The prevalence rate of E. multilocularis in wild foxes has dramatically increased over the past few decades [30] (Fig. 2). In 1985, less than 10% of foxes were reportedly infected, however, by 1998 this figure had risen to 58.4%, and it has continued to increase over the past few years. Due to the high prevalence rate of echinococcosis in wild red foxes, as well as their increasing population, they are considered to be the major definitive hosts of the parasite in Japan [32]. Although most of the parasite biomass occurs in foxes, other definitive hosts may also serve as sources of infection [35], however, a mathematical model of egg excretion dynamics has suggested that foxes have a higher mean biotic potential than any other known definitive hosts [22].

In addition to the high prevalence rate of echinococcosis in wild red foxes, the invasion of red foxes into urban areas has also been of concern because it indicates that an increasing infection pressure for densely populated areas is inevitable. For these reasons, it is expected that an urban cycle of E. multilocularis will eventually occur in the most populous city of Hokkaido, Sapporo, and most surveys conducted within the city, either by necropsy or coproantigen detection, have already registered the presence of Echinococcus-infected foxes in parks and woodlands [43]. Further, large amounts of E. multilocularis coproantigen positive fox feces have also been reported in urban areas adjacent to recreational parks [43]. Other suitable intermediate hosts have also been trapped in recreational parks, however none of these were found to be infected [43]. This phenomenon has also been documented in other endemic countries in which an increasing number of infected foxes have been found foraging in cities and villages [7]. For example, in Europe, prevalence rates in foxes have risen in many agricultural dominated areas in France, The Netherlands, Germany, Austria, Slovakia and Poland, however, the life cycle has also been established in many urban areas in which foxes are present in high population densities, which presents an increased risk of infection for large human populations [4,35].

In Japan, raccoon dogs have also been found to be infected with E. multilocularis [48]. Based on experimental infection studies they were suggested to be capable of playing a significant role in the epidemiology of alveolar echinococcosis [41]. Although their low population in Hokkaido indicates that they may have a lesser role in transmission of the disease, raccoon dogs have a greater reproduction potential than foxes, and their impact may increase with the effects of global warming [41].

Prevalence studies of echinococcosis in dogs in Japan have been very limited; however, a 30 year survey by necropsy examinations of 9,930 dogs from 1966 conducted by the Hokkaido government revealed 98 infected dogs. Recently, annual examinations of less than 10 dogs conducted by the Hokkaido government registered zero incidence rate. However, the Forum on Environment and Animals (FEA), from April 2004 through August 2005, examined a total of 1,460 fecal samples obtained from domestic dogs nationwide by animal clinics found 4 (0.27%) dogs that were positive for echinococcosis based on coproantigen and PCR assays [18]. It was estimated that nearly ten thousand pet dogs are transported between Honshu and Hokkaido by plane and ferry every year, and that this includes up to 30 E. multilocularis-infected animals [8]. Further, it has been reported that 2 out of 69 dogs that were moved from Hokkaido to Honshu were found to be positive for E. multilocularis based on coproantigen examination [21]. This has raised concerns that echinococcosis might spread to Honshu as a result of pet dog translocation. In addition, a dog in Saitama prefecture on the main island (adjacent to Tokyo) was found to have E. multilocularis eggs in its feces [47].

Additionally, a recent survey of dogs transported through ferry ports in Hokkaido found 2 dogs (2/183) that were positive for E. multilocularis coproantigen [28], and one of these dogs was a non-resident of Hokkaido that had been permitted to roam freely for only a few hours during its 5 day stay. Taken together, these findings suggest that there has been a rise in the infection rate of domestic dogs in Hokkaido, and that these dogs have the potential to spread the disease throughout Japan. It should be noted, however, that infection in domestic dogs can only be spread by way of highly contaminated rodent intermediate hosts, which are closely associated with the high infection prevalence rates observed in wild foxes [12]. Nevertheless, it has been demonstrated that dogs have a high biotic potential and contribution to the transmission cycle of echinococcosis [22], and the lifetime incidence in dogs being infected with E. multilocularis at least once can reach about 10% even in a population that has only a low prevalence rate of infection [7].

Echinococcosis infection in cats has also been reported, however, in all cases the tapeworms collected were immature in form [46], therefore, it has been suggested that cats only play a minor role in the maintenance of E. multilocularis in endemic areas [41]. In spite of this, in Japan a cat was recently shown to be excreting taeniid eggs that were confirmed to be E. multilocularis by PCR. This recent finding is similar to observations in Europe, which indicate that cats are potential sources of infective eggs.

Rausch et al. [33] conducted the first study on anthelmintic treatment in a 10-year trial involving monthly deworming of dogs with praziquantel in a village in St. Lawrence, Alaska a hyperendemic area to E. multilocularis. Although discontinued recently because of the cost of the drug, the replacement of sled-dogs by machines decreased remarkably the incidence of alveolar echinococcosis in the island. However, with the presence of other definitive hosts, i.e. arctic foxes, different dimensions of the problem are expected to surface.

It has since been asserted that there is no reliable, cost-effective method for the sustainable control or eradication of E. multilocularis in the sylvatic cycle [10]. In spite of this assertion, our reference laboratory has been deworming red foxes in Koshimizu (200 km²), Hokkaido, Japan since 1997. To accomplish this, a survey was initially conducted to locate fox dens, and then fox feces were collected from the vicinity of the dens and examined for the presence of taeniid eggs and Echinococcus coproantigen. The following year, anthelmintic-fortified bait, which consisted of commercial fish sausages (1.5 cm long) embedded with one half of a 25 mg praziquantel tablet (Droncit; Bayer, Germany), was distributed manually in approximately half of the total area around the fox dens on a monthly basis on foot. The baiting trial showed that there was a decrease in the taeniid egg infection rate in foxes in the baited area after one month, and that this suppressive effect persisted in the following years, despite a decrease in the number of times the bait was distributed. The trial also showed that intermediate host rodents born following the bait distribution had a significantly lower prevalence of infection than the overwintered older rodents [42].

In a follow-up study conducted in April 2001, praziquantel-fortified bait was distributed throughout the entire area of Koshimizu alongside roads, at intersections and at wind-shield forests by local residents using cars to allow for faster mobility. The bait was made from fish-waste products, using the same procedure that is used for manufacturing "kamaboko" (Fig. 3) fortified with praziquantel (50 mg/ piece of bait). Based on a comparison of feces collected from foxes within the treatment area to feces collected from foxes outside the treatment area, which were used as a control, taeniid egg infection rates and coproantigen infection rates were significantly decreased in foxes inside the treatment area. This significant reduction of taeniid egg infection rates, however, was not observed until six months after the start of bait distribution, and the lower coproantigen positive rates were not observed for almost a year. A recent study found that, after continuous annual distribution of bait manufactured from fish-waste products the prevalence rate of coproantigen positive feces in 2006 was reduced to zero. In this study, local residents used cars to distribute bait annually alongside roads, at intersections between roads, and in wind-shield forests, and this proved to be a rapid method that did not require a large number of personnel and was highly effective at suppressing the infection rate of E. multilocularis in wild red foxes [19].

A control strategy initiated by local residents was also conducted in Kutchan, Hokkaido, Japan, which is another echinococcosis endemic area. A baseline study, in which fox feces were collected from a 100 km² study area, was conducted in July, September and November of 2005, prior to the distribution of bait. The prevalence rates of taeniid egg and E. multilocularis coproantigen positive feces were 7% (19/268) and 21% (55/268), respectively. Between May and November of 2006, a monthly distribution of approximately 1,500 pieces of bait was conducted throughout the study area by volunteers comprised of local residents of Kutchan. The bait was distributed with the use of GIS- based maps to identify the foraging habitat of wild foxes. Remarkably, the prevalence rates of taeniid egg and coproantigen positive feces dropped to 0% and 2% after less than a year of baiting. The results obtained when this strategy was used indicate that distribution of praziquantel-fortified bait using GIS-based maps could allow bait distribution to be restricted to only areas commonly visited by foxes, thereby cutting costs and time.

The baiting system implemented in Japan varies from that of Germany, which started wild fox deworming in 1990. Although the vegetation, quantity of snowfall, the species of voles involved and their habitat in Hokkaido are different from those of Europe, the primary difference in baiting systems used is that the post-deworming prevalence rates of infection in Europe are evaluated by hunting the foxes. In a baiting campaign that utilized planes to distribute bait over a large area, stronger effects were observed in the 156 km² core area than in the 6 to 10 km border area. It has been suggested that the border effect observed in this campaign may have occurred as a result of immigration of young, infected foxes [14,36]. Similarly, following control trials in northern Germany, the prevalence rate of infection recovered unexpectedly and rapidly, reaching pre-control levels 15 months after the end of the baiting campaign [13]. It is believed that this occurred as a result of young foxes being dispersed due to hunting pressure upon foxes in the border area, which resulted in there being vacant territories available for younger generations of infected foxes. While some studies have indicated that there were no significant age-dependent differences in the rate of E. multilocularis infection, other studies have found juvenile foxes to be more frequently infected than adults, and infection rates in young foxes have been found to be significantly greater under highly-endemic conditions than low-endemic conditions [11,40]. In addition, it has been reported that subadult foxes carry significantly higher worm burdens than adult foxes [15]. Taken together, these findings indicate that invasion of young infected foxes into the territories left by the hunted foxes could maintain or increase the prevalence rate of infection.

In Japan, however, the ecological niches of the foxes being treated were not disturbed because the efficacy of deworming was assessed using coproantigen detection in fox feces collected from the environment of baited areas instead of hunting. The differences observed in the efficacy of the Japanese treatment method and the German method indicate that the use of fox culling or hunting for evaluation of control efficacy is actually detrimental to the success of the baiting campaign [20]. Similarly, anthelmintic baiting of foxes against urban contamination with E. mutilocularis using intravital diagnosis for the assessment of efficacy was highly successful in Switzerland [14]. In the Swiss study, a pronounced reduction of E. multilocularis prevalence rates was observed in both the definitive and intermediate hosts when an approach combining anthelmintic baiting and coproantigen diagnosis was used.

Taken together, these results indicate that the use of intravital diagnosis, such as coproantigen [1] or copro-DNA [26] examination provides a superior means for assessing control interventions while preserving both the animals being treated and their environments.

"Endogenous development" involves building on local resources, enhancing in situ development, maximizing local control of the development process, and recognizing the needs and the values of local residents [34]. As the Dag Hammarskjöld Report [5] puts it, such development relies on what a human group has: its natural environment, its cultural heritage, the creativity of the men and women who constitute it, becoming richer through exchange between them and with other groups and entails the autonomous definition of development styles and of life styles.

In all of the baiting campaigns reviewed in this paper that were conducted in Japan, the endogenous initiative of local residents, which was facilitated by NonProfit Organization (NPO), was highly instrumental. Zoonotic diseases are of concern not only to public health personnel but also to individual residents who are at risk of infection. Moreover, the use of local resources, including local residents for fecal collection and bait distribution, locally produced fish-waste products for bait manufacturing, and local fund-raising to support the deworming program was found to be imperative for the success of a sustainable program for the control and prevention of echinoccocosis.

In a follow-up campaign that has been ongoing in Koshimizu since 2002, bait distribution, fecal collection and monitoring of echinococcosis in wild foxes was conducted by Okhotsk Sanctuary, a local NPO. Endogenous control initiatives introduced by this NPO provided a necessary solution that allowed the prolongation of the baiting campaign, thereby enabling a sustainable approach to the suppression of the E. multilocularis infection in wild red foxes, as indicated by the most recent prevalence rate of coproantigen positive feces of wild foxes in Koshimizu, which was 0%.

Another control initiative introduced by another NPO such as WAO in Kutchan, Hokkaido [31] also produced favorable results. To finance a long term preventive measure, WAO organized an echinococcosis control sticker sale campaign, in which stickers containing information regarding the life cycle of E. multilocularis and the threat of echinococcosis to public health are sold at approximately US $4. The proceeds of the sticker sales are then used to finance the program, including the costs of the anthelmintic-fortified baits and fecal examinations. In addition, WAO, with the help of the OIE reference lab, spearheaded an information dissemination initiative that involved local residents, including university students and children in elementary school.

The results of our research project, entitled "Prevention on the spread of areas that are endemic for zoonotic parasitic diseases" disclosed a strong possibility that dogs infected with Echinococcus could transmit the infection to their owners. Based on these findings, the Ministry of Health, Labor and Science, Japan directed the Hokkaido Prefectural Government to take measures to prevent the infection of pet owners from occurring, including campaigns to make the public aware of the potential threat [19].

In addition, an amendment to the Infectious Disease Law in Japan was made that required inclusion of certain specific zoonotic diseases in the 4th category (diseases which must be reported). The diseases added to this category included echinococcosis in dogs as well as bacterial dysentery in primates and West Nile fever in birds. In addition, during their 20th session, the Infectious Disease Evaluation Committee of the Ministry of Health, Labor, and Science passed a resolution that made it mandatory for veterinarians to report cases of echinococcosis in dogs to the health authorities. Thus, a national reporting system for dogs infected with E. multilocularis has been used by veterinarians since October 2004 [19].

The Ministry of Health, Labor, and Science, with the assistance of our laboratory, has also published guidelines regarding standard procedures and diagnostic measures to be taken when reports are submitted by veterinarians, and these guidelines have been distributed to local health offices, as well as to practicing veterinarians throughout Japan. The following three criteria for diagnosis are stipulated in the national reporting system, and a positive result in any of these should be reported to health authorities [19,30]:

It is believed that research institutions have an important role to play in extending practical assistance to the public as the knowledge and understanding of the disease advances, especially with regards to its control and prevention. Thus, according to Zinsstag [49] "Although there is no doubt that progress in animal health research must continue, it must also respond to societal needs and lead to solutions that can be delivered quickly." Therefore, in 1999, this OIE reference laboratory organized a scheme called the FEA, Japan. This scheme is able to link important organizations including government offices, academic institutions, international agencies (e.g. the OIE), veterinary associations and non-governmental organizations, such as NPOs comprised of local residents, that all have the primary goal of controlling echinococcosis in Hokkaido.

The FEA is presently serving as a hub for private veterinarians involved in small animal practice throughout Japan for the confirmatory diagnosis of echinococcosis. Veterinarians who suspect Echinococcus infection in dogs, cats, or other susceptible definitive hosts send fecal samples to the FEA, which then conducts laboratory examinations. Further, the FEA has assisted with the endogenous initiative of NPOs by providing them with technical expertise, laboratory examinations of fox feces, anthelmintic- fortified baits and necessary materials such as the "eki-bin" (echi-bottle), which are containers used for the collection of fecal samples safely. In addition, the FEA provides intravital diagnosis using ELISA (EmA9) to determine the prevalence rate of echinococcosis in foxes during pre- and post-baiting campaigns. Overall, the FEA has enhanced the connection between laboratory findings and field applications through accurate diagnosis and proper monitoring of echinococcosis in Japan.

Recently, "Full-Echinococcus", a database of full-length cDNAs obtained from a human parasite, E. multilocularis, was produced in cooperation with this reference laboratory. The full-length cDNA library was produced using the Vector-trapper method on hydatid cysts developed in cotton rats that were infected with E.multilocularis. A total of 10,966 5'end-one-pass sequences were compared with the non-redundant database, DDBJ/Genbank/EMBL, using the BLAST and TBLASTX programs. Two-thirds of the sequences were considered to be derived from Echinococcus, while the remaining one-third represented host genes. Many of the former clones represent full-length cDNAs that are expressed in the larval stage, and these clones are available for further analysis and experiments.