Home Protozoa
Giardia PDF Print E-mail
Monday, 08 February 2010 00:00
Lucy Robertson
Andreas Nocker

  • Flagellated protozoan parasite in the phylum Sarcomastigophora, order Diplomonadida. Has a worldwide distribution.
  • 6 different species of Giardia currently recognised, of which one, Giardia duodenalis (syn. G. lamblia, syn. G. intestinalis), is infective to humans and a range of mammalian hosts. Only this species is considered in these pages.
  • Different assemblages, or genotypes, within the species Giardia duodenalis, although apparently morphologically identical, have different host specificity, with assemblages A and B infectious to humans and various other mammals.
  • Many mammals are infected with assemblages of Giardia that are not infectious to humans.
  • Infection with Giardia may be asymptomatic, or may result in giardiasis, usually manifest as a diarrhoeic disease, possibly with other abdominal symptoms and steatorrhoea. 
  • Although treatment is usually simple and effective, some infections are apparently refractory to treatment. In some cases, symptoms may also continue after successful elimination of the parasite.
  • Simple two-stage lifecycle; trophozoites are motile, active feeding stage in the small intestine of the mammalian host. Cysts are the transmission stage, excreted in the faeces of infected hosts.
  • Infectious dose is low, and due to the multiple cycles of asexual binary division, the excretion rate is high.
  • Cysts are ovoid, 11-15 µm in length and 7-10 µm width. They are immediately infectious upon excretion, but also very robust and can survive for prolonged periods in the environment, especially at low temperatures, thus lending themselves to the possibility of waterborne transmission.
  • Cysts in water may be derived from human infections (e.g. from sewage) or from animal infections. Due to animals usually being infected by host-specific assemblages of Giardia that are not infective to humans, cysts in water derived from animal infections may not be of public health significance. 
  • Surveys have shown that Giardia cysts may be detected in raw drinking water worldwide.
  • Analysis methods are usually based on concentration of relatively large volumes of water (10 L and above), followed by separation (often immunomagnetic separation), and identification usually by using monoclonal antibodies with a fluorescent label.
  • Molecular methods are not often used for detection, but more frequently for assessing the genotype of the cysts detected.
  • Giardia cysts are considered to be less resistant to water treatment regimes than Cryptosporidium oocysts, but are nevertheless resistant to chlorination at the levels usually employed in water treatment, particularly at low temperatures and high pH.
  • They are less resistant against ozonisation, and many studies suggest that UV disinfection is the water treatment method of choice for Giardia cyst inactivation. However, UV disinfection has provided less promising results for inactivation of Giardia cysts during wastewater treatment.
  • Photocatalytic disinfection is a treatment that appears to be particularly promising.
Since first being discovered in 1681, Giardia have been identified from a wide range of vertebrate hosts, including mammalian, avian, reptilian and amphibian species ( Kutz et al. 2009). At one point, over 50 different species of Giardia were described, but currently 6 different species are recognised; Giardia agilis from amphibians, Giardia ardeae and Giardia psittaci from birds, Giardia muris and Giardia microti from rodents, and Giardia duodenalis (syn. G. lamblia, syn. G. intestinalis) from a range of mammalian hosts, including livestock, companion animals, rodents, artiodactyls and humans.

Molecular studies have divided G. duodenalis into various assemblages or genotypes, which, whilst being apparently morpholocially identical, demonstrate host specificity patterns as well as various other pheyontypic differences (Monis et al. 2009). Thus G. duodenalis is currently described as a species complex, consisting of 7 genotypes (A-G), of which only A and B are infectious to humans (as well as other mammals), and it has been proposed that the current taxonomy should be reorganised, such that the different genotypes are recognised as separate species. However, this suggestion has currently not been accepted and is under debate ( Robertson et al. 2010; https://community.eupathdb.org/). The existence of different genotypes with different host specificity patterns is important for considering zoonotic potential; although animals may sometimes be reservoirs of Giardia infectious to humans, cysts derived from a considerable proportion of animal infections are probably of no public health significance as they are of a genotype that is not infective to humans.

Lifecycle. The lifecycle of G. duodenalis is generally considered as very simple, being composed of only two stages, the cyst and the trophozoite. Following ingestion of the cyst, excystation occurs in the duodenum, releasing 2 trophozoites. These are a motile, active, feeding stage, but non-invasive, and replicate in the duodenum and upper jejunum by binary fission. Although Giardia replication is generally considered to be asexual, there is some evidence of sexual reproduction also, although the mechanism, frequency, and significance remain currently unknown ( Birky CW 2010). As the trophozoites transit towards the large intestine, they undergo encystation due to physiological changes and triggers, resulting in the production of environmentally resistant cysts, which are excreted in the faeces. During diarrhoeic episodes, trophozoites might also be excreted in the faeces, but they are not environmentally resistant and do not transmit the infection further. The cysts are immediately infectious, but are also resistant to environmental pressures and therefore may survive in the environment for considerable periods.

Infection. Giardia infections are usually associated with diarrhoea, with symptoms commencing between 1 and 14 days after infection (usually about 1 week), but can be asymptomatic or responsible for a broad clinical spectrum, ranging from acute to chronic ( Robertson et al. 2010). Diarrhoea is often foul-smelling and frothy, due to steatorrhoea, and may be accompanied by nausea, vomiting, and weight loss, and children might suffer more serious consequences. Some individuals suffer long-term sequelae even after resolution of infection. Giardiasis may be self-limiting, but treatment is usually recommended. At least six different drug classes are available, but 5-nitroimidazole compounds are usually the agents of choice. Although usually effective, treatment failure is not unknown. The reason for the spectrum of symptom patterns has been debated, and may be due to either host factors or parasite factors or both ( Robertson et al. 2010).

Although most cases of giardiasis are probably sporadic, due to the longevity of Giardia cysts in the environment, there is the potential for communitywide outbreaks due to the contamination of water or food. Foodborne outbreaks are relatively rare and usually involve relatively few individuals ( Smith et al. 2007), whilst waterborne outbreaks often involve hundreds, sometimes thousands, of individuals, and over 130 waterborne outbreaks were documented up until 2003 ( Karanis et al. 2007). The largest outbreak of waterborne giardiasis in the last 10 years, occurred in Bergen, Norway during the fall/winter of 2004/2005, although the contamination event probably occurred in late summer (August) 2004 ( Robertson et al. 2006). In this outbreak, in which there were 1262 laboratory-confirmed cases, though the actual number of individuals with symptomatic infection has been estimated to have been in excess of 5000 (http://bgrg.uib.no/), it is thought that heavy rain contributed to washing Giardia cyst-contaminated sewage from a broken wastepipe into the water intake of a reservoir. This outbreak was the first in which genotyping was used to disprove a suspected source of contamination ( Robertson et al. 2006), and, whilst having a serious negative impact on the lives of many of Bergen’s residents, also provided the opportunity for a considerable number of environmental and clinical investigations (http://bgrg.uib.no/).

There have been many reports on the occurrence and concentrations of Giardia cysts in different water types (raw water, drinking water, recreational water) from different countries around the world. The results obtained are obviously affected by the efficiency of the method used in detecting Giardia cysts in that particular water type; different methods and different efficiencies obviously affect the usefulness of comparing results from different places or different sources. From 1994 onwards, the majority of studies have used methods based, either loosely or directly, on US EPA Method 1623 (http://www.epa.gov/nerlcwww/1623de05.pdf), in which filtration of water is followed by elution from the filter, separation of the cysts by immunomagnetic separation (IMS), with detection usually by immunofluorescent microscopy (IFAT), in which the cysts are labelled using monoclonal antibodies bound to a fluorochrome such as FITC. Variations in the method include, for example, flocculation for concentrating cysts from the initial large volumes of water.

It should be noted that these usually applied methods for detecting Giardia cysts in environmental samples do not provide information on the infectivity, pathogenicity, or virulence of the cysts detected. Also, although, genotyping studies may sometimes be used retrospectively, it is often not possible to determine the source from which the Giardia cysts are derived. This lack of information has been considered to present a major challenge for the future monitoring of waterborne Giardia ( Karanis et al. 2007).

Examples of studies reporting prevalence and concentrations of Giardia cysts in water samples are given below. Details on occurrence, concentrations, and (where investigated) genotype are given from studies from different countries, published from 2008 onwards. A brief summary is provided afterwards of some earlier studies.

  • Nine 50 L water samples from a recreational lake in Malaya ( Lim et al. 2009) were analysed for both Cryptosporidium oocysts and Giardia cysts, using capsule filtration, followed by IMS, with detection by IFAT. Giardia cysts were detected in 7 samples (77.8 %), at concentrations of between 0.17 to 1.1 cysts per litre. Genotyping by PCR and sequencing at the SSU rRNA gene demonstrated that at least some of the cysts were of assemblage A, and thus of zoonotic potential.
  • One hundred and sixteen 50 L water samples were collected from 29 different sampling points on the River Tambre, NW Spain ( Castro-Hermida et al. 2009) and analysed for both Cryptosporidium oocysts and Giardia cysts using cartridge filtration, followed by IMS, with detection by IFAT. Giardia cysts were detected in 78 samples (67.2 %), with positive samples from all sampling points, at concentrations of between 2 to 722 cysts per litre. Genotyping by PCR and sequencing at the β giardin gene demonstrated the presence of cysts of assemblage A, and thus of zoonotic potential, and also of assemblage E (not infectious to humans).
  • One hundred and seventy five water samples (water sample volume not reported) consisting of 69 untreated water samples and 106 treated water samples from Portugal ( Lobo et al. 2009) were analysed for both Cryptosporidium oocysts and Giardia cysts following US EPA Method 1623, with IMS separation,  and detection by IFAT. Giardia cysts were detected in 40 of the raw water samples (58.0 %) and in 27 of the treated water samples (25.6 %). Cyst concentrations were not reported. Genotyping by PCR and sequencing at the ß giardin gene demonstrated the presence of cysts of assemblage A (and thus of zoonotic potential), in one raw water sample.
  • Forty 20 L water samples were collected from 2 recreational streams in Switzerland ( Wicki et al. 2009) and analysed for Giardia cysts using concentration by aluminium sulphate flocculation, followed by IMS separation, and detection by IFAT. Giardia cysts were detected in 39 of the samples (97.5 %), at concentrations of between 1 to 261 cysts per 20 litres.  Genotyping was not attempted.
  • Thirty 1,000 L water samples consisting of 15 untreated water samples and 15 treated water samples from Paraná, Brazil ( Nishi et al. 2009) were analysed for both Cryptosporidium oocysts and Giardia cysts by sequential membrane filtration steps and and detection by IFAT of the concentrated filter eluate. Giardia cysts were detected in 3 of the raw water samples (26.6 %) and none of the treated water samples (0 %). Cyst concentrations were 0.2 cysts per litre for all positive samples. Genotyping was not attempted.
  • One hundred and sixty two 20 L water samples from river water drinking sources around Paris and surrounding areas in France ( Mons et al. 2009) were analysed for both Cryptosporidium oocysts and Giardia cysts, using capsule filtration, followed by IMS, with detection by IFAT. Giardia cysts were detected in 152 samples (93.8 %), at concentrations of between 0.05 to 51.1 cysts per litre. Genotyping was not attempted.
  • Twenty six water samples (of between 5 and 20 L) consisting of 16 untreated surface water samples at the intake to water treatment plants, 7 stream/duct water samples, and 3 recreational water samples from Hungary ( Plutzer et al. 2008) were analysed for both Cryptosporidium oocysts and Giardia cysts using concentration by sulphate flocculation or filtration, followed by IMS separation, and detection by IFAT. Giardia cysts were detected in 10 of the raw water samples (62.5 %), 4 of the stream/duct water samples (57.1 %), and in 1 of the recreational water samples (33.3 %), at concentrations of between 0.1 to 17.4 cysts per litre. Genotyping by PCR and sequencing at the SSU rRNA and GDH genes demonstrated the presence of cysts of assemblages A and B in raw water samples and stream/duct water (and thus of zoonotic potential).
  • Twenty one water samples (of between 60 and 200 L) consisting of 7 raw water samples, 7 clarified water samples, , and 7 treated water samples from Foggia, Italy ( Vernile et al. 2009) were analysed for both Cryptosporidium oocysts and Giardia cysts using filtration and elution, sucrose flotation, and detection by IFAT. Giardia cysts were detected in 4 of the raw water samples (57.1 %), at concentrations of between 0.01 to 0.08 cysts per litre, and in none of the other samples. Genotyping was not attempted.
  • Eighty seven 20 L water samples (of between 60 and 200 L) consisting of 32 surface water samples and 55 recreational water samples  from Amsterdam, Netherlands ( Schets et al. 2008) were analysed for both Cryptosporidium oocysts and Giardia cysts using filtration and elution, IMS, and detection by IFAT. Giardia cysts were detected in 31 of the surface water samples (96.9 %), at concentrations of between 0.1 to 16.7 cysts per litre, and in 20 of the recreational water samples (36.4 %), at concentrations of between 0.1 to 1.1 cysts per litre. Genotyping was not attempted.
  • Many earlier studies were performed in Canada and USA on raw and finished water samples. In these studies genotyping was not attempted and, in many of the studies, filtration of large volumes were used (100 L or more) using yarn-wound filters was used. Most studies used IFAT for detection purposes, although DAPI staining is rarely included (if ever). Additionally, all studies performed prior to 1996 did not use IMS (had not been developed), and thus the recovery efficiencies of the methods may be relatively low. Nevertheless studies generally report that Giardia cysts occur relatively often in surface waters, particularly those under the influence of sewage effluents or agricultural run off, although cyst concentrations are generally low. Some of these earlier reports investigate seasonality of occurrence ( LeChevallier et al. 1991; Akin and Jakubowski 1986; Ongerth JE 1989; Rose et al. 1988; Rose et al. 1986; Isaac-Renton et al. 1996; Rose et al. 1991; Wallis et al. 1996; Ong et al. 1996; Ongerth et al. 1995).
There are several reasons why waterborne transmission of Giardia is important, including the low infectious dose, the high excretion rate, and, perhaps even more importantly, because the transmission stage of Giardia, the cyst, is very resistant to many of the disinfectant regimes commonly used in the water industry. The resistance of Giardia cysts to environmental pressures and disinfection regimes is considered to be due to the filamentous cyst wall, which contains carbohydrate and protein in the ratio 3:2, with the carbohydrate component of the filaments a unique (ß1-3)-linked N-acetylgalactosamine (GalNAc) homopolymer ( Gerwig et al. 2002) for which degradative enzymes have not been found, although presumably must exist as there is no apparent irreversible accumulation, and may, indeed, by involved in excystation, although this could simply involve proteolysis of the protein in the cyst wall.  Conformational studies have demonstrated that the highly insoluble nature of the cyst wall is not due the conformational properties of a single GalNAc polysaccharide chain, but is a result of strong inter-chain interactions in which the potential covalent linkages between GalNAc polymers and the wall protein might contribute.

Outbreaks of waterborne giardiasis have resulted in considerable interest in ascertaining which water treatments are effective at inactivating Giardia cysts in water. However, as Cryptosporidium oocysts are generally considered to be even more robust than Giardia cysts ( Betancourt and Rose 2004), and also lend themselves to transmission by the waterborne route, more research has been directed to Cryptosporidium inactivation. Additionally, as Giardia cysts are larger than Cryptosporidium oocysts, they are more readily removed by physical processes such as filtration.

The multiple-barrier approach is generally accepted as the guiding principle for providing drinking safe water, and processes such as catchment protection, filtration, flocculation, and sedimentation, may all be important contributors to the reduction of viable, infectious Giardia cysts in drinking water supplies, knowledge on the impacts of individual disinfection procedures on Giardia cyst inactivation is also important in provision of water supplies. The most important disinfectants in the water industry include: chlorine, chloramine, chlorine dioxide, ozone, ultraviolet (UV) irradiation, and photocatalytic disinfection. Some of the studies of the effects of some of these disinfectants/disinfectant regimes on Giardia cysts in water are summarised below:

Chlorination: Data published during the 1980s ( Jarroll et al. 1981; Rice et al. 1982; Hoff et al. 1985; Leahy et al. 1987, Hibler et al. 1987) demonstrated that although Giardia cysts could be inactivated by chlorine if temperatures were sufficiently high, they were resistant to chlorination at low temperatures and high pH. These studies also demonstrated that C·t’ products required for 99 % inactivation of Giardia cysts were many orders of magnitude higher than those necessary for 99 % inactivation of bacterial and viral pathogens or indicators. Thus, it was concluded that not only is judging the microbiological safety of water solely on the basis of absence of coliforms not a good practice, but also that although chlorine is able to inactivate Giardia cysts, this only  occurs at concentrations well above these employed in routine water treatment practices. The fact that outbreaks of waterborne giardiasis have occurred when drinking water has been treated by chlorination (including the 2004 outbreak of waterborne giardiasis in Bergen, Norway with over 1200 laboratory-diagnosed cases, and probably many more actual cases ( Robertson et al. 2006) provides further evidence that this treatment is insufficient to inactivate Giardia cysts. Chlorine is particularly useful as a drinking water treatment against bacteria and viruses, as the residual is protective against these potential pathogens in the distribution network, but its limited potency against protozoan parasites (Giardia and Cryptosporidium) is broadly accepted.

Chlorine dioxide and chloramination: Use of N-halamine compounds were investigated as an alternative to chlorine. In experiments using Giardia cysts derived from dogs (either naturally or experimentally infected), Kong et al. 1988 demonstrated that the compound 3-chloro-4,4-dimethyl-2-oxazolidinone reduced excystation at lower concentrations or after shorter contact times at a given total chlorine concentration than did free chlorine. Experiments using a commercially available chlorine dioxide preparation, designed for the sterilization of small volumes of drinking water, demonstrated that under the conditions recommended by the manufacturer, there was limited inactivation of Giardia cysts, as assessed by vital dye staining and excystation (Winiecka-Krusnell & Linder, 1998). However, prolonged exposure at higher concentrations was more effective. Despite the apparent superiority of chloramination over chlorination, the apparent lack of oocysticidal effect of chloramine on Cryptosporidium ( Korich et al. 1990) has meant that research has focused largely on other disinfection approaches.

Ozonisation: Initial investigations on the use of ozone of inactivate Giardia cysts produced relatively promising results ( Wickramanayake et al. 1984), as Giardia cysts seemed considerably more susceptible to ozonisation than chlorination, requiring a C·t’ value to achieve 99 % inactivation (as assessed by in vitro excystation) only twice or three times that needed for inactivation of bacterial and virus, compared with the much higher relative C·t’ values needed for chlorination. As with chlorination, the efficacy of ozonisation was also apparently reduced at lower temperatures, although the temperature effect was apparently not so high as with chlorination. However, a study by Finch et al. 1993, that used infectivity to assess the susceptibility of both G. muris cysts and G. duodenalis cysts to inactivation by ozone demonstrated not only that both species were of similar sensitivity to ozonisation, but also that the simple C·t’ value necessary for 3 log unit inactivation was over twice that recommended by the Surface Water Treatment Rule. Similar values were obtained in the study by Widmer et al. 2002, and this study also demonstrated, by a variety of approaches, that ozone inactivates Giardia cysts by either direct or indirect degradation of proteins (Widmer et al. 2002). The requirement for higher activation energy for protozoa than for bacteria, especially at low temperatures, requires significantly higher ozone exposures, and these may result in the formation of disinfection by-products, of which bromate is considered to be of most concern, due to its potentially carcinogenic effects, and occurs during ozonisation of bromide-containing waters ( von Gunten U 2003). To avoid this, further manipulation of the water treatment processes are recommended (e.g. alteration of pH). Another aspect that has been indicated for consideration regarding water disinfection and Giardia cyst inactivation by ozonisation, is that the disinfection efficacy apparently reduces as cyst concentration decreases ( Haas and Kaymak 2003). As most bench-scale trials use large numbers of cysts, but the concentrations of Giardia cysts in water are rather low, this is potentially very important as it implies that in real-world water treatment situations the disinfection efficacy of ozone against Giardia may be less than assumed from the results of laboratory studies.

UV irradiation: UV technology was first utilized in water treatment in Marseille in 1910 (Hijnen et al. 2005) and is considered to have several advantages as a disinfection process ( Betancourt and Rose 2004). These include: (a) it is a physical process that does not rely on the use of chemical additions; (b) it requires relatively short contact times; and (c) UV disinfection by-products have not been identified to date. The disadvantages of UV treatment are: (i) variations in output amongst the different types of UV lamps, and reactor design and scale-up issues; (ii) difficulties in measuring dose in practice; (iii) interference due to turbidity/particulate matter; and (iv) lack of residual disinfection effect. UV irradiation was first identified as a potentially useful technology for inactivating Cryptosporidium oocysts in drinking water in 1995 ( Campbell et al. 1995), followed up by work by Clancy et al. 1998, and since then there have been various studies on the inactivation of Giardia cysts by UV. These have generally demonstrated that Giardia cysts are susceptible to UV disinfection at doses that are applicable to the water ( Craik et al. 2000; Belosevic et al. 2001; Campbell and Wallis 2002; Linden et al. 2002; Mofidi et al. 2002; Shin et al. 2009). Microorganisms have two mechanisms by which they might repair DNA damage resulting from UV exposure, photoreactivation and dark repair. The results of experiments investigating whether these may occur in Giardia cysts are conflicting: whereas Craik et al. 2000 and Linden et al. 2002 concluded that although G. muris and G. duodenalis cysts lose their infective ability permanently as a result of UV damage, Belosevic et al. 2001 demonstrated some DNA repair after UV radiation and Kruithof et al. 2005 also demonstrated in vivo reactivation of G. muris cysts, and Shin et al. 2005 showed that there is some repair also of G. duodenalis cysts following UV exposure.  Some research has also indicated that some isolates of Giardia may be more or less sensitive to UV irradiation ( Li et al. 2007).
Wastewater is often also treated by UV, but infection studies using UV-treated wastewater treatment plant samples ( Neto et al. 2006) indicated that the infectivity of the naturally occurring Giardia cysts was not completely abrogated by this treatment, possibly because of the relatively high turbidity and quantities of particulate matter, and in a similar study in Canada, Li et al. 2009 demonstrated that Giardia cysts in secondary effluent both upstream and downstream of UV reactors in wastewater treatment plants could cause infections in gerbils, indicating that the inactivation provided by the UV systems was less than anticipated from the UV dose responses published in the literature. The authors suggest various alternative hypotheses for these results including adhesion of bacteria to the cysts and/or absorption of colloidal material hindering penetration of UV irradiation, repair of damage by the cysts to an extent that a proportion regained infectivity, and the UV dose being lower than recorded.

Solar disinfection: batch solar disinfection (SODIS) has been suggested as an appropriate household water treatment technology for use as an emergency intervention in aftermath of natural or man-made disasters and has been demonstrated to be effective against Giardia cysts ( McGuigan et al. 2006).

Photocatalytic disinfection: Photocatalytic disinfection is a nanotechnology system of disinfection based on the interaction between light and solid semiconductor particles. First suggested as a promising source of hydroxyl radicals for disinfection of water (Matsunaga et al. 1985), photocatalysis using titanium dioxide (TiO2) as a sensitiser, was first investigated for inactivation of Cryptosporidium oocysts in 2002, with promising results ( Curtis et al. 2002). Experiments by Lee et al. 2004 suggested this technology could also be applicable to inactivation of Giardia cysts, and Yu and Kim 2004 proposed that a photoreactor involving TiO2 immobilized optic fibre reactor could be usefully applied to Giardia inactivation. A further study using TiO2 and modified catalyst silver loaded TiO2 (Ag- TiO2) ( Sökmen et al. 2008) indicated not only that photocatalytic disinfection might be an environmentally friendly technology for water disinfection, but that TiO2 thin film coated materials (glass or PET) may be more promising for total inactivation of Giardia, and that the photocatalytic action of TiO2 could be improved by catalyst modification. The authors recommend that feasible photoreactor design should be seriously investigated for online water treatment. A more recent study ( Navalon et al. 2009) demonstrates the efficacy of this technique in inactivating Giardia cysts (and Cryptosporidium oocysts) using a a commercial fibrous ceramic TiO2 photocatalyst. The efficiency was enhanced by addition of a small concentration of chlorine and the authors consider that the photocatalytic process is suitable for safe and complete water disinfection.

How the viability of Giardia cysts can best be assessed has been a matter of research interest for many years. As only viable Giardia cysts are of public health significance, the occurrence of non-viable Giardia cysts in a treated water supply (for example) is of less importance (although obviously is indicative of the potential for contamination). The gold-standard for viability of Cryptosporidium parvum oocysts is infectivity in an animal model, but for Giardia cysts this is more problematic as some cyst isolates that are known to be infectious to humans do not easily establish in animal models such as gerbils (for example, Giardia of assemblage B; see Benere et al. submitted/in press). Additionally, infectivity models generally rely on large numbers of clean cysts, and these are seldom encountered in environmental samples. Thus, although infectivity models may be useful for investigating the effects of different parameters on cyst viability, they are not so useful for assessing whether cysts in environmental samples are viable.

As well as animal infectivity, various different surrogate models for assessing Giardia cyst viability have been developed and used in different studies in order to investigate survival of Giardia cysts under different conditions. These models include: 1) in vitro excystation ( Smith and Smith 1989); 2) cyst morphology by light microscopy ( Schupp and Erlandsen 1987; Thiriat et al. 1998); 3) uptake or exclusion of fluorogenic dyes ( Smith and Smith 1989; Taghi-Kilani et al. 1996; Dowd and Pillai 1997; Thiriat et al. 1998); 4) fluorescent in situ hybridisation (FISH) ( Graczyk et al. 2003); 5) electrorotation ( Dalton et al. 2001); 6) heat shock mRNA analysis ( Abbaszadegan et al. 1997; Lee et al. 2009) 7) Quantification of mRNA transcripts during excystation ( Bertrand et al. 2009). None of these different surrogate models is considered to be perfect, each having their own difficulties. Whilst in vitro excystation may underestimate viability, the use of fluorogenic dyes and/or FISH may overestimate viability (Robertson et al. 2008). Nevertheless, despite the limitations of these methods, they may provide useful indications on the survival of Giardia cysts under different conditions, although relatively few data have been published. Giardia muris cysts are frequently used instead of G. duodenalis for survival experiments, as their infectivity to mice provides a useful handle, they are relatively easily propagated and obtained, and their survival characteristics are likely to be very similar to those of G. duodenalis cysts.

An important determinant of Giardia cyst survival in natural waters is ambient temperature. Lower temperatures down to the freezing point are generally beneficial for survival, but freeze-thaw cycles damage cyst integrity ( Robertson and Gjerde 2006). Two very early studies that used eosin dye exclusion as a marker for viability, reported thermal death points of Giardia cysts to be 50°C ( Cerva L 1955) and 64°C ( Boeck WC 1921). Based on excystation capability, Schaefer et al. 1984 determined the thermal death point of Giardia muris cysts to be approximately 54°C.
Selected studies on survival are summarized in the following:

  • Robertson and Gjerde 2006 investigated the fate of Giardia cysts and Cryptosporidium overwintering in a natural river environment in Norway. Purified cysts and oocysts (app. 5 x 104 per ml) were submerged in filter chambers at a depth of app. 1 m in a river. Temperature at this depth ranged between app. 1 and 7°C during the study period, and freezing of the cyst/oocyst suspension was not observed. Viability was monitored during winter based on morphology and exclusion/inclusion of dyes (PI and DAPI). Viable Giardia cysts were not detected after 1 month (29 days), but cysts stored in distilled water at 4°C could be detected up to 39 days. Loss of viability in the river was more rapid for Giardia cysts than for Cryptosporidium oocysts. In a previous similar study in a terrestrial winter environment, loss of viability had been more rapid, probably due to fragmentation due to freeze-thaw cycles.
  • Karim et al. 2004 compared the persistence of indicator microorganisms and waterborne pathogens in sediment and water column as occur in wetlands used for wastewater treatment in Arizona, USA. Giardia muris cysts were used to study die off rates in the two environments, with viability determination by in vitro excystation.  Survival was more prolonged in the water column, with die-off rates of 0.028 log10 per day in water and 0.37 log10 per day in sediment. The authors proposed that the more rapid die-off in sediment may be due to biological antagonism or the presence of organic substances.
  • DeRegnier et al. 1989 studied the viability of Giardia muris cysts submerged in perforated vials (containing fecal biomass) in lake, river, Minneapolis tap water, and refrigerated distilled water (5-7°C) as a control. Samples were removed from each water environment at 3,7,14, 28, 56, and 84 days. Viability was assessed by exclusion of propidium iodide, mouse infectivity, and cyst morphology. Cysts suspended in lake water remained viable for up to 56 days at a depth of 30 feet (water temperature 6-7°C), but only of 28 days at 15 feet (temperature 17-20°C). Cysts from the river water environment in a fall and winter experiment could infect mice for at least up to 28 days and 56 days, respectively. Water temperatures ranged between 19 and 27°C in fall and 0 and 2°C in winter. Tap water (20-28°C) seemed to inactivate cysts within 14 days (viability dropped to <2% at day 7; residual chloramine level in the range between 2.9 to 3.6 mg per liter), whereas cysts in the refrigerated control remained viable for up to 56 days. Among nine water quality parameters tested (dissolved oxygen, pH, total hardness, nitrate, nitrogen, phosphorous, turbidity, and color), only temperature correlated with cyst viability, with more prolonged survival at lower temperatures. However, in another study investigating survival of Giardia cysts (and other waterborne pathogens) in marine waters, factors other than temperature (including water quality, salinity, and the presence of light) also impacted on cyst survival ( Rose et al. 1997).
The theoretical infectious dose for Giardia is ingestion of a single infectious cyst, and cysts are immediately infectious upon excretion and are able to survive for considerable periods in the environment. In Rendtorffs classic infection study from the 1950s ( Rendtorff RC 1954), doses ranging from 1 to 106 were ingested by volunteers, and at a dose of 10 cysts was reported to result in infection in 100 % (2 out of 2) volunteers. Thus, it is traditionally accepted that the infectious dose for Giardia is relatively low. However, it is worth noting that although in this study 53% of the volunteers became infected, and changes in bowel motions were observed, none of the volunteers developed symptoms of giardiasis. Thus, although the individuals were infected, they did not have classical giardiasis. The infection-to-illness ratio varies between isolates, as shown by the different response of volunteers subjects to two different isolates from symptomatic human infections in a study by Nash et al. 1987.

The probability of infection has since been described (Rose et al. 1991) by two exponential models: 

                         Psingle = 1 – exp(–rN)      and       Pannual = 1 – (1 – Psingle)EF

The variables in these models are defined as follows: Psingle = probability of infection for a single event, Pannual = annualized probability of infection, r = fraction of organisms ingested that initiate infection, and N = average number of ingested organisms. The value for r developed by Rose et al. was 0.01982 (95% confidence interval, 0.009798 to 0.03582).

Rose et al. 1991 estimated that an exposure to an annual geometric mean of 0.0007 cysts per 100 liters would result in a 1/10,000 annual risk of infection assuming that 2 liters of water are consumed daily. However, it should be noted that this would be a slight overestimation of risk as not all the cysts in the water might be viable, not all species might infect humans and a 2 liter daily water consumption might not generally apply. Underestimation of risk, on the other hand, might be caused by detection method inefficiencies, peak contamination levels, and prolonged duration of exposure. Additionally, this exponential model assumes that the microorganisms are distributed randomly in a given environmental medium (e.g., sediment) and follow the Poisson distribution. It also assumes that the probability of infection per ingested organism does not vary. However, with the overestimation and underestimation taken together, it was suggested that the model may be useful for estimating the probability of infection. The same authors presented data on cyst levels during five waterborne outbreak of giardiasis associated with unfiltered chlorinated surface waters. The following relationships between cyst levels and attack rates in different waterborne outbreaks were presented:  

Table: Relationship between Giardia cyst levels during five waterborne outbreaks and attack rates (data from Rose et al. 1991). ‘Attack rates’ were defined by the number of illnesses in the exposed population and probably underestimate the rate of infection as they do not account for unreported cases or asymptomatic infections.

In general lower cyst levels were associated with lower levels of infection. In addition to the degree of contamination, also the cyst viability and effectiveness of inactivation through chlorine and the length of exposure were reported as primary factors influencing the rate of infection.

In an estimation of the risk of becoming infected with Giardia (and other gastrointestinal pathogens) from accidental ingestion of contaminated sediments over a 1-year period from a river in New Jersey, USA ( Donovan et al. 2008a; Donovan et al. 2008b), the r value of 0.01982 developed by Rose et al. 1991 was also included in a Monte Carlo uncertainty analysis, which assumed an empirical distribution based on likelihood confidence intervals developed by Rose et al. Eight of 16 samples contained Giardia, and  these sampling data were used in the risk assessment, in which three potential exposure scenarios were considered: visitor, recreator (e.g. swimmer), and homeless person. Single-event risk was first evaluated for the three individual exposure scenarios; overall risk was then determined over a 1-year period using Monte Carlo techniques to characterize uncertainty. Based upon best estimates of the average concentration of Giardia cysts at three different point of the river, annual risk estimates for Giardia infection among recreators were 0.29, 0.64, and 0.14, respectively. For visitors, annual risk estimates for Giardia infection at these same places were 0.03, 0.10, and 0.01, respectively, based upon average concentrations of Giardia. The homeless person scenario presented the highest Giardia risks. Average Giardia concentrations of 4.64 cysts/g and 18.35 cysts/g at two places, respectively, resulted in annual risk estimates of 0.56 and 0.87, respectively, whilst at the third place, exposure to an average concentration of 1.62 cysts/g was associated with an annualized probability of infection of 0.30. The parameter that significantly contributed to the uncertainty in the risk estimates was exposure frequency. Across all locations and scenarios, exposure frequency contributed approximately 25 to 65% to the total uncertainty. The relative contribution of other factors appeared to be somewhat dependent upon the Giardia concentration. Thus, the human health risk assessment indicated that there was substantial risk (defined as risk greater than 1 in 10,000) associated with exposure to the Giardia-contaminated sediments at each location on this river.

In another study, the risk estimates associated with Giardia in drinking water derived from the dose-response parameter published in the literature were compared with the incidence of acute digestive conditions (ADC) measured in the framework of an epidemiological study in a general population ( Zmirou-Navier et al. 2006). In this study, a daily follow-up of digestive morbidity among a panel of 544 volunteers was combined with a microbiological surveillance of tap water, with Giardia dose estimated as being the product of drinking water intake (in L) multiplied by the logarithmic value of cysts concentrations. Using an Odds Ratio for one unit of dose [OR = 1.76 (95 % CI: 1.21, 2.55)] a very good consistency with the risk assessment estimate was demonstrated, provided that a 20 % abatement factor was applied to the cysts counts. The abatement parameter encompassed uncertainties associated with cyst viability, infectivity and virulence in natural settings. Thus, using these parameters, a daily water intake of 2 L and a Giardia cyst concentration of 10 cysts/100 L, yielded an estimated relative excess risk of 12 % according to the Rendtorff model, (against 11 % when multiplying the baseline rate of ADC by the corresponding OR). The authors concluded that the dose-response function for waterborne Giardia risk derived from clinical experiments is consistent with epidemiological data, but note that data are lacking regarding key characteristics that may have a significant impact on quantitative risk assessment results.

Despite considerable research on molecular methods for detecting and identifying Giardia cysts in water and environmental samples, the gold standard and the most widely used method remains the immunofluorescence antibody test (IFAT). In this assay, the water sample concentrate is labelled with a monoclonal antibody (mAb) against the Giardia cyst wall. The mAb binding is itself visualised by being bound to a fluorophore, commonly fluorescein isothiocyanate (FITC), but other fluorochromes may be used. 4’6 diamidino-2-phenyl indole (DAPI) staining is often used in addition, which highlights the nuclei within the cyst. Both the mAb staining and the DAPI staining are visualised using fluorescence microscopy. Light microscopy, usually using DIC/Normarski optics, is often used in addition to examine the morphology of objects that are stained with the mAb, as cross-reactivity often occurs with algae and other particulates in water sample concentrates. IFAT may also be screened using an automated method or by flow cytometric separation.

Molecular methods are more commonly used for genotyping any cysts identified in order to determine whether they are of assemblage A or B (and therefore are of public health significance as they have the potential to infect humans), and also, if they belong to another assemblage, to attempt to determine their origins. One of the challenges with using molecular methods for identifying Giardia cysts from water/environmental samples is the efficiency of the DNA extraction rate from, frequently, relatively few organisms. Additionally, for those methods based on PCR, the presence of inhibitors, such as humic acids, fulvic acids, and phenols, in the matrix also poses a problem. Comparisons of DNA extraction methods for Giardia cysts in environmental matrices have been conducted, particularly without the use of IMS, which is an expensive step in the procedure. A combination of freeze-thaw, sonication, and kits have been recommended ( Guy et al. 2003), and the use of Chelex 100 chelating resin post-lysis ( Anceno et al. 2007; Yu et al. 2009).

Some advances in the use of molecular detection of Giardia cysts in water/environmental analysis are outlined below: 

  • LAMP: Loop mediated isothermal amplification (LAMP) is based on auto cycling strand displacement DNA synthesis by Bst polymerase. In testing the use of this technology on water and environmental samples ( Plutzer and Karanis 2009), of 10 surface water samples that were positive by IFAT, 7 were positive using LAMP, with primers targeting the Elongation Factor 1 alpha (EF-1α) gene, and of 15 sewage water samples that were positive by IFAT, 9 were positive by LAMP. Some samples that were positive by LAMP were considered to contain relatively few cysts (1 in 10 L), but other samples that were negative by LAMP had relatively high concentrations of cysts (400 cysts in 0.5 L). Use of standard PCR on the same samples found 4 of the surface water samples positive and 10 of the sewage water samples positive using primers targeting the SSU rRNA gene, and 3 of the surface water samples positive, and 9 of the sewage water samples using primers targeting the glutamate dehydrogenase (GDH) gene. Samples positive by LAMP were not always those that were positive by PCR, and vice versa. The authors considered LAMP to be superior to PCR due to its relative ease and cheapness (not requiring a thermal cycler) and also because it is not affected by inhibitors in the sample. The lack of quantification and problems with false negatives are problems that suggest that this technology is not yet sufficiently developed for routine analysis. However a later publication (Plutzer et al. 2010) using spiked drinking water samples did not apparently have this problem, and cyst concentrations down to 1 cyst in 10 L were detected, even in high turbidity water samples. The authors emphasise that the lack of requirement for immunomagnetic separation means that this is a cost-effective method, as is the lack of necessity for a fluorescence microscope (need for IFAT) or a thermal cycler (required for PCR). However, the reagents for extraction of DNA are necessary, as well as those for the LAMP reaction. A positive reaction is detected as a white turbidity, due to the formation of magnesium pyrophosphate, a by-product of the amplification reaction, that is produced in proportion to the amount of amplified products. The authors suggest that use of a turbidimeter can be used to quantify the results.
  • RT-PCR: Reverse transcription polymerase chain reaction (RT-PCR) targeting the heat shock protein gene was proposed over 12 years ago as a method for detecting only viable Giardia cysts in water samples ( Abbaszadegan et al. 1997).  However, it had low sensitivity (103 cysts/100 μl). More promising results were achieved by Kaucner and Stinear 1998, but was not taken up by other laboratories. A yet more recent publication, ( Lee et al. 2009), using a new primer set against the same gene, has suggested a sensitivity of 1 cyst per 100 µl water concentrate can be achieved when heat shock treatment is applied.
  • Real-time PCR: Real-time PCR (qPCR) for detection of Giardia in water/environmental samples was first described by Guy et al. 2003, using primers targeting the ß-giardin gene. It was considered to be sensitive down to a single cyst, but in environmental samples the quantitative evaluation was dependent on inhibitors. A later publication ( Bertrand et al. 2004) sought to improve the specificity, and was specifically directed towards detection and quantification of Giardia cysts in wastewater, and used probes targeted to the elongation factor 1A gene. A sensitivity of 18 cysts in 200 μl purified suspension (or 180 cysts/l of wastewater assuming that the PCR efficiency is not decreased in environmental samples) was achieved. However, when applied to 6 wastewater samples, only 5 produced amplification curves, and quantification by IFA always gave higher results than quantification by qPCR. Nevertheless, the authors were of the opinion that the qPCR assay provided a good indication of the level of Giardia contamination.
  • PCR: The publication by Rochelle et al. 1997 provided the first real information on the potential for using PCR for detecting waterborne Giardia cysts, although the authors were careful to emphasize both the requirement for refinement of the technique (primers, PCR optimisation) and the need for parallel use of conventional techniques. Since then, PCR has been used in a variety of different studies for detection and identification of Giardia cysts in water and environmental samples. Rimhanen-Finne et al. 2002 published a paper in which the detection limit for Giardia cysts by PCR (coupled with IMS) was 50 cysts in 2 litres of surface water, and used this method to detect Giardia contamination in one out of 54 surface waters. As cysts are often in low concentrations, detection sensitivity must be maximized and this can be increased by using nested PCR. A study on detection of Giardia cysts using a nested PCR targeting the triosephosphate isomerase gene demonstrated that for pure isolates, in which the cysts were transferred directly into the PCR tubes, 80 % of 50 samples gave a positive result, whilst 100 % of 50 replicates containing 10 cysts gave a positive result ( Miller and Sterling 2007).
The following pictures are part of the Public Health Image Library (PHIL) from the Centers for Disease Control and Prevention (CDC). The photos are in the public domain and thus free of copyright restrictions. We would like to express our appreciation for providing these images.


Figure 1:

                                  Giardia_photo 1         

This photomicrograph depicts a Giardia lamblia cyst using an iodine staining technique.

Source: http://phil.cdc.gov/phil/home.asp
Photo ID: 3742
Content provider(s): Centers for Disease Control and Prevention

Figure 2:

                                  Giardia_photo 2

This photomicrograph depicts Giardia lamblia parasites using an indirect immunofluorescence test for giardiasis.

Source: http://phil.cdc.gov/phil/home.asp
Photo ID: 3746
Content provider(s): Centers for Disease Control and Prevention/Dr. Visvesvara

Figure 3:

                                   Giardia_photo 3

This scanning electron micrograph (SEM) reveals some of the external ultrastructural details displayed by a flagellated Giardia lamblia protozoan parasite. G. lamblia is the organism responsible for causing the diarrheal disease "giardiasis". Once an animal or person has been infected with this protozoan, the parasite lives in the intestine, and is passed in the stool. Because the parasite is protected by an outer shell, it can survive outside the body, and in the environment for long periods of time.

Source: http://phil.cdc.gov/phil/home.asp
Photo ID: 8698
Content provider(s): Centers for Disease Control and Prevention/ Janice Carr

Figure 4:

                                   Giardia_photo 4    

This digitally-colorized scanning electron micrograph (SEM) depicted a Giardia lamblia protozoan that was about to become two, separate organisms, as it was caught in a late stage of cell division, producing a heart-shaped form.

Source: http://phil.cdc.gov/phil/home.asp
Photo ID: 11652
Content provider(s): Centers for Disease Control and Prevention/ Dr. Stan Erlandsen
Credit: Dr. Stan Erlandsen 

Last Updated on Tuesday, 20 September 2011 13:33


Hepatitis A
Hepatitis E


Copyright © 2017 Waterborne Pathogens. All Rights Reserved. Powered by SuSanA