OUP user menu

Actinomycete-flora associated with submersed freshwater macrophytes

Debra L. Wohl, J Vaun McArthur
DOI: http://dx.doi.org/10.1111/j.1574-6941.1998.tb00499.x 135-140 First published online: 1 June 1998


Aquatic vegetation from three stream sites located within the US Department of Energy's Savannah River Site, South Carolina was sampled to determine the presence, distribution and diversity of actinomycetes on submersed macrophytes. At individual sampling sites, small pieces of submersed aquatic plants (Sparganium americanum, Ranunculus pusillus and Microanthemum umbrosum) were clipped and placed in sterile sampling containers. In the laboratory, plant fragments were subjected to a wide array of selective measures (5 pretreatments×8 media×2 antibiotic regimes×2 temperatures) in order to obtain the greatest diversity and best isolation methods. A total of 433 actinomycete colonies was isolated, which included 32 distinct strains. The streams, plants and plant tissue types (healthy, necrotic or dead) all had an effect on the presence, distribution and diversity of the actinomycetes.

  • Actinomycete
  • Freshwater macrophyte
  • Stream
  • Selective medium
  • Isolation technique

1 Introduction

The presence, distribution and diversity of actinomycetes associated with submersed freshwater plants has not been investigated. The few existing surveys of aquatic actinomycetes (reviewed by [13) have concentrated on littoral zones of reservoirs and rivers where high levels of decomposing organic matter persist. Such studies are primarily focused on tastes, discolorations and odors associated with drinking water [411. More recent surveys of actinomycetes in aquatic environments are directed at their potential value as producers of novel enzymes and metabolites [1,12,13. Despite recent efforts, most of the studies cited above still refer to these microorganisms as a group of filamentous bacteria rather than attempting to understand their diversity or ecological role.

Several reasons may exist for why the diversity and ecology of actinomycetes in aquatic environments have been neglected. Primarily, isolation of actinomycetes relative to other bacteria tends to be difficult. The filamentous nature of actinomycetes allows them to strongly adhere to their substratum, making them difficult to remove from field materials and in subsequent culture transfers. Also related to their filamentous form and spore production, isolation and culturing methods for these microorganisms often require isolation techniques more similarly aligned to those used for filamentous fungi (e.g., baiting, leaf washing) than bacteria. Furthermore, the long incubation times that actinomycete colonies require often result in overgrowth by other bacteria and fungi or for plates to be discarded before their appearance. Some researchers [1,1416 have argued that many actinomycetes are not truly aquatic, but are rather an artifact of runoff and spore survival. In contrast, the findings of Makkar and Cross [17], Aumen [18], Johnston and Cross [15], Raschke et al. [8], Willoughby [19], Higgins and Silvey [20], and Roach and Silvey [21] support the existence of aquatic actinomycetes. Their slow growth and inability to dominate plates rapidly has often led to them being overlooked. The lack of data regarding actinomycetes from aquatic environments may lead to the general assumption that they are ecologically unimportant.

As the need to better understand microbial communities and interactions, and to find novel enzymes and metabolites increases, we believe that actinomycetes from freshwater habitats should not be overlooked. This study was conducted as an initial survey of their presence, distribution and diversity on submersed vegetation in two streams near Aiken, South Carolina. As aquatic actinomycetes may have important differences from their terrestrial counterparts, a wide array of isolation techniques was used to determine which methods were most suitable for collection of aquatic actinomycetes.

2 Materials and methods

During early July through late September 1995, samples were collected from three sites within two small southeastern streams located on the coastal plain near Aiken Co., South Carolina at the US Department of Energy's Savannah River Site. Each stream site is a low gradient (<2%), sandy bottomed stream with aquatic plants patchily distributed (Table 1). At individual sampling sites within each stream, small pieces of submersed aquatic plants were clipped and placed in sterile sampling containers. The three aquatic plants sampled were (1) Sparganium americanum (Bur-reed), a long, slender, single blade leaf; (2) Microanthemum umbrosum (Shade mudflower), a small, bright green, orbicular leafed plant; and (3) Ranunculus pusillus, a broad heart-shape leafed plant. Collections of plant material were transported to the laboratory within 3 h of collection.

View this table:

Physical variables of the stream sites sampled on the US Department of Energy's Savannah River Site, South Carolina for the collection period of July through September 1995

StreamTinkerUpper Three RunsUpper Three Runs
Latitude33° 22′ 30″33° 21′ 28″33° 17′ 00″
Longitude81° 32′ 30″81° 37′ 16″81° 17′ 30″
Temperature range20.6–23.520.6–2418–22.6
pH range6.0–6.54.8–5.04.8–5.3
Macrophyte(s)S. americanumS. americanumM. umbrosum
R. pusillus

In the laboratory, collected plant material was aseptically cut into 0.5×1 cm fragments and recorded as to tissue type. Three tissue types were defined: healthy, necrotic/yellowing, and dead. Plant fragments were then subjected to an array of selective treatments in order to detect the greatest actinomycete diversity and document the best means for future collections from this environment.

One of five pretreatments was applied to the plant material prior to plating. The pretreatments were: (1) incubation at 100°C for 0.2 h, (2) incubation at 60°C for 2 h, (3) bath sonication for 30 s, (4) tip sonication for 30 s and (5) rinsing the plant material in a concentrated antibiotics solution (200 mg ml−1 cycloheximide and 50 mg ml−1 polymixin B sulfate). The five pre-treated types of plant tissue were then placed on 8 media, each with antibiotics and without (Table 2). Plates were incubated at either room temperature (∼18°C) or 27°C for up to 35 days to allow for actinomycete growth.

View this table:

Media employed in this study

Actinomycete isolation agara,bDifco®
Chitin agara,b25 g finely ground practical grade chitin, 0.5 g yeast extract, 15 g agar, 40 mM KH2PO4, 40 mM K2HPO4, 1 l dI; Adjust pH to 6.75, [23]
Colloidal chitin agara,b2 g colloidal chitin, 0.02 g CaCO3, 0.01 g FeSO4-7H2O, 1.71 KCl, 0.05 g MgSO4-7H2O, 1.63 g Na2SO4, 15 g agar, 1 l dI
M3 agara,c0.73 g Na2SO4, 0.47 g KH2PO4, 0.2 g sodium proprionate, 0.1 g KNO3, 0.1 g MgSO4-7H2O, 0.02 g CaCO3, 0.2 mg FeSO4-7H2O, 0.18 mg ZnSO4-7H2O, 0.02 mg MnSO4-4H2O, 980 ml dI. Add filter sterilized: 0.04 g cycloheximide per 10 ml dI and 0.04 g thiamine HCl per 10 ml dI, [14]
Nutrient agara,d8 g nutrient broth (Difco®), 15 g agar, 1 l dI; Adjust pH to 6.9
Starch casein agara,d10 g potato starch, 1 g casein, 0.5 g K2HPO4, 15 g agar, 1 l dI
Trypticase soy agara,dDifco®
Water agara,e0.0001 g yeast extract, 15 g agar, 1 l dI
  • Media were prepared with: a no antibiotics; b 100 mg ml−1 cycloheximide; c 20 mg ml−1 polymixin B sulfate; d 50 mg ml−l nystatin; e 75 mg ml−1 cycloheximide and 20 mg ml−1 polymixin B sulfate.

Whenever possible, actinomycetes were classified to genera. Initial screening of all isolates was based on the Gram-stain test. All Gram-positive and Gram-variable isolates were maintained. Taxonomic determinations were done on the basis of morphology using a Nikon Microphot FXA microscope in conjunction with the chemotaxonomic methods of Lechevalier and Lechevalier [22]. Briefly, actinomycetes were grown in yeast-dextrose broth (10 g yeast extract, 10 g dextrose, 1 l dI water), harvested, and analyzed chromatographically for isomeric diaminopimelic acid configurations and for whole-cell sugar composition. Chromatography was performed on flexible cellulose-coated thin-layer chromatography plates (Selecto Scientific®, Norcross, GA). Twenty percent of the distinct isolates were later confirmed by an outside agency (Microbial ID®, Inc., Newark, DE).

3 Results and discussion

During the course of this study, 433 actinomycete colonies were isolated. Morphological and chemotaxonomic characterizations according to Lechevalier and Lechevalier [22] identified 32 distinct actinomycete strains (Table 3). More than 45% of all actinomycete colonies isolated were Streptomycetes. Of the 32 distinct strains identified, 34% were strains of Streptomyces, while Pseudonocardia, Nocardia, Micromonospora and Actinoplanes each comprised an additional 10% of the diversity.

View this table:

The number of distinct actinomycete strains per plant type isolated between July and September 1995 at the Savannah River Site, South Carolina

LocaleSite 1Site 2Site 3Total number of distinct strains per genera
Plant typeS. americanumS. americanumM. umbrosumR. pusillus
  • Asterisk (*) indicates that the strain was not found at any other locale.

The presence, distribution and diversity of actinomycetes collected were dependent not only on the aquatic vegetation sampled, but also on the streams and plant tissue types sampled. Despite equal sampling efforts at all three sites, most actinomycetes were collected from site 1. The most evident difference between site 1, located within Tinker Creek, and the two other sites is higher pH. Unlike sites 2 and 3, water pH in Tinker Creek was typically between 6.0–6.5 (Table 1). The greater actinomycete presence at site 1, despite the presence of the same type of macrophyte at site 2, indicates that pH or other factors, such as stream size, water chemistry, soil chemistry and watershed drainage, influence the actinomycete-flora.

Although 68% of all actinomycete isolates and 83% of the diversity was derived from S. americanum at site 1, all sites demonstrated distinct actinomycete associations with both plant type (M. umbrosum, R. pusillus and S. americanum) and tissue type (healthy, necrotic or dead). At site 3, M. umbrosum and R. pusillus grow interspersed. However, less than 2% of the actinomycete colonies were isolated from M. umbrosum and those 8 isolates only represented 2 Streptomyces strains (Table 3). Even though a broad array of selective measures was used to isolate actinomycetes from M. umbrosum, plates were typically dominated by the Gram-positive bacterium, Bacillus subtillus, and several filamentous fungi and oomycotous fungi. From the same locale, R. pusillus had greater actinomycete diversity. Thus, it appears that actinomycetes generally do not occur on M. umbrosum.

The actinomycete-flora on R. pusillus had a very distinct distribution associated with its plant tissue type. Unlike M. umbrosum, 108 actinomycete isolates were collected from R. pusillus. Commonly isolated from both healthy and necrotic plant tissues were a number of strains of Streptomyces and Nocardia. However, Pseudonocardia was only isolated from areas of R. pusillus classified as necrotic or dead. Furthermore, the relationship of Pseudonocardia with necrotic or dead plant tissues was true not only for R. pusillus, but also for S. americanum.

As previously described, most actinomycete isolates were collected from Sparganium americanum (Table 3). Although healthy and necrotic S. americanum tissues demonstrated no discernible trend in species composition or density, dead decomposing tissues had greater densities of actinomycetes. Common isolates from plant tissues classified as dead, particularly from site 1, included Nocardiodes, Pseudonocardia, Streptomyces, Nocardia, and Micromonospora.

From these findings, information regarding isolation methods for actinomycetes was also obtained. The filamentous nature and spore production by many actinomycetes permitted several selective measures for reduction of microbial flora without complete removal or death of actinomycetes. Fig. 1 displays the percent of total colonies isolated from each of the pretreatments. Sonication before plating yielded the greatest likelihood for successfully isolating actinomycetes. Although tip sonication and bath sonication only differed slightly in actual rates of success, bath sonication was easier to use and therefore preferred. Incubation at 60°C for 2 h, which reduced the viability of vegetative hyphae and bacterial cells, also resulted in relatively high likelihood for successful isolation of actinomycetes (Fig. 1). Although this method was particularly useful for isolation of Streptomycetes, it resulted in lower overall diversity than other pretreatments.


The percent of total actinomycete colonies isolated from each of the pretreatments.

In addition to selective pretreatments, several other factors determined actinomycete diversity and isolation success. For example, the media on which actinomycetes were first isolated appeared to be very important. Despite pretreatments, nutrient agar with and without antibiotics and tryptic soy agar without antibiotics were not selective enough for initial isolation of actinomycetes. On these media, other microorganisms rapidly overgrew actinomycetes. The few colonies of actinomycetes that did grow, did not produce spores, making subsequent transfers extremely difficult. However, in the presence of antibiotics, tryptic soy agar and starch casein agar resulted in the greatest numbers of actinomycete colonies (Fig. 2). Water agar with antibiotics was too selective and yielded little growth of any microorganism. In the absence of antibiotics, however, water agar and actinomycete isolation agar both were relatively useful (Fig. 2). Similarly, chitin agar and colloidal chitin agar were useful as well, although yeasts also grew well on these plates. It is important to note, in support of findings by Makkar and Cross [17] and Willoughby [19], that Actinoplanes was only isolated from starch casein agar and colloidal chitin agar in the absence of antibiotics.


The percent of total actinomycete colonies isolated on each of the media. Media abbreviations are as follows: A = actinomycete isolation agar, C = chitin agar, CC = colloidal chitin agar, M3 = M3 agar, N = nutrient agar, SC = starch casein agar, TS = tryptic soy agar and W = water agar.

In addition to media, temperature and length of incubation also affected actinomycete isolation success. All collections were made from streams ranging from 18–24°C (Table 1). In the laboratory, plates were maintained at room temperature or 27°C. Plates maintained at 27°C were rapidly overgrown by non-actinomycete bacteria. However, while plates maintained at ambient temperatures initially exhibited high levels of bacterial growth other than actinomycetes, the actinomycetes eventually cleared many of the microorganisms from their vicinity. The time required for actinomycete colonies to appear and substantiate themselves was important. Many actinomycete colonies were not discernible in <7 days and several other isolates including both strains of Microbiospora and Actinomadura required as many as 3.5 weeks to appear. Once the colonies became apparent, however, they could subsequently be transferred to new plates with lower doses of antibiotics, on which they exhibited more rapid growth.

These findings indicate that actinomycetes are associated with submersed freshwater macrophytes. By using a wide array of selective measures, a relatively diverse group of actinomycetes were collected. Furthermore, it is evident that the presence, distribution, and diversity of actinomycetes are dependent on the streams, plants, and tissue types from which they were isolated. Given the economic importance of this group of microorganisms both in industry and as a cause of discoloration and odors in drinking water, we believe further studies are needed to understand the ecology of the aquatic actinomycetes.


This research was supported by the US Department of Energy, Award number DE-FC09-96SR18546. We thank David R. Bowne, M. Steven Doggett, M.A. Moran and two anonymous reviewers for their helpful and insightful comments.


View Abstract