Previous Page TOC Next Page

7. Characteristics of the virus and its method of production and dispersion

Rabbit haemorrhagic disease virus was first reported in the Peoples Republic of China in 1984 as the cause of a new highly transmissible and lethal pandemic of rabbits (Blancou, 1991). This disease, which is called RCD in Australia and New Zealand, was reported in the Republic of Korea in 1985 and had spread throughout the country by 1987. The disease first appeared in Europe in southern Italy in 1986 and also in Spain the following year. By 1988 it had broken out simultaneously in several other European countries, including France, Germany and Denmark. It also occurred in Mexico in 1988. In 1989 64 million farmed rabbits died of the disease in Italy and in 1990 it was recorded in United Kingdom. Reports from Asia and Europe were of a rapidly spreading condition causing high morbidity and mortality in adult rabbits. It is now present in more than 40 countries.

Death due to rabbit haemorrhagic disease was first reported in a shipment of German Angora rabbits three days after arrival in China. The Chinese reported that the disease subsequently spread to millions of Chinese rabbits. The early history of the disease is unknown and open to speculation. Possibilities are:

  • it existed as a lethal disease of rabbits, but in an area where its activity went unreported;
  • it existed as a non-lethal disease or inapparent infection of rabbits; and that a mutation occurred so that it evolved into a lethal disease;
  • it existed as an infection of another species, and that a mutation occurred so that it became infectious for rabbits and was lethal.
Smith suggests that when the disease was diagnosed it was the first occurrence of the disease as it was dramatic in effect and could not have gone unnoticed in either Germany or China. The disease then spread rapidly through Chinese rabbits and therefore is not presumed to be of Chinese origin.

"The most reasonable explanation for the sudden appearance and high mortality is that within 24-48 hours after arriving in China the entire shipment of Angora rabbits were exposed to a new agent from some unknown animal source either in their food, or their environment.


This rationale argues for a non-rabbit source of the rabbit haemorrhagic disease agent, and against host specificity."

He concludes that:

"Host specificity has to be proven before rabbit haemorrhagic disease agent can be approved as a biological control agent."

The nature of the causative agent and therefore the correct classification of the virus has been disputed and this continues to a limited extent. Chinese research workers, Ji, Du and Xu (1991), reported virus growth in cell culture but persistent infections have not been maintained by other workers. Because of the difficulty in growing RCD virus in culture Koch’s postulates have not been confirmed and therefore some submitters have challenged the identity of the virus. It is, however, not unusual for virologists to have difficulty in culturing viruses and confirming Koch’s postulates. Many examples of such cases exist in the literature.

The liver has yielded the highest virus concentration and most workers have extracted virus from this organ for virologic studies. Results of virus and nucleic acid extraction from different laboratories have varied widely. Korean workers and early reports from the People’s Republic of China (Xu and Chen, 1989) report a picornavirus. More recent work from China and also the United States of America reports a parvo-like virus (Xu, 1991. Gregg and House, 1989. Gregg per comm.), while the Europeans and Australians (Ohlinger et al. 1990, Rodak et al. 1990, Parra and Prieto, 1990, Meyers et al. 1991, and Chasey et al. 1992) believe the virus to be a member of the family Caliciviridae.

However, research workers in Europe and Australia have characterised the infectious agent found in liver preparations, thus strongly supporting the conclusion that the RCD virus is the cause of infection. I find that the following evidence provides the most compelling reasons for believing that a calicivirus is responsible for the disease:

  • the viral capsid protein of RCD virus is well characterised and has been cloned and sequenced. It has been sequenced in the baculovirus system and the protein self assembles into virus-like particles similar in size and morphology to native virus and is antigenically indistinguishable (Nagesha, et al. 1995).
  • The capsid determines the serotype of rabbit calicivirus, not any other calicivirus.
  • Recombinant capsid protein produces an immune response which protects against challenge with virulent RCD virus.
  • Monoclonal antibodies have been epitope-mapped to RCD virus capsid protein (Capucci et al. 1995). These antibodies neutralise in vivo haemagluttination experiments (Collins et al. 1996).
  • Purified RNA from liver preparations is infectious.
Parra, a virologist from Spain, also believes that the disease is caused by a calicivirus:

"Concerning the agent responsible for rabbit haemorrhagic disease (RHD), it is true that at first it has been identified by several authors as a member of different viral families; Parvoviridae, Picornaviridae and Caliciviridae. This was a consequence of the general lack of knowledge on the virus structure. Never the less, after the work of several European groups it very soon became clear that RHD virus is a calicivirus and not a parvo or picornavirus. With respect to the possibility of more than one viral agent being responsible for RHD, our results show that highly purified RHD virus virions are able to reproduce the disease in rabbits, showing all the clinical signs of RHD. Additionally, the disease can also be produced injecting the genetic material of the virus only (not the whole virus). Finally, the animals can be protected against RHD by immunization with a single protein component of RHD virus. It is clear to me that RHD is the result of the infection with a single type of virus."

CSIRO-AAHL has looked for evidence of parvovirus but have found none (Lenghaus, pers com.). Recent work by Capucci (pers comm) has shown that 60 hours after infection of the rabbit the RCD virus loses the outer shell of protein. It takes on a smooth appearance and is smaller in size and mass than before. This may explain the apparent discrepancy in descriptions of the causative virus and the belief that there is more than one virus. Shope believes that the evidence that it might be contaminated with a parvovirus is weak and should be discounted.

A newly identified non pathogenic calicivirus has been isolated from rabbits. It contains significant differences from the previously characterised RCD virus isolates in terms of pathogenicity, viral titre, tropism, and primary sequence of the structural protein. Cross protection studies, antigenic data and sequence comparisons demonstrate that the new virus is more closely related to RCD virus than the European brown hare syndrome virus. The existence of this nonpathogenic calicivirus, which the authors propose to name rabbit calicivirus, provides an explanation for the early discrepancies found in the course of serological surveys of the rabbit population in European countries. Infection with this calicivirus stimulates the production of antibody protecting rabbits against challenge with the pathogenic strain (Capucci et al. 1996).

Chasey and others from the Central Veterinary Laboratory and Central Science Laboratory in Britain reported at the European Society for Veterinary Virology, Calicivirus Symposium in October of 1996 that antibody to RCD virus "has existed for many years in UK rabbits" and that 64% of a sample 900 adult wild rabbit sera showed a protective level of antibodies before the disease spread to the UK. The immunity has presumably been induced by infection with the non-pathogenic virus identified by Capucci. See also European observations in section 3.1.

Several European studies have also shown RCD seroreactive rabbits at least 12 years before the recognition of RCD in China (Rodak et al. 1990; Rodak et al. 1991)

I am satisfied that rabbit calicivirus disease is caused by a calicivirus even without Koch’s postulates being fulfilled.

The concern about importing and eventually releasing a strain of RCD virus that contains contaminating agents is valid, but can be addressed by submitting the imported material to careful laboratory analysis. The techniques are the same as apply to a vaccine virus for veterinary or human use. Such a vaccine virus is rigidly tested for contaminating agents by the manufacturer and the government control authorities before being released. This has been done at AAHL with the Czechoslovakia strain which would be the source of virus for New Zealand. The risk of contamination has been addressed in the disease risk assessment.

The classification and structure of RCD virus is described in the Application. RCD virus is a single stranded positive sense RNA virus of 7437 nucleotides. A single major capsid protein of approximately 60 kDa has been identified.

Genetic change of RCD virus

Many submitters, including those with qualified support for the introduction of the virus, expressed concern about the possibility of the virus changing its behaviour so that it infected other species. There was also some expectation that the virus would lose its virulence and thereby its ability to kill rabbits.

Several submitters have expressed concern about introducing new virus diseases to New Zealand especially where is an apparent lack of knowledge as to how the virus would behave. Van Roy (673) emphasises that New Zealand has a choice over the introduction of this virus.

"If we took time to study all those hundreds of different RNA viruses currently in New Zealand, there would be some which we would definitely rather not have, and would not choose to take the risk of introducing if we had the choice. The threshold of certainty with respect to safety on the introduction of a new virus must always be considerably higher than that which would trigger concern and expenditure on research with a virus already in the country." (673)

Domingo also expresses concern:

"As a virologist having worked with RNA viruses for many years I feel a deep concern about the evolutionary potential of RCD in the field. The origin of this, as well as of other viruses, is uncertain. Host specificity is not a fixed trait in viruses. Viruses such as RCD mutate continuously and may also undergo recombination with genetic material of cellular and viral RNAs. This may result in changes of host specificity in ways that at present are unpredictable. Vaccines may no longer protect when a change in virulence or host specificity of RCD has occurred. Most RNA viruses are highly variable and they must not be released in the environment under any circumstances. It is playing with fire."

Smith also claims that an epidemic of RCD virus in a naive population of rabbits in New Zealand would expose non-target species to a much higher level of mutants and viral burdens than would be expected from low-level endemic diseases.

Steinhauer and Holland (1987) describe methods of genetic changes in RNA viruses as:

  • mutation
  • recombination
  • genetic re-assortment.

Burke is not aware of any published studies that address recombination between caliciviruses but evidence of nucleotide variation is known for all members of the Caliciviridae where it has been examined. As RCD virus is not a segmented virus, genetic re-assortment is not a viable method of change for this virus.

Mutation

Mutation is a change in individual bases of the RNA or DNA genomes. While virus genomes may be either RNA or DNA, other organisms are based only on DNA. As stated previously RCD virus is an RNA virus.

RNA viruses mutate at a higher frequency than DNA viruses and these mutations may lead to a change in the biological properties of the virus. The possible biological consequences of possible changes in viral RNA are:

  • antigenic change,
  • host cell specification changes, and/or
  • changes in disease patterns and virulence.

While DNA mutation rates have been estimated to be between 10-7 to 10-11 per base pair, per replication, the error rate of RNA polymerases average 10-4 per site following a limited number of replications (Steinhauer and Holland, 1987).

Scientists who have determined the sequence of RCD virus report a 1-4% change in the nucleotides over a 10 year period. For example, Boga et al (1994) showed that the 2,483 nucleotide sequence at the 3' end of the genome in the Spanish isolate of RCD virus (AST/89) was 95.4% similar to the corresponding section of the German isolate sequenced by Meyerset al. (1991). The structural proteins of the two strains differed in only 10 amino acid residues.

The outcomes of these changes in viral RNA are unpredictable and they may be successful or unsuccessful for the virus. Those that are successful may lead to a change in the biological properties of the virus. An unsuccessful mutation results in the virus being unable to replicate itself, so it becomes a crippled virus. Although a given RNA virus mutates at a high frequency, this does not necessarily mean that the virus will evolve (change) rapidly. This only happens when the mutations are successful.

The vast majority of these new genome combinations will be non-viable or non-competitive and will generally disappear. Most of the viable and competitive new ones will be unremarkable in their biological activities (Steinhauer and Holland, 1987).

The concern expressed by expert viral evolutionists is the potential for adverse evolution. This potential is present for RCD virus as it is for any other RNA virus that occur in New Zealand (including hepatitis A, rubella, measles, mumps, parainfluenza, the common cold, and Norwalk virus (a calicivirus)). In controlled laboratory experiments, scientists can select for a change in virulence, and also for a change in host range but in nature there are other selection pressures that reduce the possibility of these changes persisting.

The effect of mutation on virulence of RCD virus for the rabbit needs to be considered in the planning of a control programme. (See section 4.3) Natural selection of less virulent virus by repeated passage is a well described phenomenon, and can be expected with RCD virus. Similarly, the rabbit can be expected to undergo natural selection to resist disease caused by the virus. It has been suggested that regular reseeding of virulent RCD virus could be used to counter potential changes in the virus and its target host (Shope). Lektos suspect that the most useful period for use of the virus as a biological control agent would be in the first ten years.

Te Puni Kokiri requested that the reviewers quantify the risk of a mutation causing an adverse effect, but they were unable to do this. Burke and Shope responded in the following manner:

"In my opinion it is not possible to assess numerically the risk of an adverse mutation of RCV (RCD virus). Scientists who have determined the sequence of RCV report a 1-4% change in the nucleotides over a 10 year period. Many of these changes are silent in that they do not modify the make-up of the proteins of the virus. Those nucleotide changes that resulted in changes in the protein over this period have not modified the virulence or switched the host range of RCV as far as we know. (Shope)


Therefore, although we cannot assign a numerical risk value, we can say on a historical basis that such a mutation is highly unlikely. It is no more or less likely than that a change will occur in any one of the hundreds of other RNA viruses that infect animals and humans in New Zealand." (Shope)


"Given the current multi-continent geographic distribution of wild RCD virus (Europe and Australia), release of RCD virus in New Zealand is unlikely to add appreciably to the very low risk that RCD virus might - somewhere in the world - jump species barrier to cause a new disease." (Burke)


"The risk of an adverse epidemiological consequence of the intentional release of RCD virus is low but it is not zero." (Burke)

In view of the rarity of such events it is likely that the mathematical expression of such a risk, if it could be calculated, would have little meaning.

7.1 Host range

The European rabbit (Oryctolagus cuniculus) is the only known natural host of RCD virus but it is possible that other hosts exist. For example, only 7 of 48 species of rabbits and hares have been tested for susceptibility (Shope).

Several species of animal were tested at CSIRO-AAHL to determine the host range of RCD. Studies were conducted according to a protocol approved by a working party of the Australia and New Zealand Environment and Conservation Council. Animals were selected to provide a broad representation of a number of native and domestic animal species in Australia and New Zealand. Two species, the kiwi and short-tailed bat were selected by DoC for testing. A total of four animals from each species was tested except for the bats, where seven were tested. Consideration was given to the logistical problems associated with testing a large number of different animals and birds under conditions of high biological security.

Tests for host specificity at CSIRO-AAHL were carried out by intramuscular inoculation of 33 different animal species. These results have been analysed by the Bureau of Resource Sciences (BRS, 1996).

The pathological finding seen in the kiwis have been described by Buddle, et al:

"There were no gross or histological lesions suggestive of RCD virus infection observed in the two kiwis which were killed. However, several small white lesions (1-2 mm in diameter) were observed in the lung and liver of this bird and a parasite fragment was seen in the centre of one of these hepatic granulomas. The lung and liver lesions were consistent with a parasitic infection. This bird also had mild multiple foci of gliosis and degeneration of axons in the grey matter of the spinal cord with an accompanying mild meningitis. A focal granuloma was found in the white matter of the brain of this bird, just beneath the meninges of the cerebrum. There was a mild focal chronic myositis at the site of inoculation both birds."

There were no lesions suggestive of infection in the bats (Buddle, et al.)

Antibody was detected in some of the exposed Australian and New Zealand animals, notably kiwis where antibody persisted for at least 136 days. The ELISA results in the antibody tests for the foxes, chickens, falcons, seagulls, lizard and wombat are not considered indicative of a developing immune response to rabbit calicivirus as they do not show a consistent rise in titres. Results obtained for kiwis and mice were considered consistent with a developing immune reaction (BRS, 1996).

Tissues taken from two kiwi sacrificed for further study were processed and injected into susceptible rabbits. None of the 49 rabbits which received tissue extracts from the kiwi showed any clinical signs of RCD, none seroconverted to RCD, and none showed histological evidence of RCD infection. In addition, no RCD virus genetic material was detected by PCR assay of liver tissue from any of the 49 rabbits examined. No evidence of virus multiplication was found in kiwis or mice. The antibody detected may variably be induced by a sufficient antigenic mass without infection (i.e. replication of virus) occurring (Shope).

A panel of medical and veterinary virologists consulted on this issue concluded that ‘no further testing of kiwi is required and that kiwi can be regarded as not being susceptible to infection with the RCD virus.’

In spite of the criticism of the methodology of the host specificity testing and the inherent limitations of the trial there was no convincing evidence that any of the animals in the trial were infected with RCD virus.

The conclusion of the host specificity trial of RCD virus was based on the results of all of the tests that were applied. They included the use of sentinel rabbits, serology to detect antibodies to rabbit calicivirus, polymerase chain reaction to detect rabbit calicivirus genome, gross pathology, histopathology to detect viral lesions, and immunohistology to detect rabbit calicivirus antigen.

Experiments under different laboratory conditions, outside of Australia, also failed to show convincing evidence of infection in non-target species except that some authors have reported RCD virus causing disease in hares (Du, 1990, Nowotny, N., et al. 1997). However, many more authors insist that cross transmission attempts have been unsuccessful (BRS, 1996).

Antibody responses have been seen following the administration of RCD virus. Six foxes given oral doses of homogenised liver from rabbits that died from RCD showed pronounced antibody responses at 7 days and these persisted for 14 days before diminishing. Low titres were still evident in three foxes after 6 months, when the study finished. It was not determined if the virus infected and replicated in the exposed foxes. The foxes remained clinically healthy throughout the study but were not finally killed and examined pathologically (Leighton et al. 1995). A similar serological response was detected in dogs given virus by the intra ocular and intravenous routes and again the animals remained healthy (Simon, et al. 1994.) The slight increase in antibodies was attributed to the inoculation of foreign antigenic material. Simon et al. (1994) also showed that dogs fed livers from rabbits infected with RCD excreted the virus in their faeces.

Dissatisfaction was expressed by many submitters with the scientific data that was presented in the Application and in particular the laboratory studies of the host range conducted in Australia. It was felt that the claims for host specificity were not supported by sufficient data or were poorly founded. There are claims by the submitters that insufficient time and research effort has been put into these matters, with several commenting on the fact that the original research programme through to 1998, intended prior to the Australian release, has not been completed.

Numerous submitters point to the inadequate coverage of New Zealand species in the testing programme so far, and identify particular species for further research as the focus of their concerns: NZ dotterel, weka, terns; hawks and ground nesting birds; invertebrates and reptiles; waterfowl, pheasants, partridge, quail; cats; marine species; sportfish and game fish.

One submitter (673) suggested that there were three questions that needed to be answered:

" Has RCD ever infected any other species other than the European rabbit?

Is it now infecting any other species?

Could it be expected to infect any other species?"
The proof for a lack of replication, i.e. the proof of a negative can be very difficult. However, Smith raises the valid point that:

"the question was and is, can rabbit haemorrhagic disease virus infect any non-rabbit species?. .... The controversy of infection or simple antigenic response could have been easily resolved with very simple laboratory tests using killed rabbit haemorrhagic disease virus." (376)

The critics held that: virus might have been present early on after inoculation and was not tested for at that time; antibody studies by competitive ELISA were imperfectly standardised; and the route of infection was not a natural route. Each of these points has some merit, but the laboratory inoculation studies should be interpreted within the limits of what was done, not what was not done (Shope).

Burke notes that the method used to detect infection may not have been sufficiently sensitive to measure infections that were abortive or low-level but nonetheless genuine "infections." The term "productive infection" is sometimes used in an attempt to differentiate between low-level and high-level infections, but this term simply reflects the technology employed to detect "infection". It follows logically that if the definition of "infection" is imprecise and technology dependent, then the definition of "transmission" - infection of one individual by another - is also imprecise.

It is the impression of Burke that there are no known RNA viruses for which there is proven to be a true single host species specificity. Even for relatively host species-specific viruses, low level infections occur when the virus is inoculated at high levels into a different but related host species. Virtually all human RNA virus pathogens can infect chimpanzees and other primates. Most known positive strand RNA viruses belonging to the Picornaviridae, Flaviviridae, Alphaviridae, and Coronaviridae can even be passed successively through mice by serial tissue homogenate transfers (Burke).

Smith and Matson have pointed out that some caliciviruses such as the San Miguel Sea Lion Virus, have a broad host range, and that this was established only after 20 or more years of research. On the other hand, there are examples of RNA viruses, measles for instance, in other virus families that have an extremely restricted host range. The host range of measles (humans and higher primates) has been studied intensively for over 50 years and has not changed (Shope).

Tests on non target species have shown no convincing evidence of infection but it cannot be proven that they were not infected. Likewise, it has not been ruled out that subclinical infections occur. The fox and dog are possible examples of infections of this type.

A few submitters addressed the merits of field studies of the host range by comparing them with the laboratory studies. One of the submitters expressed the view that:

"field evidence tends to provide more valid reassurance, that this agent is less likely to produce infection in other species, to the extent that predictions can be made." (477)

Since it is impractical to test all non-target species in the laboratory, one is left with observations in nature, i.e. in Australia where the natural experiment is now going on.

7.2 Stability of the host relationship.

Many of the submitters have raised concerns about the possibility that RCD virus will jump species barriers or host switch. These terms apply when the normal host range of a virus changes and a new species is infected where it previously could not be.

For a virus to cross from one species to cause an epidemic in another, two thresholds must be crossed:

  • Infection: the virus infects one individual of the new host species.
  • Transmission: the virus is transmitted from one individual of the new host species to infect another individual.

Available evidence suggests that it is a very rare event for viruses to extend their host range. The only documented case of a host change, canine parvovirus, is under debate as it is argued that the virus could always infect dog tissues.

The caliciviruses have very stable host relationships in spite of the acknowledged facility of this group of viruses to change genetically. The close contact of human beings with the very common caliciviruses of cats (cat snuffles) without any evidence of adverse effects on human beings is also noted, although antibodies have been detected in human sera (Cubitt, pers comm.) The significance of the latter is unknown but it may be an example of a low level infection or a response to antigen.

7.3 Clinical and pathological effects.

Rabbit calicivirus is known to infect susceptible rabbits via the oral, nasal and parental (by injection) routes. The virus is present in the blood of infected rabbits and is present in a wide range of tissues during progression of the disease.

The incubation period is typically 24-48 hours with a maximum of 3 days. Death occurs 12-48 hours after clinical signs appear and these include anorexia, fever, apathy, dullness, prostration and lateral recumbency. Convulsions (Xu and Chen, 1989) and a terminal squeal have been reported (Fuller et al. 1993). According to Lenghaus et al. (1996) there appears to be minimal distress and no signs of pain. In 95% of cases rabbits die within24 - 72 hours of infection. Of the remainder, a few will recover and the others usually die within six days with signs of jaundice. The disease seen in rabbits which were exposed to direct contact with infected rabbits, or which consumed contaminated rabbit feed, was similar to that seen after direct inoculation of virus.

Under laboratory conditions the mortality rate is close to 100% (Lenghaus et al. 1995). Under field conditions, in the Yunta and Blinman regions of Australia, Newsome and Mutze (1995) estimated the mortality rates may have been greater than 80%. However, mortality rates in subsequent outbreaks in these areas have been very variable.

In a study by Lenghaus, 23 baby rabbits, in three litters, up to ten days old, survived RCD virus infection, although they did develop histological lesions and excreted sufficient virus to kill an in-contact adult rabbit. Six of 13 five-week old rabbits survived natural infection. Only two of more than 200 rabbits older than about eight weeks, survived after challenge with the virus.

There is a marked age difference in susceptibility to RCD virus. Rabbits up to eight weeks of age have an inherent immunity to the virus.

Researchers have not been able to show that a carrier state is established in survivors of RCD, as judged by the inability of these survivors to re-infect other susceptible rabbits (Lenghaus et al. 1996). This in itself raises questions as the disease appears able to recur in the same area 10 months later. The question must be raised as to how the virus is able to re-appear after that period of time. A carrier state is characteristic for some of the other caliciviruses such as feline calicivirus and smooth round-structured viruses.

Survivors of RCD appear to have a life long immunity, with high stable antibody titres which are maintained for 12-18 months. Passive maternal antibody in offspring waned progressively, so that offspring of immune does were fully susceptible to infection by 12 weeks of age (Lenghaus et al. 1995). Field evidence in Australia suggests that this immunity to infection persists for at least two months.

Necropsy findings

The most consistent gross findings are an enlarged firm dark spleen, a pale swollen liver, wet lungs which fail to collapse and clear stable froth in the trachea. In dead rabbits left for some hours before necropsy, the froth becomes progressively more blood-stained and sometimes exudes from the nostrils. Occasionally the kidneys are also swollen and a dark red colour. Recent field experiences in Australia suggests that field diagnosis of RCD from gross pathology is difficult.

Histologically, there is a characteristic coagulative necrosis of hepatocytes in periportal areas of the liver with hepatocytes of the mid-zone and centrilobular areas more sporadically affected. There are numerous thrombi in hepatic sinusoids. There is massive destruction of the fixed histiocytes lining the sinusoids of the spleen. Normal sinusoidal architecture is almost completely destroyed. Specific stains have demonstrated that degeneration andnecrosis of hepatocytes and splenic histiocytes is related to the presence of RCD virus. Death is considered to be due to respiratory failure and heart failure following a disseminated intravascular coagulation (DIC), with numerous fibrin thrombo-emboli present, particularly in blood vessels of the heart, lungs and kidneys (Lenghaus et al. 1996).

Pathogenesis

Little work has been published on the pathogenesis of RCD and Parra raises concerns about using the virus without knowing about the cell type affected by RCD virus and the receptor it uses to infect the susceptible cells. Matson too, believes that when a new agent is to be released into the environment the burden of proof for safety is much higher and knowing the pathogenesis of the agent is a prerequisite.

7.4 Epidemiology.

It has been proposed that the transmission of virus occurred mainly through the faecal-oral route (Morrise et al. 1991) or through contact between rabbits (Cancellotti & Renzi, 1991). Gregg and House (1991) showed that rabbits died after exposure to the faeces of rabbits which had recovered from RCD for two and four weeks. However, Collins et al. (1996) were unable to find viral antigen in faeces and confirm faeces as a source of infection.

Spread of infection via aerosols is considered an unlikely means for transmission as there is little evidence for the formation of aerosols from coughing or saliva (Gregg, et al. 1991). Studies at CSIRO-AAHL have not shown the transfer of virus between rabbits 50 cm apart. However, the potential transfer may have been inhibited by the lack of air movements within the building (Lenghaus, et al. 1995).

Lenghaus and others, 1994 (cited in CSIRO-1 report, 1996), were able to show that pens which had housed RCD infected rabbits remained contaminated for up to four weeks after removal of the infected rabbits.

Subsequent studies showed that rabbits which had died of RCD and held at 22oC for up to 20 days were a potential source of virus (CSIRO-1 report, 1996). It is not clear what significance to attach to this finding but the carcase is clearly a potential source of virus for rabbits, for mechanical spread by carrion eaters such as dogs, foxes and birds, and for the contamination of insects.

Insects are not considered to be vectors of the virus in Europe (Villafuerte et al, 1995), however, recent work in Australia provides some evidence that insects play a role in the distribution of the virus there (Cooke, 1996. CSIRO-1 report, 1996).

The European rabbit flea Spilopsyllus cuniculi, the Spanish rabbit flea Xenopsylla cunicularis and Culex annulirostris mosquitoes were able to spread RCD between laboratory rabbits, by feeding from infected and then control rabbits. Generally, RCD virus could only be spread if the transmission occurred immediately (Lenghaus et al. 1995). However, New Zealand does not have either of the two rabbit fleas.

Bushflies, Musca vetustissima, have been shown in the laboratory to mechanically carry the virus between infected rabbits and transmit the disease when they seek out moist discharge from around the eyes and nose (CSIRO-1 report, 1996). They have also been apparent in large numbers during outbreaks of RCD in South Australia.

Blowflies have been shown in the laboratory to carry RCD virus from dissected infected rabbits and presumably contaminate the environment even though they probably do not contact live rabbits directly.

The investigation of the spread of RCD virus from Wardang Island through mainland Australia (Cooke, 1996) provides some evidence for the transmission of virus by insects during the months of September and October. Free-flying or wind-assisted insects, bushflies, mosquitoes and blowflies could have been responsible for carrying the virus. Windspeed, wind direction, air temperature and rainfall are the major factors influencing the activity and distribution of insects.

Field experience in Australia has shown that RCD outbreaks in Victoria coincided with abundant mosquitoes and in Western Australia they coincided with large numbers of mosquitoes and blow flies (Norbury, 1996). The CSIRO-1 report, 1996, shows that the number of new outbreaks each month was highest in spring and autumn and lowest in summer and winter. This also suggests that insects may be involved in the spread because the activity of flies and mosquitoes is favoured by temperate conditions and restricted by low temperatures in winter and heat and aridity in summer.

Despite reasonable evidence that insects are involved in the transmission of RCD virus, Wardhaugh and Rochester (1996) state that the slowing in the rate of spread and the small number of new cases during summer of 1995-96 could not be explained by a lack of vectors. In fact, field samples show that substantial numbers of flies and mosquitoes persisted throughout the summer. It is conjectured that the hot summer conditions reduced the survival times of the virus in the environment and this reduced the rate of spread. In north eastern South Australia, for example, air temperatures during summer commonly reach 45oC and soil temperatures can be 60oC or more for up to six hours each day (Cooke, 1996). Although virus may survive 3 to 7 days at 37oC it is destroyed in 1 to 120 minutes at 56oC. Consequently survival times for virus in carcases or in the soil surface or on day flying insects is likely to be very short. Temperatures in rabbit burrows in these conditions are unlikely to be less than 27oC.

Smid and others (1991) has shown that the virus in suspensions of liver, spleen, lung and kidney from infected rabbits can survive storage on dry cloth at room temperature (approximately 20oC) for up to 105 days and for at least 105 days at 4oC.

Thirteen of 16 pools of insects including bush flies, blowflies and mosquitoes collected in areas where RCD was found were shown to be positive for the presence of RCD virus by PCR and one of these pools was shown to contain infectious virus (CSIRO-1 report, 1996).

If winged insects are involved then it is likely that the virus will spread more readily in some seasons than in others. Insect transmission can be expected to add to the underlying contact transmission within social groups of rabbits. However, in the absence of vectors, and with high summer temperatures the number of diseased rabbits may decline. The disease may die out locally because it kills its host so rapidly.

In the relatively mild summers of 1993, 1994, 1995 of the United Kingdom, RCD has a seasonal incidence in farmed rabbits and in the unusually warm summer of 1995 there was a greater spread of the disease than the earlier summers (Trout pers. comm with Cooke, 1996).

The inadvertent spread of virus by people should not be underestimated. Rabbit fibre pedlars, who sheared and purchased fibre even from infected rabbits and carcases, were thought to have been responsible for the spread of the disease in the People’s Republic of China. They travelled long distances with contaminated hands, clothes, shoes and scissors (Xu, 1991). Fuller et al. (1993), in describing the first known clinical cases of RCD in the United Kingdom, report that the virus was probably spread from a rabbitry in Berkshire to another in Hampshire by moving infected kits to a foster mother.

Crosby and McLennan (1996) reviewed the potential vectors for RCD virus in New Zealand and identified eight species of fly in two categories:

  • potential direct-contact vectors (transferring the virus to healthy rabbits), and
  • potential indirect contact vectors (contaminating rabbit grazing areas with the virus).

Three flystrike species of blowflies have the potential to be both direct-contact and indirect-contact vectors. Four carrion species of blowflies could also be indirect-contact vectors. One species of fleshfly has the potential to be a vector. These flies are more common in warmer months, but little is known of their seasonality and relative abundance in the most rabbit-prone areas. Crosby and McLennan considered that mosquitoes and blackflies are unlikely to be potential vectors because they carry virus for a short term only and are not abundant in rabbit prone areas.

There are a number of predatory and scavenging vertebrates which could transport the virus. The most important were reported as the ferret and dominican gull (Crosby and McLennan, 1996). However, humans were regarded as the most significant potential vector and there is the potential for strategic release of virus where the disease is needed to control rabbits.

In areas with high summer temperatures, transmission rates are predicted to be highest in the spring and autumn, as is observed in Australia. But in areas with moderate summer temperatures, which enable survival of the virus, there may be high summer transmission rates, as is observed in the United Kingdom.

A number of possible transmission mechanisms are reported in the literature, including contact with rabbits, their body fluids and excreta, vectors and aerosols. However, their importance in the field and their role in the epidemiology of RCD is unknown.

There is ample anecdotal evidence, however, that the virus can spread rapidly and over long distances. If RCD virus is to be used as a biological control agent it is important to know how the disease is spread and how it would behave in New Zealand. This knowledge is also important in protecting non-target populations such as commercial, pet and research rabbits. While supporters of the introduction claim that the Australian experience will provide the answers this may not be so, as there are environmental differences between the two countries and the pool of vector candidates in New Zealand is smaller than Australia.

The questions of vectors and survival of the virus outside of the rabbit in varying local and seasonal environments and in or on any carrier animals are not fully resolved for Australia and remain open for the New Zealand environment.

The range, number and quantitative presence of potential vectors in New Zealand is likely to be less than Australia, and possibly Europe.

Salman identified a number of key epidemiological questions that remain to be answered before a sound scientific decision can be made in favour of an introduction:

  • How is the virus transmitted from rabbit to rabbit?
  • What is the most effective mode of transmission?
  • What is the role of vectors in the transmission of the virus?
  • What are the host factors for the spread of the disease in the rabbit population?
  • What are the favourable environmental conditions for the spread of the disease in rabbit population?
  • What is the relationship between host susceptibility and dose of the virus?
  • Is there an interaction between the exposure of this virus and other disease agents in the rabbit population or other animal populations?
  • What are the limitations of the current diagnostic procedures for the disease in the rabbit population in terms of its sensitivity and specificity (i.e. accuracy)?
  • What is the potential for RCD virus to be a carrier of other viral agents?
  • How does RCD virus persist in the field?

My inability to confidently answer these questions highlights my uncertainty as to how the virus would behave in New Zealand.

7.5 Ease of contamination and method for multiplication and deployment.

The Application describes how the virus sourced from stocks in Australia is prepared for the inoculation of rabbits and although Figure 5.1 covers inoculum distribution and security there is little detail about the method of multiplication and deployment. Discussions with the Applicant Group reveal their intention to source subsequent supplies of virus from an Australian supplier. See also section 4.4.

In Australia the protocol for release required that feral rabbits were captured from the farm that was to be targeted, the rabbits were injected with the virus inoculum and the rabbits were to be released in the same area in which they had been captured. The latter is important as ‘foreign’ rabbits are likely to be rejected from a warren and these infected rabbits would not likely be a source of RCD infection. It is conjectured that the protocols for release were not always adhered to in Australia and this may have contributed to the variable results.

There are a number of additional explanations for the variable success of a deliberate release and these include:

  • release at the wrong time in relation to the possible interrelationship of the following factors: season, temperature/humidity, insect activity
  • immunity due to an earlier undetected epidemic of RCD
  • young rabbits in the population with maternal antibody
  • inadequate pre and post monitoring
  • very low rabbit population density
  • inoculated rabbits were spread too diffusely

The Application also makes reference to the use of oral baits for the release of the virus but there is no information as to how they would be prepared or used.

There has been variable success in initiating field infections by releasing inoculated rabbits in Australia.

If the virus is introduced into New Zealand there is no authority to control its use outside of the RPMS’s.

Conclusion

The scientific evidence leads me to believe that a calicivirus causes the disease and that the disease risk assessment will address the purity of the inoculum.

I consider the risk of adverse mutation to be no greater or less than other RNA viruses already in New Zealand but it is not zero.

Questions remain about host specificity and subclinical infections.

There is much that is unknown in regard to the pathogenesis and epidemiology of the disease and further work would be required to answer important questions such as the role of potential vectors in New Zealand and survival of the virus in the field. Without this information I am unable to predict the behaviour of RCD in New Zealand and therefore assess its potential as a biological control agent.

Previous Page TOC Next Page

Contact for Enquiries

Manager, Strategic Science Team
MAF Biosecurity New Zealand
PO Box 2526
Wellington
NEW ZEALAND

Phone: +64 4 894 0115
Fax: +64 4 894 0731
Contact this person

 
Print this page Print this page




WebSite survey