Microbial Ecology Fungi in the Hyporheic Zone of a Springbrook F. Ba¨rlocher1, L.G. Nikolcheva1, K.P. Wilson2 and D.D. Williams2 (1) Department of Biology, Mount Allison University, 63B York Street, Sackville, NB, Canada E3L 1G7 (2) Department of Life Sciences, University of Toronto at Scarborough, 1265 Military Trail, Scarborough, ON, Canada M1C 1A4 Received: 3 March 2005 / Accepted: 7 March 2005 / Online publication: 15 August 2006
Abstract
Eight bimonthly sediment core samples (n = 6) were collected, to a depth of 64 cm, from the hyporheic zone of a springbrook in southern Ontario, Canada. Sediment cores were divided into three to four sections, and organic matter was subdivided into six different categories. Twigs were the most common substrate, followed by roots, cedar leaves, wood, grass, and deciduous leaves. The contributions of deciduous and cedar leaves declined with depth, whereas that of wood increased. On each sampling date and from each section, three randomly chosen substrates 93 cm were examined for conidia of aquatic hyphomycetes. The number of identified species significantly decreased with depth, and was highest on deciduous leaves and lowest on wood. Season had no significant effect on species numbers. DNA from substrates was extracted, amplified with fungal primers, and differentiated into phylotypes with denaturing gradient gel electrophoresis (DGGE). Absence/presence patterns of phylotype were significantly affected by season but not by section level. Both season and section level significantly affected relative densities of the bands of the 10 most common phylotypes. Our data suggest that aquatic hyphomycetes and other fungi readily disperse within the hyporheic zone, and that their relative scarcity in this habitat is due to a lack of suitable substrates.
Introduction
In streams with porous bed materials, there is exchange between surface and subsurface water masses. Ecological processes within the former are now quite well understood [1, 8, 21, 41], but those in the subsurface are less well known. Environmental conditions within the interstices are unlike those found both on the streambed surface (benthic habitat) and in the true groundwater. InteracCorrespondence to: F. Ba¨rlocher; E-mail:
[email protected]
708
DOI: 10.1007/s00248-006-9102-4
tions of these water masses produce a three-dimensional subsurface region with unique physicochemical characteristics, termed the hyporheic zone [13]. This region supports distinct communities of invertebrates and microorganisms that extend, in a ragged-edged ecotone, downward and laterally into the bed [41]. These organisms combine to make the zone one of high biological activity intimately linked to the nature of the water passing through it. The hyporheic zone is becoming increasingly recognized for its role as a contributor/transformer/sink of nutrients [8, 11, 17, 38, 39] . In many streams, autumn-shed leaves represent the major input of energy that fuels lotic foodwebs. Aquatic fungi and, to a lesser extent, bacteria, play a significant role in the Bconditioning^ of this litter before its consumption by invertebrates [38, 40]. However, the degree and nature of processing of such food material within the interstices are less well known, although a limited number of studies have addressed the issue [9, 19, 23] and predictions have been made based on analogous aquatic habitats [37]. For example, interstitial bacteria growing in biofilms on substrate particle surfaces have been predicted to create a variety of microniches that may allow the coexistence of a high number of species and perhaps promote the activity of some otherwise poor competitors. Further, a variety of anaerobic respiratory pathways (e.g., nitrate, ferric ion, sulfate, and methanogenic) may exist, with biofilm dynamics allowing these to take place even in aerobic sediments. Anaerobic pathways may well account for a substantial amount of total hyporheic organic matter mineralization [37]. A significant knowledge gap exists in terms of the role of fungi in hyporheic dynamics. It is hypothesized that they will be important to the breakdown of buried particulate organic matter, which can account for a large proportion of total stream organic matter [22]. Several studies have indeed reported the occurrence of fungal hyphae in hyporheic habitats or aquifers [9, 10, 25, 36], but their identities or roles are virtually unknown. There is one report on the occurrence of aquatic hyphomycete conidia on glass beads
& Volume 52, 708–715 (2006) & *
Springer Science+Business Media, Inc. 2006
F. BA¨RLOCHER
ET AL.:
FUNGI
IN THE
HYPORHEIC ZONE
OF A
SPRINGBROOK
buried 10 cm deep in the sediment of a stream [6], and representatives of the same group also occur in groundwater wells [23]. Before its functional role can be meaningfully addressed, the structure of the hyporheic fungal community must be determined. This was the objective of the current study: we evaluated the occurrence of aquatic hyphomycetes by identifying conidia observed on particulate organic matter collected at different levels within the hyporheic zone. We complemented this traditional assay by molecular techniques allowing the detection of nonreproductive fungal structures. Based on what is known about aquatic hyphomycetes, we expected them to be closely tied to the occurrence of leaf material in oxygenated microhabitats. Methods Study Site. Valley Spring, Southern Ontario, Canada (43-450 N, 79-150 W), is a small springbrook, õ60 m long, 0.5–1.2 m wide, and 2–3.5 cm deep. It has a surface flow ranging between 1800 and 2300 L hj1 that arises from several sources, and a bed substratum comprising a mixture of coarse sand/silt, with õ20–30% macrophyte cover (chiefly Nasturtium officinale). Valley Spring is at an elevation of 152 m, in a mixed coniferous and deciduous woodland, and set in a temperate climate. For a full description of the study site and its ecology, see [20, 42]. The section sampled was approximately 7 m from the spring source and consisted of the main channel and bank habitats. Sample Collections. Substrates for fungal analyses were sampled using clear, acrylic core tubes (1.3 m long) with a 3.5 cm inner diameter. Samples were taken bimonthly, from June 2002 to August 2003, randomly along two transects that ran at right angles to the main channel and into each bank (three per transect; n = 6). To take a sample, the core tube was placed on the stream or bank surface and hit repeatedly with a 1 kg sledgehammer to drive it into the desired depth—a piece of wood placed over the top of the pipe prevented cracking. A small hand vacuum pump fitted with Tygon tubing (0.8 mm inner diameter) and a rubber stopper was then used to seal the top of the core tube and pumped until a vacuum was created. This prevented the sediment from falling out when the tube was extracted from the bed. A ruler was used to measure the height of the core tube protruding from the bed surface so that the penetration depth could be determined. Also, the height of the top of the sediment from the top of the core tubes was measured so as to provide a measure of sediment compression. The tube was then pulled out of the bed, laid on a sheet of clear plastic wrap, and the sediment was carefully pushed out with a 3.2-cm-diameter wooden dowel. The length of each core was immediately measured for volumetric calculations, wrapped in the plastic, and placed into a cooler. Samples were brought back to the laboratory within 1.5 h for processing.
709
In the laboratory, each core was sliced longitudinally with a sterile knife so that descriptions of each sediment layer could be made. Each visibly different sediment layer (based on color and grain size) was sliced width wise, denoted a numbered section (starting with 1 at surface), and its length was recorded. All organic matter 92 mm was removed from each layer with forceps and rinsed with Nanopure water (18 W) to remove invertebrates. After rinsing, the organic matter was placed on a Nanopure moist 110-mm-diameter Whatman no. 1 filter. A second moist filter was placed on top of the organic matter in preparation for its transport to Mount Allison University, New Brunswick, where sporulation and DNA analyses were conducted. Each layer was placed in a small Ziploc bag (õ1000 mL), with all of the layers from one sediment core being placed into a larger 4-L Ziploc bag. Transport via courier from Scarborough to Sackville typically took 2 days; keeping the substrates moist between filter papers minimized loss of viability of aquatic hyphomycetes. Once the organic matter subsamples had been removed from each core layer, the remaining sediment was placed in a small Ziploc bag and its wet weight was determined. After weighing, the sediment layer samples were placed into a 50-C oven until a constant dry weight was reached. After dry mass determination, each layer was sieved through a series of mesh sizes (2.0 mm, 1.0 mm, 600 mm, 250 mm, 125 mm, 90 mm, 45 mm, and G45 mm) for grain size analysis. Sieve data were converted to units to calculate mean sizes (T1 SD) from cumulative frequency distribution curves [27]. Grain sizes were also classified according to the Wentworth grain size classification scale [7]. Sporulation Analyses. Samples collected on June 12, August 19, October 25, December 13, 2002; and February 10, April 20, June 23, and August 22, 2003 were examined for sporulating aquatic hyphomycetes. The samples of organic matter received at Mount Allison University were grouped into six categories: leaves (mostly belonging to Acer sp.), twigs (diameter e3 mm), grass, cedar leaves (Thuja occidentalis L.), wood, and roots. Samples differed by transect (two levels), date (eight levels), and section (three or four levels). The significance of these three factors on relative abundances of the six substrates was analyzed with canonical analysis of principal coordinates (CAP [3]; in contrast to other multivariable analyses, this approach allows hypothesis testing). Data were transformed into Bray
