SEARCH

SEARCH BY CITATION

Fig. S1. Flow diagram summarizing the methodology. For the thin section method (conventional TEM) the samples were high-pressure-frozen, freeze-substituted, resin-embedded and then sectioned. The sections were first CARD-FISH-labelled and subsequently first observed under the epifluorescence microscope and then under the TEM. For the cryogenic method the TEM grids were prepared through coating with formvar and carbon, poly-L-lysine and colloidal gold. Samples were then plunge frozen on these grids, followed by cryogenic TEM (cryo-TEM). After TEM, the samples were thawed and fixed, followed by CARD-FISH and confocal laser scanning microscopy (CLSM).

Fig. S2. Cryo-TEM/CARD-FISH overlay images of C. crescentus as a function of electron dose. Each row contains a set of cryo-TEM and CLSM-correlated images. The electron dose used for the high-magnification TEM image within each set is listed to the left. Image columns 1 and 2 (the two left columns) represent cryo-TEM images of C. crescentus cells. Images in column 1 were obtained at 20 k× or 15 k×, and different electron doses as indicated. The images in columns 2, 3 and 4 represent exactly the same field of view. The images in column 2 are low-dose defocused diffraction cryo-TEM images, all obtained at approximately the same dose (∼2.3 e- nm−2), as a first step in the mapping of suitable targets onto the cryo-TEM finder grid. Images in columns 3 and 4 (the two right columns) were obtained by confocal scanning light microscopy (CLSM). Bright-field images are shown in column 3 and were used to visualize the colloidal gold (black dots) in order to overlay the cryo-TEM images with the confocal data. Column 4 (on the right) represents maximum intensity projections (z projections) of the fluorescence images with the signal from the EUB338 probe shown in green. Depending on the electron dose the strength of the observed fluorescence signal varies widely. The gain used for CLSM images was not constant, but rather optimized for each experimental run, and this is clearly reflected in different levels of background noise in displayed images. Yellow squares indicate the target cells: cells imaged by HR cryo-TEM, correlated with FISH. In several epifluorescent images (right column; rows B–E, H–L) this same area, centred in the cell of interest, is repeated as an inset at the bottom right corner. To detect low FISH signals it was necessary to use high gain and visualize the target cells in images with high noise, within a small displayed view (as in the insets).

Fig. S3. Cryo-TEM/CARD-FISH overlay images of acid mine drainage (AMD) biofilm at different electron doses. Each row contains a set of cryo-TEM and CLSM-correlated images. The electron dose used for the high-magnification TEM image within each set is listed to the left. Image columns 1 and 2 (the two left columns) show cryo-TEM images of bacteria and archaea. Rows A–D present low-dose defocused diffraction imaging of areas never exposed to high-magnification imaging. The shape and size of the microorganisms are clear, along low resolution features and strong labelling possible. High-magnification images in column 1 were obtained at 20 k× or 15 k×, and at different electron doses as indicated. The images in columns 2, 3 and 4 represent exactly the same field of view. The images in column 2 are low-dose defocused diffraction cryo-TEM images obtained at a dose of approximately 2.3 e- nm−2 or lower (same for A–D), as a first step in the mapping of suitable targets onto the cryo-TEM finder grid. Images in columns 3 and 4 (the two right columns) were obtained by confocal scanning light microscopy (CLSM). Bright-field images shown in column 3 were used to visualize the colloidal gold (black dots) in order to overlay the cryo-TEM images with the confocal data. Images in column 4 are presented as maximum intensity projections (z projections; the EUB338 probe is shown in green, the ARC915 probe in red and DAPI in blue). Depending on the electron dose the strength of the observed fluorescence signal varies widely. In rows G and I archaeal cells are indicated with read arrows, adjacent to more visible bacteria; while in row G the archaeal cells, but not the bacterium, give a nice epifluorescence, in row I it is the opposite (identical electron dose). In row K, the low-magnification images include two bacteria; the bacterium imaged at high magnification gives no epifluorescence signal. In row L the bacterium imaged with a higher dose gives a nice epifluorescence signal. Row H was arbitrarily chosen for display of a control with three channels: green, red and blue. Yellow squares are as in Fig. S2.

Table S1. Growth medium for the biofilm samples.

Movie S1. Cryo-TEM tomographic reconstruction of an AMD archaeal cell. The microorganisms shown here and in main text Fig. 2I are the same species, and can be unequivocally identified based on the observed ultrastructure. They were imaged on different grids; to optimize the quality of tomographic reconstructions, Lacey Carbon-coated Formvar grids (Ted Pella 01881-F) were used for the three-dimensional (3D) cryo-TEM data. Due to the high electron dose used in 3D cryo-TEM, this organism cannot be used for FISH labelling. In general, the high-quality 3D structural information obtained by cryo-ET, from species or strains determined by correlative cryo-EM/CARD-FISH, can be mapped to special distribution and association patterns in resin sections (Figs 1 and 2).

FilenameFormatSizeDescription
EMI4_275_sm_supp_info.doc101KSupporting info item
EMI4_275_sm_movieS1.mpg8032KSupporting info item
EMI4_275_sm_fS1.tif564KSupporting info item
EMI4_275_sm_fS2.tif8707KSupporting info item
EMI4_275_sm_fS3.tif10376KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.