Land use intensity controls actinobacterial community structure

Abstract

Actinobacteria are major producers of secondary metabolites; however, it is unclear how they are distributed in the environment. DNA was extracted from forest, pasture and cultivated soils, street sediments (dust and material in place), and sediments affected by animal activity (e.g. guano, vermicompost) and characterised with two actinobacterial and a bacterial-specific 16S rDNA primer set. Amplicons (140/156) generated with the two actinobacterial-specific and amplicons (471) generated with bacterial-specific primers were analysed. Amplicons from actinobacterial-specific primer were disproportionately actinomycetal from animal-affected (soil) samples and street sediments and either verrucomicrobial (i.e. non-actinobacterial) and from a novel non-actinomycetal actinobacterial group for soils. Actinobacterial amplified ribosomal DNA restriction analysis and terminal restriction fragment length polymorphism fingerprints clustered by land use, with cultivated soils clustering apart from uncultivated soils. Actinobacterial amplicons generated with eubacterial primers were overwhelmingly from (116/126) street sediments; acidobacterial amplicons from soils (74/75). In two street samples, >90% of clones were actinomycetal. Actinomycetes are selected in terrestrial soils and sediments by cultivation, urbanisation and animal activity.

Figure 1

Neighbour-joining tree of 16S sequences generated with the F-Act/R-Bact primers of Heuer et al. [15]. Bootstrap values below 50 are not shown. Samples are colour coded as: forest/pasture soils (blue), insect- and earthworm-associated sediments (orange), street sediments (red). Unifrac lineage analysis was carried out on nodes A and B. For node A the P value was <0.0002 and observed/expected occurrences were for soils 53/37.7, street sediments 2/6.6 and animal affected sediments 16/26.7. For node B the P value was <0.0000 and observed/expected occurrences were for soils 17/35.1, street sediments 11/6.1 and animal affected sediments 37/28.8

Figure 2

Neighbour-joining tree of 16S clones generated with the F-Act/R-Act primers of Monciardini et al. (2003). Bootstrap valules below 50 are not shown. Simulated T-RFLP values are shown for all matches were sequence length allows. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green), insect- and earthworm-associated sediments (orange), street sediments (red). Unifrac lineage analysis was carried out on nodes A and B. For node A, the P value was <0.0000 and observed/expected occurrences were for soils 19/43, street sediments 25/14.5 and animal-affected sediments 42/28.5. For node B, the P value was <0.0000 and observed/expected occurrences were for soils 58/29, street sediments 0/9.9 and animal-affected sediments 3/19.1

Figure 3

UPGMA tree of amplified ribosomal DNA restriction analysis (ARDRA) patterns generated using the F-Act/R-Bact primers of Heuer et al. [15] and TaqI digestion. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green), insect- and earthworm-associated sediments (orange), street sediments (red). For sample descriptions, see Tables 1 and 2

Figure 4

UPGMA tree of terminal restriction fragment polymorphism (T-RFLP) patterns generated using the A3R primer of Monciardini et al. [30] F-Act/R-Act primers and HhaI digestion. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green), insect- and earthworm-associated sediments (orange), street sediments (red). For sample descriptions, see Tables 1 and 2

Figure 5

UPGMA tree of terminal restriction fragment polymorphism (T-RFLP) patterns generated using the F-243 primer of Monciardini et al’s [30] F-Act/R-Act primers and HhaI digestion. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green), insect- and earthworm-associated sediments (orange), street sediments (red). Two sequenced clones are also included for comparison. For sample descriptions, see Tables 1 and 2

Figure 6

Redundancy analysis ordination plots of fingerprint patterns for all characterised samples. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green), insect- and earthworm-associated sediments (orange), street sediments (red). a Amplified ribosomal DNA restriction analysis (ARDRA) patterns for all characterised samples. b Reverse terminal restriction fragment polymorphism (T-RFLP) patterns for all characterised samples. c Forward terminal restriction fragment polymorphism (T-RFLP) patterns for all characterised samples. d Amplified ribosomal DNA restriction analysis (ARDRA) patterns for all soils. Samples are colour coded as: forest/pasture soils (blue), cultivated soils (sea green). e Reverse terminal restriction fragment polymorphism (T-RFLP) patterns for all soils. f Forward terminal restriction fragment polymorphism (T-RFLP) patterns for all soils

Figure 7

Neighbour-joining tree of Actinobacterial 16S clones generated with the F-Act/R-Act primers of Marchesi et al. [28]. Bootstrap values below 50 are not shown. Samples are colour coded as: soils (blue), insect- and earthworm-associated sediments (orange), street sediments (red)

Figure 8

Unifrac Jackknife environmental clustering of the 12 eubacterial clone libraries generated with the primers of Marchesi et al. (1998) using the weighted Unifrac algorithm. Jackknife values below 50 are not shown. Samples are colour coded as: forest soils (blue), cultivated soils (sea green), street sediments (red)