w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/watres
Review
Bacteriophages infecting Bacteroides as a marker for microbial source tracking Joan Jofre*, Anicet R. Blanch, Francisco Lucena, Maite Muniesa Department of Microbiology, University of Barcelona, Diagonal 643, Annex, Floor 0, 08028 Barcelona, Spain
article info
abstract
Article history:
Bacteriophages infecting certain strains of Bacteroides are amid the numerous procedures
Received 19 November 2013
proposed for tracking the source of faecal pollution. These bacteriophages fulfil reasonably
Received in revised form
well most of the requirements identified as appropriate for a suitable marker of faecal
30 January 2014
sources. Thus, different host strains are available that detect bacteriophages preferably in
Accepted 1 February 2014
water contaminated with faecal wastes corresponding to different animal species. For
Available online 11 February 2014
phages found preferably in human faecal wastes, which are the ones that have been more extensively studied, the amounts of phages found in waters contaminated with human
Keywords:
fecal samples is reasonably high; these amounts are invariable through the time; their
Microbial source tracking
resistance to natural and anthropogenic stressors is comparable to that of other relatively
Bacteroides
resistant indicator of faecal pollution such us coliphages; the abundance ratios of somatic
Bacteriophages
coliphages and bacteriophages infecting Bacteroides thetaiotaomicron GA17 are unvarying in
Somatic coliphages
recent and aged contamination; and standardised detection methods exist. These methods are easy, cost effective and provide data susceptible of numerical analysis. In contrast, there are some uncertainties regarding their geographical stability, and consequently suitable hosts need to be isolated for different geographical areas. However, a feasible method has been described to isolate suitable hosts in a given geographical area. In summary, phages infecting Bacteroides are a marker of faecal sources that in our opinion merits being included in the “toolbox” for microbial source tracking. However, further research is still needed in order to make clear some uncertainties regarding some of their characteristics and behaviour, to compare their suitability to the one of emerging methods such us targeting Bacteroidetes by qPCR assays; or settling molecular methods for their determination. ª 2014 Elsevier Ltd. All rights reserved.
Contents 1. 2. 3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Microbial source tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Bacteriophages infecting Bacteroides as a marker for source tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.1. General characteristics of bacteriophages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.2. General characteristics of bacteriophages infecting Bacteroides species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
* Corresponding author. Tel.: þ34 934021487; fax: þ34 934039047. E-mail address: [email protected] (J. Jofre). 0043-1354/$ e see front matter ª 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2014.02.006
2
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
4.
5. 6. 7.
1.
3.3. Do phages infecting Bacteroides fulfil the requirements for a suitable marker of faecal sources? . . . . . . . . . . . . . . . 4 3.4. Host-specificity of bacterial strains for detection of phages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.5. Detection and enumeration methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Levels in faecally impacted surface waters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.1. Temporal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.2. Geographical stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.3. Resistance to natural and anthropogenic stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.4. The ratio between markers: the need of a suitable “partner” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Source tracking studies using Bacteroides phages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Final remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Uncited reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 AcknowledgmentsAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Introduction
Water-borne diseases transmitted by the faecaleoral route make a significant contribution to the burden of diseases worldwide (WHO, 2008). Consequently, faecal pollution of surface and ground waters is a global public health concern. For example, in the USA the pathogen load is one of the main causes of fresh water body impairment (USEPA, 2004). Faecal pollutants can reach water bodies through direct discharge of faecal waste or raw wastewater, secondary effluents from wastewater treatment works, combined sewer and sanitary overflows, animal husbandry activities, the wastewaters from abattoirs and the meat industry, and wildlife (USEPA, 2004). The contribution of each of these faecal sources to the pathogen load varies strongly among watersheds. Determining the source of faecal contamination in aquatic environments is essential for estimating the health risks associated with faecal pollution, facilitating measures to remediate polluted waterways, and resolving legal responsibility for remediation. Microbial source tracking should enable investigators to uncover the sources of faecal pollution in a particular water body (Malakoff, 2002). Different approaches can be used to track the faecal sources. Eukaryotic mitochondrial DNA to differentiate sources in faecally contaminated surface water has been explored (Caldwell et al., 2007). The appropriateness of chemical markers has concentrated on the distribution pattern of chemicals such as caffeine, fragrance substances, fluorescent whitening agents, and faecal sterol isomers (Leeming et al., 1996; Standley et al., 2000; Hagedorn and Weisberg, 2011). As well, many different microorganisms have been tested for source tracking purposes. When using microorganisms, the techniques are then considered microbial source tracking (MST).
2.
Microbial source tracking
Candidate microbes, both pathogenic and commensal organisms, determined in many different ways, both by culture-dependent methods and molecular methods, or by library dependent and library independent methods, have been investigated and reviewed as potential tools for the identification of human faecal sources (Scott et al., 2002;
Simpson et al., 2002; Field and Samadpour, 2007; Blanch et al., 2008; Hagedorn et al., 2011). However, field studies using most of the numerous chemical and microbiological methods available to track sources of faecal contamination have shown that the existing methods are incomplete and that “tool boxes” of a variable number of different markers are needed, and consequently more research into suitable markers or combinations of markers is still worthwhile (Simpson et al., 2002; Stewart et al., 2003; Blanch et al., 2008). In our opinion the requirements for an ideal marker are summarized in Table 1. From the many studies performed, all markers investigated so far fail in one or more of the requirements, and this is the reason why a “toolbox” approach is recommended.
3. Bacteriophages infecting Bacteroides as a marker for source tracking Among the numerous tested procedures for microbial source tracking, there are three based on bacteriophages (Jofre et al., 2011) that are library independent methods. The first one is based on the frequencies distribution of the 4 serotypes/genotypes of F-specific RNA bacteriophages (Furuse, 1987; Hsu et al., 1995; Jofre et al., 2011), since serotypes/genotypes II and III predominate in humans and I and IV in non-humans studied so far. The second procedure is based on bacteriophages infecting selected strains of Bacteroides ssp. (Tartera and Jofre, 1987; Tartera et al., 1989). Bacteroides species are among the most abundant species in the feces of warm blooded animals (Allsop and Stickler, 1985); as well, since the early work of Bernhard and Field (2000) q-PCR assays targeting Bacteroidetes associated to certain animal species have become increasingly prominent for MST (Wuertz et al., 2011). And the third is based on bacteriophages infecting certain Enterococcus host strains (Bonilla et al., 2010; Purnell et al., 2011; SantiagoRodriguez et al., 2013). The following text is concerned with bacteriophages, particularly those infecting Bacteroides spp.
3.1.
General characteristics of bacteriophages
Bacteriophages, also called phages, are viruses that infect bacteria. They consist of a nucleic acid molecule (genome)
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
Table 1 e Requirements for an ideal marker of faecal sources. Requirement
Comments
Host-specificity
This is the most important attribute they should have Feasible, cost effective, fast and that permit numerical analysis
Appropriate detection and enumeration methods Easily measurable amounts in water environments Temporal stability
Geographic continuity Suitable resistance to natural and anthropogenic stressors
Have a “partner”
Their concentrations should be detectable by the respective method of measurement in any environmental matrix studied They should not have seasonal distribution and their prevalence should be constant through the years Useful anywhere or at least in extensive geographical areas Persistence in the environment and resistance to stressors neither very low, as is the case of E. coli, nor very high as is the case of spores of sulphite reducing clostridia. If possible, to have a persistence and resistance similar to the “partner” By “partner” we understand a cohort being a non-discriminating indicator of faecal pollution that is needed to associate the concentrations of the marker of faecal sources with the extent of faecal contamination and a marker of faecal sources
surrounded by a protein coat called capsid. Many phages also contain additional structures such as tails and spikes. Much less frequently they contain lipids. These particular characteristics mean that in terms of composition, structure, morphology and capsid size, phages share many properties with animal viruses. According to electron microscope observations, phages are more abundant than bacteria in most environments (Weinbauer, 2004). Whithman et al. (1998) estimated that in the biosphere there are around 5 1030 bacteria. Extrapolating this figure, the number of bacteriophages on Earth may be estimated at approximately 1031. Bacteriophages can only replicate inside susceptible and metabolizing host bacteria. A particular phage can only infect certain bacteria, a process mainly determined by the presence or absence of receptor molecules on the surface of the bacteria. Phage receptors had been described in different parts of the bacteria (capsule, cell wall, flagella and pili). Phages attaching to the receptors located in the cell wall are typically known as somatic phages. Thus, bacteriophages infecting Escherichia coli through receptors in the cell wall are named somatic coliphages (Hayes, 1968). Phages are often grouped according to their host range, based on morphology, nucleic acid, strategies of infection, morphogenesis, phylogeny, serology, sensitivity to physical and chemical agents, and dependence on properties of hosts and environment. The present classification, which has been
3
adopted by the International Committee on Taxonomy of Viruses, is mostly based on phage morphology and characteristics of the nucleic acid (Fauquet el al., 2005).
3.2. General characteristics of bacteriophages infecting Bacteroides species Bacteriophages infecting strains of Bacteroides fragilis, Bacteroides tethaiaotaomicron, Bacteroides ruminicola and Bacteroides ovatus had been detected in faeces and wastewater contaminated with faecal wastes (Booth et al., 1979; Tartera and Jofre, 1987; Klieve et al., 1991; Paya´n et al., 2005). Replication of bacteriophages infecting B. fragilis outside of the gut of warm blooded animals is unlikely; possibly due to requirements of the host strain regarding anaerobiosis and the need of certain nutrients that will unlikely coincide in natural environments, preventing as a consequence phage replication (Tartera and Jofre, 1987), which is a requirement for indicators. Bacteriophages infecting different Bacteroides species described so far have all been tailed and the vast majority have the morphology Siphoviridae (Booth et al., 1979; Diston et al., 2012; Klieve et al., 1991; Queralt et al., 2003; Paya´n, 2006; Ogilvie et al., 2012), this is with an icosahedron-shaped head and a flexible tail, showing an elevated degree of morphological homogeneity (Queralt et al., 2003; Diston et al., 2012; Ogilvie et al., 2012). The genome of the few Bacteroides infecting phages studied consists of double stranded DNA, corresponding to that of Siphoviridae (Puig and Girone´s, 1999; Hawkins et al., 2008; Ogilvie et al., 2012). The first genome sequence of a phage B40-8 (ATCC 51477-B1 eB40-8), infecting B. fragilis HSP40, isolated by Tartera and Jofre (1987), has been completed. It has 44,929 base pairs with a G þ C content of 38.7% and forty-six putative open reading frames (Hawkins et al., 2008). Later, bacteriophage FB124-14, which infects strain GB-124 of Bacteroides, has 47,159 bp with an average G þ C content of 38.7% and 68 putative open reading frames (Ogilvie et al., 2012). Phages infecting Bacteroides are supposed to infect the host through the cell wall and therefore technically somatic bacteriophages. The phage receptors identified are cell wall proteins (Puig et al., 2001). The amount of capsule of the host seems to play a role in phage infectivity presumably by impairing accessibility to the receptor sites on the bacterial surface (Booth et al., 1979; Klieve et al., 1991). Importantly, most phages infecting Bacteroides have a narrow host range (Cooper et al., 1984; Kory and Booth, 1986; Tartera and Jofre, 1987; Paya´n, 2006).
3.3. Do phages infecting Bacteroides fulfil the requirements for a suitable marker of faecal sources? The majority of the information currently available and reviewed in the following text refers to bacterial hosts that preferably detect bacteriophages in water contaminated with human faecal wastes. However, recent reports in the literature also refer to bacterial hosts that detect bacteriophages in water contaminated with animal faecal wastes (Go´mezDon˜ate et al., 2011; Wicki et al., 2011).
4
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
3.4. Host-specificity of bacterial strains for detection of phages In addition to the narrow host range mentioned previously, strains of Bacteroides spp. differ in their ability to recover phages in municipal sewage or from water bodies polluted with municipal sewage (Puig et al., 1999; Paya´n et al., 2005). Bacteroides strains differ also in their capability to detect bacteriophages in the faecal material of different animal species, including humans, and hence in their ability to determine the origin of faecal contamination in a given sample. Thus, some strains of B. fragilis, such as RYC2056 and VPI3625, detect phages both in human and non-human faecal wastes (Kator and Rhodes, 1992; Puig et al., 1999; Blanch et al., 2006), whereas other strains are more source-specific. The characteristics of available host strains of Bacteroides that detect phages mostly in human faecal wastes are summarized in table 2. They belong to different species, differ in the geographical area where they are applicable and differ in the numbers of phages that they detect in municipal wastewaters. However, they also detect phages in a very low percentage of samples that a priori never received human faecal wastes. But, as it will be pointed out, their concentrations in these samples not contaminated with human faecal wastes are very low and their ratios with respect to faecal indicators, such as faecal coliforms, or somatic coliphages or E. coli, are much lower than in human sources. Interesting enough to mention is that (meta) genomic analysis of 2 bacteriophages (B40-8 and FB124.14) infecting strains B. fragilis HSP40 and strain Bacteroides sp. GB124 respectively, specific of strains detecting preferably bacteriophages from human origin, indicate the human gutspecific nature of both phages (Ogilvie et al., 2012). In contrast, information regarding host bacteria detecting bacteriophages from animal faecal sources is scarce. Recently, Go´mez-Don˜ate et al. (2011) have reported the isolation of several Bacteroides strains from pig, cattle and poultry faeces, able to detect specifically phages in faecal sources that coincide with their own origin. The isolated strains belonged to B.
fragilis and Bacteroides thetaiotaomicron. These host strains detect phages in samples contaminated with faecal material originating from each source (70e100% of the samples), and show no detection or very low percentages of detection of phages in samples from other animal origins (from 0 to 20% of the samples). Most frequent concentrations of bacteriophages detected with the host specific host strain in wastewater of abattoirs slaughtering specifically pigs, cattle or poultry ranged from 5 104 to 5 105 PFU, from 5 102 to 5 103 and from 5 103 to 5 104 respectively Go´mez-Don˜ate et al. (2011). Only host strains isolated from pig wastewater detected phages in 50% of human sewage samples, though at concentrations several orders of magnitude lower than the concentrations of phages detected by human-specific host strains. In addition, Wicki et al. (2011) have reported the isolation of strain host associated for bovine (1 strain) and horses (2 strains) that did not detect phages in municipal wastewater, and detect about 102 PFUs per 100 ml in a 67% of abattoir wastewater samples. The reasons for this narrow host range of Bacteroides phages and for this animal host association are not fully understood. This animal host association is also observed for Bacteroidales species according to studies performed with gene markers (Wuertz et al., 2011). A potential explanation is that obligate anaerobes and phages capable of infecting them might have co-evolved more separately within animal hosts than the facultative bacteria present in the intestinal microbiota. In fact, a considerable number of interactions between Bacteroides and the animal host have been described for human and Bacteroides species (Xu et al., 2003; Ley et al., 2008).
3.5.
Detection and enumeration methods
Standardized plaque assay and enrichment methods for bacteriophages infecting B. fragilis are available (Araujo et al., 2001; ISO, 2001). These methods are similar to those described for enumerating bacteriophages of most Gram negative bacteria, such as for example somatic coliphages, the difference being
Table 2 e Amounts of bacteriophages detected by discriminating Bacteroides host strains in municipal wastewaters of different geographical areas. Bacteroides host strain B. fragilis HSP40
B. tethaioataomicron GA17 Bacteroides sp. GB124 B. fragilis HB13 Bacteroides sp. ARABA 84 a b
Geographical area
Concentrations in municipal wastewaters (PFU/100 ml)
Spain, France, Israel, South Africa
5 103e104
USA, Great Britain, Sweden, Tailand Spain, France, Cyprus, Colombia, Tunisia United Kingdom, Sweden Great Britain Spain Spain Colombia Switzerland
2 105, and a median >2 105. The percentage of values below >104 is only 4%. Therefore, the probability of having a value >5 102 in samples contaminated only with human faecal wastes and a value
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/watres
Review
Bacteriophages infecting Bacteroides as a marker for microbial source tracking Joan Jofre*, Anicet R. Blanch, Francisco Lucena, Maite Muniesa Department of Microbiology, University of Barcelona, Diagonal 643, Annex, Floor 0, 08028 Barcelona, Spain
article info
abstract
Article history:
Bacteriophages infecting certain strains of Bacteroides are amid the numerous procedures
Received 19 November 2013
proposed for tracking the source of faecal pollution. These bacteriophages fulfil reasonably
Received in revised form
well most of the requirements identified as appropriate for a suitable marker of faecal
30 January 2014
sources. Thus, different host strains are available that detect bacteriophages preferably in
Accepted 1 February 2014
water contaminated with faecal wastes corresponding to different animal species. For
Available online 11 February 2014
phages found preferably in human faecal wastes, which are the ones that have been more extensively studied, the amounts of phages found in waters contaminated with human
Keywords:
fecal samples is reasonably high; these amounts are invariable through the time; their
Microbial source tracking
resistance to natural and anthropogenic stressors is comparable to that of other relatively
Bacteroides
resistant indicator of faecal pollution such us coliphages; the abundance ratios of somatic
Bacteriophages
coliphages and bacteriophages infecting Bacteroides thetaiotaomicron GA17 are unvarying in
Somatic coliphages
recent and aged contamination; and standardised detection methods exist. These methods are easy, cost effective and provide data susceptible of numerical analysis. In contrast, there are some uncertainties regarding their geographical stability, and consequently suitable hosts need to be isolated for different geographical areas. However, a feasible method has been described to isolate suitable hosts in a given geographical area. In summary, phages infecting Bacteroides are a marker of faecal sources that in our opinion merits being included in the “toolbox” for microbial source tracking. However, further research is still needed in order to make clear some uncertainties regarding some of their characteristics and behaviour, to compare their suitability to the one of emerging methods such us targeting Bacteroidetes by qPCR assays; or settling molecular methods for their determination. ª 2014 Elsevier Ltd. All rights reserved.
Contents 1. 2. 3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Microbial source tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Bacteriophages infecting Bacteroides as a marker for source tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.1. General characteristics of bacteriophages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.2. General characteristics of bacteriophages infecting Bacteroides species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
* Corresponding author. Tel.: þ34 934021487; fax: þ34 934039047. E-mail address: [email protected] (J. Jofre). 0043-1354/$ e see front matter ª 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2014.02.006
2
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
4.
5. 6. 7.
1.
3.3. Do phages infecting Bacteroides fulfil the requirements for a suitable marker of faecal sources? . . . . . . . . . . . . . . . 4 3.4. Host-specificity of bacterial strains for detection of phages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.5. Detection and enumeration methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Levels in faecally impacted surface waters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.1. Temporal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4.2. Geographical stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.3. Resistance to natural and anthropogenic stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.4. The ratio between markers: the need of a suitable “partner” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Source tracking studies using Bacteroides phages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Final remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Uncited reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 AcknowledgmentsAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Introduction
Water-borne diseases transmitted by the faecaleoral route make a significant contribution to the burden of diseases worldwide (WHO, 2008). Consequently, faecal pollution of surface and ground waters is a global public health concern. For example, in the USA the pathogen load is one of the main causes of fresh water body impairment (USEPA, 2004). Faecal pollutants can reach water bodies through direct discharge of faecal waste or raw wastewater, secondary effluents from wastewater treatment works, combined sewer and sanitary overflows, animal husbandry activities, the wastewaters from abattoirs and the meat industry, and wildlife (USEPA, 2004). The contribution of each of these faecal sources to the pathogen load varies strongly among watersheds. Determining the source of faecal contamination in aquatic environments is essential for estimating the health risks associated with faecal pollution, facilitating measures to remediate polluted waterways, and resolving legal responsibility for remediation. Microbial source tracking should enable investigators to uncover the sources of faecal pollution in a particular water body (Malakoff, 2002). Different approaches can be used to track the faecal sources. Eukaryotic mitochondrial DNA to differentiate sources in faecally contaminated surface water has been explored (Caldwell et al., 2007). The appropriateness of chemical markers has concentrated on the distribution pattern of chemicals such as caffeine, fragrance substances, fluorescent whitening agents, and faecal sterol isomers (Leeming et al., 1996; Standley et al., 2000; Hagedorn and Weisberg, 2011). As well, many different microorganisms have been tested for source tracking purposes. When using microorganisms, the techniques are then considered microbial source tracking (MST).
2.
Microbial source tracking
Candidate microbes, both pathogenic and commensal organisms, determined in many different ways, both by culture-dependent methods and molecular methods, or by library dependent and library independent methods, have been investigated and reviewed as potential tools for the identification of human faecal sources (Scott et al., 2002;
Simpson et al., 2002; Field and Samadpour, 2007; Blanch et al., 2008; Hagedorn et al., 2011). However, field studies using most of the numerous chemical and microbiological methods available to track sources of faecal contamination have shown that the existing methods are incomplete and that “tool boxes” of a variable number of different markers are needed, and consequently more research into suitable markers or combinations of markers is still worthwhile (Simpson et al., 2002; Stewart et al., 2003; Blanch et al., 2008). In our opinion the requirements for an ideal marker are summarized in Table 1. From the many studies performed, all markers investigated so far fail in one or more of the requirements, and this is the reason why a “toolbox” approach is recommended.
3. Bacteriophages infecting Bacteroides as a marker for source tracking Among the numerous tested procedures for microbial source tracking, there are three based on bacteriophages (Jofre et al., 2011) that are library independent methods. The first one is based on the frequencies distribution of the 4 serotypes/genotypes of F-specific RNA bacteriophages (Furuse, 1987; Hsu et al., 1995; Jofre et al., 2011), since serotypes/genotypes II and III predominate in humans and I and IV in non-humans studied so far. The second procedure is based on bacteriophages infecting selected strains of Bacteroides ssp. (Tartera and Jofre, 1987; Tartera et al., 1989). Bacteroides species are among the most abundant species in the feces of warm blooded animals (Allsop and Stickler, 1985); as well, since the early work of Bernhard and Field (2000) q-PCR assays targeting Bacteroidetes associated to certain animal species have become increasingly prominent for MST (Wuertz et al., 2011). And the third is based on bacteriophages infecting certain Enterococcus host strains (Bonilla et al., 2010; Purnell et al., 2011; SantiagoRodriguez et al., 2013). The following text is concerned with bacteriophages, particularly those infecting Bacteroides spp.
3.1.
General characteristics of bacteriophages
Bacteriophages, also called phages, are viruses that infect bacteria. They consist of a nucleic acid molecule (genome)
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
Table 1 e Requirements for an ideal marker of faecal sources. Requirement
Comments
Host-specificity
This is the most important attribute they should have Feasible, cost effective, fast and that permit numerical analysis
Appropriate detection and enumeration methods Easily measurable amounts in water environments Temporal stability
Geographic continuity Suitable resistance to natural and anthropogenic stressors
Have a “partner”
Their concentrations should be detectable by the respective method of measurement in any environmental matrix studied They should not have seasonal distribution and their prevalence should be constant through the years Useful anywhere or at least in extensive geographical areas Persistence in the environment and resistance to stressors neither very low, as is the case of E. coli, nor very high as is the case of spores of sulphite reducing clostridia. If possible, to have a persistence and resistance similar to the “partner” By “partner” we understand a cohort being a non-discriminating indicator of faecal pollution that is needed to associate the concentrations of the marker of faecal sources with the extent of faecal contamination and a marker of faecal sources
surrounded by a protein coat called capsid. Many phages also contain additional structures such as tails and spikes. Much less frequently they contain lipids. These particular characteristics mean that in terms of composition, structure, morphology and capsid size, phages share many properties with animal viruses. According to electron microscope observations, phages are more abundant than bacteria in most environments (Weinbauer, 2004). Whithman et al. (1998) estimated that in the biosphere there are around 5 1030 bacteria. Extrapolating this figure, the number of bacteriophages on Earth may be estimated at approximately 1031. Bacteriophages can only replicate inside susceptible and metabolizing host bacteria. A particular phage can only infect certain bacteria, a process mainly determined by the presence or absence of receptor molecules on the surface of the bacteria. Phage receptors had been described in different parts of the bacteria (capsule, cell wall, flagella and pili). Phages attaching to the receptors located in the cell wall are typically known as somatic phages. Thus, bacteriophages infecting Escherichia coli through receptors in the cell wall are named somatic coliphages (Hayes, 1968). Phages are often grouped according to their host range, based on morphology, nucleic acid, strategies of infection, morphogenesis, phylogeny, serology, sensitivity to physical and chemical agents, and dependence on properties of hosts and environment. The present classification, which has been
3
adopted by the International Committee on Taxonomy of Viruses, is mostly based on phage morphology and characteristics of the nucleic acid (Fauquet el al., 2005).
3.2. General characteristics of bacteriophages infecting Bacteroides species Bacteriophages infecting strains of Bacteroides fragilis, Bacteroides tethaiaotaomicron, Bacteroides ruminicola and Bacteroides ovatus had been detected in faeces and wastewater contaminated with faecal wastes (Booth et al., 1979; Tartera and Jofre, 1987; Klieve et al., 1991; Paya´n et al., 2005). Replication of bacteriophages infecting B. fragilis outside of the gut of warm blooded animals is unlikely; possibly due to requirements of the host strain regarding anaerobiosis and the need of certain nutrients that will unlikely coincide in natural environments, preventing as a consequence phage replication (Tartera and Jofre, 1987), which is a requirement for indicators. Bacteriophages infecting different Bacteroides species described so far have all been tailed and the vast majority have the morphology Siphoviridae (Booth et al., 1979; Diston et al., 2012; Klieve et al., 1991; Queralt et al., 2003; Paya´n, 2006; Ogilvie et al., 2012), this is with an icosahedron-shaped head and a flexible tail, showing an elevated degree of morphological homogeneity (Queralt et al., 2003; Diston et al., 2012; Ogilvie et al., 2012). The genome of the few Bacteroides infecting phages studied consists of double stranded DNA, corresponding to that of Siphoviridae (Puig and Girone´s, 1999; Hawkins et al., 2008; Ogilvie et al., 2012). The first genome sequence of a phage B40-8 (ATCC 51477-B1 eB40-8), infecting B. fragilis HSP40, isolated by Tartera and Jofre (1987), has been completed. It has 44,929 base pairs with a G þ C content of 38.7% and forty-six putative open reading frames (Hawkins et al., 2008). Later, bacteriophage FB124-14, which infects strain GB-124 of Bacteroides, has 47,159 bp with an average G þ C content of 38.7% and 68 putative open reading frames (Ogilvie et al., 2012). Phages infecting Bacteroides are supposed to infect the host through the cell wall and therefore technically somatic bacteriophages. The phage receptors identified are cell wall proteins (Puig et al., 2001). The amount of capsule of the host seems to play a role in phage infectivity presumably by impairing accessibility to the receptor sites on the bacterial surface (Booth et al., 1979; Klieve et al., 1991). Importantly, most phages infecting Bacteroides have a narrow host range (Cooper et al., 1984; Kory and Booth, 1986; Tartera and Jofre, 1987; Paya´n, 2006).
3.3. Do phages infecting Bacteroides fulfil the requirements for a suitable marker of faecal sources? The majority of the information currently available and reviewed in the following text refers to bacterial hosts that preferably detect bacteriophages in water contaminated with human faecal wastes. However, recent reports in the literature also refer to bacterial hosts that detect bacteriophages in water contaminated with animal faecal wastes (Go´mezDon˜ate et al., 2011; Wicki et al., 2011).
4
w a t e r r e s e a r c h 5 5 ( 2 0 1 4 ) 1 e1 1
3.4. Host-specificity of bacterial strains for detection of phages In addition to the narrow host range mentioned previously, strains of Bacteroides spp. differ in their ability to recover phages in municipal sewage or from water bodies polluted with municipal sewage (Puig et al., 1999; Paya´n et al., 2005). Bacteroides strains differ also in their capability to detect bacteriophages in the faecal material of different animal species, including humans, and hence in their ability to determine the origin of faecal contamination in a given sample. Thus, some strains of B. fragilis, such as RYC2056 and VPI3625, detect phages both in human and non-human faecal wastes (Kator and Rhodes, 1992; Puig et al., 1999; Blanch et al., 2006), whereas other strains are more source-specific. The characteristics of available host strains of Bacteroides that detect phages mostly in human faecal wastes are summarized in table 2. They belong to different species, differ in the geographical area where they are applicable and differ in the numbers of phages that they detect in municipal wastewaters. However, they also detect phages in a very low percentage of samples that a priori never received human faecal wastes. But, as it will be pointed out, their concentrations in these samples not contaminated with human faecal wastes are very low and their ratios with respect to faecal indicators, such as faecal coliforms, or somatic coliphages or E. coli, are much lower than in human sources. Interesting enough to mention is that (meta) genomic analysis of 2 bacteriophages (B40-8 and FB124.14) infecting strains B. fragilis HSP40 and strain Bacteroides sp. GB124 respectively, specific of strains detecting preferably bacteriophages from human origin, indicate the human gutspecific nature of both phages (Ogilvie et al., 2012). In contrast, information regarding host bacteria detecting bacteriophages from animal faecal sources is scarce. Recently, Go´mez-Don˜ate et al. (2011) have reported the isolation of several Bacteroides strains from pig, cattle and poultry faeces, able to detect specifically phages in faecal sources that coincide with their own origin. The isolated strains belonged to B.
fragilis and Bacteroides thetaiotaomicron. These host strains detect phages in samples contaminated with faecal material originating from each source (70e100% of the samples), and show no detection or very low percentages of detection of phages in samples from other animal origins (from 0 to 20% of the samples). Most frequent concentrations of bacteriophages detected with the host specific host strain in wastewater of abattoirs slaughtering specifically pigs, cattle or poultry ranged from 5 104 to 5 105 PFU, from 5 102 to 5 103 and from 5 103 to 5 104 respectively Go´mez-Don˜ate et al. (2011). Only host strains isolated from pig wastewater detected phages in 50% of human sewage samples, though at concentrations several orders of magnitude lower than the concentrations of phages detected by human-specific host strains. In addition, Wicki et al. (2011) have reported the isolation of strain host associated for bovine (1 strain) and horses (2 strains) that did not detect phages in municipal wastewater, and detect about 102 PFUs per 100 ml in a 67% of abattoir wastewater samples. The reasons for this narrow host range of Bacteroides phages and for this animal host association are not fully understood. This animal host association is also observed for Bacteroidales species according to studies performed with gene markers (Wuertz et al., 2011). A potential explanation is that obligate anaerobes and phages capable of infecting them might have co-evolved more separately within animal hosts than the facultative bacteria present in the intestinal microbiota. In fact, a considerable number of interactions between Bacteroides and the animal host have been described for human and Bacteroides species (Xu et al., 2003; Ley et al., 2008).
3.5.
Detection and enumeration methods
Standardized plaque assay and enrichment methods for bacteriophages infecting B. fragilis are available (Araujo et al., 2001; ISO, 2001). These methods are similar to those described for enumerating bacteriophages of most Gram negative bacteria, such as for example somatic coliphages, the difference being
Table 2 e Amounts of bacteriophages detected by discriminating Bacteroides host strains in municipal wastewaters of different geographical areas. Bacteroides host strain B. fragilis HSP40
B. tethaioataomicron GA17 Bacteroides sp. GB124 B. fragilis HB13 Bacteroides sp. ARABA 84 a b
Geographical area
Concentrations in municipal wastewaters (PFU/100 ml)
Spain, France, Israel, South Africa
5 103e104
USA, Great Britain, Sweden, Tailand Spain, France, Cyprus, Colombia, Tunisia United Kingdom, Sweden Great Britain Spain Spain Colombia Switzerland
2 105, and a median >2 105. The percentage of values below >104 is only 4%. Therefore, the probability of having a value >5 102 in samples contaminated only with human faecal wastes and a value
