, on the other hand, been
0723-2020/02/25/03-360 $ 15.00/0Phenotypic characterisation of Halomonas species 361
transferred recently to the genus Chromohalobacter as 20, 25 and 30% (w/v). The pH growth range was determined in
Chromohalobacter canadensis, Chromohalobacter israe- a similar way on MH media in which the pH had been adjusted
lensis and Chromohalobacter salexigens respectively to 5, 6, 7, 8, 9 or 10 with HCl or NaOH. The temperature range
(ARAHAL et al., 2001a, 2001b). was determined as above by incubating the bacteria at tempera-
tures of 4, 15, 20, 25, 32, 37 and 45 °C. The ability to grow
Our aim has been to describe phenotypically the type
anaerobically was evaluated on solid MH medium incubated in
strains of 21 species of the genus Halomonas, studying a jars with the Gas Pak Anaerobic System (BBL).
large number of characteristics by means of a reliable,
consistent, reproducible approach. For nutritional assays Biochemical tests
we used both classical and automated (BIOLOG) tests Oxidase reaction was performed according to KOVACS, 1956.
and therefore our results have allowed us to evaluate the Catalase was determined by adding 10 volumes of H202 to each
application of automated methods to the phenotypic strain culture (18 h) on solid MH medium.
identification of such bacteria. We have also selected Acid production from carbohydrates (adonitol, L-arabinose,
those tests which we consider to be most valuable for D-fructose, D-galactose, D-glucose, myo-inositol, lactose, mal-
tose, D-mannitol, D-mannose, D-melezitose, L-rhamnose, su-
identifying new isolates belonging to the genus
crose, D-salicin, D-sorbitol, sorbose and D-trehalose) was de-
Halomonas and for differentiating species from each tected in liquid MH medium. Doubtful results were checked in
other. The results of our study should go a long way to Leifson medium (1963) modified for marine bacteria and con-
clarifying the classification of bacteria included in this taining large quantities of sugars. The consumption of glucose,
heterogeneous genus. either by oxidation or fermentation, was studied in MH medium
to which 1% (w/v) glucose had been added. The medium was
put into Weimberg tubes and the inoculum deposited at the bot-
Materials and Methods tom of the tube with the aid of a Pasteur pipette; after inocula-
tion the tubes were sealed with 2 ml of agar (2% w/v) and 2 ml
Bacterial strains of paraffin. Respiration in fumarate, nitrate and nitrite was
We used the type strains of 21 validly published species be- studied by the same procedure, the medium being supplemented
longing to the genus Halomonas: Halomonas aquamarina DSM with potassium nitrate (2 g/l), potassium nitrite (2 g/l) or sodium
30161T; Halomonas campisalis ATCC 700597T; Halomonas cu- fumarate (0.8 g/l) (CALLIES and MANNHEIM, 1978).
pida CECT 5001T; Halomonas desiderata DSM 9502T; H2S production was tested in liquid MH medium supple-
Halomonas elongata CECT 4279T; Halomonas eurihalina mented with 0.01% (w/v) of L-cysteine; the indicator was a strip
ATCC 49336T; Halomonas halmophila ATCC 19717T; of paper impregnated with lead-acetate placed in the neck of the
Halomonas halodenitrificans CECT 5012T; Halomonas halodu- tube (CLARKE, 1953). To assay nitrate reduction 0.2% (w/v)
rans LMG 10144T; Halomonas halophila CCM 3662T; KNO3 was added to the liquid MH medium. Nitrite was detect-
Halomonas magadiensis NCIMB 13595T; Halomonas marina ed with naphthylamine/sulphanilic acid reagents and residual ni-
ATCC 25374T; Halomonas marisflavi JCM 10873; Halomonas trate with zinc dust; gas production was detected in inverted
maura CECT 5298T; Halomonas meridiana DSM 5425T; Durham tubes (SKERMAN, 1967).
Halomonas pacifica ATCC 27122T; Halomonas pantellerensis Gelatine hydrolysis was tested by flooding cultures on solid
DSM 9661T; Halomonas salina CECT 5288T; Halomonas sub- MH medium containing 1% (w/v) gelatine, with the Frazier
glaciescola DSM 4683T; Halomonas variabilis DSM 3051T and reagent (1926). Casein hydrolysis was indicated by a clear zone
Halomonas venusta ATCC 27125T. To determine intraspecific around bacterial growth on double-strength solid MH medium
variation for some selected species we studied 38 strains of plus an equal quantity of skimmed milk (BARROW and FELTHAM,
Halomonas halophila (QUESADA et al., 1980); 26 strains of 1993); negative results were confirmed using the Frazier
Halomonas salina (VALDERRAMA et al., 1991); 19 strains of reagent. Starch hydrolysis was tested by flooding cultures on
Halomonas eurihalina (QUESADA et al., 1984) and 4 strains of solid MH medium containing 1% (w/v) starch with Lugol’s io-
Halomonas maura (BOUCHOTROCH et al., 2001). dine (BARROW and FELTHAM, 1993). Hydrolysis of Tween 20
and Tween 80 was tested on solid MH medium supplemented
Maintenance and basal medium with 1% (w/v) of separately autoclaved Tween 20 and Tween
MH medium (QUESADA et al., 1983) with 7.5% (w/v) sea-salt 80 (SIERRA, 1957). Phosphatase production was tested on solid
solution (RODRÍGUEZ-VALERA et al., 1981) was used to maintain MH medium with 10 ml/l of a filtered, sterilised solution of
the strains, to prepare the inocula and as the basal medium for sodium phenolphthalein diphosphate (1%) by exposure to a
phenotypic tests, unless otherwise stated. The pH was adjusted concentrated ammonia solution (BAIRD-PARKER, 1963). DNAse
to 7.2 with 1M NaOH except for the media used for growing H. activity was revealed on DNA agar supplemented with 7.5%
magadiensis and H. campisalis, which were adjusted to pH 9. (w/v) sea-salt solution (RODRÍGUEZ-VALERA et al., 1981) by a
clear zone after flooding with 1N HCl (JEFFREIS et al., 1957).
Morphological tests Haemolysis was studied in solid MH medium supplemented
Gram staining and motility were determined in 24-hour cul- with 5% (v/v) lamb’s blood. Growth on MacConkey or Cetrim-
tures in liquid MH medium. Poly-β-hydroxybutyric acid (PHB) ide-agar was assessed on these media prepared with 7.5 %
staining was done in 5-day cultures using the method of STANIER (w/v) sea-salt solution. The lecithovitellin test was carried out
et al. (1966). The morphology and size of colonies and the pro- according to LARPENT and LARPENT-GOURGAND (1957) using
duction of diffusible and non-diffusible pigments and ex- solid MH medium. Tyrosine hydrolysis was revealed by clear
opolysaccharide were examined in MH medium after 3 days’ in- zones in cultures on solid MH medium containing 5 g/l tyrosine
cubation. (GORDON and SMITH, 1955); pigment production was also de-
tected in this medium. Aesculin hydrolysis was determined by
Physiological tests adding 0.1% (w/v) aesculin and 0.05% (w/v) ferric citrate to
Growth at different salt concentrations was determined by MH medium (BARROW and FELTHAM, 1993).
spreading 20 µl of each inoculum onto the surface of solid MH Selenite reduction was determined according to LAPAGE and
media with sea-salt concentrations of 0, 0.5, 1, 3, 5, 7.5, 10, 15, BASCOMB (1968) with a 4 g/l concentration of sodium hydrogen362 J. A. MATA et al.
selenite. Gluconate oxidation was tested according to SHAW and Determination of test scores
CLARKE (1955). The ONPG test was carried out by adding 250 To determine which tests were the most diagnostically con-
ml of a solution containing 6 g/l O-nitrophenyl-β-D-galactopy- clusive in the numerical analysis for differentiating between
ranoside in 0.01 M Na2HPO4 to 750 ml of liquid MH medium phena we applied Sneath’s formula (1980):
(LOWE, 1962). Indol production was tested in liquid MH medi-
um using Kovacs’ reagent (1928). Methyl red and Voges-
Proskauer were tested using methyl red and Barrit’s reagent Score =
(1936) respectively. Urea hydrolysis was detected with Chris-
tensen’s medium (1946) and phenylalanine deamination was as-
sessed on phenylalanine agar (Difco), both media prepared with where PiA, PiB and PiC are the probabilities of character i, in
7.5% (w/v) sea-salt solution (RODRÍGUEZ-VALERA et al., 1981). phena A, B and C, and q the number of taxa. The tests with
Lysine and ornithine decarboxylases were tested according to scores higher than 0.5 were considered to have most value for
BARROW and FELTHAM (1993). differentiating between the strains. With these tests we made a
second numerical analysis using the same methodology as that
Nutritional tests described above.
Nutritional assays were made using both classical and minia-
turized automated methods. For the former we used Koser Chi-squared analysis and contingency coefficient
medium (1923) as modified by VENTOSA et al. (1982), (%, w/v): To determine to what extent the phenotypic description of a
NaCl, 7.5; KCl, 0.2; MgSO4.7H2O, 0.02; KNO3, 0.1; given type species represents all the strains included in the species
(NH4)2HPO4, 0.1; and KH2PO4. 0.05 filter-sterilised substrate we performed a chi-squared analysis and the contingency coeffi-
(0.1% w/v) was added to this medium, with the exception of cient (Milton 1994) on four selected species previously described
carbohydrates, which were used at 0.2% (w/v). When amino by us: H. halophila, H. eurihalina, H. salina and H. maura.
acids were used as substrates the basal medium contained nei-
ther KNO3 nor (NH4)2HPO4. The organic compounds used as
sole source of carbon and energy or sole source of carbon, nitro- Results
gen and energy were the following: aesculin, L-arabinose, D-cel-
lobiose, D-fructose, D-galactose, D-glucose, lactose, maltose, D- The type strains of 21 species belonging to the genus
mannose, D-melezitose, D-raffinose, L-rhamnose, ribose, D- Halomonas were submitted to extensive morphological,
salicin, starch, D-trehalose, D-xylose; adonitol, ethanol, glyc- physiological, biochemical, nutritional and antimicrobial
erol, myo-inositol, D-mannitol, sorbitol; acetate, citrate, for- susceptibility tests, using the same growth conditions for
mate, fumarate, gluconate, malonate, propionate, succinate; L- all the strains, except for H. magadiensis and H. camp-
alanine, L-cysteine, L-histidine, DL-isoleucine, L-lysine, L-me- isalis, as we have indicated above. These conditions were
thionine, L-serine and L-valine.
chosen as being suitable for the growth of the wide range
Nutritional tests on 95 substrates were also performed with
the BIOLOG GN2 automated identification system (Hayward,
of species in question. We also incorporated into our
California). Strains were grown on solid MH medium contain- study other strains of H. halophila, H. eurihalina, H. sali-
ing 7.5 % (w/v) total salts at 32 °C for 24 h and then suspended na and H. maura to make intraspecific comparisons.
in a sterile solution of 7.5 % (w/v) salts. Cell density was adjust- The results of the numerical analyses are shown in Fig-
ed to an OD590 of between 0.34 and 0.39. The BIOLOG GN2 ures 1 and 2. The estimated test error was less than 3% in
microplates were immediately inoculated with 125 µl of cell sus- either case, which would not affect the cluster analysis sig-
pension per well and incubated at 32 °C for 24 h. Substrate oxi- nificantly. No test was found to be so irreproducible (test
dation was measured with a microplate reader at 590 nm. variance ≥ 0.2) as to be excluded from the data matrix.
Nevertheless, we did not use the following parameters:
Antimicrobial susceptibility tests
colony size and appearance, rod-shape and size, and opti-
Antimicrobial susceptibility was assayed using the diffusion
mum salt concentration for growth due to their being
agar method (BAUER et al., 1966). The antimicrobial com-
pounds (BBL) were: amoxicillin 25 µg, ampicillin 10 µg, carbeni- somewhat subjective and difficult to evaluate. The tests re-
cillin 100 µg, cefoxitin 30 µg, cefotaxime 30 µg, chlorampheni- sponsible for most errors were those referring to lysine and
col 30 µg, erythromycin 15 µg, kanamycin 30 µg, nalidixic acid ornithine decarboxylases, gluconate oxidation and acid
30 µg, nitrofurantoin 300 µg, polymyxin B 300 UI, rifampycin production from L-arabinose, L-rhamnose and sorbose.
30 µg, streptomycin 10 µg, sulphamide 250µg, tobramycin 10µg Figure 1 contains the numerical analysis carried out
and trimetroprim-sulphametoxazol 1.25 µg–23.75 µg. using 205 differential phenotypic data of type strains
(Table 1). At a 62% similarity level most of the type
Numerical analyses strains were grouped into three phena (A, B and C), the
Differential characteristics of type strains were coded in binary main difference between them being the capacity of the
form; positive and negative results were coded as 1 and 0 respec- bacteria included within phenon A to produce acids from
tively; non-comparable or missing characteristics were coded as all the sugars tested. Halomonas campisalis, Halomonas
9. Data were submitted to cluster analysis using a simple match-
desiderata and Halomonas subglaciescola formed a sepa-
ing coefficient (SSM) (SOKAL and MICHENER, 1958) and clustering
was achieved by the unweighted-pair-group method of associa-
rate cluster (phenon C) due to their using few substrates
tion (UPGMA) (SNEATH and SOKAL, 1973). The TAXAN program when tested with either classical nutritional methods or
(Information Resources Group, Maryland Biotechnology Insti- the BIOLOG GN2 system. H. campisalis and H. desider-
tute, University of Maryland, College Park, HD20742, USA) was ata showed high similarity (more than 88%) between
used for computer analysis. Test error was evaluated by examin- each other. H. marisflavi did not cluster with the rest of
ing 10 strains in duplicate (SNEATH and JOHNSON, 1972). the strains at this 62% similarity level.Phenotypic characterisation of Halomonas species 363 Figure 2 contains the numerical analysis carried out teine. α-OH-butyric acid, α-keto-butyric acid, and α- using the 57 most conclusive diagnostic tests (those with keto-valeric acid from BIOLOG GN2 were also negative scores of 0.5 or above; Table 2). At a similarity level of for all strains. 62%, we obtained the same clustering as in Figure 1, al- Most of the type strains of Halomonas species were ca- though some strains changed their position slightly in re- pable of using numerous carbon (or carbon and nitrogen) lation to the other species. For example, H. marisflavi be- sources. Twenty-one compounds of the BIOLOG GN sys- comes more closely related to the species included in phe- tem were metabolised by more than 70% of the strains. non A because it is able to produce acids from sugars. On As we have mentioned above, Table 1 shows the differ- the other hand, H. eurihalina, H. variabilis, H. venusta ential characteristics of type strains of Halomonas species and H. magadiensis appear more closely related to each ordered according to the groups shown in Figure 1. In other. this table we refer to assays never made elsewhere, and All the strains studied were Gram-negative rods, pro- also to those results which do not agree with previously duced catalase, reduced selenite, and grew at pH values published data. The variations in the phenotypic features of between 8 and 10; at sea-salt concentrations of be- of the different strains of H. eurihalina, H. halophila, H. tween 3% and 15% (w/v) and at temperatures between salina and H. maura are indicated as percentages of 20 °C and 37 °C. Colonies on MH medium containing strains that give the same responses as the type strains. 7.5% (w/v) salts were circular, convex and slightly The results of the chi-squared test and contingency coeffi- opaque with completely smooth edges. They were 2 to 3 cient were 0.6746, 0.6727, 0.6695 and 0.6699 respec- mm in diameter after 72 h at 32 °C, except those from tively. EPS-producing bacteria, which formed larger colonies We also point out those phenotypic features that clear- (more than 5 mm). They tested negative for the follow- ly distinguish species from each other (Table 3), among ing: diffusible pigment, Indol, methyl red, Voges- which we indicate some single characteristics that differ- Proskauer, respiration on fumarate, growth with L-cys- entiate between given species, such as PHB accumulation, Fig. 1. Dendrogram showing the clustering of the type strains of 21 species of Halomonas on the basis of 205 phenotypic characteristics (95 studied by the BIOLOG System). The simple matching (SSM) coefficient and unweighted-pair- group method with average (UPGMA) cluster- ing were used.
364 J. A. MATA et al.
which separates H. halodurans from the other species, proved quite difficult for various reasons. In some cases the
phenylalanine deaminase, which was only positive for phenotypic description of a given species has been made on
Halomonas desiderata, and casein hydrolysis for the basis of a relatively low number of characteristics, as oc-
Halomonas halodurans. Halomonas marina was the only curs for example with Halomonas halodenitrificans
species resistant to rifampycin. As far as Halomonas (KOCUR, 1984), H. marina (COBET et al., 1970) and several
elongata is concerned, it can be clearly differentiated by other species described some years ago. Furthermore, the
its capacity to metabolise sebacid acid. Finally, characteristics selected by some authors to describe a
Halomonas marisflavi could easily be distinguished from species might be different from those selected by other au-
the other species because of three features: haemolysis thors; a case in point, the old Deleya species (BAUMANN et
(positive), lecithinase (positive) and its resistance to al., 1972; 1983) was described using nutritional data alone.
trimetroprim-sulphametoxazol, all of which explains why In addition, the methodology followed to determine a given
this species did not cluster at a 62% similarity level with characteristic may not always be the same, as for example
the rest of the species. with acid production from sugars. And finally, some au-
thors have not described the methods used; for example, H.
halmophila was originally described in 1940 as Flavobac-
Discussion terium halmephilum (VOLCANI, 1940); later, in 1990, DOB-
SON et al. added to the description of this microorganism by
In the course of our taxonomic studies we can verify that means of nearly a hundred further tests without, unfortu-
the majority of halophilic microorganisms isolated from hy- nately, describing the techniques used.
persaline habitats are Gram-negative rods with a respirato- In the review of halophilic microorganisms published
ry metabolism and are related to the genus Halomonas. by VENTOSA et al. (1998), the authors included a table
Nevertheless, the differentiation of Halomonas species on summarising the differential features among Halomonas
the basis of their phenotypic characteristics has so far species. For their purpose they relied upon data from
Fig. 2. Dendrogram showing clustering of type
strains of species of Halomonas on the basis of
57 phenotypic characteristics with scores of > 0.5.
The simple matching (SSM) coefficient and un-
weighted-pair-group method with average
(UPGMA) clustering were used.Table 1. Phenotypic characteristics of type strains of species of the genus Halomonas. 1. Halomonas aquamarina; 2. Halomonas meridiana; 3. Halomonas cupida; 4. Halomonas pantelleriensis; 5. Halomonas halmophila; 6. Halomonas elongata; 7. Halomonas variabilis; 8. Halomonas eurihalina; 9. Halomonas magadiensis; 10. Halomonas salina; 11. Halomonas halodenitrificans; 12. Halomonas maura; 13. Halomonas halophila; 14. Halomonas paci- fica; 15. Halomonas venusta; 16. Halomonas halodurans; 17. Halomonas marina; 18; Halomonas campisalis; 19. Halomonas desiderata; 20. Halomonas subglaciescola; 21. Halomonas marisflavi.
Table 1. Continued.
Table 1. Continued.
Table 1. Continued.
Table 1. Continued. Data shaded in grey indicate characteristics not analysed in previous descriptions; asterisks indicate results that are not in accordance with previous descriptions; bold characteristics indicate useful tests for distinguishing a particular species. a % of strains that react in the same way as the type strain. b When supplied as the sole source of carbon and energy or as the sole source of carbon, nitrogen and energy.
Table 2. Phenotypic characteristics with scores of 0.5 or above of type strains of species of the genus Halomonas. 1. Halomonas aquamarina; 2. Halomonas meridiana; 3. Halomonas cupida; 4. Halomonas halmophila; 5. Halomonas pantelleriensis; 6. Halomonas marisflavi; 7. Halomonas elongata; 8. Halomonas eurihalina; 9. Halomonas variabilis; 10. Halomonas venusta; 11. Halomonas magadiensis; 12. Halomonas halodenitrificans; 13. Halomonas maura; 14. Halomonas salina; 15. Halomonas halodurans; 16. Halomonas marina ; 17. Halomonas halophila; 18; Halomonas pacifica; 19. Halomonas campisalis; 20. Halomonas desiderata; 21. Halomonas sub- glaciescola.
Table 2. Continued.
a
When supplied as the sole source of carbon and energy.Table 3. Useful tests for distinguishing the type strains of species of the genus Halomonas.
1. Halomonas aquamarina; 2. Halomonas meridiana; 3. Halomonas cupida; 4. Halomonas halmophila; 5. Halomonas pantelleriensis; 6. Halomonas marisflavi; 7. Halomonas elonga-
ta; 8. Halomonas eurihalina; 9. Halomonas variabilis; 10. Halomonas venusta; 11. Halomonas magadiensis; 12. Halomonas halodenitrificans; 13. Halomonas maura; 14. Halomonas
salina; 15. Halomonas halodurans; 16. Halomonas marina; 17. Halomonas halophila; 18; Halomonas pacifica; 19. Halomonas campisalis; 20. Halomonas desiderata; 21. Halomonas
subglaciescola.
a
When supplied as the sole source of carbon and energy or as the sole source of carbon, nitrogen and energy.Phenotypic characterisation of Halomonas species 373
original publications of these species and therefore 40% phena shown in Figure 1 by means of the analysis de-
of the species lacked data for more than 30% of the dif- scribed by SNEATH (1980). Using the selected tests (those
ferential characteristics cited. ARAHAL et al. (2002) have with scores of 0.5 or above, cf. Table 2), we made a sec-
recently published a phylogenetic study of the Halomon- ond numerical analysis, from which we obtained the
adaceae family, in which they point out the need for phe- same clustering of species (Figs 1 and 2). This will allow
notypic data to be used together with their results in a re- future researchers to reduce the number of phenotypic
classification of the genus Halomonas. data needed to conduct a numerical taxonomic study of
For all these reasons we have characterised the type such bacteria.
strains of 21 species of Halomonas via 234 phenotypic We then went on to make intraspecific comparisons
tests. Moreover, we have used the same methods through- between the type strains of four species described by us,
out, performing the phenotypic tests under consistent, H. halophila (QUESADA et al., 1984), H. eurihalina (QUE-
standard conditions to obtain reproducible results (VAN- SADA et al., 1990, H salina (VALDERRAMA et al. 1991) and
DAMME et al., 1996). It is well known for example that H. maura (BOUCHOTROCH at al., 2001) and other strains
the salt concentration might modify the results of some of these species. The chi-squared test and that of the coef-
phenotypic tests; in this way, H. halodurans has been ficient of contingency revealed a high degree of homo-
shown to suffer some phenotypic changes when the salt geneity between the type strains and the rest of the strains
concentration was lowered from 8% to 1.7% (w/v) (HER- classified within these species, since the contingency coef-
BERT and VREELAND, 1987) Thus we chose to carry out ficient was in all cases close to 0.67 in an interval from 0
our assays at a salt concentration of 7.5 % (w/v), the to 0.7. Furthermore, the strains belonging to the same
most appropriate concentration for the growth of the species were highly similar to each other, as the results
wide range of Halomonas strains we are dealing with. shown in Table 1 indicate.
As can be seen, most of the tests gave the same results Our work here represents a thorough phenotypic anal-
as those published elsewhere by the authors who original- ysis and the first numerical taxonomic study for the type
ly described these strains. Nevertheless, some of our re- strains of Halomonas species. To the best of our knowl-
sults (indicated with asterisks in Table 1) did not coincide edge this is the first time that the BIOLOG system has
with the previously published data, although these cases been applied to this group of halophilic bacteria. The oxi-
only amounted to 4% of the total. Most of these differ- dation of some substrates included in BIOLOG, together
ences might be due to the different methods used. with acid production from sugars, leads us to group
We determined some characteristics by a second tech-
Halomonas species into three distinct phena. Further to
nique to confirm doubtful results and those of our results
this, enzyme-activity assays and biochemical and physio-
which were not in accordance with data obtained from
logical tests allow us to distinguish quite clearly between
the original reports. For example, to test acid production
individual species.
from sugars, we used two different culture media: Leifson
medium, with a low peptone and high carbohydrate con- We believe that the data contained in this report
tent (LEIFSON, 1963), and MH basal medium (QUESADA et should be very useful to other researchers working with
al., 1983), which had more peptones and a lower carbo- halophilic microorganisms. Firstly, the data included in
hydrate content. In this way our results indicated that H. the Tables should be of great use to future researchers for
elongata was not able to ferment glucose, just as other identifying an isolate of Halomonas either as a new taxon
authors such as VENTOSA et al. (1998) have also observed. or a member of an already described species. Secondly,
We also found some differences in the nutritional tests we have determined how to carry out a reliable numerical
carried out using classical and automated methods. These analysis without using an inordinate number of tests.
discrepancies in the results were probably due to their dif- Many of the tests selected for such an analysis can be
ferent operational modes: increased growth by assimila- made in BIOLOG systems and are therefore easy, reliable
tion of carbon sources (classical method) or oxidation of and reproducible (Table 2). Finally, we have been able to
substrates via an electron transport chain (BIOLOG discover many essential phenotypic characteristics that
method) (RUGER & KRAMBECK, 1994). So we distinguished distinguish Halomonas species from each other (Table 3).
two sets of data, substrate use and substrate metabolism, These characteristics can be determined by tests which
as recommended by RUGER & KRAMBECK (1994). are relatively simple but at the same time are highly effi-
In the case of Halomonas halodurans we studied two cient for species differentiation. The tests reveal physio-
cultures from two different collections, DSM and LMG. logical, biochemical and nutritional features as well as
At first we worked with H. halodurans DSM 5160T but antimicrobial susceptibility.
we found quite significant differences between our initial Our study also provides the phenotypic data neccesary
results and the data published by the authors of the first to complement phylogenetic studies already carried out
report, such as acid production from some sugars, H2S by other authors working in the field, since it is becoming
production and aesculin and Tween 80 hydrolysis and clear that the taxonomic position of some of species of
thus we decided to use Halomonas halodurans LMG the genus Halomonas should be reconsidered (ARAHAL et
10144T to carry out this study. Other authors have also al., 2002).
observed similar discrepancies between strains from dif-
ferent sources JAMES et al. (1990). Acknowledgements
We subsequently determined which tests gave the The authors thank A. VENTOSA and D. R. ARAHAL for
most conclusive results in differentiating between the supplying some of the strains and A. VENTOSA for his crit-374 J. A. MATA et al.
ical reading of our manuscript and valuable discussions. CHRISTENSEN, W. B.: Urea descomposition as a means of differ-
This research was supported by a grant from the Direc- entiating Proteus and paracolon cultures from each other and
ción General de Investigación Científica y Técnica from Salmonella and Shigella. J. Bacteriol. 52, 461–466
(BOS2000-1519) and from the Plan Andaluz de Investi- (1946).
CLARKE, P. H.: Hydrogen sulphide production by bacteria. J.
gación, Spain. We also thank A. L. TATE for revising the
Gen. Microbiol. 8, 397–407 (1953).
English text and M. VALDERRAMA for helping us with the COBET, A. B., WIRSEN, C., JONES, G. E.: The effect of nikel in
statistical interpretation of our data. marine bacterium. Arthrobacter marinus sp. nov. J. Gen. Mi-
crobiol. 62, 159–169 (1970).
DOBSON, S. J., FRANZMANN, P. D.: Unification of the genera De-
References leya (BAUMAN et al., 1993), Halomonas (VREELAND et al.,
1980), and Halovibrio (FENDRICH 1988) and the species
AGUILAR, A.: Extremophiles research in the European Union: Paracoccus halodenitrificans (ROBINSON and GIBBONS 1952)
from fundamental aspects to industrial expectations. FEMS into a single genus Halomonas, and placement of the genus
Microbiol. Rev. 18, 89–92 (1996). Zymobacter in the family Halomonadaceae. Int. J. Syst. Bac-
AKAGAWA, M., YAMASATO, K.: Synonymy of Alcaligenes aqua- teriol. 46, 550–558 (1996).
marinus, Alcaligenes faecalis subsp. homari, and Deleya DOBSON, S. J., JAMES, S. R., FRANZMANN, P. D., MCMEEKIN, T.
aesta: Deleya aquamarina comb. nov. as the type species of A.: Emended description of Halomonas halmophila (NCMB
the genus Deleya. Int. J. Syst. Bacteriol. 39, 462–466 (1989). 1971T). Int. J. Syst. Bacteriol. 40, 462–463 (1990).
ARAHAL, D. R., GARCÍA, M. T., LUDWIG, W., SCHLEIFER, K. H., DUCKWORTH, A. W., GRANT, W. D., JONES, B. E., MEIJER, D.,
VENTOSA, A.: Transfer of Halomonas canadensis and MÁRQUEZ, M. C., VENTOSA, A.: Halomonas magadii sp. nov.,
Halomonas israelensis to the genus Chromohalobacter as a new member of the genus Halomonas isolated from a soda
Chromohalobacter canadensis comb. nov. and Chromo- lake of the East African Rift Valley. Extremophiles. 4, 53–60
halobacter israelensis comb. nov. Int. J. Syst. Evol. Microbiol. (2000).
51, 1443–1448 (2001a). FENDRICH, C.: Halovibrio variabilis gen. nov. sp. nov., Pseu-
ARAHAL, D. R., GARCÍA, M. T., VARGAS, C., CÁNOVAS, D., NIETO, domonas halophila sp. nov. and a new halophilic aerobic coc-
J. J., VENTOSA, A.: Chromohalobacter salexigens sp. nov., a coid eubacterium from Great Salt Lake, Utah, USA. System.
moderately halophilic species that includes Halomonas elon- Appl. Microbiol. 11, 36–43 (1988).
gata DSM 3043 and ATCC 33174. Int. J. Syst. Evol. Microbi- FRANZMANN, P. D., BURTON, H. R. Y MCMEEKIN, T. A.:
ol. 51, 1457–1462 (2001b). Halomonas subglaciescola, a new species of halotolerant bac-
ARAHAL, D.R. LUDWIG, W., SCHLEIFER, K.H., VENTOSA, A.: Phy-
teria isolated from Antarctica. Int. J. Syst. Bacteriol. 37,
logeny of the family Halomonadaceae based on 23S and 16S
27–34 (1987).
rDNA sequence analyses. Int. J. Syst. Evol. Microbiol. 52,
FRANZMANN, P. D., WEHMEYER, U., STACKEBRANDT, E.:
241–249 (2002).
Halomonadaceae fam. nov., a new family of the class Pro-
BAIRD-PARKER, A. C.: A classification of micrococci and staphy-
lococci based on physiological and biochemical tests. J. Gen. teobacteria to accommodate the genera Halomonas and De-
Microbiol. 30, 409–427 (1963). leya. System. Appl. Microbiol. 11, 16–19 (1988).
BARRITT, M. M.: The intensification of the Voges-Proskauer re- FRAZIER, W. C.: A method for the detection of changes in gelatin
action by the adition of α-naphthol. J. Pathol. Bacteriol. 42, due to bacteria. J. Infect. Dis. 39, 302–309 (1926).
441–445 (1936). GALINSKI, E. A., TINDALL, B. J.: Biotechnological prospects for
BARROW, G. I., FELTHAM, R. K. A. (eds.): Cowan and Steel’s. halophiles and halotolerant microorganisms. In: Molecular
Manual for the identification of medical bacteria, 3rd ed., biology and biotechnology of extremophiles (R. D. HERBERT,
Cambridge, Cambridge University Press 1993. R. J. SHARP eds.). London, Blackie 1992.
BAUER, A. W., KRIBY, W. M. M., SHERRIS, J. C., TURCK, M.: An- GORDON, R. E., SMITH, M. M.: Rapidly growing, acid fast bacte-
tibiotic susceptibility testing by a standardized single disk ria. II. Species description of Mycobacterium fortuitum Cruz.
method. Am. J. Clin. Pathol. 45, 493–496 (1966). J. Bact. 69, 502–507 (1955).
BAUMANN, L., BAUMANN, P., MANDEL, M., ALLEN, R. D.: Taxon- HEBERT, A. M., VREELAND, R. H.: Phenotypic comparison of
omy of aerobic marine eubacteria. J. Bacteriol. 110, 402–429 halotolerant bacteria: Halomonas halodurans sp. nov., nom.
(1972). rev., comb. nov. Int. J. Syst. Bacteriol. 37, 347–350. (1987).
BAUMANN, L., BOWDITH, R. D., BAUMANN, P.: Description of De- HORIKOSHI, K.: A new microbial world-Extremophiles. Ex-
leya gen. nov. created to accomodate the marine species Al- tremophiles 1, 1 (1997).
caligenes aestus, A. pacificus, A. cupidus, A. venustus and JAMES, S. R., DOBSON, S. J., FRANZMANN, P. D., MCMEENKIN,
Pseudomonas marina. Int. J. Syst. Bacteriol. 33, 793–802 T.A.: Halomonas meridiana, a new species of extremely halo-
(1983). tolerant bacteria isolated from Antartic saline lakes. System.
BERENDES, F., GOTTSCHALK, G., HEINE-DOBBERNACK, E., MOORE, Appl. Microbiol. 13, 270–278 (1990).
E. R. B., TINDALL, B. J.: Halomonas desiderata sp. nov., a new JEFFRIES, C. D., HOLTMAN, D. F., GUSE, D. G.: Rapid method for
alkaliphilic, halotolerant and denitrifying bacterium isolated determining the activity of microorganisms on nucleic acids.
from a municipal sewage works. System. Appl. Microbiol. J. Bacteriol. 73, 591–613 (1957).
19, 158–167 (1996). KOCUR, M.: Genus Paracoccus Davis 1969, 384al, pp. 399-402.
BOUCHOTROCH, S., QUESADA, E., MORAL DEL, A., LLAMAS, I., In: Bergey’s manual of determinative bacteriology (N. R.
BEJAR, V.: Halomonas maura sp. nov., a new moderately KRIEG, HOLT, J. G., eds.), vol. 1, Baltimore, Md, The
halophilic, exopolysaccharide-producing bacteria. Int. J. Syst. WILLIAMS and WILKINS Co. 1984.
Evol. Microbiol. 51, 1625–1632 (2001). KOSER, S. A.: Utilization of the salts of organic acids by the
CALLIES, E., MANNHEIM, W.: Classification of the Flavobacteri- colon-aerogenes group. J. Bacteriol. 8, 493–520 (1923).
um-Cytophaga complex on the basis of respiratory quinones KOVÁCS, N.: Eine Ereinfachte Methode zum Nachweiss der In-
and fumarate respiration. Int. J. Syst. Bacteriol. 28, 14–19 dolbildung durch Bakterien. Z. ImmunForsch. Exp. Ther. 55,
(1978). 311–314 (1928).Phenotypic characterisation of Halomonas species 375 KOVÁCS, N.: Identification of Pseudomonas pyocyanea by the SNEATH, P. H. A.: Basic program for the most diagnostic proper- oxidase reaction. Nature (London) 178, 703–704 (1956). ties of groups from an identification matrix of percent posi- LANE, D. J.: 16S/23S sequencing, pp. 115-175. In: Nucleic acid tive characters. Computers and Geosciences 6: 21–26 (1980). techniques in bacterial systematics (E. STACKEBRANDT, GOOD- SNEATH, P. H. A., JOHNSON, R.: The influence on numerical tax- FELLOW, M. eds.), Chichester, John Wiley and Sons 1991. onomic similarities of errors in microbiological test. J. Gen. LAPAGE, S. P., BASCOMB, S.: Use of selenite reduction in bacterial Microbiol. 72: 377–392 (1972). classification. J. Appl. Bacteriol. 31, 568–580 (1968). SNEATH, P. H. A., SOKAL, R. R.: Numerical taxonomy. The prin- LARPET, J. P., LARPENT-GOURGAND, M.: Mémento technique de ciples and practice of numerical classification, San Francisco, microbiologie, Paris, Technique et Documentation 1957. Freeman, Williams and Wilkins Co. 1973. LEIFSON, E.: Determination of carbohydrate metabolism of ma- SOKAL, R. R., MICHENER, C. D.: A statistical method for evaluat- rine bacteria. J. Bacteriol. 85,1183–1184 (1963). ing systematic relationships. Univ. Kansas Sci. Bull. 38, LOWE, G. H.: The rapid detection of lactose fermentation in 1409–1438 (1958). paracolon organisms by the demonstration of β-D-galactosi- STANIER, R. Y., PALLERONI, N. J., DOUDOROFF, M.: The aerobic dase. J. Med. Lab. Technol. 19, 21 (1962). pseudomonadads: a taxonomic study. J. Gen. Microbiol. 43, MILTON, J. S.: Statistical methods in the biological and health 159–271 (1966). sciences. Londres. MacGraw-Hill Companies Inc. 1994. VANDAMME, P., POT, B., GILLIS, M., DE VOS, P., KERSTERS, K., MORMILE, M. R., ROMINE, M. F., GARCIA, T., VENTOSA, A., BAI- SWINGS, J.: Polyphasic taxonomy, a consensus approach to LEY, T. J., PEYTON, B. M.: Halomonas campisalis, sp. nov., a bacterial systematics. Microbiol. Rev. 60, 407–438 (1996). denitrifying, moderately haloalkalophilic bacterium. System. VALDERRAMA, M. J., QUESADA, E., BEJAR, V., VENTOSA, A., Appl. Microbiol. 22, 551–558 (1999). GUTIERREZ, M. C., RUIZ-BERRAQUERO, F. Y RAMOS-CORMEN- QUESADA, E., VALDERRAMA, M.J., BÉJAR, V., VENTOSA, A., ZANA, A.: Deleya salina sp. nov., a moderately halophilic gram- GUTIERREZ M.C., RUIZ-BERRAQUERO, F., RAMOS-CORMEN- negative bacterium. Int. J. Syst. Bacteriol. 41, 377–384 (1991). ZANA, A.: Volcaniella eurihalina gen. nov., sp. nov., a moder- VENTOSA, A., NIETO, J.: Biotechnological applications and po- ately halophilic nonmotile gram-negative rod. Int. J. Syst. tentialities of halophilic microorganisms. World Journal of Bacteriol. 40, 261–267 (1990). Microbiology and Biotechnology 11, 85–94 (1995). QUESADA, E., VENTOSA, A., RUIZ-BERRAQUERO, F., RAMOS-COR- VENTOSA, A., GARCÍA, M. T., KAMEKURA, M., ONISHI, H., RUIZ- MENZANA, A.: Deleya halophila, a new species of moderately BERRAQUERO, F.: Bacillus halophilus sp nov., a moderately halophilic bacteria. Int. J. Syst. Bacteriol. 34, 287–292 halophilic Bacillus species. Syst. Appl. Microbiol. 12, (1984). 162–166 (1989). QUESADA, E., VENTOSA, A., RODRÍGUEZ-VALERA, F., MEGÍAS, L., VENTOSA, A., NIETO, J.J., OREN, A.: Biology of moderately RAMOS-CORMENZANA, A.: Numerical taxonomy of moderate- halophilic aerobic bacteria. Microbiol. Mol. Biol. Rev. 62, ly halophilic Gram-negative bacteria from hypersaline soils. J. 504–544 (1998). Gen. Microbiol. 129, 2649–2657 (1983). VENTOSA, A., QUESADA, E.,RODRÍGUEZ-VALERA, F., RUIZ-BERRA- RODRÍGUEZ-VALERA, F., RUIZ-BERRAQUERO, F., RAMOS-CORMEN- QUERO, F., RAMOS-CORMENZANA, A.: Numerical taxonomy of ZANA, A.: Characteristics of the heterotrophic population in moderately halophilic gram-negative rods. J. Gen. Microbiol. hypersaline environments of different salt concentration. Mi- 128, 1959–1968 (1982). crobiol. Ecol. 7, 235–243 (1981). VOLCANI, B.E.: Studies on the microflora of the Dead Sea. Ph D. ROMANO, I., NICOLAUS, B., LAMA, L., MANCA, M. C., GAMBA- Thesis, Jerusalem, The Hebrew University. 1940. CORTA, A.: Characterization of a haloalkalophilic strictly aer- VREELAND, R. H., LITCHFIELD, C. D., MARTIN, E. L., ELLIOT, E.: obic bacterium, isolated from Pantelleria island. System. Halomonas elongata, a new genus and species of extremely salt- Appl. Microbiol. 19, 326–333 (1996). tolerant bacteria. Int. J. Syst. Bacteriol. 30, 485–495 (1980). ROSENBERG, A.: Pseudomonas halodurans sp. nov., a halotoler- YOON, J-H., CHOI, S. H., LEE, K-C., KHO, Y. H., KANG, K. H., ant bacterium. Arch. Microbiol. 136, 117–123 (1983). PARK, Y-H.: Halomonas marisflavae sp. nov., a halophilic RÜGER, H-J., KRAMBECK, H-J.: Evaluation of the BIOLOG sub- bacterium isolated from the Yellow Sea in Korea. Int. J. Syst. strate metabolism system for classification of marine bacteria. Evol. Microbiol. 51, 1171–1177 (2001). System. Appl. Microbiol. 17, 281–288 (1994). YOON, J-H., LEE, K-C., KHO, Y.H., KANG, K.H., KIM, C-J., SHAW, C., CLARKE, P. H.: Biochemical classification of Proteus PARK, Y-H.: Halomonas alimentaria sp. nov., isolated from and Providence cultures. J. Gen. Microbiol. 13, 155–161 jeotgal, a traditional korean fermented seafood. Int. J. Syst. (1955). Evol. Microbiol. 52, 123–130S (2002). SIERRA, G. A.: Simple method for the selection of lypolytic activ- ity of microorganisms and some observations on the influence Corresponding author: of the contact between cells and fatty substrates. Antonie van Dr. E. QUESADA, Dpto. de Microbiología, Facultad de Farmacia, Leewenhoek 23, 15–22 (1957). Campus Universitario de Cartuja s/n., 18071 Granada, Spain SKERMAN, V. B. D.: A guide to the identification of the genera of Tel.: ++34-958-243871; Fax: ++34-958-246235; bacteria, 1st ed, Baltimore, Williams and Wilkins Co. 1967. e-mail: equesada@ugr.es
You can also read
