Characterisation of Streptomyces spp. isolated from water-damaged buildings

Abstract

Saprophytic Streptomyces spp. common in soil and producing biologically active compounds have been related to abnormal microbial growth in buildings where occupants may have health problems. We characterised 11 randomly selected water-damaged building isolates. The 16S rDNA sequence similarity was over 95.4% between strains so that seven, three, and one sequences had greater than 99.8, 99.7 and 99.7% similarity with those of Streptomyces griseus ATCC 10137 (Y15501), Streptomyces albidoflavus DSM 40455^T (Z76676), and Streptomyces coelicolor A3(2) (Y00411), respectively. Although differences in morphology, pigmentation, fatty acids, biological activity and pH tolerance indicated that strains did not necessarily match with three single phenotypes, they all appeared to belong to two or three branches of Streptomyces spp. most common environmental isolates.

Streptomyces spp., Phylogeny, Water-damaged building, 16S rRNA, Fatty acid

1 Introduction

Gram-positive, filamentous bacteria of the genus Streptomyces within the class Actinobacteria[1] are regarded as common saprophytic soil bacteria with a complex life cycle. They typically occur in soil as spores, which germinate and produce substrate mycelium under favourable nutritional conditions. Growth restriction is related to the aerial mycelium formation and sporulation, and the induction of secondary metabolite biosynthesis, including antibiotics and other biologically active compounds [2–4].

Mesophilic actinomycetes, mainly Streptomyces species, are frequently found in the indoor air of water-damaged buildings with abnormal microbial growth associated with a number of respiratory health problems in occupants, but not in references, when collected directly on tryptone–yeast extract–glucose (TYG) agar by Andersen six-stage impactor [5–10]. Their spores may induce biological responses in macrophages, including the production of reactive oxygen metabolites, NO and cytokines [11–13]. Because of the indicator role of Streptomyces spp. in water-damaged buildings [5,6], it was of interest to characterise a randomly selected, morphologically heterogeneous group of 11 strains differing in biological activity [11–12,13] by determining fatty acid profiles, pigmentation and 16S rRNA gene sequences, widely used for microbial identification and evolutionary studies [14–18].

2 Materials and methods

2.1 Bacterial strains and culture conditions

Streptomyces strains used in the study were isolated from indoor environments of separate water-damaged buildings by Andersen six-stage impactor (Graseby Andersen, Atlanta, GA, USA) on TYG agar (Bacto Plate Count Agar, Difco, Detroit, MI, USA) as presented [5]. Plates were incubated at room temperature of 20±2°C for 5–14 days. Reference samples were collected from outdoors, and from buildings not suffering water damage and abnormal microbial growth. To correct for the effect of outdoor microbial levels, indoor/outdoor ratios were used to statistically compare reference and water-damaged homes [5]. The buildings were located in different towns in central Finland. Streptomyces strains VTT E-99-1326 to VTT E-99-1331 were from the study of Nevalainen et al. [5]. Strain VTT E-99-1332 was isolated from a terrace house, strain VTT E-99-1333 from a single-family house, and strain VTT E-99-1334 from a University building. In exception, strains VTT E-99-1335 and VTT E-99-1336 were isolated directly from the building material of the University Hospital and a block of flats, respectively, using dilution plates. The identification of strains was based on growth characteristics, morphology, pigmentation, cycloheximide tolerance, and 16S rRNA gene sequences. Strains VTT E-99-1326 and VTT E-99-1331 were identified as Streptomyces californicus and Streptomyces anulatus, respectively, at the German Collection of Microorganisms and Cell Cultures (DSMZ) on the basis of partial 16S rDNA sequencing and physiological tests. Strains are stored at VTT Culture Collection (VTT, Technical Research Center of Finland, Biotechnology and Food Research). The media used in morphology and pigmentation studies were TYG agar, starch–casein–KNO₃ agar, actinomycete isolation agar with 0.01 g l⁻¹ of FeSO₄.7H₂O instead of 0.001 g l⁻¹[2], glycerol–arginine agar [19], and tryptone–soy agar (Unipath, Hampshire, UK). To prepare inocula Streptomyces spp. were grown on TYG agar until they sporulated. The plates were flooded with sterile, distilled water, and the resulting spore suspension was harvested [20]. The spore density was determined by acridine orange staining and microscope counting [21]. Six parallel 5-μl spots of a high density spore suspension (>5×10⁷ ml⁻¹) were inoculated on solidified media, and results on morphology and pigmentation were recorded after 5 and 14 days.

2.2 Isolation of chromosomal DNA

The spore suspension of 40 μl was inoculated to 10 ml of 2×TY medium containing 1.2% tryptone and 0.6% yeast extract, and the flask was shaken (200 rpm, 20±2°C) for 3 days. Cells were harvested by centrifugation (3000×g, 20 min), and lysed in two volumes of 10 mM TE buffer (pH 8.0) by incubating for 20 min at 65°C. Cell debris was removed by centrifugation, and the supernatant was extracted with phenol and chloroform:isoamyl alcohol (24:1 v/v) [24]. The DNA was precipitated by adding 1/10 volume of 3 M potassium acetate and two volumes of ethanol, separated by centrifugation (10 000×g), and dried. An aliquot of water-solubilised DNA was analysed in agarose gel electrophoresis to assess the DNA content [24].

2.3 PCR amplification of the 16S rDNA

The amplification of 16S rDNA was performed in an automated PTC-100™ programmable thermal controller (MJ Research, Watertown, MA, USA). The 50-μl reaction mixture contained c. 20 ng of template DNA, 31 pmol of each primer, 25 μmol of each deoxynucleoside triphosphate, 5 μl of 10×PCR buffer (200 mM Tris–HCl, pH 8.5; 100 mM (NH₄)₂SO₄; 20 mM MgSO₄; 1% Triton^® X-100; 1 mg ml⁻¹ bovine serum albumin) for Pfu DNA polymerase (Stratagene, Amsterdam, The Netherlands) and 10% (v/v) of dimethyl sulfoxide. The 5′-phosphorylated primers used were 5′-AGA GTT TGA TCC TGG CTC AG-3′ (positions 7–26, Streptomyces ambofaciens numbering) [22] and 5′-AAG GAG GTG ATC CAG CCG CA-3′ (positions 1525–1506) [23]. After 5 min heat denaturation at 98°C, 30 amplification cycles of 45 s denaturation at 95°C, 45 s of primer annealing at 58°C, and 3 min of primer extension at 72°C were performed. Finally, 72°C was maintained for 5 min, followed by cooling to 4°C.

2.4 Isolation and cloning of amplified 16S rDNA

The amplified fragments were purified in a 0.75% (w/v) low melting temperature agarose (Promega, Madison, WI, USA) gel in TBE buffer (0.045 M Tris-borate, 0.001 M EDTA) and stained with 0.67 μg ml⁻¹ of ethidium bromide [24]. The agarose containing the PCR product was excised and 100 μl of 10 mM TE buffer was added. The sample was incubated at 68°C for 10 min, extracted with 400 μl of phenol, and centrifuged at 13 000×g for 15 min. The aqueous layer was extracted with 400 μl of phenol:chloroform:isoamyl alcohol (25:24:1, v/v). After centrifugation at 13 000×g for 15 min, the DNA of water phase was precipitated with 1/10 volume of 3 M sodium acetate and two volumes of ethanol at −20°C overnight. The DNA precipitate was separated by centrifugation at 4°C for 15 min, washed with 1 ml of 70% ethanol, dried in vacuum, resuspended in 10 mM TE buffer, and stored at −20°C.

The ligation reaction mixture contained 85 ng of PCR-amplified, 5′-phosphorylated blunt-end fragment as an insert, 40 ng of dephosphorylated and SmaI-digested pUC19 vector (MBI Fermentas, Vilnius, Lithuania), and 1.5 U of T4 ligase in ligation buffer (0.04 M Tris–HCl pH 7.8, 0.01 M MgCl₂, 0.01 M dithiothreitol, 0.5 mM ATP) (MBI Fermentas). The mixture of 10 μl was incubated at 12°C for 16 h.

For transformation, a single colony of Escherichia coli DH5α was grown in 10 ml of 2×TY medium (200 rpm, 37°C). This suspension (100 μl) was inoculated to 10 ml of 2×TY medium, and grown as above until the absorbance OD₆₀₀ was 0.2–0.3. After centrifugation at 6000×g for 10 min at 4°C, the pellet was resuspended in 5 ml of 50 mM CaCl₂, and incubated on ice for 20 min. The centrifugation was repeated for 5 min, the pellet resuspended in 5 ml of CaCl₂, and incubated on ice for 15 min. After centrifugation (6000×g, 5 min) the pellet was resuspended in 1 ml of 50 mM CaCl₂, and incubated for 1 h on ice. The ligation mixture (5 μl) was added into 300 μl of cells, incubated on ice for 40 min, at 42°C for 2 min, and on ice for 2 min. 2×TY medium (0.5 ml) was added, the suspension was shaken at 100 rpm for 1 h at 30°C, and spread on the plate for incubation overnight at 30°C. The clones were screened for α-complementation by using X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) and IPTG (isopropyl-β-D-thiogalactopyranoside) [24].

2.5 Sequencing of 16S rDNA

Plasmids containing 16S rDNA fragment were isolated with Wizard Plus SV minipreps DNA Purification System (Promega) according to the manufacturer's instructions. Three clones of each strain were sequenced (A.I. Virtanen Institute, Kuopio, Finland) with the Thermosequenase Fluorescent Labeled Primer Cycle Sequencing kit with 7-deaza-gGTP, RPN 2538 (Amersham Pharmacia Biotech, Uppsala, Sweden), and A.L.F. or A.L.F.express DNA sequencer (Amersham Pharmacia Biotech). The forward sequencing primers were 5′-GTA AAA CGA CGG CCA GT-3′ (M13/pUC sequencing primer, MBI Fermentas), 5′-TGC CAG CAG CCG CGG TAA TA-3′ (nt 488–507, S. ambofaciens nomenclature) [22], and 5′-TGT TGG GTT AAG TCC CGC AA-3′ (nt 1055–1074), and the reverse primers were 5′-CAG GAA ACA GCT ATG AC-3′ (M13/pUC reverse sequencing primer, MBI Fermentas), 5′-CCG CCT ACG AGC TCT TTA-3′ (nt 560–543), and 5′-TTG CGG GAC TTA ACC CAA CA-3′ (nt 1074–1055). The 16S rDNA sequences of three independent clones of each strain were sequenced in both directions.

2.6 Phylogenetic analyses

The length of 16S rDNA sequences was 1519 nt for strains VTT E-99-1329, E-99-1330, E-99-1333 and E-99-1336, and 1518 nt for the other strains. The sequences were compared with the FASTA algorithm [25] to other rRNA gene sequences obtained from the Ribosomal Database Project [26], and the EMBL and GenBank databases. Sequences were aligned with PILEUP [27], and distance matrices were calculated according to the Kimura two-parameter option [28] with the DNADIST program of the PHYLIP 3.57c program package [29]. The phylogenetic tree was constructed by the neighbour-joining method [30], and the tree was plotted with a TREEVIEW program [31].

2.7 Nucleotide sequence accession numbers

The 16S rDNA sequences of Streptomyces strains VTT E-99-1326 to E-99-1336 have accession numbers AF429390–AF429400, respectively.

2.8 Fatty acid analysis

To determine fatty acid compositions, strains were grown on TYG agar (n=3) and the mycelium was collected. The fatty acid profiles were determined after saponification, methylation, and extraction by GLC-MS as presented previously [32,33].

3 Results

Streptomyces spp. were randomly selected from the collection of strains obtained in the course of a survey on abnormal microbial growth in homes, schools and offices. Although all strains produced extensively branching primary mycelium transformed during the life cycle into aerial mycelium bearing spores typical for streptomycetes, they clearly had species-specific differences in growth morphology and pigmentation and, consequently, in phenotype (Table 1). Part of the strains formed a secondary growth zone around the primary colony, or excreted water droplets, exudates. The most common mycelial colours were yellow and brown, in addition to green, red, grey and blue. The spores of all strains contained white, and occasionally grey and green.

1

The pigmentation and macroscopic morphological features of Streptomyces spp.

Strain	1326	1327	1328	1329	1330	1331	1332	1333	1334	1335	1336a
Mycelium colours
yellow	+	+	+	+	+	+	+	+	+	+	+
brown	+	+	+	+	+		+	+		+	+
green		+	+	+	+	+			+
grey									+
red				+				+
blue								+	+
Spore colours
white	+	+	+	+	+	+	+	+	+	+	+
grey	+			+	+			+	+
green			+	+		+
Macroscopic morphology
Colony appearance
smooth	+	+	+	+		+	+	+			ND
hairy, spiny, warty	+	+	+	+	+	+	+	+	+	+	ND
rugose	+	+	+			+	+	+		+	ND
Circle of secondary growth	+	+	+	+	+	+		+	+		ND
Exudates	+	+	+	+				+	+		ND

ND, not determined.

3.1 Fatty acids

Streptomyces strains 1332 and 1335 with white spores and no green pigment differed in fatty acid profiles from the other Streptomyces spp. (Table 2). Strain 1332 had the highest amounts of iso-branched and unsaturated fatty acids i-16:0, i-16:1 and i-17:1, whereas strain 1335 had the greatest proportions of 16:1, 17:1, i-15:0 and i-17:0. These two strains contained only low amounts of i-14:0 and a-15:0, which were typical fatty acids for the other strains. In strain 1336 the proportions of 16:1 and a-17:1 were higher and that of i-16:0 lower when compared to the other strains.

2

The fatty acid profiles of Streptomyces spp. isolated from water-damaged buildings

Strain	i-14:0	14:0	i-15:0	a-15:0	15:0	i-16:1	i-16:0	16:1	16:0	i-17:1	a-17:1	i-17:0	a-17:0	17:1	17:0	18:1	18:0
1326	6.5±0.3	0.3±0.1	8.2±0.2	23.5±0.4	1.1±0.1	3.7±0.1	25.9±0.8	5.1±0.1	5.3±0.1	4.2±0.1	3.7±0.1	3.1±0.2	6.3±0.2	2.0±0.2	0.3±0.1	0.8±0.1	0.1±0.1
1327	5.9±0.4	0.2±0.1	4.7±0.2	27.7±0.6	0.7±0.1	4.4±0.2	27.1±0.4	4.3±0.3	2.1±0.2	4.9±1.5	6.2±0.3	0.7±0.1	8.8±0.6	1.6±0.3	tr	0.6±0.1	tr
1328	6.3±1.1	0.5±0.1	7.0±0.1	27.2±1.2	1.2±0.3	5.4±0.6	21.9±1.7	7.5±0.9	3.9±0.3	2.6±1.5	5.3±0.5	1.1±0.2	7.0±0.7	2.5±0.6	tr	0.8±0.1	0.1±0.1
1329	7.4±1.4	0.5±0.1	8.6±0.2	21.1±1.8	1.3±0.3	10.1±1.3	19.9±2.6	5.2±0.4	8.5±0.8	3.3±0.7	4.3±0.2	1.5±0.3	5.3±0.9	2.5±1.1	tr	0.5±0.3	0.1±0.1
1330	9.1±1.6	0.5±0.1	8.1±0.4	21.6±1.7	1.4±0.1	9.7±0.8	18.1±2.2	3.6±0.4	9.9±0.4	2.7±0.3	4.1±0.5	1.2±0.2	5.1±0.7	4.1±1.1	0.2±0.1	0.5±0.3	0.2±0.1
1331	6.7±0.6	0.4±0.1	5.0±0.2	23.7±0.9	0.6±0.1	3.4±0.2	29.1±0.4	3.6±0.1	3.2±0.1	3.1±2.4	6.1±0.7	1.2±0.3	6.0±0.2	6.6±0.2	tr	1.4±0.1	0.1±0.1
1332	0.9±0.1	0.1±0.1	7.6±0.2	0.7±0.1	0.4±0.2	22.8±0.2	38.1±0.5	6.5±0.3	0.7±0.1	7.8±1.2	0.7±0.1	3.9±0.1	4.8±0.1	4.1±0.1	tr	0.7±0.1	0.1±0.1
1333	7.1±0.3	0.6±0.1	6.8±0.2	22.5±0.8	1.7±0.1	6.2±0.5	26.3±0.9	5.6±0.1	6.1±0.1	2.8±0.1	3.8±0.1	1.8±0.2	7.2±0.4	1.1±0.1	0.2±0.1	0.3±0.1	0.1±0.1
1334	11.3±0.1	0.3±0.1	6.2±0.2	28.2±0.9	2.2±0.2	7.7±0.3	18.8±0.3	7.3±0.3	4.7±0.1	3.2±0.3	5.0±0.2	0.7±0.1	2.4±0.1	2.1±0.3	tr	0.1±0.1	tr
1335	1.1±0.1	0.7±0.1	12.7±0.7	1.3±0.6	3.6±0.1	5.0±0.5	29.3±1.0	14.5±0.8	3.9±0.6	3.7±0.5	tr	7.9±0.2	2.9±0.2	12.4±1.0	1.0±0.1	tr	0.2±0.1
1336	3.3±2.0	0.3±0.1	7.0±0.1	24.8±1.3	1.6±0.2	9.2±1.1	14.0±1.1	14.8±1.0	5.6±0.5	3.4±0.2	7.4±0.3	0.8±0.1	5.8±0.6	1.9±0.2	0.1±0.1	tr	tr

All data are given as %±S.D.

3.2 Phylogenetic analysis

The phylogenetic tree constructed on the bases of 11 almost complete Streptomyces spp. 16S rRNA gene sequences detected three phyletic lines with close sequence similarity to Streptomyces griseus ATCC 10137 (Y15501), Streptomyces albidoflavus DSM 40455^T (Z76676) and Streptomyces coelicolor A3(2) (Y00411) (Fig. 1). The similarities in 16S rDNA sequences between strains affiliating with S. griseus and S. albidoflavus, S. griseus and S. coelicolor, and with S. albidoflavus and S. coelicolor were 95.4–95.7%, 96.6–96.7%, and 97.6–97.7%, respectively. The overall 16S rRNA gene sequence similarity of 11 Streptomyces spp. was greater than 95.4% (Table 3), and the G+C content was 58.2–59.4 mol%. No differences were detected between sequences of three independent clones of each strain.

1

The neighbour-joining tree based on nearly complete 16S rRNA sequences of 27 streptomycetes. The numbers at the nodes indicate the levels of bootstrap support based on a neighbour-joining analysis of 500 resampled data sets. The 16S rDNA sequences of Streptomyces strains VTT E-99-1326 to E-99-1336 have accession numbers AF429390–AF429400, respectively.

3

Similarity matrix for 16S rRNA gene sequences of Streptomyces spp., indicators of water-damaged buildings, and of strains S. griseus ATCC 10137 (Y15501), S. coelicolor A3(2) (Y00411), and S. albidoflavus DSM 40455^T (Z76676)

Strain	1331	1334	1326	1328	1327	1332	1335	S. g.	1333	S. c.	1329	S. a.	1330	1336
1331	100.0	100.0	99.9	99.9	99.9	99.9	99.9	99.9	96.6	96.3	95.6	95.5	95.7	95.7
1334		100.0	99.9	99.9	99.9	99.9	99.9	99.9	96.6	96.3	95.6	95.5	95.7	95.7
1326			100.0	99.8	99.8	99.8	99.8	99.8	96.7	96.4	95.6	95.5	95.7	95.7
1328				100.0	100.0	100.0	100.0	99.9	96.6	96.2	95.4	95.3	95.5	95.5
1327					100.0	100.0	100.0	99.9	96.6	96.2	95.4	95.3	95.5	95.5
1332						100.0	100.0	99.9	96.6	96.2	95.4	95.3	95.5	95.5
1335							100.0	99.9	96.6	96.2	95.4	95.3	95.5	95.5
S. griseus ATCC 10137								100.0	96.4	96.0	95.3	95.3	95.4	95.4
1333									100.0	99.7	97.6	97.5	97.7	97.7
S. coelicolor A3(2)										100.0	97.2	97.2	97.3	97.3
1329											100.0	99.9	99.9	99.7
S. albidoflavus DSM 40455T												100.0	99.9	99.7
1330													100.0	99.7
1336														100.0

The 16S rRNA gene sequences of strains 1326, 1327, 1328, 1331, 1332, 1334 and 1335 were closely affiliated with S. griseus (Fig. 1), the similarity index being 99.8–99.9% (Table 3). However, all sequences had A instead of G of S. griseus at 449 nt (S. ambofaciens nomenclature [27]) (Fig. 2). The identical 16S rRNA gene sequences of strains 1327, 1328, and 1335 had a second base change from C to T at nt 184, and strain 1332 additionally had a deletion at nt 79. The identical 16S rDNA sequences of strains 1331 and 1334 had a replacement of T of S. griseus with C at nt 77, and a deletion at nt 79. They grouped on a slightly separate branch with the sequence of strain 1326 which had an additional base change from A to G at nt 200 (Fig. 2).

2

Sequence alignment of 16S rRNA encoding DNAs for 11 strains of Streptomyces spp., indicators of water-damaged buildings, and for S. griseus ATCC 10137 (Y15501), S. albidoflavus DSM 40455^T (Z76676), and S. coelicolor A3(2) (Y00411). The sequences are aligned with the sequence of S. griseus ATCC 10137. The nucleotides that differ from those of S. griseus ATCC 10137 are shown; identical nucleotides are indicated by dots, and deletions by dashes. The numbers on the sequence indicate positions from the 5′ end of the 16S rRNA (S. ambofaciens nomenclature [22]).

The second 16S rDNA sequence cluster of Streptomyces strains 1329, 1330 and 1336 had over 99.7% homology to that of S. albidoflavus (Fig. 1). The 16S rDNA sequence of strain 1336 had 5′-CGTCTGC-3′ at the position of nt 177–183 instead of sequence 5′-TGTCCAT-3′ (Fig. 2), and strains 1329, 1330 and 1336 had T, C, and T, respectively, at nt 1247 where the S. albidoflavus sequence has N.

The 16S rDNA sequence of strain 1333 clustered next to that of S. coelicolor (Fig. 1) with a sequence identity of 99.7% (Table 3), due to the sequence 5′-GGT-3′ at the position nt 600–602 where the sequence of S. coelicolor contains 5′-CCA-3′ (Fig. 2). The only difference observed in sequences outside the variable regions was due to the presence of A in strain 1333 at nt 358 instead of G of S. coelicolor.

4 Discussion

The 11 Streptomyces spp. of this study were randomly selected from the collection of strains isolated from diverse geographical areas in Finland in the course of surveys on water-damaged buildings with abnormal microbial growth, including homes, schools and offices. Streptomyces spp. are frequently found in Finnish homes suffering water damage and connected mould growth, but not in references [5]. The 11 Streptomyces spp. characterised formed three phylogenetic clusters with seven, three and one 16S rDNA sequences similar to those of S. griseus ATCC 10137, S. albidoflavus DSM 40455^T and S. coelicolor A3(2), which belong to the subgroups Kitasatospora setae, S. albidofavus, and S. coelicolor, respectively, in the Ribosomal Database Project. In the numerical classification presented by Kämpfer et al [34] and including species defined by Williams et al. [35], S. griseus, S. albidoflavus and S. coelicolor all belong only to the first great cluster out of 15 major, 34 minor, and 40 single-member clusters. Although the major groupings are in line with the genetic data [1], the taxonomic situation within the genus Streptomyces is complex and yet unresolved, such as phylogenetically closely related strains may have the rich chemical, morphological and physiological diversity. Several strains of especially S. griseus and S. coelicolor have been characterised which may, in fact, reflect limited criteria in species identification with little evidence of novelty of strains and, on the other hand, the common occurrence of these strains in environmental isolates [1,34,35]. Therefore, Streptomyces spp. isolated from water-damaged buildings and related with mould problem cannot be regarded as a phylogenetically diverse group, but seemed to belong to the most common species of Streptomyces isolated from the environment.

The 11 strains were typical Streptomyces spp. with greater than 95.4% similarity in 16S rRNA gene sequences between each other and previously published Streptomyces spp. sequences (EMBL GenBank Data Libraries, Ribosomal Database Project). The almost complete 16S rRNA sequences were 1518–1519 nt in length, and had a G+C content of 58.2–59.4 mol%, in contrast to the total genomic DNA of 72–75 mol%. The fatty acid profiles were similar to those earlier determined for Streptomyces spp. [22,32,36–44]. Despite the fact that the strains formed three clusters with only small differences in 16S rDNA sequence within the clusters, clear strain-specific differences in growth morphology, pigmentation, biological responses induced by the spores [11–12,13] and pH tolerance [4] were apparent. S. griseus-like strains 1326 and 1331 were identified as S. californicus and S. anulatus, respectively, on the basis of biochemical tests (German Collection of Microorganisms and Cell Cultures, DSMZ). The fatty acid profiles of strains 1332 and 1335 differed from the other five S. griseus-like strains, whereas S. coelicolor A3(2)- and S. albidoflavus DSM 40455^T-like strains had fatty acid profiles quite similar to five S. griseus-like ATCC 10137 strains (Table 2,Fig. 1). The lack of correlation in the clustering between 16S rDNA sequence phylogeny and fatty acid analyses has also earlier been observed [45]. Altogether, the results indicate that Streptomyces spp. isolated from buildings did not necessarily match with three similar phenotypes.

Streptomycetes are not regarded as common indoor microbes, but are frequently found in buildings suffering from moisture problems [5,6]. According to the results Streptomyces spp. isolated from separate, randomly selected water-damaged buildings seemed to belong to the most common environmental isolates. Streptomycetes generally have a pronounced biological capacity to degrade complex organic compounds, such as starch, pectin and casein etc. [2], which may enable their growth on a variety of wetted building materials, and on fungal mycelium commonly associated with abnormal microbial growth [10]. The fact that streptomycetes often produce compounds controlling e.g. growth of other bacteria and fungi [2,3,6] enables a competition advantage for Streptomyces spp., and might be responsible for biological responses induced by spores in cell cultures [11–12,13]. It is not known whether there is a selective pressure for certain types of streptomycetes in water-damaged buildings, or this selection is biased by the method of sample collection and cultivation on TYG agar. The culture-based isolation methods are selective, do not differentiate between active and resting cells, and fast-growing microbes easily overgrow slow-growing actinomycetes [46]. The percentage of isolates obtained generally represents only a small part of the diversity in a natural community [47]. Nevertheless, independent of the degree of selection, the cultivation of indoor environment samples on TYG agar revealed that elevated levels of the most common streptomycete environmental isolates may occur in water-damaged buildings, and can thus be used as one method for screening indoor microbial problems.

Acknowledgements

The technical assistance of Mrs Ulla Kukkonen and the late Mrs Tuula Wallenius are gratefully acknowledged. This work was supported by the Finnish Work Environment Fund (Grants 96028 and 98103).